UNIVERSITY OF CALIFORNIA LIBRARY •-TjTf I'^T T TTF' ^T Digitized by the Internet Archive in 2007 with funding from IVIicrosoft Corporation http://www.archive.org/details/engichemiOOstilrich ENGINEERING CHEMISTRY. M Published by |^ The Chemical Publishing Co. | Easton, Penna. S Publishers of Scientific Books M Engineering Chemistry Portland Cement g Agricultural Chemistry Qualitative Analysis g Household Chemistry Chemists' Pocket Manual g Metallurgy, Etc. g Engineering Chemistry A Manual of Quantitative Chemical Analysis for the Use of Students, Chemists and Engineers. FIFTH EDITION BY THOMAS B. STILLMAM, M.Sc, Ph.D. Late Professor of Engineering Chemistry in the Stevens Institute of Tech- nology. Member : Societe Chimique de France ; Deutsche Chemische Gesellschaft, Berlin; American Chemical Society; Internat- ional Society for Testing Materials for Construction, Zurich ; American Electro-Chemical Society ; Society of Chemical Industry, London. WITH ONE HUNDRED AND FIFTY ILLUSTRATIONS EASTON. PA. : THE CHEMICAL PUBLISHING CO. 1916 LONDON, ENGLAND : WILLIAMS & NORGATE 14 HENRIETTA STREET, COVENT GARDEN, W. C. Copyright. 1895. by Edward Hart Copyright, 1900. by Edward Hart Copyright, 1905, by Edward Hart Copyright, 1910, by Edward Hart Copyright, 1916, by Edward Hart PREFACE. "The master is gone. The master's tvork is here." As this, the 5th edition of "Engineering Chemistry," was ap- proaching completion, the author, Thomas B. Stillman, Ph.D., died, August 10, 191 5. That the final completion was possible is due to the hearty co-operation and earnest work of many good friends to whom the editors extend herewith deepest thanks — namely to Clarence Carr, Captain U. S. N., I^ewis F. Lyne, Jr., Oil Specialty and Supply Co., E. G. Bashore, Chief Chemist, Babcock and Wilcox Company, Professor William Main, George H. Gibson, Harrison Safety Boiler Works, W. H. Fulweiler, Chemist, United Gas Improvement Co., R. E. Brueckner, M. E., R. Vuilleumier, Chief Engineer, Pintsch Compressing Co., and to the Kennecott Lab- oratories. AlvBKRT E. STII^IvMAN, E. M., M.Sc, Thomas B. Stili^man, M. E., Editors. August 10, 19 16. ^°\ CONTENTS. PAGE The Examination and Analysis of Coal and Coke i The Examination and Analysis of Limestone 65 The Examination and Analysis of Iron Ores 71 The Analysis of Iron Pyrites 78 The Commercial Sampling of Iron Ores 80 The Analysis of Blast Furnace Slacr 88 The Analysis of Manganese Ores 90 Methods for Copper, Lead and Zinc 91 Graphic Method for Blast Furnace Charges 95 The Blast Furnace as a Power Plant 98 Cast Iron Analysis no Foundry Chemistry 115 The Examination and Analysis of Steel 126 Analysis of Tin Plate 153 Alloys 160 The Chemical and Physical Examination of Portland and Natural Cements 195 Concrete 242 Analysis of Clay, Kaolin, Fire Sand, Building Stones, etc 250 Asphalt 273 Methods of Testing Coal Tar and Refined Tars, Oils and Pitches... 349 The Examination of Lubricating Oils 362 Remarks on Lubricants and Lubrication 421 Oils Used for Illumination 441 Linseed Oils 453 Fuel Oil 455 Ultimate Analysis of Oils 472 Soap Analysis 475 The Analysis of Paris Green 494 Paint Analysis 496 The Chemical and Physical Examination of Paper 544 Water Analysis 570 Filtration of Water 600 Water for Locomotive Use 606 • Feed Water Heaters 611 Fuel Economizers 623 Gas Analysis 627 Flue Gas Analysis 627 Analysis of Illuminating Gas 655 Gas Calorimetry 665 CONTr:NTS V PAGE Manufacture of Water Gas 670 Natural Gas 677 Acetylene 679 Valuation of Coal for Gas Production C90 Manufacture of Oil Gas C95 Practical Photometry 705 Pyrometry 718 Appendix 728 Anah'sis of Cylinder Deposits 728 Analysis of Cyanides 729 Analysis of Welsbach Mantles ; 730 Analysis of Gelatine Dynamites 731 Tables 732 Determination of Phosphorus Pentoxide 736 Iron Determinations 738 Index 745 LIST OF ILLUSTRATIONS. FIG. PACK 1. Abbe pulverizer 19 2. Emerson calorimeter 35 3. Emerson calorimeter 37 4. Ignition cup 39 4a. Crucible for coke analysis 55 4b. Coke shatter test apparatus 64 5. Apparatus for limestone analysis 67 6. Carbon dioxide apparatus 69 7. Bunsen valve stopper 71 8. Allen apparatus for ferrous oxide 74 9. Parallel system of ore sampling 80 10. Ziz zag system of ore sampling 80 11. Rope net system of ore sampling 81 12. Diagram for moisture sampling 82 13. Cone sampling 83 14. Face sampling 84 15. Graphic method for calculating blast furnace charges 96 16. Apparatus for filtration in determination of carbon by the direct combustion method 131 17. Vanier combustion train 133 18. Colorimeter 135 19. Camp's agitator 141 20. Apparatus for determination of sulphur by the evolution method 145 21. Method of sampling tin plate 155 2.2,. Le Chatelier specific gravity apparatus 210 23. Vicat apparatus 213 24. Details for briquette 216 25. Details for gang mold , 217 26. Mold for compression test pieces 217 2.T. Form of clip 220 28. Riehle U. S. standard 1000-pound automatic cement tester 221 29. Ball bearing block for testing machine 122, 30. Apparatus for making accelerated test 225 31. Appearance of ball for different consistencies of cement paste.. 230 32. Soundness pat showing shrinkage cracks 231 33. Soundness pat showing disintegration cracks 231 34. Soundness pat with top surface flattened for determining time of setting 231 35. Correct method of molding cement pat 232 36. Method of mounting Gilmore needles 233 37. Correct method of filling briquette mold 234 38. Correct method of troweling surface of briquettes 234 39. Hydraulic compression machine for cement and concrete testing 243 40. Curves showing relative compressive strength of concretes 244 41. Impact test on oil mixed concrete ; 248 42. Riehle U . S. standard testing machine 255 43. Tagliabue freezing apparatus 258 44. Standard, automatic transverse brick testing machine 261 45. Standard rattler 265 46. Three gang abrasion cylinder 269 47. Olsen standard impact tester 270 48. Sohmer hydrometer 282 I,IST OF ILI^USTRATIONS Vll FIG. PAGE 49. New York testing laboratory oven 284 50. Apparatus for determining soluble bitumen 286 51. Apparatus for determining paraffine scale 294 52. Penetrometer 296 53. Instrument for determining the consistency of road binders.... 300 54. Smith ductility machine, electrically driven 303 55. Smith ductility machine, hand -power 304 56. Dow briquette mold 307 57. N. Y. State tester 312 58. Reeve centrifuge extractor 316 59. Recovery apparatus 319 60. Dulin rotarex 320 61. Sieve shaker 321 62. Apparatus for water in tar 349 63. Melting point apparatus 351 64. Breaking point apparatus 352 65. Light oil distillation 353 66. Special apparatus for heavy oil analysis 357 67. Tagliabue hydrometer 364 68. Chart for oil analysis by specific gravities 365 69. Williams-Westphal balance 366 70. Modified Westphal balance 367 71. Araeo picnometer of Eichhorn 368 ^2. Apparatus for cold test of oils 370 73. Engler viscosimeter 373 74. Saybolt Standard Universal viscosimeter 375 75. Tagliabue's improved viscosimeter 376 75a. Doolittle viscosimeter 379 76. Apparatus for determining the flashing and burning points of combustible liquids, Cleveland cup 383 TJ. Pensky-Martens tester 385 78. Separatory funnel for oils 392 79. Apparatus for determining congealing point of fatty acids from oils 394 80. Apparatus for determining congealing point of fatty acids from oils 394 81. Apparatus for melting points of fatty acids from oils 395 82. Apparatus for melting points of fatty acids from oils 395 83. Graduated tube 401 84. "Gray" carbon residue flask 402 85. Machine for determining coefficient of friction of oils 410 86. Martens' lubricant friction machine 412 87. Wisconsin testing apparatus for oils 443 88. Tagliabue open tester for illuminating oils 446 89. Tagliabue closed tester for illuminating oils 446 90. The Foster automatic oil tester 446 91. The Saybolt electric oil tester 449 92. The Stammer colorimeter 452 93. Microphotograph — Linen fiber before treatment 547 94. Microphotograph — Linen fiber after treatment 547 95. Alicrophotograph — Poplar wood fiber 547 96. Microphotograph — Poplar wood fiber 547 97. Microphotograph — Spruce wood fiber 548 Vlll LIST OF ILLUSTRATIONS FIG. PAGE 98. Microphotograph — Spruce wood fiber (after pulping) 548 99, Microphotograph— Cotton fiber 548 100. Microphotograph— Coniferous fiber 548 lOi. Wendler machine for paper testing 559 102. Apparatus for absorption of blotting paper 564 103. Apparatus for testing resistance to elongation 565 104. Water evaporating apparatus 575 105. Apparatus for ammonia determination 591 106. Wolff's colorimeter 592 107. Dervaux water purifier 601 108. Dervaux water purifier 601 109. Filter press 603 no. Filter press 604 111. Heater and filter press 605 112. Kennicott water softening plant 610 113. Goubert closed feed water heater 612 114. Cochrane feed-water heater 616 115. Oil separator and drain 617 116. Cochrane metering heater 619 117. Green fuel economizer 624 118. Modified Elliott apparatus 628 1 19. Orsat apparatus 632 120. Hankee pipette 635 121. Improved Hankee pipette 635 122. Hahn apparatus 636 123. Continuous sample method for gas 638 124. Heat carried away by chimney gases — chart 650 125. Per cent. CO2 in gas — chart 651 126. Complete gas analyzing apparatus 655 127. Standard U. G. I. Hempel burette 656 128. Double absorption cuprous chloride pipette 657 129. Tutwiler and Bond hygrometer 658 130. Diagram of illuminating gas anabasis 661 131. Junkers' gas calorimeter 666 132. Junkers' gas calorimeter 667 133. Lowe water gas apparatus 672 134. General acetylene generator 681 135. Automatic acetylene generator 682 136. Acetylene buoy 687 137. Sun valve 688 138. Newbigging's gas production plant 692 139. Pintsch gas plant, side elevation 697 140. Pintsch gas plant, ground plan 698 141. Bunsen photometer 707 142. Standard bar photometer 715 143. Harcourt pentone lamp 716 144. Candle-power computer 717 145. Electrical resistance pyrometer 719 146. Heraeus quartz glass thermometer 721 147. Le Chatelier electric pyrometer ^2.'>i 148. Indicating and recording unit — Bristol pyrometer 724 149. Fery radiation pyrometer 725 A. Bunsen valve 74i ENGINEERING CHEMISTRY. THE PROXIMATE ANALYSIS OF COAL AND COKE. Determination of Moisture, Volatile and Combustible Matter, Fixed Carbon, Ash, and Sulphur. Take a weighed platinum crucible (capacity about 25 cc.) and weigh in it 1.5 grams of the powdered coal. Transfer to a dry- ing oven and heat to 103° C. for 40 minutes; cool in a desic- cator/ and weigh. Loss is moisture. Grams Crucible -|- cover -f coal 26. 1 1 7 Crucible + cover 24.617 Coal taken 1500 Crucible -j- cover + coal, before drying 26.117 Crucible + cover ~|- coal, after drying 26.109 -o 2 " Moisture 0.008 0.008 X 100 , . ^ =; 0.53 per cent, moisture. The crucible containing the dried coal is now heated over a Bunsen burner for y/2 minutes, then over the blast-lamp for 3I/2 minutes more, taking care that the cover of the crucible fits closely. Cool in the desiccator. Loss in weight equals volatile and combustible matter plus ^ of the sulphur. Grams i Crucible -\- cover + coal, before heating 7 minutes. . . . 26.109 Crucible + cover -j- coal, after heating 7 minutes 25.569 Volatile and combustible matter -f ^ S 0.540 0.540 X 100 1-5 -f- V2 s. 36 per cent, volatile and combustible matter 1 The desiccator should contain H2SO4 (C. P.), not CaClg, as finely pulverized coal is very hydroscopic even in pressure of CaClo. i:nginkering chemistry The crucible and contents are now heated over a Bunsen burner (lid of crucible removed) until all carbonaceous matter is consumed. Where the combustion is extremely- slow, it can be expedited by introducing into the crucible a slow current of oxygen gas so regulated that the contents of the crucible are not disturbed. Replace cover of crucible when ignition is complete, cool in desiccator and weigh. Grams. Crucible + cover + coal, before complete combustion 25.569 Crucible + cover -j- residue, after complete combus- tion 24.669 Fixed carbon + ^ S 0.900 o.Qoo X 100 ^ ^ -, , -^ 1=3 60 per cent, fixed carbon + >^S. f Crucible + cover + residue of coal, after complete combustion (ash) 24.669 Crucible and cover 24.61 7 Ash 0.052 O OS2 X 100 — ^ =,: 1.46 per cent. ash. 1.5 o -+ i- Resume. Per cent. Moisture 0.53 Volatile and combustible matter -f- ^ S 36.00 Fixed carbon + ^ S 60.00 Ash 3.46 Total ■. 99-99 It is necessary, now^, to determine the percentage of the sul- phur present in the coal and subtract it from the amounts of volatile and combustible matter and fixed carbon. The method is as follows : Determination of Sulphur in Coal by the Eschka-Fresenius Method. Preparation of Sample and Mixture.^ — Thoroughly mix on glazed paper i gram of coal and 3 grams of Eschka mixture. The mixture is prepared by thoroughly incorporating tv^o parts of 1 Amer. Soc. Testing Materials, 1914. ENGINEERING CHEMISTRY 3 magnesium oxide with one part of sodium carbonate by passing through a 40-mesh screen. By this method of preparation the mixture attains a uniformity comparable with that of the labo- ratory sample of coal and thorough incorporation is, therefore, more easily effected. Transfer to a No. o Royal Berlin crucible or platinum cruci- ble of similar size and cover with about i gram of Eschka mix- ture. Ignition. — On account of the amount of sulphur contained in artificial gas, it is preferable to heat the crucible over an alcohol, gasoline or natural gas flame or in an electrically heated muffle (procedure (a) below). The use of artificial gas for heating the coal and Eschka mixture is permissible, provided the crucibles are heated in a muffle (procedure {h) below). (a) Heat the crucible, placed in a slanting position on a triangle, over a very low flame to avoid rapid expulsion of the volatile matter, which tends to prevent complete absorption of the products of combustion of sulphur. Heat the crucible slowly for 30 minutes, gradually increasing the temperature and occasionally stirring until all black particles disappear, which is an indication of the completeness of the procedure. {h) Place the crucible in a cold gas muffle and gradually raise the temperature to 870 to 925° C. (cherry-red heat) in about I hour. Maintain this maximum temperature for about i^ hours and then allow the crucible to cool in the muffle. Subsequent Treatment. — Remove and empty the contents into a 300 cc. beaker and digest with 100 cc. of hot water for y^ to ^ of an hour, with occasional stirring. Filter and wash the in- soluble matter by decantation. After several washings in this manner, transfer the insoluble matter to the filter and wash five times, keeping the mixture well agitated. Treat the filtrate, amounting to about 250 cc, with 10 to 20 cc. of saturated bromine water, make slightly acid with hydrochloric acid and boil to expel the liberated bromine. Make just neutral to methyl orange with sodium hydroxide or sodium carbonate solution, then add i cc. of normal HCl. Boil again and add slowly from 4 ENGINEERING CHEMISTRY a pipette with constant stirring lo cc. of a lo per cent, solution of barium chloride (BaClo.2H20). Continue boiling for 15 minutes and allow to stand for at least 2 hours at a temperature just below boiling. Filter through an ashless filter paper and wash with hot distilled water until a silver nitrate solution show^s no precipitate with a drop of the filtrate. Place the wet filter containing the precipitate of barium sulphate in a weighed plat- inum or alundum crucible, allowing a free access of air by fold- ing the paper over the precipitate loosely to prevent spattering. Smoke the paper off gradually and at no time allow it to burn with flame. After the paper is practically consumed raise the temperature to approximately 925° C. and heat to constant weight. Thus : Grams. Amount of coal taken 1.016 Crucible -f BaS04 16.533 Crucible 16.51 1 BaSO^ 0.042 S = 0.0057 gram. 0.0057 X 100 , ^ ^ iiLJi:! — 0.56 per cent. S. 1. 016 ^ ^ Taking this amount and subtraction one-half of it from the volatile and combustible matter of the coal, and one-half from the fixed carbon, the coal analysis will be : Per cent. Moisture 0.53 Volatile and combustible matter 35-72 Fixed carbon 59-72 Sulphur 0.56 Ash 3.46 Total 99-99 In most cases the sulphur in coal exists combined with iron to form ferrous sulphide; it also occurs as calcium sulphate, or both forms may be present in the same coal. To determine the sulphur trioxide combined with the lime, take 10 grams of the finely powdered coal and digest at a gentle ENGINEERING CHEMISTRY 5 heat, 2 hours, in a solution of sodium carbonate (i.io). It is fihered, washed with hot water, the filtrate made acid with hydrochloric acid, boiled 5 minutes and the sulphur trioxide pre- cipitated with barium chloride solution. Determination of Sulphur by the Peroxide Fusion Method. This method is most conveniently carried out in the bomb which is a part of the Parr calorimeter/ the fusion resulting from a heat deter- mination being especially well suited to this purpose. The charge con- sists of 0.5 gram of the air-dry laboratory sample of coal, i gram of potassium chlorate pulverized to about 20-mesh, and 10 grams by measure of sodium peroxide of the grade regularly prescribed for calorimetric purposes. For mixtures intended only for sulphur determinations, oven- drying is unnecessary. The coal and potassium chlorate are first added to the bomb or fusion cup and thoroughly mixed, being careful to break down any lumps that may form. The sodium peroxide is then added, the container closed and ingredients thoroughly mixed by shaking. After igniting and cooling the charge, dissolve the fusion in a covered beaker, using 150 cc. of water. Add concentrated hydrochloric acid just past the neutral point. This will require from 25 to 30 ^c. of acid. Add I cc. of concentrated HCl (specific gravity 1.19) in excess. Filter and wash with hot water, making the final bulk of the solution approximately 250 cc. Heat to boiling and precipitate the sulphate by adding 10 cc. of a hot 10 per cent, solution of barium chloride. Continue the boiling for 15 minutes and allow to stand for at least 2 hours at a temperature just below boiling. Filter, wash and ignite as described under the Eschka-Fresenius method. Particular care should be taken in washing the precipitate obtained by this method in order to remove all soluble salts which are found in the fusion process. Determination of Sulphur in the Washings from an Oxygen Bomb Calorimeter. After the combustion, the bomb is washed out thoroughly with dis- tilled water, and the washings collected in a 250 cc. beaker. Six to 8 cc. of dilute (1:1) hydrochloric acid containing some bromine water are then added and the solution heated to boiling. The insoluble matter is filtered off and washed free from sulphates with hot water. The filtrate and washings, which should have a total volume of 200 cc, are made just neutral to methyl orange with sodium-hydroxide or carbonate solu- * A simpler, inexpensive bomb is described in Journal Am. Chem. Soc, Vol. 25, p. 184, 1903; see also Noyes, "Organic Chemistry for the Eaboratory," p. 21. 6 Engine;b:ring che^mistry tion, I cc. of normal HCl, is added, and the procedure is completed as to time with the Eschka-Fresenius method.) (If an odor of SO2 is detected in the escaping gases from the bomb, the washings cannot be used for the sulphur determination. In such cases a higher oxygen pressure is required. Twenty to 25 atmospheres of oxygen is usuallj'^ sufficient to completely oxidize all sulphur to SO3 in bombs of 400 to 600 cc. capacity. Some difficulty may be experienced in securing complete oxidation of all sulphur in bombs of less than 300 cc. capacity. The analyst should in all cases check his results from time to time with the Eschka-Fresenius method.) Determination of Phosphorus in Coal and Coke. To the ash from 5 grams of coal in a platinum capsule is added 10 cc. of nitric acid and 3 to 5 cc. of hydrofluoric acid. The liquid is evaporated and the residue fused with 3 grams of sodium carbonate. If unburned carbon is present 0.2 gram of sodium nitrate is mixed with the carbonate. The melt is leached with water and the solution filtered. The residue is ignited, fused with sodium carbonate alone, the melt leached and the solution filtered. The combined filtrates, held in a flask, are just acidified with nitric acid and concentrated to a volume of 100 cc. To the solution, brought to a temperature of 85° C, is added 50 cc. of molybdate solution and the flask is shaken for 10 minutes. The precipitate is washed sik times, or until free from acid, with a 2 per cent, solution of potassium nitrate, then returned to the flask and titrated with standard sodium hydroxide solution. The alkali solution may well be made equal to 0.00025 gram phosphorus per cubic centimeter, or 0.005 per cent, for a 5-gram sample of coal, and is 0.995 of i/5 normal. Or the phosphorus in the precipitate is determined by reduction and titration of the molybdenum with per- manganate. Directions for Sampling Coal, or Preparing Samples for Chemical Analysis. Containers for Shipment to Laboratory. Samples in which the moisture content is important should always be shipped in moisture-tight containers. A galvanized iron or tin can with a screw top which is sealed with a rubber gasket and adhesive tape is best adapted to this purpose. Glass fruit jars sealed with rubber gaskets may be used, but require careful packing to avoid breakage in transit. Samples in which the moisture content is of no importance need no especial protection from loss of moisture. Preparation oe Laboratory Samples. The method of preparing a suitable sample for the various analytical Engine:e:ring chemistry determinations that are required in coal analysis should conform as nearly as practicable to the following requirements : 1. A uniform distribution of coal and impurities must be maintained throughout the process of reducing to the final powdered sample. This should be insured by thorough mixing between each dividing or quarter- ing process, and by having due regard to the ratio of size of largest impurities and weight of sample. 2. Unrecorded changes in moisture during the procedure of sampling must be reduced to a minimum. Coal, especially when in a pulverized condition, is exceedingly sus- ceptible to change in moisture content. The general tendency is loss of moisture on dividing to finer sizes. This may amount to several per cent, in coal that has not been previously air-dried. The equilibrium point of the moisture in pulverized coal varies with the temperature and humidity of the air. Coal that has reached equili- brium with respect to moisture in an atmosphere of low humidity will reabsorb moisture if placed in an atmosphere of higher humidity.^ 3. Due regard must be given to the tendency of coal to absorb oxygen and deteriorate in heating value. The time of air-drying must, therefore, be as short as possible and should correspond to the statements given under Method No. i, below. Methods of Sampung, The following alternate methods of preparing laboratory coal samples are recommended as conforming to the theoretical requirements set forth in the preceding paragraphs, and as being commercially practicable for technical coal analysis : Method No. i. — Samples of coal received by the laboratory which exceed 5 pounds in weight, or 4-mesh (length of openings in sieve not to exceed 0.20 inch) in size should be rapidly crushed to 4-mesh, mixed and reduced to not less than 5 pounds. This portion is then transferred to a weighed sheet-metal pan, spread out to a depth of i inch and at once weighed. The pan is placed in a special drier" and the coal allowed to air-dry in circulating air at 10° to 15° C. above the sampling-room tem- perature, until the rate of moisture loss is less than o.i per cent, per hour, as shown by 2 weighings made at intervals of 2 to 4 hours. In most cases Appalachian bituminous coal and anthracite will be air-dry ^ For experimental data on moisture changes in coal samples, see N. W. Lord, "Experimental Work of the Chemical Eaboratory," Bulletin No. 28, Bureau of Mines, pp. 13-16, 1911. ^ For details of air-drying oven see Bownocker, lyord and Somermeier, "Coal," Bulletin No. 9, 4th Series, Ohio Geological Survey, p. 312, 1908; or F. M. Stanton and A. C. Fieldner, "Methods of Analyzing Coal and Coke," Technical Paper No. 8, Bureau of Mines, p. 4, 19 12; or E. E- Somermeier, "Coal, Its Composition, Analysis, Utilization and Valuation," p. 71, McGraw-Hill Book Co., 1912. 8 ENGINEERING CHEMISTRY if left in the drier over night. IlHnois coals may require 48 hours and lignites "jz hours for air- drying. Immediately after the last weighing has been made, the entire sample should be rapidly pulverized to lo-mesh size, mixed and reduced to 500 grams with an enclosed riffle sampler^ whose sub-divisions are not more than ^ inch apart. This 500 gram portion is at once transferred to the porcelain jar (8.95 inches in diameter and 9.65 inches high) of an Abbe ball mill, sealed air-tight and then pulverized to 60-mesh. Bituminous coals require about Yz hour and anthracites about 2 hours to pulverize to 60-mesh. The jar should contain about one-third of its volume of i inch, well- rounded flint pebbles, and should be rotated at about 60 revolutions per minute. The coal is removed from the porcelain jar by emptying the contents on a }/^-inch screen, which is vigorously shaken a moment to detach the coal from the pebbles. The sample is then reduced to the final laboratory sample of approximately 60 grams by successively halving it with a small, enclosed riffle sampler. All of the final sample should then be put through the 60-mesh sieve, and at once transferred to a 4-ounce wide-mouthed bottle which is securely ^zlosed with a well-fitting rubber stopper. To avoid moisture change, the sieve should be covered while sifting. Usually a few particles of coarse material remain on the sieve. These must be rubbed down on a bucking board or mortar to 60-mesh, and thoroughly mixed with the sample. (If one could be certain that all of each sample would pass the 60-mesh sieve it would be preferable to omit sieving, since it has a tendency' to segregate the particles of slate and pyrite and offers an opportunity for change in moisture content. On the other hand, if the sieving is omitted there is great danger of rather coarse particles of slate and pyrite being present in the final sample.) The mixing and reducing of the sample after removal from the ball mill should be done rapidly to minimize loss or absorption of moisture. The total time elapsing from the opening of the ball mill jar to the stoppering of the laboratory sample bottle need not exceed 2 or 3 minutes. The total loss in weight of samples while air-drying is reported as air-drying loss. The moisture in the coal "as received" = moisture in . , . , , , ^ 100 — air-drying loss , . , . an air-dned coal X — r air-drying loss. Method No. 2. — Samples of coal if larger than 4-mesh (0.20 inch) should be rapidly reduced to 5 pounds at 4-mesh or finer as in Method No. I. ^ For details of riffle sampler see Bulletin No. 9, 4th Series, Ohio Geological Survey, p. 313, 1908; or E. E. Somermeier, "Coal, Its Composition, Analysis, Utili- zation and Valuation," p. 73, McGraw-Hill Book Co., 19 12. KNGINEERING CHEMISTRY This 5-poimd portion is quickly passed through a suitable crushing apparatus, — rolls or enclosed coffee-mill type of grinder, — adjusted to crush to 10 or 20-mesh size. A 60-grani moisture sample should be taken, without sieving, immediately after the material has passed through the crushing apparatus. This sample should be taken with a spoon from various parts of the 10 or 20-mesh product, and should be placed directly in a rubber-stoppered bottle. The main portion of the sample is further pulverized until all passes through a 20-mesh sieve. It is then thoroughly mixed and reduced on the riffle to about 120 grams, which is pulverized to 60-mesh by any suitable apparatus without regard to loss of moisture. After this sample has been passed through the 60-mesh sieve it is again mixed and divided on a small riffle to 60 grams. The final sample should be transferred to a 4-ounce rubber-stoppered bottle. Moisture is determined at 105° C. on i gram portions of the 60-mesh sample and on 5 gram portions of the 20-mesh sample by the method described in this report. In the latter case the drying is continued 1^/2 hours. The analysis of the 60-mesh coal, which has become partly air- dried during sampling, is calculated to the dry basis by dividing each result by i minus its content of moisture. The analysis of the coal "as received" is computed from the "dry-coal" analysis by multiplying by i minus the total moisture found in the 20-mesh sample. Coal containing visible superficial moisture should be spread out in weighed pans and allowed to air-dry as in Method No. i, or at room temperature ; otherwise considerable loss of moisture will take place while crushing to lo-mesh size. The percentage of loss in weight is recorded and the analysis of the air-dried sample corrected to the "as received" basis, as described in Method No. i. Notes on the Two Methods oe Sampling. The first method is preferable for the preparation of laboratory samples that are intended for highly accurate analysis. The unavoidable loss of moisture during sampling is less than by the second method, especially in the case of samples of wet or freshly mined coal. Such samples lose moisture rapidly on exposure to air. Air-drying should not be unneces- sarily prolonged, as otherwise an appreciable loss of heating value from oxidation takes place. The second method can be more readily adapted to the apparatus at hand in the ordinary laboratory, as it does not require the special air- drier or ball mills for pulverizing the coal. The method admits hand- ling a large number of samples in a short time. The moisture obtained by this method is usually somewhat less than that obtained by the first method. In the case of coals that have lost part of their moisture con- tent through being exposed to the atmosphere, like the usual commercial lo engine:e:ring che:mistry shipments, this difference need not exceed 0.5 per cent. Wet samples must be partly air-dried. Coals which are high in sulphur and slate should preferably be pul- verized to 80-mesh. The disk pulverizer is not adapted to the fine grinding of coke and anthracite ; the abrasive action of the coke on the iron surface of the disk pulverizer seriously contaminates the sample ; and anthracite is heated by the rubbing surfaces to a degree that may change the com- position of the sample. A chipmunk jaw crusher is well adapted to crushing the sample re- ceived at the laboratory to 4-mesh size, and a roll crusher for reducing the 4-mesh material to 10 to 20-mesh size. The rolls have one disad- vantage, in that with some coals, flakes are formed which must be broken up by rubbing through a sieve before the sample can be reduced on the riffle to quantities less than 500 grams. On the other hand, the rolls have a large capacity and are easily cleaned. Coffee or bone mill types of grinders may be used for grinding to 10 or 20-mesh size. They should be entirely enclosed and provided with a covered hopper and receptacle of sufficient capacity to hold the entire 5-pound sample. A new porcelain jar ball mill and pebbles should always be tested for abrasion before use. This may be done by grinding 500 grams of sugar for a period of 2 hours, and then determining the ash in the sugar; or by keeping a record of the loss in weight of jar and pebbles and the weight of coal ground. Determination of Moisture. The report of the sub-committee II on moisture in coal states : In view of our own experience and that of the chemists who co-operated with a sub-committee of the International Committee on Analyses^ it seems needless to strive at present in ordinary work for a very high degree of refinement in the determination of moisture. So sensitive are coals to humidity changes of the air that it is evidently only by chance that 2 or more analysts will reach the same results for moisture in a given coal, especially if they live in different cities or make the tests on different days in the same place. To the truth of this the report of the above-mentioned International Committee bears abundant testimony. The variations therein shown are probably due, in part, to lack of realization on the part of many of the analysts of the magnitude of the changes in moisture content that may arise during the transfer of the coal from the containing vessel to the drying receptacle and during the weighing ^ Proceedings, Eighth International Congress of Applied Chemistry, Vol. 25, p. 4, 1912. i:ngine;e:ring chemistry ii operation. Nevertheless the chances of variation are so serious that the opening statement above is fully justified. I. Approximate Method. Use a pair of shallow weighing capsules with ground caps or other well-fitting covers. Suitable forms are indicated below. Heat these under the conditions at which the coal is to be dried, stopper or cover, cool over concentrated sulphuric acid for 30 minutes and weigh. Dip out with a spoon or spatula from the container two portions of coal of about i gram weight each, put these quickly into the drying vessels, close, and weigh at once. An alternative procedure (more open to error), after transferring an amount sli"-htly in excess of I gram, is to bring to exactly i-gram weight (±0.5 milligram) by quickly removing the excess weight of coal with a spatula. The utmost dispatch must be used in order to minimize the exposure of the coal until the weight is found. When the 20-mesh, 5-gram sampling is used, it is to be weighed in a similar measure with an accuracy of 2 milligrams. Further procedure: Quickly place the vessels open in a preheated oven (at 104 to 110° C.) through which passes a current of air dried by concentrated sulphuric acid. Close the oven at once and heat for i hour, or in the case of the 20-mesh, 5-gram sampHng, for i>^ hours. Then open the oven, cover the capsules quickly and place in a desiccator over concentrated sulphuric acid. When cool, weigh. Notes. I. Although watch glasses ground to fit and with clamp are most effec- tive drying vessels on account of their shallowness, other vessels will be found more convenient. The form that commends itself most, because the ash determination can be made on the moisture sample without transfer, is a porcelain cup of the size and shape represented by No. 1716 in the 1913 catalog of Eimer & Amend and by No. 1338 (No. 2 Royal Meissen porcelain capsule, % inch deep and i^ inches in diameter) in the catalog of the Scientific Materials Co. of Pittsburgh. At the Bureau of Mines this cup is used with a well-fitting aluminum cover. The cup is 20 to 22 millimeters deep and 38 to 40 millimeters wide. Glass capsules, as used by S. W. Parr and recommended by the International Committee, are likewise suitable. Those tried at the Bureau of Standards are 15 millimeters deep and 25 millimeters wide, somewhat shallower than those of Parr. The shallower the drying vessels are, consistent with convenient handling, the quicker and more perfect is the drying. The Parr capsules have the upper part of the wall ground on the. outside and the cap is ground on the inside, leaving a smooth edge, a feature which facilitates 12 ENGINEERING CHEMISTRY transfer of the coal to the ashing vessel if the same sample is to be used for determination of the ash. 2. The oven must be so constructed as to have a uniform temperature in all parts and a minimum of air space. The air current must be rapid enough to renew the gas in the oven frequently when the oven holds from 6 to 12 vessels of coal. The cylindrical form of oven shown in the Technologic Paper No. 8, of the Bureau of Mines, page 5, and holding 6 crucibles, is well adapted for the purpose ; also a new rectangular oven of the same Bureau, holding 12 crucibles and measuring inside 12J/2 inches high by 4 inches wide and 141^ inches long. The approximate air space in each of these ovens is 0.05 cubic foot. With this type of oven, of small air space, it has been found at the Bureau of Mines that the air must be renewed from two to four times a minute to obtain the maximum loss in weight from the coal in I hour. The International Committee in its report recommends, when using air, to remove one of the capsules from the oven at the expiration of 30 minutes, to continue heating the other for 30 minutes longer, and to accept the higher loss in weight, if a difference is shown, as may happen with certain coals. Our committee has not deemed it advisable to prescribe this precaution, both on account of the added labor involved in testing four portions of a sample instead of two and because of the probable error in the determination of the true moisture content is so large as to render the precaution of doubtful value. It may, however, serve a useful purpose at times and is a permissable modification of the procedure given in the approximate method. If used, however, a statement to that effect should accompany the report of analysis. II. Methods of Greater Accuracy. Method No. i. — This method is like the approximate method, but in- stead of air use a current of dry carbon dioxide gas. After the hour's heating, open the oven, cover the capsules, place them in a vacuum desic- cator over concentrated sulphuric acid and exhaust the desiccator. When cool, slowly admit dry air and weigh at once. Note. — Exhaustion of the desiccator is necessary in order to avoid serious error in weight from the presence of carbon dioxide in the capsules when these are weighed. Although carbon dioxide is absorbed by coal at room temperature there is no absorption above 100° C. Nitrogen gas is to be preferred to carbon dioxide because its density is so near that of air that it will be unnecessary to displace it from the capsules before final weighing. If this gas is used, a vacuum is unnecessary. Method N^o. 2. — This method is applicable to the test of only i sample at a time. It is like Method No. i, but uses a current of dry ENGINKERING CHEMISTRY I3 nitrogen gas and instead of a shallow capsule, a U-tube with well-ground stoppers, and any special form of oven in which the tube can be hung at a temperature of 104 to 110° C. Fill the tube with dry nitrogen before taking its weight empty, and weigh always with a counterpoise tube of about the same displacement and weight. Introduce about a gram of the coal quickly through a short and wide-stem funnel without attempt to secure a weight of exactly i gram. Before hanging the tube in the preheated oven pass nitrogen to displace all air, and continuously while heating. When the last trace of moisture has disappeared from the outlet of the tube, remove from the oven and let cool with the gas still passing. When cool close the cocks, hang in the balance case for 15 minutes and after opening one cock weigh with counterpoise. The counterpoise need not be filled with nitrogen. As a check the water given off may be collected in sulphuric acid and weighed, care being taken to keep the absorption vessel full of nitrogen. The weight of water thus obtained is a little higher than that found indirectly. Method No. 3. — This method is for use when time does not press. Dry in a vacuum desiccator over sulphuric acid of maximum concentra- tion for 3 and 7 days, longer if necessary. The vacuum should be high — not over 3 millimeters of mercury press- ure — and should be checked by a manometer. The capsules mentioned above may be used. Before evacuating, fill the desiccator with an inert gas, and before opening the desiccator carefully let in air dried by sul- phuric acid. Weigh immediately. This method is easy of execution and is sound of principle, since a possible error due to loss of gaseous constituents is negligible. It is important, however, when using a high vacuum, to produce this gradually since, if suddenly produced before most of the moisture and air have escaped there may be projection of the coal from the capsule. It has not been deemed advisable to recommend the xylene method of Constam (boiling a large weight of coal with xylene and collecting and measuring the water that distils over) since, though promising, the method has not been subjected to exhaustive test. The same statement applies to a method said to be in use in Germany, which consists in heating the coal in a vacuum at the temperature of boiling alcohol for an hour. So far as tests made at the Bureau of Standards allow of judging, the method justifies the claims that have been made for it, and it will be tested further. An article on the subject by P. Schlaflfer has been recently published.^ ^ Z. angew. Chem., Vol. 27, p. 52 (1914). 14 EnginE£:ring chemistry Determination of Volatile Matter. I. MuFFi,E Method. It is recommended that for volatile matter determinations a lo gram platinum crucible be used having a capsule cover, or one fitting closely enough so that the carbon from bituminous or lignite coals does not burn awa}' from the under side. The capsule cover fits inside of the crucible and not on top. The crucible with i gram of coal is placed in a muffle maintained at approximately 950° C. for 7 minutes. With a muffle of the horizontal type, the crucible should not rest on the floor of the muffle but should be supported on a platinum or nichrome triangle bent into a tripod form. After the more rapid discharge of the volatile matter, well shown by the disappearance of the luminous flame, the cover should be tapped lightly to more perfectly seal the crucible and thus guard against the admission of air. II. ALTERNATE Method. One gram of coal is placed in a platinum crucible of approximately 20 cc. capacity (35 millimeters in diameter at the top and 35 millimeters high). The crucible should have a tightly fitting cover, as above. The crucible is placed in the flame of a Meker burner, size No. 4, having approximately an outside diameter at the top of 25 millimeters and giving a flame not less than 15 centimeters high. The temperature should be from 900 to 950° C. determined by" placing a thermo couple through the perfo- rated cover, which for this purpose may be of nickel. The junction of the couple should be placed in contact with the center of the bottom of the crucible. Or the temperature may be indicated by the fusion of pure potassium chromate in the covered crucible (fusion of KsCrOs, 940° C). The crucible is placed in the flame about i centimeter above the top of the burner and the heating is continued for 7 minutes. After the main part of the gases have been discharged the cover should be tapped into place as above described. When the gas pressure is variable it is well to use a U-tube attachment to the burner to show the pressure. For lignites a preliminary heating of 5 minutes is carried out, during which time the flame of the burner is played upon the bottom of the crucible in such a manner as to bring about the discharge of volatile matter at a rate not sufficient to cause sparking. After the preliminary heating the crucible is placed in the full burner flame for 7 minutes as above described. For coke or anthracite a capsule cover or nested crucible should always be used. ENGINKERING CHEMISTRY 15 Determination of Ash. One gram of coal, either freshly weighed or that which has been used for the moisture determination, is ignited in a shallow porcelain capsule.^ A low temperature should at first be used, obtained by placing the capsule just above the tip of a Bunsen flame turned down to 2 or 3 inches in height. Frequent stirring with a platinum or nichrome wire is necessary. After a considerable part of the carbon is burned off the flame should be turned up and the heat increased to low redness. The capsule should finally be transferred to a muffle maintained at dull or cherry-red tem- perature between 700° and 750° C. From 20 to 30 minutes will ordinarily be required for the first part of the process, while 10 minutes should be ample for the heating in the muffle. If a muffle is used for the whole process, the heating should be started with the muffle cold or on the hearth at a low temperature. Corrected Ash. The application of a correction for sulphur present in the iron pyrites depends largely upon the use to be made of the results. For technical purposes it may well be omitted. For comparative purposes, especially where use is to be made of the pure coal or unit values, it should be applied. Five-eighths of the sulphur present in the pyritic form, if added to the ash, would restore the iron sulphide to the original form as weighed. While with certain types of coal, especially those extensively used for steaming purposes, averaging 15 to 20 per cent, ash, it is evident that there is a volatile ash constituent of considerable importance due to hydration of clayey material, in our present state of information as to the uniform distribution of this constituent it does not seem advisable to incorporate it in technical analyses. For a comparative study, however, a correction for this type of ingredient cannot be avoided. The factor which has received extended application is an increase of 8 per cent, of the ash as weighed to represent this volatile constituent. AllowabIvR Variations. No carbonates present Carbonates present Coals with more than 12 per cent, ash vSame analyst, per cent. 0.2 05 Different analyst, per cent. 0-3 o.i The analyses of a few representative coals are here given : ^ Such dishes as are listed under the name of "Gluh-Schalchen," No. 5837 in Greiner and Friedrichs, catalog, 19 12. i6 ENGINEERING CHEMISTRY "Bog Head Cannel'' Coae. Per cent. Moisture 0.60 Volatile and combustible matter 71.30 Fixed carbon 21.20 Sulphur 0.30 Ash 6.60 Total 100.00 "Pittsburgh Bitetminous" Coae. Per cent. Moisture 1.28 Volatile and combustible matter 37-36 Fixed carbon 57-33 Sulphur 0.72 Ash 3.31 Total 100.00 "Penn Anthracite," Wiekes-Barre, Dee. & Hudson Canae Co.'s "Vein No. 5." Per cent. Moisture 4.182 Volatile and combustible matter 4-283 Fixed carbon 85.320 Sulphur 0.794 Ash 5.421 Total 100.00 It is found in practice that coal from the same vein or seam varies in composition with the size of the coal, the percentage of ash increasing as the size of the coal diminishes. Thus, samples collected from the Hauto Screen Building of Lehigh Coal and Navigation Co., Pa., gave the following: size of coal Moisture Volatile matter Fixed carbon Sulphur Ash Total Egg Stove Chestnut. . . Pea Buckwheat. 1.722 1.426 1.732 1.760 1.690 3-518 4.156 4.046 3.894 4.058 88.489 83.672 80.715 79045 76.918 0.609 0.572 0.841 0.637 0.714 5.662 10.174 12.666 14.664 16.620 100 100 100 100 100 These coals are separated into different sizes according to th( ENGINKERING CHE:mISTRY 17 mesh of the screen over which they pass. The sizes noted in the above table passed over and through sieve meshes of the follow- ing dimensions : Broken or grate size Egg Stove " Chestnut " Pea ' • Buckwheat No. i. Buckwheat No. 2. (Rice) Buckwheat No. 3. (Barle} through 4.00 in. over 2.50 m 2.50 1.75 '•75 T.25 1-25 0.75 0.75 050 50 0.25 Analysis of a sample of ash of a Welsh coal, by J. A. Phillips, gave: Per cent. Silica 26.87 Akimina and iron oxide 56.95 Lime 5.30 Magnesia 1.19 Sulphuric acid 7.23 Phosphoric acid 0.74 Undetermined 1.72 Total 100.00 An analysis, by Gautier, of the ash of a sample of English coke, gave the following: Per cent. Silica 42.10 Alumina 3440 Calcium Carbonate 4.80 Magnesium carbonate 0.40 Calcium sulphate 12.55 Ferric oxide 5.28 Total 99-53 Method of Testing Coal, for Amount of Slate. A quick and useful method of determining the amount of slate in the small sizes of prepared coal is employed by the Dela- l8 ENGINEERING CHEMISTRY ware, I^ackawanna & Western Coal Department at its mines in Pennsylvania. When the railroad car is being loaded, samples of coal are collected which aggregate lo pounds. About one- fourth of this lo-pound sample is set apart at the testing house for the slate determination. The method is as follows : A solution is prepared by mixing sulphuric acid with water until the mixture shows specific gravity of 1.7 by hydrometer test. This solution is placed in an earthenware jar. A perfo- rated copper vessel, of several times the capacity of the coal sample, is suspended in the solution. On the sample being poured into the copper receptacle and agitated, the slate sinks, while the coal floats on the solution. The coal is skimmed off, washed, weighed and compared with the total weight of the coal and slate. This leaves nothing to the discrimination of an inspector as to what should be classed as slate.^ Resume. In selecting a sample of coal for analysis, it is absolutely essential that it be a representative one and quickly taken, during delivery of the coal, to prevent loss of surface moisture by evaporation. The sample should be collected in an air-tight vessel. The coal and vessel are weighed, the coal spread upon a non-absorbent surface and dried at 70° F. for 24 hours. It is then weighed with the receptacle — : the difference in weight represents surface moisture. The coal is passed through a crusher rapidly and then trans- ferred to an Abbe pebble pulverizer.^ This machine is air-tight and reduces all the coal so that it passes through a 200-mesh sieve. This pulverizer has the advantage that during the grind- ing of the coal no moisture is lost or absorbed. After pulveriz- ation the coal is immediately transferred to a bottle, tightly stop- pered — ready for the analysis. An example of the above operation can be stated as follows : 1 Consult: "Coal Ash," by John W. Cobb, B. Sc. (London), 7. Soc. Chem. Ind., Jan. 12, 1904, pp. 11-15. ''Fig. 1. KNGINKKRING CHEMISTRY 19 Grams. Weight of coal and can before air-drying at 70° F 4327. Weight of coal and can after air-drying at 70° F 4206. Weight of can 225. :. Weight of coal wet less can 4102. :. Weight of coal dry less can 3981. Moisture in coal after air-drying 24 hours at 70° F 121. Percentage of moisture after air-drying 2.9 Percentage of moisture in the air-dried coal, as per direction, page 18 2.1 Total moisture 5.0 It will be seen from this example that 2.9 per cent, of moisture with the coal is surface moisture and of no value to the pur- chaser. It is of value, however, to the contractor, for unless this Fig. I. surface moisture is determined and deducted, he is paid for water. Suppose the consignment of coal is 1,000 tons; 2.9 per cent, of this is 29 tons, in other words 58,000 pounds of surface water are present, costing the purchaser, with coal selling at $4 per ton. $116. Referring now to specification page 22, it will be noticed that moisture is allowed jn coal up to 1.5 per cent., all amounts over this the contractor pays a rebate or penalty for each ^ per cent. 20 ENGINEERING CHEMISTRY of moisture. This rebate is generally arranged (when large quantities of coal are sold) between the contractor and pur- chaser, and varies greatly. Taking the determination of moisture in the analysis given on page i8, amounting to 5.0 per cent, total, the rebate would be upon 3.5 per cent, moisture. SPECIFICATIONS FOR ANTHRACITE COAL. Department of Docks and Ferries, New York City, N. Y. Quality ArticeE I. All the coal to be furnished under this of Coal. Contract shall be free burning white ash and of some of the following grades : Susquehanna Coal Company, "Susquehanna"; Lehigh Valley Coal Company, "Wyoming" ; Delaware, Lackawanna & Western, "Scranton" ; Delaware & Hudson, "Lackawanna" ; Pennsylvania Coal Company, "Pittston" ; Lehigh & Wilkes-Barre, "Wilkes-Barre" ; or some other grade equal thereto and satisfactory to and approved by the Engineer. When demanded by the Engineer, satisfactory evidence shall be furnished of the name of the mine from w^hich the coal fur- nished is mined. All the coal required, except in cases of emergency, must be delivered direct from boats, and must be of the best quality of white ash anthracite coal. It shall be of the size hereafter mentioned, well screened, dry, clean, fresh mined and in good merchantable condi- tion. When it becomes necessary to deliver coal from yard or pocket, permission of the Engineer must be obtained. Bills of Article 2. An original Bill of Lading for every Lading. cargo of coal supplied must be furnished the Engineer by the Contractor, also an original Bill of Lading must be furnished the Engineer by the shipper from the shipping port. KNGINKERING CHEMISTRY 21 Unit ot AritclE 3- A ton in this contract shall be taken Weight. to mean a ton of 2,240 pounds. Size of Article 4. Unless otherwise directed or allowed Coal. all coal furnished shall be known as "pea" coal and conform to the following standard : That will pass through a screen having 3/J-inch square meshes and will pass over a screen having 5^ -inch square meshes. Analysis. ArticeE 5. All coal shall be subject to chemical analysis at any time the Engineer shall require 11. When analyzed it must contain not more than the fol- lowing per cents, of impurities : Per cent. Ash 12.00 Sulphur 1. 00 Moisture 1.50 In case the ash exceeds 12 per cent., but is less than 16 per cent., or the moisture exceeds 1.5 per cent., but is less than 3 per cent., the Engineer may, at his discretion, accept the deliv- ery for an amount as many per cent, less than the actual delivery as the per cent, of ash exceeds 12 per cent, or the per cent, of moisture exceeds 1.5 per cent.; or in case both ash and moisture arc excessive, for as many per cent, less than the actual delivery as the sum of the excess in per cent, of ash and moisture above 12 per cent, and 1.5 per cent., respectively, is of the total delivery, but no coal which contains more than 16 per cent, of ash, or 3 per cent, of moisture will be accepted, except at the discretion of the Engineer. Coal to be analyzed shall be taken by judicious sampling, and if such coal analyzed be not accepted by the department, the Con- tractor shall pay the cost of such analyses, and all other expense and damage to which the party of the first part is put by reason thereof : otherwise said expense shall be borne by the said party of the first part. 22 Engine:e:ring chemistry SPECIFICATIONS FOR BITUMINOUS COAL. Interborough Rapid Transit Co., N. Y. PrEuminary Specifications for Bituminous Coal. Kind of Coal must be a good steam, coking, run of mine, Coal. bituminous coal, free from all dirt or excessive dust, a dry sample of which will approximate the Transit Company's standard in heat value and analysis. Carbon Volatile matter Ash B. t. u. per lb. Sulphur l^fo 20% 9% 14100.O 1.5% The table of heat or B. t. u. values^ upon which the bonuses or deductions will be made is as follows : B. t. u. Tables of Values. For coal which is found by test to contain per pound of dry coal, from 15501 and above 28c. per ton above standard 15541 to 15500, both inclusive 27c. 15401 to 15450 15351 to 15400 " 15301 to 15350 " 15251 to 15300 " 15201 to 15250 " 15151 to 15200 " 15101 to 15150 " 15051 to 15100 " 15001 to 15050 " 14951 to 15000 " 14901 to 14950 " 14851 to 14900 " 14801 to 14850 " 14751 to 14800 " 14701 to 14750 " 14651 to 14700 " 14601 to 14650 " 14551 to 14600 " 14501 to 14550 " 14451 to 14500 " 14401 to 14450 " 14351 to 14440 " 14301 to 14350 " 14251 to 14300 " 14201 to 14250 " 14151 to 14200 " 14101 to 14150 " ^ For methods of determination of the B. t. .26c. •25c. .24c. " .23c. .22c. " .2IC. .20c. .19c. .i8c. .17c. . i6c. .15c. . 14c. .13c. . I2C. " .IXC. " .IOC. . 9c. . 8c. . 7c. . 6c. . 5c. . 4c. . 3c. . 2C. " . IC. standard value of coal see page 27 Knginee:ring chemistry 23 14051 i4CX)i 13951 I390I I385I 1 380 1 I375I I370I 1 365 1 I360I I355I I350I 1 344 1 I340I I335I I330I I325I 1 320 1 I3I5I I3IOI 1 305 1 I300I I295I 1 290 1 I285I 1 280 1 12751 I270I 1265 1 1 260 1 I255I I250I 1245 1 1 240 1 I235I I230I 1225 1 1 2201 I215I I2I0I 1 205 1 12001 12000 to 14100, both inclusive ic. per ton below standard to 14050 " 2C. " " to 14000 " 3c. " " to 13950 " 4c. to 13900 " 5c. to 13850 " 6c. " " to 13800 " 7c. " " to 13750 " ....8c. to 13700 " 9c. " " to 13650 " IOC. " " to 13600 " lie. to 13550 " I2C. to 13500 " 13c. to 13450 " 14c. to 13400 " 15c. to 13350 " i6c. to 13300 " 17c. " to 13250 " i8c. to 13200 " 19c. " to 13150 " 20c. to I3IOO " 2IC. " " to 13050 " 22c. to 13000 " 23c. to 12950 " 24c. " " to 12900 " 25c. " " to 12850 " 26c. to 12800 " 27c. to 12750 " 2Sc. to 12700 " 29c. to 12650 " 30c. to 12600 " 31C. to 12550 " 32c. to 12500 " 33^- to 12450 " 34c. to 12400 " 35c. to 12350 " 36c. to 12300 " 37c. to 12250 " 38c. to 12200 " 39c. to 12150 " 40c. " to I2IOO " 4IC. to 12050 " 42c. " and below 43^. 24 ENGINEERING CHEMISTRY Penalized Coal. Coal which is shown by analysis to contain less than 20 per cent, of volatile matter; 9 per cent, of ash; or 1.50 per cent, of sul- phur, will be accepted without a deduction from the bidder's price, plus or minus an amount for excess or deficiency of B. t. u. value, as herein provided. Where the analysis gives amounts for any or all elements in excess of these quantities, deductions will be made from the bidder's price in accordance with the tables of values of volatile matter, ash and sulphur below given, plus or minus the amount for excess or deficiency of the standard B. t. u. value, in addition to any other deductions which may be made as herein provided. Table of Volatile Matter. For coal which is found by test to contain per pound of dry coal : Over 20 per cent, and less than 20.5 per cent 2c. per ton 20.5 per cent, and over, and less than 21.0 per cent 4c. " 21.0 21.5 22.0 22.5 23.0 23-5 24.0 21.5 22.0 22.5 23.0 23-5 24.0 6c. 8c. IOC. I2C. 14c. i6c. i8c. Table of Ash. For coal which is found by test to contain per pound of dry coal : Over 9 per cent, and less than 9.5 per cent 2c. per 9.5 per cent, and over, and less than lo.o per cent 4c. ton lo.o 10.5 II.O 11-5 12.0 12.5 130 13-5 10.5 II.O 11.5 12.0 12.5 130 13-5 6c. 8c. IOC. I2C. 14c. i6c. i8c. 23c. ENGINEERING CHEMISTRY 2^ Table of Sulphur. For coal which is found by test to contain per pound of dr}^ coal : Over 1.50 per cent, and less than 1.75 per cent 6c. per ton 1.75 per cent, and over, and less than 2.C0 per cent loc. " 2.00 " " " 2.25 " 14c. " 2.25 " " " 2.50 " i8c. 2.50 " " 20c. " Should any lighter of coal delivered at the Company's docks contain less than 700 tons, a deduction of 7 cents per ton will be made from the price as determined by the B. t. u. value and analysis, in addition to any other penalty provided for herein. Should any lighter of coal delivered at the Company's docks be rejected by the Superintendent on account of excessive amount of coal dust, then a deduction of 25 cents per ton will be made from the price as determined by the B. t. u. value and analysis, for the coal taken from said lighter, in addition to any other penalty which may be made as herein provided. Should any lighter of coal be delivered in other than self-trimming lighters as herein provided, a deduction of 7 cents per ton will be made from the price, as determined by the B. t. u. value and analysis, exclusive of any other penalty which may be made as herein provided. Weighing. The Contractor's Bill of Lading will be checked by the Com- pany's scales. Should there be a deficiency of i per cent, or more between the bill of lading and the Company's weights, then the Company's weights will be taken as correct. The following is a convenient form for recording a coal analysis : — Report on Coal Analysis. 26 Kngine:e:ring chemistry The sample of coal received from you ,, marked tests as follows : Per cent. Moisture in coal by air-drying 24 hours at 70° F Moisture in air-dried coal, pulverized, heated Yz hour at 212° F. Total moisture Volatile and combustible matter Fixed carbon Sulphur Ash Total B. t. u. per pound As above stated, after the determination of the surface moist- ure, the coal is pulverized in an Abbe pulverizer and from this a portion is taken for the determination of the moisture (at 212° F.), the volatile and combustible matter, fixed carbon, sul- phur and ash, the result being : Per cent. Moisture (at 212° F.) 2.16 Volatile and combustible matter 8.56 Fixed carbon 77 ■'^1 Sulphur 0.46 Ash 11.55 Total , 100.00 These are the percentages referred to coal from which surface moisture has been removed. The surface moisture in this case amounts to 2.90 per cent. ; this, if added to the above, brings the total to 102.90 per cent. Hence a correction is necessary for the determinations: This is, 100 — 2.90 = 97.1 per cent., so that e:ngine:e:ring chemistry 27 each of the above determinations (except moisture, air-dried) multiplied by the factor 0.971 gives the following percentages on the original sample: Per cent. Moisture, by air-drying 2.90 Moisture at 212° F. after drying 2.10 Total moisture 5.00 Volatile and combustible matter 8.31 Fixed carbon 75-03 Sulphur 0.44 Ash 11.22 Total 100.00 Determination of the "B. t. u." in Coal by Combustion in Oxygen Calorimeter. The specifications are to be of two classes (a) and (b). The pro- cedure specified under (a) may be followed in tests where a tolerance of at least i per cent, is allowed. The procedure under (b) is to be used in all cases where the Hmit of tolerance is less than i per cent, and is to be followed in all cases of dispute. Under (b) any 3 determinations made at the same time on the same sample may be required to fall within a range of 0.3 per cent. Combustion Bombs. — The Atwater, Emerson, Mahler, Peters, Williams or similar bombs may be used. For (a) the lining material of the bomb need not be specified. The Parr calorimeter may also be used, but only in condition that both parties to the contract agree to its use. For (b) the burnt shell has a lining of platinum, gold, porcelain, enamel or other material which is not attacked by nitric and sulphuric acids, or other products of combustion. Calorimeter Jacket. — The calorimeter (except the Parr) must be pro- vided with a water jacket, having a cover to protect the calorimeter from air currents. The jacket must be kept filled with water. For (b) the water in the jacket must be kept within 2 or 3° C. of the temperature of the room and should be stirred continuously by some mechanical stirring device. Stirring of the Calorimeter Water. — The water in the calorimeter must be stirred sufficiently well to give consistent thermometer readings while the temperature is rising rapidly. The speed of stirring should be kept con- stant. For (b) a motor driven screw or turbine stirrer is recommended and the speed should not be sufficient to hold the temperature of the 28 ENGINEERING CHEMISTRY calorimeter more than 0.3 or 0.4° C. above that of the jacket, when the stirrer is allowed to run continuously. Accurate results cannot be ob- tained when too much energy is supplied by the stirring device or when the rate of stirring is too irregular. The portion of the stirring device immersed in the calorimeter should be separated from that outside by non-conducting material, such as hard rubber, to prevent conduction of heat from the motor or outside air. Thermometers. — Thermometers used shall have been certified by a government testing bureau and shall be used with the corrections given on the certificate. This shall also appl}^ to electrical resistance or thermo- electric thermometers. For (b) correction shall also be made for the temperature of the emergent stem of all mercurial thermometers, and for the "setting" of Beckmann thermometers. For accurate work either Beckmann or special calorimetric thermometers graduated to o.oi or 0.02° C. are required. Such thermometers should be tapped Hghtly just before each reading to avoid errors due to the sticking of the mercury meniscus, particularly, when the temperature is falling. A convenient method is to mount a small electric buzzer directly on the top of the thermometer and connect it up with a dry cell and a push button. The button should be pressed for a few seconds immediately before each reading. Oxygen. — Oxygen used for combustions shall be free from combustible material and for (b) it shall not contain more than 5 per cent, nitrogen and argon together. The total amount of oxygen contained in the bomb for a combustion shall not be less than 5 grams per gram of coal. But the combustion must be complete as shown by the absence of any sooty deposit on opening the bomb after firing. Firing Wire. — The coal in the bomb may be ignited by means of either iron or platinum wire. If iron wire is used, it should be of about No. 34 B. & S. gauge and not more than 10 centimeters (preferably 5 centimeters) should be used at a time. A correction of 1,600 calories per gram weight of iron wire burned is to be subtracted from the observed number of calories. Except, however, that this correction may be omitted from both the standardizations of bomb and coal combustions, provided the same amount of wire is used in all cases. Standardisation. — The water equivalent of a calorimeter can best be determined by the use of the standard combustion samples supplied by the Bureau of Standards. The required water equivalent is equal to the weight of the sample multiplied by its heat of combustion per gram and divided by the corrected rise in temperature. The calorimeter shall be standardized b}' the combustion of standard samples supplied by the Bureau of Standards, and used according to the directions given in the certificates which accompany them. A standardiza- tion shall consist of a series of not less than 5 combustions of either ENGINEERING CHEMISTRY 29 the same, or different standard materials. The conditions as to amount o£ water, ox3-gen, firing wire, method of correcting for radiation, etc., under which these combustions are made shall be the same as for coal combustions. For (b) in the case of any disagreement between contract- ing parties a check standardization shall be made at the time of test, but such standardization may consist of two or more combustions of standardizing samples. Manipulation. 1. Preparation of Sample. — The ground sample, which is in approxi- mate moisture equilibrium with the atmosphere, is to be thoroughly mixed in the bottle and an amount, approximately i gram, is to be taken out and weighed in the crucible in which it is to be burned. Coals which are likely to be blown out of the crucible should be briquetted. Standardiz- ing samples are also to be briquetted. After w^eighing, the sample should preferably be immediately placed in the bomb and this closed. This procedure is necessary to avoid sublimation when naphthalene is used. 2. Preparation of the Bomb. — The firing wire, if iron, should be meas- ured and coiled in a small spiral and connected between the platinum terminals, using, if necessary, a piece of platinum wire somewhat heavier than the iron wire, to make the connection. The platinum and the iron must both be clean. x\bout 0.5 cc. of water should be placed in the bottom of the bomb to saturate, with moisture, the oxygen used for combustion. When the crucible is put in place in the bomb, the firing wire should touch the coal or briquette of standard material. For the combustion of standardizing samples iron wire is preferable to platinum. 3. Filling the Bomb with Oxygen. — Oxygen from the supply tank is to be admitted slowly to avoid blowing the coal from the crucible, and the pressure allowed to reach 20 atmospheres for the larger bombs or about 30 atmospheres for the smaller bombs, so that the bomb shall contain an amount of oxygen sufficient for complete combustion, namely, at least 5 grams per gram of coal, or other combustible. When feasible, the bomb may be exhausted before filling to remove the nitrogen of the air, thus reducing the amount of the nitric acid formed. 4. Calorimeter Water. — The calorimeter is to be filled with the required amount of water, depending upon the type of calorimeter. The amount may be determined either by measurement in a standardized flask or by weighing. For {b) distilled water should be used and the amount de- termined by weighing. The amount must be kept the same as that used in standardization of the apparatus, or a correction applied for the differ- ence in weight. 5. Temperature Adjustments. — The initial temperature in the calori- meter should be so adjusted that the final temperature after the combus- tion, will not be more than 1° C. preferably about 0.5° C, above that of 30 e:ngine:e:ring chemistry the jacket, under which conditions the total correction for heat gained from or lost to the surroundings will be small when the rise of tempera- ture is 2 or 3° C. and the effect of evaporation will also be small. 6. Firing Current. — The electric current used for firing the charge should be obtained from storage, or dry cells having an electromotive force of not more than 12 volts. The circuit should be closed by means of a switch which should remain closed for not more than 2 seconds. When possible, it is recommended that an ammeter be used in the firing circuit to indicate when the firing wire has burned out. For {h) the electro- motive force of the firing battery shall not exceed 12 volts, since a higher voltage is liable to cause an arc between the firing terminals, introducing additional heat, which cannot be measured with certainty. 7. Method of Making an Observation. — The bomb when ready for firing, is to be placed in the calorimeter, the firing wires connected, the cover put in place and the stirrer and thermometer so placed as not to be in contact with the bomb or container. The stirrer is then started and after the thermometer reading has become steady, not less than 2 minutes after the stirrer is started, temperatures are read at i minute intervals for 5 minutes and the charge is then fired, noting the exact time of firing. Observations of temperature are then made at intervals depending upon the method to be used for computing the cooling correction. When the temperature has reached its maximum and is falling uniformly, a series of thermometer readings is taken at i minute intervals for 5 minutes to determine the cooling rate. 8. Titration. — After a combustion the bomb is to be opened, after allowing the gas to escape, and the inside examined for traces of un- burned material or sooty deposit. If these are found, the observations shall be discarded. If the combustion appears complete, the bomb is to be rinsed out and the washings titrated to determine the amount of acid formed. A correction of 230 calories per gram of nitric acid should be subtracted from the total heat observed. If the sulphur content of the coal is de- termined, the amount of sulphuric acid should be computed and an addi- tional correction of 1,220 calories per gram of H2SO4 should be sub- tracted, for the excess of the heat of formation of the sulphuric acid over that of nitric acid. Computation of Rksui^ts. The following method of computation is recommended, to take the place of the Pfaundler or other similar formulas for computing the cool- ing correction (radiation correction). Observe (i) the rate of rise (n) of the calorimeter temperature in degrees per minute for 4 or 5 minutes before firing; (2) the time (a) at which the last temperature reading is made immediately before firing; (3) the time (b) when the rise of temperature has reached six-tenths of KNGINEERING CHEMISTRY 3I its total amount (this point can generally be determined by adding to the temperature observed before firing, 60 per cent, of the expected^ temperature rise, and noting the time when this point is reached) ; (4) the time (c) of a thermometer reading taken when the temperature change has become uniform some 5 minutes after firing, (5) the final rate of cooling (ro) in degrees per minute for 5 minutes. The rate n is to be multiplied by the time b — a in minutes and tenths of a minute, and this product added (subtracted if the temperature was falling at the time (a) to the thermometer reading taken at the time a. The rate ^2 is to be multiplied by the time c — b and this product added (subtracted if the temperature was rising at the time c and later) to the thermometer reading taken at time c. The difference of the 2 ther- mometer readings thus corrected, provided the corrections from the cer- tificate have already been applied, gives the total rise of temperature due to the combustion. This multiplied by the water equivalent of the calorimeter gives the total amount of heat hberated. This result, cor- rected for the heats of formation of nitric and sulphuric acids observed and for the heat of combustion of the firing wire, when that is included, is to be divided by the weight of the charge to find the heat of combustion in calories per gram. Calories per gram multiplied by 1.8 give the British thermal units per pound. (See example.) The results should be reduced to calories per gram or British thermal units per pound of dry coal, the moisture being determined upon a sample taken from the bottle at about the same time as the combustion sam,ple is taken. For an accurate comparison of coals of different hydrogen content, by means of observation with the combustion bomb, the results which are obtained at constant volume should be reduced to heat of combustion at constant pressure, and to "net" instead of total heat. The former reduction is usually omitted as it is not of great importance ; the latter is, however, of considerable importance as the water formed by the com- bustion of coal in the bomb is all condensed and its latent heat measured, while in industrial practice the water usually passes off uncondensed and the latent heat is lost. The correction for water condensed may 1 When the temperature rise is not approximately known beforehand, it is only necessary to take thermometer readings at 40, 50, 60 seconds (and possibly 70 seconds with some calorimeters) after firing, and from these observations to find when the tem- perature rise had reached 60 per cent, of the total. Thus, if the temperature at firing was 2.135°, at 40 seconds 3.05°, at 50 seconds 3.92°, at 60 seconds 4.16°, and the final temperature were 4.200°, the total rise was 2.07° ; 60 per cent, of it was 1.44°. The temperature to be observed was then 2.07° + 1.44° = 3/05°. Referring to the observations at 40 and 50 seconds, the temperatures were respectively 3.05 and 3.92°. The time corresponding to the temperature of 3.51° was therefore ^.Sl — "? OS 40 4- -^ '-^- X 10 = 45 seconds. 40 + 3.92 - 3.05 32 ENGINKERING CHEMISTRY amount to nearly 5 per cent, for some bituminous coals while for anthra- cites and coke it is negligible. The results of determinations of calorific power shall be stated either as "total" heat of combustion or "net" heat of combustion. Example. Observations. Water equivalent = 2550 grams Weight of charge = 1.0535 Approximate rise of temperature = 3.2° 60 per cent, of approximate rise = 1.9° Time Temperature Corrected temperature 10-21 15.244° (Thermometer corrections from the certificate) 22 .250 23 .255 24 .261 25 .266 (a) 26 .272 15.276° Charge fired (b) 27-2 17.2°' (c) 31 18.500° 18.497° 32 .498 33 .497 34 496 35 .494 36 .493 Computation. Vx = 0.028° H- 5 = 0.0056° per minute; b — a = 1.2 minutes. The corrected initial temperature is 15.276° -r 0.0056° X 1.2 = 15.283°. r^ = 0.007° -^ 5 = 0.0014° per minute ; c — b =1 3.8 minutes. The corrected final temperature is 18.497° + 0.0014 X 3-8 = 18.502° Total rise 18.502° — 15.283° = 3.219° Total calories 2,550 X 3.219 = 8,209 Titration, etc = — 7 Calories from 1.0535 grams coal 8,202 Calories per gram 7,785 or British thermal units per pound 14,013 In practice, the time b — a will be found so nearly constant for a given calorimeter with the usual amounts of fuel that b need be determined only occasionally. Enginee:ring chemistry 33 Ali.owabi,e Variations. Per cent. *^ame analyst 0.3 Different analysts 0.4 Total heat of combustion shall refer to results computed as described above. Net heat of combustion at 20° shall refer to results computed as follows : The amount of watef in grams per gram of coal formed by combustion, multiplied by 580 is to be subtracted from the "total" heat in calories per gram to give the "net" heat in calories per gram, or the amount of water in pounds per pound of coal multiplied by 1,040 is to be subtracted from the total heat in British thermal units, to give the net heat in British thermal units, per pound. Combustion oi? Anthracites and Coke. For anthracites and coke, which have a high ash content and do not readily burn completely, the following procedure is recommended : The inside of the crucible is lined completely wdth ignited asbestos in a thin layer pressed well down into the angles. The coal is then sprinkled evenly over the surface of the asbestos. Otherwise the procedure is as previously described. Parr Cai,ori meter. The essential conditions for the operation of the Parr or peroxide calorimeter are as follows : The coal should be finely pulverized. While 6o-mesh is sufficient for bituminous coals, anthracites should be ground to at least lOO-mesh. The sodium peroxide used should be received in solder sealed tins and of a size suitable for emptying completely into the container for use, preferably a glass jar with level sealed cap. In addition to the reaction the peroxide serves as a diluent and the ratio necessary for a quiet reaction should be maintained, preferably 0.5 gram of coal to approximately 10 grams of peroxide. One gram of pulverized potassium chlorate is also used to advantage. A thorough and uniform mixing with the peroxide is secured by shaking in the closed cartridge. Coals with moisture above 2 or 3 per cent, must be oven-dried at 110° C. in the usual manner after weighing out, and before mixing with the chemicals. The correction factors to be subtracted are as follows : 1 The initial temperature is 15.27°; 60 per cent, of the expected rise is 1.9°. The read- ing to observe is then 17.2°. 3 34 ENGINEERING CHEMISTRY Deg. Cent. For each per cent, of ash 0.00275 For each per cent, of sulphur 0.005 For I gram of KCIO3 0.130 For electric fuse wire 0.008 For oxygen of bituminous coals for 0.5 gram 0.025 For oxygen of brown Hgnites for 0.5 gram 0.050 For oxygen of benzoic acid for 0.5 gram 0.124 The products of combustion, CO2 and H2O, combine with the chemical with the formation of heat, which amounts in each case to 27 per cent, of the total heat of the reaction. The corrections for ash, fuse wire, etc., in terms of the temperature rise together with radiation and thermometer corrections must first be subtracted from the indicated rise in temperature. The formula for the final calculation then becomes : Corrected thermometer rise X o-73 X total water , .^ . ^ — L2^^ = calorific value. 0.5 g. coal The Emerson Fuel Calorimeter. The scarcity of fuel, and the continually increasing price, have brought about conditions which reveal the fact that coal can no longer be bought by its name only, such as "New River" or 'Tocohontas," etc., as in nearly every district where the high grade coal is produced there are seams of the poorest variety of coal within a few miles distance, thus making the buying and selling of coal from the district name only, straightforward enough from the sellers' point of view, but possibly misleading to the uninformed buyer. The agreement between the seller and the consumer to fix the price of fuel as so much per ton for a fuel giving a given num- ber of British thermal units per pound, with a reduction or in- crease of price per 50 or 100 B. t. u. less or more than the speci- fied number has been adopted, and is at present operating satis- factorily among users of coal in large quantities. The buying of fuel on a guarantee to reach or excel a certain specified British thermal units per pound is adopted in some in- stances, and gives very satisfactory results. The fuel calorimeter here described is the so-called bomb calo- ENGINKERING CIIICMISTRY 35 36 ENGINEKRING CHEMISTRY rimeter of the Berthelot type, meaning that the combustible dur- ing ignition is retained in a stout receptacle in which is inserted an excess of oxygen gas under pressure to carry on the combus- tion. The determination of the heat of combustion is made by a calorimeter method, the bomb being placed in a watei calorim- eter during the combustion. The product of the rise in tem- perature in the calorimeter and the water plus the water equiv- alent of the calorimeter and its contents gives us directly the quantity of heat in calories per given weight of combustible. The combustible is ignited by means of a fine platinum wire rendered incandescent by the passage of an electric current. One terminal of the circuit is introduced into the interior by means of an insulated plug, the other terminal is grounded in the bomb. The combustible is introduced in a finely divided condition to insure complete combustion, and is held by a pan on a wire sup- port. Apparatus. Bomb. — The bomb is made of carbon steel, consisting of two cups joined by means of a heavy steel nut. The two cups are machined at their contact faces with a tongue and groove, the joint being made tight by means of a lead gasket inserted in the groove. The lining is of sheet nickel, spun in to fit, or of a double process high temperature porcelain. The bomb is made up tight with a milled wrench or spanner. The oxygen valve at the top of the bomb is made of steel. The pan holding the combustible is of platinum, and the supporting wire of nickel. The fuse wire should be platinum. Calorimeter. — The jacket is a double-walled copper tank be- tween the walls of which water is inserted. The calorimeter is made as light as possible, of sheet brass. Stirring Device. — The stirrer is directly connected to a small series motor and is enclosed in a tube to facilitate its action in cir- culating the water. The stirrer is mounted on a post on the calorimeter jacket as is the thermometer holder. ^B Fig. 3. 38 i:ngine:e:ring che:mistry The motor is driven by a iio-volt circuit, and should be placed in series with a i6-candle-power lamp. (55 watts, taking y^ am- pere.) If so desired, a motor driven by a battery can be speci- fied in ordering the apparatus. Oxygen Piping. — The piping for the insertion of oxygen under pressure is made especially strong and durable. The piping of small internal bore is made of heavy brass. The system is fitted with a hand nipple at one end to make the connection with the bomb, and the other end has a special fitting to grasp the oxygen supply tank. Immediately after each run the inside of the bomb should be washed out with a cloth moistened with a dilute solution of caustic soda and then with water. The linings should be frequently removed and the inner sur- face of the bomb under the linings should be coated slightly with oil. (This oil will in no way affect the operation of the bomb and can be left when the same is in operation.) The bomb is of steel and plated with nickel. This plating cannot be made an absolute protection against corrosion being placed as it is directly upon the steel and care should be taken that the entire surface should be covered with a slight film of oil after using the apparatus. MANIPUI.AT10N. Heat of Combustion of Solid fuels. — Place the lower half of the bomb in the holder and the platinum pan in the wire support after having wired the fuse, wire according to the accompany- ing sketch and following directions. To place the platinum wire, twist one end of the same into the small hole at edge of pan and extend the wire across the pan through the hole in mica, allowing it to dip sufficiently to be in contact with the fuel which is afterward placed in the pan. After passing through the mica, the wire is led to the side of bomb, where it is grounded at the binding post. The wire must in no case touch the pan except at the edge where the twisted contact is made. The fuse wire should be placed in series with e:ngine:e:ring chemistry 39 two 32 candle-power lamps in parallel when no- volt power circuit is used for firing. The fuel used is sampled, crushed, and powdered according to directions given below. Fill a test-tube or convenient weighing vial with the prepared sample and weigh the same accurately to ^/^^ of a milli- gram. Pour from this into the pan in the bomb until the pan is approximately half full. Weigh the vial again and the dif- ference of the above weighing gives the net quantity of fuel in the bomb. This weight should be greater than ^/^^ of a gram, and not more than iVio grams. For hard coal the maximum charge should be not greater than i gram. Hard coal should not be as finely divided as soft coal. (Through an 80-mesh sieve is sufficient.) JJTtca^ JPlatinum, Fig. 4. The upper half of the bomb is placed in position and the nut screwed down as far as may be by hand, care being taken not to cross the threads. The shoulder on the upper half of the bomb over which the nut makes bearing contact should be thoroughly lubricated with graphite and oil. Extreme care should be taken that no oil or grease is deposited on the lead gasket, as the bomb, when working properly, closes without the upper half turning on the gasket on account of the contact friction of the nut. Any oil on the lead gasket would tend to hinder the proper action in this respect. 40 ENGINEERING CHEMISTRY The large wrench is used to make the joint tight. The bomb is now ready to be filled with oxygen, and this is accomplished by means of the spindle valve at the top of the bomb. The nipple is coupled to the oxygen piping by means of the attached hand union. In handling the bomb, care should be taken not to tip or jar the same, as fuel may be thrown from the pan. The spindle valve on the bomb need only be opened one turn, and then the valve on the oxygen supply tank is very cautiously opened. The pressure gauge should be carefully watched and the tank valve so regulated that the pressure in the system shall rise very gradually. When the pressure reaches 300 pounds per square inch, the tank valve is closed, and then the spindle valve immediately after. The bomb should be immersed in water immediately to detect any possible leakages. (Preferably a glass jar, as slight leaks are detected by looking from the various sides.) The bomb is now ready for the calorimeter, which is prepared as follows : Nineteen hundred grams of distilled water are placed in the calorimeter can at a temperature about i^° below the jacket temperature (which temperature should h& approximately of the room temperature). The bomb is then placed in the calorimeter and the stirrer and thermometer are lowered into position as indicated by the preceding illustration. The ther- mometer is immersed about 3 inches in the water. The bulb of the thermometer should not touch the bomb. The terminals of the electric circuit used for firing should now be attached, one to the bomb and the other to the can ; this latter making contact with the pin in the plug at the bottom of the bomb. Care should be taken that the bomb does not touch the sides of the can. The Operation. The stirrer is now started, and allowed to run 3 or 4 minutes to equalize the temperature throughout the calorimeter. Readings of the thermometer are now taken for 5 minutes ENGINEERING CHEMISTRY 4I (reading to the i-iooth degree every ^ minute) at the end of which time the switch is turned on for an instant only ; which will be found sufficient to fire the charge. In course of a few seconds the temperature begins to rise rapidly and readings are taken as before, every half minute from the time of firing. Af- ter a maximum temperature is reached and the rate of change of temperature is evidently due only to the radiation to or from the calorimeter, we then continue our readings for an additional 5 minutes, reading every j^^ minute. These readings before the firing and after the maximum temperatures are necessary in the computation of the cooling correction. The time elapsed from the time of firing to the maximum temperature should be in no case more than 6 minutes. When through with run, replace bomb in the holder and allow the products of combustion from within to escape through the valve at the top of the bomb. Unscrew the large nut and clean the interior of the bomb. The inside of the nut should be kept greased; and also the threaded part at the top of the lower cup. The pan may be cleaned by boiling in dilute hydrochloric acid. Any slag clinging to the pan may be fused with sodium car- bonate. The fused mass dissolves in hot water. Computation. The data obtained during the run is used as follows : The difference between the temperature at maximum and the temperature at firing gives directly the apparent rise in tem- perature in the calorimeter. To this apparent rise, however, we must apply a cooling correction which is computed as follows : The change in temperature during the preliminary 5 minutes of reading divided by the time (5 minutes) gives the rate of change of temperature per minute due to radiation to or from the calorimeter and also any heating due to stirring, etc. This factor is called Rj, in like manner the readings taken after final temperature give Ro. These two rates of change of temperature also give the existing conditions in the calorimeter at the start and at the finish of the run. Therefore, the algebraic sum of the two rates divided by two will give the mean (or average) 42 ENGINEERING CHEMISTRY value of the rate of change of temperature during the entire run due to radiations to and from the calorimeter. This value mul- tiplied by the time from firing to maximum will give the total cooling correction. The cooling correction thus determined has been found by long experience to be a very close approximation to the radiation effects encountered when working under these above conditions. This latter quantity is either added to or subtracted from the apparent rise taken from the data of the run, accordingly as the balance of heat radiation is to the surroundings or from the surroundings. This is at once determined from an inspection of the data. Cooling correction is expressed : R, ± R, X time from firing to maximum. The correction rise of temperature divided by the weight of fuel used will give the rise per gram of fuel. This rise per gram times the weight of water plus "water equivalent" will give immediately the calories per gram of fuel, which is the result to be obtained. The result in calories per gram of fuel multiplied by the factor 1.8 gives the B. t. u. per pound of fuel. The water equivalent is the quantity of heat necessary to raise the metal parts of the calorimeter and the bomb 1° C. and is equal to the sum of the product of their weights times their specific heats. This water equivalent factor may be checked by burning in the bomb a fuel or combustible of standard heating value, the same having been determined accurately. Extreme care should be taken that such standardizing substances should be of prac- tically 100 per cent, purity and absolutely free from chemically or physically combined water. The value of such a standard substance in calories per gram is divided by the rise in temperature in the calorimeter per gram of sample and the result is the water plus the water equiva- ENGINKE^RING CHE:mISTRY ' 43 lent of the apparatus. The water being known, the water equiva- lent is thus determined. With a combustible of absolute purity this determination will check the value of the water equivalent as figured from the weights and specific heats of the material included in the acting parts of the calorimeter. Heavy Oils, Coke, Hard Coal, Etc. — The determination of the heat of combustion of heavy oils such as crude petroleum, and also of coke and extremely hard coals is best made by burning the same mixed with a ready burning combustible, such as a high-grade bituminous coal or C. P. carbon. This auxiliary com- bustible facilitates the complete combustion of the whole mixture in the case of coke and hard coal and with the heavy oil it acts as a holder and prevents rapid evaporation of the oil. The auxiliary combustible should be placed at the bottom of the pan and the coke, coal or oil sprinkled over it. The C. P. carbon or other auxiliary combustible should be dried with extreme care and carefully standardized as to the resulting rise in temperature per gram in the calorimeter when the same is completely burned. Fu^I. TESTING. Sampling. — The sampling is a most important element in the process of the determination of the heating value of a fuel. The rules which are stated below should be used with due considera- tion and judgment. The working conditions under which the fuel has to be sampled vary widely. For instance, the large coal pile of 10,000 tons, the shipment of 20 or 30 carloads, the whole mine, or a few tons may be the subject of test at hand and the method of sampling is slightly different in each case. If sampling is to be done at the mine, from a map of the mine several points for samples should be chosen in such manner as to give a fair sample of the whole. These points should be near the working face. A cut across the face from 3 to 6 inches in width and i inch in thickness should be made at each point. This cut should be 44 ENGINKERING CHEMISTRY taken out complete except for that part which would be rejected by the mine worker. The samples taken from these several points should be thrown together, crushed and mixed. The crushing should bring the size down to about 3/2 -inch cubes or less. The sample thus procured is spread out on an oilcloth or canvas and mixed. Then by drawing lines through the sample at right angles the fuel is systematically reduced in bulk and at the same time sampled by the usual process known as "quarter- ing," that is, the opposite quarters are taken out and the rest discarded. The sample taken away from the mine need not exceed that which may be contained in an ordinary 2-quart jar. This 2-quart sample is ground to a smaller size either with heavy mortar and pestle or proper grinding machine and then spread on a clean surface of glazed paper or cloth and after mixing the quartering is repeated. Regrinding and quartering are repeated until the sample is reduced to about 100 cc. This sample is finely powdered and the whole passed completely through a lOO-mesh sieve. This finely powdered sample is spread on a glazed paper and quartered for a small sample for the weighing tube (5 or 6 grams). The remainder should be tightly sealed and kept for future reference. If sampling is attempted on a large pile, say one of 10,000 tons or like, a grab sample with shovel if made with due care and judgment gives accurate results. Shovelfuls should be taken at intervals from the base of the pile to the top, and should be taken 18 inches or 2 feet under the surface of the pile if the pile has been long exposed to the weather. A considerable part of the sample should be taken from the larger pieces which are invariably found at the bottom of the pile. Any pieces which are too large to be handled with the shovel should be broken with a maul and parts of the fragments used for sample. Pieces which are often encountered containing practically nothing but slate or other forms of rock should not be entirely neglected, and it is herein that the whole operation relies on the judgment of the sampler. It is purely a matter of observation of the apparent e:ngine:e:ring chemistry 45 percentage of such material present in the pile that governs the sampler as to how much he shall include of the same in his sample. His ability to accomplish the above determines the suc- cess or failure of his work. The total sample taken from a pile need not exceed an amount which may be contained in a box 3 feet on a side. This sample is crushed, ground, mixed and quartered as was described above. If sample is to be procured from carload shipment a few well chosen shovelfuls from each car give the desired results : Sampling from shipment from vessels is accomplished by tak- ing a grab sample from the hold of the ship or by having the stokers instructed to lay aside a shovelful, at certain intervals in their work of filling the bucket hoists. Sampling from boiler tests is accomplished by saving an aver- age shovelful from each stoke period by the fireman. The same rules and machines which are used in the sampling of ore are equally good in the sampling of fuel. Moisture:. If the moisture is to be determined a quartered sample may be placed on a dried and weighed watch glass, and the weights taken before and after heating in the drying closet. When removing watch glass from closet it should be cooled in a desiccator. The temperature of the drying closet should be approximately 105° C. and the sample should be kept in the closet not less than 40 minutes. SAMPLE RUN. Sample No. 128 (dried). Run No. 2. Thermometer used, No. 11,258. Weight of tube and coal =: 7.8047 grams Weight of tube = 7.0742 grams Coal taken = 0.7305 gram Weight of water = 2,000 grams 46 ENGINEERING CHEMISTRY Readings of Thermometer. Time Temp. Time Temp. Time Temp. O.O 19.300 0.30 20.050 11.00 21.695 0.30 19.300 6.00 21.050 0.30 21.695 1. 00 19.300 0.30 21.420 12,00 21.690 0.30 19-305 7.00 21.570 0.30 21.690 2.CX3 19-305 0.30 21.650 13.00 21.690 0.30 19.310 8.00 21.690 0.20 21.690 3.00 19.310 0.30 21.700 Max. Temp. 0.30 19-315 9.00 21.700 4.00 19.320 0.30 21.700 0.30 19.320 10.00 21.700 5-00 19.325 Firing Temp. 0.30 21.700 0.002 Apparent rise in temperature = 2.375. Rate of change of temperature before firing = 0.005 = Ri- Rate of change of temperature after maximum temperature = R2. Average rate of change of temperature during run = 0.0015. Total cooHng correction := [0.0015 X 3-5 (min.)] = 0.005 [subtractive]. Total corrected rise in temperature = 2.370. Rise per gram of sample = 3.244. The water equivalent of bomb, calorimeter can, stirrer, etc. =r 434. Gram calories per gram of sample = (2,000 + 434) X 3.244 = 7,895. British thermal units per pound of sample == 7,895 X I-8 = 14,310. GOVERNMENT SPECIFICATIONS FOR PURCHASE OF COAL. Anthracite Coal. Price and Payment. Payment will be made at the price named in the proposal for the coal specified, corrected as indicated in Table IVa for variations in ash above and below the standard, as shown by analysis. Bituminous Coal. Price and Payment. On certification from the coal-inspection section of the United States Geological Survey that samples have been procured from deliveries con- stituting a completed order, payment will be made at once of 90 per cent, of the amount of the bill, 10 per cent, being withheld to protect the Gov- ernment against the delivery of coal of inferior quality and to offset any deduction that may be determined by the United States Geological Survey ENGINKKRING CHEMISTRY 47 from the average quality of all the deliveries on the order. On receipt of the Survey's report on the quality of the coal final settlement will be made. TABIvE IVa. — Price Corrections, in Cents per Ton, for Anthracite CoAE, Due to Variations in Ash in "Dry Coae" Above and Beeow the Established Standard. size of coal Ash in "dry coal" per cent. Furnace and egg stove Chestnut Pea Buckwheat 6. ox to 6.50 6.51 to 7.00 7.01 to 7.50 7.51 to 8.00 24 21 18 15 27 24 21 18 15 27 24 21 18 15 15 12.5 10 7.5 5 8.01 to 8.50 8.51 to 9.00 9.01 to 9.50 9.51 to 10.00 Contract price 10.01 to 10.50 10.51 to 11.00 ii.oi to 11.50 11.51 to 12.00 Contract price 12.01 to 12.50 12.51 to 13.00 13.01 to 13.50 13.51 to 14.00 15 18 21 24 Contract price 14.01 to 14.50 14.51 to 15.00 15.01 to 15.50 15.51 to 16.00 15 18 21 24 27 Contract price 12 10 8 16.01 to 16.50 16.51 to 17.00 15 18 21 24 27 6 4 17.01 to 17.50 17.51 to 18.00 18.01 to 18.50 18.51 to 19.00 5-0 7-5 10. 12.5 15.0 Contract price 19.01 to 19.50 19.51 to 20.00 20.01 to 20.50 20.51 to 21.00 21.01 to 21.50 21.51 to 22.00 4 8 14 21 32 48 Note. — Figures above heavy line represent cents per ton to be added to contract price; figures below heavy line represent cents per ton to be deducted from contract price. If the 10 per cent, withheld should prove insufficient to satisfy the claim of the Government on account of the delivery of fuel of low grade, the balance shall be deducted from the next succeeding order or orders. 48 i:ngine:e:ring chemistry Payment will be made at the price named in the proposal for the coal specified therein, corrected for variations in heating value and ash, as shown by analysis, above and below the standard established by the contractor in his proposal. The correction in price for variations in British thermal units is a pro rata one and is determined by the following formula : Delivered B. t. u. ^ . ^ , . • . u -j 5 — ;— r; X contract price = price to be paid. Standard B. t. u. "^ ^ ^ ^ For example, if a coal delivered on a contract guaranteeing 14,000 B. t, u. in coal "as received" at a price of $3 per ton shows by calorific test 14,300 B. t. u. in coal "as received," the price to be paid is found, by substitution in the formula, to be 14,300 X |3 = $3,064. 14,000 The price will also be further corrected for the percentage of ash. For all coal which by analysis contains less ash than that established in this proposal a premium of i cent^ per ton for each whole per cent, less ash will be paid, x^n increase in the ash content of 2 per cent, over the standard established by contractor w-ill be tolerated without deduction. Price Corrections, in Cents Per Ton, for Bituminous Coal, Due TO Variations in Ash in "Dry Coal" Above and Below THE Established Standard. Contract stan- dard, per cent of ash in "dry coal" Maximum limit for ash (per cent.) Schedule of variations in ash (percent.) Below- Higher limits 5 6 7 8 9 10 II 12 13 14 15 16 17 18 12 13 14 14 15 16 16 17 18 19 19 20 21 22 7 8 9 10 II 12 13 14 15 16 i7 18 19 20 Deduction ( cts . per ton ) 7-8 8-9 9-10 lO-II 11-12 12-13 13-14 14-1S 15-16 16-17 17-18 18-19 19-20 20-21 8-9 9-10 10- 1 1 11-12 12-13 13-14 14-15 15-16 16-17 17-18 18-19 19-20 20-21 21-22 9-10 lO-II II-I2 12-13 13-14 14-15 15-16 J6-I7 17-18 18-19 19-20 20-21 21-22 22-23 10-11 11-12 12-13 I3-H 14-15 15-16 16-17 17-18 18-19 19-20 20-2 1 21-22 22-23 12 II-I2 12-13 13-14 14-^5 15-16 16-17 i7-:8 18-19 19-20 20-21 21-22 22-23 18 12-13 13-H 14-15 15-16 16-17 17-18 18-19 19-20 20-21 21-22 25 13-14 14-15 15-16 16-17 17-18 35 In the specifications for the fiscal 1910-11 this premium will be 2 cents. m \-^ .- 1 i g s ? 1 ^P l8 . <5 1 « ON ~ * 1 = 1! B3 =s £t SI 1 « o Si' ?= K i> eg K ^ ^ S' S' 3^ H S •- S ro rr; ir> <-o -.00 !s rs M ►4 '^^^ uE- " " " " (N t^ (N 80 H « "3 « ^ 4 ^ 4 ^ r': C3N CO On 00 >-' CO ''-i « rr, r 06 88 d d t ^^iiii^^-^ ^ 8 ?> 8 t^ 2.^?. ?-.8 §€ ^1 § § ^ 88 x ^ ^ ^ .^ < : ^ = =.S^ J3AO H ^ :5t J W l* IT 4^ ^ « « < qSnoaqx 1 1 ^ 1 1 4 ^ "^ fe 1 1 1 ' ;5?t ^„ 1 1^ t^ t^ g-s 1 f nq 6 2 W s >< ■< o^ ' di _0 III 1^1 ill "« iiil ma 1=1111^11 Bi CU Oi u. Oi J- hja, ^D cias S a pu K2 "C be J O; K 5 11 s n c^ ^ Z : ■ fiilii! 1! 1 II S S-^ SM 0« ^ u a *' aappia ^ n V Q '4; '*<;&. 0 u^ 1, aj^^ aaHBtn sy ON 1 1 < V) 9 aii.uioA qsv 00 00 00 3N C o o n 2 ^ V '0 s -5 »5 ^ S 60 OS a So; 2 ^M-o i ° -^ -. ^ = tsit^ O 2 u = o (s 5 1 § 8 1 « ^ -5 ■^ in 00 !== = -S- !/j i; "^ i7 X .^■B i 14 = ^ = * >> u ~ ^ oxO" •§1 ^ = ^8 1^ -l^aj ^s 11 '0 Q.02 a- 5 (S (0^ f ^ \ n 50 e:ngine:e:ring che:mistry . VALUE OF COAL AS FUEL.* Requirements of Use. Coal is now burned for power production in gas producers and in boiler furnaces. For coals and lignites high in moisture or high in ash, the gas producer, used in connection with a gas engine, is best adapted to develop power, but for the generation of steam, which can be used for heating as well as for power, coal may be more conveniently burned in a specially constructed furnace under a boiler. Coal is burned under boilers for producing power, for drying various materials, or for warming buildings. The most valuable coal, therefore, is that which gives up the most heat to the boiler for a given weight burned. The value of a coal is indicated by the number of heat units it con- tains. This heating value is expressed in terms of British thermal units (abbreviated B. t. u.) per pound of coal, and is determined by means of a special apparatus called a calorimeter. In purchasing coal for any power plant the aim should be to obtain a fuel which, all things considered (such as equipment, price of coal, and cost of labor and repairs), will produce horse-power for the least cost. Experiments seem to indicate that almost any fuel may be burned with reasonable efficiency in a properly designed apparatus. The recognized requirements are as follows: (i) A uniform and continuous supply of fuel to the furnace; (2) an air supply slightly in excess of the theoreti- cal amount required for complete combustion; (3) a temperature suffi- ciently high to ignite the gases that are driven off from the fuel; (4) a complete mixture of these gases with the air supplied before they reach a cooling surface, such as the shell or tubes of a boiler. Factors Affecting Value. General Statement. — Some of the factors that may influence the com- mercial results obtained in a boiler are cost of the coal as determined by price and heating value, care in firing, design of the furnace and boiler setting, size of grate, formation of excessive amounts of cHnker and ash, available draft, and size of the coal. Moisture. — Coal as mined contains more or less moisture. It is ex- posed to the air in shipment and may either dry out or be drenched by rain. The moisture in the coal deUvered is worthless to the purchaser and really costs him a considerable amount in freight and cartage and in the loss of the heat required for its evaporation in the furnace. If all coal contained the same proportion of moisture, or if the moisture in coal delivered by a given dealer were constant in amount, the purchaser's problem, so far as this factor is concerned, would be simpHfied. * Bulletin 428, Dept. Interior, U. S. Geological Survey. i:ngine:e:ring che:mistry 51 Under present conditions the moisture is an important element in the valuation of a ton of coal. It is evidently necessary to consider the coal just as it is received in order to determine its value to the consumer, but chemical reports should be made on both "dry coal" and "coal as received." The report on dry coal is convenient for comparing several coals to de- termine the relation of each element to the others; this report is im- portant because the moisture in the same coal varies from day to day. The dry coal report is also convenient for comparing the performance of boilers burning the same or similar coals. Of several coals having a similar composition, the one that has the least moisture and the least ash will generate the most steam when burned under a boiler. Ash. — Earthy matter and other impurities that will not burn are classed as ash. In commercial coals the proportion of ash may range from 4 to 25 per cent. Coals containing small percentages of ash are the most valuable, not only because of their correspondingly higher heating ca- pacity but because they offer less resistance to the free and uniform distri- bution of air through the bed of coal in the furnace. The labor and cost of managing the fires and of handling the ashes are also correspondingly less and are items to be considered in the choice of a coal. With the ordinary furnace equipment there may be a considerable loss of efficiency and capacity through a large percentage of ash. With some kinds of equipment it has been found that as the ash increases there is a decided drop in both efficiency and capacity. In some experiments made to de- termine the influence of excessive amounts of ash, coal containing as high as 40 per cent, would generate no steam when fired on a chain grate, and therefore the efficiency and capacity of the plant would be zero.* Such coal is not only worthless, but its use involves a direct expense, due to the cost of handling it. Whether the result would be similar with equipment other than a chain grate has not yet been determined. How- ever, coals so high in ash that they are unsuited to boiler furnaces can be utilized in gas producers. Volatile Matter and Fixed Carbon. — The volatile part of some coals, shown in the analyses, may be all combustible, but it generally contains some inert matter. The amount of this differs in different coals, and therefore it is impossible to determine the heating value of any coal from its proximate analysis alone. Moreover, different coals that contain the same proportion of volatile matter do not behave ahke in the furnace. In order to determine the value of one coal as compared with another for the same purpose it is important to know both the chemical composition and the British thermal units. Of two coals of different character, the one that contains the higher proportion of fixed carbon is most easily burned, so as to give the ^ Abbott, W. ly., Some characteristics of coal as aflfecting performance with steam boilers, a paper read before the Western Society of Engineers, Chicago, III. 52 ENGINEERING CHEMISTRY maximum efiiciency. However, if the coal containing the higher volatile matter is properly burned in a suitably designed furnace it may be made equally efficient. Sulphur and Clinker. — Sulphur may be present in the free state, or, as is more common, in combination with iron or other elements. Other impurities with sulphur often form a clinker that shuts out the air and increases the labor of handling the furnaces. It is possible, however, to burn coals containing up to 5 per cent, of sulphur without great difficulty from clinkers. A little steam introduced under the grate will relieve much of the trouble. Clinker may be due to other causes than sulphur, as any constituents of the ash which are easily fusible may produce it. There is need of further investigation to determine the influence of sulphur and the elements that form ash in furnace fires during combustion. Size of Coal. — The size of the coal influences the capacity of any given equipment, owing to its effect on the draft. With a poor draft fine coal can not be burned in sufficient quantities to maintain the rated capacity. If thin fires are resorted to, the efficiency is usually lowered as a result of an excessive supply of air through holes in the fire. As a rule, when dust and very fine coal are fed into the furnace they either check the flow of air or are taken up by the draft and after being only partly burned are deposited back of the bridge wall ; or they may pass up the stack, to the annoyance of people in the vicinity of the plant. If this dust is completely burned in passing through the furnace there is of course no loss of fuel. Coal of uniform size forms the most satisfactory fuel, as it does not pack so closely as coal of different sizes mixed. In general it may be said that in any market the coal obtainable at the lowest price is the most economical, provided the furnace equipment is suitable. If the furnace is not so designed as to permit the use of the cheaper coal it should be changed. Heat Units. — The tests tend to show that, other conditions being equal, coals of similar composition are of value in proportion to the British thermal units, and the determination of these units in any coal will give approximately its value. It should be remembered, however, that the value of a coal for any particular plant is influenced by the character of the furnace, for all furnaces are not equally suitable for burning the many grades of coal. Aside from this factor, coals may be compared in terms of the British thermal units obtained for i cent, or on the cost per million heat units. Summary. — In 'the purchase of coal, then, attention should be given to the character of the furnace equipment and the load, the character of coal best suited to the plant conditions, the number of heat units obtain- able for a unit price, the cost of handling the coal and ash, and the possibility of burning the coal without smoke or other objectionable features. ENGINKKRING CHKMISTRY 53 COKE. Coke is the residue left from the destructive distillation of coal. Its composition is principally carbon and ash, but sulphur and phosphorus may be present in small amounts. Composition of Cokr. Carbon . . Hydrogen Oxygen . . Nitrogen . Sulphur . . Ash Moisture • • German Per cent. 84.76 0.90 0.34 1.38 0-93 9.42 2.27 English Per cent. 9^-75 0.45 1.20 0.60 5-50 French Per cent. 8435 0.67 1-35 0.40 1.06 12.17 American Per cent. 81.34 o 57 2,40 0.89 1.04 13.76 Composition or Pocohontas Coke. Per cent. Moisture 0.70 Volatile matter 1.25 Fixed carbon 91.16 Ash 7.59 Sulphur 0.632 Phosphorus 0.005 Compression strength 236 pounds per square inch. I ton occupies about 85 cubic feet. Comparative Coke Analysis. Bee hive Retort Vertical ovens coke ovens retort 0.35 1.25 1-35 0.34 1. 61 1-73 92.69 86.66 87.40 5.89 10.48 952 0.74 0.77 0.99 1.83 1.90 1.82 52.07 49-49 59-25 47.93 50.51 40.75 Horizontal retort Moisture Volatile compounds Fixed carbon Ash Sulphur Real gravity Per cent, coke Per cent, cells 2.57 3-84 86.05 7-54 0.96 1.73 53.89 46.11 REPORT OF THE COMMITTEE ON STANDARD METHODS FOR DETERMINING THE CONSTITUENTS OF FOUNDRY COKE. (American Foundry Association.) Sampling. Each carload of coke shall be considered as a unit. While the car is being unloaded, full length pieces of coke shall be taken 54 ENGINKERING CHEMISTRY at about equal intervals and a sample approximately the size of an egg taken from each end and also from the middle of each piece, until 25 to 40 pounds are obtained. Should it be necessary to sample from a stock pile, 25 to 30 pounds of sample obtained as above directed, shall be taken for each 50 tons in the pile, care being used to get the piece from different places which will give a fair average sample. Preparing the Sample. Crush the sample between hardened surfaces, preferably of manganese or chrome steel, until all the material passes through a ^-inch mesh sieve. Quarter this; reserve one portion for moisture determination and crush the other portion until it will all pass through a ^-inch mesh sieve, and again quarter down until about 2 pounds remain. Crush this until it will pass a No. 20-mesh sieve, and quarter down to about 20 grams. Grind this until it all passes through a No. lOO-mesh sieve. Moisture. Dry I kilogram of y^-'mch mesh sample to constant weight at 104° to 107° C. The loss in weight shall be calculated to per cent, moisture. Moisture shall be determined on the ground sample by getting the loss in weight when I gram sample is heated in an open platinum crucible of about 20 cc. capacity for I hour at 104° to 107° C. The moisture on the ground sample shall be used to calculate the other results gotten from the ground sample to percentages in the coarse undried sample. Volatile Matter. Cover the crucible containing the dried sample, with another crucible (either platinum or porcelain) of such a size that it will fit closely to the sides of the outer crucible, and its bottom will rest Ys inch to ^ inch above the bottom of the outer crucible. Ignite 3^ minutes with the Bunsen burner and 3^ minutes with the blast lamp. I^et cool, remove the inner crucible and reweigh the outer crucible with contents. The loss of weight is volatile matter. Ash and Fixed Carbon. Ignite the sample upon which the volatile matter was deter- e:ngine;kring chemistry 55 mined until all the carbon is burned, having the crucible open and inclined. The ash should be tested for unburned carbon by moistening it with alcohol, which will show black any carbon remaining. After all carbon is burned, the weight of the cru- cible and ash minus the weight of the crucible, gives the amount of ash in the sample. Fig. 4a. The amount of fixed carbon is obtained by subtracting the weight of the crucible and ash from the weight of the crucible and residue from the volatile matter determinations. Sulphur. Crucible. — A soft steel or nickel crucible of about 40 cc. capacity, the lid being perforated with a small hole for the in- troduction of the igniting wire. Crucible Stand. — Any arrangement suitable for holding the crucible firmly in place and out of contact with the beaker during the peroxide combustion. Determination. — To the dry crucible add first 12 grams of sodium peroxide and 0.5 gram of powderd potassium chlorate, then exactly 0.7 gram of coke (80-mesh) and mix thoroughly by means of a small spatula. Place the covered crucible on its stand in a 20-ounce beaker containing enough water to immerse the lower half of the crucible. Ignite the crucible contents by thrusting in, for a moment, a red hot wire through the lid hole. Wait 2 minutes or longer for the mass to cool somewhat, remove the stand and tip over the crucible on its side in the water. After the fusion dissolves, rinse and remove the crucible. 56 ENGINEERING CHEMISTRY Acidify the solution with hydrochloric acid, then add ammonia in slight excess, filter and wash. To the filtrate add a drop of methyl orange, then hydrochloric acid from a graduated pipette or burette until 0.5 cc. in excess. Bring to boiling, add drop wise about 10 cc. of barium chloride solution, continue boiling at least 15 minutes longer, and allow it to stand in a warm place for not less than 2 hours, filter, wash until the silver nitrate test shows no chlorides, ignite and weigh as BaS04. Grams BaSO^ X i9-6 = per cent, sulphur. Phosphorus. Ignite 5 grams of coke in a platinum dish or large platinum crucible until all the carbon is burned off, then add 10 cc. hydro- chloric acid (i-i) and 20 cc. hydrofluoric acid and evaporate to dryness and ignite at a dull red heat. Fuse the residue with about iy2 grams of sodium carbonate and 2 grams of potassium nitrate. Cool, place the dish in a beaker of water and boil. Clean and remove the dish. iVcidify the solution with hydro- chloric acid, precipitate with ammonia, boil, filter and wash with hot water. Wash the filter w^ith warm dilute nitric acid to dis- solve the precipitate. Should it not dissolve, wash with warm dilute hydrochloric acid until dissolved. In the latter case, it will be necessary to evaporate to about 5 cc, add 30 cc. nitric acid (1.20 specific gravity) ; again evaporate to about 5 cc. and add 30 cc. nitric acid (1.20 specific gravity). After heating the solu- tion to between 70° and 90° C, add 50 cc. of molybdate solu- tion. Agitate the solution a few minutes, then filter, and wash five times with a 3 per cent, nitric acid solution, and five times with a 0.1 per cent, potassium nitrate solution. Transfer the precipitate and filter to the flask in which the precipitate was made. Add 30 cc. water, then NaOH (N/5) from a burette until in excess, keeping the solution agitated. When the yellow precipitate is all dissolved add o.i cc. of phenolphthalein solution as indicator, and then titrate with HoSO^CN^). cc. (N/5) NaOH ^ cc. (N/5) H,SO, X 0.0054 = per cent, phosphorus. To make the molybdate solution, add 100 grams molybdic KNGTNKERING CHEMISTRY 57 acid to 250 cc. water, and to this add 150 cc. ammonia. Stir until all is dissolved and add 65 cc. nitric acid (1.42 specific gravity). Make another solution by adding 400 cc. concentrated nitric acid to 1,100 cc. water, and when the solutions are cool, pour the first slowly into the second with constant stirring and add a couple of drops of ammonium phosphate. Dr. Andrew A. Bl>^ "^ . John Fulton, M. E., gives the following as the standard for the chemical and physical properties of coke : Coke. Method of Manufacture To be Used For Style Charge of in oven Size pounds Yield Time in per of cent, cooking Kind Size of of fur- furnace nace 11^X5^6^' 48 Bee-hive 7600 63 and Iron blast. 70^ X 16^ 12^ X 6^ 72 c ^ R a; III 11:^ tfl CJ CI m CO Chemical analysis IT. 1" §1 g 'S •55 1 u -a 5 s Dry Wet Dry Wet Coke Cells 15-47 23.67 58.98 87.34 49.96 50.04 301 120 I 2.5 1.89 87.46 0.49 11.32 0.69 0.029 o.on LIMESTONE. The composition of the various limestones and the methods of analysis of the same are of importance to the Chemical En- gineer. The following scheme of analysis is arranged as a sim- plified method. 5 B o . 3 22 ctf 3 X^ -o a . a « W -JO O o-" o 12 £2 ^^"" fe * o < 2c; ^8 vt •-■ a . rt'O . , eao t >> s •O j ca^ po-'-Oh o; ca «^ iT,'*- _ _ I £ o ij^ a a -- /^ 3 tfl ■> a j/5 o "5 '^.bJDrt-X o j:^ u a 4^3 £ a v-rt ^•o«;■M o «2 ;< o cd CO Ho 5 f' S ^' Sj'S a - O^ o a V «ox ^ to On CO »' f e C fe '^ ii •- Si"° a . js "^ - O ««aa rt s M o a'5 9 4^ a a'S 5 a « rt CO4J 3 i^Cfl O o _^ 3 tn K cO» CO K ;M U» ^O 4/ caZ u a o 5 "^ "i II o £?i-Si2M SBvS i: tn X a "O _ di a 2-o:::sss « ctf S5 "^ & g8<£-a^i5 I § e g ^1^ 5 a^S-^S - O eft » p ^ .2 a *j O. X a "p. 11 a — On On | V SZ P, (u S rt O " O V O i> - be a Q ca o •a a 3 ii:a§ii O bo (LI ^ ^ bc"^ bti ^{^ * o ca i» 2 o ^-^ - •S3 Ha ■- -"""o'a -^3 11 a S a rt tn a 3 ca: I- bo d d O f a o a 5 .2?.t o 3 u o 6? u ^ P «) ♦J ^ c8"* 3=-£ a oS^ » «3 4^ « a.5.5.t: a'Z 41 4J ■" ° S 3: s-M a 41 3 re— - . O = 41 bOT3 p-^^ cc w =« nli art ■= i -ti P -^ p:f.5?iy ""O , _ u p = ti p o. - ca »- s 41 *j 41 .3 u ^•" 1^ a o a tir: o 3 •^ ^ y rt -« rt , — 0.C8 a i) 5 ^ "^ ^ ? "^ 41 4IM « n i:ngine:e:ring che:mistry 67 68 e:ngine:e:ring chemistry Determination of Carbon Dioxide in Limestone. The U-tube B (Fig. 5) contains water acidified with sulphuric acid. No more of the mixture should be placed in the tube than just sufficient to make a seal at the base of the U-tube. The U-tubes C and D contain granulated calcium chloride. As this chemical often contains CaO it is always advisable before connecting these tubes with the apparatus to first pass carbon dioxide gas through them to saturate any CaO, and then aspirate with air, to exhaust all free carbon dioxide. The U-tubes E and F contain soda-lime granulated, medium size, and are weighed carefully before using the apparatus. The U-tube G contains calcium chloride to absorb any moist- ure that might enter from the water in the aspirator. Two grams of the limestone are transferred to the flask A, and the flask connected with the apparatus shown in Fig. 5. Dilute hydrochloric acid (50 cc.) is allowed to run into the flask A from the funnel tube, and heat is gradually applied until the liquid in the flask begins to boil. Connect the Bennert^ drying apparatus with the funnel tube of flask A, the aspirator with G,^ and slowly aspirate air through the entire apparatus. The carbon dioxide is all absorbed by the soda-lime. During the absorption of the carbon dioxide by the soda-lime, the tube E becomes heated. It must be cooled to the surrounding temperature before weighing. After aspirating about 4 liters of air, weigh the soda lime tubes. Replace the tubes E and F with the apparatus and as- pirate about 2 liters of air and again weigh. This must be re- peated until the weight of the tubes remains constant. Grams. Soda-lime tubes and CO2 48.2270 Soda-lime tubes 47.4307 CO2 0.7963 °-^^'^^X'°° = 39-81 per cent. CO,. 1 The first cylinder contains a strong solution of potassium hydrate, the second cylin- der H2SO4 concentrated, and the large U-tube granulated calcium chloride (anhydrous). 2 Any other means of exhaust can be used that can be regulated. e:ngini:e:ring che:mistry 69 The carbon dioxide may also be determined as follows : Weigh I gram of the powdered limestone and transfer it to the carbon dioxide apparatus (Fig. 6) through the opening in C. Fill the tube A half full with strong sulphuric acid. Fill the tube B with hydrochloric acid, dilute (1:1), then weigh the ap- paratus and contents. Remove the upper stopper in B and care- fully allow the hydrochloric acid to run into the flask and dis- solve the limestone; close the stop-cock as soon as the hydro- chloric acid has entered the flask; carbon dioxide will be evolved and pass out through the-tube A, in which tube the sulphuric acid acts as a drier on the evolved carbon dioxide. Upon solution of the limestone, air is slowly aspirated through B, C, and A, by connecting the copper outlet of A with an aspirator. When the carbon dioxide is all displaced by air in A, B and C, the apparatus and contents are allowed to cool, then weighed. The difference between this weight and the first weight represents the amount of carbon dioxide evolved from the i gram of limestone. Thus, I -gram limestone taken : Grams. Weight of apparatus after evolving CO2 17.267 Weight of apparatus before evolving 16.871 CO. 0.396 0.3960 X 100 r . r^r^ -^2. — d 31^ 39.6 per cent. COj. This apparatus is easily and rapidly operated giving results agreeing within i per cent, and in many commercial analyses the carbon dioxide can be determined with it instead of using the more complex apparatus for accurate work shown in Fig. 5. 70 engine:e:ring che:mistry Limestone. Resume. Pe^ cent. Organic matter 2.02 Silica 4.80 Iron and aluminum oxides 1,40 Lime 42.16 Magnesia 7.31 Sulphur trioxide 2.50 Carbon dioxide 39.81 100.00 The SO3 is united with CaO to form CaSO*. SO3 : CaSO* : : 2.50 : x X = 4.25 Subtracting the 1.75 CaO used to unite with the SO3 there remains 40.41 CaO to unite with CO2. CaO : CaCOa : : 40.41 : x X = 72.17 MgO : MgCOa : : 7.31 : ^' ■v = 15-36 Per cent. Organic matter 2.02 Silica, etc 4.80 Iron and aluminum oxides 1,40 Calcium sulphate 4.25 Calcium carbonate 72.17 Magnesium carbonate 15.36 100.00 The analyses shows the limestone to be a dolomite or magne sian limestone. The following is an analysis of high-grade lime- stone : Per cent. Silica 0.87 Iron and aluminum oxides 0.12 Calcium carbonate 98.60 Magnesium carbonate 0.22 99.81 It is seldom that phosphoric acid is determined in limestone, since it usually amount to less than 0.02 per cent. It is essential however, in cases where the limestone is to be used in blast-fur- naces making Bessemer pig iron. d o B oj= o o GO d < s *o B B o O I CO w c« o sis •^ o CO5H 4J 5 ^ C O o c ^5^^,.:^9W 2-^ S.b-3 «j:.2 o w,; o ""u t-^ ^.<3 Bgo'^?SSS-3.0--g| ^•«3g . p n 1 sfii5>.ii+| S .3 5 ^-0:^; V o ir°3=5«a.s-^ l.:irtQ-5^--J'3-ai£-^^'S .2 «•« ?i ^ 3 vtD ^ 3 o'z: bo's* ^ >. •= . «tf3 3 ^ :^ 01 72 ENGINKERING CHEMISTRY EXAMPI^E. Ten grams of iron ore taken. Grams. Insoluble residue and crucible 10.551 Crucible 10.301 0.250 0.25 X 100 . , , = 2.50 per cent, insoluble matter. Solution = 500 cc. Phosphorus pentoxide ( 100 cc. ) . See appendix. Iron. Fifty cc. reduced with zinc or SnCl2 requires 34.65 cc. stand- ard K.Cr^O- solution. One cc. K^CrgO^ corresponds to 0.0168 gram iron, 34.65 X 0.0168 = 0.58212 gram iron in 50 cc. of the iron solution. „, 0.58212 X 10 X 100 o ^ . , Then ^^ ^ ^ =58.21 per cent. Fe in the ore. 10 ^ r = 83.16 " FegOa in the ore. Sulphur trioxide (50 cc). Grams. Crucible + BaSOi 11. 126 Crucible ii.oii BaS04 0.015 BaS04 : SO. :: 0.015 : x X = 0.0051 0.0051 X 10 X 100 ^^ ^ ^^^ ^ = 0.51 per cent. SO3. Alumina (50 cc. from 250 cc. = Vs of 100 cc. of original solution, ) Grams. Crucible + AL-Oa.Fe.Os 12.6614 Crucible 12.3160 Al,0..Fe,03 0.3454 e:ngine:ering che:mistry 73 Fifty cubic centimeters of the iron solution, in (4), by titration, gave 0.58212 gram of iron or 0.3326 gram of ferric oxide for 50 cc. of the 250 cc. solution of Fe203,Al203 in (3) of the scheme on page 71. Subtract this weight (0.3326) from the weight of alumina and ferric oxide (0.3454) in the 50 cc. The remainder equals 0.0128 gram alumina. 0.0128 X 25 X 100 , _ ^ ^ -^ =^ 3,20 per cent. Al^Og. Another method of determination of alumina in presence of ferric oxide, where the aluminum oxide is in small amount, is to fuse the weighed oxides with potassium hydroxide in a silver capsule, and extract with water. The alumina forms a soluble salt whereas the ferric oxide remains undissolved. Filter off ferric oxide, wash, ignite, weigh and subtract weight from the former weight of both oxides. Difference is the weight of alumina. Manganese oxide (100 cc). Grams. Crucible + Mn:i04 12.166 Crucible 12. 131 Mn304 0.035 0.035 X 5 X. 00 _ , ^5 p,, ,,„t Mn30.. Lime (100 cc). Grams. Crucible + CaO 8.936 Crucible 8.929 CaO 0.007 0.007 X 5 X 100 ^ ^ ^ ^ ^^ ^ ^^ = 0.35 per cent. CaO. Magnesia ( 100 cc. ). Grams. Crucible + Mg.P.Or 8.929 Crucible 8.919 MgoPaOi o.oio Mg2P207 : (MgO)2 : : o.oio : x X = 0.0036 o.oo-iS X 10 X 100 _ . ,, ^ ^ ^ — = 0.18 per cent. MgO. 74 e:ngine:icring chemistry Water of hydration. Grams. Amount of ore taken 1.267 CaCl2 tube + H-.0 29.065 CaClo tube 28.963 H2O 0.102 — C^-" ^ 8,05 per cent. H.^O (hydrated.) Resume. Carbon dioxide absent. Percent. Insoluble mineral matter 2.50 AI2O3 3.20 Fe^Os ' 83.16 Mn304 1.75 P2O5 0.12 SO3 0.51 CaO 0.35 MgO 0.18 H2O (hydrated) 8.05 Total 99.82 If the ore is a magnetite, the iron exists as FeO,Fe203. There are several methods of determining the FeO in presence of FegOa- The one recommended by Whittlesay and Wilbur^ is 1 Chemical Nervs, 19, 270. kngine:e:ring che:mistry 75 frequently used, but the method of Allen is simpler and is to be preferred. It is as follows : One gram of the very finely powdered iron ore is heated in a small sealed combustion tube, half full of fuming hydrochloric acid (25 cc. of the acid being sufficient). The heating is first performed in the water-bath for 2 or 3 hours, then in a hot-air oven at 150° C. for 4 hours more. The ore is thus completely decomposed, and after cooling the tube, it is broken under water in a beaker, and the ferrous oxide immediately determined by titration with standard solution of potassium bichromate. The amount of ferrous oxide subtracted from the total oxides, determined in another sample of the ore, gives the amount of ferric oxide. Iron Ores Insoluble in Acids. Some iron ores resist solution in acids in which case the scheme is modified as follows : Two grams of the finely-pulverized ore are fused with 15 grams of fusion mixture (Na^gCOs -|- K2CO3) in a large platinum crucible for ^ hour. After cooling, the fused mass is treated with boiling water, the contents transferred to a 4-inch porcelain capsule, made acid with hydrochloric acid (carefully), and evap- orated to dryness, 50 cc. hydrochloric acid added, warmed until solution of iron is complete, then 50 cc. of water added, and the solution filtered from the silica, etc. The analysis can now be finished by scheme for iron ore. Determination of Chromium in Chrome Iron Ore.^ Take 0.5 gram of the very finely divided mineral and inti- mately mix it with 12 grams of a mixture containing equal parts of dry sodium carbonate and barium dioxide transfer to a large platinum crucible, and fuse over the Bunsen burner for J^ hour. At the end of this time a quiet fusion is obtained and the decom- position is completed. The crucible is placed in a beaker covered with water, and hydrochloric acid added, a little at a time, till the mass is completely disintegrated. The crucible is then removed, 1 Process of Donath modified by I,. P. Kinnicutt and G. W. Patterson. J. Anal. Chem., 3, 151- 76 ENGINEERING CHEMISTRY the solution made strongly alkaline with caustic potash, and lo cc. of a 5 per cent, solution of hydrogen dioxide added to oxidize the small amount of chromium sesquioxide that may be present. The solution is now boiled for 20 minutes to remove any excess of hydrogen dioxide, made acid with hydrochloric acid, and the amount of chromic acid determined by the aid of a standardized solution of ferrous chloride, i cc. of which corresponds to 0.015 gram Cro03. Reference. Consult : Johnson, p. 140. The following analyses indicate the varying amounts of chro- mium sesquioxide in the chrome iron ores : Place FeO MgO CroOg ALO3 SiO,. Analyst 1. Chester Co., Pa 2. Baltimore Md. 35.14 36.54 18.97 20.13 24.00 25.66 35.68 21.28 8.42 30.04 32.93 34.66 9.96 7-45 5T36 15.03 18.13 6 68 51.66 39-51 44.91 60.04 53.00 54.08 45.90 49-75 64.17 63.37 52.13 63-38 9.72 13.00 13.85 11.85 12.00 9.02 3.20 11.30 10.83 1-95 10.S4 2.09 = 99.61 10.06= 99.11 11.82= 98.51 - = 99.47 Mn. 10.00, 1. 00 = 100. 4.83= 98.95 - = 99.81 - = 100.46 19.91 = lOI.OI Ca. 2.21, 201 = 99.58 4.75 = 100.65 Ni. - = 104.29 Seybert. 3. " ma.ssive 4- " cryst 5. Siberia 6. Roraa.s, Nor 7. Bolton, Ga . 8. I,ake Memphramagog, U. S. . 9. Beresof , Sib 10. Baltimore, Md 11. Voltena, Tuscany, 12. Texas, Pa Abich. Langier. Hunt. Moberg. Rivot. Bechi. Garret. Determination of Titanium in Iron Ores. The method of BetteU is generally used. Fuse about 0.5 gram of the finely powdered ore with 6 grams of pure potassium bisulphate in a platinum crucible at a gentle heat, carefully increased to redness, and continued till the mass is in tranquil fusion. Remove from the source of heat, allow to cool, digest for some hours in 150 cc. of cold distilled water (not more than 300 cc. are to be used, as it generally causes a pre- cipitation of some titanic acid) ; filter off from the silica, dilute to 1,200 cc, add sulphurous acid until all the iron is reduced, then boil 6 hours, replacing the water as it evaporates. 1 Crookes, "Select Methods," p. 194. e:ngine;e:ring che:mistry 77 ■ ^^^^P The titanic acid is precipitated as a white powder, which is ^^^^low filtered off, washed by decantafion, a Httle sulphuric acid ^B being added to the wash-water to prevent it carrying away ti- tanic acid in suspension. Dry, ignite, allow to cool, moisten with solution of ammonium carbonate, reignite, and weigh. The ti- tanic acid is invariably obtained as a white powder with a faint yellow tinge, if the process has been properly carried out. Reference. (Iron Ores.) "The Iron Ores of the United States." Proceedings of the Iron and Steel Institute, special volume, 1890, pp. 68-91. Composition of Various Iron Ores. u o ate §5 o h .T5 O o o ObO FeO. .... Fe^O, . . . . MnO., . . . AI2O3 . . - . CaO MgO .... SiO, .... CO, P2O5 ••.. SO3 H2O. TiO., . . . Cr^Oa- .. { Organic ( Matter Total 90.52 trace 1-39 0.70 0.42 4.76 0.26 0.05 1.90 69-93 3.12 1.53 1.62 13-45 0.25 10.21 26.52 63.18 0.12 3.28 0.38 6.68 0.05 o.oi 20.13 11.85 7.45 60.04 22.39 53-71 0.25 0.50 23.72 45-86 0.40 0.96 5-86 1-37 1.S5 10.88 31.02 0.21 o.io 0.90 2.72 40.77 0.90 0.72 10. 1 2 26. 41 17-38 47-96 9-50 3.12 39-19 99-47 100.57 99.41 99.02 99-77 Z° = Mi; ^-^ «j: •^ i M-" 0.0 4ix^« >.o~ o i^ o s ^i ± r: ^ ^ J ° 0° o'.2 c3 ^ - — noiS«Jao-«' ^ 3 J « u 2: K c 3^.Sft25^« ^ -Cflrrii _ C ? o "0.5: ^•■^ -s ■""C n k s 11 ^ O 5 5 '^ S! aS-^ 2 § if o B o^-5.N ^tJ I ^3 I ^^53^ S t o,. "i.'O >W3^^§ ^^.S * C rt rt 3 s ii "2 o • - g t; " 2-^"k&-SS' e (LI u "5 ^ -i .. .-'OK > a o I "^ ^- '*'. ,-> t; 7l ^ '-. ai 'u 3ia a^-ra ^ ^ o — Si) rt Q 0.« ■;;oa«stci:3*>o > U5 .:: ^ 11 ^ o _ fe « H ^ CO te be O M— ^'■^ii^boS'^ 3 H •r I is J a 02 p ji s s ^3^ g'!!-:^^.^ 3 rtrt : « ce _ o J <"— ''ti 3 ; c8 II « o SOS s ^x •;; be £ t/j 3 J3 "^ o O ill* 08 « N TO O a, ::: t: > z> ofe _ >* 4; S X. o K/ 3 t- 'o'reW 11 o 5j 0^0 11 HO 08 be I/O 4, a " j: P be y p.:- ctf 'i: u o ii SO '- H K '11"" ENGINEERING CHEMISTRY 79 -L "-* * "o o^ ^ oqO «c/3 O^ ^• cpO *: n Jil ffi (DO a3 ^ ^ ca *J O) u 03 aos i'^ 1 g=^ ^^ •Ss 10 v_^ 1^ CO „ .so d Vi -.;^l-ii-i)-i a;(uajo ci « 10 d d 10 Tf d ' ' ' g: *C« CO Cfl — m pel .S CO CS «J T:f -l i-i v-> u> ;^ 52^ t u — • •• ..^. . _ J Fig. II. 82 Kngine:e:ring chemistry A moisture must be taken for each train load from all open pit mines; a train consists of 40 to 45 of the 50-ton steel cars, or 60 to 68 of the 25-ton wooden cars. A moisture sample must also be taken for all the cars loaded as a shaft or stockpile during each lo-hour shift. Samples must be taken from three places on top of each car as shown by the following diagram. Fig. 12. Care should be taken to secure the sample from well under- neath the surface as soon as practicable after loading, maintain- ing the true proportion of lump and fine ore. The sample as taken must be immediately placed in a can with a tightly fitting lid, and brought to the crusher house. It is optional to take a moisture sample from the regular sample, provided it has been taken from well underneath the surface. Cargo Sampling at Consumer's End. Cargo ores present the most serious obstacles to a uniform method of sampling. The boats vary in size from 3,000 to 12,000 tons with I or 2 decks, and in the number of hatches from 6 to 36 widths varying from 12 to 24 feet. The grabs at the different unloading points vary in number, kind, size, and the rapidity of their operation. The ores vary from the very fine to the all lump, from the so-called mixed ores such as the groups to the mixed cargoes, consisting of different ores in the same boat, and with different ways of loading boats. Grab Sampling. — An excellent method of cargo sampling where the entire cargo will be represented, and particularly adapt- able for fine ores. A sampler with a small scoop attached to a handle of suitable length and holding a definite amount of ore, a quarter, a half or a whole pound, takes a scoopful from each kngine:e:ring che:mistry 83 grab as it rises above the deck. The disadvantage of this system is its increased cost due to the extra number of samplers, one being required for each grab during the entire time of unloading. The general plan for the sampling of all cargoes is to first sample the tops of the piles, before the grabs have started to unload; this is called sampling of the cones. After the grab has removed from the hatch all the ore v^ithin reach, the ex- posed faces standing on each side are sampled; this method is known as face sampling. Or when the latter practice is im- practicable owing to the operation of the grabs, then the method of rounds is followed. In sampling a small shovel or garden trowel is used, the total length of which, including the handle, is 12 inches, and it also constitutes a measure. It is the aim to take equal sized samples from each of the points selected. When lump ore or rock is en- countered at the point determined by the measure a portion is broken off equal to the amount regularly taken. In the sampling of the cones, at a point midway between the side and the center of the boat, directly under the edge of the hatch, the first sample is taken and sampling continued i shovel length apart up the surface of the cone, over its apex and down the opposite surface to a corresponding point under the other edge of the hatch. This line is crossed from corres- ponding opposite points under the hatch as shown in the follow- ing sketch. Not more than one-tenth of the total sample is to be taken in the sampling of the cones. CONE SAMPLING. HATCH. ///•V/;'-^\\V \\ Fig. 13. 84 ENGINEKRING CHEMISTRY Face Sampling. — After a grab has removed from a hatch all the ore within reach and has moved to another hatch, the sam- pler shall measure 2 shovel lengths from the side of the boat and start up the exposed face of the ore, taking samples i shovel length apart all the way to the top, using a ladder if necessary. The next vertical line is measured 4 shovel lengths from the first and the samples taken each shovel length apart on this line as be- fore and so on for each succeeding line across the boat. This is re- peated on the opposite ore face, and the entire procedure con- tinued until the ore faces of all the hatches are sampled that the character of the boat and the operation of the grabs will permit. When a bulkhead occurs only the face opposite to it is to be sampled. FACE SAMPLING. DECK. Round Sampling. — When the operation of the grabs makes sampling by the face method impracticable, or with boats having 24-foot hatch centers, and decks furnishing protection to the samplers, then sampling shall be done while the ore is being re- moved by the grabs. When 5 or 6 feet of the face of the ore have been exposed, the sampler shall enter the hatch and meas- uring 2 shovel lengths from the side of the boat or edge of the face, take successive samples up the face i shovel length apart. The next vertical line is measured ■ 4 shovel lengths from the first, and samples are taken all the way to the top as before, and so on across the entire face of the ore. This procedure is re- peated on the opposite face, one-third of the total weight of the KNGINEKRING CHEMISTRY 85 sample to be taken in the first round. When all the ore within reach of the grab is removed, the second round is taken, using the measurements as above, and the remaining two-thirds of the sam- ple are secured. Part of the regular sample is to be taken for the moisture sam- ple, and for the fineness sample when such is desired. Car Sampling at Consumer's End. When the ore is received in cars the greatest possible number are represented in the samples, and not less than lo equal sized samples are taken from each car. When cars are loaded with fine ore with the piles in opposite ends, at least 5 samples are taken from each pile, the first one at the apex of the pile, and the other 4 at points symmetrically arranged around the sides of the pile, two-thirds of the distance from the apex to the base of the pile or sides of the car. With cars loaded in the center, the system is the same, except that the center of the side of the pile length- wise of the car, is the first point, the other four being symmetri- cally arranged around this point. When the 10 points are located in a car, each of them is sup- posed to represent a definite area, equal to one-tenth of the ore surface of the car. If the car contains all fine ore, then 10 equal sized samples are taken, one from each of the points. If the car contains a mixture of fine and lump ore, with varying amounts of each in the areas included in the different divisions, then each area is judged separately and sampled accordingly. The fine and lump ore are taken each in its proper proportion, the former with the trowel, the latter being chipped, or selected small pieces being taken, each about the size of the first joint of the thumb. The combined sample of fine, chipped and selected pieces from each area, equals the amount taken were it all fine ore. If the con- tents of the car are all lump ore, the proper sized pieces are chipped from four or five of the lumps in each of the 10 areas making 40 or 50 pieces from each car, the total amount of chipped pieces from each of the areas equalling the amount that would be taken were it all fine ore. All samples of fine ore are taken from well underneath the surface to obtain the ore in its natural state. 86 e:ngine:e:ring che:mistry A proportionate amount of the main sample is retained in a tightly closed can for the moisture determination. Preparation of Samples in General. In the preparation of the sample for analysis, the ideal prac- tice is to crush and quarter alternately until the desired quantity with the requisite degree of fineness is attained. A more ex- peditious and equally efficient method, is to crush the entire sample to the desired degree of comminution, then reduce the quantity by successive quartering as before, until the desired amount remains. It should ever be our purpose to approach as closely as possible to either of these two methods in the prepara- tion of all samples for analysis. Preparation of Sample at Producer's End. Samples before being quartered are brought into the crusher house where they are dried, if necessary, at ioo° C. until the ore can be well mixed. Care must be taken to prevent over heating when other than low pressure steam is used for the purpose, es- pecially with ores containing a large quantity of limonite. When sufficiently dry the larger lumps are crushed, if necessary, so that the entire sample will pass through a 3^-inch mesh sieve; a finer sieve may be used if desired. The sample is thoroughly mixed on an iron top table, then spread out evenly about ^ of an inch in depth and alternately quartered and mixed until % of the original sample remains. It is now crushed until fine enough to pass a ^-inch mesh sieve, then mixed and quartered as before until about 2 pounds remain. It is again crushed until fine enough to pass a 20-mesh sieve, and spread out evenly about ^ of an inch in depth. About 3 ounces of this are taken from all over the pile, with a small spatula, dried in a small pan at 100° C. and crushed on a chrome steel plate, until it will pass through a lOO-mesh sieve. After being thoroughly mixed this is trans- ferred to a bottle, and constitutes the sample for analysis. The following method for the preparation of the sample is optional. All the ore is passed through a ^-inch sieve, thor- oughly mixed on a suitable cloth and quartered in the usual Engine:e:ring chemistry %y way, diagonally opposite portions being rejected until about 5 pounds remain. The sample is again crushed, if need be, to pass a ^-inch mesh sieve, and mixed and quartered as before until 1^/2 to 2 pounds remain. After being dried at 100° C. for 20 to 30 minutes, it is crushed to a fineness of 20-mesh, mixed and quartered until about 3 ounces remain, and the entire sam- ple is crushed on a chrome steel plate and passed through a 100- mesh sieve. The sample is spread out in a shallow pan, dried at 100° C. for 30 minutes, again mixed and transferred to a 3-ounce bottle or can for the analysis. Preparation of Sample at Consumer's End. The aggregate sample is dried at 100° C. and crushed before any quartering whatsoever, so that the entire mass will pass through a ^-inch mesh sieve. This is reduced by successive quartering and crushing until its weight is from 4 to 8 ounces, and it will then all pass through an 8-inch mesh sieve. Or the entire sample is crushed so it will pass through an 8-inch mesh sieve, and then quartered as before until from 4 to 8 ounces remain. And this final quantity is then further crushed with a chrome steel bucking board and muller and all passed through a loo-mesh sieve. A sufficient amount of this powder for all the needs of the analyst is placed in a small air-tight con- tainer, dried for i hour at 100° C. and when cool it constitutes the sample for analysis. A separate, larger portion of the same sample is retained for further needs. Resume:. Per cent. Lime (CaO) 36.46 Magnesia (MgO) 2.12 Silica (SiOO 43-25 Alumina (AI2O3) 15.94 Ferrous oxide (FeO) 0.31 Sulphur (S) 1.53 Manganese oxide (Mn02) o.io Phosphoric acid (P2O3) 0.09 Undetermined 0.20 Total 100.00 ENGINEERING CHEMISTRY a w 2 Z 8 c 1: ^ t- i) V .t: "o a * a i; .2 ^ W J2 Sg c 5 ^ .5 c O-Og 'O rs -;: >- i: o o cd j: 2 ^ ;:: p u c >. 8 .i^ _ - * t P 8 -5 :g -g ii i; -M It O 2 01 CI "^ f— i ii S rt 3 rt o O 73 .- C .5 S c £ ^ o 2 X y a "S g o o > j^ 2 .•-_!. '^•O -vr, ^^ 5'^ It c Crucible + Mg2P207 = 11.90253 grams ■ earl filte lefi ps tere -a °'0 to : HoO, make solution n idd NaCoHgOo, boil and ron and aluminum. Tl 250 cc. flask, a few dro e twelve hours, then fil "Z •;; tr. .3 MgoPoOr = 0.02353 d ^ 4; .- « MgaPgO; :'(MgO)o : : 0.02353 : ^ • |5-2^d ^ = 0.00843 S 2-S.^Z^D. 00843 X 5 X 100 . „ ^ Sg-C^coaj 2 -2.13 per cent. MgO. fa cu JicoZ be .^ * Crucible + CaO = 12.02484 grams. c^ •♦ - =11.87900 " v.-^-B « i5 £ = >."s ;t* .0.14584x5X100 ^ , •1 uS-cO , = 36.46 per cent. CaO. cs MgO. .. add 100 h NaoCOg oxides oi iferred to led, set as (S cocoSr: (J ii «rtO Crucible + Mn304 = 11.87936 grams. CaO, 50 cc le wit rated trans le add iShed, ^ 'Scoo Mn304 = 000036 d -j^-='^---pi s «j;;^(^(i( 0.00036 Mn304 = 0.00041 Mn02. •- i-_,T)rt 0.00041 X 5 X 100 gsp'gc^ 3^ — -^-^ = o.io- per cent. MnOg. « S coSo n g rt ZX^ CO -a 'g Crucible + AlsOg+FeaOg == 11.94415 grams. S « 1 Subtract FeoOg ) „ ^„„ .. 1 found by titration / = °°"i39 9. ^< . Crucible + AI2O3 = 11.94276 " -53 Crucible = 11.8790 " A^l§ AI0O3 = 0.006376 " < 6 r2 ii .i: 0.006376 X 5 :^ 100 . . , ^ 3 "1.5 ^ = 15.94 per cent. AloOg. rt S «•= i"^ « ,■" 50 cc. require 0.058 cc, KgCroOT solution. ""^ nO^ - "^ I cc. K-jCroO: solution = 0.0168 eram Fe. «u d S-= co-tio'c 50 cc. solution of slag = 0.000974 " " C 0.2 .ti a;-- £• 250CC. " =0.004870 " " 6 °o3&.5^CJ4i Fe . FeO :: 0.00487 : ;r ^ , CO ^ sam- from 1 and sphor- rected Crucible + MgaPgOr = 11.00935 grams. " — = 11.00879 '* one 50 cc. lutioii r pho; as di ore MgoPoOy = 0.00056 MgoPoO; . P2O5 • . 0.00056 : :r ^u^ g-^-oS X = 0.0003594 CO ° t? y C H (u w 2 ca- 0.0003594 X 5 X too ^ ^ ^ ^^^^ — ^— = 0.09 per cent. P0O5 cu aSS.y.? . ^S ■«» -^^ Crucible + SiOo = 17.585 gr^ms. .^-S " — = 16.720 -§ ^^ .. 0.8 SiOo = 0.865 6* ci5 65 X 100 ^ r.-^ -^ = 43-25 per cent. SiOo. « rt':2 ENGINEERING CHEMISTRY 89 Form of Bi.ank Used for Reporting Blast-Furnace Si^ag Anai^yses. S1.AG. j)ate Orf^ impri "^Jr> nf irnn .... .... Lime ( CaO ) • ... . . Silica (SiOg) Oxide of iron ( FeO ) Calcium sulphide (CaS) ' Examples of Blast-Furnace Slag Analyses. FeO SiOa AI2O3 CaO MgO MnO., Sulphur f sulphide of 1 Calcium \ calcium j Phosphoric acid (P2O5) • Undetermined loss No. i.i No. 2.2 Per cent. Per cent. 0.270 0.436 45.460 35.000 16.590 14.362 32.805 45-370 1.080 1-398 0.083 trace I.571 1.875 1-963 1.500 0.008 0.059 1. 1 70 — 100.000 100.000 Some varieties of slag are soluble in hydrochloric acid, in which case the solution can be made in HCl. This applies also to open-hearth slags, refinery slag, tap-cinder, mill-cinder and converter slag. 1 Slag made during the run of Alice Furnace, on mixture containing Enterprise ore. - Slag made at the Sloss Furnace in June, 1886, on No. i foundry iron. (Consult Trans. A.I. M. E., 16, p. 148.) 90 ENGINEERING CHEMISTRY Basic slags, from the Thomas-Bessemer process, often con- tain over 20 per cent, of phosphoric acid and require a some- what different process of analysis. Thus: One gram of the finely pulverized slag is fused with excess of sodium carbonate in a platinum crucible. Extract with water, acidify the solution with nitric acid and evaporate in a porcelain dish to dryness. Take up with hydrochloric acid, dilute to Y^ a liter, and precipitate the phosphoric acid by the acetate process. The precipitate is filtered, dissolved in hydrochloric acid, ex- cess of nitric acid added, and the solution concentrated until the hydrochloric and acetic acids are expelled. The nitric acid so- lution is diluted to 500 cc. and two portions are taken (each 250 cc.) and the phosphoric acid determined in these by the molybdate method. THE FORD-WULIAMS METHOD FOR HIGH GRADE MANGANESE ORES. Ores in Which all of the Manganese is Soluble in HCl. — Weigh 0.5 gram of finely powdered ore, transfer to No. 4 beaker, and add 20 cc. HCl. Digest until residue is white and flotant, evaporate to a syrup, add no cc. strong HNOj and boil until red fumes are nearly gone. Now add ^ teaspoonful of fine as- bestos fiber using care so as to prevent boiling over. All red fumes will disappear in about i minute after adding asbestos. Continue boiling, add 8 grams KCIO3, boil 10 minutes longer, remove from the source of heat, allow to settle and filter, using an asbestos filter. Wash as in the original method. Weigh 0.750 gram of C. P. oxalic acid, dissolve in hot water in a small beaker and transfer this solution to the beaker in which the MnO^ was precipitated. Add 50 cc. of warm dilute H2S04> dilute with warm water to 200 cc. ; and drop in the asbestos filter containing the MnO,2 precipitate. Stir and heat gently (not above 80° C. until all Mn02 is dissolved. Titrate the solution with standard KMnO^. Also titrate 0.750 gram oxalic acid with the same KMn04. Subtract from this value the i:ngine:e:ring chemistry 91 number of cubic centimeters used in the first titration. The differ- ence is equal to the KMn04 equivalent to the manganese in the ore. Knowing the iron value of the KMnO^, we can calculate its man- ganese value, for Fe value X ^^/ii2 = Mn value. Multiply differ- ence between first and second titrations by Mn value of KMn04 and this result by 200, which gives percentage Mn in ore. Where all of the Manganese of the Ore is not Soluble in HCl, filter from this residue and evaporate the main solution. In the meantime treat the residue with HFl and a few drops of H2SO4. Evaporate off the HFl, add a little HCl, heat until dissolved and add to the main solution. Proceed as in the case of ore in which all manganese is soluble in HCl. This method gives results checikng well with the phosphate method. — /. Jas. Skinner, Chemist, Longdale Iron Co., Virginia. PROVISIONAL METHODS FOR COPPER, LEAD AND ZINC OF THE COMMITTEE ON UNIFORMITY IN TECHNICAL ANALY- SIS OF THE WESTERN ASSOCIATION OF THE TECHNICAL CHEMISTS AND METALLURGISTS. The Determination of Copper. Solutions. — I. A solution of sodium thiosulphate ; 20 grams of the salt in i liter of water. 2. A solution of starch; made by shaking up i gram of finely powdered starch in a few cc. of water and pouring it into 200 cc. of boiling water. This should be made fresh every few days. 3. A solution of potassium iodide containing 300 grams to the liter. Standardization. — Dissolve 0.2 to 0.5 gram of copper foil in 5 cc. of nitric acid and evaporate to 2 or 3 cc. Add 5 cc. of hot water and 6 cc. of ammonia. Boil a few minutes and cool. Dilute to 75 or 80 cc, add 8 cc. of acetic acid and 10 cc. of the potassium iodide solution. Shake until all the copper is pre- cipitated and titrate the free iodine as follows : The thiosulphate 92 ENGINEERING CHEMISTRY is run in until the brown of the iodine changes to yellow ; then add 4 or 5 cc. of the starch solution and carefully run in more thiosulphate until the blue caused by the starch disappears, when the titration is finished. Copper in Ores. — Dissolve 0.5 gram of ore in 2 cc. of nitric acid, 3 cc. of hydrochloric and 4 cc. of sulphuric acid. Evap- orate to copious white fumes. Cool and take up with 25 cc. of cold water. Add a piece of sheet aluminum and boil until all the copper is precipitated. Add 10 cc. of hydrogen sulphide water to insure complete precipitation of the copper.^ Filter, washing three times by decantation. Pour 55 cc. strong nitric acid over the aluminum and copper. Remove the aluminum and wash with a minimum hot water. Place the beaker under the filter and pour strong bromine water over the same to dissolve any copper or sulphide of copper that may run over. Wash with hot water and evaporate the solution to 2 or 3 cc. Add 5 cc. of hot water, 6 cc. of strong ammonia and boil. Add 8 cc. of acetic acid and 10 cc. of the potassium iodide solution. Shake until all the copper is precipitated and titrate. The Determination of Lead. Solutions. — A solution of ammonium molybdate containing 8.64 grams of the molybdate to the liter. A solution of ammonium acetate containing 200 grams to the liter. A solution of tannic acid containing i part of the acid in 300 parts of water. This should be made fresh every few days. Standardization. — Dissolve 2 pieces of pure lead foil weighing about 0.3 and 0.5 gram respectively in 10 to 15 cc. of (i : i) nitric acid. When the lead is dissolved add 20 cc. of (i : i) sulphuric acid. Stir thoroughly and allow to settle. Decant on the filter-paper and wash by decantation three or four times with water containing 2 per cent, sulphuric acid, always decanting as closely as possible. Wash once with a little cold water, keeping ^ The copper is not generally precipitated completely by aluminum, hence the need of hydrogen sulphide water. Arsenic and antimony are precipitated by the hydrogen sulphide water, and are oxidized by bromine, otherwise the results would be high. Kngine:e:ring chemistry 93 I I^B^as much of the precipitate in the beaker as possible. Dissolve the I^V lead sulphate that is on the filter-paper by pouring over it 50 cc. of the hot ammonium acetate solution. Pass this solution through the filter a second or third time if necessary, then wash the paper with hot w^ater. Pour the hot ammonium acetate over the main bulk of the precipitate. Heat until this is dissolved, dilute to 200 cc, make barely acid with acetic acid, and titrate. Titration. — The ammonium molybdate is run in with constant stirring, testing from time to time by placing a drop of the solu- tion upon a drop of the tannic acid solution on a spot plate. When this gives a yellow color the titration is finished. Lead in Ores. — Dissolve 0.5 to i.o gram of ore in 10 cc. of strong nitric, then add 10 cc. of strong sulphuric acid. Evap- orate to white fumes, cool, dilute to 50 cc, and boil. Decant on the filter-paper and wash by decantation 3 or 4 times with hot water containing 2 per cent, sulphuric acid, then once with a little cold water. Now, with the aid of the wash-bottle, transfer the sulphates from the filter back to the beaker, add 50 cc. of ammonium acetate solution and boil thoroughly.^ Filter again through the original filter, wash with hot water, make the filtrate slightly acid with acetic acid, and titrate. The Determination of Zinc. Solutions Required. — A solution of ammonium chloride made by dissolving 60 grams of the salt in 250 cc of water and adding 150 cc. of strong ammonia. A solution of potassium ferrocyanide, 21.6 grams of the pure salt to the liter. A solution of uranium acetate, 4.5 grams to 100 cc. of water and 2 cc of acetic acid. A saturated solution of potassium chlorate in nitric acid. Standardization. — Dissolve 0.2 to 0.3 gram of freshly ignited zinc oxide in 10 cc. of hydrochloric acid, make just neutral with ammonia. Add 6 cc. of concentrated hydrochloric acid, dilute 1 This procedure is necessary on account of the fact that lead sulphate is not readily soluble in ammonium acetate when other sulphates such as barium and calcium are present. 94 KNGINKE^RING CHE:mISTRY to i8o cc. with water, boil and filter. Check, using C. P. zinc instead of zinc oxide. Titration. — Dissolve the solution into two equal parts and quickly titrate one part, then add the remainder and quickly run in enough to almost finish the titration, then add 2 drops at a time, testing after each addition by placing a drop of the solution upon a drop of the uranium acetate on a spot plate. When this gives a yellow color the titration is finished.^ Zinc in Ores. — Dissolve 0.5 to i.o gram, according to the rich- ness of the ore, in 4 to 8 cc. of hydrochloric and an equal amount of nitric acid. Evaporate to one-third of the original volume. If gelatinous silica is present dilute a little and filter.^ Add 15 cc. of the solution of potassium chlorate in nitric acid. Evaporate to dryness, but do not bake,^ cool, add 40 cc. of the ammonium chloride solution and boil, being sure that all clots are broken up. Filter and wash with hot solution of ammonium chloride. If much iron is present the precipitate should be dis- solved in a minimum of hot dilute (1:3) hydrochloric acid, and the treatment with ammonium chloride solution repeated.* Neutralize the filtrate and add 15 cc. of strong hydrochloric acid, 15 grams of granulated lead, boil until all copper is pre- cipitated, and titrate. The amount of solution, of free hydrochloric acid and of am- monium chloride should always be the same and the solution should always be titrated at the same temperature, which should be near the boiling point. ^ By taking care to complete the titration by adding only 2 drops of ferrocyanide at a time, a delicate end reaction may be obtained. If several successive spots develop the yellow color it is then possible to deduct from the burette reading the value of the number of spots added after the end color was first obtained. 2 Gelatinous silica combines with zinc in alkaline solution. 3 If the dry mass is heated much above the boiling point, the zinc is rendered difficultly soluble in ammonium chloride. Furthermore zinc chloride is volatile, in the presence of hydrochloric acid at a comparatively low temperature. ■* It is very difficult, if not impossible, to wash all the zinc out of the ferric hydroxide. ENGINEE^RING CHE:mISTRY 95 GRAPHIC METHOD FOR CALCULATmG BLAST-FURNACE CHARGES. The rule consists o£ 2 equal scales at right angles (Fig. 15), one of which (a) is fixed to a small board, while the other (b) is fixed at right angles to a upon a block, c, which is capable of sliding motion in a groove parallel to a.^ The point A, given by the intersection of the zeros of the scales, is marked upon the board, and from it a line AB parallel to the groove is drawn. With A as a center, lines AC, AD, AE, are also drawn, making with AB, angles whose tangents are equal to the ratios between the v/eight of the silica to weight of base in the respective silicates which it is desirable to produce in order to form the typical fusible slags ordinarily met within blast-furnace practice. The lines AC, AD, AE, are marked with the names of the bases for which they have been calculated. Thus AC makes an angle of 28 degrees 10 minutes with AB — this angle having a tangent whose value is 0.5357, which is the ratio of the atomic weight of silica to twice the atomic weight of lime, and corresponds to calcium silicate: this line, therefore, is marked "Lime." Similarly the line AD makes an angle of 36 degrees 52 minutes with AB, the value of whose tangent is 0.75, or the ratio of the atomic weight of silica to the atomic weight of 2MgO; hence it is marked "Magnesia." Also the line AE at an angle of 41 degrees 25 minutes, and this having a tangent corresponding to the ratio of the atomic weight of 3Si02, to that of 2AI2O3, makes the line correspond to the value of the component parts of silica and of alumina in aluminum silicate, and so it is marked "Alumina." With such a scale it is a very simple matter to at once read off either the excess of silica in any ore, or else the amount required to properly flux off the earthy bases present. ^ H. C. Jenkins: Iron and Steel Institute, 1891. 96 ENGINEERING CHEMISTRY , l- 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 ® bo I 5 I w I S ^; o .0 *- J3 bo T3 .t; >. c ^ !«! rt u O O t/i C -1-1 S rt 2 b "I .& 'i 2 ^ <■> i: 'i 2 c o I- S ^ o U « -S CU > O rt . -H •- <" ^ C zi E t* *-! frt 3 S S 4; I .2 ^. S Is >. o « "3 I E t^ « ?> ^ 3 E 1- V E *" S S* o '- ^ w T3 i5 g (u 2.S bo ua .1 y C • "d 4J C - -d -^ .s 2 ° K V c4 0< bo .2-5 S ^ .S 6 1 ^-1 134 Engin^e;ring chemistry barium carbonate is thrust, by means of a long glass rod, into a flask, removing any adhering particles from the sides of C, by a stream of water from a wash bottle. An excess of the standard acid is now added from a burette or pipette, using a portion to wash out M, and after the contents of the flask have been thor- oughly agitated by shaking, the excess of acid is titrated against the standard alkali, using 3 drops of the methyl orange indicator. Notes. The operation of filtering can be carried out very rapidly after a little practice. Glass wool should on no account be used as a substitute for the quartz, on account of the probability of errors arising from its attack by the alkali or acid. It is well to wash out the rubber tubes connected to the Meyer tube with a little water each day before beginning work. Determination of Carbon by the Colorimetric Method^ This method depends upon the color given to nitric acid (specific gravity 1.2) when steel is dissolved therein; the carbon present producing a light-brown or dark-brown coloration to the liquid in proportion as the carbon is in small or large amounts. The apparatus (Fig. 18) is well arranged for this test. It con- sists of a series of graduated tubes, of glass, each 27.5 centimeters long, 15 millimeters in diameter, and graduated to hold 30 cc. divided by 0.20 cc. The back plate of the apparatus is of white porcelain, 25.5 centimeters wide, 27 centimeters high, and 3 milli- meters thick, and I found it much better than the various cameras to obtain correct comparisons of colors of solutions in the differ- ent tubes. Three standard steels are required, one containing I per cent, combined carbon, for tool steels, etc., one containing 0.4 per cent, carbon, for tires, rails, etc., and 0.2 per cent., carbon, for soft steels, these percentages of carbon having been very accurately determined by combustion. The process is as follows : Two-tenths gram of the standard steel is transferred to one of the graduated tubes, and 0.2 gram ^A. I,add Colby. ENGINDE^RING CHEMISTRY 135 of the steel in which the amount of carbon is to be determined is transferred to another graduated tube and nitric acid (specific gravity 1.20) added and the tubes placed in cold water to pre- vent energetic action of the acid. The amount of nitric acid to be used is as follows : steels with less than 0.3 per cent, carbon. Fig. 18. 3 cc. acid; 0.3 to 0.5 per cent, carbon, 4 cc. acid; 0.5 to 0.8 per cent, carbon, 5 cc. acid ; and so on.^ After a few minutes interval the tubes are placed in warm water, and the latter gradually raised to the boiling point and maintained at that temperature about 20 minutes or until the steel dissolves. 1 "Quantitative Analysis for Mining Engineers," by Prof. E. H. Miller, 1904. 136 ENGINEERING CHEMISTRY Suppose the standard steel contained 0.6 per cent, carbon, the amount of nitric acid required would be 5 cc. After solution and diluting with water the standard to 8 cc. it matches the solution containing the other steel diluted to 13 cc. The amount of carbon in the unknown steel would be 0.977. Eight cubic centimeters contains 0.6 per cent.; 13 cc. standard contains 0.97 per cent. The use of Eggertz' color test for combined carbon requires that steels should have been subjected to a similar physical treat- ment to which the standard steels have been subjected in order to secure accurate results. A steel shows less carbon by color, when hardened than when unhardened, and less unannealed than when annealed. Several modifications of the process have been submitted by various chemists, but they offer no special advan- tages. Stead renders the nitric acid solution of the steel alkaline with sodium hydroxide, which dissolves the carbon, producing a solution about two and a half times stronger in color than the solution in nitric acid. The precipitated iron oxide is filtered off, and a measured quantity of the colored filtrate is transferred to a Stead chromometer and the color compared with a standard steel under similar conditions ; except where the carbon is present in minute quantity only is the process of any advantage over the Eggertz method. Albert I^add Colby states: "Hardened steel should be thor- oughly annealed, preferably in lime." The standard steel used in this case should have been annealed before drillings or turnings were taken. The amount of nitric acid used in dissolving the steel should vary with the carbon present. The following table showing the amount of acid used, the per cent, of carbon present in the standard steel used for steels of varying carbon, and the method of calculating the per cent, of carbon in the sample, is the result of a long experience with the Eggertz method : ^NGINEJERING CHEMISTRY 137 Range No. Weight taken Grams Per cent. Standard To calculate per cent. in cc. carbon in diluted in carbon in steel tested, the per cent. acid standard comparison reading of graduated carbon added steel tube to tube should be 1.20 — 0.80 7 0,02 1.04 , 20.8 CC. Divided by 20 0.79—0.60 6 0.02 0.68 13.6 CC. Divided by 20 0.59—0.50 5 0.02 0.58 II. 6 CC. Divided by 20 0.49 — 0.40 5 0.02 0.49 9.8 CC. Divided by 20 0.39-0.23 4 0.02 0.34 6.8 CC. Divided by 20 0.22—0.14 4 0.02 0.201 6.7 CC. Multiplied by 0.03 0.13— O.IO 3 0.02 0. 114 5.7 CC. Multiplied by 0.02 0.09 — 0.06 3 0.02 0.082 4.1 CC. Multiplied by 0.02 Special apparatus has been devised which greatly facilitates the application of the Eggertz method to the rapid determination of carbon in consecutive "blows" of Bessemer steel. The sample of each *'blow" should be sent to the laboratory in the form of a flat bar about }i inch by ^ inch by 8 inches long. These bars may be most conveniently obtained by passing the small test ingot through a set of small rolls. At some mills the test ingot is hammered into a bar. In either case the bar should be allowed to cool slowly on refractory bricks. It should not come in con- tact with a cold metal surface during cooling. The solution of the drillings may be materially hastened by heating in a calcium chloride bath, kept at 110° C. by high pres- sure steam, instead of in boiling water; and when dissolved the solution may be rapidly cooled by transferring the test tubes to an unglazed earthenware vessel filled with water, which is kept cold by evaporation from the exterior surface of the vessel. Determination of Manganese by the Bismuthate Method. S01.UT10NS Re:quire:d. Nitric Acid. — Mix 500 cc. of nitric acid (specific gravity 1.42) and 1,500 CC. of distilled water. Nitric Acid for Washing. — Mix 30 cc. of nitric acid (specific gravity 1.42) and 970 cc. of distilled water. Stock Sodium Arsenite. — To 15 grams of arsenious acid (AS2O3) in a 300-cc. Erlenmeyer flask, add 45 grams of sodium carbonate and 150 cc. of distilled water. Heat the flask and con- 138 ENGINEERING CHEMISTRY tents on a water bath until the arsenious acid is dissolved, cool the solution and make up to 1,000 cc. with distilled water. Standard Sodium Arsenite. — Dilute 300 cc. of stock sodium arsenite solution to 1,000 cc. with distilled water and titrate against potassium permanganate solution (about N/io), which has been standardized by using Bureau of Standards sodium oxalate.^ Adjust the solution so that i cc. is equivalent to o.io per cent, of manganese, when a i-gram sample is taken. The factor NaaCjO^ »-^ Mn = 0.16397 (using the 1913 atomic weights). Method. In a 300-cc. Erlenmeyer flask dissolve i gram of steel in 50 cc. of the nitric acid, and boil to expel the oxides of nitrogen. Cool, and add about j^ gram of sodium bismuthate and heat for a few minutes, or until the pink color has disappeared, with or without precipitation of manganese dioxide. Add small portions of ferrous sulphate (or any suitable reducing agent) in sufficient quantity to clear the solution, and boil to expel the oxides of nitrogen. Cool to 15° C, add an excess of sodium bismuthate and agitate for a few minutes. Add 50 cc. of 3 per cent, nitric acid and filter through an alundum filter or asbestos pad, wash- ing with 3 per cent, nitric acid. Titrate immediately with stand- ard sodium arsenite solution to the disappearance of the pink color, each cubic centimeter required representing o.io per cent, manganese. Notes. In the method, the preliminary treatment with sodium bismuthate has been found by a number of investigators to be apparently unnecessary; however, the available data to confirm this position are not considered sufficient to warrant its omission. In making the asbestos filter pad it is advisable to have a thin bed, and as much surface as possible. This insures rapid filtration, and the filter, may be used until it becomes clogged with bismuthate. The filtrate must be perfectly clear, since the least particle of bis- muthate carried through the filter will vitiate the results. ^ Circular No. 40, Bureau of Standards, Oct. i, 19 12. ENGINEERING CHEMISTRY I39 Determmation of Manganese by the Persulphate Method. (Routine.) S01.UT10NS Required. Nitric Acid. — Mix 1,000 cc. of nitric acid, specific gravity 1.42,, and 1,200 cc. of distilled water. Silver Nitrate. — Dissolve 1.33 grams of silver nitrate in 1,000 cc. of distilled v^ater. Stock Sodium Arsenite. — To 15 grams of arsenious acid (AS2O3) in a 300 cc. Erlenmeyer flask, add 45 grams of sodium carbonate and 150 cc. of distilled water. Heat the flask and con- tents on a water bath until the arsenious acid is dissolved, cool the solution and make up to 1,000 cc. with distilled water. Standard Sodium Arsenite. — Dilute a sufficient quantity of stock sodium arsenite solution with distilled water, and stand- ardize against a steel of known manganese content, as deter- mined by the bismuthate method. This solution should be of such strength that each cubic centimeter will be equivalent to o.io per cent, of manganese on the basis of the weight of sample taken. Method. In a small Erlenmeyer flask or large test tube, dissolve o.i to 0.3 gram of steel, depending on the manganese content of the sample, in 15 cc. of the nitric acid. Boil gently until the solution is complete and the liquid is clear. Add 15 cc. silver nitrate solu- tion and I gram of ammonium persulphate, and continue heating the solution for y^ minute after the oxidation begins and bubbles rise freely. Cool in running water and complete the determina- tion by either of the following procedures : (a) C olo rime trie . — Compare the color of the solution with that of a standard steel treated under like conditions. {h) Titration. — Titrate with standard sodium arsenite solution to the disappearance of the pink color, each cubic centimeter required representing o.io per cent, of manganese. Notes. In order to obtain reliable results by the colorimetric procedure, the standard should be of the same kind and of approximately the same I40 e;ngine;e)ring chemistry chemical composition as the sample steel. The manganese content of the standard steel is determined by the bismuthate method. The ammonium persulphate should be kept in moistened condition by small additions of distilled water at required intervals. Determination of Phosphorus by the Molybdate Magnesia Method. Nitric Acid. — Mix i,ooo cc. of nitric acid, specific gravity 1.42, and 1,200 cc. of distilled water. Nitric Acid for Washing. — Mix 20 cc. nitric acid, specific gravity 1.42, and 1,000 cc. of distilled water. Potassium Permanganate. — Dissolve 25 grams of potassium permanganate in 1,000 cc. of distilled water. Ammonium Bisulfite. — Dissolve 30 grams of ammonium bisul- fite in 1,000 cc. of distilled water. Ammonium Hydroxide, Approximately 10 Per Cent. — Mix 1,000 cc. of ammonium hydroxide, specific gravity 0.90, and 2,000 cc. of distilled water. Ammonium Molybdate. Solution No. I. — Place in a beaker 100 grams of 85 per cent, molybdic acid, mix it thoroughly with 240 cc. of distilled water, add 140 cc. of ammonium hydroxide, specific gravity 0.90, filter, and add 60 cc. of nitric acid, specific gravity 1.42. Solution No. 2. — Mix 400 cc. of nitric acid, specific gravity 1.42, and 960 cc. of distilled water. When the solutions are cold, add solution No. i to solution No. 2, stirring constantly; then add o.i gram of ammonium phos- phate dissolved in 10 cc. of distilled water, and let stand at least 24 hours before using. Magnesia Mixture. — Dissolve 50 grams of magnesium chloride and 125 grams of ammonium chloride in 750 cc. of distilled water, and then add 150 cc. of ammonium hydroxide, specific gravity 0.90. Method. In a 300 cc. Erlenmeyer flask dissolve 5 grams of steel in 75 cc. of the nitric acid. Heat to boiling; while boiling add about 12 cc. of the potassium permanganate solution, and continue e:ngine:e;ring chemistry 141 boiling until manganese dioxide precipitates. Dissolve this pre- cipitate by additions of the ammonium bisulphite solution, boil until clear and free from brown fumes, cool to 35° C, add 100 cc. of the ammonium molybdate solution at room tem- perature, let stand i minute, shake or agitate for 3 minutes,^ filter on a 9-centimeter paper and wash the precipitate at least three times with the 2 per cent, nitric acid solution to free it from iron. Fig. 19. Treat the precipitate on the filter with the 10 per cent, am- monium hydroxide solution, letting the solution run into a loo-cc. beaker containing 10 cc. of hydrochloric acid, specific gravity 1.20, and 0.5 gram of citric acid; add 30 cc. of ammonium hydroxide, specific gravity 0.90, cool, and then add 10 cc. of the magnesia mixture very slowly, while stirring the solution vigorously. Set aside in a cool place for 2 hours, filter and wash with the 10 per cent, ammonium hydroxide solution. Ignite and weigh. Dis- solve the precipitate of magnesium pyrophosphate with 5 cc. of nitric acid, specific gravity 1.20, and 20 cc. of distilled water, filter and wash with hot water. Ignite and weigh. The differ- ^ Use of Camp's agitator is suggested (Fig. 19). 142 ^ngine;ering chemistry ence in weights represents pure magnesium pyrophosphate con- taining 27.84 per cent, of phosphorus. Note. The ^mmonium molybdate solution should be kept in a cool place and should always be filtered before using. Determination of Phosphorus by the Alkalimetric Method. (Routine.) S0I.UTIONS Require^d. Nitric Acid. — Mix 1,000 cc. of nitric acid, specific gravity 1.42, and 1,200 cc. of distilled water. Nitric Acid for Washing. — Mix 20 cc. of nitric acid, specific gravity 1.42, and 1,000 cc. of distilled water. Potassium Permanganate. — Dissolve 25 grams of potassium permanganate in 1,000 cc. of distilled water. Ammonium Bisulfite. — Dissolve 30 grams of ammonium bisul- fite in 1,000 cc. of distilled water. Ammonium Molybdate. Solution No. i. — Place in a beaker 100 grams of 85 per cent, molybdic acid, mix it thoroughly with 240 cc. of distilled water, add 140 cc. of ammonium hydroxide, specific gravity 0.90, filter and add 60 cc. of nitric acid, specific gravity 1.42. Solution No. 2. — Mix 400 cc. of nitric acid, specific gravity T.42, and 960 cc. of distilled water. When the solutions are cold, add solution No. i to solution No. 2, stirring constantly; then add o.i gram of ammonium phos- phate dissolved in 10 cc. of distilled water and let stand at least 24 hours before using. Potassium Nitrate, i Per Cent. — Dissolve 10 grams of potas- sium nitrate in 1,000 cc. of distilled water. Phenol phthalein Indicator. — Dissolve 0.2 gram in 50 cc. of ,95 per cent, ethyl alcohol and 50 cc. of distilled water. Standard Sodium Hydroxide. — ^Dissolve 6.5 grams of purified sodium hydroxide in 1,000 cc. of distilled water, add a slight excess of i per cent, solution of barium hydroxide, let stand for ENGINEERING CHEMISTRY I43 24 hours, decant the liquid, and standardize it against a steel of known phosphorus content, as determined by the molybdate magnesia method, so that i cc. will be equivalent to o.oi per cent, of phosphorus on the basis of a 2-gram sample (see notes). Protect the solution from carbon dioxide with a soda lirfte tube. Standard Nitric Acid. — Mix lo cc. of nitric acid, specific gravity 1.42, and 1,000 cc. of distilled water. Titrate the solu- tion against standardized sodium hydroxide, using phenol- phthalein as indicator, and make it equivalent to the sodium hydroxide by adding distilled water. Method. In a 300-cc. Erlenmeyer flask dissolve 2 grams of steel in 50 cc. of the nitric acid. Heat the solution to boiling and while boil- ing add about 6 cc. of the potassium permanganate solution and continue boiling until manganese dioxide precipitate. Dissolve this precipitate by additions of the ammonium bisulfite solution, boil until clear and free from brown fumes, cool to 80° C, add 50 cc. of the ammonium molybdate solution at room temperature, let stand i minute, shake or agitate for 3 minutes, and filter on a 9-centimeter paper. Wash the precipitate three times with the 2 per cent, nitric acid solution to free it from iron, and continue the washing with the i per cent, potassium nitrate solution until the precipitate and flask are free from acid. Transfer the paper and precipitate to a solution flask, add 20 cc. of distilled water, 5 drops of phenolphthalein solution as indicator, and an excess of standard sodium hydroxide solution. Insert a rubber stopper and shake vigorously until solution of the precipitate is complete. Wash oif the stopper with distilled water and determine the excess of sodium hydroxide solution by titrating with standard nitric acid solution. Each cubic centi- meter of standard sodium hydroxide solution represents 0.0 1 per cent, of phosphorus. Notes. The ammonium molybdate solution should be kept in a cool place and should always be filtered before using. 144 ENGINEERING CHEMISTRY All distilled water used in titration should be freed from carbon dioxide by boiling or otherwise. Bureau of Standards Standard Steel No. 19 (a) is recommended as a suitable steel for standardization of the sodium hydroxide solution. Determination of Sulphur by the Oxidation Method. Solution Required. Barium Chloride. — Dissolve 100 grams of barium chloride in 1,000 cc. of distilled water. Method. In a 400-CC. beaker dissolve 5 grams of the steel in a mixture of 40 cc. of nitric acid, specific gravity 1.42, and 5 cc. of hydro- chloric acid, specific gravity 1.20, add 0.5 gram of sodium car- bonate and evaporate the solution to dryness. Add 40 cc. of hydrochloric acid, specific gravity 1.20, evaporate to dryness and bake at a moderate heat. After solution of the residue in 30 cc. of hydrochloric acid, specific gravity 1.20, and evaporation to sirupy consistency, add 2 to 4 cc. of hydrochloric acid, specific gravity 1.20, and then 30 to 40 cc. of hot water. Filter and wash with cold water, the final volume not exceeding 100 cc. To the cold filtrate add 10 cc. of the barium chloride solution. Let stand at least 24 hours, filter on a 9-centimeter paper, wash the pre- cipitate first with a hot solution containing 10 cc. of hydrochloric acid, specific gravity 1.20, and i gram barium chloride to the liter, until free from iron ; and then with hot water till free from chloride. Ignite and weigh as barium sulphate. Keep the washings separate from the main filtrate and evap- orate them to recover any dissolved barium sulphate. Note. A blank determination on all reagents used should be made and the results corrected accordingly. Determination of Sulphur by the Evolution Titration Method. (Routine.) Apparatus. Use a 480-CC. flask with a delivery tube and a 300-cc. tumbler of tall form (Fig. 20). e;ngine:e:ring chemistry 145 Capacity Iboz. =480cc Fig. 20. — Apparatus for determination of sulphur by the evolution method. SoivUTlONS RE)QUIRED. Dilute Hydrochloric Acid. — Mix 500 cc. of hydrochloric acid, specific gravity 1.20, and 500 cc. of distilled water. Ammoniacal Cadmium Chloride. — Dissolve 10 grams of cad- mium chloride in 400 cc. of distilled water and add 600 cc. of ammonium hydroxide, specific gravity 0.90. Potassium lodate. — Dissolve 1.116 gram of potassium iodate and 12 grams of potassium iodide in 1,000 cc. of distilled water. Standardize with a steel of known sulphur content. Each cubic centimeter should be equivalent to o.oi per cent, of sulphur, when a 5-gram sample is used (see notes). Starch. — To 1,000 cc. of boiling distilled water, add a cold 146 ENGlNEEiRING CHEMISTRY suspension of 6 grams of starch in 100 cc. of distilled water; cool, add a solution of 6 grams of zinc chloride in 50 cc. of dis- tilled water, and mix thoroughly. Me:thod. Place 5 grams of steel in the flask and connect the latter as shown in Fig. 17. Place 10 cc. of the ammoniacal cadmium chloride solution and 150 cc. of distilled water in the tumbler. Add 80 cc. of the dilute hydrochloric acid to the flask through the thistle tube, heat the flask with its contents gently until the solu- tion of the steel is complete, then boil the solution for ^ minute. Remove the tumbler which contains all the sulphur as cadmium sulphide, and to it add 5 cc. of starch solution and 40 cc. of the dilute hydrochloric acid, titrating immediately with potassium iodate solution to a permanent blue color. Notes. Extremely slow or rapid evolution of hydrogen sulphide is to be avoided. Bureau of Standards Standard Steel No. 8 (a) is recommended for standardizing the potassium iodate solution. Determination of Silicon by the Nitro Sulphuric Method. S01.UT10NS Required. Nitro-Sulfuric Acid. — Mix 1,000 cc. of sulphuric acid, specific gravity 1.84, 1,500 cc. of nitric acid, specific gravity 1.42, and 5,500 cc. of distilled water. Dilute Hydrochloric Acid. — Mix 100 cc. of hydrochloric acid, specific gravity 1.20, and 900 cc. of distilled water. METHOD. Add cautiously 80 cc. of the nitro sulphuric acid to 4,702 grams of steel, in a platinum or porcelain dish of 300 cc. capacity, cover with a watch glass, heat until the steel is dissolved and evaporate slowly until copious fumes of sulphuric acid are evolved. Cool, add 125 cc. of distilled water, heat with frequent stirring until all salts are dissolved, add 5 cc. of hydrochloric acid, specific gravity 1.20, heat for 2 minutes, and filter on a 9-centimeter paper. Wash the precipitate several times with hot water, then ENGINEERING CHEMISTRY I47 with hot hydrochloric acid and hot water alternately to complete the removal of iron salts, and finally with hot water until free from acid. Transfer the filter to a platinum crucible, burn off the paper carefully with the crucible covered, finally igniting over a blast lamp or in a muffle furnace at i,ooo° C. for at least 20 minutes ; cool in a desiccator and weigh. Add sufficient sulphuric acid, specific gravity 1.84, to moisten the silica and then a small amount of hydrofluoric acid. Evaporate to dryness, ignite and weigh. The difference in weights in milligrams divided by 100 equals the percentage of silicon. Note. A blank determination on all reagents used should be made and the results corrected accordingly. Determination of Silicon by the Sulphuric Acid Method. (Optional.) S01.UT10N Required. Dilute Hydrochloric Acid. — Mix 100 cc. of hydrochloric acid, specific gravity 1.20, and 900 cc. of distilled water. Method. To 2,351 grams of steel, in a beaker of low form of 500 cc. capacity, add 60 cc. of distilled water, and then cautiously 15 cc. of sulphuric acid, specific gravity 1.84. Cover with a watch glass, heat until the steel is dissolved and evaporate until copious fumes of sulphuric acid are evolved. Cool, add 100 cc. of dis- tilled water and heat with frequent stirring until the salts are in solution. Filter on a 9-centimeter paper, wash the precipitate several times with cold water, then with cold dilute hydrochloric acid until free from iron, and finally with cold water until free from acid. Ignite and weigh. Add sufficient sulphuric acid, specific gravity i .84, to moisten the silica and then a small amount of hydrofluoric acid. Evaporate to dryness, ignite and weigh. The difference in weights in milligrams divided by 50 equals the percentage of silicon. Note. A blank determination on all reagents used should be made and the results corrected accordingly. 148 ENGINEERING CHEMISTRY Determination of Copper. S01.UT10NS Required. Sulphuric Acid. — Mix 200 cc. of sulphuric acid, specific grav- ity 1.84, and 800 cc. of distilled water. Potassium Perrocyanide. — Dissolve 10 grams of potassium ferrocyanide in 100 cc. of distilled water. Standard Copper Nitrate. — Dissolve 2 grams of purest electro- lytic copper in 20 cc. of nitric acid (i : i), and dilute to 1,000 cc. with distilled water. Each cubic centimeter is equivalent to 0.02 per cent, of copper on the basis of a lo-gram sample. Method. In a 300-cc. beaker dissolve 10 grams of the steel in 75 cc. of the sulphuric acid, and then add 150 cc. of distilled water. Heat the solution and saturate with hydrogen sulphide, filter and wash the precipitate free from iron with i per cent, sulphuric acid containing hydrogen sulphide. Incinerate the paper with its contents in a porcelain crucible and fuse with 0.5 gram of acid sodium sulphate. Extract with hot water, filter, and complete the determination colorimetrically as under i (a) or i {h) , or electrolytically as under 2, as follows : 1. Evaporate the filtrate to about 25 cc, make faintly am- moniacal, filter into a loo-cc. Nessler tube and wash with hot water. {a) If the solution is a strong blue, to another 100 cc. Nessler tube add 50 cc. of distilled water, 5 cc. of ammonium hydroxide, specific gravity 0.90, and from a burette the standard copper- nitrate solution until the blue colors match. {h) If the solution is a faint blue, to the filtrate in a Nessler tube add the dilute sulphuric acid to faint acidity and then a few drops of the potassium ferrocyanide solution. To another loo-cc. Nessler tube add 50 cc. of distilled water, a few drops of the potassium ferrocyanide solution, and from a burette the standard copper nitrate solution until the reddish brown colors match. 2. Make the filtrate slightly acid with sulphuric acid, dilute with distilled water to a suitable volume, and determine the copper electrolytically. ENGINEERING CHEMISTRY I49 Determination of Nickel by the Gravimetric Dimethylglyoxime Method. Solutions Required. Hydrochloric Acid. — Mix 500 cc. of hydrochloric acid, specific gravity 1.20, and 500 cc. of distilled water. Dimethylglyoxime. — Dissolve i gram of dimethylglyoxime in 100 cc. of 95 per cent, ethyl alcohol. Method. In a 150-cc. beaker dissolve i gram of the steel in 20 cc. of the hydrochloric acid, and add about 2 cc. of nitric acid, specific gravity 1.42, to oxidize the iron. Filter the solution and add to the filtrate 6 grams of tartaric acid, and water till the volume is 300 cc. Make the solution faintly ammoniacal, then faintly acid with the hydrochloric acid and heat nearly to boiling; add 20 cc. of the dimethylglyoxime solution and then ammonium hydroxide, specific gravity 0.90, drop by drop till faintly alkaline, stirring vigorously. After standing i hour, filter on a weighed gooch crucible, wash with hot water, dry at no to 120° C. and weigh. The precipitate contains 20.31 per cent, of nickel. Notes. In making dimethylglyoxime solution, methyl alcohol may be substi- tuted for ethyl alcohol. The weight of sample taken should be varied according to the nickel content. Determination of Nickel by the Volumetric Dimethylglyoxime Method. {Routine.) S01.UT10NS Required. Hydrochloric Acid. — Mix 500 cc. of hydrochloric acid, specific gravity 1.20, and 500 cc. of distilled water. Dimethylglyoxime . — Dissolve i gram of dimethylglyoxime in 100 cc. of 95 per cent, ethyl alcohol. Silver Nitrate. — Dissolve 0.5 gram of silver nitrate in 1,000 cc. of distilled water. 150 ENGINEERING CHEMISTRY Potassium Iodide. — Dissolve 20 grams of potassium iodide in 100 cc. of distilled water. Standard Potassium Cyanide. — Dissolve 2.29 grams of potas- sium cyanide in 1,000 cc. of distilled v^ater. Standardize this solution by the procedure described below, against a steel of known nickel content as determined by the gravimetric dimethyl- glyoxime method, so that each cubic centimeter is equivalent to 0.05 per cent, of nickel on the basis of a i-gram sample (see notes). Method. In a 150-CC. beaker dissolve i gram of the steel in 20 cc. of the hydrochloric acid, and add about 2 cc. of nitric acid, specific gravity 1.42, to oxidize the iron. Filter the solution and add to the filtrate 6 grams of tartaric acid, and water until the volume is 300 cc. Make the solution faintly ammoniacal, then faintly acid with the hydrochloric acid, and cool thoroughly. Add 20 cc. of the dimethylglyoxime solution and then ammonium hydroxide, specific gravity 0.90, drop by drop, till faintly alkaline, stirring vigorously. After standing for a few minutes, filter on a gooch crucible and wash with hot water. Dissolve the pre- cipitate on the filter with 10 to 20 cc. of nitric acid (hot), specific gravity 1.42, added drop by drop, and then wash five times with hot water, using suction. To the solution in a 500-cc. beaker add 3 grams of ammonium persulphate and boil for 5 minutes. Cool, make distinctly ammoniacal, add 10 cc. each of the silver nitrate and potassium iodide solutions, and titrate with the standard potassium cyanide solution to a faint turbidity. Notes. In making dimethylglyoxime solution, methyl alcohol may be substi- tuted for ethyl alcohol. Bureau of Standards Standard Steel No. 33 is recommended for standardizing the potassium cyanide solution. The weight of sample taken should be varied according to the nickel content. ENGINEE^RING CHEMISTRY I5I Determination of Chromium. SoivUTioNS Re:quire:d. Hydrochloric Acid. — Mix 500 cc. of hydrochloric acid, specific gravity 1.20, and 500 cc. of distilled water. Sodium Carbonate. — A saturated solution; approximately 60 grams of sodium carbonate and 100 cc. of distilled water. Barium Carbonate. — Ten grams of finely divided barium car- bonate suspended in 100 cc. of distilled water. Standard Sodium Chromate. — Dissolve 2.6322 grams of sodium chromate in 1,000 cc. of distilled water. Each cubic cen- timeter is equivalent to 0.02 per cent, of chromium, when a 5-gram sample is used. Standard Potassium Permanganate. — Dissolve 2 grams of potassium permanganate in 1,000 cc. of distilled water,. Stand- ardize by using Bureau of Standards sodium oxalate,^ and dilute the solution with distilled water so that i cc. is equivalent to 0.02 per cent, chromium, when a 5-gram sample is taken. The factor Na^C^O^ »-► Cr = 0.2584 (using the 1913 atomic weights). ferrous Sulphate. — Dissolve 25 grams of ferrous ammonium sulphate in 900 cc. of distilled water and 100 cc. of sulphuric acid (i : i). Method. In a 300-cc. Erlenmeyer flask, covered, dissolve 5 grams of steel in 50 cc. of the hydrochloric acid. When completely dis- solved, add gradually the saturated solution of sodium carbonate until practically all the free acid is neutralized; finish the neu- tralization with the barium carbonate suspension, using an excess of about I gram of the carbonate. Boil the solution in the flask for 10 or 15 minutes, with the cover on. Filter the precipitate rapidly on paper and wash twice with hot water. Transfer the filter to a platinum crucible and after burning off the paper, fuse the residue for 10 minutes with a mixture of 5 grams of sodium carbonate and 0.25 gram of potassium nitrate. Dissolve the ^ Circular No. 40, Bureau of Standards, Oct. i, 19 12. 152 DNGINEKRING CHEMISTRY fusion in water, transfer to a beaker, add 2 cc. of 3 per cent, hydrogen peroxide, boil a few minutes and filter. Complete the determination of chromium in the filtrate by either of the follow- ing procedures : 1. If the solution is a strong yellow, add 10 cc. of sulphuric acid (i : i), and then the ferrous sulphate solution in measured excess. Cool thoroughly and titrate with the standard potassium- permanganate solution. The number of cubic centimeters of the potassium permanganate solution obtained, subtracted from the number corresponding to the volume of the ferrous sulphate solution used, will give the volume of the potassium perman- ganate solution equivalent to the chromium in the sample. 2. If the solution is a light yellow, cool the solution and transfer to a icmd-cc. Nessler tube. To another Nessler tube add distilled water, and from a burette add the standard-sodium chromate solution until the yellow colors match. Note. If procedure No. i is used, all hydrogen peroxide must be destroyed by boiling before acidifying, otherwise chromic acid will be reduced at this stage. STEEL. References. "Chemical Analysis of Special Steels, Steel-making Alloys and Graphites," by Charles Morris Johnson, N. Y., 1914. "The Determination of Chromium and Manganese in Steel," by Frd. C. T. Daniels, Jour. Ind. and Eng. Chem., Aug., 1914. "Determination of Carbon in Steel by the Direct Combustion Method," by Wm. Brady, Jour. Ind. and Bng. Chem., Oct., 1914. "Improved Method for the Determination of Nitrogen in Steel," by L. E. Barton, Jour. Ind. and Bng. Chem., Dec, 1914. "Manganese Steel," by John H. Hall, Jour. Ind. and Eng. Chem., Feb., 191 5. "Determination of Copper in Steel," by W. D. Brown, Jour. Ind. and Eng. Chem., Mar., 191 5. "Modern Steel Analysis," by J. A. Pickard, London, 1914. "The Chemical Analysis of Iron," by A. A. Blair, N. Y., 1913 e:ngine:kring chemistry 153 Analysis of Tin Plate. Dissolve 5 grams of tin or terne plate in 100 cc. hydrochloric acid (specific gravity i.io), in a 500 cc. graduated flask, with exclusion of air. When dissolved, cool, and fill up to the mark with water and thoroughly mix. Transfer 50 cc. to a beaker, and after adding starch paste, titrate the tin with standard iodine solution. A convenient strength of iodine is made by dissolving 5.38 grams of pure iodine in strong aqueous solution of potassium iodide and diluting to i liter. For the iron determination add mercuric chloride in excess to 50 cc. of tin plate solution, and titrate the iron with standard bichromate. The determination of manganese is quite important, since it shows whether iron or steel has been tinned. Treat 4 grams of tin plate, cut into small pieces, with hot dilute sulphuric acid for about 15 minutes. When the iron has dissolved, leaving the layers of tin and lead, add a little zinc and allow to stand for about 2 minutes. Filter and dilute to 20 cc. Take one-half of this filtrate, add 5 cc. nitric acid (specific gravity 1.20), and treat in the ordinary way with lead peroxide. The lead in tin plate is best determined as sulphate after first separating the tin by nitric acid. However, for ordinary work, it is sufficiently accurate to take lead by difference, allowing 0.25 per cent, for phosphorus, carbon, sulphur, silicon, etc., in addi- tion to the tin, iron and manganese previously determined. In order to test the accuracy of the iodine method for tin, a weighed quantity of pure tin, together with about forty times as much iron, was dissolved and the tin titrated. The result was as follows : Tin (gram) Amount taken 0.1255 Amount found 0.1266 The following are a few analyses that were made of British terne plate used for roofing : 154 DNGINEJ^RING CHEjMlSTRY Tin Lead Iron Manganese • • Carbon ] Phosphorus ' Sulphur i' Silicon, etc. J 1.58 7-97' 89.84 0.36 0.25 2.08 7.13' 90.23 0.31 0.25 2.40 8.89 ^8.10 0.31 0.25 99-95 3.37 11.98 84.18 035 0.25 100.13 1.60 2.481 95-31 0.38 0.25 100.00 2.54 7.48' 8935 0.36 0.25 100.00 1.97 8.12^ 89.29 0.37 0.25 VIII 1.96 7.09 90-55 0.32 0.25 100.17 IX 2.56 I0.231 86.64 0.32 0.25 The iodine method may be used for determining tin in all alloys which contain no metals that affect iodine. However, when the percentage of tin exceeds 10 per cent., as in the case of solder, the following method, although not quite so simple or rapid, is somewhat more accurate. In principle the scheme is simply a revision of the well known stannous chloride titration method for iron. Dissolve 5 grams of the tin alloy in strong hydrochloric acid in a 500 cc. graduated fiask, as in the case of tin plate. After diluting to the mark, fill a 50 cc. burette with the solution. Transfer 10 cc. of a standard ferric chloride solution (10 grams of iron in i liter) to a 4-ounce flask and heat to boiling. While boiling run the tin alloy solution cautiously into the ferric chloride until the yellow disappears. Cool and determine the excess of stannous chloride with stand- ard iodine solution (Fe^Cle + SnCl^ .= 2^eC\^ + SnClJ. Proc. Eng. Soc. W. Pa., 82, 182. Method of Sampling and Analysis of Tin, Terne and Lead-Covered Sheets.* Me:thod of SaMPIvING. Four 2 by 4-in. pieces are cut, one from each side of the sheet, parallel with the sides and equidistant from the ends, as shown in Fig. 21. One sheet from each grade or shipment is taken for analysis. These samples, before weighing, should be thoroughly cleaned ^ By difference. * Proceedings Amer. Soc. Testing Materials, J. A. Aupperle. ENGINEERING CHEMISTRY 155 with chloroform, carbon tetrachloride or gasoline. Each piece is then cut in half, marking one half "A" and the other half "B." The four pieces comprising lot A are then accurately weighed together, cut into small pieces about Ys inch square, thoroughly mixed, and used for the determination of tin and lead. The four pieces comprising lot B are reserved for the analysis of base metal and the direct determination of coating as a check on the analysis of lot A. "^ k-.--4-- 28' ■2 n i<-2>| /4 ■■ j-r M s : TT <::> CV4 i i j 1 1 it- " 1 •> Fig. 21. A templet should be provided, made preferably from steel ys inch thick and exactly 2 by 4 inches. A scribe is used to ac- curately mark the sections to be cut. The templet is then used to subdivide the 2 by 4-inch specimens into two pieces, 2 by 2 inches. The sections for analysis are then cut with tinner's shears. Determination of Tin. Three 5-gram portions of the finely cut sample of lot A are placed into three 300-cc. Erlenmeyer flasks, each fitted with a one-hole rubber, stopper containing a glass tube bent twice at right angles, one end of which projects through the rubber stopper for a short distance, the other end being long enough 156 Engine;e)ring chemistry to reach almost to the bottom of a beaker, placed on a level with the flask, containing about 300 cc. of dilute sodium-bicarbonate solution. Add 75 cc. of concentrated hydrochloric acid, connect the flask with the stopper containing the glass tube, and place the flask on a hot plate. Heat gradually at first until most of the metal is in solution. The long end of the glass tube, in the meantime, is submerged in the beaker. The hydrochloric acid solution is finally brought to boiling and when all the metal is dissolved the beaker containing dilute sodium bicarbonate solu- tion is replaced by one containing a saturated solution of the same. Remove the beaker and flask to a cool place. This will cause a small amount of the sodium bicarbonate to enter the flask and exclude the air. The solution is finally brought to a low temperature, preferably with ice water. This solution is then diluted to about 200 cc. with oxygen- free water which con- tains several cubic centimeters of starch solution, and titrated with N/20 iodine solution. We have found this strength of iodine solution to be the most satisfactory for this method. The distilled water free from oxygen is obtained in any of three ways : ( i ) By passing carbon dioxide through the cold distilled water; (2) by boiling vigorously and cooling; or (3) by adding a few cubic centimeters of concentrated hydrochloric acid to the water and then about 2 grams of sodium bicarbonate, stirring vigorously. By running this determination in triplicate, the first titration serves as a control to indicate the number of cubic centimeters of iodine required, whence the two succeeding titrations may be made very rapidly and should check very closely. Standardizing the Iodine Solution. — About o.i gram of pure tin and 4 grams of iron filings are dissolved in 75 cc. of concen- trated hydrochloric acid, etc., as under the determination of tin. One cubic centimeter of N/20 iodine ^ 0.002975 gram of tin. Calculation. — Weight of tin: Wt. of tin on 5 g. X W t. (g.) of 16 sq. in. ^ _ . _ ■ ~~ X 0.0421 — o number of pounds per case of 112 sheets, 20 by 28 in. DNGlNEEiRING CHEJMISTRY 1 57 Determination of Lead. Dissolve lo grams of the finely cut sample of lot A in 150 cc. nitric acid ( i : i ) . Heat until free from brown fumes and dilute to I liter and mix thoroughly. Take 100 cc. of this solution, add 10 cc. of concentrated nitric acid, electrolyze at a tempera- ture of 50 to 60° C, using I to 2 amperes and 2.3 to 2.5 volts. The weight of Pb02 is multiplied by 0.866. Calculation. — Weight of lead : PbO^ found (g.) X 0.866 X 20 = Pb ; Pb X Wt. (g.) of 16 sq. in. ^ ^ ^ X 8.6421 := 10 ^ number of pounds per case of 112 sheets, 20 by 28 in. Direct Determination of the Weight of Coating. The remaining four pieces representing lot B are used for the analysis of the base metal and incidentally can be used for the direct determination of the weight of coating. The four 2 by 2-inch pieces are carefully weighed together and each piece is wrapped with a stiff platinum or nickel wire in such a manner that it may be placed in the acid in a horizontal position. Heat 60 cc. of concentrated sulphuric acid contained in a 400-cc. Jena glass beaker to at least 250° C, immerse each piece separately in the hot acid for exactly i minute, and remove to a 600-cc. Jena beaker containing 50 cc. of distilled water. Immerse momentarily and rub the surface while washing with about 50 cc. more of distilled water, using a wash bottle for this purpose. The four samples are thoroughly dried, reweighed, and used for the analysis of base metal. ^ The loss in weight represents the coating and some iron. The sulphuric acid contained in the 400 cc. beaker is cooled and com- bined with the washings in the 600 cc. beaker. Two hundred cubic centimeters of concentrated hydrochloric acid are added and the solution boiled for a few minutes. The solution is cooled, poured into a graduated 500 cc. flask and filled to the mark with distilled water. ^ The methods of analysis of the base metal are outside the scope of this paper and will not be given. 158 e:ngine:e;ring chemistry Determination of Iron. Place 100 cc. of this solution in a 300-cc. Erlenmeyer flask, add I cc. of a saturated solution of potassium permanganate to oxidize the iron and tin, heat to boiling and reduce with a few- drops of stannous chloride. Cool, pour into a liter beaker con- taining 400 cc. of distilled water, add 25 cc. of mercuric chloride, followed by 10 cc. of phosphoric acid and manganese sulphate solution, and titrate with N/io potassium permanganate. Calculation. — Grams. Four pieces 2 by 2 in. weigh 28.5686 Same after stripped in acid 24.1620 Loss, coating plus iron 4.4066 Iron as found by titration 0.4887 Weight of coating 3.9179 3.9179 X 8.6421 z= number of pounds per case of 112 sheets, 20 by 28 in. Tin in 100 cc. X 5 X 100 ^ .^. . ^ = percentage of tin. Weight of coating '^ PbOo (in 100 cc.) X 0.866 X 10 X 100 , -, , ^-^ — „, . ' \. — — — ■ = percentage of lead. Weight of coatmg In the analysis of tin plate, the weight of coating is expressed in pounds per box, which is a half case, or 112 sheets 14 by 20 inches; hence to obtain the weight of coating per box on tin plate, the number of pounds as obtained above is divided by 2. The remainder of the solution which has been used for the determination of iron can be used for the determination of tin as follows : Place three portions of 100 cc. each in three 300-cc. Erlenmeyer flasks. If any of the lead sulphate should or should not be removed in any of these portions, the accuracy of the tin determination is not affected. Add i gram of powdered anti- mony, connect with rubber stopper and glass tube described in the method of determination of tin in the sample of lot A, place on a hot plate, using dilute sodium bicarbonate solution as a trap, and heat until the solution becomes decolorized. Replace ENGINE^ERING CHE:mISTRY 159 the dilute sodium bicarbonate solution with a saturated solution of the same, remove from the hot plate, cool, dilute and complete the determination as described under the first method. Specifications. Ternepi^ate (Rooeing Tin). All roofing tin to be made of best quality soft steel as a basis, resquared, 112 sheets to the box. Black plate from which made to weigh per 112 sheets net in the black Tin when finished to weigh per 112 sheets net IC 14 by 20 inches Pounds 95 to 100 115 to 120 IC 28 by 20 inches Pounds 195 to 200 235 to 240 IX 14 by : inches Pounds 125 to 130 145 to 150 IX 28 by 20 inches Pounds 250 to 255 290 to 295 1. Coating on all roofing tin to be a mixture of pure new tin and pure new lead, thoroughly mixed and having a proportion of not less than 20 per cent, of tin and the balance lead ; coating to be thoroughly amal- gamated with the black plate by the palm oil process. 2. This coating must be applied so that the sheets be evenly and equally coated on both sides and the coating distributed equally over each sheet. 3. After the plate has been cleansed in a weak acid solution, it is to be thoroughly washed with water, after which nothing is to be brought in contact with the black plate but pure palm oil, pure new tin, and pure new lead. 4. Every sheet so coated must be free from all defects, blisters, bad edges and corners, and bare or imperfectly coated spots. Each sheet to be stamped with the brand, thickness of the plate, and name of the manufacturer. An affidavit to the above must be furnished by both the successful bidder and the superintendent of the works where the plates are made, which affidavit must accompany the delivery of the roofing tin. Tinned Pirate (Bright Tin). All tin to be made of best quahty soft steel as a basis, 112 sheets to the box. i6o ENGINEERING CHEMISTRY Black plate from which made to weigh per 112 sheets net in the black Tin when finished to weigh per 112 sheets net IC 14 by 20 inches Pounds 103 108 IXX 14 by 20 inches Pounds 155 160 IXXXXi4by2o inches Pounds 200 A margin of 2^2 per cent, less than that specified will be allowed, provided it can be shown by the contractor that he has endeavored to comply with the specifications regarding the weight of tin required. The tin is to be of the best quality of Straits, Malacca, or Australian. If other size sheets are required, the sample proportions of black plate and tin should be observed. The coating is to be thoroughly amalgamated with the black plate. This coating must be applied so that the sheets be evenly and equally coated on both sides and the coating distributed equally over each sheet. Every sheet so coated must be free from all defects, blisters, bad edges and corners, and bare or imperfectly coated spots. ALLOYS. This subject may be divided into three classes: 1. Alloys composed principally of copper and zinc, or of cop- per, tin and zinc, or tin and lead. 2. Alloys or compositions in which copper, tin, lead, or anti- mony are constituents. 3. Alloys not included in the first two divisions. Alloys of the first class may comprise brass, bronze, bell metal, gun metal, Muntz's metal, bar solder (i^ lead, ^ tin), etc. The analysis may be performed as follows (if composed of copper and zinc only) : Transfer i gram of the brass to a No. 3 beaker covered with a watch glass, and add gradually 25 cc. nitric acid ; when solution is complete, remove watch glass, after washing, allow solution to cool, transfer it to a 250-cc. flask, and add water to the containing mark. Mix thoroughly (the solution being at 15° C), and transfer 50 cc. of the solution to a No. 3 beaker, dilute sufficiently with water and precipitate the copper ENGINEE^RING CHEMISTRY . l6l electrically. Upon complete precipitation of the copper, the plati- num cone and spiral are removed from the solution, washed with water, and the washings added to the solution in the beaker. Add a few drops of nitric acid to the solution, boil and precipitate the zinc with a slight excess of sodium carbonate. Boil, filter, wash well with hot water, dry, ignite, and weigh as ZnO. Example : One gram brass turnings taken. Solution 250 cc. Fifty cubic centimeters of solution taken : Grams Platinum cone -f- Cn 28.175 Platinum cone 29.995 Cu 0.160 0.160 X 5 X 100 - ^ ^ — 80 per cent. Cu. Grams Porcelain crucible -f ZnO 17.655 Porcelain crucible 17.605 ZnO 0.050 0.050 X 65 ^ 0.040 X 5 X 100 - — ^ ^ =0.040 Zn. — = 20 per cent. 81 I ^ Per cent. Cu ; 80 Zn 20 Total 100 Where tin is also a component, the above method is varied as follows : Take I gram of the fine turnings and digest with nitric acid as above. Evaporate nearly to dryness, add 50 cc. warm water, filter by decantation into a 250-cc. flask, washing the precipitate thoroughly with hot water, dry it, ignite and weigh as SnOg, and calculate to Sn. The filtrate is made up to 250 cc. with water (15° C), thor- oughly mixed, and 50 cc. taken for copper and zinc as before. II i62 ^ngine;e;ring chejmistry Grams Porcelain crucible -f- Sn02 16.6743 Porcelain crucible 16.5220 Sn02 0.1523 Sn = 12 per cent. Grams Platinum cone + Cu 28.115 Platinum cone 27.995 Cu 0.120 Cu = 60 per cent. Grams Porcelain crucible + ZnO . . . '. 17.6750 Porcelain crucible 1 7.6052 ZnO 0.0698 Zn = 28 per cent. Resume. Per cent. Sn 12 Cu • 60 Zn 28 Total 100 For a method for the complete analysis of brass including iron, and antimony, consult article by. Albert J. Hall, Electro- chemical & Metallurgical Industry, Nov., 1908, pp. 444-446. The Estimation of Copper in Copper Slags.^ Weigh 5 or 10 grams of the slag into a No. 3 beaker, add 100 cc. hot water, cover and boil on iron plate. Add 30 cc. of HCl, 5 cc. at intervals of about a minute, rotating the beaker to prevent sticking. When the HCl is all added continue the boil- ing for 5 minutes, remove from the plate, add an additional 100 cc. of hot water and conduct into the solution a rapid stream of H^S for 5 minutes. Filter under exhaust through a porcelain gooch crucible, using a paper disc in the bottom thereof and wash three times with water containing a little H2S. It is not 1 Thorn Smith (Research Chemist, Ducktown Copper Works, Tenn.), Chem. Engineer. ENGINE^ERING CHEMISTRY 163 necessary to remove all of the precipitate from the beaker. Put the gooch crucible and contents back into the beaker, add 10 cc. HNO3 containing a little Br and nitrous acid, then add 20 cc. of ammonia and titrate with cyanide, filtering, if desirable first. Or better, neutralize the acid solution with ammonia, add 2 cc. each of HNO3 and H^SO^ and electrolyze. (Most slags of this character filter through the gooch crucible with difficulty owing to the small amount of colloidal silica present. In such cases the addition of a few drops of i to i HF will expedite the filtra- tion.) ExAMPi.Es oj? Alloys oe the First Ceass. Bell metal Brass Brass (yellow) Speculum metal Delta metal or "Sterro Muntz metal Mosiac gold Gun metal Pinchbeck { Mannheim ) ; Gold / Tin Parts 33.4 Cu. 65.0 91.0 83.0 80.0 Copper Parts 78.0 72.0 60.0 66.6 60.0 60.0 Zn. 35 -o 17.0 20.0 zinc Parts 28.0 40.0 38.2 (*i.8 Fe) 40.0 Sn. 90 The Good Effect of Deoxidizing Brass and Bronze Scrap and New Metal by Magnesium. Magnesium has now become one of the agents for deoxidizing metals, and manufacturers are beginning to realize that it is advantageous in melting various scrap. On account of the fact that it is a metal and not a non-metallic element like phosphorus, a slight excess over and above that necessary to deoxidize the metal is not injurious. Some experiments were recently made in Germany on the ef- fect of magnesium upon both scrap brass and bronze and new metal, and the results indicate that the castings were greatly improved. These results are given below. 164 ENGINEERING CHEMISTRY The quantity of magnesium used, 0.05 per cent., has been found ample for the deoxidizing. It is an error to employ more, and many of the unsatisfactory results have been brought about by using an excess. TABI.E Showing Results oe Tests Made on Brass, Red Metai. and Bronze, Both With and Without Magnesium No Alloy Composition of the Alloy Original Dimen- sions of Test Bars Tensile Strength Elon- gation Per Cent. Diameter in inches SecMArea in sq. in. I.bs. Lbs. sq. in. As A5 Brass 90% Scrap Brass, Gates and Sprues 10% Zinc .976 .988 .748 .766 11,968 12,540 16,000 16,300 3.2 2.8 As melted A 6 A 6 925 948 .672 705 15,378 15,136 22,800 21,400 3.5 4.5 Deoxidized with 0.05^0 Magnesium C I C I Brass All Scrap, Consisting of Chips, Filings and Grindings 972 .976 .742 .748 12,144 13,112 16,300 17,500- I.O 2.0 As melted C 2 C 2 .976 .976 .748 .748 20,280 20,416 27,100 27,200 7-5 6.5 Deoxidized with 0.05% Magnesium B 4 B 4 Red Brass 95% Scrap Red Brass in Form of Gates Sprues, etc. 5% Zinc 968 976 736 748 15,202 13,640 20,600 18,200 5.0 3-0 As melted B 3 B 3 .976 .976 •748 .748 21,846 20,790 29,200 27,800 8.0 7.0 Deoxidized with 0.05^ Magnesium D I D I Bronze 90% Copper and 10% Tin .968 .964 .736 .728 18,700 18,458 25,400 25,300 lO.O 8.0 As melted D 2 D 2 .964 .964 .728 .728 23,100 23,232 31,700 31,900 13-0 12.0 Deoxidized with 0.05 9& Magnesium ENGINEE^RING CHE^MISTRY 165 According to these tests, the strength of the test bars was increased from 30 to 40 per cent., and the elongation from 40 to 60 per cent. Alloys of the second class may comprise Babbitt metal, Britannia metal, type metal, white metal, camelia metal, Tobin bronze, ajax metal, car-box metal, manganese bronze, magnolia metal, etc. ExAMPi^Es OF Ar.i.0Ys OF THE Second Class. Iron Tin Anti- mony I,ead Copper Zinc Bis- muth Phos. Babbitt metal 45-5 90.0 80.0 85.0 77.8 5U.O 40.0 0.9 lO.O 12.40 4-75 22.90 4-25 10.98 45-50 8.00 16.6 82.0 3-0 13.00 10,00 14.5 19.4 5-0 15.0 ^4.38 40.0 1-5 Pewter ••• 20.0 A Qhhnrv m*^tn 1 ...... . 2.8 Soft solder 50.0 550 0.4 9-5 227 80.0 27.10 H.75 7-37 84-33 0.2 61.2 79.70 82.67 trace 37.3 PViosr^Virtr Vtrnnyf^ .... 0.8 0.005 Deoxidized bronze • • • IVTjjcrnol 1 a mftal . . ... 0.20 2.45 0.25 50.0 "•55 70.20 81.28 34.1 77.0 4.4 6 10.20 0.68 20.4 Aiav TTiftal 0.37 Oar-V»r»Y mptnl ....... o.6i Pjir«jr>n'« ■tjeViitf mftal . "B" alloy, P. R. R... •....• 15.00 trace 80.0 White metal 12.0 15.0 82.0 99-7 Tvne " Shot " ars.0.3 Analysis of Babbitt Metal.^ Two grams of drillings in an 8-ounce beaker are treated with 30 cc. nitric acid (specific gravity 1.20) and heated till decom- position is complete and the free acid nearly all evaporated. When about 5 cc. of the solution remain, add 15 cc. of water, and then add concentrated sodium hydroxide solution till nearly neutral ; 50 cc. of sodium sulphide solution are then added, the mixture well stirred, and then boiled gently for ^ hour. ^ Some varieties of Delta metal contain ^ Method of E. M. Bruce, modified. to 2 per cent, of tin. l66 ' ENGINEJDRING CHEMISTRY The solution then contains the tin and antimony. The precip- itate, which contains the sulphides of lead and copper, is filtered on a 9-centimeter Swedish filter, and washed thoroughly with water containing i per cent, of the above sodium sulphide solu- tion. The filtrate is received in a 300 cc. beaker. Tin and Antimony. — The filtrate is diluted to 200 cc. and boiled. Crystals of oxalic acid, C. P., are cautiously added till the sodium sulphide is all decomposed and a milky separation appears, mixed with a precipitate which is usually at first black. Boil for 20 minutes. Pass hydrogen sulphide for 10 min- utes. Filter rapidly on a gooch crucible and wash with hot water. Dry and heat crucible and contents in a stream of carbon dioxide to a temperature above 300° C. for one hour. Cool in carbon dioxide, remove crucible and weigh as Sb^Sg. The gooch crucible containing the Sb^Ss + S may be treated with alcohol, then carbon disulphide, (in order to remove the sulphur), then alcohol dried and weighed, instead of igniting in carbon dioxide. SKSs X 0.71390 z= Sb. The filtrate from the SbgSg is treated with 30 cc. concentrated sulphuric acid and boiled down till all oxalic acid is decomposed and strong fumes of sulphuric acid come off. Cool. Dilute cau- tiously to 200 cc, mix well and filter quickly. Dilute filtrate to 300 cc, warm slightly and pass hydrogen sulphide. Filter stan- nous sulphide and wash with hot water. Dry, ignite, and weigh as stannic oxide in porcelain crucible. SnO^ X 0.788 ■=^ Sn. The copper and lead sulphide precipitate is washed off the filter, treated with dilute nitric acid, warmed till decomposed, and the sulphur filtered off. The lead is then separated as sul- phate by evaporation with sulphuric acid. The lead sulphate is filtered on a gooch crucible, washed with water containing 5 per cent, sulphuric acid, dried, and ignited over a Bunsen burner. PbSO^ X 0.68298 ,= Pb. The copper is separated from the filtrate by hydrogen sulphide. The sulphide is decomposed by nitric acid, and the resulting solution titrated or electrolyzed. Sodium sulphide solution for Babbitt analysis is made up as DNGINEEJRING CHEMISTRY 167 follows : One pound sodium sulphide crystals are dissolved in 2 liters of water. Portions of this are from time to time saturated with hydrogen sulphide gas and filtered for use. Separation of Tin and Antimony in Alloys. Mengin treats the alloy (for instance, anti-friction metal) with nitric acid (1.15), collects the insoluble oxides of tin and anti- mony, w^ashes, carefully ignites and weighs them, =1 M. The mixed oxides are next suspended in hydrochloric acid and water and a ball or plate of pure tin added, whereupon the antimony is reduced to metal and the tin converted into chloride; the reac- tion is best accelerated by heat, about 3 hours being necessary for 2 grams of the oxides. The precipitated antimony is washed by decantation with water, then with alcohol, dried and weighed =^ A. There is no appreciable oxidation of the antimony and the method is very exact. The tin is estimated by difference. M — Ax 1.262 = weight of tin oxide; the latter multiplied by 0.7888 gives the weight of tin in the alloy. An alternative method for the estimation of the tin is to precipitate the latter by zinc. The following figures (indicating grams) of an analysis, show the accuracy of the method : Samples taken Oxides found Metals found Sn 1. 162 ) ( Sn 1. 154 Sb 1.312 ^ -^- ^ I sb ..1.309 Third Class May Comprise. Aluminum bronze Al 7.3, Si 6.5, Cu 86.2, or Al 10, Cu 90 Ferro-aluminum Al 1.25, Fe, etc., 99.75, or Al 12.50, Fe, etc., 87.50 Ferro-tungsten Fe 43.4, W 53.1, Mn 3.5 German silver Cu 50, Ni 14.8, Sn 3.1, Zn 31.9 Rosine Ni 40, Ag 10, Al 30, Sn 20 Metalline Co 35, Al 25, Cu 30, Fe 10 Aluminum "bourbounz" Al 85.74, Sn 12.94, Si 1.32 Silicon bronze Fe, etc., 86.59, Si 13.41 Guthrie's "Eutectic" Cd 14.03, Sn 21.10, Pb 20.55, Bi 50 Arsenic bronze Cu 79.70, Sn 10, Pb 9.50, As 0.80 Manganese bronze Cu 88, Sn 10, Mn 2 Packf ong Cu 44, Ni 16, Zn 40 The following new alloys are mentioned by A. M. Farlie, i68 Engine:ering chemistry chemist, Tennessee Copper Co., in the journal of Metal Industry, Sept., 1906: "Cupro Magnesium" Copper 90%, Magnesium 10% Hydraulic bronze Copper 75%, Zinc 14%, Tin 11% Hardware metal. .. Copper 50%, Zinc 34.9%, Aluminum 0.10%, Nickel 15% "Phono-Electric" Mare Copper 98.55%, Tin 1.4%, Silicon 0.05% Sterline Copper 68.52%, Zinc 12.84%, Nickel 17.88%, Iron 0.76% Platinoid Copper 54.04%, Zinc 20.42%, Lead 0.15%, Nickel 24.77% Iron 0.47%, Manganese 0.15% Manganese resistance metal Cu 85%, Iron 3%, Manganese 12% Manganin. . .Copper 82.12%, Nickel 2.29%, Iron 0.57%, Manganese 15.02% Trolley wheel bronze Copper 92%, Zinc 2%, Tin 6% Hydraulic metal Copper 83.05%, Nickel 6%, Iron 10.81%, Lead 0.10% Acid-resisting metal Copper 82%, Zinc 2%, Tin 8%, Lead 8% Victor metal Copper 49.94%, Nickel 34.27%, Zinc 15.40% Aluminum 0.11%, Iron 0.28% Needle metal Copper 84.96%, Zinc 5.31%, Tin 7.96%, Lead 1.77% Pattern bronze Copper 90%, Zinc 2.50%, Tin 6.00%, Lead 1.50% Turbine wheel mixture. .Copper 86.77%, Zinc 3.48%, Tin 8.68%, Lead 1.07% Aluminum silver Copper 57%, Zinc 20%, Nickel 20%, Al 3% Cuprum magnesium is used in the proportion of i pound to 100 pounds of copper as a deoxidizing agent. Impairs the con- ductivity of copper less than silicon or any other deoxidizer. ''Hydraulic bronze" is a metal especially adapted for steam uses. ''Phono-electric wire" is used for trolley wire, telephone wire, etc. It has a much higher tensile strength than pure copper, but only 40 per cent, of its conductivity. In the manufacture of the alloy, the silicon is nearly all slagged off, consequently little or none can be found in the finished wire by chemical analysis. "Sterline" a white metal, but classed among the copper alloys on account of its high copper tenor. It is used as an imitation silver. New White Metai, Ai^lgys. Kayserzinn Copper 1.58%, Tin 92.98%, Antimony 5.44% Tempered lead Tin 0.98%, Lead 98.51%, Antimony 0.11%, Sodium 1.3% Improved Britannia metal Copper 2.31%, Tin 90.10%, An 7.44% Mn 0.15% Soft bearing metal Copper 0.42%, Tin 11.40%, Lead 80.65% Antimony 7.53% Alkali-resisting alloy Nickel 5.0%, Iron 95% White brass Copper 2%, Zinc 34%, Tin 64% EjNGINIvERING CHEMISTRY 169 Platinoid is used largely in the manufacture of electrical in- struments. ' Manganese resistance metal is used as a resistance material in place of German silver. The electrical conductivity is only 3 to 4.5 per cent, that of copper. Manganin is used as resistance material. The nickel increases the melting point, permitting a higher heat v^ithout danger of fusing. The nickel also decreases the temperature coefficient of the electrical resistance. Trolley Wire Bronze. — The name implies the use. The zinc is added to give solidity to the casting. The introduction of lead into the alloy results in increased wear. Hydraulic metal resists the action of acid mine water better than either red brass, muntz metal, copper-tin bronze, or man- ganese bronze. It is remelted once before using. Acid-resisting metal, said to be the best strong metal for resisting the action of acid, the lead alloys being too soft for many purposes. Phosphorus is added in the form of phosphor- tin, 0.125 pound of 5 per cent, phosphor-tin to every loo pounds of the alloy. Any excess of phosphor-tin over the amount speci- fied with result in blow holes. The phosphor-tin is added last, and the metal is cast at the lowest temperature at which it will run. The mixture is reported to be particularly useful for sul- phite pulp mill fittings. It will resist the action of nitric acid. Victor metal, another white metal high in copper. It is whiter than German silver, but cannot be rolled. It withstands the action of salt air and water and consequently is used largely for marine work. To make the alloy, melt the nickel and copper to- gether under borax, then add 2 ounces of aluminum for each 100 pounds of alloy, and finally add the zinc. Needle metal, so-called on account of its fluidity. Pattern Bronze. — The name implies its use for making pat- terns. It files well and casts sharply. Turbine-Wheel Mixture. — Inferior in strength to manganese- bronze, but less liable to shrinkage spots and porous areas, which 170 ENGINEERING CHEMISTRY are particularly objectionable in turbine wheels. The metal should be poured cool. Aluminum Silver. — A strong white metal strong in copper. As it does not tarnish in the air, it can replace steel in many in- stances, e. g., parts of typewriters, adding machines and similar contrivances. The metal is cast and remelted before use. Kayserzinn, — This is practically Britannia metal under a new name. It appeared in 1903, and was imported to this country from Berlin. Tempered Lead. — To make this alloy, melt the lead, and push the sodium in small pieces into the molten metal. Ingots of the alloy should be coated with paraffin to prevent oxidation. It is much harder than pure lead. Soft bearing metal, used for lining car boxes. Alkali-resisting alloy, sometimes contains as much as 10 per cent, nickel. Used in the manufacture of machinery which comes in contact with soap, washing soda, bluing or starches. Alloys containing zinc, tin, lead, aluminum, antimony or silicon are easily corroded by caustic alkali. White Brass. — A good anti- friction metal having double the electrical conductivity of Babbitt, and therefore useful in elec- trical machinery. Analysis of Aluminum Bronze. Take i gram of bronze in fine turnings, transfer to a No. 3 beaker and add gradually 25 cc. of aqua regia. Evaporate to dryness, to render the silica insoluble, take up with 25 cc. hydro- chloric acid, 25 cc. water, warm, filter, and wash well. The resi- due is dried, ignited and weighed as SiO^, and calculated to Si. The filtrate from the silica is diluted to 250 cc, thoroughly mixed and 100 cc. transferred to a No. 3 beaker and the copper pre- cipitated with hydrogen sulphide, filtered, washed with hydrogen sulphide water, the cupric sulphide dissolved in nitric acid, and the copper determined by electrolysis. The filtrate from the cupric sulphide is boiled to expel hydrogen sulphide, a few drops of nitric acid added, the solution made alkaline with ENGINEEJRING CHEMISTRY I7I ammonia, and the alumina determined as AUOg, and calculated to Al. Determination of Manganese in Manganese Bronze.^ Dissolve 5 grams of drilling in nitric acid of 1.20 specific grav- ity, using a large beaker to avoid frothing over. An excess of acid must be avoided as it interferes with the precipitation of the copper by hydrogen sulphide. When solution is complete, trans- fer to a 500 cc. cylinder without filtering out the precipitated stannic oxide. Make up to 300 cc. and pass a rapid current of hydrogen sulphide from a Kipp's apparatus until the supernatant liquid is colorless. Decant off through a dry filter, 180 cc. cor- responding to 3 grams of sample, and boil rapidly down to about 10 cc. Transfer to a small beaker and add 25 cc. of strong nitric acid. Boil down one-half, make up with strong nitric acid, boil, and add i spoonful of potassium chlorate. Boil 10 minutes and add another spoonful of potassium chlorate. Boil until free from chlorine, cool in water, and filter on asbestos, using filter pump. Wash with strong nitric acid through which a stream of air has been passed. When free from iron, wash with cold water until no acid remains. Place the felt and precipitate in the same beaker and dissolve in ferrous sulphate, using 5 cc. at a time. Titrate back with permanganate until a pink color remains. Deduct the number of cubic centimeters used in titrating back, from the number of equivalents of ferrous sulphate used, and the remainder shows the manganese in the amount of sample taken. Permanganate Solution. — Dissolve 1.149 grams of potassium permanganate in i,ocx) cc. water; i cc. equals o.ooi gram manga- nese; check by dissolving 0.1425 gram ammonio ferrous sulphate in a little water and acidulating with sulphuric acid. This should precipitate 10 milligrams of manganese. If not, apply factor of correction. Ferrous Sulphate Solution. — A solution of ferrous sulphate in 2 per cent, sulphuric acid so dilute that 5 cc. corresponds to 10 cc. permanganate solution. This is best made by trial and solution. ^ Jesse Jones: /. Am. Chem. Soc, 15, 414- ^•5 ^;5 O rt C J ^'a o a rt-O cs i* rot* do ^„- (U ^o5 4J 0*j -' cd ^ N 1 :ta a; O u 0-u) rt jt: o'-o 822 :?1 ClJ o c J L 1^ rttn (LI'S tec Ill's >><«' O 0; cja CO J: -,„ « 4^ w W I -y; co^.^ CO C^ 'w o 2 bag CO a; o CO cj o "S - e f rt tn n " CO j^ M*- CO ;- "^ CO CO"*- t; -; .. lU en 3 . m 11 1- C O — 00, = 3 o -' i; - b* h «-' o ^ o • - y •50 = co g CO ^T3 - CO i' :oijd^^ ■SirtSio'o"'o.o^.5f'H J^ •' CO 1, i^TcO „ .. .J 1) >-. E 5'^5 5 2 -2 ~ £ rt « |s o5S:5 =«S ^■ o ?i^ Sii'5-°\5-^ c 5 J! E-^ Sc - 5"Sii: P ot« •2 «^o t;'=^ii " »^ o Sf o •5x3 1;.^ "J 5i^ rt s iS--^ cOcct^o bfo-Cso -agJii Ctj:^ I. o ^co^-Scoaa^c-^-^x.^^ ^^ O ^ P- C CO 7- CO CO ^'^ u*i cOtC^ u n^-;- B 80 •3 3 ojE - ^la®* i.J^o'^'- S^-^^^ T3 s CO f, — .tT'- w!';.os_cBa — ^'f cfl£ -Ss Sts ^ S c bo's II c'u .5.2 ^go "" U3 5 CO (U o bco2 2 (« W Q jj '•Q w 'm "O u niirt o-o-r "'^ya 3— [fl a ih.-^jco c •-"o O"- oi cti"" >>35^5^-;;.e-'S >ffi U|.M .- co;;a .- 'o 4/ CO J v:: ^3 bo (T 3 CO w . .= a.j5.5«'S?;ai «^«J?i-,3f^ - -2 >.•"" a o o ? CI «|'Sii^^ltf8 3 "^^ a ^ -"'^ e ^ .y o :sr.y^ 4^ 2^5 CO § --o Soo^Sn-E.^ |.i'So^4.gx:& a— o*" " cd S^o-5S ENGINEERING CHEMISTRY I73 Analysis of Ferro Aluminum. Five grams of the ferro aluminum are transferred to a 500 cc. beaker and dissolved in 75 cc. sulphuric acid (specific gravity 1.30), then evaporated to dryness. The residue is treated with 50 cc. dilute sulphuric acid diluted to 300 cc. and mixed well; 100 cc. of the solution (= 1.666 grams) are filtered off into a graduated 100 cc. measure; this is then poured into a 250 cc. beaker; about 5 grams of pure iron wire are now added and the solution boiled, so as to reduce any ferric salt formed ; the excess of acid is carefully neutralized with solution of sodium carbonate and the mixture gradually poured into 150 cc. of a boiling solu- tion containing 30 grams of potassium hydroxide and 60 grams of potassium cyanide; the mixture of potassium cyanide with iron precipitated as hydroxide is diluted up to 500 cc. in a gradu- ated measure, and 300 cc. (= i gram of sample) filtered off into a 6-inch evaporating dish ; 200 cc. of a standard solution of am- monium nitrate are now added and the mixture heaied 40 min- utes; filter and wash the precipitated alumina with hot water, redissolve in 25 cc. of dilute hydrochloric acid, dilute to 2CX) cc, neutralize with ammonium hydroxide, add a slight excess, boil, filter, and wash with hot water, dry ignite, and weigh as Al^Og. The weight obtained multiplied by 0.534 X 100 ^ percentages of aluminum. This amount subtracted from 100 per cent, gives the percentage of iron. (Phillips.) German silver, rosine, aluminum *'bourbounz," Guthrie's "eutectic" and arsenic bronze all require solutions in nitric acid to render the tin insoluble, which is then separated by filtration from the other components. The determination of phosphorus in phosphor-tin presents some difficulty on account of the insoluble compound which phos- phoric acid forms with stannic oxide. HempeP states as follows : The ordinary way of analyzing phosphide of tin by dissolving it in aqua regia and precipitating with hydrogen sulphide is not satisfactory, as a considerable quantity of phosphorus is thrown down with the precipitated sulphide as a basic phosphate of tin. ^Ber. d. chem. Ges., 22, 2478; /. Anal. Chem., 4, 83. 174 ENGINEERING CHEMISTRY It is easily analyzed according to Wohler's method, by treating with chlorine, the chlorides of tin and phosphorus formed being collected in about lo cc. of concentrated nitric acid. If the appa- ratus be rinsed with a solution of i part concentrated nitric acid and 2 parts water, no trace of stannic oxide is precipitated. The phosphoric acid is now easily precipitated in the usual way by molybdic acid. If dilute nitric acid is taken, a part of the phosphorus separates with the stannic oxide and the result will be too low. This also applies to the determination of phosphorus in phosphor bronze. dualitative Tests of Alloys of Lead, Copper, Tin, and Antimony.^ For lead, dissolve in aqua regia. If much lead be present, it will separate on cooling as chloride; if only a small amount is present it will be detected by the addition of 4 volumes of 95 per cent, alcohol.^ For tin, dissolve in concentrated hydrochloric acid, and before the portion of alloy taken is completely dissolved, pour off the supernatant solution, cool to separate lead as chloride, add 4 vol- umes of alcohol, filter, and to the filtrate add a slight excess of bromine to convert stannous to stannic chloride; heat to expel free bromine, dilute, and pass hydrogen sulphide gas, when if tin is present it will be obtained as yellow stannic sulphide. For antimony, treat alloy with concentrated hydrochloric acid. Almost all the antimony is left undissolved. Decant, wash the residue with water, after which dissolve in hydrochloric acid with potassium chlorate, boil to expel free chlorine, pass hydrogen sulphide, obtaining a precipitate of Sb^aSg, if antimony is present. If copper is also present, it will be precipitated as copper sulphide and may obscure the color of the antimonic sulphide ; if so, filter and treat the precipitate with potassium hydroxide solution, which will dissolve the antimonic sulphide. Filter and acidify filtrate, when the pure color of antimonic sulphide will be ob- served if antimony is present. ^ Communicated to the author by G. W. Thompson, Chemist, National Lead Co., N.Y. 2 Consult, The Journal of Industrial and Engineering Chemistry, Vol. I, p. 520 (Aug., 1909). e:ngine;e;ring che:mistry 175 For copper, treat the alloy with dilute nitric acid in a porce- lain dish and evaporate to dryness; if copper is present, it will show as a green ring where it crystallizes out as nitrate on the edge of the residue. For arsenic, dissolve alloy in hydrochloric acid with addition of potassium chlorate in an Erlenmeyer flask, boil to expel chlo- rine, add some more concentrated hydrochloric acid and 2 ^rams of sodium thiosulphate, connect flask with a condenser and distil, following in principle the method first proposed by Fischer. Arsenic, if present, will be found in the distillate by passing through it hydrogen sulphide gas. Cluantitative Analysis of Alloys Containing Copper, Lead, Antimony, and Tin.^ One gram of the finely divided alloy is dissolved by boiling in from 70 to 100 cc. of the following solution, in a covered beaker. The solution is made by dissolving 20 grams of potassium chloride in 500 cc. of water, adding 400 cc. concentrated hydro- chloric acid, mixing, and then adding 100 cc. nitric acid of 1.40 specific gravity. No decomposition between hydrochloric acid and nitric acid takes place in this solution in the cold. If com- plete solution of the alloy is difficult in the amount of solution taken, more is added as required. Continue boiling until solution is evaporated to about 50 cc. Cool by placing beaker in cold water until the bulk of the lead has crystallized out as chloride, and then add slowly with constant stirring, icx) cc. 95 per cent, alcohol. Allow to stand about 20 minutes, filter through a 9-centimeter filter paper into a No. 4 beaker; wash by decanta- tion three times with mixture (4 to i) of 95 per cent, alcohol and hydrochloric acid, concentrated, and wash filter paper twice with the same mixture. Wash the lead chloride on the paper into a beaker, and wash filter paper several times with hot water, allowing washings to flow into the beaker with the rest of the chloride. Finally wash twice with solution of ammonium acetate, hot (the ammonium 1 Method of G. W. Thompson. 176 ENGINEERING CHEMISTRY acetate solution is made by taking one volume of ammonia water (specific gravity 0.900), adding to it i volume of water and then acetic acid strong until the reaction is slightly acid to litmus), heat until the lead chloride is dissolved, then add 15 cc. of a saturated solution of potassium bichromate, and warm until pre- cipitate is of good orange color. Filter on weighed gooch cru- cible, wash with water, alcohol, and ether, dry at 110° C, and weigh. Evaporate filtrate from lead chloride by heating on a hot plate and finally to dryness on water-bath; add 10 cc. solution potas- sium hydroxide (i gram to 5 cc.) and after a few minutes 20 cc. hydrogen peroxide; heat on water-bath for 20 minutes, add 10 grams ammonium oxalate, 10 grams oxalic acid, and 200 cc. of water. Heat to boiling, pass hydrogen sulphide with solution near boiling for 45 minutes; filter at once and wash precipitate with hot water. Boil filtrate to expel hydrogen sulphide, con- centrate if necessary, and electrolyze over night, using a current of about 0.5 ampere. Usually by morning the solution will have become alkaline, in which case it may be taken for granted that the tin is all precipitated on the cylinder. The cylinder is re- moved, washed twice with water and then with 95 per cent, alcohol, dried, and weighed. The precipitate of antimony and copper sulphides on paper is washed back into the beaker with the least amount of water possible, and treated with 10 cc. potas- sium hydroxide solution (i : 5), heated on a water-bath until un- dissolved matter is distinctly black; it is then filtered through the same paper it was washed from, into a 12-ounce Erlenmeyer flask, washed, etc. On the filter the copper is obtained as sulphide with a small amount of lead which failed of precipitation as chloride. If it is desired to determine this lead, it can be done by separation from the copper as usual ; if not, dry and ignite pre- cipitate in a small casserole, dissolve in nitric acid, boil to expel nitrogen dioxide, neutralize with sodium carbonate, add a few drops of ammonia, and determine volumetrically with potassium cyanide standardized against pure copper. The solution of anti- mony sulphide in potassium hydroxide should not amount to over ENGINEERING CHEMISTRY 1 77 40 cc. Add I gram potassium chlorate, 50 cc. concentrated hy- drochloric acid, boil until solution is colorless and free chlorine is driven off; filter through mineral wool ; if sulphur has separated into a similar flask, wash out original with concentrated hydro- chloric acid, cool, add i gram of potassium iodide, i cc. carbon disulphide, and titrate tor antimony with tenth-normal sodium thiosulphate, i cc. of which equals 0.0060 gram antimony. This systematic method assumes the absence of other metals than lead, tin, antimony, and copper. For the determination of other metals we offer the following suggestions : If arsenic is present it will be separated with the antimony and will liberate iodine, as does antimony. One cc. of tenth-normal thiosulphate equals 0.00375 gram of arsenic. Arsenic is preferably determined on a separate portion by dissolving in hydrochloric acid and potassium chlorate, boiling to expel free chlorine, and distilling after the addition of sodium thiosulphate as a reducing agent, passing hydrogen sul- phide through the distillate and weighing as As^Sg, or dissolving in potassium hydroxide and determining volumetrically as in the case of antimony. Bismuth^ and cadmium sulphides would remain with copper after treatment with potassium hydroxide — this renders this method very suitable for fusible metals. Zinc would interfere with this method, but as zinc does not alloy with lead, we will not speak of it further. Nickel and cobalt alloy but slightly with tin, and if present, should be sought for both in the precipitate left by potassium hydroxide and in the tin precipitated on a cylinder. Iron will also be precipitated with tin if present in an oxalic acid solution. Phosphorus is best determined by Dudley's method. - In alloys containing only lead and tin, with the tin under 20 per cent., the two constituents can best be determined by treat- ment with dilute nitric acid in a pocelain dish, evaporating to dryness on a water-bath, etc., and determining lead as chromate and tin as stannic oxide. In samples free from iron and copper, ^ Consult, "Note on the solubility of bismuth sulphide in sodium sulphide, etc.," Stillman, J. Atner. Chem. Soc, Vol. i8, Aug., 1896. 2 Am. Eng. and R. R. /., 8, 128 (1914)- 12 lyS EJNGIN^ISRING CHEMISTRY antimony may be determined directly by solution in hydrochloric acid and potassium chlorate, boiling to expel chlorine, and titra- ting as with pure antimony. Antimony in solders may be deter- mined very accurately by dissolving in hydrochloric acid without access of air and filtering out the undissolved antimony on a weighed gooch crucible. I have not found that a weighable amount of antimony was lost as stibine by this treatment. In the analysis of alloys of lead and tin, Richards' scales which are accurate within i per cent., may be used. In the examination of the various classes of alloys described at the beginning of this paper, various steps in their analysis may be left out with the absence of the respective metals. Rapid Method for Determination of Sb, in Presence of Sn, and Pb. Dissolve sample i to 5 grams according as content is o.i to i per cent, or over with chemically pure hydrochloric acid in 6 oz. flask. Pour off supernatant solution carefully. Dissolve Sb residue (washing not necessary) in dilute HCl and small amount of chlorate of potash. Boil, cool and titrate with thiosulphate of sodium potassium iodide and CS^ indicator. Shake up well with CS;2- Color reaction is very accurate (Louis Ruprect). A Method for the Rapid Quantitative Analysis of Bronze and Brass.^ (Pb, Cu, Sn, Sb, Fe, and Zn.) Introduction. The method of analysis reported in this paper is the outgrowth of an investigation that was undertaken as the result of a request made to one of us sometime ago to take charge of some "control" work for a large company manufacturing bronze metal in various forms. Most of the metal is sold on specification. The alloy supplied by this firm varies in composition as follows : Cu, 65 to 69 per cent. ; Pb, 25 to 30 per cent; Sn, 5 to 6 per cent. ; Fe, o.i to 0.3 per cent.; Zn, 1.5 to 3 per cent. The requirements were that the analysis be accurate to 0.2 per cent, and that the time consumed in making the complete analysis should not exceed I hour, preferably 45 minutes. ^ Presented at the 49th meeting of the American Chemical Society, Cincinnati, April 6-10, 19 14, by Richard Edwin Lee, John P. Trickey and Walter H. Fegely. e:ngine:e:ring chemistry 179 Part I. — Method of Analysis. De:termination of Lead. Procedure. — To 0.5 gram of the alloy in a 300 cc. Erlenmeyer flask, add 12 cc. of water containing 4 grams of tartaric acid, and then 4 cc. of concentrated HNO3. Place on a hot plate to dis- solve the alloy quickly. (The solution should be perfectly clear; if not, reject it and repeat the procedure.) Remove the solution from the hot plate, allow to cool from i to 2 minutes, add 10 cc. concentrated 11^804, and then heat on hot plate until all nitrous fumes are expelled. {Caution: Care must be exercised not to carry the procedure beyond this stage or the tartaric acid will be decomposed.) Dilute with 100 cc. of cold water, add 75 cc. of ethyl alcohol, shake, allow to stand for 2 or 3 minutes, filter through a weighed gooch crucible, wash with water to which a little H,2S04 has been added, until the precipitate is white and then wash out acid with alcohol. Reject the filtrate. Dry the precipitate in crucible on hot plate, and then heat to dull redness with the Bunsen burner for a few minutes. Cool and weigh as PbSO^. Wt. PbSO^ X 0.683 = wt. Pb Notes. 1. In order to prevent occlusion and adsorption of the Cu salts by the lead sulphate precipitate, it was found necessary to make the solution from which the lead precipitate is separated by filtration relatively large. The solution of the lead sulphate is prevented by the addition of alcohol. 2. In case the alloy contains a small percentage of lead, it is advisable, of course, to use a relatively large amount of the sample. If this is done, the quantities of the reagents employed, including water and alcohol, should be increased proportionately. The reasons for this procedure are obvious. In the first place, unless the volume of the solution is made larger the concentration of the Cu and other metals present becomes very large owing to the use of larger amounts of sample. This will increase the error due to occlusion and adsorption as mentioned in Note i. In the second place, in order to preserve the proper concentration of sulphate ions for the complete precipitation of Pb in this larger volume of solu- tion it is necessary to use larger amounts of the reagent, sulphuric acid. DETERMINATION OF CoPPER. Procedure. — Place 0.5 gram of the alloy in a 406 cc. beaker. i8o e;ngine;e;ring chemistry add 5 to lo cc. of dilute HNO3 (1:1), cover beaker with a watch glass until violent action ceases, then remove watch glass and evaporate on hot plate to sirupy consistency. Dilute to about 200 cc. and add KOH solution until a small precipitate of copper hydroxide persists after thorough stirring. Now add acetic acid until the copper precipitate is completely dissolved, then add a small excess of the acid. Cool the mixture to tap- water temperature, then add 40 cc. of the KI solution (100 grams to I liter) and 5 cc. of starch solution, and titrate immediately with standard Na^sSoOg solution until the blue color disappears. Notes. 1. It has been pointed out by Walker and Whitman' that the results obtained by following Low's iodide method" are uniformly a little low. They add "This error is not due to the method which gives exceedingly accurate results, but to the fact that nearly 6 per cent, of the copper is not precipitated as cuprous oxide. This loss is uniform, for if we add 6 per cent, of the copper determined the result will be the per cent, of the copper in the alloys." This is undoubtedly true. Their further state- ments, however, that when an alloy containing 5 per cent, of copper is decomposed by nitric acid, evaporated to dryness, taken up with nitric acid and filtered, the error in determining the copper in the filtrate will frequently be 0.5 to 0.7 per cent., cannot be confirmed by our experience. By following the proposed method of standardization and analysis, which is essentially Low's method, we have been able to check within o.i per cent, repeatedly. 2. Provision has not been made in the method as formulated for the separation of copper from any other metals, yet care must be exercised to exclude from the solution prepared for titration any substances which will either liberate or absorb iodine. Therefore, free CI, free Br, nitrous oxides, ferric ions, and As and Sb in the "ous" condition must be absent. Ferric ions may be removed by adding ammonium fluoride, which interacts with the former to produce ferric fluoride. This latter substance which is only slightly ionized has little or no oxidizing power and, there- fore, cannot liberate iodine under the existing conditions. The trivalent arsenic and antimony, if present, must be oxidized to the pentavalent condition by the addition of bromine. Excess of bromine must be removed by boiling the solution before titrating. No other elements interfere with the procedure. 3. Pb, Bi, and Cd, if present, interact with the KI and form the corresponding yellow insoluble iodides. This causes no trouble, however ; '^Journal of the Industrial and Engineering Chemistry, 1, (1909), 519. ^Jour. Am. Chem.Soc, 24, (1902), 1082. Engine:ering che:mistry i8i in fact many chemists regard the presence of one or more of these metals as a distinct advantage as the presence of the yellow precipitate assists the operator in securing uniform end points. In this laboratory it is cus- tomary to add a few cubic centimeters of a solution of lead acetate to the solution to be titrated if it is known that Pb is not present. 4. In order that the liberated iodine may be held in solution it is necessary to use rather large excess of KI. This procedure increases the speed of the reaction. 5. The presence of an excess of inorganic acid interferes- with the procedure. It should be remembered, however, that unless the solution contains a sufficient excess of acetic acid the end point will not be sharp, 6. The solution should always be cooled to tap water temperature just before the KI solution is added. 7. It should be remembered that the Sn present will make an insol- uble residue in the solution prepared for titration. Determination of Tin (and Sb). Procedure. — Weigh 0.5 gram of the alloy into a 400 cc. beaker, add 5 to 10 cc. of dilute HNO3 (1:1), cover with a watch glass until violent action ceases, then remove cover and evaporate to a paste on hot plate. Add 15 cc. of dilute HNO3, boil for several minutes and dilute to about 200 cc. Boil for a few min- utes, filter through a weighed gooch crucible, wash with hot nitric acid wash, and then with hot water. Pour the filtrate into a 500 cc. beaker and reserve for check determination of copper. Dry the crucible and contents on hot plate and then ignite at red heat for 10 minutes. The ignited residue consists of SnO,2 and any Sb (as Sb204) which may have been present in the alloy. Wt. SnOj (and Sb^OJ X 0.788 = wt. Sn (and Sb) Notes. 1. The results obtained by this method are usually a little high, owing to the fact that the precipitate frequently contains traces of the oxides of Cu, Sb, and Pb. However, the digestion of the dried residue in dilute nitric acid and the final separation of the precipitate from a large volume •of solution, tend to reduce errors from this source. The precipitate will also contain any phosphorus that is present in the sample. 2. Much time and energy were consumed in an effort to obtain a volumetric method for determining tin in a separate portion of the alloy. One of the most attractive volumetric methods for making this deter- mination is Walker and Whitman's modification^ of Low's idimetric ^Journal of the Industrial and Engineering Chemistry, 1, (1909), 519. i82 e:ngine:e:ring che;mistry method. Our efforts, however, to adapt It to our scheme of analysis were without success. The chief obstacle to its application to the rapid analysis of bronze is the presence of a relatively large percentage of Cu in the alloy. The method, however, was found to give excellent results when used in making analyses of Babbitt metal if the percentage of Cu in the alloy was small; but if the percentage of Cu was large the results came high, ow.ing to the fact that during the reduction of the Sn the Cu was reduced to cuprous chloride which takes up a portion of the iodine when the solution is finally titrated. Our results in regard to the errors intro duced by the titration of Sn in the presence of Cu agree with those obtained by Ibbotson and Aitchison.^ (Check Dete;rmination of Copper.) Procedure. — Cool the filtrate from the Sn determination, add KOH solution until a small precipitate of copper hydroxide per- sists after thorough stirring. Add acetic acid until the copper precipitate is completely dissolved, then add a small excess of the acid. Follow directions as given under Determination of Copper. Determination of Antimony. Procedure. — Weigh out 0.5 gram of the fine drillings of the alloy into a 300 cc. Kjeldahl flask, add 25 cc. of concentrated sulphuric acid and heat over the bare flame of a Bunsen burner. Keep the acid at its boiling point until the solution is clear or the residue is white. Cool, add 100 cc. of water, boil for several minutes and transfer the contents of the flask to a 400 cc. beaker. Dilute to 200 cc, heat to 70° C. (158° F.) and titrate with a standard KMn04 solution. The permanganate solution should be added rapidly until the permanganate color persists, then add several cubic centimeters in excess. Stir the solution vigorously, and then titrate with a standard solution of ferrous ammonium sulphate until the pink color just disappears. Notes. I. Although antimony is not found in the majority of bronzes and brasses it frequently occurs alloyed with variable percentages of lead, tin, copper and iron. Therefore, in order to make the present scheme of analysis as wide as possible in its application and thereby increase its usefulness, it was deemed advisable to incorporate a method for the rapid determination of Sb. 1 Chem. News, 107 (1913), 109; also C. A., 7 (1913), 2005; 8 (1914), 476. ENGINEERING CHEMISTRY 183 2. The method is at once recognized as a modification of Low's well known method/ The chief difficulty we experienced in fitting it to our scheme of analysis was the matter of securing Sb in a suitable condition in sulphuric acid solution. Alloys containing a high percentage of copper resist solution by the usual procedure. Nitric acid is ehminated as a solvent because of its oxidizing action ; and HCl and KCIO3 are ineffective unless the treatment is greatly prolonged. Finally, the not entirely satis- factory method of decomposing the alloy in concentrated sulphuric acid in a Kjeldahl flask exposed to the bare flame of a Bunsen burner was adopted. Complete decomposition is usually effected in 10 to 20 minutes, after which the determination may be readily finished in 10 minutes. After one or two trials, it will probably be found that the blue color imparted to the solution by the presence of the copper does not hinder the determination of the exact end point when the permanganate is added. 3. Demorest has pointed out in an excellent paper" that it is necessary to employ a large excess of potassium permanganate to complete the oxidation of the Sb. It is not best to have HCl present when the anti- mony is titrated as the end point is made very transient by its presence. Determination of Iron and Zinc. Procedure. — To 0.5 gram of the alloy in a 400 cc. beaker add sufficient dilute HNO3 (1:1) to dissolve the sample. Heat on hot plate until the alloy is thoroughly decomposed, then evaporate the solution just to dryness. Add 10 cc. of concentrated HCl and 100 cc. of water, heat to about 70° C. and pass H^S through the mixture until all the Pb, Cu, Sn, and Sb (Cd, etc.) are pre- cipitated. Filter, using a Buchner funnel with an asbestos mat, wash the precipitate with water. The filtrate which contains the iron and zinc should be transferred to a 500 cc. beaker. Note;. A drop of this filtrate should be transferred to a spot plate and tested with a drop of potassium ferrocyanide for the presence of Cu and Fe which are interfering substances and if present in weighable quantities they should be removed. If Cu is present, which will be indicated by the presence of red coloration, treat the filtrate again with hydrogen sulphide and filter ; if Fe is present, which will be indicated by the appear- ance of a blue coloration, proceed as directed under (2), Iron and Zinc. I, Zinc, if Fe is Absent. — Dilute the filtrate to 200 cc, heat to 70° C. and titrate with standard potassium ferrocyanide, using ^Journal of the Industrial and Engineering Chemistry, 5 (1913), 842. 2 Ibid.. 5 (1913), 842. 184 KNGINEKRING CHEJMISTRY ferric chloride or uranyl nitrate as an indicator. The titration should be performed slowly and with constant stirring in order to obtain the most satisfactory results. Continue to kdd the ferrocyanide until a drop of the solution in the beaker shows a bluish green tinge (brown tinge when uranyl nitrate is used as an indicator) when tested on a white porcelain plate with a drop of ferric chloride after standing a few seconds. The quantity of the standard solution which is acquired to produce a good end point in the blank determination made at the time of standard- izing the solution, must be subtracted from the amount of standard used in making the determination. Notes. 1. Correction for Blank. — As the indicators are not very sensitive under the imposed conditions, it is necessary to determine the excess of standard solution required to effect the color change of the indicator used. A "blank" must be run, therefore, using the same quantities of reagents under corresponding conditions of volume, temperature and acidity. 2. If the solution turns blue during the titration, it is an indication of the presence of small quantities of iron. 2. Iron and Zinc. — Add 2 cc. of concentrated HNO3 to the filtrate to oxidize the Fe, heat to boiling add 25 cc. 5/N NH^Cl and then NH4OH until the odor of the reagent barely persists after boiling the mixture for i minute. Filter off the Fe(OH)3 and ignite. Weigh as Fe^O^. Notes. 1. Zinc is completely precipitated from HCl solutions by potassium ferrocyanide as white zinc ferrocyanide. Such metals as Pb, Cu, Sn, Fe and Mn are also precipitated by this reagent, and therefore must be removed before the Zn is titrated. 2. The acid solution must not contain free CI, free Br or the oxides of chlorine as these substances decompose ferrocyanide. 3. Care must be taken to conduct the standardization as well as all determinations under corresponding conditions with particular reference to volume, temperature, acidity, amount of ammonium salts and the rate of titration. Furthermore, it is imperative that the titration be conducted slowly and with constant stirring of the solution. If this precaution is no observed the end point will be reached apparently before all the zinc is precipitated. DNGINEERING CHE:mISTRY 185 Specifications for Inspection of Material, Copper, Brass and Bronze. Used in the construction of machinery coming under the cognizance of the Bureau of Steam Engineering, Navy Department, U. S. (1908). Specifications for Miscellaneous Brass Castings. [88-10-2 Mixture.] For ali. Purposes for which no other Aeloy is Specified. The inspection of these castings shall conform to the "General instructions for the inspection of copper, brass, and bronze material com- ing under the cognizance of the Bureau of Steam Engineering." The composition must be made of materials of the purest commercial quality. The naval inspector will take drillings for chemical analyses. The analysis must show that the metal contains not less than 87 per cent, nor more than 89 per cent, copper ; not less than 9 per cent, nor more than 11 per cent, tin and the remainder zinc. Chemical Analysis. Government Analysis. — Drillings for analysis must be fine, clean, dry, and free from scale. The inspector may take them from any test piece, or from any part of the material, provided in this* last case that by so doing the material will not be rendered unfit for use. Unless otherwise requested, the chemist will make determinations of those elements only which are limited by the specifications. Specifications for Seamless Brass Pipe, Iron Pipe Sizes, Mads to Correspond with Iron Pipe and to Fit Iron Pipe Fittings. Material. — Pipe shall be made of material of purest commercial quality, compounded from 60 to 70 per cent, of pure copper and from 40 to 30 per cent, of pure zinc, and not more than 0.5 of i per cent, of lead, the manufacturer being allowed this variation of composition in order to get the material best suited for the purpose for which it is intended. The naval inspector will take drillings for chemical analyses. The pipe will be inspected for surface defects and it must be free from cracks, seams, and defects generally. Each pipe must withstand an internal hydraulic pressure which will subject the metal to a stress of 7,000 pounds per square inch without showing weakness or defects, in accordance with the formula for thin hollow cylinders under tension where i86 e:nginekring chemistry p = safe internal pressure ; d = inside diameter of pipe in inches ; J = safe tensile strength of material = 7,000 pounds per square inch; t = thickness of pipe in inches ; but no pipe will be tested beyond 1,000 pounds per square inch per gauge, unless specially directed. All pipe, unless ordered "hard," is to be annealed sufficiently to pre- vent fire cracking and to stand the physical tests. "Brass pipe for radia- tors for heating system, and similar purposes, to be semi-annealed, and to stand bending 180° around a diameter of i ^/le inches." When the pipe is finished (ready for shipment), the naval inspector will subject i per cent, of the lot, taken at random, to the following physical tests : (a) The end of each test pipe must stand being flattened by ham- mering until the sides are brought parallel, with a curve on the inside at the ends not greater in diameter than twice the thickness of the metal in the pipe, without showing cracks or flaws. (b) Each test pipe shall have a piece 3 inches long cut from it, which piece when split must stand opening out flat without showing cracks or flaws. (c) Each test pipe must stand threading in a satisfactory manner with the usual thread for the size of the pipe. When the pipe is ordered "hard," the (a) and (b) tests shall be made on annealed test specimens. These (a), (b), (c) tests shall be made on each of the test pipes, and the test specimens shall be furnished at the contractor's expense. If any of these pipes selected for tests fail, the naval inspector will select two extra pipes from the same lot and put them through the same test as the pipe that failed, and both of these pipes must be found satisfactory in order that the lot may be passed. The failure to pass satisfactorily any one of the tests marked (a), (b), (c) will reject the lot. All pipe shall be up to the gauge ordered. Each large single pipe, or bundle of small pipes, must be marked with the name of the vessel for which it is intended, or with the number of the order. The standard weight for seamless drawn brass pipe will be 0.3 pound per cubic inch of material, but a tolerance not to exceed 5 per cent, over weight will be allowed. Specifications for Seamless Pipe of ''Benedict Nickel" or Other Equivalent Metal. Material — They shall be made of material of purest commercial quality compounded from at least 15 per cent, nickel and the remaining ENGINEEJRING CHEMISTRY 187 metal pure copper. The naval inspector will take drillings for chemical analyses. The pipes will be inspected for surface defects. They must be free from cracks, seams, and injurious defects generally. Each pipe shall be subjected to an internal water pressure equivalent to a tensile stress of 14,000 pounds per square inch without showing weak- ness or defects, in accordance with the formula for thin hollow cylinders under tension where p = safe internal pressure ; d = inside diameter of pipe in inches ; j- izz safe tensile strength of material = 14,000 pounds per square inch; t = thickness of pipe in inches ; but no pipe shall be tested beyond 2,000 pounds per square inch per gauge, unless specially directed. When the pipe is finished (ready for shipment), the naval inspector will subject i per cent, of the pipe, taken at random, to the following physical tests : (a) Test specimens selected from these shall not show less than 65,000 pounds tensile strength per square inch. (b) The end of each test pipe must stand being flattened by ham- mering until the sides are brought parallel, with a curve on the inside at the ends not greater in diameter than the thickness of the metal in the pipe, without showing cracks or flaws. (c) Each test pipe shall have a piece 3 inches long cut from it, which piece when split must stand opening out flat without showing cracks or flaws. (d) Each test pipe must stand threading in a satisfactory manner with the usual thread for the size of the pipe. (e) The end of a pipe, cold, must stand having a taper pin, taper i^ inches to the foot, driven into it until the end of the piece stretches to i^ times the original diameter without showing cracks or flaws. (/) A piece of the pipe one diameter long must stand crushing in the direction of its axis under a hammer until shortened to three gauges of the pipe in height, without showing cracks or flaws. (g) A piece of pipe must stand flanging cold, the width of the flang- ing to be one-fourth of the inside diameter of the pipe. (h) Hot or Fire-Crack Test. — Pieces, 2 feet long, cut from each test specimen, shall be heated to about 300° F., and must then stand plunging into ice-cold water without showing cracks. l88 DNGINEIERING CHEMISTRY (i) Cold-Bending Tests — "Each, test specimen, cold, shall be sawed through the axis for a distance of i foot, then the split parts folded back across the grain, flat on themselves, without showing fracture. These tests shall be made on each of the i per cent, test pipes, and the test specimens shall be furnished at the contractor's expense. If any one of these pipes selected for test fails, the naval inspector will select two extra pipes from the same lot and put them through the same test as the pipe that failed, and both of these pipes must be found satisfactory in order that the lot may be passed. The failure to pass satisfactorily any one of the tests will reject the lot. All pipe shall be up to the gauge ordered. Each large single pipe or bundle of small pipes will be marked with the name of the vessel for which it is intended, or with the number of the order. The standard weight for seamless "Benedict nickel" pipe will be 0.314 pound per cubic inch of material. Specifications for Seamless Tubes for Condensers and Feed-Water Heaters. Material. — The tubes may be made of any one of the following com- positions, as required by machinery specifications or ordered on requisi- tion, viz., 70 copper, 29 zinc, i tin, which need not be tinned ; or 60 copper and 40 zinc, which must be tinned inside and outside ; or Benedict nickel, which must not be tinned. In every case all the metals used must be of the purest commercial quality. Test Specimens. — The naval inspector shall select at random from the finished tubes, when ready for shipment, a number of tubes equal to I per cent, of the entire order, for tests. These test tubes will be exclusive of the number required on the order and 'will be furnished at the con- tractor's expense. Weight. — All of the test tubes are to be weighed together, and their average weight taken to represent the weight of the whole order. All tubes must be up to the required gauge on the thinnest side. An excess of weight of not more than 5 per cent, will be allowed for all tubes. Tubes of 70-29-1 mixture must weigh not less than 0.308 pound per cubic inch of metal. Tubes of 60-40 mixture must weigh not less than 0.298 pound pei" cubic inch. Benedict nickel tubes shall weigh 5 per cent, more than the 60-40 mixture. Hot or Fire-Crack Test. — Pieces 2 feet long cut from each test tube will be heated to about 300° F., and at that temperature must stand plunging in ice-cold water without showing cracks. EJNGINKKRING CHEjMISTRY 189 Cold-Bending Tests. — Pieces 2 feet long cut from each test tube and sawed longitudinally for a distance of i foot must stand opening out flat and folding flat back cold across the grain without showing signs of fracture. Surface Inspection. — All tubes must be seamless, true to form, of an equal thickness throughout, of workmanlike finish, free from cracks, seams, and other defects, and stiff enough to lie straight when resting on supports 6 feet apart. Tinning and Annealing.— AW tubes of 60 per cent, copper and 40 per cent, zinc must, after final drawing, be annealed, acid cleaned, dipped in molten tin, and then immediately wiped with hempen tow inside and out to insure their being smooth on both surfaces. All tubes of 70 per cent, copper and 29 per cent, zinc and i per cent, tin must be annealed and acid cleaned. Benedict nickel tubes shall not be annealed. • The amount of annealing required for condenser tubes shall be suffi- cient to enable the tubes to pass the physical tests, and shall be sufficient to permit the tubes to be properly packed in the tube sheet without distortion. Tubes for feed-water heaters and gland stock and such other tubes (except Benedict nickel) as are larger and thicker than condenser tubes, must be sufficiently annealed to prevent cracking. Benedict nickel tubes are distinguished from brass by their greater hardness and density and by their color, which is like that of tin, and is uniform throughout the material. Water-Pressure Tests. — All finished tubes shall be subjected to 1,000 pounds internal water pressure without showing weakness or defect. Specifications for White Metal. For Anti- Friction Linings. Material. — The composition shall consist of 3.7 per cent, of the best refined copper, 88.8 per cent. Banca tin, 7.5. per cent, of the regulus of antimonj^ and shall be well fluxed with borax and resin mixing. When practicable, the weighing and mixing of the metals will be witnessed by a Government inspector. Otherwise as many chemical analyses will be taken as, in the judgment of the naval inspector, will show that the material is of the proper composition. If by reason of scarcity Banca tin cannot be procured, another stand- ard brand of tin may be proposed, subject to the approval of the Bureau of Steam Engineering. Specifications for Manganese-Bronze Castings. The castings must be made in accordance with the drawings and specifications — sound, clean, free from blowholes, porous places, cracks, 190 DNGINS^RING CHEMISTRY or any other defects which will materially affect their strength or appear- ance or which indicate an inferior quality of metal. In the case of screw propellers coupons will be cast attached to the hub and to each blade; the coupons will be cast in a horizontal position, and those on the blades will be attached at half the distance from the root to the periphery. The coupons will be cast 2 inches in diameter and turned down as required by specification. The coupons are to have no treatment other than machining to reduce them to the proper diameter. For castings weighing over 200 pounds test pieces or coupons shall be taken in such number and from such parts of the casting as will thor- oughly exhibit the quality of the metal. Castings weighing less than 200 pounds may be tested by lots, each lot to be represented by two test pieces. If the castings are too small for the attachment of coupons, the test pieces may be cast separately from the same metal under as nearly as possible the same condition as the casting. Coupons shall not be detached from castings until they are stamped by the inspector. If test pieces are cast separately from the casting, they must be cast in the presence of the inspector and stamped by him as soon as they are taken out of the molds. The test pieces shall show an ultimate tensile strength of not less than 60,000 pounds per square inch, an elastic limit of not less than 30,000 pounds per square inch, and an elongation of not less than 20 per cent, in 2 inches. The color of the fractured section of the test pieces and the grain of the metal must be uniform throughout. Specifications for Journal Boxes and Guide Gibs, and Other Parts of the Same Composition. Material — All metals used for this composition must be of the purest commercial quality. Analyses. — The naval inspector will take drillings for analyses and these must show that the composition consists of 83 per cent, copper, 13.5 per cent, tin and 3.5 per cent, zinc, no component to vary more than I per cent, above or below the amounts specified. Specifications for Brazing Metal. Used Principai^ly for Copper Pipe. The inspection of this metal shall conform to the "General instruc- tions for the inspection of copper, brass, and bronze material coming under the cognizance of the Bureau of Steam Engineering." Material. — This composition shall be made of materials of the purest commercial quality. Engindkring chemistry 191 The naval inspector will take drillings for analyses. The analyses shall show that the composition consists of not less than 84 per cent, nor more than 86 per cent, copper and the remainder zinc. Specifications for Rolled Copper, Muntz Metal, and Brass Sheets, Plates, and Rods. The inspection must conform to the "General instructions for the inspection of material coming under the cognizance of the Bureau of Steam Engineering." Material.— AW metal used either alone or in the manufacture of alloys must be of the purest commercial quality. The copper must be best Lake copper, or its equivalent. Analyses. — The naval inspector will take drillings for analyses. An analysis of the copper sheets, plates, and rounds must show that they contain not less than 99.5 per cent, pure copper. An analysis of Muntz metal must show not less than 59 per cent, copper and the remainder zinc. An analysis of brass must show that it is of the specified com- position, no component varying more than i per cent, in amount above or below that specified. Surface Inspection. — The sheets, plates, and rounds must be free from all surface defects ; the sheets and plates must be cut to the dimen- sions ordered. They are not to be less than the calculated weight, taking the weight of i cubic inch of hot-rolled copper to be 0.320 pound, i cubic inch of cold-rolled copper 0.323 pound, i cubic inch of rolled Muntz metal 0.296 pound, and i cubic inch of rolled brass 0.297 to 0.313 pound, accord- ing to its composition. A variation of 2^/2 per cent, under gauge will be allowed. Tolerance for Bxcess of Weights. — An excess of weight of 5 per cent, will be allowed on all sheets up to 60 inches wide. For all sheets above 60 inches wide a tolerance of 8 per cent, will be allowed. Specifications for Copper Pipes. The pipe must be made of best Lake copper, or its equivalent, and a chemical analysis must show that the metal is 99.5 per cent, pure copper. The naval inspector will take drillings for analyses. The pipe must be free from indentations, cracks, flaws, or other sur- face defects, inside and outside, perfectly round, of the specified diameter and thickness in all parts, except as provided in special cases. All straight sections of pipe 6 inches in diameter (inside) or less shall be seamless drawn. Each pipe must withstand an internal hydraulic pressure which will subject the metal to a stress of 6,000 pounds per square inch, the test pressure being calculated by the following formula for thin hollow cylin- 192 KNGINEDRING CHEMISTRY ders, but in no case will a test pressure of over 1,000 pounds per square inch per gauge be required: in which p := safe internal pressure ; d = inside diameter in inches ; s ^= safe tensile strength of material = 6,000 pounds per square inch; t =z thickness of pipe in inches. X /■ y^ 8"- A. Every pipe must be perfectly tight under pressure and show no signs of bulging cracks, flaws, porous places, or other defects. A strip i^ inches wide will be taken from each lot of 2,000 pounds or less of pipe and must stand the following tests : (a) If less than ^ inch thick, it must stand bending flat back cold after being annealed. (b) If yi inch or over, it must bend back after being annealed until the ends are parallel and the inner radius of the bend is equal to the thickness of the piece. (c) In every case the ends of the bending test pieces shall stand hammering down hot to a knife edge without showing signs of cracks. The pipe to stand flanging without defects. Pipes of 2 inches inside diameter and over, for steam or feed pipes or other such high pressure, are to be subject to tensile tests, one piece of pipe from each lot of 1,000 pounds or less being selected to represent the lot. If the pipes are from 2 to 6 inches inside diameter, they will be cut circumferentially. The test pieces will be heated to a cherry red and straightened when hot, then machined to the shape shown in the sketch, care being taken to have the brazed seam, if any, between the measuring points. For thickness up to and including ^ inch, the width of the narrow part of the test piece shall be about i>4 inches. For thicker pieces the width shall be such as to give a cross section of about half a square inch, but the breadth shall not in any case be less than the thickness. The rolled surfaces are not to be machined, but to be left in their original condition. The test piece must show an ultimate tensile strength after being ENGINEERING CHEMISTRY I93 annealed of at least 28,000 pounds per square inch for all pipe, and an elongation of at least 25 per cent, in 8 inches in the case of seamless pipe. Thickness. — Every pipe must be of the specified thickness in the thinnest part. Weight. — The weight of every pipe must be at least equal to the calculated weight, on the basis of i cubic inch of copper pipe weighs 0.323 pound. An excess of weight equal to 5 per cent, of the calculated weight will be allowed. Specifications for Phosphor Bronze. Used Principai^ly for Pump Rods and Vai.ve Springs Exposed. TO THE Action oe Sea Water. Rounds, whether cast, rolled, or forged, shall have an ultimate tensile strength and elongation of 50,000 pounds and 25 per cent, respectively. Note. — The test pieces are to be as nearly as possible of the same diameter as the rounds, or else they are to be not less than 14 inch in diameter and taken at a distance from the circumference equal to one-half the radius of the round. Phosphor bronze spring wire shall be hard and elastic. The naval inspector will take drillings for analyses, and these shall show not less than 85 per cent, copper, not less than 3 per cent, tin, and not less than o.oi per cent, phosphorus, the balance to be made up of whatever components the manufacturers consider best suited to produce a composition of the maximum strength, and incorrodible in sea water. Specifications for Rolled Bronze. (Plates, Shapes, and Rounds.) The inspection must conform to the "General instructions for the inspection of copper, brass, and bronze material coming under the cog- nizance of the Bureau of Steam Engineering." The naval inspector will take drillings for chemical analyses. The analyses must show that the alloy is composed of not less than 60 per cent, nor more than 63 per cent, copper; not less than J^ of i per cent, nor more than 1J/2 per cent, tin and the remainder zinc, with such small quantities of other ingredients as a manufacturer may think necessary in the case of proprietary materials. All bars are to be cleaned and straightened and must stand : (a) Being hammered hot to a fine point. (b) Being bent cold through an angle of 120° and to a radius equal to the diameter or thickness of the bars. If the metal is to be rolled into rods for bolts or other important 13 194 ENGINEERING CHEMISTRY parts subject to stress, one test piece for every lot of 400 pounds or less shall show the following results : Ultimate tensile strength per square inch Elastic limit Elongation per cent, in 2 inches Not less than 60,000 pounds. At least one-half ultimate tensile strength Not less than 25 per cent. In the case of large lots the number of test pieces to be left to the judgment of the naval inspector. Various composition materials, otherwise conforming to the specifica- tions but manufactured under proprietary processes or having proprietary names, will be accepted as rolled bronze. ~ Sheet Brass (Composition). Sixty to 70 per cent, pure copper. Forty to 30 per cent, pure zinc, and not more than five-tenths of i per cent. lead. MoNEi. MetaIv. Chemical composition of monel metal to be not less than 60 per cent, nickel and the balance copper, other elements in small percentages without being detrimental to its physical qualities. Navy mixture No. i, for rods, tubes, and similar material. Minimum tensile strength per square inch, 80,000 pounds ; minimum elastic limit, 50,000 pounds ; minimum elongation in 2 inches, 20 per cent., and to stand cold bend of 180°, about an inner diameter of i inch. Navy mixture No. 2, for castings and similar material. Minimum tensile strength per square inch, 60,000 pounds; minimum elastic limit, 35,000 pounds ; minimum elongation in 2 inches, 25 per cent., and to stand cold bend of 180°, about an inner diameter of i inch. Specifications for Pig Lead. 1. Pig lead will be required for either as No. i or No. 2. No. i grade is for foundry use, and No. 2 grade is for weights and ballast. 2. No. I pig lead to be of the best quality and practically free from all impurities. To show on analysis not less than 99.5 per cent, of metallic lead. 3. No. 2 pig lead to be of good commercial grade and to show on analysis not less than 97.5 per cent, of metallic lead. References. "Rapid Analysis of Alloys for Tin, Antimony and Arsenic," by F. A. Stief, Jour. hid. and Bng. Cheni., Mar., 191 5. ENGINEERING CHEMISTRY I95 "The Electrolytic Separation of Zinc, Copper and Iron from Arsenic," by A. K. Balls and C. C. McDonnell, Jour. Ind. and Bng. Chem., Jan., 1915. "Determination of Tin in Metal Foil of Pb, Sn and Sb," Chem. Abstracts, 1 914, p. 1464. "Self-lubricating Bearing Metals," Chem. Abstracts, 1914, p. 11 58. THE CHEMICAL AND PHYSICAL EXAMINATION OF PORTLAND AND NATURAL CEMENTS. The definition of Portland cement as adopted by the Verein Deutscher Portland Cement Fabrikanten is, that it is a product obtained by crushing after heating to the sintering point, a mix- ture of limestone, marl, chalk or hydraulic limestone with clay and is to be distinguished from "slag" cement (a form of Port- land) as the latter is formed by the following processes: (i) Granulation of the slag, (2) drying of the slag, (3) mixing with a suitable proportion of slaked lime, (4) grinding of the mix- ture. The Portland cement of the United States comprises those cements which are produced by the burning to the sin- tering point and grinding of artificial mixtures of limestone (or marl, chalk, or hydraulic limestone) and clay, or slag sand.^ After manufacture it is practically CagSiOg, and is quite dis- tinct from another product made and largely consumed here called "hydraulic cement, or natural cement." Experience has shown that Portland cements containing over 5 per cent, of magnesia (MgO) are inferior in lasting qualities, and by the gradual absorption of water produce cracking and disintegration. The "Ecole Nationale," of Paris, rejects all cements containing over 2.5 per cent, of sulphuric acid. Thus, if upon chemical analysis, magnesia is found present in amounts over 5 per cent., carbonic and sulphuric acids in amounts over 2.5 per cent., the cement can be condemned at once ivithout any mechanical tests. Therefore, it is evident that a careful test of a Portland cement requires : ( i ) a chemical analysis to determine the proportion ^J. Soc. Chem. Ind., 1901, p. 1212; "Historical Sketch of Slag Cement," by Pro- fessor William Kendrick, Scientific American Supplement, May 25, 1901. 196 ENGINEERING CHEMISTRY of the ingredients, and (2) the mechanical or physical tests to determine fineness, tensile strength, and resistance to crushing.^ The scheme on page 197 is arranged to show the method of making a cement analysis. In order to more fully explain the scheme for the analysis of Portland and natural cements, the following analysis of a Port- land cement is given : Grams Amount of cement taken 2.000 (i) Crucible + SiO. 11.205 Crucible 10.721 SiO, 0484 0.484 X 100 ^ c,r^ — ^ ^ = 24 20 per cent. SiO.^. Grams (2) Crucible + AI2O3 (in soluble residue) 10.743 Crucible 10.721 AI2O3 0.022 =1.10 per cent. AloOj in insoluble residue. Grams (3) Crucible + FcoO. 10.745 Crucible 10.721 Fe^Oa 0.024 0.024 X 2.50 X 100 4. T? r^ ^-^ ^^—^ z= 3.00 per cent. Fe^O^. Grams (4) Crucible + AW, 10.762 Crucible 10.721 AI2O3 0.041 0.041 X 2.5 X 100 ^ A1 ^ — -^—^ — ^^-^^ =5.12 per cent. AljO.,. 5.12 per cent. + i.io per cent, (from (i)) = 5.12 per cent. AI2O3. Grams (5) Crucible -\- CaO 12.2223 Crucible 11. 7210 CaO 0.5013 0-5013 X 2.50 X 100 . . ^^^ ^^„, p, p, = 62.67 V^^ cent. CaU. ^ "Portland Cement Manufacture," by Edwin C. Eckle, Municipal Engineertng, January, 1904. . o Kb . a -3 as s ^ g 5^ O £•= ^o 4^ ^ JC to E = E 2 ^>> N o bfl V CO is S^ 05 'S^g'H^.-:ls>'1 i-^"-^?-?t-'5:;2S<5' «3 J3kT M^ ^3 C CO ^ o 50 o -o CO in o V o ■E X .S *i (LI 10 O ■^ ts t? _ J5 o .5 if! 7 ^H o ^ 11 *"■ in O 2^'ns- Ui O — ^"O ^ g- o 00;^ ri O fj 01 "^ fl rt a w > •52 o'l-'sa . 0:0 u o o Q « 'f'^ 0.25 ' c p •« .2n > ^- 4; ■P y p o -ti _-0«S4i a c w -a - .2 i) n c fe i* k_r M 4. lT'o "CO c:i >> a "5 ^ -« --^ s o -aw ^ t-u- V 4^^ l«5-; -o^ "«ii- CO > *-> O Tj CO CO nrj j: 4> ^\: I ■* - tn o • lO 3 CO o. ^u ^.ij a.-:i «. a = CO 3 CO *j •2 TO^'<« ^ >>■ «i^co.i!rP Q ^ bfi. ,4^ M .5 o £ i3> H •M C ., ■«• ' CO *^ Q CO 1^ fi •- bfl XI CO « a A c 5 s 2 H 3 u o o - a 3 a. V -6.3 2i sis CO w __ "55 -^ "O 198 DNGINEKRING CHEMISTRY Grams (6) Platinum dish + MgSO* + NaoS04 33820 Platinum dish 33755 MgS04 + Na2S04 0.065 Determining the Magnesia as 1 Mg,PA. Crucible + Mg^P^O^ Crucible 10.7483 10.7210 MgS04 0.0294 Mg-AO; 0.0273 '1- 0.0273 gram MgaPaO^ corre- sponds to 0.0294 gram of MgSO^ — which is subtracted from 0.065 gram Na,SO, + MgSO,. J Na^SO, 0.0356 The 0.0294 gram MgS04 corresponds to 0.0098 gram of MgO. 0.0098 X 2.5 100 ,, ^ ^—^ — = 1.22 per cent. MgO. The 0.0356 gram NaaSO* corresponds to 0.0155 gram NaaO. 0.0,55x^2.5x100 _ ^^ p^^ ^^^^ j^^^Q ■ Grams Crucible + BaS04 10.729 Crucible 10.721 BaS04 0.008 SOs = 0.0027 gram 0.0027 X 5 X 100 0.67 per cent. SO3 re:sum]§;. SiO. AI2O3 Fe^Oa CaO MgO Na>0 1.96 SO. 0.67 Per cent. , 24.20 6.22 , 3.00 , 62.67 1.22 Total 99-94' ^ Consult Stevens Institute Indicator, "Determination of Alkalies in Portland Cement," by Thomas B. Stillman, October, 1901. ENGINEERING CHEMISTRY 199 The following well-known brands of cement were analyzed in my laboratory : SiOg • . • . AI2O, . . . • Fe,03.... CaO-... MgO . . . . K2O NagO . . . . SO3 CO2 . •• Total Burham's (per cent. 21.70 6.82 2.37 62.26 1.48 T.84 0.98 1.20 1.30 99-95 Dyckerhoff's (per cent.) 19-05 7.90 5-48 63.62 1.87 0.78 0.36 0.94 Saylor's (per cent.) 21.25 4.21 8.25 61.25 I -50 I. or 0.99 1.38 99.84 In some cements quartz is a constituent in amounts varying from 0.5 to 6 per cent. It can be separated from combined silica by the method of Fresenius.^ Where carbonic acid has been indicated by the qualitative analysis, the quantitative analysis, for this constituent, should be made upon at least 8 grams of the cement. The carbonic acid rarely reaches i per cent., and while it is generally absent in well-burned cements, it is by no means an uncommon constituent to the amount of 0.15-0.30 per cent., as the following table of analyses of German cements will show : CaO SiOg . • • Fe^O, ... A\,0,.... MgO . . . . Alkalies . SO3 CO2 Insoluble 61.99 23.69 2.71 8.29 0.47 0.95 0.69 0.27 0.44 62.89 22.80 3-40 7.70 1.20 1.30 0.71 63.71 25-37 3-14 4.31 1.25 0.84 0.87 63.27 19.80 3.22 6.73 2.02 1.48 1.08 0.23 1.38 65.59 22.85 2.76 5-51 1.24 0.92 1.69 59-96 23.70 3-'5 8.20 1. 00 1.05 0.88 0.20 0.80 64.51 22.38 2.24 9-45 1.44 60.81 22.63 2.42 7.06 2.89 2.83 0.47 033 Free Lime in Portland Cement. Free lime is sometimes found in Portland cement. Keiser and Forder give a process for determining it in Amer. Chem. Jour- nal, 31, 153. Consult also: "A Rapid Method for the Deter- ^ "Quantitative Chemical Analysis," p. 259. 200 V, N r. f N K f% K f X (> ( ' f f i: M f S'f k Y mination of JJme in Cement," by liernard Knright, Jour Amer. Chem. Soc, August, 1904, pp. jcxj3-ioo6. Slag cement can be distinguished from Portland cement by the method of Seger and Cramer (Chem. Zcitung, 2y, 879; which is as follows: Pass the material through a ifxj-mesh sieve and bc^il 50 grams with 100 cc, of water for 3 hours, kcej^ng uj> the volume of water by additions from time to time, and agitating occasionally to prevent formation of lumps. Filter, wash twice with hot water and dry at 110° to 120^ C, In i gram or more of the ma- terial determine loss on ignition {i. e., water of hydration). Portland showed 10.19 to 13.10 per cent, (average 11.46 per cent.) slag cement gave 0.70 to 0.84 (average 0.78 per cent.).' The amount soluble in water also differs. Shaking i to 1.5 grams with 3 liters of freshly boiled water, and collecting and weighing the residue after ignition, indicated that there was dissolved out from Portland cement (28.7 to 42.8) an average of 37.15 per cent., but from the slag cement 1.2 to 4.56 C average 2.33) per cent. 'the rarer constituents found in 3 samjjies of J^ehigh I'cHtland cement by Meade, "Portland Cement," p. 32, were as follows: No. I No. a No. 3 (p«r cent.) (per cent.) (per cent.) Titanic acid Ferrous oxitlc Manganous oxide tjtrontiuin oxide Calcium sulphide Potash Soda Phosphorus pentoxide • . 0.28 0.23 0,06 6.08 0,18 0.50 0.26 0.25 0.27 0.16 0,08 0.09 0.48 0.31 0.31 0.32 o. 1 1 0,09 a.07 0.59 0.38 0.29 Rapid Determination of Lime (CaO), Without Separation of Silica, Etc.,' (Volumetric). Weigii 0.5 gram of cement in a dry 5cx; cc. beaker and add, with constant stirring, 20 cc. of cold water. Break up the lumps and when all the sample is in suspension, except the heavier 1 R. Waller: The School of Mines Quarfrrlv, Aptil \u',a p mo. 8 Chemical Etigineer, 1, p. 21. ENGINEERING CHEMISTRY 201 particles, add 20 cc. of dilute (1:1) hydrochloric acid and heat until solution is complete. This usually takes from 5 to 6 min- utes. Heat to boiling, add dilute ammonia carefully to the solu- tion until a slight permanent precipitate forms. Heat to boiling, add 10 cc. of a 10 per cent, solution of oxalic acid. Stir until the oxides of iron and aluminum are entirely dissolved and only a slight precipitate of calcium oxalate remains. Add 200 cc. of boiling water and 20 cc. saturated solution of ammonium oxalate to completely precipitate the lime. Boil for a few minutes, re- move from the heat, allow the precipitate to settle and filter on a 1 1 -centimeter filter. Wash the precipitate and filter ten times with hot water using not more than lo cc. each time. Remove the filter from the funnel, open and lay against the sides of a beaker, wash from the paper into the beaker, with hot water, add dilute sulphuric acid, 5 cc, heat to 80° C. and titrate with standard permanganate until a pink color is obtained. (5.64 grams KMnO^ in i liter of water, i cc. corresponds to 0.005 gram CaO.) Standard Specifications for Cement. American Society for Testing Materials. Generai, Observations. 1. These remarks have been prepared with a view of pointing out the pertinent features of the various requirements and the I)recautions to be observed in the interpretation of the results of the tests. 2. The committee would suggest that the acceptance or re- jection under these specifications be based on tests made by an experienced person having the proper means for making the tests. Specific Gravity. 3. Specific gravity is useful in detecting adulteration. The results of tests of specific gravity are not necessarily conclusive as an indication of the quality of a cement, but when in com- bination with the results of other tests may afford valuable in- dications. 202 ENGINEERING CHEMISTRY g W a Q < o (X, w o OS o s ^ >'S o 13 2 % to en O 0'*-^»OTf *"S l-JcSI-lMMMI-HMdM oO u'Jl t^^' 00vOrO0'^f^OC«»-J'O r^ t^\£> vO t^ t^oo vo 00 r^ CO TtOO rt lo O CO r^ O O d U ID •= o s a SocT! g o ^ >> 4| a; cc . 2 o o bo ^^ O O UPQ 1 3 Booth, Garrett and Blair Manufacturer's analysis c Slo U'J) CJ ON rOOO »0 ro -^ OnvO vO lO lO -^ ►H CO M t^ C7\ t^v£) 00 O ^ M -' «■ M ci w d •-; d "-^ Crv TT « rtvo lo rOVO PJ t^ CTx . w 00 r^oo Tj- -I o\ ^ t^ CO CO d^^ t^vd r^ r^vd ds lo P) vT) -<■ ro d pi P- p O > O O CO O O u < w W u o ? o DC (/2 W PQ < engineering chemistry 203 Fineness. 4. The sieves should be kept thoroughly dry. Time of Setting. 5. Great care should be exercised to maintain the test pieces under as uniform conditions as possible. A sudden change or wide range of temperature in the room in which the tests are made, a very dry or humid atmosphere, and other irregularities vitally affect the rate of setting. Constancy of Voi^ume. 6. The tests for constancy of volume are divided into two classes, the first normal, the second accelerated. The latter should be regarded as a precautionary test only, and not infalli- ble. So many conditions enter into the making and interpreting of it that it should be used with extreme care. 7. In making the pats the greatest care should be exercised to avoid initial strains due to molding or too rapid drying-out during the first 24 hours. The pats should be preserved under the most uniform conditions possible, and rapid changes of tem- perature be avoided. 8. The failure to meet the requirements of the accelerated tests need not be sufficient cause for rejection. The cement may, however, be held for 28 days, and a retest made at the end of that period, using a new sample. Failure to meet the require- ments at this time should be considered sufficient cause for rejection, although in the present state of our knowledge it can- not be said that such failure necessarily indicates unsoundness, nor can the cement be considered entirely satisfactory simply because it passes the tests. Specifications. GENERAL CONDITIONS. 1. All cement shall be inspected. 2. Cement may be inspected either at the place of manufac- ture or on the work. 3. In order to allow ample time for inspecting and testing, 204 ENGINEERING CHEMISTRY the cement should be stored in a suitable weather-tight building having the floor properly blocked or raised from the ground. 4. The cement shall be stored in such a manner as to permit easy access for proper inspection and identification of each ship- ment. 5. Every facility shall be provided by the contractor and a period of at least 12 days allowed for the inspection and necessary tests. 6. Cement shall be delivered in suitable packages with the brand and name of manufacturer plainly marked thereon. 7. A bag of cement shall contain 94 pounds of cement net. Each barrel of Portland cement shall contain 4 bags, and each barrel of natural cement shall contain 3 bags of the above net weight. 8. Cement failing to meet the 7-day requirements may be held awaiting the results of the 28-day tests before rejection. 9. All tests shall be made in accordance with the methods proposed by the Committee on Uniform Tests of Cement of the American Society of Civil Engineers, presented to the Society January 21, 1903, and amended January 20, 1904, and January 15, 1908, with all subsequent amendments thereto. (See adden- dum to these specifications.) 10. The acceptance or rejection shall be based on the follow- ing requirements : NATURAL CEMENT.^ 11. Definition. This term shall be applied to the finely pul- verized product resulting from the calcination of an argillaceous limestone at a temperature only sufiicient to drive off the carbonic acid gas. 'Hoffmann" "Cummings" .... "Buffalo" "California" "Norton" James River, Virginia "Napanee" SiOa 27.30 26.69 24.30 24-34 27.98 25- 15 19.90 AloO, 7.14 7.2t 2.61 8.56 7.28 8.00 5.92 FeO 1.80 1.30 6.20 2.08 1.70 3-28 1. 14 CaO 35-98 43 12 39-45 61.62 37-59 49-53 46.75 MgO 18.0 1955 6.16 0.40 15.00 13.7S 16.00 K..O Na«0 6.S0 I-I3 5-30 2.00 7.96 802 COoHoO. and ios.s 1. 00 13.23 0.80 2-49 0.26 2.27 A few analyses here given will indicate the variation of composition. e:ngine:ering chemistry 205 Fineness. 12. It shall leave by weight a residue of not more than 10 per cent, on the No. 100, and 30 per cent, on the No. 200 sieve. Time of Setting. 13. It shall not develop initial set in less than 10 minutes; and shall not develop hard set in less than 30 minutes, or in more than 3 hours. TENS11.E Strength. 14. The minimum requirements for tensile strength for bri- quettes I square inch in cross section shall be as follows, and the cement shall show no retrogression in strength within the periods specified : Neat Cement. Age strength 24 hours in moist air 75 lbs. 7 days (i day in moist air, 6 days in water) 150 lbs. 28 days (i day in moist air, 27 days in water) 250 lbs. I Part Cement, 3 Parts Standard Ottawa Sand. 7 days (i day in moist air, 6 days in water) 50 lbs. 28 days (i day in moist air, 27 days in water) 125 lbs. Constancy of Voi^ume. 15. Pats of neat cement about 3 inches in diameter, ^ inch thick at center, tapering to a thin edge, shall be kept in moist air for a period of 24 hours. (a) A pat is then kept in air at normal temperature. (b) Another is kept in water maintained as near 70° F. as practicable. 16. These pats are observed at intervals for at least 28 days, and, to satisfactorily pass the tests, shall remain firm and hard and show no signs of distortion, checking, cracking, or disinte- grating. PORTLAND CEMENT. 17. Definition. This term is applied to the finely pulverized product resulting from the calcination to incipient fusion of an intimate mixture of properly proportioned argillaceous and cal- careous materials, and to which no addition greater than 3 per cent, has been made subsequent to calcination. 2o6 e:ngink^ring chkmistry Spe)cific Gravity. i8. The specific gravity of cement shall not be less than 3.10. Should the test of cement as received fall below this requirement, a second test may be made upon a sample ignited at a low red heat. The loss in weight of the ignited cement shall not exceed 4 per cent. Finkne:ss. 19. It shall leave by weight a residue of not more than 8 per cent, on the No. 100, and not more than 25 per cent, on the No. 200 sieve. TiMK OF SivTTlNG. 20. It shall not develop initial set in less than 30 minutes ; and must develop hard set in not less than i hour, nor more than 10 hours. Te:nsii.e: Strength. 21. The minimum requirements for tensile strength for bri- quettes I square inch in cross section shall be as follows, and the cement shall show no retrogression in strength within the periods specified : Neat Cement. Age Strength 24 hours in moist air 1 75 lbs. 7 days (i day in moist air, 6 days in water) 500 lbs. 28 days (1 day in moist air, 2"] days in water) 600 lbs. I Part Cement, 3 Parts Standard Ottawa Sand. 7 days (i day in moist air, 6 days in water) 200 lbs. 28 days (i day in moist air, 27 days in water) 275 lbs. Constancy of Volume. 22. Pats of neat cement about 3 inches in diameter, ^ inch thick at the center, and tapering to a thin edge, shall be kept in moist air for a period of 24 hours. (a) A pat is then kept in air at normal temperature and ob- served at intervals for at least 28 days. (&) Another pat is kept in water maintained as near 70° F. as practicable, and observed at intervals for at least 28 days. ENGINE^ERING CHEMISTRY 20/ (c) A third pat is exposed in any convenient way in an atmos- phere of steam, above boiUng water, in a loosely closed vessel for 5 hours. 23. These pats, to satisfactorily pass the requirements, shall remain firm and hard, and show no signs of distortion, checking, cracking, or disintegrating. Sulphuric Acid and Magnesia. 24. The cement shall not contain more than 1.75 per cent, of anhydrous sulphuric acid (SO3), nor more than 4 per cent, of magnesia (MgO). Methods for Testing Cement.^ Recommended by the Special Committee on Uniform Tests of Cement of the American Society of Civil Engineers. Sampung. 1. Selection of Sample. — The selection of samples for testing should be left to the engineer. The number of packages sampled and the quantity taken from each package will depend on the importance of the work and the facilities for making the tests. 2. The samples should fairly represent the material. When the amount to be tested is small it is recommended that i barrel in 10 be sampled ; when the amount is large it may be impractic- able to take samples from more than i barrel in 30 or 50. When the samples are taken from bins at the mill for each 50 to 200 barrels will suffice. 3. Samples should be passed through a sieve having 20 meshes per linear inch, in order to break up lumps and remove foreign material; the use of this .sieve is also effective to obtain a thorough mixing of the samples when this is desired. To de- termine the acceptance or rejection of cement it is preferable, when time permits, to test the samples separately. Tests to de- termine the general characteristics of a cement, extending over a long period may be made with mixed samples. * Accompanying Final Report of Special Committee on Uniform Tests of Cement of the American Society of Civil Engineers, dated January 17, 19 12. 2o8 ENGINEERING CHEMISTRY 4. Method of Sampling. — Cement in barrels should be sampled through a hole made in the head, or in one of the staves midway between the heads, by means of an auger or a sampling iron similar to that used by sugar inspectors; if in bags, the sample should be taken from surface to center; cement in bins should be sampled in such a manner to represent fairly the contents of the bin. Sampling from bins is not recommended if the method of manufacture is such that ingredients of any kind are added to the cement subsequently. Chemical Anai^ysis. 5. Significance. — Chemical analysis may serve to detect adul- teration of cement with inert material, such as slag or ground limestone, if in considerable amount. It is useful in determining whether certain constituents, such as magnesia and sulphuric anhydride, are present in inadmissible proportions. 6. The determination of the principal constituents of cement silica, alumina, iron oxide, and lime, is not conclusive as an in- dication of quality. Faulty cement results more frequently from imperfect preparation of the raw material or defective burning than from incorrect proportions. Cement made from material ground very finely and thoroughly burned may contain much more lime than the amount usually present, and still be per- fectly sound. On the other hand, cements low in lime may, on account of careless preparation of the raw material, be of dan- gerous character. Furthermore, the composition of the product may be so greatly modified by the ash of the fuel used in burn- ing as to affect in a great degree the significance of the results of analysis. 7. Methods. — The methods to be followed, except for deter- mining the loss on ignition, should be those proposed by the Committee on Uniformity in the Analysis of Materials for the Portland Cement Industry, reported in the Journal of the Society for Chemical Industry, Vol. 21, page 12, 1902; and published in Engineering News, Vol. 50, p. 60, 1903; and in Engineering Record, Vol. 48, p. 49, 1903, and in addition thereto, the follow- ing: ENGINEERING CHEMISTRY 209 (a) The insoluble residue may be determined as follows : To a i-gram sample of the cement are added 30 cc. of water and 10 cc. of concentrated hydrochloric acid, and then warmed until effer- vescence ceases, and digested on a steam bath until dissolved. The residue is filtered, washed with hot water, and the filter paper and contents digested on the steam bath in a 5 per cent, solution of sodium carbonate. This residue is filtered, washed with hot water, then with hot hydrochloric acid, and finally with hot water, and then ignited at a red heat and weighed. The quantity so obtained is the insoluble residue. (b) The loss on ignition shall be determined in the following manner: One-half gram of cement is heated in a weighed plat- inum crucible, with cover, for 5 minutes with a Bunsen burner (starting with a low flame and gradually increasing to its full height) and then heated for 15 minutes with a blast lamp; the difference between the weight after cooling and the original weight is the loss on ignition. The temperature should not ex- ceed 900° C, or a low red heat; the ignition should preferably be made in a muffle. Specific Gravity. 8. Significance. — The specific gravity of cement is lowered by adulteration and hydration, but the adulteration must be con- siderable to be detected by tests of specific gravity. 9. Inasmuch as the differences in specific gravity are usually very small, great care must be exercised in making the determin- ation. 10. Apparatus. — The determination of specific gravity should be made with a standardized Le Chatelier apparatus. This con- sists of a flask (D), Fig. 22, of about 120 cc. capacity, the neck of which is about 20 centimeters long; in the middle of this neck is a bulb (C), above and below which are two marks (F) and (H) ; the volume between these two marks is 20 cc. The neck has a diameter of about 9 millimeters, and is graduated into tenths of cubic centimeters above the mark (P). 11. Benzene (62° Baume naphtha) or kerosene free from water should be used in making the determination. 14 2IO Kngine:kring ch:e:mistry 12. Method. — The flask is filled with either of these liquids to the lower mark (B), and 64 grams of cement, cooled to the tem- perature of the liquid, is slowly introduced through the funnel {B), (the stem of which should be long enough to extend into the flask to the top of the bulb (C), taking care that the cement Fig. 22. — Le Chatelier's specific gravity apparatus. does not adhere to the sides of the flask, and that the funnel does not touch the liquid. After all the cement is introduced, the level of the liquid will rise to some division of the graduated neck; this reading, plus 20 cc, is the volume displaced by 64 grams of the cement. e;ngine:e;ring chemistry 211 13. The specific gravity is then obtained from the formula, weight of cement, in grams, Specific gravity displaced vOiUme in cubic centimeter 14. The flask, during the operation, is kept immersed in water in a jar (A), in order to avoid variations in the temperature of the liquid in the flask, which should not exceed ^° C. The results of repeated tests should agree within o.oi. The deter- mination of specific gravity should be made on the cement as received; if it should fall below 3.10, a second determination should be made after igniting the sample in a covered dish, preferably of platinum, at a low red heat not exceeding 900° C. The sample should be heated for 5 minutes with a Bunsen burner (starting with a low flame and gradually increasing to its full height) and then heated for 15 minutes with a blast lamp; the ignition should preferably be made in a muffle-. 15. The apparatus may be cleaned in the following manner: The fiask is inverted and shaken vertically until the liquid flows freely, and then held in a vertical position until empty ; any traces of cement remaining can be removed by pouring into the flask a small quantity of clean liquid benzene or kerosene and repeating the operation. Fineness. 16. Significance. — It is generally accepted that the coarser par- ticles in cement are practically inert, and it is only the extremely fine powder that possesses cementing qualities. The more finely cement is pulverized, other conditions being the same, the more sand it will carry and produce a mortar of a given strength. 17. Apparatus. — The fineness of a sample of cement is deter- mined by weighing the residue retained on certain sieves. Those known as No. 100 and No. 200, having approximately 100 and 200 wires per linear inch, respectively, should be used. They should be 8 inches in diameter. The frame should be of brass, 8 inches in diameter, and the sieve of brass wire cloth conform- ing to the following requirements : 212 ENGINE^^RIXG CHEMISTRY No. of sieve Diameter of wire (inches) Meshes, per linear inch Warp Woof lOO 200 0.0042 to 0.0048 0.0021 to 0.0023 95 to lOI 192 to 203 92 to 103 1 90 to 205 The meshes in any smaller space, down to 0.25 inch, should be proportional in number. 18. Method. — The test should be made with 50 grams of cement, dried at a temperature of 100° C. (212° F.). 19. The cement is placed on the No. 200 sieve, which, with pan and cover attached, is held in one hand in a slightly inclined position, and moved forward and backward about 200 times per minute, at the same time striking the side gently, on the up stroke, against the palm of the other hand. The operation is continued until not more than 0.05 gram will pass through in I minute. The residue is weighed, then placed on the No. icx) sieve, and the operation repeated. The work may be expedited by placing in the sieve a few large steel shot, which should be removed before the final i minute of sieving. The sieves should be thoroughly dry and clean. NORMAI, C0NSISTE:nCY. 20. Significance. — The use of a proper percentage of water in making pastes^ and mortars for the various tests is exceedingly important and affects vitally the results obtained. 21. The amount of water, expressed in percentage by weight of the dry cement, required to produce a paste of plasticity desired, termed "normal consistency," should be determined with the Vicat apparatus in the following manner : 22. Apparatus. — This consists of a frame (A), Fig. 23, bear- ing a movable rod {B), weighing 300 grams, one end (C) being I centimeter in diameter for a distance of 6 centimeters, the other having a removable needle (Z)), i millimeter in diameter, 6 millimeters long. The rod is reversible, and can be held in any ^ The term "paste" is used in this report to designate a mixture of cement and water, and the word "mortar" to designate a mixture of cement, sand and water. KNGINKKRING CHE:m1STRY 213 desired position by a screw (B), and has midway between the ends a mark (F) which moves under a scale (graduated to milh- meters) attached to the frame (A). The paste is held in a conical, hard-rubber ring (G), y centimeters in diameter at the base, 4 centimeters high, resting on a glass plate (H) about 10 centimeters square. \B 9 9 F ^E Fig. 23. — Vicat apparatus. 23. Method. — In making the determination, the same quantity of cement as will be used subsequently for each batch in making the test pieces, but not less than 500 grams, with a measured quantity of water, is kneaded into a paste, as described in para- graph 45, and quickly formed into a ball with the hands, com- 214 ^ENGINEERING CHEMISTRY pleting the operation by tossing it six times from one hand to the other, maintained about 6 inches apart; the ball resting in the palm of one hand is pressed into the larger end of the rubber ring held in the other hand, completely filling the ring with paste ; the excess at the larger end is then removed by a single move- ment of the palm of the hand; the ring is then placed on its larger end on a glass plate and the excess paste at the smaller end is sliced off at the top of the ring by a single oblique stroke of a trowel held at a slight angle with the top of the ring. Dur- ing these operations care must be taken not to compress the paste. The paste confined in the ring, resting on the plate, is placed under the rod, the larger end of which is brought in con- tact with the surface of the paste ; the scale is then read, and the rod quickly released. 24. The paste is of normal consistency when the cylinder settles to a point 10 millimeters below the original surface in Yz minute after being released. The apparatus must be free from all vibrations during the test. 25. Trial pastes are made with varying percentages of water until the normal consistency is obtained. 26. Having determined the percentage of water required to produce a paste of normal consistency, the percentage required for a mortar containing, by weight, i part of cement to 3 parts of standard Ottawa sand, is obtained from the following table, the amount being a percentage of the combined weight of the cement and sand. Percentage of Water for Standard Mortars. One cement. One cement. I One cement. Neat three standard Neat three standard i Neat three standard Ottawa sand Ottawa sand 1 Ottawa sand 15 8.0 i ^3 9-3 31 10.7 16 8.2 24 9-5 32 10.8 17 8.3 25 9-7 33 II. 18 8.5 26 9.8 34 II. 2 19 8.7 27 10. 35 11-3 20 8.8 28 10.2 36 11-5 21 9.0 29 10.3 37 II. 7 22 9.2 30 10.5 38 II. 8 engineering chemistry 21$ Time oe Setting. 2y. Significance. — The object of this test is to determine the time which elapses from the moment water is added until the paste ceases to be plastic (called the "initial set"), and also the time until it acquires a certain degree of hardness (called the "final set" or "hard set"). The former is the more important, since, with the commencement of setting, the process of crys- tallization begins. As a disturbance of this process may produce a loss of strength, it is desirable to complete the operation of mixing or molding or incorporating the mortar into the work before the cement begins to set. 28. Apparatus. — The initial and final set should be determined with the Vicat apparatus described in paragraph 22. 29. Method. — A paste of normal consistency is molded in the hard-rubber ring, as described in paragraph 23, and placed under the rod (B), the smaller end of which is then carefully brought in contact with the surface of the paste, and the rod quickly released. 30. The initial set is said to have occurred when the needle ceases to pass a point 5 millimeters above the glass plate ; and the final set, when the needle does not sink visibly into the paste. 31. The test pieces should be kept in moist air during the test; this may be accomplished by placing them on a rack over water contained in a pan and covered by a damp cloth; the cloth to be kept from contact with them by means of a wire screen; or they may be stored in a moist box or closet. 32. Care should be taken to keep the needle clean, as the col- lection of cement on the sides of the needle retards the penetra- tion, while cement on the point may increase the penetration. 33. The time of setting is affected not only by the percentage and temperature of the water used and the amount of kneading the paste receives, but by the temperature and humidity of the air, and its determination is, therefore, only approximate. Standard Sand. 34. The sand to be used should be natural sand from Ottawa, 111., screened to pass a No. 20 sieve, and retained on a No. 30 2l6 KNGINEERING CHEMISTRY sieve. The sieves should be at least 8 inches in diameter; the wire cloth should be of brass wire and should conform to the following requirements : No of sieve Diameter of wire (inches) Meshes, per I,inear Inch Warp Woof 20 30 0.016 to o.ory o.oii to 0.012 19.5 to 20.5 29-5 to 30.5 19.0 to 21.0 28.5 to 31.5 Fig. 24.. — Details for Briquette. ENGINEERING CHEMISTRY 217 Sand which has passed the No. 20 sieve is standard when not more than 5 grams pass the No. 30 sieve in i minute of contin- uous sifting of a 500-gram sample.- Form of Test Pieces. 35. For tensile tests the form of test piece shown in Fig. 24 should be used. 36. For compressive tests, 2-inch cubes should be used. Molds. 37. The molds should be of brass, bronze, or other non-cor- rodible material, and should have sufficient metal in the sides to prevent spreading during molding. 38. Molds may be either single or gang molds. The latter are preferred by many. If used, the types shown in Figs. 25 and 26 are recommended. / Fig. 25. — Details for gang mold. Fig. 26. — Mold for compression test pieces. 39. The molds should be wiped with an oily cloth before using. Mixing. 40. The proportions of sand and cement should be stated by weight; the quantity of water should be stated as a percentage by weight of the dry material. 41. The metric system is recommended because of the con- venient relation of the gram and the cubic centimeter. ^ This sand may now (1912) be obtained from the Ottawa Silica Co., at a cost of two cents per pound, f. o. b. cars, Ottawa, 111. 2i8 e:nginee)ring che:mistry 42. The temperature of the room and of the mixing water should be maintained as nearly as practicable at 21° C. (70° F.). 43. The quantity of material to be mixed at one time depends on the number of test pieces to be made; 1,000 grams is a con- venient quantity to mix by hand methods. 44. The Committee has investigated the various mechanical mixing machines thus far devised, but cannot recommend any of them for the following reasons: (i) the tendency of most cement is to "ball up" in the machine, thereby preventing work- ing it into a homogeneous paste; (2) there are no means of ascertaining when the mixing is complete without stopping the machine; and (3) it is difficult to keep the machine clean. 45. Method. — The material is weighed, placed on a non-absorb- ent surface (preferably plate glass), thoroughly mixed dry if sand be used, and a crater formed in the center, into which the proper percentage of clean water is poured; the material on the outer edge is turned into the center by aid of a trowel. As soon as the water has been absorbed, which should not require more than i minute the operation is completed by vigorously kneading with the hands for i minute. During the operation the hands should be protected by rubber gloves. M01.DING. 46. The Committee has not been able to secure satisfactory results with existing molding machines ; the operation of machine molding is very slow; and is not practicable with pastes or mor- tars containing as large percentages of water as herein recom- mended. 47. Method. — Immediately after mixing, the paste or mortar is placed in the molds with the hands, pressed in firmly with the fingers, and smoothed off with a trowel without ramming. The material should be heaped above the mold, and, in smoothing off, the trowel should be drawn over the mold in such a manner as to exert a moderate pressure on the material. The mold should then be turned over and the operation of heaping and smoothing off repeated. 48. A check on the uniformity of mixing and molding may be ENGINEERING CHEMISTRY 219 afforded by weighing the test pieces on removal from the moist closet; test pieces from any sample which vary in weight more than 3 per cent, from the average should not be considered. Storage oi^ the Test Pieces. 49. During the first 24 hours after molding, the test pieces should be kept in moist air to prevent drying. 50. Two methods are in common use to prevent drying: (i) covering the test pieces with a damp cloth, and (2) placing them in a moist closet. The use of the damp cloth, as usually carried out, is objectionable, because the cloth may dry out unequally and in consequence the test pieces will not all be subjected to the same degree of moisture. This defect may be remedied to some ex- tent by immersing the edges of the cloth in water; contact be- tween the cloth and the test pieces should be prevented by means of a wire screen, or some similar arrangement. A moist closet is so much more effective in securing uniformly moist air, and is so easily devised and so inexpensive, that the use of the damp cloth should be abandoned. 51. A moist closet consists of a soapstone or slate box, or a wooden box lined with metal, the interior surface being covered with felt or broad wicking kept wet, the bottom of the box being kept covered with water. The interior of the box is provided with glass shelves on which to place the test pieces, the shelves being so arranged that they may be withdrawn readily. 52. After 24 hours in moist air, the pieces to be tested after longer periods should be immersed in water in storage tanks or pans made of non-corrodible material. 53. The air and water in the moist closet and the water in the storage tanks should be maintained as nearly as practicable at 21° C. (70° F.). TensiIvE Strength. 54. The tests may be made with any standard machine. 55. The clip is shown in Fig. 27. It must be made accurately, the pins and rollers turned, and the rollers bored slightly larger than the pins so as to turn easily. There should be a slight clear- 220 KNGINKERING CHEMISTRY ance at each end of the roller, and the pins should be kept prop- erly lubricated and free from grit. The clips should be used without cushioning at the points of contact. W{ SECTION A-B Roller turned and accurately bored to easy turning fit Fig. 27. — Form of Clip. 56. Test pieces should be broken as soon as they are removed from the water. Care should be observed in centering the 'test pieces in the testing machine, as cross strains, produced by im- perfect centering, tend to lower the breaking strength. The load should not be applied too suddenly, as it may produce vibration, the shock from which often causes the test pieces to break before the ultimate strength is reached. The bearing surfaces of the e;nginee;ring chemistry 221 clips and test pieces must be kept free from grains of sand or dirt, which would prevent a good bearing. The load should be applied at the rate of 600 pounds per minute The average of the results of test pieces from each sample should be taken as the test of the sample. Test pieces which do not break within 34 inch of the center, or are otherwise manifestly faulty, should be excluded in determining average results. Fig. 28. — Riehle U. S. Standard 1,000-pound automatic cement heater. Description and Operation. It is composed entirely of metal. The beam is brought to a 222 e;nginee:ring chemistry balance by pouring shot into the cone-shaped bucket on the left of the machine, thus counterbalancing the weight on the right- hand side of the machine. The test briquette is then placed in the grips and by means of the handwheel under the lower grip, the slack is taken up. A piston valve (Patented Nov. 8, 1904) in the bucket is then lifted by throwing the latch over and the shot flows out of the bucket causing the weight to overbalance the bucket and load thus to be applied to the specimen. When a sufficient weight of shot has flowed out of the bucket, the un- balanced force of the weight is sufficient to break the briquette, and then the lightened bucket is moved upward by the weight and the piston valve in it closed, causing the flow of shot to cease. To change the speed of the test the flow of shot can be regulated by means of the knurled screw at top of the piston valve. The weight of shot which has flowed out is a measure of the force required to break the briquette, and this shot is caught in a scoop on a scale which is graduated to read directly the stress on the briquette. If for any reason the main beam should touch the buffer before the specimen of cement is broken, the valve automatically closes and the flow of shot ceases. The operator then raises the beam by means of the crank through the worm and worm gear, and the test continues. If it is desired to make a test with the beam in a horizontal position, it can be kept level by means of the crank and worm wheel. In place of the spring balance, any form of scale may be used. Dimensions. Extreme length 30 in. Extreme width 15 in. Extreme height 2 ft. 4 in. Weight 115 lbs. Shipping weight 1 50 lbs. Shipping measurements 10 cu. ft. Compressive: Strength. 57. The tests may be made with any machine provided w4th Engine:rring chemistry 223 means for so applying the load that the line of pressure is along the axis of the test piece. A ball-bearing block for this purpose is shown in Fig. 29. Some appliance should be provided to facilitate placing the axis of the test piece exactly in line with the center of the ball-bearing. Fig. 29. — Ball-bearing block for testing machine'. 58. The test piece should be placed in the testing machine, with a piece of heavy blotting paper on each of the crushing faces, which should be those that were in contact with the mold. Constancy of Voi^umf. 59. Significance. — The object is to detect those qualities which tend to destroy the strength and durability of a cement. Under normal conditions these defects will in some cases develop 224 ENGINEERING CHEMISTRY quickly, and in other cases may not develop for a considerable time. Since the detection of these destructive qualities before using the cement in construction is essential, tests are made not only under normal conditions but under artificial conditions created to hasten the development of these defects. Tests may, therefore, be divided into two classes : ( i ) Normal tests, made in either air or water maintained, as nearly as practicable, at 21° C. (70° F.) ; and (2) Accelerated tests, made in air, steam or water, at temperature of 45° C. (113° F.) and upward. The committee recommends that these tests be made in the following riianner : 60. Methods. — Pats, about 3 inches in diameter, }^ inch thick at the center, and tapering to a thin edge, should be made on clean glass plates (about 4 inches square) from cement paste of normal consistency, and stored in a moist closet for 24 hours. 61. Normal Tests. — After 24 hours in the moist closet, a pat is immersed in water for 28 days and observed at intervals. A similar pat, after 24 hours in the moist closet, is exposed to the air for 28 days or more and observed at intervals. 62. Accelerated Test. — After 24 hours in the moist closet, a pat is placed in an atmosphere of steam, upon a wire screen i inch above boiling water, for 5 hours. The apparatus should be so constructed that the steam will escape freely and atmospheric pressure be maintained. Since the type of apparatus used has a great influence on the results, the arrangement shown in Fig. 30 is recommended. 63. Pats which remain firm and hard and show no signs of cracking, distortion, or disintegration are said to be "of constant volume" or "sound." 64. Should the pat leave the plate, distortion may be detected best with a straight-edge applied to the surface which was in contact with the plate. 65. In the present state of our knowledge it cannot be said that a cement which fails to pass the accelerated test will prove de- fective in the work ; nor can a cement be considered entirely safe simply because it has passed these tests. Engine;e)ring chemistry 225 15 226 ENGINEERING CHEMISTRY Interpretation of Results of Tests. ^ Chemicae. The composition of normal Portland cement has been the sub- ject of a great deal of investigation and it can be said that the quantities of silica, alumina, oxide of iron, lime, magnesia, and sulphuric anhydride can vary within fairly wide limits without materially affecting the quality of the material. A normal American Portland cement which meets the standard specifications for soundness, setting time and tensile strength has an approximate composition within the following limits : Per cent. Silica 19-25 Alumina 5-9 Iron oxide 2-4 Lime 60-64 Magnesia 1-4 Sulphur trioxide I-I-75 Loss on ignition 0.5-3.00 Insoluble residue o. i-i .00 It is also true that a number of cements have been made both here and abroad which have passed all standard physical tests in which these limits have been exceeded in one or more par- ticulars, and it is equally true that a sound and satisfactory cement does not necessarily resiilt from the above composition. It is probable that further investigation will give a clearer understanding of the constitution of Portland cement, but at present chemical analysis furnishes but little indication of the quality of the material. Defective cement usually results from imperfect manufacture, not from faulty composition. Cement made from very finely ground material, thoroughly mixed and properly burned, may be perfectly sound when containing more than the usual quantity of lime, while a cement low in lime may be entirely unsound due to careless manufacture. The analysis of a cement will show the uniformity in compo- sition of the product from individual mills, but will furnish little ^ United States Government Specifications for Portland Cement, Circular No. 33, Bureau of Standards, May i, 19 12. ENGINEERING CHEMISTRY 227 or no indication of the quality of the material. Occasional analy- sis should, however, be made for record and to determine the quantity of sulphuric anhydride and magnesia present. The ground clinker as it comes from the mill is usually quick setting which requires correction. This is usually accomplished by the addition of a small quantity of more or less hydrated calcium sulphate, either gypsum or plaster of Paris. Experience and practice have shown that an addition of 3 per cent, or less is sufficient for the purpose. Three per cent, of calcium sulphate (CaS04) contains about 1.75 per cent, sulphuric anhydride (SO3), and as this has been considered the maximum quantity necessary to control time of set, the specification limits the SO3 content to 1.75 per cent. The specification prohibits the addition of any material sub- sequent to calcination except the 3 per cent, of calcium sulphate permitted to regulate time of set. Other additions may be diffi- cult or impossible to detect even by a careful mill inspection during the process of manufacture^ but as the normal adulterant would be ground raw material, an excess of "insoluble residue" would reveal the addition of silicious material, and an excess in "loss on ignition" would point to the addition of calcareous material when either is added in sufficient quantity to make the adulteration profitable. The effect of relatively small quantities of magnesia (MgO) in normal Portland cement, while still vmder investigation, can be considered harmless. Earlier investigators believed that as magnesia had a slower rate of hydration than lime, the hydration of any free magnesia (MgO) present would occur after the cement had set and cause disintegration. The effect of magnesia was considered especially injurious when the cement was exposed to the action of sea water. More recent investigation has shown that cement can be made which is perfectly sound under all conditions when containing 5 per cent, of magnesia and it has also been found that the lime in Portland cement exposed to sea water is replaced by magnesia. The maximum limit for magnesia has been set at 4 per cent., 228 E;NGINE;e:rING CHi^MISTRY as it has been established that this quantity is not injurious and it is high enough to permit the use of the large quantities of raw material available in most sections of the country. Physicai.. Specific Gravity. — If the Le Chatelier apparatus is used for the determination of specific gravity, the clean volumenometer flask is filled with benzene free from water (which can be obtained by placing some calcium chloride or caustic lime in the benzene storage jar) to a point on the stem between zero and i cc. The flask is then placed in a constant temperature bath until volume is constant. The usual method is to introduce 64 grams of cement into the flask, taking care that the powder does not adhere to the tube above the liquid, and to free the cement from air by rolling the flask in an inclined position. The flask is then replaced in the constant temperature bath until a constant volume is recorded. The specific gravity is obtained from the formula : weight of cement in grams specific gravity -p — ; -^ — ^ ■. —r^ . displaced volume in cubic centimeter The specific gravity of a Portland cement is not an indication of its cementing value. It will vary with the constituents of the cement, especially with the content of iron oxide. Thus the white or very light Portland cements, containing a fraction of a per cent, of iron oxide, usually have a comparatively low specific gravity ranging from 3.05 to 3.15, while a cement containing 3 to 4 per cent, or more of iron oxide may have a specific gravity of 3.20 or even higher. It is materially affected by the temper- ature and duration of burning the cement, the hard-burned cement having the higher specific gravity. A comparatively low specific gravity does not necessarily indicate that a cement is underburned or adulterated, as large percentages of raw materials could be added to a cement with a normally high specific gravity before the gravity would be reduced below 3.10. If a Portland cement fresh from the mill normally has a com- paratively low specific gravity, upon aging it may absorb suf- ENGINEERING CHEMISTRY 229 ficient moisture and carbon dioxide to reduce the gravity below 3.10. It has been found that this does not appreciably affect the cementing value of the material; in fact, many cements are un- sound until th€y have been aged. Thus a redetermination is permitted upon a sample heated to a temperature sufficient to drive off any moisture which may be absorbed by the cement subsequent to manufacturing, but would not drive off any car- bon dioxide nor correct underburning in the process of manu- facturing the cement. The value of the specific gravity determination lies in the fact that it is easily made in the field or laboratory, and when the nor- mal specific gravity of the cement is known, any considerable variation in quality due to underburning or the addition of for- eign materials may be detected. Fineness. — Only the extremely fine powder of cement called flour possesses appreciable cementing qualities and the coarser particles are practically inert. No sieve is fine enough to deter- mine the flour in a cement, nor is there any other means of ac- curately and practically measuring the flour. Some cements grind easier than others, thus, although a larger percentage of one cement may pass the 200-mesh sieve than another, the for- mer may have a smaller percentage of actual flour due to the dif- ference in the hardness and the character of the clinker, and the method used in grinding. Thus the cementing value of different cements can not be compared directly upon their apparent fine- ness through a 200-mesh sieve. With cement from the same mill, with similar clinker and grinding machinery, however, it is probable that the greater the percentage which passes the 200- mesh sieve the greater the percentage of flour in that particular cement. Normal Consistency. — The quantity of water used in making the paste from which the pats for soundness, tests of setting, and the briquettes are made, is very important and may vitally affect the results obtained. The determination consists in measuring the quantity of water required to bring a cement, to a certain state of plasticity. 230 ENGINEERING CHEMISTRY In determining the normal consistency by the ball method, after mixing the paste it should be formed into a ball with as little working as possible and a new batch of cement should be mixed for each trial paste. In order to obtain just the requisite quantity of paste to form a ball 2 inches in diameter, a measure made from a pipe with a 2-inch inside diameter cut i^ inches 23.5 5^ ONE PER CENT. ABOVE NORMAL. 24.5$^ TWO PER CENT. ABOVE NORMAL Fig. 31. — Appearance of ball for different consistencies of cement paste. long will be found convenient. The section of pipe should be open at both ends, so that it can be pushed down onto the paste on the mixing table and the excess paste cut off with the trowel. The appearance of the ball using the correct percentage of water for normal consistency as compared with a less and greater quantity of water is shown in Fig. 31. ENGINEERING CHEMISTRY 23I Mixing. — The homogeneity of the cement paste is dependent upon the thoroughness of the mixing, and this may have con- siderable influence upon the time of setting and the strength of the briquettes. Soundness. — The purpose of this test is to detect those qual- ities in a cement which tend to destroy the strength and dura- bility. Unsoundness is usually manifested by a change in vol- ume w^hich causes cracking, swelling, or disintegration. If the pat is not properly made, or if it is placed where it will be subject to any drying during the first 24 hours, it may develop what are known as shrinkage cracks, which are not an indication of 32. — Soundness pat showing Fig. 33. — Soundness pat showing shrinkage cracks. disintegration cracks. Fig. 34. — Soundness pat with top surface flattened for determining time of setting. unsoundness and should not be confused with disintegration cracks, as shown in Figs. 32 and 33. No shrinkage cracks should develop after the first 24 or 48 hours. The failure of the pats to remain on the glass nor the cracking of the glass to which the pat is attached does not necessarily indicate unsoundness. In molding the pats, the cement paste should first be flattened on the glass and the pat formed by drawing the trowel from the outer edge toward the center, as shown in Fig. 35. Time of Setting. — The purpose of this test is to determine the time which elapses from the moment water is added until the 232 ENGINEKRING CHEMISTRY paste ceases to be plastic and the time required for it to obtain a certain degree of hardness. The determination of the "initial set" or when plasticity ceases is the more important, as a dis- turbance of the material after this time may cause a loss of strength and thus it is important that the mixing and molding or the incorporating of the material into the work be accomplished within this time. The time of setting is usually determined upon Fig- 35- — Correct method of molding cement pat. one of the pats which is to be used for the soundness test, the top surface being flattened somewhat, as shown in Fig. 34. In using the Gillmore needles care should be taken to apply the needles in a vertical position and perpendicular to the surface of the pat. Fig. 36 shows an arrangement for mounting the Gillmore needles so that they are always perpendicular to the surface of the pat. The rate of setting and hardening may be materially affeted by slight changes in temperature. The per- centage of water used in gauging and the humidity of the moist ENGINEERING CHEMISTRY 233 closet in which the test pieces are stored may also affect the setting somewhat. Tensile Tests. — Consistent results can only be obtained by exercising great care in molding and testing the briquettes. The correct method of filling the mold is shown in Figs. 37 and 38. In testing, the sides of the briquettes and the clips should be -ci. 3 t'ig. 36, — Method of mounting Gillmore needles. thoroughly cleaned and free from grains of sand or dirt which would prevent a good bearing, and the briquette should be care- fully centered in the clips so as to avoid cross strains. It may be considered good laboratory practice if the individual briquettes of any set do not show a greater variation from the mean value than 8 per cent, for sand mixtures and 12 per cent, for neat mix- tures. 234 DNGINKKRING CHEMISTRY Fig. 37. — Correct method of filling briquette mold. Fig. 38. — Correct method of troweling surface of briquettes. E:NGlN]ei;RING CHE)MISTRY 235 Bureau of Standards, Sieve Specifications. Wire cloth for standard sieves for cement and sand shall be woven (not twilled) from brass, bronze, or other suitable wire, and mounted on the frames without distortion. The sieve frames shall be circular, about 20 centimeters (7.87 inches) in diameter, 6 centimeters (2.36 inches) high, and pro- vided with a pan about 5 centimeters (1.97 inches) deep and a cover. No. 100 Cement Sieve, o.oojj-Inch Opening. — The No. 100 sieve should have 100 wires per inch and shall conform to the following specifications of diameter of wire and size of mesh : The diameter of the wires in the sieve should be 0.0045 i'^ch and the average diameter of such wires as may be measured shall not be outside of the limits 0.0042 to 0.0048 inch for either warp or shoot wires. The number of warp wires per whole inch, as measured at any point of the sieve, shall not be outside the limits 98 to loi per inch, and of the shoot wires 96 to 102 per inch. For any interval of 0.25 to 0.50 inch in which the mesh may be measured the mesh shall not be outside the limits 95 to loi wires per inch for the warp wires and 93 to 103 wires per inch for the shoot wires. A^o. 200 Cement Sieve, 0.002g-Inch Opening. — The No. 200 sieve should have 200 wires per inch and shall conform to the following specifications of diameter of wire and size of mesh : The diameter of the wires in the sieve should be 0.0021 inch, and the average diameter of such wires as may be measured shall not be outside the limits 0.0019 to 0.0023 inch for either warp or shoot wires. The number of warp wires per whole inch, as measured at any point of the sieve, shall not be outside the limits 195 to 202 per inch, and of the shoot wires 192 to 204 per inch. For any interval of 0.25 to 0.50 inch in which the mesh may be measured the mesh shall not be outside the limits 192 to 203 wires per inch for the warp wires and 190 to 205 wires per inch for the shoot wires. No. 20 Sand Sieve, o.oj^^-Inch Opening. — No. 20 sieves shall have between 19.5 and 20.5 wires per whole inch of the warp 236 e:nginekring chemistry wires and between 19 and 21 wires per inch of the shoot wires. The diameter of the wire should be 0.0165 inch and the average as measured shall not vary outside the limits 0.0160 to 0.0170 inch. No. JO Sand Sieve, 0.022^-Inch Opening. — No. 30 sieves shall have between 29.5 and 30.5 wires per whole inch of the warp wires and between 28.5 and 31.5 per whole inch of the shoot wires. The diameter of the wire should be o.oiio inch and the average as measured shall not vary outside the limits 0.0105 to 0.0115 inch. Chemical Analyses. Reprint of Report Authorized by the Committee, New York Section Society for Chemical Industry. Solution. — One-half gram of the finely powdered substance is to be weighed out, and, if a limestone or unburned mixture, strongly ignited in a covered platinum crucible over a strong blast for 15 minutes, or longer if the blast is not powerful enough to effect complete conversion to a cement in this time. It is then transferred to an evaporating dish, preferably of platinum for the sake of celerity in vaporation, moistened with enough water to prevent lumping, and 5 to 10 cc. of strong HCl added and digested with the aid of gentle heat and agitation until solution is complete. Solution may be aided by light pressure with the flattened end of a glass rod.^ The solution is then evaporated to dryness, as far as this may be possible on the bath. Silica (SiO^). — The residue without further heating is treated at first with 5 to 10 cc. of strong HCl which is then diluted to half strength or less, or upon the residue may be poured at once a larger volume of acid of half strength. The dish is then cov- ered and digestion allowed to go on for 10 minutes on the bath, after which the solution is filtered and the separated silica washed thoroughly with water. The filtrate is again evaporated to dry- ness, the residue without further heating taken up with acid and water and the small amount of silica it contains separated on an- 1 If anything remains undecomposed it should be separated, fused with a little Na2C02, dissolved and added to the original solution. Of course a small amount of separated non-gelatinous silica is not to be mistaken for undecomposed matter. ENGINEERING CHEMISTRY 237 other filter paper. The papers containing the residue are trans- ferred wet to a weighed platinum crucible, dried, ignited, first over a Bunsen burner until the carbon of the filter is completely consumed, and finally over the blast for 15 minutes and checked by a further blasting for 10 minutes or to constant weight. The silica, if great accuracy is desired, is treated in the crucible with about 10 cc. of HFl and 4 drops of H0SO4, and evaporated over a low flame to complete dryness. The small residue is finally blasted, for a minute or two, cooled and weighed. The difference between this weight and the weight previously obtained gives the amount of silica.^ Alumina and Iron {AIJD^ and Fefi^). — The filtrate about 250 cc. from the second evaporation for SiOg, is made alkaline with NH4OH after adding HCl, if need be, to insure a total of 10 to 15 cc. strong acid, and boiled to expel excess of NH3, or until there is but a faint odor of it, and the precipitate iron and aluminum hydrates, after settling, are washed once by decantation and slightly on the filter. Setting aside the filtrate, the precipi- tate, is dissolved in hot dilute HCl, the solution passing into the beaker in which the precipitation was made. The aluminum and iron are then reprecipitated by NH^OH, boiled and the second precipitate collected and washed on the same filter used in the first instance. The filter paper, with the precipitate, is then placed in a weighed platinum crucible, the paper burned off and the precipitate ignited and finally blasted 5 minutes, with care to prevent reduction, cooled and weighed as Al^Og + Fe203.^ Iron {F^2^z)' — "^^^ combined iron and aluminum oxides are fused in a platinum crucible at a very low temperature with about 3 to 4 grams of KHSO4, or, better, NaHS04, the melt taken up with so much dilute H12SO4 that there shall be no less than 5 grams absolute acid and enough water to effect solution on heat- ing. The solution is then evaporated and eventually heated till acid fumes come off copiously. After cooling and redissolving in water the small amount of silica is filtered out, weighed and 1 For. ordinary control in the plant laboratory this correction may, perhaps, be neglected ; the double evaporation never. ' This precipitate contains TiOo, P2O5, Mn304. 238 Engine:ering che:mistry corrected by HFl and H,2S04/ The filtrate is reduced by zinc, or preferably by hydrogen sulphide, boiling out the excess of the latter afterwards while passing CO2 through the flask, and titrated with permanganate.^ The strength of the permanganate solution should not be greater than 0.0040 gram Fe^Og per cubic centimeter. Lime (CaO). — To the combined filtrate from the AI2O3 + FcisOg precipitate a few drops of NH4OH are added, and the solution brought to boiling. To the boiling solution 20 cc. of a saturated solution of ammonium oxalate are added, and the boil- ing continued until the precipitated CaQ04 assumes a well- defined granular form. It is then allowed to stand for 20 min- utes, or until the precipitate has settled, and then filtered and washed. The precipitate and filter are placed wet in a platinum crucible, and the paper burned off over a small flame of a Bunsen burner. It is then ignited, redissolved in HCl, and the solution made up to 100 cc. with water. Ammonia is added in slight excess, and the liquid is boiled. If a small amount of AUOg separates this is filtered out, weighed, and the amount added to that found in the first determination, when greater accuracy is desired. The lime is then reprecipitated by ammonium oxalate, allowed to stand until settled, filtered and washed,^ weighed as oxide by ignition and blasting in a covered crucible to constant weight, or determined with dilute standard permanganate.* Magnesia (MgO). — The combined filtrates from the calcium precipitates are acidified with HCl and concentrated on the steam bath to about 150 cc, 10 cc. of saturated solution of Na (HN4)HP04 are added, and the solution boiled for several minutes. It is then removed from the flame and cooled by plac- ing the beaker in ice water. After cooling, NH4OH is added 1 This correction of Al203Fe203 for silica should not be made when the HFl correction of the main silica has been omitted, unless that silica was obtained by only one evapora- tion and filtration. After two evaporations and filtrations i to 2 milligrams of SiO are still to be found with the AloOg FcoOg. 2 In this way only is the influence of titanium to be avoided and a correct result obtained for iron. 3 The volume of wash-water should not be too large ; vide Hillebrand. * The accuracy of this method admits of criticism, but its convenience and rapidity demand its insertion. e)ngine:e:ring chemistry 239 drop by drop with constant stirring until the crystalline ammo- nium magnesium orthophosphate begins to form, and then in moderate excess, the stirring being continued for several minutes. It is then set aside for several hours in a cool atmosphere and filtered. The precipitate is redissolved in hot dilute HCl, the solution made up to about 100 cc, i cc. of a saturated solution of Na (NH4)HP04 added, and ammonia drop by drop, with constant stirring until the precipitate is again formed as described and the ammonia is in moderate excess. It is then allowed to stand for about 2 hours, when it is filtered on a paper or a gooch crucible, ignited, cooled and weighed as MgsPaO^. Alkalies (K^^O and Na.fi). — For the determination of the alkalies, the well-known method of Prof. J. Lawrence Smith is to be followed, either with or without the addition of CaCOg with NH.Cl. Anhydrous Sulphuric Acid (SO^). — One gram of the sub- stance is dissolved in 15 cc. of HCl, filtered and residue washed thoroughly.^ The solution is made up to 250 cc. in a beaker and boiled. To the boiling solution 10 cc. of a saturated solution of BaCU is added slowly drop by drop from a pipette and the boiling con- tinued until the precipitate is well formed, or digestion on the steam bath may be substituted for the boiling. It is then set aside over night, or for a few hours, filtered, ignited and weighed as BaSO^. Total Sulphur. — One gram of the material is weighed out in a large platinum crucible and fused with NagCOg and a little KNO3, being careful to avoid contamination from sulphur in the gases from source of heat. This may be done by fitting the crucible in a hole in an asbestos board. The melt is treated in the crucible with boiling water and the liquid poured into a tall, narrow beaker and more hot water added until the mass is disintegrated. The solution is then filtered. The filtrate contained in a No. 4 beaker is to be acidulated with HCl and made up to 250 cc. with dis- ^ Evaporation to dryness is unnecessary, unless gelatinous silica should have separated and should never be performed on a bath heated by gas; vide Hillebrand. 240 ENGINEERING CHEMISTRY tilled water, boiled, the sulphur precipitated as BaS04, and allowed to stand over night or for a few hours. Loss on Ignition. — Half a gram of cement is to be weighed out in a platinum crucible, placed in a hole in an asbestos board so that about three-fifths of the crucible projects below, and blasted 15 minutes, preferably with an inclined flame. The loss by weight, which is checked by a second blasting of 5 minutes, is the loss on ignition. May, 1903.- — Recent investigations have shown that large errors in results are often due to the use of impure distilled water and reagents. The analyst should, therefore, test his distilled water by evaporation and his reagents by appropriate tests before pro- ceeding with his work. Specifications for Quicklime.^ 1. Quicklime is a material the major part of which is calcium oxide or calcium and magnesium oxides, which will slake on the addition of water. 2. Quicklime is divided into two grades :' (a) Selected — Shall be a well-burned, picked free from ashes, core, clinker or other foreign materials. (b) Run-of-Kiln — Shall be a well-burned lime without selection. 3. Quicklime is shipped in two forms : (a) Lump Lime — Shall be the size in which it comes from the kiln. (6) Pulverized Lime — Shall be lump limed reduced in size to pass a 54-inch screen. 4. Quicklimes are divided according to their chemical composition into four types : (a) High-Calcium — Shall be quicklime containing over 90 per cent, of calcium oxide. {h) Calcium — Shall be quicklime containing not under 85 per cent. and not over 90 per cent, of calcium oxide. (c) Magnesium — Shall be quicklime containing between 10 and 25 per cent, of magnesium oxide. {d) Dolomitic — Shall be quicklime containing not under 25 per cent, of magnesium oxide. 5. The particular grade, form and type shall be specified by purchaser. ^Tentative Specifications, Amer. Soc. Testing Materials, 1914. PP- 370-372. ^NGINEEIRING CHEMISTRY 24I I. Chemical Properties and Tests. (A) Sampling. 6. When quicklime is shipped in bulk, the sample shall be so taken that it will represent an average of all parts of the shipment from top to bottom, and shall not contain a disproportionate share of the top and bottom layers, which are most subject to changes. The samples shall comprise at least 10 shovelfuls taken from different parts of the ship- ment. The total sample taken shall weigh at least 100 pounds and shall be crushed to pass a i-inch ring, and quartered to provide a 15-pound sample for the laboratory. 7. When quicklime is shipped in barrels, at least 3 per cent, of the number of barrels shall be sampled. They shall be taken from various parts of the shipment, dumped, mixed and sampled as specified in Sec- tion 6. 8. All samples to be sent to the laboratory shall be immediately transferred to an air-tight container in which the unused portion shall be stored until the quicklime shall finally be accepted or rejected by the purchaser. (J5) Chemical Tests. 9. (a) The grade, type and chemical properties of quicklime shall be determined by standard chemical methods of analysis. (b) Selected quicklime shall contain not under 90 per cent, of calcium and magnesium oxides and not over 3 per cent, of carbon dioxide. (c) Run-of-kiln quicklime shall contain not under 85 per cent, of calcium and magnesium oxides, and not over 5 per cent, of carbon dioxide. II. Physical Properties and Tests. 10. An average 5-pound sample shall be put into a box and slaked by an experienced operator with sufficient water to produce the maximum quantity of lime putty, care being taken to avoid "burning" or "drowning" the lime. It shall be allowed to stand for 24 hours and then washed through a 20-mesh sieve by a stream of water having a moderate pres- sure. No materials shall be rubbed through the screens. Not over 3 per cent, of the weight of the selected quicklime nor over 5 per cent, of the weight of the run-of-kiln quicklime shall be retained on the sieve. The sample of lump lime taken for this test shall be broken to pass a i-inch screen and be retained on a ^-inch screen. Pulverized lime shall be tested as received. III. Inspection and Rejection. 11. (a) All quicklime shall be subject to inspection. (b) The quicklime may be inspected either at the place of manufac- ture or the point of delivery as arranged at the time of purchase. 16 242 e:ngine:ering chemistry (c) The inspector representing the purchaser shall have free entry at all times while work on the contract of the purchaser is being per- formed, to all parts of the manufacturer's w^orks which concern the manufacture of the quicklime ordered. The manufacturer shall afford the inspector all reasonable facilities for inspection and sampling, which shall be so conducted as not to inter- fere unnecessarily with the operation of the works. (d) The purchaser may make the tests to govern the acceptance or rejection of the quicklime in his own laborator}^ or elsewhere. Such tests, however, shall be made at the expense of the purchaser. 12. Unless otherwise specified, an}- rejection based on failure to pass tests prescribed in these specifications shall be reported within 5 days from the taking of the samples. 13. Rehearing. — Samples which represent rejected quicklime shall be preserved in air-tight containers for 5 days from the date of the test report. In case of dissatisfaction with the results of the tests, the manu- facturer may make claim for a rehearing w'ithin that time. CONCRETE. Some Field and Laboratory Tests of Concrete.^ One of the important checks instituted within the past year on the concrete work of the New York State Highway Commis- sion was in the testing of the finished product. Engineers in charge of concrete work are required to make 6-mch cubes from the mixed concrete as deposited in the work. Two cubes are taken from every 500 cubic yards of concrete laid; this in the case of a concrete highway 16 feet wide and of our standard thickness represents about 1,700 linear feet of roadway. These specimens are stored in moist sand near the highway for 21 days at which time they are sent to the laboratory, where at 28 days they are tested for compressive strength. Some of the tests so made are tabulated below. Table I is the record of some very high compression breaks obtained on 6-inch cubes made in the field from material being placed on the high- way. There is some slight variation in the age of the different cubes due to delay in shipping, but a large proportion of the material is from 28 to 30 days old. This is not an average high- "^ Engineering News, Jan. 21, 1915, by H. S. Mattimore, First Assistant Engineer, New York State Highway Commission, Albany, N. Y. DNGINEE^RING CHl^MISTRY 243 way; in fact, it is one of the best. The two interesting features I Fig. 39. — Hydraulic compression machine for making compression tests of Portland cement and concrete cubes. of these tests are the consistent uniformity of the breaks and the 244 ENGINEERING CHEMISTRY fact that the compressions are the highest I have ever seen re- corded on plain concrete. I wish to call attention to the tests on the two sands from different banks, used in this concrete. Both are coarse sands with voids slightly below the average, they are comparatively clean and show a good compressive strength in mortar. They would not be called sharp sands as many of the grains are rounded. This material is from an old lake deposit and unfortunately rather limited. 4000 CUBSS) ?8D^YS Age o-f Concre+e 6WEEKS Fig. 40. — Curves showing relative compressive strength of i: 1^:3 concrete using different aggregates. (Six-inch cubes made in field from concrete being used in highway work and stored in field 21 days.) It will be noted the last two cubes tabulated were mixed with stone screenings substituted for sand. This concrete was not used in the highway, but was made for experimental purposes only. It may be rather an unfair comparison as only two cubes were made from this material, but there is such a great difference in compressive strength that we do feel thankful that sand rather than stone screenings was used in the concrete placed in the highway. The curves shown in the accompanying diagram give a good illustration of concrete obtained in acttial practice. I believe this is a good, fair comparison of stone and gravel concrete, as both of these aggregates were of high grade, particular attention being paid to obtain a clean, uniform product passing all requirements for this class of concrete. ENGINEERING CHEMISTRY 245 TABLE I.— Compression Tests on i : i>^ : 3 Concrete Cubes, Made From Material Being Placed on N. Y. State Highway. Break lb. per Break lb. per 1 Age sq. in., aver, of 2 cubes Sand used Age sq. in., aver, of 2 cubes Sand used 30 5480-;- 33 5555 -T- 34 4365 'S 28 4500 7 45 4«i5 u 28 4680 ^ 36 3980 n3 28 4050 " 28 4810 c 30 4665 c 29 4925 ^ 30 4420 S 30 4675 '- 28 5030 M 28 4625 28 4450 d 28 4875^ ;?; 32 33 4450 1945* Z * The concrete lepresented by the last two cubes was made of stone and screenings substituted for sand. It was made for experiment, not used on highway. Tests of Sand Used IN Above Concrete. No. I. Voids = 31.2 5< gradation ;4 Loam = 1.5 ft compression break No. 2. Voids = 28.6 fo gradation ^ Loam = 2.5 ^i compression break Passing % in. loo.o ] Passing No. 6, 93.3 | Passing No. 20, 41.7 I Passing No. 50, 10.5 [ Passing No. 100, 1.8 j Passing No. 200, 0.9 J I : 3 Mortar Ottawa, 1420 Natural, 1987 Washed, 1802 Passing % in. loo.o ) Passing No, 6, 94.0 | Passing No. 20, 28.7 ! Passing No. 50, 9-1 f Passing No. 100, 1.6 | Passing No. 200, i.i J I : 3 Mortar Ottawa, 1520 Natural, 1750 Washed, 1685 The material made with screenings, although show^ing a fair compression strength, is much below that obtained with a good sand. In Table II are given other comparative results of the com- pressive strength of concrete with different kinds of aggregate. TABLvE II.— Comparative Tests on 1 : 2>^ : 5 Concrete Using Different Aggregates (6-in. Cubes) Aggregates Slag and sand Stone and sand Gravel and sand Stone and screenings Coated gravel and sand unwashed Age Days 28 28 28 28 28 No. of cubes 16 134 142 28 48 Compression break, lb. per sq. in. 2,000 1.990 1.895 1,740 1,170 246 ENGINEERING CHEMISTRY Table III is given to show the importance of proper sampHng of sand sources. Both of these sands were from the same bank, being sampled by different engineers. Sand No. i is well within our specifications for use in the best grade of concrete while sand No. 2 is of a poor grade for any concrete. TABLE III. — Table to Show Effect of Sand Gradation on Compressive Strength of Concrete. Gradation Passing ^ in. • Passing No. 6 Passing No. 20 Passing No. 5C1 - Passing No. 100 Passing No. 200 Voids per cent. Loam per cent. Comparative strength (lb. per sq. in.) Ottawa Natural Natural Ottawa in per cent 157-5 No. I No. 2 per cent. per cent. 1 00.0 100. 94 -o 74-5 38.4 30-4 14.4 26.7 2.0 8.4 1.4 4.5 30.9 25-5 3-6 4.0 1.445 1.975 2,275 875 44.2 Specifications for Concrete Pavement and Curb Foundations. Bureau of Highways, Borough of Manhattan, N. Y., 1914. (Partial.) 1. Concrete for pavement and for curb foundations shall be composed of I part Portland cement, 3 parts of sand and 6 parts of broken stone or gravel, or a mixture of both, measured by volume. For the purpose of determining these proportions i bag of cement shall be considered a cubic foot and the other materials shall be measured in approved receptacles. 2. The term Portland cement shall signify the finely pulverized product resulting from the calcination to incipient fusion of an intimate mixture of properly proportioned argillaceous and calcareous materials, and to which no addition greater than 3 per cent, has been made subse- quent to calcination. 3. It shall be finely ground, of uniform color, and free from lumps and cakes. It shall weigh not less than 380 pounds to the barrel, 4 bags of 95 pounds each, being considered equivalent to i barrel. 4. The sample shall be a fair average of the contents of the package e:ngine:e:ring che:mistry 247 taken from surface to center. Samples must be submitted at least 10 days (Sundays and holidays excluded) before using, for the inspection and approval of the engineer. 15. The sand shall be clean, sharp, free from dirt, mica and vegetable matter, and shall contain not more than 5 per cent, of clay. It shall con- tain both coarse and fine particles and be so graded that not more than 10 per cent, shall be retained on a No. 4 sieve and all retained on a No. 100 sieve. Sand which does not fulfil the above requirements in its natural condition shall be screened, w^orked or mixed with other sand to produce a result in accordance with said requirements. 16. If approved b}^ the engineer an intimate mixture of sand and crusher screenings may be used instead of sand alone. This mixture will generally be of equal parts, though the proportions of screenings to sand may be materially increased at the discretion of the engineer. Crusher screenings shall be free from dirt, clay, mica and vegetable matter, and all shall pass a No. 4 sieve and be retained on a No. 100 sieve. 22. The concrete foundation shall be 6 inches thick, unless otherwise ordered and shall withstand such tests a^ may be deemed necessary, and the contractor shall furnish such samples as may be required for the purpose. "Oil-mixed Portland cement concrete"^ by Logan Waller Page, Director, Office of Public Roads, U. S. Dept. of Agriculture, gives a series of important tests, with the following summary of conclusions : ( I ) . The tensile strength of i : 3 oil mixed mortar is very little different from that of plain mortar, and shows a substantial gain in strength at 28 days and 6 months over that at 7 days. (2) The times of initial and final set are delayed by the addi- tion of oil ; 5 per cent, of oil increases the time of initial set by 50 per cent, and the time of final set by 47 per cent. (3) The crushing strength of mortar and concrete is decreased by the addition of oil to the mix. Concrete with 10 per cent, of oil has 75 per cent, of the strength of plain concrete at 28 days. At the age of i year the crushing strength of i : 3 mortar suffers but little with the addition of oil in amounts up to 10 per cent. (4) The toughness or resistance to impact is but slightly af- fected by the addition of oil in amounts up to about 10 per cent. 1 Office of Public Roads, Bulletin No. 46, U. S. Dept. of Agriculture. 248 ENGINEERING CHEMISTRY Fig. 41. — Impact test on oil-mixed concrete. ENGINEERING CHEMISTRY 249 (5) The stiffness of oil-mixed concrete appears to be but little different from that of plain concrete. (6) Elasticity. — Results of tests for permanent deformation indicate that no definite law is followed by oil mixed concrete. (7) Absorption. — Oil mixed mortar and concrete containing 10 per cent, of oil have very little absprption and under low pressures both are water-proof. (8) Permeability. — Oil-mixed mortar containing lo per cent, of oil is absolutely water-tight under pressures as high as 40 potmds per square inch. Tests indicate that oil-mixed mortar is effective as a waterproofing agent under low pressures when plastered on either side of porous concrete. (9) The bond tests show the inadvisability of using plain bar reinforcement with oil-concrete mixtures. The bond of de- formed bars is not seriously weakened by the addition of oil in amounts up to 10 per cent. Note. — A public patent has been granted for mixing oil with Portland cement concrete and hydraulic cements giving an alkaline reaction, and therefore anyone is at liberty to use this process without the payment of royalties. References. "Proportioning Aggregates for Portland Cement Concrete," by Albert Moyer, Amer. Society Testing Materials, July, 1914. The author states : "One of the principal results obtained in these investigations was that arbitrary specifications without previous knowledge of the character of the aggregates that are to be used are wrong, and that such stated pro- portions as 1 : 2 :4 or 1 :3 :6, etc., are meaningless. "It was found that 94 cubic feet, or 3.8 cubic feet per barrel, as a unit of measurement is incorrect. Investigations prove that it takes no pounds of Portland cement to make i cubic foot of paste. "The paper gives various methods of carrying on investigations, so that with a given sand and a given stone or gravel, proportions can be stated by the engineer which will make a concrete of maximum density and maximum strength." "Test Results with Concrete Water-proofing Materials," Bng. News, Jan. 21, 1915. "Standard Practice Instructions for Concrete Testing Laboratory," by Ralph E. Goodwin, in charge of tests of concrete and concrete aggregates. Public Service Commission, New York City, Bng. News, Feb. 4, 1915. 250 ENGINEERING CHEMISTRY ANALYSIS OF CLAY, KAOLIN, FIRE SAND, BUILDING STONES, ETC. The following are to be determined: Silica (total, combined, free, hydrated), alumina, lime, magnesia, potash, soda, ferrous or ferric oxide, manganous oxide, titanic oxide, sulphur trioxide, and combined water. The total silica is determined by fusing i gram of the clay (previously dried at 100° C.) with 10 parts of an equal mixture of sodium and potassium carbonates, in a large platinum cruci- ble. Fusion must be complete and maintained at a red heat 30 minutes. Allow to cool, treat with an excess of boiling water, make acid with hydrochloric acid, transfer solution to a 4-inch porcelain capsule and evaporate to dryness. Take up with 25 cc. hydro- chloric acid, add water, boil, and filter upon ashless filter. Wash well with boiling water, dry, ignite, and weigh as silica (total). The forms of combination of the silica in the clay are deter- mined as follows :^ Let A represent silica in combination with bases of the clay. Let B represent hydrated silicic acid. Let C represent quartz sand. Dry 2 grams of the clay at a temperature of 100° C, heat with sulphuric acid, to which a little water has been added, for 8 or 10 hours, evaporate to dryness, cool, add water, filter out the undissolved residue, wash, dry, and weigh A + B + C. Now transfer it in small portions at a time to a boilmg solution of sodium carbonate (i: 10) contained in a platinum dish, boil for some time, filter ofif each time, still very hot. When all is transferred to the dish, boil repeatedly with strong solution of sodium carbonate, until a few drops of the fluid, passing through the filter, finally remains clear on warming- with ammonium chloride. Wash the residue, first with hot water, then (to in- sure the removal of every trace of sodium carbonate which may still adhere to it) with water slightly acidified with hydrochloric acid, and finally with w^ater. This will dissolve A 4- B, and leave a residue C of sand, which dry, ignite, and weigh. ^ From Fresenius' "Quantitative Analysis," Cairns, p. 68. ENGINEERING CHEMISTRY 2"^ To determine B, boil 4 or 5 grams of the clay (previously dried at 100° C.) directly with a strong solution of sodium carbonate, in a platinum dish as above, and filter and wash thoroughly with hot water. Acidify the filtrate with hydro- chloric acid, evaporate to dryness and determine this silica. It represents B or the hydrated silicic acid. Add together the weights of B and C thus found and subtract the sum from the weight of the first residue A + B + C. The difference will be the weight of A or silica in combination with bases in the clay. If the weight of A + B + C found here be the same as that of the silica found by fusign, in another sample of tlie clay of the same amount, the sand is quartz, but if the weight of A -f- B -f C be greater, then the sand contains silicates. The weight of the bases combined with silica to silicates can be found by subtracting the weight of total silica found in i gram, by fusion, from the weight of A + B + C in i gram. Alumina, Ferric Oxide, Manganese Dioxide, Lime, and Magnesia. The hydrochloric acid filtrate from the silica (by fusion) is made nearly alkaline with sodium carbonate, then excess of sodium acetate added, the solution boiled 5 minutes then filtered bv decantation and washed well. Residue , AI2O3 . Fe203 . Dissolve in hot dilute H0SO4 and divide into two equal portions. First Portion — Make alkaline with NH4OH boil and filter, wash dry, ignite, and weigh as Al203.Feo03. Second Portion'^— Ti trate for iron calcu- late Feo found to Fe203, and this subtracted from weight of AloO;. l^'^sOs gives weight of theAloOg. Both weights to be multiplied by 2. Filtrate, Transfer to a flask, add a few drops of Br, set aside 12 hours, filter and wash. Residue, MnOg. Dry, ignite, weigh as Mn304 and Filtrate. Add a few drops of ammonia (reaction of solution must be alkaline), then an excess of solution of ammonium oxalate, set aside 4 hours, filter and wash. Residue, CaC204. Dry. ignite and weigh as CaO. Filtrate. Add solution of so- dium phosphate with stirring, set aside 4 hours, fi 1 ter, wash with ammoniacal water, dry and ignite, weigh as Mg2P207, and calcu- late to MgO. 1 If iron is present in small amount, fuse 3 grams of the clav with Na2C03, dissolve in HoO, acidify with HCl, evaporate to dryness, take up with HCl. precipitate the FejOg AI2O3 in the filtrate, filter, wash precipitate well with water, dissolve in dilute HJSO4, transfer to 200 cc. flask with Bunsen valve, reduce with zinc and titrate with standard permanganate of potash solution. 252 Engine:e:ring che;mistry Potash and Soda. Take i gram of the dried clay, transfer to a 3-inch platinum capsule, add 10 cc. sulphuric acid and 20 cc. hydrofluoric acid and heat gently until the silica is completely vaporized and the excess of acid added driven ofif. Allow to cool, add 20 cc. warm hydrochloric acid, then 25 cc. water, transfer contents of plati- num capsule to a No. 3 beaker, add 2 cc. nitric acid, and boil. Add ammonia to alkaline reaction, boil, filter off the alumina and ferric oxide, and to the filtrate add ammonium oxalate to precipi- tate the lime; allow to stand 4 hours, then filter; the magnesia is separated in the filtrate by ammonium phosphate, and the fil- trate from the magnesium phosphate precipitate is evaporated to dryness and ignited to expel ammonium salts. The residue is treated with hydrochloric acid and the potash precipitated by solution of platinic chloride as usual, and weighed as KsPtClg on counterpoised filters. The alcoholic washings and filtrate are evaporated to dryness, the platinum compound decomposed by heating to redness with oxalic acid, boiled with water, filtered, a few drops of sulphuric acid added, then evaporated to dryness, ignited to constant weight as sodium sulphate, and then cal- culated to NagO. Sulphur Trioxide. This is determined by fusing i gram of the clay with sodium and potassium carbonates, separating the silica as usual, and precipitating the sulphur trioxide by solution of barium chloride in the acid filtrate. Titanic Oxide. Fuse 5 grams of the dried clay with an excess of a mixture of sodium fluoride and sodium bisulphate, in a platinum crucible for 30 minutes at a red heat. Treat the cold mass with cold water, about 200 cc, add potassium hydroxide in excess, filter off the titanic oxide, wash, dry, and ignite and ftise this titanic oxide with about twelve times its weight of acid sodium sulphate; allow to cool, and treat with concentrated sulphuric acid. This ENGINEERING CHEMISTRY 253 is now added to 600 cc. of water, boiled for i hour, and the precipitated titanic oxide filtered, dried, and weighed. Water of HydratioiL Take 2 grams of the clay, dried at 100^ C, transfer to a cov- ered platinum crucible and ignite over a blast lamp at a red heat to constant weight. The loss represents the combined w^ater. Composition of Some Representative Fire Clays. SiOj (com'd) A1,0, H,0 KjO NajO CaO MgO Fe,0, SiOj (freei .- Moisture. TiO,.-... SO, Org. matter . 50.46 35-90 12.74 0.13 0.02 1-50 Tot a] 50-15 35-60 13-61 0.07 O.Il 0.16 083 0.14 56.42 26.35 10-95 0.48 0.60 055 1-33 2.80 1.15 65.10 22.22 7.10 0.18 0.14 0.18 1.92 2.18 058 39-94 36.30 14.52 0.42 0.19 0.19 0.46 4.90 3.26 0.72 0.35 0.14 0.22 0-18 98-31 40.33 38.54 13.00 0.66 0.08 0.38 0.90 5-'5 29.67 20.87 8.61 1.55 0.30 1.45 36.41 1.14 44.20 39-14 14-05 0.25 0.45 0.20 0.90 1.05 100.75100.67 100.63' 99.60 99. i& 99.92 99.24 ICO. 00 100.24 Clays or fire sands that are to be used in the manufacture of fire-bricks, retorts, etc., should contain only small amounts of easily fusible materials, such as potash, soda, or iron, less than I per cent, of either alkali, or 2 per cent, of iron oxide being allowable in the best fire clays. No. I. — Mt. Savage fire clay, Md. Xo. 2. — Fire clay, Clearfield County, Pa. Xo. 3. — Glenboig clay, England. Xo. 4- — Stourbridge clay, England. Xo. 5. — Saaran clay, Germany. Xo. 6.— "Dinas," England.^ Xo. 7. — ^Zettlitz clay, Bohemia. Xo. 8- — Stoneware clay, X. J. Xo. 9. — Paper clay, N. J. * Used in making the celebrated "Dinas** fire-bricks, noted for their endurance at high heats and for swdUnc and making ti^t roofs for furnaces.. 254 ENGINEERING CHEMISTRY Building stone, such as granite, limestone, sandstone, slate, brick, etc., are generally subjected to certain mechanical or phys- ical tests in addition to a chemical analysis to determine their relative value. These physical tests generally comprise: 1. Crushing strength. 2. Absorptive power. 3. Resistance to the expansion of frost, by saturating the stone with water and freezing a number of times to produce an effect similar to frost. 4. Microscopical examination. Crushing Strength of Various Building Stones. Kind of stone. Granite Trap rock of New Jersey Marble Limestone Sandstone Common red brick • • . Ultimate crushing strength. Pounds per square inch. Minimum. 12,000 20,000 8,000 7,000 5,000 2,000 Maximum. 21,000 24,000 20,000 20,000 15,000 3,000 Tons per .square foot. 860 1,440 580 500 360 144 Maximum. 1,510 1,730 1,440 1,440 i,oSo 216 1. Crushing Strength. The crushing strength is generally determined by applying a measured force to i-inch or 2-inch cubes of the material until they are crushed. These compression tests are comparative only and give no idea of the crushing strength of the material in large masses. A Riehle U. S. Standard automatic and autographic testing machine is used for this. purpose (Fig. 42). In the specifications for granite block for street pavement, the following is selected as a portion of the requirements : Quality. — The granite from which the blocks are cut shall be medium grained, showing uniformity in quality and texture, without seams, scales or discolorations indicating disintegration, an even distribution of con- stituent minerals, and free from mica or feldspar. No outcrop, soft, brittle or seamy stones will be accepted. i;ngine:e:ring chemistry 255 Dimensions. — The size of the blocks shall be as follows: Not less than 6 inches or more than 10 inches long; not less than 3^ inches or more than 4I/2 inches wide, and not less than 4^4 inches or more than 5^ inches deep. Inspection before Delivery. — The paving blocks will be inspected at the quarry or at the dock as unloaded, and if the percentage of blocks failing to conform to these specifications found in 1,000 blocks, as unloaded, shall exceed 15 per cent, the whole cargo will be condemned and shall not be used on the work. 256 ENGINEERING CHEMISTRY % 00 10 t^ rO Tj- Tt O o o '^^ "2 I *^^°° "0 ^ ^, '^ ", I ^ 't *":: I i-T hT. pT hh m i-T i-T i-T « «," m" i-T 00 M "-I M I ^1 5 - S X o P O d" ro d" I 00 I vo O 10 «_ n 00 O^ fO O 10 (^ t^ On •I ^ ON Tt fO O I ONOO "^ I I •^00 O NO 00 C» M O t^ , r-'OO r^NO « »OVO , OO-^i Tt.vOOOOOvOVO ,VOOO , »0 O ON O ON £, C •Sr rj o 0»0 t^ONiOOOOOO"-! vOO"-! — ci I Tt -.' t-I d 1^ OnOO" 00 00" On VO NO NO NO vO t~<.NO NO NO vO t^ rO 00 10 Tj-NO On I tOfONO' MM fSSK S (U ii-2i^So2 ■^ i» o nJ a> - — ^^ .^ s-o 2 ^ 5 « PQ 2 S H S P^ o o W W H^ is ^ c o o ill' .222^ o a >• • .^ >-< ~ u rt CO Sals ' O O cfl o t: oj tfl n a CO c CO . aC'-* e -So S « - C (L» a; -M c ^ 2 "^ c^ 'S J2 tn CO is ™ c n *j '^ rt rt 3{t: j- en M O o w CO 'S CO u I ^^51. DNGINEKRING CHEMISTRY 257 Compression Test of Granite Cxtbes. Designation of specimen Dimension specimen Crushing load I^ength inches Width inches Breadth inches Area sq. in. Actual lb. I^b. per sq. in. I.— w 2.— W T V 1.97 1.97 1.98 1.98 1.968 1. 961 1.978 1.967 1. 961 1.967 I.97S 1,980 3.859 3.857 3.912 3.895 89,360 100,000 95,170 100,000 2.^,164 25,930 24.330 25,670 2.— Y 2. Absorptive Power.^ This is determined by drying the sample and weighing it, then soaking it in water for 24 hours and weighing again. The increase of weight represents the amount of water absorbed. A close, fine-grained stone absorbs less water than a coarse-grained one, and generally the less the absorption, the better the stone. Absorptive Power of Stone, Brick, and Mortar, Kind of material Granite . • Marble . . . Limestone Sandstone Brick .... Mortar • . • Rate of absorption Maximum Minimum I— 150 I-I50 1—20 1-500 I— 15 1—240 1-5 1—50 1 — 2 I — JO The following is the method for absorption in use at the Road Material Laboratory, Bureau of Chemistry, U. S. Dept. of Agriculture: The method used for determining the absorptive- ness of rock is not intended to give the porosity, but merely to obtain the number of pounds of water absorbed by a cubic foot or rock in 96 hours, determined from small samples. A smoothly worn stone, between 20 and 60 grams in weight, which has been through the abrasion test is used. Afte;r being weighed ^ Thus, if 150 units of dry granite weigh after immersion in water 151 units, absorption is i in 150 stated i — 150. 17 the 258 ENGINEJERING CHE;MISTRY in air it is immersed in water and immediately re-weighed in water. The absorption is obtained by the following formula : Number of pounds of water absorbed by a cubic foot of rock zi: C — B A— B X 62.5, in which A is equal to the weight in air, B the weight in water irnmediately after immersion, C the weight after absorption for 96 hours, and 62.5 the weight of a cubic foot of water. From these weights, the specific gravity and the weight per cubic foot of the rocks are determined.^ 3. Freezing Test. Samples of the weighed material, preferably cut in 2-inch cubes, are saturated with water, then placed in a Tagliabue freez- Fig- 43. ing apparatus (Fig. 43 and maintained at a temperature of ^ Transactions American Society for Testing Materials, 1903, p. 300. KNGINKERING CHEMISTRY 259 10° F. for 4 hours. They are then removed, allowed to thaw gradually to a temperature of about 65° ; then moistened with water and placed again in the freezing apparatus and maintained at a temperature of 10° F. for 4 hours. This process is repeated at least ten times, when, after the samples have acquired the temperature of the room, the moisture is wiped from them, they are then dried, and their weight carefully determined. The loss of weight represents the material broken off by the expansive action of freezing the contained water. The following method of making the frost test of building stones is from "Uniform Methods of Procedure in Testing Building and Structural Materials" by J. Bauschinger (Mechanisch-technischen I^abora- torium, Miinchen).^ The examination of resistance to frost is to be determined from samples of uniform size, inasmuch as the absorption of water and action of frost are directly proportional to the surface exposed. The test sample should be a cube of 7 centimeters (2.76 inches) length on edges. The frost test consists of : a. The determination of the compressive strength of saturated stones, and its comparison with that of dried pieces. b. The determination of compressive strength of the dried stone after having been frozen and thawed out twenty-five times, and its comparison with that of dried pieces not so treated. c. The determination of the loss of weight of the stone after the twenty-fifth frost and thaw ; special attention must be paid to the loss of those particles which are detached by the mechanical action, and also those lost by solution in a definite quantity of water. d. The examination of the frozen stone by use of a magnify- ing glass, to determine particularly whether fissures or scaling occurred. For the frost test are to be used : Six pieces for compression tests in dry condition, three normal and parallel to the bed of the stone, six test pieces in saturated ^ Standard Tests and Methods of Testing Materials: Trans. Am. Soc. Mech. Eng., 14, 1294. 26o ENGINEERING CHEMISTRY condition, not frozen, however; three tested normal to, and three parallel to, bed of stone. Six test pieces for tests when frozen, three of which are to be tested normal to, and three parallel to, bed of stone. When making the freezing test the following details are to be observed : a. During the absorption of water, the cubes are at first to be immersed by 2 centimeters (0.77 inch) deep, and are to be low- ered, little by little, until finally submerged. h. For immersion distilled water is to be used at a tempera- ture of from 15° to 20° C. c. The standard blocks are to be subjected to temperatures of from 10° to 15° C. d. The blocks are to be subjected to the influence of such cold for 4 hours, and they are to be thus treated when completely saturated. e. The blocks are to be thawed out in a given quantity of dis- tilled water at from 15° to 20° C. The Testing of Brick. — i. When testing bricks as found in a delivery, the least burnt are always to be selected for investi- gation. 2. Bricks are to be tested for resistance to compression in the shape of cubical pieces, formed by the superposition of two half bricks, which are to be united by a thin layer of mortar consist- ing of pure Portland cement, and the pressure surfaces are also to be made smooth by covering them with a thin coating of the same material. At least six pieces are to be tested. 3. The specific gravity is to be determined. 4. In order to control the uniformity of the material, the porosity of the bricks is to be determined; for this purpose they are to be thoroughly dried and then submerged in water until saturated. Ten pieces are to be thoroughly dried upon an iron plate and weighed ; then these bricks are to be immersed in water for 24 hours, in such a way that the water-level stands at half the thickness; after this they are to be submerged for another 24 hours, then to be dried superficially and again weighed; thus ENGINEERING CHEMISTRY 261 the average quantity of water absorbed is determined. The porosity is always to be calculated by volume, though the per cent, of water absorbed is always to be stated in addition. Fig. 44. — Standard automatic transverse brick testing machine. 5. Resistance against frost is to be determined as follows: a. Five of the bricks, previously saturated by water, are to be tested by compression. b. The other five are put into a refrigerator which can produce a temperature of — 15° C. at least, and kept therein for 4 hours; then they are removed and thawed in water of a temperature of 10° C. Particles which might possibly become detached are to remain in the vessels in which the brick is thawed until the end 262 ENGINEERING CHEMISTRY of the Operation. This process of freezing is repeated twenty- five times, and the detached particles are dried and compared by weight with the original dry weight of brick. Particular atten- tion, by using a magnifying glass, is to be given to the possible formation of cracks or laminations. c. After freezing, the bricks are to be tested by compression. For this test they are dried, and the result obtained is to be com- pared with that of dry brick not frozen. d. Thus, freezing the bricks does not give a knowledge of the absolute frost-resisting capacity; the value of the investigation is only relative, because by it can only be determined which brick can be most easily destroyed by the action of frost. 6. To test bricks for the presence of soluble salts, five are selected, and again those which are least burnt, and then such which have not yet been moistened. Of these, again, the interior parts only are used, for which reason the bricks are split in three directions, thus producing eight pieces, of which the corners lying innermost in the brick are knocked off. These are then powdered until all passes through a sieve of 900 meshes per square centi- meter (about 5,840 per square inch), from which the dust is again separated by a sieve of 4,900 meshes per square centimeter (about 31,360 per square inch), and the particles remaining on the latter are examined. Twenty-five grams are lixiviated in 250 cc. of the distilled water, boiled for about i hour, however, replenishing the quantity of water evaporated, then filtered and washed. The quantity of soluble salts present is then determined by boiling down the solution and bringing the residue to a red heat for a few minutes. The quantity of soluble salts present is to be given in per cent, of the original weight of brick. The salts obtained are to be submitted to a chemical analysis. 7. Determinations of the presence of calcium carbonate, pyrites, mica, and similar substances are to be made on the un- burned clay, for which purpose unburned bricks are to be fur- nished. These are soaked in water and the coarse particles are separated by passing the whole material through a sieve having 400 meshes per square centimeter. The sand thus obtained is to ENGINEEIRING CHEMISTRY 263 be examined by the magnifying glass and with hydrochloric acid to determine its mineralogical composition. When impurities, such as carbonate, pyrites, etc., are found, then pieces of brick, such, for instance, as remained from the determination of soluble salts, are to be examined in a Papin's digester for their detele- rious influence. They are to be so arranged in a Papin's digester that they are not touched by the water directly, but are subjected to the action of the generated steam alone. The pressure of steam shall be ^ atmosphere, and the duration of test 3 hours. Possibly occurring disintegration is to be determined by means of the magnifying glass. 4. Microscopical Examination. This consists in examining under the microscope thin sections of the building stone. Important results are often obtained, especially so if the substances used as matrix are indicated — the presence and amount of injurious substances, such as iron pyrites, mica, etc. Nearly all reports upon samples of building stone now in- clude the microscopical examination. The first and most essential test applied to building stone is to determine the structure and character of a stone, to know whether it be of granite, syenite, sandstone, quartzite, or some- thing else. Although an expert can usually determine at a glance to which, if any, of these groups a particular stone be- longs, it is frequently possible to determine the precise litholog- ical character only by a microscopical examination. Thus, for instance, there is a class of Cambrian rocks commonly called Potsdam sandstone, that are not sandstones at all, but are hard, compact rocks known as quartzites, which have been derived from sandstones by metamorphic action. The essential differ- ence between a sandstone and a quartzite lies in the presence of secondary silica between the quartz granules comprising the latter ; the presence of this secondary silica or quartz can be de- termined for a certainty only by microscopical means. The microscope is not only useful in determining the structure of a 264 ENGINEERING CHEMISTRY stone, but it has even greater practical value in making it pos- sible to detect the presence of deleterious substances, such as pyrite and marcasite, or other minerals whose chemical com- position is effected by atmospheric agencies and thus exert a del- eterious effect upon the stone. ^ The Testing of Paving Bricks. This consists in the following tests: 1. Cross breaking. 2. Crushing. 3. Impact (the rattler test). 4. Absorption. 5. Chemical analysis. 1. Cross Breaking.^ 1. Support the brick on edge, or as laid in a pavement, on a hardened steel knife rounded longitudinally to the radius of 12 inches, and traversely to the radius of ji inch, and bolted in position so that the screw-span of 6 inches applied to load in the middle of the top shall pass through the steel knife edge, straight, longitudinal, and rounded transversely to a radius of 1/16 inch. 2. Apply the load to the middle of the top face through a har- dened steel knife-edge, straight, longitudinally, and rounded transversely to a radius of 1/16 inch. 3. Apply the load in a uniform rate of increase until fracture ensues. 4. Complete the modulus of rupture by the formula F = ^ .^ in which F = modulus of rupture in pounds per square inch ; W = the total brick load in pounds ; Iv = the length span in inches, 6 ; B = breadth of brick in inches; D = depth of brick in inches. 5. Samples for test must be free from all visible irregularities of surface, or deformities in shape, and their upper and lower faces must be practically parallel. ^ H. t,ynwood Garrison; Trans. Amer. Soc. Civil Eng., 33, 88. 2 "Street Pavements and Paving Materials," by Geo. W. Tillson, N. Y,, 1900, p. 273. EJNGINDERING CHEJMISTRY 265 6. Not less than 10 bricks shall be broken, and the average of all is to be taken for the standard test. 2. Crushing Test. I. The crushing test should be made of half-brick loaded edge- wise, or as they are laid on the street. If the machine used is unable to crush the full half -brick, the area may be reduced by chipping off, keeping the form of the piece to be tested as nearly prismatic as possible. A machine (Fig. 45) of at least 100,000 Fig. 45. — Standard rattler. pounds capacity should be used and the standard should not be reduced below 4 square inches area in cross-section at right angles to direction of load. 2. The upper and lower surfaces should preferably be ground to true and parallel planes. If this is not done, they should be bedded in plaster of Paris while in the testing-machine, and should be allowed to harden 10 minutes under weight of the crushing plane only before the load is applied. 3. The load should be applied at a uniform rate of inrease to the point of rupture. 266 ENGINEERING CHEMISTRY 4. Not less than the average obtained from five tests of five different bricks shall constitute a standard test. 3. Impact Test (the Rattler Test). The bricks shall not lose more than an average of 23 per cent, in abrasion test conducted in the following manner : 1. Standard Rattler. — The standard shall be of the form and dimensions as adopted by The American Society for Testing Materials, 1914. 2. Standard Shot. — (Abrasive Charge.) The abrasive charge shall consist of two sizes of cast iron spheres. The larger size shall be 3.75 inches in diameter when new and shall weigh when new approximately 7.5 pounds (3.40 kilos) each, 10 to be used. The smaller spheres shall be when new 1.875 inches in diam- eter and shall weigh not to exceed 0.95 pound (0.430 kilo) each. Of these spheres so many shall be used as will bring the collective weight of the large and small spheres most nearly to 300 pounds. 3. The Brick Charge. — The number of brick per charge shall be 10 for all bricks of the so-called ''block-size" whose dimen- sions fall between from 8 to 9 inches in length, 3^4 inches in breadth and 334 and 4% inches in thickness. No block should be selected for test that would be rejected by any other require- ments of the specifications. The brick shall be clean and dried for at least 3 hours in a temperature of 100° F., before testing. 4. Speed and Duration of Revolution. — The rattler shall be rotated at a uniform rate of not less than 29^^ nor more than 30^ revolutions per minute and 1,800 revolutions shall consti- tute the standard test. A counting machine shall be attached to the rattler for count- ing the revolutions. A margin of not to exceed 10 revolutions will be allowed for stopping. Only one start and stop per test is regular and acceptable. 5. The Results. — The loss shall be calculated in percentage of the original weight of the dried brick composing the charge. In e;ngine;e:ring chicmistry 267 weighing the rattled brick any piece weighing less than i pound shall be rejected. All bricks must give modulus of rupture of not less than 1,800 when tested on their sides, being supported on knife edges 6 inches apart. The above coefficients to be calculated by the formula : 3WL ,.^ Modulus 2AD in which W equals breaking weight at the center ; L equals length between supports ; A equals area section ; D equals depth. The Board reserves the right to accept bids on bricks, which have successfully passed the above requirements in the test here- tofore referred to, made under their direction. If samples are submitted on other bricks, they will be subject to similar comparative tests as those heretofore mentioned, and if samples do not meet these requirements, bids upon such bricks will be informal and not considered. Fixing of Standards. — The percentage of loss which may be taken as the standard will not be fixed in these regulations, and shall remain within the province of the contracting parties. For the information of the public, the following scale of average losses is given, representing what may be expected of tests executed under the foregoing specifications. For bricks suitable for heavy traffic. For bricks suitable for medium traffic For bricks suitable for light traffic • • Genera average loss (Per cent.) 24 26 Maximum permissible loss (Per cent.) 24 26 28 which of these grades should be specified in any given district and for any given purpose is a matter wholly within the province of the buyer, and should be governed by the kind and amount of traffic to be carried, and the quality of paving bricks available. Bulletin 23, U. S. Department of Agriculture, Slates. 268 DNGINDERING CHE:mISTRY 4. Absorption Test. 1. Number of Brick. — The number of brick constituting sam- ple of the official test shall be five. 2. Condition of the Brick. — The brick selected for conducting this test shall be such as have been previously exposed to the rattler test. If such are not available, then each whole brick must be broken in half before the test begins. 3. Drying. — The brick shall be dried for 24 hours continu- ously at a temperature of 230°-250° F., before the absorption. 4. Soaking. — The brick shall be weighed before wet, and shall then be completely immersed for 24 hours. 5. Wiping. — After soaking, and before reweighing, the bricks must be wiped until free from surplus water and practically dry on the surface. 6. Weighing. — The samples must then be reweighed at once. The scale must be sensitive to i gram. 7. Calculation of Result. — The increase in weight due to ab- sorption shall be calculated in per cents, of the dry weight of the original bricks. Standard Abrasion Test for Road Material.^ This well-known test is similar in almost all respects to the Deval abrasion test of the French School of Road and Bridges. It has been used since 1878, and is entirely satisfactory for the purposes for which it was designed. Abrasion Test. The machine shall consist of one or more hollow iron cylin- ders; closed at one end and furnished with a tightly fitting iron cover at the other ; the cylinders to be 20 centimeters in diameter and 34 centimeters in depth, inside. The cylinders are to be mounted on a shaft at an angle of 30° with the axis of rotations of the shaft. Standard Abrasion Cylinder for Road Materials. At least 30 pounds of coarsely broken stone shall be available for a test. The rock to be tested shall be broken in pieces as ^ Adopted 1908, Amer. Soc. Testing Materials. e;ngine;e;ring chejmistry 269 nearly uniform in size as possible, and as nearly 50 pieces as possible shall constitute a test sample. The total weight of rock in a test shall be within 10 grams of 5 kilograms. All test pieces shall be washed and thoroughly dried before weighing. Ten thousand revolutions, at the rate of between 30 and 33 to a Fig. 46. — Three gang, abrasion cylinder, belt driven. Weight, 480 pounds. Length, 7 feet. Breadth, 30 inches. Height, 34 inches. minute, must constitute a test. Only the percentage of material worn off which will pass through a 0.16 centimeter (i/i6-inch) mesh sieve shall be considered in determining the amount of wear. This may be expressed either as the per cent, of the 5 kilograms used in the test, or the French coefficient, which is in more general use, may be given; that is, coefficient of wear = 20 X T^= ~%Tr " W" is the weight in grams of the detritus W W under 0.16 centimeter (1/16 inch) in size per kilogram of rock used. Standard Toughness Test for Macadam Rock.^ In the consideration of macadam road materials, toughness is understood to mean the power possessed by a material to resist fracture by impact. In testing macadam rocks under impact, it has been found best to apply a number of blows of successively increasing energy 2 Adopted 1908, Amer. Soc. Testing Materials. 270 ^ngini:e:ring chemistry and note the blow causing failure. The following test involving this principle is, therefore, recommended for determining the toughness of rock for macadam road building. Fig. 47. — Olsen standard impact tester. Length, i foot 9 inches. Height, s feet 8 inches. Breadth, 11 inches. Weight, complete with motor, 250 pounds. Tough ne:ss Test. I. Test pieces may be either cylinders or cubes, 25 millimeters ENGINEERING CHEMISTRY 2^1 in diameter, and 25 millimeters in height, cut perpendicular to the clearage of the rock. Cylinders are recommended as they are cheaper and more easily made. 2. The testing machine shall consist of an anvil 50 kilograms weight, and placed on a concrete foundation. The hammer shall be of 2 kilograms weight, and dropped upon an intervening plunger of i kilogram weight, which rests on the test piece. The lower or bearing surface of this plunger shall be of spherical shape having a radius of i centimeter. This plunger shall be made of hardened steel, and pressed firmly upon the test piece by suitable springs. The test piece shall be adjusted, so that the center of its upper surface is tangent to the spherical end of the plunger. 3. The test shall consist of a i centimeter fall of the hammer for the first blow, and an increased fall of i centimeter for each succeeding blow until failure of the test piece occurs. The num- ber of blows necessary to destroy the test piece is used to repre- sent the toughness, or the centimeter-gram of energy applied may be used. Proposed Definitions of Non-Bituminous Road Materials.^ Chert. — Compact silicioiis rock formed of calcedonic or opaline silica, or both. Crusher-Run Stone. — The product of a stone-crusher, unscreened except for the removal of the particles smaller than those remaining on a 0.32-centimeter (^-inch) screen. Dust. — Earth or other matter in fine, dry particles, so attenuated that they can be raised and carried by the wind. The product of a rock crusher passing through a fine screen. Flour. — Finely ground rocks or minerals pulverized to an impalpable powder. Granite. — A granatoid igneous rock consisting of quartz, orthoclase, more or less oligoclase, biotite and muscovite. Granitoid. — A textural term to describe those igneous rocks which are composed of recognizable minerals. Matrix. — Material used to bind together the materials in agglomerated mass. Rubble. — Rough stones of irregular shapes and sizes, broken from larger masses either naturally or artificially, as by geological action, in quarrying, or in stone-cutting or blasting. ^ Amer. Soc. Testing Materials, 19 14. 272 ENGINEERING CHEMISTRY Soil. — A mixture of fine earthy material with more or less organic matter resulting from the growth and decomposition of vegetation or animal matter. Spawl. — A piece of rock chipped off by a blow of a hammer or other tool. Stone Chips. — Small fragments of stone, irregular in shape, with sharp edges, containing no dust. Tailings. — Stones which after going through the crusher do not pass through the largest openings of the screen. * The Determination of the Apparent Specific Gravity of Rock.^ The apparent specific gravity of rock shall be determined by the fol- lowing method: First, a sample weighing between 29 and 31 grams and approximately cubical in shape shall be dried in a closed oven for i hour at a temperature of 110° C. (230° F.) and then cooled in a desiccator* for I hour; second, the sample shall be rapidly weighed in air; third, trial weighings in air and in water of another sample of approximately the same size shall be made in order to determine the approximate loss in weight on immersion; fourth, after the balances shall have been set at the calculated weight, the first sample shall be weighed as quickly as practicable in distilled water having a temperature of 25° C. (77° F.) ; fifth, the apparent specific gravity of the sample shall be calculated by the following formula : W Apparent specific gravity = —^ — — ' in which W = the weight in grams of the sample in air and Wi = the weight in grams of the sample in water just after immersion. Finally, the apparent specific gravity of the rock shall be the average of three determinations, made on three different samples according to the method above described. The Determination of the Absorption of Water per Cubic Foot of Rock.2 The absorption of water per cubic foot of rock shall be determined by the following method : First, a sample weighing between 29 and 31 grams and approximately cubical in shape shall be dried in a closed oven for I hour at a temperature of 110° C. (230° F.) and then cooled in a desiccator for i hour; second, the sample shall be rapidly weighed in air; third, trial weighings in air and in water of another sample of approxi- mately the same size shall be made in order to determine the approximate loss in weight on immersion; fourth, after the balances shall have been set at the calculated weight, the first sample shall be weighed as quickly as possible in distilled water having a temperature of 25° C. (77° F.) ; fifth, allow the sample to remain 48 hours in distilled water maintained ^Proposed Provisional Test, Amer. See. Testing Materials, 1914. * Proposed Provisional Test, Amer. Soc. Testing Materials, 19 14. e:ngine:ering chemistry 273 as nearly as practicable at 25° C. {^TJ° F.), at the termination of which time bring the water to exactly this temperature and weigh the sample while immersed in it ; sixth, the number of pounds of water absorbed per cubic foot of the sample shall be calculated by the following formula : W.. — W, Pounds of water absorbed per cubic foot = -—7 ~ X 62.24, W — Wi in which W = the weight in grams of sample in air, Wi = the weight in grams of sample in water just after immersion, W2 ^ the weight in grams of sample in water after 48 hours immersion, and 62.24 1= the weight in pounds of a cubic foot of distilled water having a temperature of 25° C. {If F.). Finally, the absorption of water per cubic foot of the rock, in pounds, shall be the average of three determinations made on three different samples according to the method above described. Mechanical Analysis of Broken Stone or Broken Slag.^ The method shall consist of, first, drying at not exceeding 110° C. (230° F.) to a constant weight a sample weighing in pounds six times the diameter in inches of the largest holes required ; second, passing the sample through such of the following sized screens having circular open- ings as are required or called for by the specifications, screens to be used in the order named: 8.89-centimeter (3^-inch), 7.62-centimeter (3-inch), 6.35-centimeter (2j/2-inch), 3.81-centimeter (i^-inch), 3.18-centimeter (i^-inch), 2.54-centimeter (i-inch), 1.90-centimeter (^-inch), 1.27-centi- meter (^-inch), and 0.64-centimeter (54-inch) ; third, determining the percentages by weight retained on each screen; fourth, recording the mechanical analysis in the following manner : Percentage passirlg 0.64-centimeter (^-inch) screen = Percentage passing 1.27-centimeter (^-inch) screen =: Percentage passing 1.90-centimeter (^-inch) screen = Percentage passing 2.54-centimeter (i-inch) screen ■=:. 100.00 ASPHALT.2 The term asphalt, as originally applied, represented a natural deposit of a bituminous substance capable of extraction from the earth in the condition of a solid, or semi-solid body of dark brownish black color of a consistency approaching solid pitch, and capable of complete combustion. ^ Proposed Provisional Method, Amer. Sec. Testing Materials. ^ "Asphalt, its Occurrence, Composition, and Commercial Uses, with Schemes fof its Analysis," by Thos. B. Stillman, Stevens Institute Indicator, Oct., 1904. 18 274 .lENGIN^ERING CHE:miSTRY Hi CO ;3 ^ J c- s .5 o ^ o (/J tM "^ t,_ • 3 rt o CO CO O ^ r^3 c3 .t: ^ 5 o .X ^ N O n eg ^ 'O -^ 4^ 4-; a; pq •;: O rt C!j c3 (LI (L» S(.VJ 03 ^ v.- ^^ A rt CO a. C O (U o o a U3 _ i, ^ ^ n nt* c3 cd cO w W 1) < 2i .2 cd .^ fc o 2 I' 2 o r- ^ -C - ?s si e:ngine:e;ring chemistry 275 O > V O :5 '5 i2 a CO a 3 3 S ii S -^ c -Y »- 2 ^3 CO ,E3 .S s pq SMHlMajLia-OHAjI c« c -5 S - s ^* g o 5 »- ^ - O OS |H a a; o ;r '■^ *-■ O (U "Tj o llCJ (D ° cj O O '^ ^ OJ C 'O > (U N M a a; OW 'T? 3 fc O ©eg p 3 h: 3 3 S o X o tfj :: 3 %> T" (L) U3 a> 3 :rt l1 J:; ij O 3 -^ «5 ^ « 7^ .ti vh P ^ 000^ oooco o CO Cm S O u ^ .'^13 3 ^ s - (55 a 3 -3 2 p. PQ ^ > tn ^ X! < = ,'§ u - 3 '. •^ n i2 o suamniig |BiDgi;jY jo suoi;noB^ 276 e;ngine:e)ring che;mistry The Latin word Asphaltum is probably derived from the Greek word Aspholtos. Homer used it and in its Latin form gave the name to the lake of Palestine, known as the Dead Sea {Locus As- phalities) and anciently as the source of the bitumen, called bitumen judiacum. Aristotle called it Asphaltos and Xenophon 400 B. c, describes the wall of Media as built of burnt bricks laid in asphalt. In later years the term asphalt and bitumen have been used rather indiscriminately, incurring more or less con- fusion in the literature of the subject. This asphalt is obtained from the island of Trinidad and occurs in two characteristic deposits. First, the lake asphalt, and, second, the Trinidad land asphalt. The difference between them consists in the fact that the lake asphalt contains more of the cementing material (petrolene) than the land asphalt which contains more of harder bitumen (asphal- tene) — and also more earthy matter. The crude Trinidad asphaltum, analyzed in my laboratory, gave the following result : Per cent. ^^^'o\en^ . ^^^ Asphaltene ) Organic matter . . > « ^ Non-bituminous... ^ Inorganic residue 39-24 Total 100.00 The following analyses were made by Booth, Garrett and Blair, consulting chemists, Philadelphia, upon various asphalts.^ Trinidad AvSphalt (Refined). Specific gravity 1.37 Per cent. Petrolene 70.12 Asphaltene 25.13 Retine ( See page 277) 3.25 Non-bituminous \ Organic matter ) ^ Total 100.00 1 Report to the Citizens' Municipal lycague, Philadelphia. ENGINEERING CHEMISTRY 277 Bermudez Asphalt (Refined). Specific gravity 1,09 Per cent. Petrolene 76.90 Asphaltene 21,08 Retine 1.02 Non-bituminous ^ Organic matter \ ^'^ Total 100.00 Mexican Asphalt (Naturae Deposit). Specific gravity 1.06 Per cent. Petrolene 68.40 Asphaltene 29.78 Retine^ Non-bituminous ^ Organic matter \ • ^'^^ Total 100.00 Uvalde Asphalt (Refined).^ Petrolene 71.78 Asphaltene 28.01 Non-bituminous ^ Organic matter ^ ' ^'^^ Total 100.00 Cuban Asphalt. Petrolene 25.46 Asphaltene 54-41 Total bitumen 79.87 Mineral matter 20.13 Dead Sea Asphalt. Per cent. Petrolene 3509 Asphaltene 63.18 Total bitumen ^,8.27 Mineral matter 1.73 1 Retine occurs in some asphalts, but in small amounts only, soluble in alcohol. Its composition varies. De Smedt gplves its formula as Ci 78.84 per cent., Hj 19.22 per cent., Si 10.78. This compound is not desirable in asphalt. Consult: Jour. Franklin Inst., 1901, p. 50; Dana's Mineralogy (Retinellite), pp. 748-749. ^ Analysis made in the chemical laboratory of Stevens Institute of Technology. 278 ^NGINEJERING CHE:mISTRY Rock asphalts and sand asphalts usually contain but small amounts of asphalt, and often are used directly for street pav- ing. The celebrated French Val de Travers Rock Asphalt con- tains as follows : Per cent. Petrolene 8.52 Asphaltene 3.92 Total bitumen 12.44 Mineral matter 87.56 Seyssei. Rock Asphai^t.^ Petrolene 7.48 Asphaltene 4.32 Total bitumen 11.80 Mineral matter 88.20 Utah Asphai^t Rock (Limestone). Petrolene 31.20 Asphaltene 12.60 Total bitumen 43.80 Mineral matter 56.20 Caueornia Sand-Rock Asphalt. Petrolene 1 1.32 Asphaltene 3.81 Total bitumen 15.13 Mineral matter 84.87 Texas Sand Asphai,t (Anderson County). Petrolene 12.09 Asphaltene 1 1.25 Total bitumen 23.34 Mineral matter 76.66 From the Uvalde bituminous limestone (Texas) is obtained "litho-carbon," used in varnish making and so largely for insu- lating purposes. Asphalt is a subject of increasing interest and value to civil and municipal engineers. Its use in construction work and allied industries is advancing in a most rapid manner not only in amount but in variety. This increase in municipal use is com- paratively of recent date, since the first use of asphalt for street paving in this country was in Newark, N. J., in 1870. ^ From the Rhone Valley, France, between the towns of Bellegrade and Seyssel. EJNGINEERING CHE)MISTRY 279 AsPHAi^T Pave:me:nt Mixture). Asphalt as used in pavement mixtures is usually mixed with a more liquid bitumen or liquid substitute, to reduce its brittleness and to increase its adhesiveness to the mineral portion of the pavement mixture. The pavement mixture generally consists of asphaltic cement, 9 to 13 per cent., sand 83 to 70 per cent., and pulverized rock 5 to 15 per cent. The proportions of these in- gredients shall be determined by weight, and depend upon their kind and quality and the traffic upon the street. The mixture should conform to the following requirements, viz. : It shall be homogeneous and tenacious, free from brittleness at ordinary- temperatures, and not be too soft to sustain the traffic in hot weather nor so hard as to be brittle in cold weather ; it must not contain less than 9 per cent, of total bitumen nor more than 12 per cent, of total bitumen, including the flux, that will dissolve in carbon bisulphide. The flux in use for this purpose is uni- formly a heavy residuum prepared by the removal of the lighter portions of petroleum by distillation. These residues naturally vary in character in the same way that the petroleums do from which they have been derived. The oils from which residuums or fluxes are prepared for use in the United States are the paraffine petroleums from the Eastern Ohio, Kentucky, Kansas, Oklahoma and Colorado fields, the asphaltic petroleum, from California, and the petroleum of mixed character from Texas, containing both paraffine and asphaltic hydrocarbons. Typicai. PARArriNE Fi.ux of 1907. Manufacturer, Solar Refining Co., Lima, O. Specific gravity 0-943 Flash point 455° F- Volatile 7 hours at 212° F o.i per cent. " " " " 235° F. (dry sample) 0.2 *' Residue at 78° F crystalline, slow flow- Bitumen insol. 88° naphtha, pitch 2.7 per cent. Per cent. sol. bit. residues by H2SO4 24.9 " Paraffine scale 6.4 " Fixed carbon 2.8 " 280 ENGINEERING CHEMISTRY CaI^IEORNIA ASPHAI.TIC PETROLEUM RESIDUUM. The petroleums of California are characterized by the fact that the residue left on distillation, if the latter is carried suf- ficiently far, is a solid bitumen resembling asphalt. The oil is said, on this account, to have an asphaltic base. If the distilla- tion is suspended at a point where the residue does not solidify on cooling but remains liquid, like a heavy and dense natural maltha, the material known as California flux is obtained which has been in use in the paving industry to a very considerable extent on the Pacific Coast. Properties : Trade Name, No. 2. Specific gravity dried at 212° F. 1.002 Flash test (N. Y. State oil tester) 254° F. r Dry substances — J Loss at 325° F. ( 7 hours) 5.0 per cent. 325 F. 1 Character of residue smooth (^ Penetration of residue at 78° F. soft Loss 400° F., 7 hours ( fresh sample) 16.7 per cent. o _> -, Character of residue smooth 400 F. I I Penetration of residue at 78° F. soft Bitumen soluble in CS2, air temperature 99.9 per cent. Difference o. i " Mineral matter 0.0 " 1 00.0 Bitumen insoluble in 88° naphtha, air temp., pitch • . 7.6 per cent. Per cent, of soluble bitumen removed by H2SO4 48.3 •• Per cent, of total bittimen as saturated hydrocarbons 47.9 Per cent, of solid paraffines 0.0 Fixed carbon 6.0 Methods for the Examination of Bituminous Road Materials.^ CivASSlFlCATlON OF BITUMINOUS ROAD MaTE^RIAI^S. For the purpose of examination bituminous road materials may be classified under the following headings : 1. Petroleums and petroleum products, including residual petroleums, fluxes, oil-asphalts, and fluxed or cut-back oil- asphalts. 2. Malthas. ^ Bulletin No. 38, U. S. Dept. Agriculture. DNGINEE^RING CHEMISTRY 281 3. Asphalts and other solid native bitumens, and asphaltic ce- ments produced by fluxing them. 4. Tar and tar products. 5. Mixtures of tar with petroleum or asphalt products, bitum- inous emulsions, and factitious asphalts. 6. Bituminous aggregates, including rock asphalts or bitum- inous rocks, bituminous concrete and asphalt or other bituminous topping. Sche^me: of Examination. All petroleum, maltha, and solid native bitumen products are subjected to the following tests: Specific gravity. Volatilization at 163° C. Bitumen soluble in carbon disulphide. Bitumen insoluble in 86° B. paraffine naphtha. Fixed carbon. Th^ SoHM^R HyDRGMEJTER for ASPHAI.TS. This is graduated from 0.85 to 1.3 at 25° C, as recommended by the committee of the American Society of Civil Engineers. The principal feature of the method is to allow the asphalt to chill in a small cylindrical vessel, which is divided into two parts, the lower part, or cup, holding exactly 10 cc, and the upper part, or sleeve, being removable from the cup by the connecting thread. The entire vessel is filled with melted asphalt, and heated for a short time at a temperature a trifle above the melting point in order to thoroughly remove air bubbles or traces of water, and after the surface is clear, the vessel is allowed to cool, at first in air temperature (in order to avoid sudden contraction and hence separation of the asphalt from the sides of the tube) and then in water of the desired temperature, which in most cases will be 60° F. The sample should be left in water a suflicient time (about Yz hour) to thoroughly adopt its temperature, and after it has reached it, the instrument is wiped dry and the upper extension part or ''sleeve" taken off. If the asphalt is so hard that it renders the unscrewing difficult, the upper part should be warmed with a Bunsen burner. 282 ENGINEERING CHEMISTRY When the "sleeve" has been removed the asphalt, which extends above the level of the lower tube or cup, is cut off with a broad knife. The cup will then contain exactly lo cc. of asphalt at 60° F. The quantity can be directly weighed out on an analytical bal- v^^V II Fig. 48. — Sohmer Hydrometer. ance, and the specific gravity ascertained by dividing the number of grams of asphalt by 10. The following method, however, simplifies the procedure : The cup (A) is filled flush, as described above; then the cover (B) is slid on it from the side, and fastened by a flange (C). The cup and its contents are then suspended from the Engine:ering chi;mistry 283 hydrometer (Fig. 48), and the whole instrument is placed in a jar containing water of 60° F. If any air bubbles should form on the instrument, it should be quickly twisted once or twice to allow these bubbles to escape. The specific gravity can then be read directly on the stem of the hydrometer without correction. The method can be applied for asphalt, road oils, tars, wax, etc.; the instrument having a range of 0.850 to 1.700 specific gravity. Volatilization Test. Equipment. I constant-temperature hot-air oven with rubber tubing. (Fig. 49.) 1 thermo-regulator. 2 chemical thermometers reading from — 10° C. to 250° C. I tin box, 6 centimeters in diameter by 2 centimeters deep. I analytical balance, capacity 100 grams, sensitive to o.i milligram. Mejthod. The object of the volatilization test is to determine the per- centage of loss which the material undergoes when 20 grams in a standard-sized container are subjected to a uniform tempera- ture of 163° C. for 5 hours, and also to ascertain any changes in the character of the material due to such heating. The oven shown in Fig. 49, known as the New York Testing Laboratory oven, is used by the Office of Public Roads, although any other form may be used that will give a uniform tempera- ture throughout all parts where samples are placed. The bulb of one of the thermometers is immersed in a sample of some fluid, non-volatile bitumen, while the other is kept in air at the same level. The first thermometer serves to show the tempera- ture of the samples during the test, while the latter gives prompt warning of any sudden changes in temperature due to irregulari- ties in the gas pressure, etc. Before making the test the interior of the oven should show a temperature of 163° C. as registered by the thermometer in air. The tin box is accurately weighed after carefully wiping with a towel to remove any grease or dirt. About 20 grams of the material to be tested is then placed in the box. The material 284 ENGINEEJRING CHE:mISTRY may then be weighed on a rough balance, if one is at hand, after which the accurate weight, which should not vary more than 0.2 gram from the specified amount, is obtained. It may be necessary to warm some of the material in order to handle it conveniently, after which it must be allowed to cool before deter- mining the accurate weight. Fig. 49. — New York Testing L,aboratory Oven. The sample should now be placed in the oven, where it is allowed to remain for a period of 5 hours, during which time the temperature as shown by the thermometer in bitumen should not vary at any time more than 2° C. from 163° C. The sample is then removed from the oven, allowed to cool, and reweighed. From the difference between this weight and the total weight before heating, the percentage of loss on the amount of material taken is calculated. The general appearance of the residue should be noted, espe- cially with regard to any changes which the material may have ENGINEEiRING CHJjMISTRY 285 undergone. Some relative idea of the amount of hardening which has taken place may be obtained from the results of a float or penetration test made on the residue, as compared with the results of the same test on the original sample. It is also frequently desirable to make the specific gravity and other tests on the resi- due for the purpose of identifying or ascertaining the character of the base used in the preparation of cut-back products. Before any tests are made on the residue, it should be melted and thor- oughly stirred while cooling. Use; of the; VoIvATilization Te;st. The volatilization test, as above described, is made on prac- tically all bitumens with the exception of tars, for which the dis- tillation test answers a similar purpose. The test is also fre- quently made at 105° C. for 5 hours, and with products contain- ing small amounts of water it is usually necessary to make a test at the lower temperature before the material can be heated at 163° C. without foaming over. In the case of emulsions it is customary to determine the loss on a 20-gram sample at room temperature for 24 hours, after which the sample is heated at 105° C. for 5 hours. This additional loss is obtained and all determinations are made on the dried residue and reported accordingly. The volatilization test is also occasionally made at 205° C. for 5 hours on a fresh sample in order to show the effect of this higher temperature as compared with the results at 163° C. Determination of Bitumen Soluble in Carbon Bisulphide. Equipment. I 100 cc. Erlenmeyer flask. I 500 cc. flask with side neck for filtering under pressure. I rubber stopper with one hole. I filter tube, 3.9 centimeters, inside diameter. I platinum gooch crucible. 1 piece of seamless rubber tubing, about 3 centimeters in diameter and 3 centimeters long. 50 grams of long fiber amphibole asbestos. 2 wash bottles ; i for solvent, i for water. I Bunsen burner. 286 e;ngine:e:ring chemistry I platinum triangle. I iron tripod. I drying oven. I desiccator with calcium chloride. I thermometer reading from — io° C. to iio° C. I vacuum pump and connections. I analytical balance, capacity lOO grams, sensitive to o.i milligram. Method. This test consists in dissolving the bitumen in carbon disulphide and recovering any insoluble matter by filtering the solution through an asbestos felt. The form of gooch crucible best adapted for the determination is 4.4 centimeters wide at the top, tapering to 3.6 centimeters at the bottom, and is 2.5 centimeters deep. For preparing the felt the necessary apparatus is arranged as shown in Fig. 50, in which a is the filtering flask, b a rubber «f-- d— Fig. so. — Apparatus for determining soluble bitumen. stopper, c the filter tube, and d a section of rubber tubing which tightly clasps the gooch crucible e. The asbestos is cut with scissors into pieces not exceeding i centimeter in length, after which it is shaken up with just sufficient water to pour easily. The crucible is filled with the suspended asbestos, which is Engine;e;ring chemistry 287 allowed to settle for a few moments. A light suction is then applied to draw off all the water and leave a firm mat of asbestos in the crucible. More of the suspended material is added, and the operation is repeated until the felt is so dense that it scarcely transmits light when held so that the bottom of the crucible is between the eye and the source of light. The felt should then be washed several times with water, and drawn firmly against the bottom of the crucible by an increased suction. The crucible is removed to a drying oven for a few minutes, after which it is ignited at red heat over a Bunsen burner, cooled in a desiccator and weighed. From 2 to 3 grams of bitumen or about 10 grams of an asphalt topping or rock asphalt is now placed in the Erlenmeyer flask, which has been weighed previously, and the accurate weight of the sample is obtained. One hundred cubic centimeters of chem- ically pure carbon disulphide is poured into the flask in small portions, with continual agitation, until all lumps disappear and nothing adheres to the bottom. The flask is then corked and set aside for 15 minutes. After being weighed, the gooch crucible containing the felt is set up over the dry pressure flask, as shown in Fig. 50, and the solution of bitumen in carbon disulphide is decanted through the felt without suction by gradually tilting the flask, with care not to stir up any precipitate that may have settled out. At the first sign of any sediment coming out, the decantation is stopped and the filter allowed to drain. A small amount of carbon disulphide is then washed down the sides of the flask, after which the pre- cipitate is brought upon the felt and the flask scrubbed, if neces- sary, with a feather or ''policeman," to remove all adhering material. The contents of the crucible are washed with carbon disulphide, until the washings run colorless. Suction is then applied until there is practically no odor of carbon disulphide in the crucible, after which the outside of the crucible is cleaned with a small amount of the solvent. The crucible and contents are dried in the hot air oven at 100° S. for about 20 minutes, cooled in a desiccator, and weighed. If any appreciable amount 288 ENGINEERING CHEMISTRY of insoluble matter adheres to the flask, it should also be dried and weighed, and any increase over the original weight of the flask should be added to the weight of insoluble matter in the crucible. The total weight of insoluble material may include both organic and mineral matter. The former, if present, is burned off by ignition at a red heat until no incandescent particles remain, thus leaving the mineral matter or ash, which can be weighed on cooling. . The difference between the total weight of material insoluble in carbon disulphide and the weight of substance taken equals the total bitumen, and the percentage weights are calcu- lated and reported as total bitumen, and organic and inorganic matter insoluble, on the basis of the weight of material taken for analysis. This method is quite satisfactory for straight oil and tar prod- ucts, but where natural asphalts are present it will be found practically impossible to retain all of the finely divided mineral matter on an asbestos felt. It is, therefore, generally more accu- rate to obtain the result for total mineral matter by direct igni- tion of a I -gram sample in a platinum crucible or to use the result for ash obtained in the fixed carbon test. The total bitu- men, is then determined by deducting from lOO per cent, the sum of the percentages of total mineral matter and organic matter insoluble. If the presence of a carbonate mineral is suspected, the percentage of mineral matter may be most accurately obtained by treating the ash from the fixed carbon determination with a few drops of ammonium carbonate solution, drying at ioo° C, then heating for a few minutes at a dull red heat, cooling, and weighing again. When difliculty in filtering is experienced — for instance, when Trinidad asphalt is present in any amount — a period of longer subsidence than 15 minutes is necessary, and the following method proposed by the committee on standard tests for road materials of the American Society for Testing Materials is recommended : From 2 to 15 grams (depending on the richness in bitumen of the substance) is weighed into a 150 cc. Erlenmeyer flask, the tare of which has been previously ascertained, and treated with 100 cc. of carbon disul- phide. The flask is then loosely corked and shaken from time to time ENGINKERING CHEJMISTRY 289 until practically all large particles of the material have been broken up, when it is set aside and not disturbed for 48 hours. The solution is then decanted off into a similar flask that has been previously weighed, as much of the solvent being poured off as possible without disturbing the residue. The first flask is again treated with fresh carbon disulphide and shaken as before, when it is put away with the second flask and not dis- turbed for 48 hours. At the end of this time the contents of the two flasks are carefully decanted off upon a weighed gooch crucible fitted with an asbestos filter, the contents of the second flask being passed through the filter first. The asbestos filter shall be made of ignited long-fiber amphibole, packed in the bottom of a gooch crucible to the depth of not over % inch. After passing the contents of both flasks through the filter, the two residues are shaken with more fresh carbon disulphide and set aside for 24 hours without disturbing, or until it is seen that a good subsidation has taken place, when the solvent is again decanted oft" upon the filter. This wash- ing is continued until the filtrate or washings are practically colorless. The crucible and both flasks are then dried at 125° C. and weighed. The filtrate containing the bitumen is evaporated, the bituminous residue burned, and the weight of the ash thus obtained added to that of the residue in the two flasks and the crucible. The sum of these weights deducted from the weight of substance taken gives the weight of bitumen extracted. Use of Totai, Bitumen Determination. This determination is made on all classes of bituminous prod- ucts. In the analysis of tars the organic matter insoluble is commonly know^n and reported as "free carbon." Determination of Bitumen Insoluble in Paraffine Naphtha. Equipment. The apparatus is the same as for bitumen soluble in carbon disulphide. Method. This determination is made in the same general manner as the total bitumen determination except that lOO cc. of 86° B. parafhne naphtha is employed as a solvent instead of carbon disulphide. Considerable difficulty is sometimes experienced in breaking up some of the heavy semi-solid bitumens; the surface of the ma- terial is attacked, but it is necessary to remove some of the in- ^ Proc. Am. Soc. for Testing Materials, 1909, Vol. IX, p. 221. 19 290 ENGINEERING CHEMISTRY soluble matter in order to expose fresh material to the action of the solvent. It is, therefore, advisable to heat the sample after it is weighed, allowing it to cool in a thin layer around the lower part of the flask. If difficulty is still experienced in dissolving the material, a rounded glass rod will be found convenient for break- ing up the undissolved particles. Not more than one-half of the total amount of naphtha required should be used until the sample is entirely broken up. The balance of the 100 cc. is then added, and the flask is twirled a moment in order to mix the contents thoroughly, after which it is corked and set aside for 30 minutes. In making the filtration the utmost care should be exercised to avoid stirring up any of the precipitate, in order that the filter may not be clogged and that the first decantation may be as com- plete as possible. The sides of the flask should then be quickly washed down with naphtha and, when the crucible has drained, the bulk of insoluble matter is brought upon the felt. Suction may be applied when the filtration by gravity almost ceases, but should be used sparingly, as it tends to clog the filter by packing the precipitate too tightly. The material on the felt should never be allowed to run entirely dry until the washing is completed, as shown by the colorless filtrate. When considerable insoluble matter adheres to the flask, no attempt should be made to remove it completely. In such cases the adhering material is merely washed until free from soluble matter, and the flask is dried with the crucible at 100° C. for about i hour, after which it is cooled and weighed. The percentage of bitumen insoluble is reported upon the basis of total bitumen taken as icx). The difference between the material insoluble in carbon disul- phide and in the naphtha is the bitumen insoluble in the latter. Thus, if in a certain instance it is found that the material in- soluble in carbon disulphide amounts to i per cent, and that 10.9 per cent, is insoluble in naphtha, the percentage of bitumen in- soluble would be calculated as follows: Bitumen insoluble in naphtha 10. 9 — i 9.9 oA ^ 1 U-. — = — = 10 per cent. Total bitumen 100 — i 99 i:ngine;e:ring chemistry 291 Us^ OF Naphtha Insolubi^k Bitumejn Determination. This test is made on all petroleums, malthas, asphalts, and other solid native bitumens and their products. It should be noted that petroleum naphthas are by no means definite compounds, but are composed of a number of hydro- carbons which vary in character and quantity according to the petroleum from which they have been distilled. Their solvent powers also vary greatly. Thus naphthas produced from as- phaltic petroleums, consisting mainly of naphthene and poly- methylene hydrocarbons, are much more powerful solvents of the heavier asphaltic hydrocarbons than are the paraffine naphthas. The density of the naphtha also affects its solvent power, for those of high specific gravity dissolve the heavier hydrocarbons more readily than those of lower specific gravity. As the main object of this test is to separate the heavier hydrocarbons of an asphaltic nature from the paraffine hydrocarbons, a paraffine sol- vent should be employed, and for ordinary purposes a paraffine naphtha of 86° B. gravity, distilling between 40° C. and 65° C, has been found to be readily obtainable and fairly satisfactory. The solvent action of 88° B. naphtha is a little lower, and there- fore preferable, but it can not be as readily obtained. The determination is also frequently made with heavier naph- thas, such as 66° B. and y2'^ B., for the purpose of grading the character of the bitumen present in the compound. A report should therefore always distinctly state the gravity and character of the solvent used. Determination of Bitumen Insoluble in Carbon Tetrachloride. Equipment. The apparatus is the same as for bitumen soluble in carbon disulphide. Method. This determination is conducted in exactly the same manner as described under "Determination of bitumen soluble in carbon disulphide," using 100 cc. of chemically pure carbon tetrachloride in place of carbon disulphide. The percentage of bitumen insoluble is reported upon the basis 292 ENGINEERING CHEMISTRY of total bitumen taken as 100, as described under "Determination of bitumen insoluble in paraffine naphtha." Use of Determination of Bitumen Insoeubee in Carbon Tetracheoride. The bitumen insoluble in carbon tetrachloride, but soluble in carbon disulphide, is commonly known as "carbenes." The test is occasionally made on petroleums, asphalts, and other solid native bitumens and their products, for the purpose of identifi- cation, or when there is any reason to suspect that the material under examination has been injured by overheating during the process of manufacture. Determination of Fixed Carbon. Equipment. I iron ring support (ring 7.5 centimeters in diameter). I platinum triangle. I Bunsen burner and rubber tubing. I platinum crucible with a tight-fitting cover (weight complete, from 20 to 30 grams). I crucible tongs. I desiccator with calcium chloride. I analytical balance, capacity 100 grams, sensitive to o.i milligram. Method. This determination is made in accordance with the method de- scribed for coal in the Journal of the American Chemical vSociety, 1899, volume 21, page 11 16. One gram of the material is placed in a platinum crucible weighing from 20 to 30 grams and having a tightly fitting cover. It is then heated for 7 minutes over the full flame of a Bunsen burner, as shown in Fig. 15. The crucible should be supported on a platinum triangle with the bottom frohi 6 to 8 centimeters above the top of the burner. The flame should be fully 20 centimeters high when burning freely, and the determ- ination should be made in a place free from drafts. The upper surface of the cover should burn clear, but the under surface should remain covered with carbon, excepting in the case of some of the more fluid bitumens, when the under surface of the cover may be quite clean. ENGINEE^RING CHEJMISTRY 293 The crucible is removed to a desiccator and when cool is weighed, after which the cover is removed, and the crucible is placed in an inclined position over the Bunsen burner and ignited until nothing but ash remains. Any carbon deposited on the cover is also burned off. The weight of ash remaining is deducted from the weight of the residue after the first ignition of the sample. This gives the weight of the so-called fixed or residual carbon, which is calculated on a basis of the total weight of the sample, exclusive of mineral matter. If the presence of a car- bonate mineral is suspected, the percentage of mineral matter may be most accurately obtained by treating the ash with a few drops of ammonium carbonate solution, drying at ioo° C, then heating for a few minutes at a dull red heat, cooling and weighing. Use of Determination for Fixed Carbon. This determination is made on all bituminous products with the exception of tars, upon which reliable results cannot be obtained, owing to the error introduced by the presence of considerable quantities of free carbon. Determination of Paraffine Scale. Eqijipment. I 200 cc. (6-ounce) glass retort (for each determination). I 150 cc. Erlenmeyer flask, I 100 cc. Erlenmeyer flask. I 500 cc. (i6-ounce) flask, with side neck for filtering under pressure. I glass funnel 65 millimeters in diameter, stem 150 millimeters long. 1 freezing mixture reservoir. 2 rubber stoppers with hole. I analytical balance, capacity 100 grams, sensitive to o.i milligram. I rough balance, capacity i kilogram, sensitive to o.i gram. I wash bottle. I quart tin cup, seamless. I package C. S. & S. 9-centimeter hardened filter papers. I vacuum pump and connections. I glass crystallizing dish 50 millimeters in diameter. I steam bath. I desiccator with calcium chloride. I 4-inch steel spatula. I Bunsen burner with rubber tubing. I iron stand with retort clamp. 294 ENGINEERING CHEMISTRY Method. One hundred grams of the material under examination should be weighed into the tared glass retort and distilled as rapidly as possible to dry coke. The distillate is caught in a 150 cc. Erlen- meyer flask, the weight of which has been previously ascertained. During the early stages of distillation a cold, damp towel wrapped around the stem of the retort will serve to condense the distillate. After high temperatures have been reached, this towel may be removed. When the distillation is completed, the distillate is al- lowed to cool to room temperature and is then weighed in the flask. This weight minus that of the flask gives the weight of the total distillate. Fig. 51. — Apparatus for determining paraffine scale. Five grams of the well mixed distillate is then weighed into a 100 cc. Erlenmeyer flask and mixed with 25 cc. of Squibb's ether. Twenty-five cubic centimeters of Squibb's absolute alcohol is then added, after which the flask is packed closely in a freezing mix- ture of finely crushed ice and salt maintained at — 18° C. in a quart tin cup. After remaining 30 minutes in this mixture, the solution is quickly filtered through a No. 575 C. S. & S. 9 centi- ENGINEERING CHEMISTRY 295 meter hardened filter paper placed in a glass funnel, which is packed in a freezing mixture, as shown in Fig. 51. Vacuum should be employed to hasten filtration. The freezing-mixture reservoir (b) shown in Fig. 51 may be made by cutting in half a round glass bottle measuring approximately 120 millimeters in diameter and using the upper half in an inverted position. Any precipitate remaining on the paper should be washed until free from oil with about 50 cc. of a i to i mixture of Squibb's ether and absolute alcohol cooled to — 18° C. After the paper has been sucked dry, it should be removed from the funnel and the adhering paraffine scale should be scraped off into a weighed crystallizing dish and dried on a steam bath. The dish and contents should then be cooled in a desiccator and weighed. The weight of the parafiine scale so obtained, divided by the weight of the distillate taken and multiplied by the percentage of the total distillate obtained from the original sample, equals the percentage of the parafiine scale. Use 01^ Paraeeine ScaeE Determination. The parafiine scale determination may be made on all native bitumens and their products which are suspected of being of a parafiine nature. It is not an extremely accurate determination, however, and is seldom employed by the Ofiice of Public Roads. Penetration Test. The object of the penetration test is to ascertain the consistency of the material under examination by determining the distance a weighted needle will penetrate into it at a given temperature. There are two machines in use for this purpose : The Dow and the New York Testing Laboratory penetrometer. The latter is used by the Ofiice of Public Roads and is thus described by Mr. C. N. Forrest, chief chemist. New York Testing I^aboratory, in a communication to the editor : The consistency of asphalt cement or of a similar material is determined by the depth to which, under a definite load and dur- ing a given time a standard needle will penetrate. 296 DNGINEKRING CHEMISTRY An instrument especially designed for this purpose is shown in the accompanying illustration, Fig. 52. The base and foot casting A can be leveled by means of the thumb screws B, and is bored to lit the standard C and also the platen D, which by means of screw on shank of platen raises or Fig. 52. — Penetrometer. lowers the revolving disc E, on which is to be placed the sample of the material to be tested. The standard C carries a bracket F, adjustable as to elevation by thumb screws R, and also the bracket G, which on back carries the clockwork H, timing the duration of the test by half-second DNGINEEJRING CHEMISTRY 297 beats, and on the front the dial J, divided into 360 degrees, with the hand K marking the number of degrees, each of which repre- sents Yio miUimeter of penetration measured by rack on sliding gauge L, engaging in pinion on shaft which actuates the hand K. The pointer M serves as a marker for the pendulum P. The beveled edge mirror N, adjustable through universal joints, serves to reflect light on the sample under test. O is a plunger or brake which holds the needle bar, representing weight of 50 grams and superincumbent weight in place until pressed inward, which movement permits needle and weight to act upon test block without friction, and is easily operated by grasping the horns Q between two fingers and pressing brake head O with thumb. S is a weight of predetermined capacity, either 50 or 150 grams. A No. 2 R. J. Roberts needle has been selected as standard, and this is set in a short brass shaft for greater convenience in handling. Five seconds is the length of time the needle should be per- mitted to penetrate ; yy^ F. or 25° C. is the standard temperature. One hundred grams is the standard weight, which includes the weight of the needle. In order to test the stability of a cement at a higher or lower temperature than jy^ F. a lighter or heavier load may be re- quired, and for this purpose the penetrometer is supplied with adjustable weights which permit of the use of a 50, 100 or 200 gram load upon the needle as desired. At 32° F. a 200-gram load, while at 100° or 115° F. but 50 grams may be required, in order to permit a measurement within the compass of the scale on the dial. It is necessary that the substance to be tested should be of a uniform and standard temperature. Bituminous substances should present a fresh surface which has been melted not longer than a day before making the test, as exposure to the air and a deposit of dust soon harden the exterior sufficiently to affect the penetra- tion. It is usually sufficient to immerse the sample of bitumen in 298 e:ngine:e:ring chkmistry water maintained at yy^ F. for j/^ hour in order to bring it to a uniform temperature, but this time may have to be lengthened if the material is either very warm or cold when it is placed in the water. If a small room is available in which the temperature may be regulated to "j^j^ F. it will facilitate the operation. The substance should be melted and poured into a 2-ounce tin sample box, about 2^ inches in diameter and ^^ inch deep, cooled by first placing in ice water and then brought to the proper tem- perature for testing {j'j^ F.) by immersing in water maintained at this temperature for about ^ hour. It is then placed upon the revolving table E of the penetrometer, and raised until the surface of the sample just touches the point of the needle. The foot of the rack L should rest upon the rod carrying the weight and the needle. The position of the hand should be noted. Re- lease the rod by pressing the plunger O for a period of 5 seconds, then again move the rack ly down, until the foot rests upon the rod, and the difference in the reading on the dial will repre- sent the depth in tenths of a millimeter which the needle has penetrated the sample, or its consistency in degrees at the standard temperature. Float Test or Fluidity Test. The consistency or fluidity of bituminous binders is of great importance, both as a first consideration to insure selection of the appropriate type and during use, uniformity and perfect con- trol of the work, just as it is in the sheet asphalt paving industry. It is obvious that a binder intended for use as a surface dressing should be of quite a different consistency from one which is to be incorporated with stone to form the wearing body of the highway. No single preparation will answer all requirements. It has been proposed that a temperature of 90° F. be consid- ered as a normal standard at which to test the consistency of road binders of a bituminous nature, because that is a fair con- dition upon a road during suitable working weather, and a com- pound should be capable of being worked to some extent, at least, at this temperature, if it is to be mixed with cool stone or swept into the interstices of a roadbed. A much more fluid ma- Enginee:ring chemistry 299 terial is required for such use than in the older forms of bitu- minous pavements, the mixture for which is taken from a cen- tral plant in a hot condition and the standard forms of pene- trometers mentioned above are not available for regulating the consistency of the same. Tests of the Consistency of Bituminous Binders for Highways. Materials Refined water gas tar.. Crude coal-tar Tarvia Tarina Texas flux oil Cahf. flux oil Headley oil, No. 2 Headley oil, No. 3. • - • Headley oil, No. 4. • • . Genasco compound Genasco compound . . • • Genasco compound Standard Oil Co Standard Oil Co Standard Oil Co Standard Oil Co Standard Oil Co N. Y. T. I,. Viscosimeter at 90° F. Min. Sec. O O I O O 3 o 2 6 ID '69 II o o o o I 20 30 30 16 51 40 35 53 56 54 55 54 6 12 23 39 24 Engler viscosimeter. 100 cc. flow at 77° F. Min. Sec. Too Stiff Too Stiff Too stiff 17 10 Too stiff Too stiff Too stiff Too stiff Too stiff Too stiff Too stiff Too stiff 5 38 Too stiff Too stiff Too stiff Too stiff 130° F. Min. Sec. 9 46 9 8 Too stiff 3_ 18 Too stiff Too stiff Too stiff Too stiff Too stiff Too stiff Too stiff I 36 3 o Too stiff Too stiff Too stiff 250° F. Min. Sec. 50 30 2 34 32 14 6 30 42 8 30 Too stiff 9 28 32 38 56 18 26 1 N. Y. T. I,aby. test at i5o°F. = 2 min. 18 sec. The Engler viscosimeter is in general use for testing fluid com- pounds, such as oils, at any desired temperature, but is not avail- able for the highly cementitious and semi-fluid asphaltic materials now preferred by road engineers, except at temperatures above 200° F. To provide an instrument for controlling the consistency of semi-fluid compounds intended for road building the author has elaborated a simple form of viscosimeter, originally suggested by Mr. E. C. Wallace, now with the Warren Brothers Company, which fills the gap for substances between those which are sufli- ciently fluid for the Engler type of viscosimeter and the semi- solid cements heretofore regulated by penetration instruments. 300 e:ngine:e:ring che;mistry The apparatus which is made by Howard & Morse, Brooklyn, N. Y., consists of two parts, an aluminum float or saucer and a conical brass collar. The two parts are shown in the drawing (Fig. 53) and are made separately for reasons of economy, so f^^ h- z H Fig- 53- — Instrument for determining the consistency of road binders. that one or two of the floats will be sufficient for an indefinite number of brass collars. In using the apparatus the brass collar is placed upon a brass plate, the surface of which has been amalgamated, and filled with ENGINEERING CHEMISTRY 3OI the bitumen under examination, after it has been softened suffi- ciently to flow freely by gentle heating. The collar must be level full, and as soon as the bitumen has cooled sufficiently to handle it is placed in ice water at 41° F. for 15 minutes. It is then at- tached to a float and immediately placed upon the surface of the water, which is maintained at 90° F., or any other temperature desired. As the plug of bitumen in the brass collar becomes warm and fluid, it is gradually forced out of the collar, and as soon as the water gains entrance to the saucer the entire apparatus sinks be- low the surface of the same. The time, in seconds, elapsing between placing the apparatus on the water and when it sinks is determined most conveniently by means of a stop watch, and is considered as the consistency of the bitumen under examination. This device has been in general use in the New York Testing Laboratory for some time. An equipment of 12 brass collars, 2 aluminum saucers, a nursery refrigerator for ice water, and an open tank holding about i quart of water and heated by a Bunsen burner is sufficient for testing a great many samples. A thermostat in the water bath will assist in maintaining a constant temperature, From the data presented in the table on page 299 it will be observed that this device is available at 90° F. for testing almost any compound (bituminous) used in road building except light oils. The Engler viscosimeter for tests at 250° F. to 350° F. and the New York Testing Laboratory device at 90° F. provide satisfactory means for making these consistency tests. Ductility Test. This test, as applied to asphalts, asphaltic cements, or bitumens, measures the distance in centimeters through which a briquette of standard size can be pulled at a given speed and a given tem- perature before rupture takes place. Generally speaking, the more ductile the material the greater the cementing value. Apart 302 DNGINEE^RING CHE^MISTRY from the nature of the bitumen itself, the more important condi- tions affecting its ductiUty are : Purity. Consistency. Size and shape of briquette. Rapidity of pull. Temperature at which test is made. Purity. — If the bituminous material contains considerable quan- tities of inert or non-bituminous matter, it will ordinarily show a lower ductility than it would if these matters were removed. Consistency. — With the same kind of bitumen, the softer the consistency the greater will be the ductility. The machine generally used for determining the ductility^ of bitumens is the "Smith Horizontal Ductility Machine" and is thus described : GenErai, Description. This machine consists of a horizontal trough L, with suitable mechanism attached for pulling apart the two ends of a briquette D of standard size until rupture takes place. Means for regu- lating the speed of the machine and for measuring the distance between the two clips at the point of rupture are provided. Two different types of machine are made : A. The hand power machine. B. The motor driven machine. In the hand driven machine the revolution of a hand wheel H operates to move the traveling carriage. The speed is regulated by means of a clicking device K attached to the driving mechan- ism. When the wheel is rotated at such a speed that the clicks produced by the clicking device synchronize with a metronome set to beat 79 beats per minute, the traveling carriage of the machine will be moved forward at the standard speed for mak- ing ductility tests at yy° F. ; vi^., 5 centimeters per minute. In the motor driven machine the gearing is so arranged that when the motor is given the full current under which it is de- signed to operate ( volts amperes), the traveling car- riage will be moved forward at the rate of 5 centimeters per KNGINEJERING CHEJMISTRY 303 minute. Where special speeds are desired, suitable interchange- able gears will be provided, thus permitting variations in speed. Operation 01^ Machine. In operating the machine the box L, is first filled with water at the required temperature to such a depth that the briquettes while they are being pulled will be completely immersed in it. The hand wheel H of the machine is then rotated so as to bring the traveling carriage B down to the point where the dis- tance between the pins P on the shelves attached to the travel- ing carriage and to one end of the machine will be approximatel)' the same as the distance between the holes in the ends of the briquette molds D when they are set up as described hereafter. Fig. 54. — Electrically driven. The briquette is then placed in the machine by slipping the rings or holes in the ends of the mold over the pins previously men- tioned. The hand wheel is then rotated in such a manner as to draw the traveling carriage away from the end of the machine until it just begins to exert a tension upon the briquette. The measuring rule E attached to the side of the machine is then moved in the loosely fitting clip A which hold it until the zero mark on the scale is exactly opposite the pointer M on the traveling carriage of the machine. The machine is then operated 304 ENGINJ^ERING CHEMISTRY ^1 pa CD DNGINDERING CHEMISTRY 305 at the required speed and the traveling carriage draws apart the two ends of the briquette. When the thread produced by puUing out the briquette breaks, the distance through which the briquette has been pulled will be indicated by the position of the pointer on the measuring scale, which is divided into centimeters, and this distance, expressed in centimeters, is reported as the ductility of the material ex- amined. A thermometer reading to i° F. inserted in a cork, should be placed in the circular hole in the traveling carriage so as to regis- ter the temperature of the water during the test. During the operation of pulling, the temperature of the water should not vary more than ^° from the standard temperature. While this test is usually conducted at yy^ F., it may also be made at 32° F. When made at this latter temperature, extreme care must be taken not to fracture the briquette, as materials of this kind are frequently very brittle at low temperatures. For this reason it is customary to pull the briquettes apart at the rate of 34 centimeter per minute while testing at 32° F., instead of 5 centimeters per minute as employed when testing them at yy° F. The traveling carriage is provided with a slip nut attachment in order to facilitate bringing it back to its original position after a test has been made. By throwing back the clips on the ends of the slip nuts and opening them out, the carriages can be moved by hand without rotating the operating wheel. Standard Conditions for Making Ductility Test. Purity. — The asphalt, if necessary, must be purified as described elsewhere herein until it consists substantially of pure bitumen. Specifications usually define the ductility required for the pure bitumen. Where the specification calls for a ductility consider- ably lower than that possessed by the bitumens under examination, it is not always necessary to go through the purifying process. If the impure bitumen shows a ductility in excess of the specification requirements for pure bitumen, it is certain that the purified bi- tumen would comply with these requirements. Where the impure bitumen shows a lower ductility than called 3o6 e;ngine:e:ring chemistry for by the specifications for pure bitumen, it will of course be necessary to purify the bitumen under examination. Consistency. — Specifications usually call for a certain ductility at 50 penetration and sometimes provide for a variation from this consistency by calling for a higher ductility if the material is softer than 50 penetration and a lower ductility if the material is harder than 50 penetration. The generally allowed ratio of increase or decrease is 2 centimeters in ductility for every 5 points in penetration. This correction is only an approximate one. If it becomes necessary to soften the asphalt under examination and to bring it to a penetration of 50 at yy° F., this is done by melting the asphalt at a temperature of approximately 300° F., and thoroughly incorporating with it, while in the molten condition, sufficient flux or residuum to bring it to the desired consistency. The flux employed should preferably be that to which it is pro- posed to use in practice. The ductility test is then made on the mixture of asphalt and flux so prepared. Si^e of Briquette. — The dimensions of this are as follows: Length over all 7^ centimeters Distance between clips 3 centimeters Width at mouth of clips 2 centimeters Width at minimum cross section, halfway between clips i centimeter Thickness throughout i centimeter Rapidity of Pull. — For ductilities taken at /"/° F., 5 centimeters per minute. For ductilities taken at 32° F., }i centimeter per minute. Temperature. — For ordinary paving work, the standard tem- perature is yy° F. For special work, ductilities are sometimes taken at 32° F. Description of M01.D. The Dow form of briquette mold is shown below. After numerous trials, this form of briquette was adopted as being the most suitable. Briquettes of other forms, including those shaped like a rod, were discarded owing to the lack of homogeneity in some bitumens which rendered their pull very KNGINEJ^RING CH15MISTRY 307 irregular unless the briquette was so shaped that it would fail at Fig. 56. some one definite point of least cross section. The molds are made of brass and are in four pieces. Preparation of Briquette:. The molding of the briquette may be done as follows : The mold should be placed upon a brasb plate. To prevent the asphalt from adhering to the plate and the inner side of the two remov- able pieces of the mold, a and o', they should be well amal- gamated. The different pieces of the mold should be held to- gether in a clamp or by means of an India rubber band. The material to be tested is poured into the mold while in a molten state, a slight excess being added to allow for shrinkage on cooling. After the briquette is nearly cooled, it is smoothed off level by means of a heated palette knife. When cooled, the clamp is taken off and the two side pieces, a and a', removed, leaving the briquette of asphalt firmly attached to the two ends of the mold, which thus serve as clips. The briquette should then be immersed in water maintained at the required tempera- ture for at least 30 minutes or until the whole mass of bitumen is at that temperature. It is then pulled apart as described above. Preliminary Treatment of Asphalts. Different asphalts vary considerably in purity and character of bitumen. Any appreciable amount of mineral matter or inert bitumen will largely affect the ductility, and for this reason the 308 DNGINEKRING CHE:mISTRY test should be made on materials of approximately the same purity as well as consistency. Asphalts to be examined for ductility are therefore usually divided into three classes and are given a preliminary treatment as described below, depending upon their classification. 1. Asphalts containing over 96 per cent, of bitumen soluble in carbon disulphide and free from lumps of inert bitumen. 2. Asphalts containing less than 96 per cent, of bitumen soluble in carbon disulphide, in which the bitumen is homogene- ous, or nearly so, i. e., contains no lumps of inert bitumen. 3. Asphalts in which the bitumen is not homogeneous; i. e., containing lumps of hard bitumen which, although soluble in car- bon disulphide, are insoluble in the softer bitumen, even in a mol- ten condition, which forms part of the bitumen under examination. Asphalts coming under the first classification need no prelim- inary treatment other than softening as previously described until they have a consistency of 50 penetration at yy^ F. Asphalts coming under the second classification should be subjected to the following treatment. Sufiicient quantity of the refined hard asphalt to yield 150 grams of pure bitumen is treated with chemically pure carbon disulphide in an Erlenmeyer flask. After standing for 2 or 3 hours; the flask is shaken until none of the asphalt is seen adhering to the sides or bottom of the flask, after which it is set aside and allowed to stand 24 hours. The solvent is then carefully decanted from the residue into a second flask. The residue is again treated with the solvent, shaken, allowed to subside, and decanted as before. This is continued until the solvent is practically colorless or of a light straw color. The combined solutions, after standing at least 24 hours after the last addition of solution, should then be carefully decanted off and the solvent distilled until only sufiicient remains to keep the extracted bitumen liquid. The residue is then poured into a large evaporating dish and as much of the remaining solvent as possible evaporated off on a steam bath. To facilitate the re- moval of the last particles of carbon disulphide from the bitumen while on the steam bath, it should be stirred from time to time. ENGINEERING CHEMISTRY 309 After this treatment on the steam bath, y^ to i cc. of water should be incorporated into the bitumen and the heating continued over a burner until all foaming ceases, after which the dish contain- ing the bitumen should be placed in a hot air oven and kept at 300° F. for 30 minutes. While heating the extracted bitumen over the burner, it should be stirred constantly with a ther- mometer, and care exercised that the temperature never exceeds 3CX)° F. The extracted bitumen is brought to a consistency of 50 penetration at yy^ F. by the addition of sufficient flux, and is then ready for testing. Carbon disulphide containing an excess of sulphur must be avoided, as with certain asphalts this sulphur reacts upon them and lowers the ductility of the extracted bitumen. Where the asphalt under examination is completely soluble in chemically pure benzole, this solvent may advantageously be substituted for carbon disulphide, as it is less inflammable and is less liable to reduce the ductility of the extracted bitumen. Asphalts coming under the third classification should be treated as follows : The asphalt, asphalt cement, or bitumen is heated in an air bath at a temperature between 300° F. and 350° F., together with a 20-mesh sieve and a 50-mesh sieve. When the material is in a thoroughly molten condition, it is first strained through the heated 20-mesh sieve and afterwards through the heated 50- mesh sieve. The molten material must not be forced through the sieves, but must run through by gravity alone. If the material thus obtained contains less than 96 per cent, of bitumen soluble in carbon disulphide, it must be still further purified by the method described for asphalts coming under the second classification; otherwise it may be softened with flux to the proper consistency and is then ready for testing. Volatilization Test. Equipment. I constant-temperature hot-air oven with rubber tubing. 1 thermo-regulator. 2 chemical thermometers reading from — 10° C. to 250° C. I tin box, 6 centimeters in diameter by 2 centimeters deep. I analytical balance, capacity 100 grams, sensitive to o.i milligram. 3io ^nginehjring chemistry Method. The object of the volatilization test is to determine the per- centage of loss which the material undergoes when 20 grams in a standard-sized container are subjected to a uniform temperature of 163° C. for 5 hours, and also to ascertain any changes in the character of the material due to such heating. The oven known as the New York Testing L^aboratory oven, is used by the Office of Public Roads, although any other form may be used that will give a uniform temperature throughout all parts where samples are placed. The bulb of one of the ther- mometers is immersed in a sample of some fluid, non-volatile bitumen, while the other is kept in air at the same level. The first thermometer serves to show the temperature of the samples during the test, while the latter gives prompt warning of any sudden changes in temperature due to irregularities in the gas pressure, etc. Before making the test the interior of the oven should show a temperature of 163° C. as registered by the thermometer in air. The tin box is accurately weighed after carefully wiping with a towel to remove any grease or dirt. About 20 grams of the material to be tested is then placed in the box. The material may then be weighed on a rough balance, if one is at hand, after which the accurate weight, which should not vary more than 0.2 gram from the specified amount, is obtained. It may be necessary to warm some of the material in order to handle it conveniently, after which it must be allowed to cool before determining the accurate weight. The sample should now be placed in the oven, where it is al- lowed to remain for a period of 5 hours, during which time the temperature as shown by the thermometer in bitumen should not vary at any time more than 2° C. from 163° C. The sample is then removed from the oven, allowed to cool, and reweighed. From the difference between this weight and the total weight before heating, the percentage of loss on the amount of material taken is calculated. The general appearance of the residue should be noted, es- pecially with regard to any changes which the material may have Engine;ering chemistry 311 undergone. Some relative idea of the amount of hardening which has taken place may be obtained from the results of a float or penetration test made on the residue, as compared with the re- sults of the same test on the original sample. It is also frequently desirable to make the specific gravity and other tests on the residue for the purpose of identifying or ascertaining the char- acter of the base used in the preparation of cut-back products. Before any tests are made on the residue, it should be melted and thoroughly stirred while cooling. USK OF THE VoivATlI.IZATlON TDST. The volatilization test, as above described, is made on prac- tically all bitumens with the exception of tars, for which the dis- tillation test answers a similar purpose. The test is also fre- quently made at 105° C. for 5 hours, and with products containing small amounts of water it is usually necessary to make a test at the lower temperature before the material can be heated at 163° C. without foaming over. In the case of emulsions it is customary to determine the loss on a 20-gram sample at room temperature for 24 hours, after which the sample is heated at 105° C. for 5 hours. This additional loss is obtained and all determinations are made on the dried residue and reported accordingly. The volatilization test is also occasionally made at 205° C. for 5 hours on a fresh sample in order to show the effect of this higher temperature as compared with the results at 163° C. Determination of Flash and Burning Points. Equipment. I New York State Board of Health oil tester with Bunsen burner. (Fig. 57.) I chemical thermometer reading from 0° C. to 400° C. I piece of 6-millimeter glass tubing, 6 centimeters in length, one end of which has been drawn to a i -millimeter opening. Soft rubber tubing for gas connection. .Me:thod. While for all ordinary purposes the open-cup method of de- termining the flash and burning points of bituminous road ma- terials is sufficiently accurate, the closed-up method described below is to be preferred. 312 ENGINEERING CHEMISTRY The oil tester shown in Fig. 57 consists of a copper oil cup a of about 300 cc. capacity, which is heated in a water or oil bath by a small Bunsen flame. The cup is provided with a glass cover, carrying a thermometer, and a hole for inserting the testing flame. Ihe testing flame is obtained from a jet of gas passed through the piece of glass tubing and should be about 5 millimeters in length. Fig. 57. — N. Y. State Tester. The flash test is made as follows : The oil cup should first be removed and the bath filled with water or cottonseed oil. The oil may always be used and is necessary for bitumens flashing at a temperature of over 100° C. The oil cup should be replaced and filled with the material to be tested to within 3 millimeters of the flange joining the cup and the vapor chamber above. The glass cover is then placed on the oil cup and the thermometer so adjusted that its bulb is just covered by the bituminous material. ENGINEERING CHEMISTRY 313 r The Bunsen flame should be appUed in such a manner that the temperature of the material in the cup is raised at the rate of about 5° C. per minute. From time to time the testing flame is inserted in the opening in the cover to about half way between the surface of the material and the cover. The appearance of a faint bluish flame over the entire surface of the bitumen shows that the flash point has been reached and the temperature at this point is taken. The burning point of the material may now be obtained by re- moving the glass cover and replacing the thermometer in a wire frame. The temperature is raised at the same rate and the material tested as before. The temperature at which the material ignites and burns is taken as the burning point. At the conclusion of this test the flame should not be blown out for danger of splashing the hot material. A metal cover or ex- tinguisher should be employed for this purpose by placing it over the ignited material. Melting Point Determination. Equipment. I iron tripod. I Bunsen burner and tubing. I piece of wire gauze 10 centimeters square. I 800 cc. Jena glass beaker, low form. I 400 cc. Jena glass beaker, tall, without lip. I iron ring support (ring 7.5 centimeters in diameter) and burette clamp. I metal cover. I object glass. I piece of wire (No. 12 Brown & Sharpe gauge) 20 centimeters in length, bent. I chemical thermometer reading from 0° C. to 250° C. I cubical brass mold. I large metal kitchen spoon. I steel spatula. Method. Since bitumens are mixtures of various organic compounds, they can have no true melting point, but an arbitrary method for determining the so-called melting point of these materials sufli- 314 ENGINEERING CHEMISTRY ciently solid to maintain their form for some time under normal conditions is of value as a means of identification and for control work. A number of methods have been tried, but the following has been selected as the most convenient and accurate for such materials. The material under examination is first melted in the spoon by the gentle application of heat until sufficiently fluid to pour readily. Care must be taken that it suffers no appreciable loss by volatilization. It is then poured into the ^^-inch brass :ubical mold, which has been amalgamated with mercury and which is placed on an amalgamated brass plate. The brass may be amalgamated by washing it first with a dilute solution of mercuric chloride or nitrate, after which the mercury is rubbed into the surface. By this means the bitumen is, to a considerable extent, prevented from sticking to the sides of the mold. The hot material should slightly more than fill the mold and, when cooled, the excess may be cut off with a hot spatula. After cooling to room temperature, the cube is removed from the mold and fastened upon the lower arm of a No. 12 wire (Brown & Sharpe gauge) bent at right angles and suspended beside a thermometer in a covered Jena glass beaker of 400 cc. capacity, which is placed in a water bath, or, for high tempera- tures, a cotton seed oil bath. The wire should be passed through the center of two opposite faces of the cube, which is suspended with its base i inch above the bottom of the beaker. The water or oil bath consists of an 800 cc. low-form Jena glass beaker suitably mounted for the application of heat from below. The beaker in which the cube is suspended is of the tall-form Jena type without lip. The metal cover has two openings. A cork, through which passes the upper arm of the wire, is inserted in one hole and the thermometer in the other. The bulb of the thermometer should be just level with the cube and at an equal distance from the side of the beaker. In order that a reading of the thermometer may be made, if necessary, at the point which passes through the cover, the hole is made triangular in shape and covered with an ordinary object glass through which the stem ENGINEERING CHEMISTRY 315 of the thermometer may be seen. Readings made through this glass should be calibrated to the angle of observation, which may be made constant by always sighting from the front edge of the opening to any given point on the stem of the thermometer below the cover. After the test specimen has been placed in the apparatus, the liquid in the outer vessel is heated in such a manner that the thermometer registers an increase of 5° C. per minute. The tem- perature at which the bitumen touches a piece of paper placed in the bottom of the beaker is taken as the melting point. Deter- minations made in the manner described should not vary more than 2° for different tests of the same material. At the beginning of this test the temperature of both bitumen and bath should be approximately 25° C. Use op MeivTing Point Determination. The melting-point determination should be made on all bi- tuminous road binders sufficiently hard to be handled at room temperature after removing from the mold. This test is not usually required for bitumens which are to be cut with a non- volatile flux before use. The Extraction of Bituminous Aggregates. Equipment for Recovering Aggregate Oni.y. I centrifuge extractor, complete with motor, speed regulator, and electrical connections. (Fig. 58.) I hot plate. I enamel ware dish approximately 2 inches deep and 9 inches in diameter. I hammer. I ^-inch cold chisel, I large metal kitchen spoon. I square foot of i /16-inch deadening felt paper. I i>4-inch stiff flat brush. I 500 cc. bottle or flask. I balance, capacity i kilogram, sensitive to o.i gram. I sheet of heavy Manila paper. 3i6 ENGINEJERING CHEMISTRY Additional Equipment for Recovering Bitumen. I iron ring support (ring lo centimeters in diameter). I iron ring support with condenser clamp. I round tin can, lo by 12 centimeters, covered with asbestos paper. I 32 candle-power incandescent lamp, with socket and connections. I asbestos hood. (Fig. 59.) I 1,000 cc. round-bottom flask, with cork. I spiral condenser, length of body 25 centimeters, with cork to fit, and rubber tubing connections. 50 centimeters of glass tubing, 8 millimeters bore. I 1,000 cc. flat-bottom flask. I porcelain evaporating dish, 11 centimeters in diameter. I watch glass, 20 centimeters in diameter. I steam bath. Method. The extractor shown in Fig. 58 was designed upon lines sitg- /• ^ X h'-' Tp^ ----.-4 —y 1. Fig. 58. — Office of Public Roads centrifuge extractor (Reeve type). gested by an examination of machines in use by A. E. Schutte and C. N. Forrest. It consists of a ^5 -horse-power 1,100 revolutions per minute vertical-shaft electric motor a, with the shaft project- ENGINEERING CHEMISTRY 317 ing into the cylindrical copper box b, the bottom of which is so inclined as to drain to the spout c. A ^/j^Q-inch circular brass plate 9^ inches in diameter is shown in d, and upon this rests the sheet-iron bowl e, which is 8^ inches in diameter by 2 ^/^g inches high, and has a 2-inch circular hole in the top. Fastened to the inner side of the bowl is the brass cup /, having a circle of ^8-inch holes for the admission of the solvent, and terminating in the hollow axle, which fits snugly through a hole at the center of the brass plate. The bowl may be drawn firmly against a felt- paper ring g, ^-inch wide, by means of the 2^-inch milled nut h, for which the hollow axle is threaded for a distance of ^ inch directly below the upper surface of the plate. The axle fits snugly over the shaft of the motor, to which it is locked by a slot and cross pin i. The aggregate is prepared for analysis by heating it in enamel- ware pan on the hot plate until it is sufficiently soft to be thoroughly disintegrated by means of a large spoon. Care must be taken, however, that the individual particles are not crushed. If a section of pavement is under examination, a piece weigh- ing somewhat over i kilogram may be cut off with hammer and chisel. The disintegrated aggregate is then allowed to cool, after which a sufficient amount is taken to yield on extraction from 50 to 60 grams of bitumen. It is placed in the iron bowl and a ring ^4 -inch wide, cut from the felt paper, is fitted on the rim, after which the brass plate is placed in position and drawn down tightly by means of the milled nut. If the bitumen is to be recovered and examined, the felt ring should be prev- iously treated in the empty extractor with a couple of charges of carbon disulphide in order to remove any small amount of grease or resin that may be present, although a proper grade of felt should be practically free from such products. The bowl is now placed on the motor shaft and the slot and pin are care- fully locked. An empty bottle is placed under the spout and 150 cc. of carbon disulphide is poured into the bowl through the small holes. The cover is put on the copper box and, after allowing the material to digest for a few minutes, the motor is 3l8 DNGINKEJRING CHElMISTRY Started, slowly at first in order to permit the aggregate to dis- tribute uniformly. The speed should then be increased suf- ficiently by means of the regulator to cause the dissolved bitumen to flow from the spout in a thin stream. When the first charge has drained, the motor is stopped and a fresh portion of disul- phide is added. This operation is repeated from four to six times with 150 cc. of disulphide. With a little experience the operator can soon gauge exactly what treatment is necessary for any given material. When the last addition of solvent has drained off, the bowl is removed and placed with the brass plate upper- most on a sheet of Manila paper. The brass plate and felt ring are carefully laid aside on the paper and, when the aggregate is thoroughly dry, it can be brushed on a pan of the rough balance and weighed. The difference between this weight and the original weight taken shows the amount of bitumen extracted. The aggregate may then be tested as occasion requires. When it is desired to recover and examine the bitumen, the apparatus shown in Fig. 59 will be found convenient and fairly safe for the distillation and recovery of such inflammable sol- vents as carbon disulphide. In the laboratory of the Office of Public Roads this apparatus is arranged so that the glass tubing passes through a stone partition between two sections of a small hood, thus keeping the distilling and receiving apparatus entirely separated. The solution of bitumen should be allowed to stand overnight in order to permit the settling of any line mineral matter that is sometimes carried through the felt ring in the extractor. The solution is then decanted into the flask a, and the solvent is driven off by means of heat from an incandescent lamp until the residue is of a thick sirupy consistency. Meanwhile the solvent is con- densed and recovered in the flask b. The residue is poured into an 1 1 -centimeter porcelain evaporating dish, and evaporated on a steam bath. The most scrupulous care must be taken at all times that no flames are in its immediate vicinity. Evaporation is carried on at a gentle heat, with continual stirring, until form- ing practically ceases. It is advisable to have a large watch ENGINEiERING CHEMISTRY 319 glass at hand to smother the flames quickly, should the material ignite. As the foaming subsides, the heat of the steam bath may be gradtially raised, and evaporation is continued until the bubbles beaten or stirred to the surface of the bitumen fail to give a blue flame or odor or sulphur dioxide when ignited by a small gas jet. The dish of bitumen should then be set in a hot- Fig. 59. — Recovery apparatus. air oven maintained at 105° C. for about an hour, after which it is allowed to cool. Its general character is noted and any tests for bitumens that are necessary are then made upon it. Grading the Mineral Aggp:egate. Equipment. I set of 18-inch stone sieves with meshes of i^, 1%, 1, ^, ]/2, %, and y^ inches, respectively. I set of 8-inch brass sand sieves of 10, 20, 30, 40, 50, 80, 100, and 200-mesh, respectively, with pan and cover. I rough balance, capacity i kilogram, sensitive to o.i gram. I ij^-inch stiff flat brush Several sheets of Manila paper. 320 ENGINEERING CHEMISTRY Fig. 60. — Dulin Rotarex, for Determining the Mineral Aggregate in Bitumen Pave- ments. The sample may be observed at all times and solvent added without removing the cover. The solvent used is non-inflammable and time for extraction is 5 minutes, leaving mineral aggregate perfectly dry so that grades may be deter- mined. Small model for samples of 10, 25 or 50 grams, for no volts, either alter- nating or direct current (except 25 cycles). Method. For aggregates containing particles too large to pass a ^-inch screen, the stone sieves are used, and are stacked in their regular order over a sheet of heavy paper, with the largest size required on top. The weighed amount of stone is placed on the largest sieve and is carefully protected from drafts which might carry away any of the fine material. The upper sieve is then removed from the stack and shaken over a large sheet of paper until no more particles come through. The material thus retained, in- cluding any fragments caught in the meshes of the sieve, is weighed and that which passes is added to the contents of the succeeding sieve. This operation is repeated with each succeed- ing sieve. ENGINEERING CHEMISTRY 321 When grading sands or fine aggregates, it is customary to take a 100-gram sample in order that the weights may give direct per- centages to tenths of i per cent. The sieves are stacked in regular order with the 200-mesh sieve resting on the pan. The sample is brushed on the top sieve, after which the cover is put on and the stack agitated for about 5 minutes with both rocking and circular shaking. Each sieve is removed in order, and shaken and tapped on a clean piece of paper until no appreciable amount ^JJ^^^^S^ j 1 ■ K I^H Fig. 61.— Sieve Shaker with Electric Motor. of material comes through. All lumps are broken up by crushing them against the side of the sieve with the finger or a small spatula. The contents of the sieve are emptied into the pan of the balance. All particles caught in the mesh are removed by brushing across the underside of the sieve and are added to the contents of the pan. As great opportunity exists for wide var- iations in the results of sand gradings made by different persons, 322 Engine:e:ring chemistry owing to the possibility of always getting a little more material to pass by continued shaking, it is well for the novice to repeat his sifting on any given mesh, after having weighed it, in order to see what further loss he can produce. If his judgment has not erred, several minutes further sifting should not produce a loss of over 0.5 gram. Where coarse aggregates have considerable material passing a ^-inch screen and it is desired to grade this material further, it should be weighed and well mixed, quartered, if necessary, and a 100-gram sample should be passed through the sand sieves. From the percentages so obtained and the weight of material passing the ^-inch sieve, the percentages of the total aggregate which these finer materials represent may be calculated. The Office of Public Roads has adopted the following recom- mendations of the Committee on Standard Tests for Road Ma- terials of the American Society for Testing Materials as to the size of wire for standard sand sieves : Diameter of wire, Meshes per linear inch: inches. 10 0.027 20 0.0165 30 0.01375 40 0.01025 50 0.009 80 .0.00575 100 0.0045 200 0.00235 Distillation Test. Equipment. I 750 cc. glass retort with tiibulature. I chemical thermometer, 0° C. to 400° C. 6 25 cc. glass cylinders, graduated to 0.2 cc. I iron tripod. I iron support with condenser clamp, I iron support with burette clamp. 1 piece of wire gauze, 10 centimeters square. 2 Bunsen burners. I asbestos hood. I pint tin cup, seamless. I rough balance, capacity i kilogram, sensitive to o.i gram. I analytical balance, capacity 100 grams, sensitive to o.i milligram. i:ngine:£:ring chejmistry 323 Me:thod. Briefly described, this test consists in distilling 250 cc. of the material under examination in a 750 cc. glass retort at a uniform rate of from 40 to 60 drops per minute and collecting the various fractions in weighed glass graduates. In preparing for the test it will be. found convenient to mark permanently on the foot of each graduate its weight to within o.i gram. Owing to the pos- sibility of varying results due to lack of uniformity in the retorts, the Office of Public Roads specifies in purchasing that "the retorts shall be of such uniform size that, when placed in a vertical posi- tion with the bulb and mouth of stem resting on a level surface, it shall require not less than 725 nor more than 800 cc. to cause an overflow into the stem." The retort should be supported with the tubulature in a vertical position on one pan of the rough balance and its tare accurately obtained. From the specific gravity of the tar taken at 25° C, the weight of 250 cc. is calculated, and this amount after warming it in a tin cup, if necessary, to make it sufficiently fluid, is poured into the tared retort. A cork stopper carrying a thermometer is then inserted in the tubulature so that the top of the bulb^ is level with the bottom of the juncture of the stem and the body of the retort. A cold, wet towel wrapped about the stem of the retort may be made to serve as a condenser for the lighter distillates. The tar should be heated gradually by means of a Bunsen burner and, if it is a crude material containing much water, great care must be taken to prevent it from foaming over. When the thermometer registers 110° C, the graduated cylinder containing the first frac- tion is replaced by another, the towel is removed, and the asbestos ^ It was formerly the practice to place the bottom of the bulb level with the bottom of the juncture of the stem and body of the retort, and distillation was stopped at ^70° C. Recent comparative tests have, however, shown that the method was adopted gives results which conform more closely to those obtained by the other principal methods in use. As compared with the former method, it may be stated that the amount of total distillate to 315° C. is approximately the same as that obtained at 270° C. by the old method. The adoption as a standard method of distillation is at present being considered by a special committee of the American Society for Testing Materials, and indications point to the fact that considerable research will be required both with regard of the type of distilling apparatus, and the standardization of the ther- mometer and method of distillation before a satisfactory standard can be recommended. 324 ENGINKERING CHEMISTRY hood placed over the retort to prevent radiation to insure a more uniform temperature. The distillation should now proceed with- out difficulty, and the rate should be maintained at from 40 to 60 drops per minute. The receiver is changed again at 170° C, after melting down by gentle heating any solid material that may have been deposited in the stem of the retort. The next fraction is collected up to 270° C, using as many graduated cylinders as may be necessary without allowing any to become filled above the 25 cc. mark. The last fractions is collected up to 315° C, after which the burner is removed, and any solid matter in the retort stem is melted and collected in the last cylinder. Any solid matter adhering to the sides of the graduates is melted down by playing the flame of a burner upon them, after which the retort and graduates are cooled to room temperature and their contents determined by volume and weight. The vol- ume of pitch remaining in the retort is found by deducting the total volume of the distillates from the original 250 cc. taken. If water was present in the tar, it will be noticed that the first frac- tion separates into two layers, the lower of ammoniacal liquor or water and the upper of oil. Note should be made of the approxi- mate volume of solids which precipitate from the distillates upon cooling. The results obtained are calculated in percentages by volume and weight to tenths of i per cent, and reported as follows : Distillate Per cent, by volume Per cent by weight 1. Water or ammoniacal liquid 2. First light oils to 1 10° C 3. Second light oils 1 10° C 4. Heavy oils i7o°C. to 270° C ■j; Heavv oils 270° to "^i '^^ C — — f\ T'itpVi rfiirJnf . ... Use of thk Distillation Test. The distillation test is made upon tars and tar products, but seldom upon other materials unless the presence of tar is sus- pected, or where a determination of water is required. In mak- ENGINEERING CHEMISTRY 325 ing the water determinations on viscous or semi-solid bituminous materials, it is usually advisable to render the samples fluid by the addition of kerosene or benzol before distillation. Bitumens and Their Essential Constituents for Road Construction and Maintenance. So much confusion exists among road engineers and others in- terested in bituminous road binders concerning the meaning of certain terms as applied to these materials that it has seemed advisable to present in brief form the definitions of such terms as at present used by the United States Office of Public Roads. It should be understood, however, that these definitions are at present more or less arbitrary, owing to wide differences of opinion held by those who are considered authorities on the subject of bitumens. Notwithstanding these facts, it is hoped that this circular will furnish highway engineers and other in- terested persons with a foundation for acquiring and systemati- cally classifying further information along the lines herein indi- cated. To aid them in this matter a brief discussion of the value of the various materials used in road construction has been given in addition to the definitions. Acid Sludge. — A mixture of sulphonated hydrocarbons result- ing from the treatment of bitumens with sulphuric acid; usually a waste or by-product obtained in this manner from the purifica- tion of tar and oil distillates. When sufficiently concentrated these sulphonated products become viscous and gummy. They are readily attacked by water and are therefore unsuitable for use as enduring road binders. Anthracene. — A waxy crystalline hydrocarbon having the chemical formula Ci^Hjo, found in tars, principally coal tars which have been produced at high temperatures. Anthracene is believed to be of no practical value in road binders. Artificial Asphalt. — See Asphalts and Oil Asphalts. Artificial Bitumens. — Hydrocarbon distillates and residues produced by the partial or fractional distillation of bitumens, and hydrocarbon distillates produced by the destructive distilla- 326 enginke:ring chemistry tion of bitumens, pyro-bitumens, and other organic materials, such as wood, bone, etc. Native bitumens which have been treated merely for the removal of water and extraneous organic and inorganic materials should not be classed as artificial prod- ucts, but as refined native bitumens. As pnmts.— Solid or semi-solid native bitumens, consisting of a mixture of hydrocarbons or complex structure, largely cyclic and bridge compounds, together with a small proportion of their sulphur and nitrogen derivatives, but free from any appreciable amount of solid paraffines, melting upon the application of heat and evidently produced by nature from petroleums containing little or no solid paraffines. Solid or semi-solid residues produced from probably similar oils by artificial processes are sometimes called asphalts, but should more properly be termed oil asphalts. The more common types of native asphalts are known by the name of the locality in which they occur, such as Trinidad, Ber- mudez, Maracaibo, Cuban, California, etc. Native asphalts with few exceptions contain water, extraneous organic or vegetable matter, and inorganic matter, such as clay, sand, etc. A large proportion of these impurities is removed by a rough refining process without otherwise changing the character of the asphalt. Native asphalts are usually too hard to be used as road binders without first fluxing them with a heavy petroleum residuum and thus producing an asphaltic cement. Artificial asphalts are, as a rule, brought to suitable consistency during the process of manu- facture. Asphaltenes. — A term commonly applied to those hydrocar- bons in petroleums, petroleum products, malthas, asphaltic ce- ments, and solid native bitumens which are soluble in carbon bisulphide but insoluble in paraffine naphtha. As a rule paraffine naphthas of different specific gravities and boiling points dis- solve different amounts of hydrocarbons in a given bitumen, and the heavier the naphtha and the higher its boiling point the greater is its solvent action. It is evident, therefore, that the percentage of asphaltenes will vary with the gravity and boiling point of the naphtha, and for this reason it would seem well to ENGINEERING CHEMISTRY 327 substitute for the term asphalenes, ''bitumen insoluble in paraffine naphtha," with a statement of the gravity of the naphtha used and the temperatures between which it boils. The presence of naphtha insoluble hydrocarbons is supposed to give body to the product in which they occur and to be accountable to a great extent for its binding value. They show no binding value, since many of them are hard and brittle, but they produce adhesive mixtures when fluxed with certain heavy oils. As a rule, for a given type of bitumen hardness increases with the percentage of bitumen insoluble in a given naphtha. The so-called asphaltenes are not found to any extent in native bitumens with a paraffine base, but occur principally in asphalts, malthas, as- phaltic petroleums, and in blown petroleum residues. They vary chemically and physically with the product in which they occur, and, therefore, do not represent definite chemical compounds. Asphaltic Petroleums. — Asphaltic petroleums, or asphaltic oils, are petroleums containing an asphaltic base — i. e., they are capable of producing residues very similar to native asphalts if evaporated or distilled down to the consistency of such asphalts. They contain little or no solid paraffines and are thus differen- tiated from paraffine petroleums. Native asphalts are probably produced from such oils by natural processes. Asphaltic Cement. — The term asphaltic cement was originally applied to a product obtained by fluxing an asphalt with a suf- ficient quantity of heavy residual oil or flux to produce a binder of suitable consistency for paving purposes. In its broadest sense it may be applied to all semi-solid bitumens of an asphaltic nature which are of suitable consistency for use as binders in street or road construction, whether prepared by fluxing a solid native or artificial bitumen or by reducing an asphaltic or semi-asphaltic petroleum by distillation or other process. Baume Gravity. — An arbitrary scale of specific gravity or density of liquids, usually expressed as degrees Baume or ° B. This scale is commonly used in connection with oil products. For liquids lighter than water the scale begins at io° B., which represents the specific gravity of water, or i.oooo. As the 328 ENGINEERING CHEMISTRY Baume degrees increase the specific gravity decreases. The fol- lowing formulae is used in converting Baume degrees for liquids lighter than water into direct specific gravity and vice versa : Specific gravity = ^^^o-g at 17-5° C. °B = ^ ^ -130 at 17.5° C. ' Specific gravity For liquids heavier than water the scale begins at 0° B., which represents the specific gravity of water, or i.ocxdo. In this scale the degrees Baume increase with the specific gravity. The fol- lowing formulae is used in converting Baume degrees for liquids heavier than water into direct specific gravity and vice versa : Specific gravity = ^ op at 15.5° C : °B=i46 — -^ r^ :- at 15.5° C. Specific gravity Benzol. — A volatile colorless fluid hydrocarbon of character- istic odor having the chemical formula CjjHg. It occurs mainly in crude coal tars and water-gas tars, and boils at 80.4° C, so that it is removed in the first fraction when these tars are sub- jected to the process of distillation. Benzol is an active solvent for most bitumens. It is sometimes called benzene, but should not be confused with benzine, which is the term applied to the lighter and more volatile fractions of petroleum. Bitumen. — Bitumens are mixtures of native or pyrogenetic hydrocarbons and their derivatives, which may be gases, liquids, viscous liquids, or solids. If solids, they melt more or less readily upon the application of heat and are soluble in carbon bisulphide, chloroform, and similar solvents. They may be divided into two main classes — (i) native bitumens and (2) artificial bitu- mens. Bitumens, being mixtures of hydrocarbons, can have no melting point, although this term is often used to denote the temperature at which they soften sufficiently to flow. I^NGINEERING CHEMISTRY 329 Bituminous. — A term applied not only to materials or objects which contain bitumen, such as bituminous rock, bituminous macadam, etc., but also to certain pyro-bitumens, such as bitu- minous coal, which give rise to the formation of bitumens upon being subjected to the process of destructive distillation. Blown Petroleum. — Blown petroleums, which are often called blown oils, are petroleum residuums through which a jet of air has been passed during or just after distillation. The blowing process causes certain chemical reactions of a complicated nature to take place and results in thickening or increasing the con- sistency of the oil to an extent depending upon its temperature and the amount of blowing which it receives. Semi-solid and solid products are thus often formed from fluid residuums. If the oil is asphaltic or semi-asphaltic in nature, asphaltic cements may be produced in this manner. Blown oils are characteristi- cally short or non-ductile when semi-solid, although they may possess considerable binding value if not originally of a paraffine nature. Blowing an oil usually increases its percentage of hy- drocarbons insoluble in any given paraffine naphtha. Carbenes. — A term commonly applied to those hydrocarbons in petroleum, petroleum products, malthas, asphaltic cements, and solid native bitumens which are soluble in carbon bisulphide but insoluble in carbon tetrachloride. The presence of an appreci- able amount of these hydrocarbons indicates that the material in which they occur has been subjected to unnecessarily high tem- peratures. Cracked oil residuums show an increase in carbenes in proportion to the extent of cracking and the formation of these products is evidently a near step to coking. But little is known of their effect upon the value of a bitumen for road construction, but they are generally looked upon with suspicion and, in certain specifications for asphaltic cements, their presence has been limited to a low percentage. Carbon Bisulphide. — This substance, sometimes called carbon disulphide, is a volatile and extremely inflammable compound of carbon and sulphur, boiling ai 47° C. and having the chemical formula CSg. Pure carbon bisulphide is a colorless mobile liquid 330 ENGINEERING CHEMISTRY having an ethereal odor. It is one of the most active solvents for bitumens and is commonly employed for this purpose in the determination of total bitumen. Carbon Tetrachloride. — A volatile non-inflammable compound of carbon and chlorine, boiling at ^6° C. It is a colorless mobile liquid with an odor similar to that of chloroform, to which it is closely related, and has the chemical formula CCI4. It is an ex- cellent solvent for bitumens, but is not usually as powerful as carbon bisulphide. It is employed in bitumen analysis for the determination of carbenes or hydrocarbons soluble in carbon bi- sulphide but insoluble in carbon tetrachloride. Coal Tar. — A mixture of hydrocarbon distillates, most un- saturated ring compounds, produced in the destructive distillation of coal. Crude coal tar is a black, more or less viscid fluid having a gassy odor and varying in specific gravity from i.io to 1.25 and sometimes higher. It always contains a certain amount of am- moniacal water which makes it unsuitable for use as a road binder. When reduced to proper consistency by distillation, coal tar makes an excellent bituminous road binder, providing it does not carry too high percentages of free carbon and naphthalene. The composition of coal tar varies according to the coal from which it is produced and the method of distillation. Tar produced at high temperatures contain a large amount of free carbon and usually run high in naphthalene, while those produced at low temperatures carry less free carbon and as a rule less naphthalene. Low temperature coal tars are therefore more suitable for the preparation of road binders. Coke-Oven Tars. — Coal tar produced from by-product coke ovens in the manufacture of coke from bituminous coal. This process of coke manufacture is essentially the same as that of coal gas. Larger charges of coal are, however, carbonized in the former, and as a rule carbonization is conducted at a lower tem- perature than in the manufacture of coal gas. The resulting tar therefore contains a smaller amount of free carbon, averaging from 3 to 10 per cent., and is better suited for the preparation of road binders than most gas-house coal tars. Engine;e;ring chemistry 331 Cracked Oil. — The term cracked oil, as applied to road binders, refers to petroleum residuums which have been overheated in the process of manufacture. Overheating causes a breaking down of certain constituents of the oil, which results first in the formation of carbenes and later of coke or free carbon. Badly cracked residuums are believed to be inferior road binders. Cracking. — The process of breaking down a hydrocarbon mole- cule by the application of heat. This may result either in the formation of other hydrocarbons molecules, at least one of which is unsaturated and shows a higher ratio of carbon to hydrogen than the original molecule, or else in the disruption of the molecule into its elements, hydrogen and carbon. In the latter case the process is said to be destructive. The more volatile and chemi- cally stable hydrocarbons can be cracked only at temperatures above their boiling points. In the distillation of oils this is ac- complished by causing condensation to take place in the still and allowing the condensed oils to fall back into the residue, the temperature of which is considerably higher than their boiling points. In carbureted water-gas manufacture, oils are cracked by vaporizing them at a much higher temperature than their boil- ing points. The heavier oils will, however, crack at temperatures below their normal boiling points, and this is particularly true of asphaltic oils, which have to be distilled very carefully, some- times under reduced pressure, in order to produce residuums which are not cracked. Cut-Back Products. — Petroleum or tar residuums which are cut-back, or fluxed, to the desired consistency with a distillate. Volatile distillates are employed for this purpose in the prepara- tion of road binders, when it is desired to have the binder in- crease in consistency or become harder after application. In such cases a residuum of proper consistency for a road binder is cut-back merely for the purpose of facilitating application. Dead Oils. — Heavy oils distilled from tars at between 170° and 270° C. with a density greater than water. These oils, if free from naphthalene, serve as an excellent flux in the preparation 332 ENGINEERING CHEMISTRY of cut-back road binders from tar pitches, which are too brittle for this purpose. Destructive Distillation. — A process of distilling organic materials in which the identity of the material distilled is de- stroyed, resulting in the formation of tarry distillates and a coke residue. Dehydrated Tar. — Crude tar from which all water has been removed by distillation and mechanical contrivances known as separators. Bmidsions. — Oily substances made miscible with water through the action of a saponifying agent or soap. Petroleums and tars may be emulsified by this means and such emulsions, if properly prepared from good materials, are often serviceable in the treat- ment of roads. The majority of road emulsions can be consid- ered only as dust palliatives and temporary binders. Fixed Carbon. — The residual coke obtained upon burning hy- drocarbon products in a covered vessel in the absence of free oxygen, according to an arbitrary method. As applied to bituminous road materials, the determination of fixed carbon would seem to be of value in connection with petroleum and asphaltic products only. Paraffine hydrocarbons produce little or no fixed carbon, while those of asphaltic character show a very considerable amount, depending upon the percentage of asphaltic compounds present and the consistency of the material. The fixed carbon determination, therefore, indicates the mechanical stability and body of such materials. It is not, however, an ex- tremely accurate determination and should not be too strongly relied upon. Since fixed carbon is a product formed by ignition, it should not be confused with free carbon, which is a material already existing in suspension. The presence of any considerable quantity of free carbon vitiates a fixed carbon determination. Flux. — As applied to road binders, this term covers fluid oils and tars which are incorporated with asphalts and semi-solid or solid oil and tar residuums for the purpose of reducing their con- sistency. Fluid petroleum residuums are commonly employed as fluxes in the preparation of asphaltic cements. A good flux pro- ENGINEERING CHEMISTRY 333 duces an absolutely homogeneous bituminous mixture. Both petroleum and tar fluxes will produce such mixtures with native and artificial asphalts, but most fluid petroleum products will not flux tar pitches satisfactorily. Free Carbon. — ^Organic matter in tars which is insoluble in carbon bisulphide. It has no binding value and serves no useful purpose in a road binder other than to act as a filler. It gives the tar in which it occurs a false consistency, reduces the binding capacity of the tar, and probably interferes with its penetration into and absorption by the road stone or road surface. The maximum allowable limit of free carbon in road binders would seem to be about 20 per cent. Gas-House Coal Tar. — Coal tar produced as a by-product in the manufacture of illuminating gas from coal. The modern gas-house coal tar is usually produced at high temperatures and, therefore, carries a percentage of free carbon varying from 20 to 30 per cent, and higher. Unless it is produced at low or medium temperatures and contains less than 20 per cent, free carbon, it is not well suited for the preparation of a dust palliative or road binder by direct distillation. High-carbon tars may, however, be combined with low-carbon tars in such proportion as to produce, when distilled to proper consistency, excellent road binders carrying less than 20 per cent, free carbon. Gilsonite. — A very pure solid native bitumen possessing many of the characteristics of asphalt. It differs from most of the native asphalts by being more brittle, having a higher melting or softening point, and being much less soluble in 86° B. paraffine naphtha. When fluxed with certain petroleum residuums it pro- duces excellent asphaltic cements. In the preparation of road binders it is extensively used for the purpose of reinforcing blown oils, with which it combines to form rubbery semi-solid mixtures. Such preparations are sometimes termed mineral rubber. Grahamite. — A pure solid native bitumen, black and brittle, which does not melt readily, but intumesces at high temperatures. It is differentiated from gilsonite and the native asphalts by the 334 ENGINEJERING CHEMISTRY fact that it is almost insoluble in paraffine naphtha. It has been produced at high temperatures, as evidenced by the percentage of carbenes which it contains, and some varieties closely approach the pyrobitumens in characteristics. It has been used to some extent in the preparation of asphaltic cements, but up to the present has been little used in the manufacture of road binders. High-Carbon Tars. — Tars containing a high percentage of free carbon — above 20 per cent. High-carbon tars are produced at high temperatures during the destructive distillation of coal and are of inferior quality for use as dust palliatives and road binders. Hydrocarbons. — Chemical compounds composed of the ele- ments hydrogen and carbon. There is practically an unlimited number of such compounds, which vary greatly in physical and chemical characteristics. Complex mixtures of hydrocarbons constitute by far the greater proportion of all bitumens. Low-Carbon Tars. — Tars containing a low percentage of free carbon — less than 10 per cent. Low-carbon tars are produced at comparatively low temperatures during the destructive distilla- tion of coal, and also by cracking oil vapors during the manu- facture of carbureted water gas. As a rule they are more suitable than high-carbon tars for use as dust palliatives and road binders, or for the preparation of such substances. Malthas. — Malthas are very viscous semi-asphaltic or asphaltic native bitumens holding an intermediate position between the petroleums of an asphaltic nature and the native asphalts. As a rule they possess excellent binding properties. They constitute the binding material of many bituminous rocks or rock asphalts, and in this capacity often serve as valuable road binders. Many malthas have a tendency to harden rapidly when exposed to atmospheric conditions, and this property, while accountable for an increase in binding value, makes them unsuitable for use as a flux in the preparation of asphaltic cements. Malthenes. — A term commonly applied to those hydrocarbons in petroleum, petroleum products, malthas, asphaltic cements, and solid native bitumens soluble in both carbon bisulphide and par- afline naphtha, but not readily volatile at temperatures lower than ENGINEERING CHEMISTRY 335 163° C. (325° F.). This class of hydrocarbons serves as a valu- able permanent fluxing medium for the so-called asphaltenes or naphtha insoluble bitumen in asphaltic cements, giving the cement any desired degree of softness when present in the right amount. It is evident, therefore, that the consistency of asphaltic bitumens, and particularly stable asphaltic cements, is largely dependent upon the relative proportion of naphtha soluble and naphtha in- soluble hydrocarbons. The same objection to the use of the term ''asphaltenes" applies to the use of the term "malthenes." Mineral Rubber. — A term sometimes applied to artificial bitu- mens of rubbery consistency, usually composed of a mixture of gilsonite and blown petroleum residuum. Naphthas. — Mixtures of hydrocarbons of low boiling points occurring rarely in nature, commonly obtained from the frac- tional distillation of certain bitumens. When this term is applied to low-boiling coal tar distillates, it is usually prefixed by the words "coal tar." The word "naphtha" by itself is generally applied to low-boiling petroleum products. Different grades of naphtha are differentiated not only by their boiling points but also by their specific gravities, which are commonly given in Baume degrees. Those of very low boiling points and specific gravities are called petrolic ethers. Naphthas vary not onl) in the two properties above mentioned but also with the type of petroleum from which they are obtained. Those derived from paraffine petroleums are quite different chemically from naphthas obtained from asphaltic petroleums. The former are much less powerful solvents for asphaltic substances than the latter. Paraffine naph- tha is used as a solvent for the separation of certain classes of hydrocarbons in asphaltic substances. Naphthalene. — A solid crystalline highly volatile hydrocarbon occurring principally in coal tars and having the chemical for- mula CioHg. Its presence in excessive quantities in road tars is believed to be detrimental, as it possesses no binding value and gradually volatilizes from the tar, leaving it hard and brittle. Native Bitumens. — Mixtures of hydrocarbons occurring in nature, which may be gases, liquids, viscous liquids, or solids, 336 ENGINEERING CHEMISTRY but if solid melting more or less readily upon the application of heat and dissolving in carbon bisulphide, chloroform, and similar solvents. The native bitumens that are of use as road materials are petroleums, malthas, asphalts, and other solid products such as gilsonite and grahamite. Native bitumens often contain im- purities such as water, inorganic matter in the form of clay, silt sand, etc., and extraneous organic or vegetable matter. Oil Asphalts. — Artificial oil pitches or asphaltic cements pro- duced as a residuum in the distillation of semi-asphaltic and as- phaltic petroleum. Many of these products are blown and are therefore known as blown oils. Oil Pitches. — More or less hard oil asphalts. Oil Tars. — Mixtures of hydrocarbon distillates, mostly un- saturated ring compounds, produced in the cracking of oil vapors at high temperatures. Oil tars are usually by-products of the manufacture of oil gas or carbureted water gas. Paraffine Naphthas. — Naphthas consisting of a mixture of light volatile hydrocarbons of the paraffine series, ordinarily obtained as light distillates of paraffine petroleum. Paraffine Petroleum. — Petroleum the base of which is com- posed principally of the paraffine or open-chain series of hydro- carbons ; it is thus differentiated from asphaltic petroleums which are composed largely of cyclic or ring hydrocarbons. Paraffine petroleums and the unaltered residues produced by their distilla- tion are of inferior value as dust palliatives and road binders. Paraffine Scale. — Solid paraffines recovered by distillation and precipitation of the distillates of petroleum and similar materials. The percentage of paraffine in bitumen is usually determined in this manner. Paraffine. — The term paraffine covers a number of greasy crystalline hydrocarbons of the paraffine series occurring as dis- solved wax in certain classes of petroleiim. When these products are recovered from petroleum, they constitute the commercial product paraffine. Paraffine is believed to be detrimental to road binders in which it occurs, and it is certain that its presence in excessive amounts indicates inferiority in the binding value of the e;ngine;e;ring che;mistry 337 material. It is probable, however, that heavy liquid hydrocarbons of the same chemical series as solid paraffine exert a much more injurious effect. Petrolenes. — An ambiguous term sometimes applied to those hydrocarbons described under malthenes, which are soluble in carbon bisulphide but insoluble in paraffine naphtha, and some- times to hydrocarbons in petroleum products volatile at or below 163° C. (325° F.). Owing to misconceptions which may occur, it would seem advisable to eliminate the use of this term. Petroleums. — Petroleums, or mineral oils, are fluid native bitu- mens of variable composition, depending largely upon the locality in which they occur. There are three general types of petroleum found in the United States: (i) Paraffine petroleums, (2) semi- asphaltic petroleums, and (3) asphaltic petroleums. Paraffine petroleums occur mainly in the eastern part of the United States and are typified by the Pennsylvania oils. The semi-asphaltic variety occurs in the southern and middle western parts of the United States. Texas is one of the main sources of this type. Asphaltic petroleums occur in the western part of the United States, particularly in California. Petroleums, if of semi-asphaltic or asphaltic character, may make excellent dust palliatives and road binders when properly treated. Petrolic Ethers. — Very light volatile naphthas obtained from petroleum. Pitches. — Semi-solid or solid residues produced in the evapora- tion or distillation of bitumens. This word is often prefixed by the name of the material from which it is derived, such as oil pitch, coal-tar pitch, etc. As a rule the term pitch is confined to the harder residuums, most of which are too hard for use as road binders unless fluxed with a more fluid product. Pyrobitumens. — Mineral organic substances which are but slightly acted upon by the solvents for the bitumens, but which, upon being subjected to destructive distillation, give rise to the formation of bitumens. Pyrobitumens are derived in nature both from bitumens and direct metamorphosis of vegetable matter. 22 338 e:nginee)ring chemistry Among the former class may be mentioned Albertite and Wurtzi- lite, and among the latter, peat, lignite, and bituminous coal. Pyrogenetic. — Originating from the action of heat. Coal tar is thus a pyrogenetic bitumen. Reduced Petroleum or Reduced Oils. — Residual oils produced from crude petroleum by the removal of water and the more vola- tile oil constituents, without chemically altering the base by cracking or other means: These residues are often made by dis- tilling the crude oil under reduced pressure. Such products are of little value for road treatment unless formed from semi-asphal- tic or asphaltic oils. Refined Tar. — A more or less viscous tar which is produced by evaporation or distillation of crude tar until the residue is of the desired consistency. This term also includes blown tars and cut-back products produced by fluxing tar pitches with volatile or non-volatile distillates. Refined tars are of value both as dust palliatives and as road binders in the treatment of macadam roads. Their binding value is proportional to their hardness within certain limits. Residual Petroleums or Residual Oils. — Heavy viscous resi- dues produced by the evaporation or distillation of crude petro- leum until at least all of the burning oils have been removed and often some of the heavier distillates as well. Residual oils grade into the artificial asphalts and oil pitches as their hardness and viscosity increase. The more fluid products, if obtained from semi-asphaltic or asphaltic petroleums, serve as excellent dust palliatives and semi-permanent road binders for the surface treatment of roads. The more viscous products are often suit- able for the surface treatment of roads if applied hot, but are seldom of value in road construction unless produced from semi- asphaltic or asphaltic oils. Residual Tars. — Heavy viscous residues produced by the evaporation or distillation of crude tar until all of the light oils have been removed. Residual tars grade into the tar pitches as their hardness and viscosity increase. If they do not contain an ENGINEERING CHEMISTRY 339 excess of free carbon, they are as a rule well adapted for use as binders in the construction of macadam roads. Rock Asphalt or Bituminous Rock. — A term applied to a great variety of sandstones and limestones more or less impregnated with maltha. Deposits of such material are widely distributed over the United States and vary from rock which is friable and wholly dependent upon the bitumen to hold the mineral fragments together to solid rock having merely its interstices filled with bitumen. The former type is of value for use as a surface binder in the construction of roads when the maltha shows good binding value and amounts to not less than 6 per cent, of the weight of rock asphalt. Semi-Asphaltic Petroleums. — Semi-asphaltic petroleums or semi-asphaltic oils are petroleums containing a semi-asphaltic base, i. e., petroleums whose residues produced by evaporation or distillation, while composed mainly of asphaltic hydrocarbons, contain also a certain percentage of paraffine wax. They thus, show a mixed base. If their percentage of heavy paraffine hydro- carbons is not excessive, they may be made to produce good dust preventives and road binders. Short. — A term applied to bituminous materials which are non- ductile. Tar Pitches. — Semi-solid or solid residual tars. Owing to the general brittleness or tar pitches, only the softer varieties are of value in their natural condition as road binders. The harder pitches may, however, be used for this purpose if fluxed to suit- able consistency with heavy or dead oil distillates of tar. Tars. — Tars are artificial or pyrogenetic bitumens produced as distillates by the destructive distillation of bitumens, pyro-bitu- mens and other organic material. Water-Gas Tars. — Mixtures of hydrocarbon distillates, mostly unsaturated ring compounds, produced by cracking oil vapors at high temperatures in the manufacture of carbureted water gas. Crude water-gas tar is a thin, oily liquid having a specific gravity lying usually between i and i.io. As a rule it contains a consid- erable quantity of water which is, however, largely removed by 340 engini:e:ring chemistry mechanical devices before the tar is placed upon the market. This water is not ammoniacal, as in the case of crude coal tars. The composition of water-gas tar varies with the character of the oil which is carbureted and with varying conditions attending the carbureting process. It always shows a low percentage of free carbon, usually less than 2 per cent., and contains little or no naphthalene unless previously used for scrubbing coal gas. Crude water-gas tar has practically no binding value and is serviceable only as a dust palliative in the surface treatment of roads. When reduced to proper consistency by distillation, however, it shows certain desirable properties for use as a road binder both for surface treatment and macadam construction. Water-gas tar may also be used in the preparation of road binders from high- carbon coal tars. When this is done, the two crude tars are mixed in such proportion that when distilled to the desired consistency the mixture will contain less than the maximum limit of free carbon allowable. Specification for Sheet Asphalt Pavement.^ Sheet Asphai,t. 1. On the concrete foundation shall be laid the asphalt pavement proper, consisting of a binder course i to i^ inches in thickness when compressed, and an asphalt wearing surface 2 inches when compressed. 2. The binder course shall be composed of : (A) Binder stone. (B) Sand. (C) Asphaltic cement. The binder stone shall be composed of hard, clean, broken stone, all of which shall pass a screen having circular meshes i inch in diameter, arid shall be graded in size from i inch down, so as to produce, when mixed with the proper proportion of sand and of asphaltic cement, the mesh composition as herein below specified for the binder mixture. If the binder stone does not contain the required amount of fine material, sound, clean, broken stone, or gravel, passing y^-'mch diameter mesh screen, and clean, sharp sand passing a lo-inch mesh screen, shall be added, in such proportions as will supply the deficiency. The binder shall be composed of broken stone and sand as above specified, mixed with asphaltic cement, complying with the requirements 1 Jersey City, N. J , 1914. Portion of specification relating to asphalt constructions only. ENGINEERING CHEMISTRY 34I hereinafter described. The binder stone -and sand shall be heated in suitable appliances, not higher than 325° F., and shall then be thoroughly mixed by machinery with asphaltic cement at 300° F., in such proportion as to thoroughly coat the stone and all fine particles of the mineral aggregate, and produce a homogeneous binder mixture having life and gloss without an excess of asphaltic cement. The binder mixture as laid shall comply with the following require- ments for percentage composition : Mineral aggregate: Retained by i-inch circular mesh, per cent, of total mixture. Mineral aggregate : Passing i-inch circular mesh, and retained by J/2-inch circular mesh, 35 to 65 per cent, of total mixture. Mineral aggregate : Passing lo-inch mesh sieve, 20 to 35 per cent, of total mixture. Bitumen : 5 to 8 per cent, of total mixture. Penetration of asphaltic cement : 50 to 65. The binder mixture, prepared as above described, shall be hauled to the work, suitably covered with canvas while in transit so as to reach the street under construction, at a temperature between 200° and 325° F. The mixture shall then be promptly spread uniformly upon the founda- tion, to such thickness that after being immediately and thoroughly com- pacted by ramming and rolling, it shall have an average thickness of i^ inches, and its upper surface shall be parallel to the surface of the pavement to be laid. Before laying the binder course, the surface of the concrete founda- tion shall be thoroughly swept and cleaned, and all dirt, foreign matter, and loose material shall be removed. No traffic, except such as may be required in depositing the surface, shall be allowed on the binder course. Any part of the binder course that shows lack of bond, or that is in any way defective or which may become loose or broken up before it is covered with the wearing surface, must be taken up and removed from the street, and replaced with good material, properly laid, in accordance with these specifications. Binder when laid, shall be followed and cov- ered with wearing surface mixture as soon as practicable, and in all cases within 24 hours after laying, in order to effect the most thorough bond between the binder and wearing surface. The binder course must be kept clean and as free from traffic as is possible under working conditions. Generally no placing of binder or wearing surface will be permitted in wet weather, but work of this character, however, may continue when overtaken by sudden rain, up to the amount which may be in transit 342 ^NGINDEJRING CHEMISTRY from the plant at the time. The plant shall, however, shut down under these conditions, and no additional material will be permitted to be laid. No binder shall be laid on concrete which has not set sufficiently to withstand properly the weight of the roller. 3. The asphalt wearing surface shall be composed of : (A) Asphaltic cement. (B) Clean sharp sand, (C) Finely powdered inorganic dust. The asphaltic cement shall comply with the requirements hereinafter described. The sand shall be hard grained, moderately sharp and clean. As used it shall be so graded in size from coarse to fine as to produce in the finished surface mixture the mesh composition herein named. The inorganic dust, or filter, shall be finely powdered limestone, Port- land cement, or other satisfactory inorganic dust. The inorganic dust as used must be thoroughly dry, and of such a degree of fineness that the whole of it shall pass a 30-mesh sieve, and not less than 66 per cent, shall pass a 200-mesh sieve. The inorganic dust shall be free from loam, clay, or earthly material, and no dust from weather rock shall be used. The wearing surface mixture shall be composed of sand, inorganic dust, and asphaltic cement, mixed as hereinafter specified in definite pro- portions by weight, depending upon their character; but whatever may be the character of the composition of the sand, dust and asphaltic cement used, the proportions of the mixture by weight shall be such as to produce in the finished pavement mixture, when laid, the percentage composition hereinafter specified. The wearing surface mixture shall not exceed the maximum per- centage, nor contain less than the minimum percentage by weight of the total mixture herein specified for mesh composition of the mineral aggre- gate and percentage of bitumen, as follows : Retained by lo-mesh sieve None Passing lo-mesh sieve Retained by 40-mesh sieve 10 to 35 per cent. Passing 40-mesh sieve Retained by 8o-mesh sieve 20 to 55 per cent. Passing 8o-mesh sieve Retained by 200-mesh sieve 10 to 30 per cent. Passing 200-mesh sieve 12 to 18 per cent. Bitumen 9^ to 12Y2 per cent. Penetration of asphaltic cement 50 to 65 The term "mineral aggregate" applied to the asphalt wearing surface mixture as used in these specifications, shall signify the entire part or EJNGINDKRING CHEMISTRY 343 percentage thereof insoluble in carbon bisulphide, including collectively the sand, inorganic dust, and such native mineral matter and insoluble matter from the refined asphalt as may be contained in the asphaltic cement. The sand and the asphaltic cement shall be heated separately, the sand to approximately 325° F. and the asphaltic cement to approximately 300° F. The maximum temperature of the sand as delivered at the mixing box shall in no case exceed 350° F. The cold, inorganic dust shall be thor- oughly mixed with the hot sand and filler at the required temperature and in the proper proportions until a homogeneous mixture is produced, in which all particles are thoroughly coated with asphaltic cement. The sand, dust and asphaltic cement comprising the charge for each batch mixed shall be proportioned by weight. The surface mixture prepared in the manner above described shall be brought to the street at a temperature ranging from 250° to 325° F., and shall be suitably covered while in transit. The temperature of the mixture within the above limits shall be regulated according to the tem- perature of the atmosphere and the character of the materials employed. It shall then be deposited roughly in place by means of hot shovels, and spread uniformly by means of hot iron rakes, in such manner that, after having received its final compression by rolling, the finished pavement shall .conform to the established grade and have a thickness of not less than 2 inches. Before the surface mixture is placed, all contact surfaces of curbs, manholes, etc., shall be well painted with hot asphaltic cement. After raking, the surface mixture shall at once be compressed* by a light steam roller, and by tamping adjacent to the curbs, after which a small amount of Portland cement shall be swept over it. It shall thqn be thoroughly compressed by a steam roller weighing not less than 10 tons; the rolling being continued until no further compression is obtained. A space of 12 inches next the curb shall be coated with hot asphaltic cement, and the same ironed into the pavement with hot smoothing irons. Definition of Asphai^t. The term "asphalt" shall signify any solid natural bitumen or the residue from the distillation of an asphaltic petroleum, or natural liquid bitumen complying with the requirements hereinafter set forth. Natural asphalt may be either in a state of purity, or in admixture with native non-bituminous matter. The word "bitumen" shall signify any natural hydrocarbon or hydrocarbons soluble in carbon bisulphide. Penetration Consistency. The word "penetration" and "consistency" of an asphalt or bitumen, as used in these specifications, shall signify "the distance, expressed in hundredths of a centimeter, that a No. 2 needle will penetrate it at 'j'j" F. in 5 seconds, under a load of 100 grams," unless otherwise specified. 344 e:ngine:e:ring chemistry AsPHALTic Petroleum. A petroleum shall be designated "asphaltic" for the purposes of this specification, if the bituminous residue therefrom prepared as hereinafter described fulfils the requirements of the tests set forth for refined asphalt. The refined asphalt shall be obtained as follows : (a) By heating solid crude natural asphalt, without the admixture of any other material, to a temperature not exceeding 400° F. until all the water has been driven off and the product is homogeneous, and complies with the requirements hereinafter set forth. The process shall be con- tinued, if necessary, until the product has a penetration of not more than 30. (b) By the distillation of asphaltic petroleum or natural liquid bitu- men without the admixture of any other material, at a temperature not exceeding 700" F., the material being constantly agitated in a closed tank during said distillation, by the aid of steam under pressure, or other approved methods and the operation continued until the residual asphalt has a penetration of not more than 60 nor less than 30. The average penetration of any shipment, within the above limits and subject to the requirement of uniformity of shipments hereinafter specified shall be as determined by the engineer. No Additions to Crude Asphalt. Nothing whatever shall be added to the crude natural material at any time, either before, during or after refining, except as hereinafter provided. ' The preparation and refining of all asphalts for use under these specifications shall be subject to such inspection at the refineries and paving plants as the engineer may direct, and said preparation and refin- ing shall in all cases be conducted in the most suitable nad approved manner. Recognized Standard. Any asphalt proposed to be used or furnished under this contract shall be equal in quality and composition to the recognized standard for its particular kind or type. Refined Asphalts. All refined asphalts, to be approved for use under these specifications, must comply with the following requirements: Penetration. It shall not vary more than 15 units in penetration from maximum to minimum. Ductility. Should the ductility of the asphalt of any part of any shipment fall below 75 centimeters, it will be required that the ductility of the asphalt Engine:ering chemistry 345 of said shipment shall not vary more than 30 centimeters in ductility from maximum to minimum. Not less than 98^ per cent, of the total bitumen of all refined asphalts shall be soluble in carbon tetrachloride. When made into an asphaltic cement by the use of such materials and methods as are described in these specifications, they must produce an asphaltic cement complying with all the requirements hereinafter set forth for asphaltic cement. Volatilization. Under the volatilization test at 325° F. the refined asphalt shall not lose more than 3 per cent, of the bitumen present, nor shall the penetra- tion after such heating be less than one-half the original penetration. This requirement for determining the "per cent, of hardening" shall apply only to asphalts having a penetration of 10 or more. Mixtures. No mixtures of asphalts either at the refinery or at the paving plant will be permitted without the written consent of the engineer, except as provided herein. The use of asphalt mixtures will be permitted only provided that : (a) The particular kinds of asphalt (each of which shall comply with these specifications) and the proportions thereof are approved by the engineer. (b) The mixtures are made at the contractor's plant under the direct supervision of the engineer. (c) The asphaltic cement resulting from such mixture complies with all the requirements for asphaltic cement as set forth under section on asphaltic cement. Fluxes. The fluxing materials shall be the residuum from the distillation by the aid of steam, or a parafiine, semi-asphaltic or asphaltic petroleum. They shall be of such character that they will combine with the asphalt used to form an acceptable and approved asphaltic cement, complying with the requirements of these specifications. In each case the proposed flux shall be tested with the asphalt to be used and its suitability or non- suitability for the proposed asphalt will be determined by the engineer. Residuums. (a) Residuums shall contain not less than 99 per cent, of bitumen soluble in carbon bisulphide at air temperature, of which bitumen not less than 98J/2 per cent, shall be soluble in carbon tetrachloride. Penetration. (b) Residuums shall have a penetration greater than 350 with a No. 2 needle, at 77° F., under 50 grams load for i second. 346 i:ngine:e:ring chemistry Flash Point. (c) Residuums shall not flash below 350° F., when tested in a New York State closed oil tester. Volatilization. (d) Residuums shall not lose more than 3 per cent, of matter by- volatilization at a constant temperature of 325° F. for 5 hours. The residue, after heating, shall be homogeneous and shall flow at 77° F. Specific Gravity. (e) Residuums shall have a specific gravity at 77° F., of not less than 0.92 nor more than i.oi. Paraffine Residuums. Residuums having a specific gravity of not less than 0.92 nor more than 0.94 at 77° F., shall be designated paraffine residuums. Semi-Asphaltic Residuums. Residuums having a specific gravity of not less than 0.94 nor more than 0.98 at 77° F., shall be designated semi-asphaltic residuums. Asphaltic Residuums. Residuums having a specific gravity of not less than 0.98 nor more than I.OI at 77° F. shall be designated asphaltic residuums. ASPHA1.TIC Cement. If the refined asphalt has the penetration required for asphaltic cement hereinafter specified, it may be used without flux, and said refined asphalt shall be considered as asphaltic cement and shall fulfil all the requirements therefor. If the refined asphalt is not of the required pene- tration, an asphaltic cement shall be prepared from refined asphalt and flux, agreeing in all respects with the requirements for each hereinbefore specified, in such proportions as to produce an asphaltic cement of not less than 40 nor more than 65 penetration, when used, except as provided for natural asphalt containing mineral matter. Penetration. The asphaltic cement shall have a penetration within the limits above specified, but the penetration shall be varied within these limits to adapt it to the character of the particular asphalt used, to the general character of the mineral aggregate of the paving mixture and to the character of the traffic on the street. The penetration of the asphaltic cement shall be as directed by the engineer in writing for each street and the asphaltic cement as used shall not vary more than 5 points plus or minus from the penetration directed. Mixing. The refined asphalt and flux in the proper proportions to produce an asphaltic cement as above specified shall be weighed separately, and then EJNGINDERING CHEMISTRY 347 be heated and melted together in a suitable tank and thoroughly agi- tated by suitable apparatus until completely blended into a homogeneous asphaltic cement to the satisfaction of the engineer. The above operation of melting and heating shall be conducted at a temperature of not less than 275° F. nor more than 350° F. During use, the asphaltic cement shall be maintained at approximately 325° F. Heating. The asphaltic cement must not be heated to a temperature exceeding 350° F, It must be kept uniform in composition and consistency, and thoroughly mixed and agitated both before and during use. Approved methods of agitation, which will not injure the cement, must be used. If kept in storage in a molten condition, any decrease in penetration, or hardening, shall be corrected by the addition and thorough incorporation of a proportionate amount of flux to produce the desired penetration before using. The refined asphalt and flux comprising the asphaltic cement, shall when required, be weighed separately in the presence of the engineer. The asphaltic cement shall comply with the following requirements : (a) It shall be completely fluxed and thoroughly homogeneous. Penetration. (b) It shall have the penetration between 40 and 65, as directed by the engineer in writing. If an asphaltic cement contains natural mineral matter, a reduction on the required penetration of fine points shall be made for each 10 per cent, of such natural mineral matter present. Volatilisation. (c) The asphaltic cement shall not lose more than 3 per cent, of the bitumen present by volatilization at a temperature of 325° F., nor shall the penetration of the residue after such heating be less than one-half the original penetration. Ductility. (d) Either the asphaltic cement as prepared at the paving plant and as used, at 50 penetration, or the purified bitumen obtained therefrom at 50 penetration, when made into a briquette having a minimum cross sec- tion of 1 square centimeter, shall have a ductility at 77° F. of not less than 30 centimeters at the time of rupture, the rate of elongation being 5 centimeters per minute. If the asphaltic cement as used complies with the requirements for ductility no test of the purified bitumen for ductility will be required. If the consistency of the asphaltic cement as used is greater or less than 50 penetration, or of the purified bitumen therefrom, above men- tioned, is greater or less than 50 penetration, the above requirement for ductility shall be as follows : 348 ENGINEERING CHEMISTRY For each increase of 5 units in penetration of the asphaltic cement as used (or of the bitumen therefrom) above 50, 2 centimeters shall be added to the requirement for ductility at 50 penetration as above estab- lished ; and for each decrease of 5 units in penetration of the asphaltic cement as used (or of the bitumen therefrom) below 50, 2 centimeters shall be subtracted from the requirement for ductility at 50 penetration as above established. Penetration at 77° and 100° F. (e) The asphaltic cement when at a penetration of 50 at 77° ¥., shall not be so susceptible to changes in temperature that the difference between its penetration at 100° F, and at 32° F. shall be more than 200; in each case, the penetration test being made with a No. 2 needle, for 5 seconds, under a load of 100 grams. If the asphaltic cement as used, has a penetration greater or less than 50 at 77° F., its penetration at 100° F. shall not exceed four times its pene- tration at 77° F., the conditions of time and load being as above established. Fluxed Asphaltic Cements. All fluxed asphaltic cements shall be prepared at the contractor's paving plant, except that, in case satisfactory reasons are given for the preparation of fluxed asphaltic cement elsewhere, the same shall be per- mitted only with the consent in writing of the engineer and under his supervision. Bidders' Samples. All bidders must deposit with the chief engineer at his oflice in the City Hall, Jersey City, N. J., and who will issue a receipt for same, samples of materials intended to be used, 5 days before the date bids or proposals are advertised to be received, labeled with the bidder's name and address as follows : (A) A sample of not less than 2 pounds of refined asphalt, together with a certificate stating the name of the asphalt and where same was mixed. (B) A sample of not less than 2 pounds of liquid asphalt flux, if any, accompanied by a certificate stating where same is mined, and giving its fire and flash tests and its specific gravity. (C) A sample of not less than 2 pounds of asphaltic cement, together with a certificate stating the formula used in its composition, all quantities being expressed in pounds. (D) A sample of not less than 2 pounds of crushed stone to be used in the binder mixture. (E) A sample of not less than 2 pounds of sand to be used in the asphalt wearing surface, together with a certificate showing what proportion of same passes a 200, 100, 80, 40 and lo-mesh screen. ENGINEERING CHEMISTRY 349 (F) A sample of not less than 2 pounds of powdered inorganic dust or filler to be used in the asphalt wearing surface mixture, together with a certificate stating the kind of material from which it is made and the percentage of Avhich passes a 200-mesh screen. METHODS FOR TESTING COAL TAR AND REFINED TARS, OILS AND PITCHES.* Determination of Water in Tar. The apparatus used is illustrated in Fig. 62. I ■ ■¥ i '1 Fig. 62. Measure 50 cc. of coal tar naphtha or light oil (which must be tested to determine that it is free from water, whenever a new supply is required) in a 250 cc. measuring cyhnder. (No ob- jection is raised to measuring the tar direct into the still or in other ways, but the measurement must be made as described in case of dispute.) Add 200 cc. of the tar. Transfer contents of cylinder to copper still and wash the cylinder with 50-75 cc. more of naphtha, adding the washings to contents of the still. Attach lid and clamp, using a paper gasket and set up apparatus as * By S. R. Church, Chief Chemist, Barrett Mfg. Co., N. Y., Jotir. of Indus, and Eng. Chemistry, April, 191 1. Position only given here by consent of author. 350 ENGINEERING CHEMISTRY shown in Fig. 62. Apply heat by means of the ring burner and distil until the vapor temperature, as indicated by the ther- mometer (in this and all other tests care must be used to have the thermometer set exactly as shown in drawings) has reached 205° C. (401° F.). The distillate is collected in the separatory funnel, to which 15-20 cc. of benzol has been previously added. This effects a clean separation of the water and oil. The reading is made after twirling the funnel and allowing to settle for a few minutes. The percentage is figured by volume. Specific Gravity. The tar is dried by taking 300-400 cc. in the apparatus used for water determination without the addition of naphtha. The distillation is carried to 170° C. (338° F.) vapor temperature. Any oil which has distilled over is separated from the water and returned to the still and thoroughly mixed in after cooling. Ap- paratus : a specific gravity bottle, Hubbard type (special), whose water capacity at 15.5° C. (60° F.) has been determined by ex- periment. Ten grams of tar are introduced at a temperature of 40-50° into the weighing bottle and the weight taken after cool- ing. Then freshly boiled distilled water is added and the bottle kept in a bath at 15.5° C. (60° F.) until no further contraction takes place. The water is then adjusted to the mark and the bottle removed from the bath and weighed. Weight of tar divided by the weight of H^O displaced gives the specific gravity. For rough determination, as of wet tar, a spindle may be used at any convenient temperature. To reduce the gravity as found to 15.5° C, 0.000685 is added for each degree C, above 15.5° C. (or 0.00038 for each degree F. above 60° F.). Melting Point. — Apparatus shown in Fig. 63. I. Pitches from 43^-77° C. (iio°-i70° F.). A clean-shaped ^-inch cube of the pitch to be formed in the mold, placed on the hook of No. 12 copper wire, and suspended in the 600 cc. beaker so that the bottom of the pitch is i inch above the bottom of the beaker. (A sheet of paper placed on bottom of beaker and con- veniently weighted will prevent pitch from sticking to the beaker when it drops off.) The pitch to remain 5 minutes in 400 cc. of Kngine:e:ring chejmistry 351 water at a temperature of 15.5° C. before heat is applied. Heat to be applied in such manner that the temperature of the water is raised 5° C. (9° F.) each minute. The temperature recorded by the thermometer at the instant the pitch touches bottom of beaker to be considered the melting point. Fig. 63. — Melting Point Apparatus. 2. Below 43° C. (110° F.) the same method can be used except that at the start the water should have a temperature of 4° C. (40° F.) 3. For pitches from yy^ C. up (170° F.), cottonseed oil should be substituted for water; otherwise the method remains the same. With these harder pitches, it may be necessary to heat the pitch in order to form a cube. Care should be used not to heat it any higher than is necessary, or to continue heating it for any length of time. A hot knife blade will often assist this manipula- tion. Note. To aid the removal of the pitch from the mold it may be greased with a very thin film of vaseline. 352 Engine:e:ring chemistry Breaking Point. A small piece of pitch is quickly melted directly on the copper disk on the steam bath to a layer of about ^/g^ of an inch. The disk is then placed in the porcelain dish and well covered with water of about ii°-i2° C. above the breaking point of the pitch. Fig. 64. — Breaking Point Apparatus. The temperature is reduced 1° per minute and tested from time to time by inserting a small, thin knife blade below the pitch and turning slightly until a point is reached at which the pitch snaps. This is taken as the breaking point. The copper disc should be held with a pair of tongs and not with the fingers. Light Oil Distillation. One hundred cubic centimeters are measured in a cylinder and transferred to the 200 cc. Jena glass distilling bulb and heated. The distillate is collected in a 100 cc. cylinder. The point where the first drop falls from the end of the condenser is noted and thereafter the cubic centimeter distilled noted at every even 10° C. continuing until 95 per cent, of the oil has distilled. Toward the end, the condensing water must be heated to avoid separation of naphthalene. Redistillation. — The extracted oil is placed in the Hempel ap- paratus (Fig. 65, apparatus for light oil distillation), and re- distilled, noting the cubic centimeters that have come over at 170° C. and 200° C, at which latter point the distillation is inter- rupted.^ 1 The fraction — 170° C. shows crude benzol, toluol and solvent; 170-200° crude heavy naphtha. ENGINEERING CHEMISTRY 353 Naphthalene. — The residue above 200° left in the flask in D is transferred to a copper beaker and cooled to 15.5° C. for 15 min- utes and the dry naphthalene determined as under creosote. CARBOI.IC Ollv. A. Specific Gravity. — Apparatus — see Fig. 65. If the oil is not limpid at 15.5° C, the gravity is taken at a higher temperature I Fig. 65. — Light Oil Distillation. No. i — Distilling Flask. No. 2 — Condenser (Special), No. 3 — Thermometer (Standard). No. 4 — ^Jena Flask (A.H.Thomas 1426, 200 cc). No. 5 — Hempel Tube (Special). No. 6 — Graduated Cylinder (100 cc). and 0.0008 added to the specific gravity for every degree above 15-5°. B. Distillation. — Same as "Light Oil." C. Tar Acids.—S^mt as 'Xight Oil." Benzoics. A. Distillation. — Same as light oil, with the water in the con- 23 354 ENGINEERING CHEMISTRY denser always cold.^ With C. P. benzol or toluol readings are taken every 0.2° C, with commercial benzols every 10° C. a reading is taken. B. Gravity. — Hydrometer at 15.5° C. C. Wash Test. — Taken only on water white grades. About 7 cc. of concentrated HoSO^ and 21 cc. of the benzol are shaken in a small glass stoppered French square bottle of 30 cc. capacity and the coloration of the acid and oil noted. Creosote Oils. A. Standard Creosoters Distillation. — Method is given in Bidl. (55, American Railway Engineering and Maintenance of Way Association. Before beginning the distillation, the retort should be care- fully weighed and exactly 100 grams of the oil placed therein, the same being weighed in the retort. The thermometer should be inserted in the retort with the lower end of the bulb ^ inch from the surface of the oil, and the condensing tube attached to the retort by a tight cork joint. The distance between the bulb of the thermometer and the end of the condensing tube should not be less than 20 nor more than 24 inches, and during the prog- ress of the distillation the thermometer must remain in the posi- tion originally placed. The distillates should be collected in weighed bottles and all fractions determined by weight. Reports are to be made on the following fractions : 0° to 170° C. 170° to 200° C. 200° to 210° C. 210° to 235° C. 235° to 270° C. 270° to 315° C. 315° to 355° C. For practical purposes there will be no need of reporting on all of these fractions. It will be sufficient to report on the fractions as follows : ^ Distillation .continued to dryness and drying point recorded. DNGINEIE^RING CHEJMISTRY 355 Below 200° C. 200° to 210° C. 210° to 235° C. 235° to 315° C. Above 315° C. Reports are to be made on individual fractions. In making such reports, it is to be distinctly understood that these fractions do not necessarily refer to individual compounds. In other words, the fractions between 210 and 235° will not necessarily be all naphthalene, but will probably contain a number of other compounds. The distillation should be a continuous one and should take about 45 minutes. When any measurable quantity of water is present in the oil, the distillation should be stopped, the oil separated from the water, and returned to the retort when the distillation should be recommended, and the previous readings discarded. In obtaining water- free oil, it will be desirable to free about 300 to 600 cc. of the oil by using the copper tar still and using 100 grams of the water-free oil for the final distillation. In the final report as to fractions, a correction must be made for the water content, so that the report may be made on the basis of a dry oil. Determination of Specific Gravity of Oil. In order to determine the specific gravity of any oil, heat the oil in a water bath until it is completely liquid. A glass stirring rod dipped into the liquid should show no solid particle on the rod when the same is withdrawn from the oil. When completely liquid, stir thoroughly and fill the hydrometer cylinder, which has previously been warmed. Insert a specific gravity hydrom- eter, taking care that the hydrometer does not touch the sides or bottom of the cylinder when the reading is taken. Take the tem- perature of the oil and make a correction for the specific gravity by reducing the same to the standard temperature of 15.5° C. or 60° F. The correct gravity is obtained by multiplying the cor- rection figure 0.0008 by the number of degrees C, or 0.00044 by 356 ENGINEE^RING CHEMISTRY the number of degrees F., the oil is found to be above 15.5° C, or 60° F. and adding the product to the observed gravity. Notes. 1. Emphasis is laid on attention to details and importance of a retort of the standard size. 2. The thermometer used must be of standard make, gas-filled and must be regularly tested for accuracy. Creosote Oil — Additional Tests. A. Drying Oil. — Apparatus as in drying tar. 500 cc. are dis- tilled up to 170° C, the water noted and the oil distilling over returned to the still after cooling. B. Tar Acids. — Apparatus shown in Drawing 7. 100 cc. of oil measured at limpid point, placed in Jena glass bulb and distilled. The distillation is continued until at least 95 per cent, has distilled off. The time from the first drop to the end should occupy about 20 minutes. The condenser tube should be kept warm enough by a flame during the operation to prevent distillate from solidifying. Warm the contents of the separatory funnel to 60° C. in water, and note reading. Add 50 cc. of a 10 per cent, caustic soda solution. Shake well and allow to settle, drawing off the clear soda, warming again to 60°, and noting the shrinkage. Add 30 cc. of soda and note any further shrinkage. Repeat, if necessary, until no further shrinkage is noted. Then the total shrinkage is the per cent, of tar acids in the heavy oil. C. Dry Naphthalene. — The extracted oil from B. is placed in a copper beaker and held at 15.5° C, for 15 minutes. The mass is filtered on a perforated funnel in a suction pump and sucked dry. The naphthalene in the filter is then pressed between paper in a letter press to remove all oil and weighed. The per- centage is figured on the weights of original oil as given by the gravity at the limpid point. D. Limpid Point. — About 5 cc. taken in a No. 4 or No. 5 test tube at 60° C. are cooled, stirring with a thermometer until the first crystals begin to form. This point is taken as the limpid point. Cool in water only, if necessary. E^NGINE^ERING CHEMISTRY 357 Creosote Oil — Special Tests. A. Distillation as in Circular 112, U. S. Department of Agri- culture, except that instead of the Hempel flask as described there, a 500 cc. Jena flask with a Hempel tube attached is used. B. Gravities. — Apparatus. This is standardized with the pipette filled with water at 60° C. Oil at 60° is drawn up to the mark in the pipette, replaced in the outer tube and weighed. This can be used only. when i or 2 cc. of oil are available. C. Sulphonation Test. — Apparatus, see Fig. 66. Apparatus for -©Ea-c^ Fig. 66. — Special Heavy Oil Analysis. No. i — Separatory Funnel (Special). No. 2^ Specific Gravity Tube (Special). No. 3 — Flask (A. H. T. 14246-500 cc). No. 4 — Hempel Tube. No. 5 — Condenser. No. 6 — Asbestos Box. No. 7 — Asbestos Pad. No. 8 — Thermometer (Stand). special heavy oil analysis. The weighed fraction distilling be- tween 305°-20° C. is warmed with concentrated H^SO^ (about 4-5 volumes) to 60° C. and the whole transferred to the separa- tory funnel. The flask is rinsed three times with more concen- 358 eNGINE:^RING CHEMISTRY trated tLSO^ and the rinsings added to the funnel. Then the funnel is stoppered and shaken, cautiously at first; afterwards vigorously, for about 15 minutes. Then let settle over night. Then the acid is carefully drawn down into the graduated por- tion to within 2 cc. of where unsulphonated residue shows. Whether any is visible or not the test should be carried further as follows : Add about 20 cc. water and let settle for ^ hour. Then draw down water as close as possible without drawing out any supernatent oil or emulsion. Then add 10 cc. strong H2SO4 and let settle for 15 to 20 minutes. Any unsulphonated residue will now settle out clear and give a distinct reading. If under 0.2 cc. it should be drawn down into the narrow part just above the stop-cock where it can be estimated to o.oi cc. The cubic centimeters are figured as percentages on the weight of the frac- tion taken. If the unsulphonated oil is dark in color it should be treated with an excess of 10 per cent, sodium hydroxide solution. If the oil is soluble in this reagent, the test is regarded as negative. Specifications for Wood Block Pavement.* The wood to be used shall be southern Longleaf pine ("Pinus Palus- tris," Miller Syn; "Pinus Australis," Michaux) according to the nomen- clature used by Chas, Mohr and Filibert Roth in Bulletin No. 13 (Revised Edition), U. S. Dept. of Agriculture, Division of Forestry, 1897; and subject to inspection in the stick before being sawn into blocks. The blocks shall be cut from what is known as prime timber as defined by the Interstate Rules of 1905, namely: All timber must be sound, well manufactured, saw-butted, all square edge, and shall be free from the following defects : unsound, loose and hollow knots, worm holes and knot holes, through shakes and round shakes that show in the surface. The blocks shall average 80 per cent, heart wood ; individual blocks may con- tain not over 50 per cent, sap wood, provided the timber has been well cured to the extent that sap wood is "live" (not brittle) and the sap wood is of a weight not less than 42 pounds per cubic foot. Timber shall be properly air dried so that blocks as cut shall not weigh more than 50 pounds per cubic foot. The annual rings in the timber used shall average not less than 8 per inch measured radially from the heart so as to include the greatest * Portions relating to the tests for coal tar used in the specifications — Borough of Manhattan, N. Y., 19 14. Engine;e:ring che:mistry 359 number of rings possible, and not over 5 per cent. o£ the stock shall show a minimum of 5 rings in any single inch of this radius. Different lots of timber of varying weights per cubic foot due to their being more or less thoroughly cured or dried, shall not be treated together in the same charge, but as nearly as possible timber weighing from 38 to 42, 42 to 44, 44 to 46, and 46 to 50 pounds per cubic foot, shall be treated in separate charges, the object being to separate the blocks for treatment according to their moisture content. The blocks shall be not less than 5 inches nor more than 9 inches in length ; and 4 inches in width. The depth of blocks parallel to fiber shall be 4]4 inches with an allowable variation of either way not exceeding 1/16 inch. They shall not vary more than ^ inch either way in width. Adjacent blocks in a course shall not vary more than % inch in width. The oil shall be sampled before treatment is begun by taking a drip sample of the completely liquefied oil, commencing after the oil has started to run freely. Where this cannot be done samples shall be taken from various depths of the storage tank. Samples of oil shall be drawn from each cylinder charge of blocks as treated, and tested at the plant as deemed advisable. Samples taken from the treating tank during the process of the work shall at no time show an accumulation of more than 2 per cent, of sawdust and dirt or other foreign matter, or more than 3 per cent, of water. Due allowance shall be made for such accumulation of foreign matter by injecting a corresponding quantity of oil into the blocks. The oil with which the blocks are treated shall be at least 75 per cent, straight coal tar product and shall comply with the following requirements : (a) The specific gravity shall not be less than 1.08 and not more than 1. 12 at 38° C. (b) It shall contain not more than 3 per cent, of matter insoluble in hot benzol and chloroform. (c) When subjected to distillation, according to the method herein- after described the amount of distillate based on water free oil shall be as follows : Up to 200° C, not more than 1.5 per cent. Up to 235° C, not more than 20 per cent. Up to 315° C, not less than 20 per cent or more than 50 per cent. The fraction distilling between 235° and 315° C. shall have a gravity of not less than 1.03 at 38° C. One hundred grams of oil shall be weighed out in a glass retort preferably made of Jena glass, having a capacity to bend of neck of 250 cc. A condensing tube, air cooled, is attached to the retort of such length that the total distance from the tubular to the end of the con- densing tube shall be approximately 60 centimeters. The tubular shall 360 i;ngine:e:ring che:mistry be fitted with a cork through which a nitrogen filled thermometer, about 40 centimeters in length graduated in single degrees and registering to 400° C, shall be inserted in such a manner that the bottom of the bulb shall be J^ inch above the liquid at the time distillation commences and during the progress of the distillation the thermometer must remain in the position originally placed. The first reading on the emergent stem of the thermometer shall be not less than the 50° point and not more than the 80° point and no correction is to be made for the emergent stem. The distillation shall be made in a place free from draughts and the bulb of the retort protected by a shield of heavy asbestos paper, shall be heated by the direct flame of an adjustable burner. The flame shall be regulated in such a manner that the rate of distillation shall continuously be not slower than i drop per second and not faster than 2 drops per second. The distillates shall be collected in weighed Erlenmeyer flasks. If water is present the amount shall be reported separately, all results being calculated on a dry oil basis. For each shipment of blocks made, the contractor or the superintend- ent of the plant of manufacture shall furnish the engineer a certificate to the effect that only oil complying with the foregoing specifications, and of the amount specified per cubic foot has been used in the treatment of the blocks shipped ; and upon completion of the contract shall furnish an affidavit that all charges of blocks manufactured have been treated with oil in compliance with the above specifications. The treatment shall consist of two operations : (a) The application of preliminary steam and vacuum. {b) The injection of a minimum average of 18 pounds of oil into each cubic foot of timber. In addition to this minimum average such additional oil shall be injected into the timber, depending on its physical condition as shall render it possible for the treated blocks to pass a 5 per cent, absorption test as hereinafter specified. Blocks cut from timber allowed under these specifications require longer or shorter periods of treatment in proportion to its being more or less thoroughly cured as to its sap and heart content, and to its pitch content. The following variations in treatment are required for charges of which the averages in weight per cubic foot are to be carefully taken. (rt) Timber — 38-42 pounds per cubic foot. — lyive steam shall be ad- mitted into the cylinder and applied to the blocks, being gradually raised during a period of i hour to 15 pounds boiler gauge pressure and about 185° F., which pressure shall be maintained for 2 hours ; then a vacuum of not less than 22 inches shall be applied for i^ hours, the temperature in cylinder maintained at not less than 155° F. Oil at not less than 180° F. nor more than 190° F. shall then be admitted and the pressure gradually raised during a period of 3 hours i;ngine:e;ring che;mistry 361 to 160 pounds, or until an average of 18 pounds of oil has been forced into each cubic foot of blocks. During this period the temperature of the oil shall not be allowed to fall below 165° F. The free oil shall then be expelled from the cylinder, (b) Timber — 42-44 pounds per cubic foot. — Live steam shall be ad- mitted into the C3^1inder and applied to the blocks, being gradually raised during a period of i hour to 18 pounds boiler gauge pressure and about 190° F., which pressure shall be maintained for a period of 3 hours ; then a vacuum of not less than 2^ inches shall be applied for 2 hours, the temperature in cylinder being maintained at not less than 150° F, Oil at not less than 180° F. nor more than 190° F. shall then be admitted and pressure gradually raised during a period of 3 hours to 165 pounds, or until an average of 18 pounds of oil has been forced into each cubic foot of blocks. During this period the temperature of the oil shall not be allowed to fall below 165° F. The free oil shall then be expelled from the cylinder. (c) Timber — 44-46 pounds per cubic foot. — Live steam shall be ad- mitted into the cylinder and applied to the blocks, being gradually raised during a period of i hour to 20 pounds boiler gauge pressure and about 190° F., which pressure shall be maintained for not less than 4 hours; then a vacuum of not less than 24 inches shall be applied for 2 hours, at temperature in cylinder being maintained at not less than 140° F. Oil at not less than 180° F., nor more than 190° F. shall then be admitted and pressure gradually raised during a period of 3^/2 hours to 165 pounds or until an average of 18 pounds of oil has been forced into each cubic foot of blocks. During this period the temperature of the oil shall not be allowed to fall below 165° F. The free oil shall then be expelled from the cylinder. (d) Timber — 46-50 pounds per cubic foot. — Live steam shall be ad- mitted into the cylinder and applied to the blocks, being gradually raised during a period of 2 hours to 25 pounds boiler gauge pressure and 220° F., which pressure shall be maintained for not less than 5 hours ; then a vacuum of not less than 24 inches shall be applied for 2^ hours, the temperature in cylinder being maintained above 140° F. Oil at not less than 180° F., nor more than 190° F, shall then be admitted and pressure gradually raised during a period of 3>4 hours to 170 pounds or until an average of 18 pounds of oil have been forced into each cubic foot of blocks. During this period the temperature of the oil shall not be allowed to fall below 165° F, The free oil shall then be expelled from the cylinder. In applying the treatment specified, variations and changes may be made from time to time in duration of treatment and in temperatures and pressures used, to suit various gravities of oil and different varie- ties of timber. Such may be specified by the engineer, but shall not be 362 ENGINEE^RING CHEMISTRY continued unless their use is clearly warranted by an improvement in the quality of the blocks manufactured. This shall be demonstrated by tests to be made on samples of each charge treated, as hereinafter specified. Upon the completion of treatment, charges shall be allowed to remain in cylinders for from 30 minutes to i hour, and shall then be withdrawn. The blocks shall be protected from the sun after manufacture, and shall be loaded within 48 hours thereafter for shipment. An inspector appointed by the president will inspect the lumber and other materials used in the manufacture and treatment of the blocks. Any material and blocks not in compliance with these specifications shall be rejected. The inspection shall include the making of such tests upon materials and samples of treated blocks as may be desired by the engineer. The manufacturer shall afford the engineer every facility requested for measuring tanks, cylinders, cages, etc., and for taking and analyzing samples as often as may be deemed necessary, including the use of laboratory and such apparatus as he may require. References. "Investigations on Coal Tar and Some of Its Products." A. R. Warner and W. B. Southerton, Jour. Gas Lighting, Feb. 27, 1912. "Methods for the Examination of Bituminous Road Materials." J^ulletin 38, Office of Public Roads, U. S. Dept. Agr. THE EXAMINATION OF LUBRICATING OILS. The generally accepted conditions of a good lubricant are as follows : 1. Body enough to prevent the surfaces to which it is applied from coming in contact with each other. 2. Freedom from corrosive acids, either of mineral, animal, or vegetable origin. 3. As fluid as possible, consistent with "body." 4. A minimum coefficient of friction. 5. High "flash" and "burning" points. 6. Freedom from all materials liable to produce oxidation or "gumming," or addition of "artificial thickeners." 7. Must not be easily thinned or vaporized by heat or thick- ened by cold. The examinations to be made to verify the above are both chemical and mechanical, and are usually arranged in the follow- ing order : I. Specific gravity. ENGINEERING CHEMISTRY 363 2. Cold test. 3. Viscosity. 4. Iodine absorption. 5. Flash and fire tests. 6. Acidity. 7. Maumene's test. 8. Identification of the oil, whether a simple mineral oil, animal oil, vegetable oil, or a mixture. 9. Coefficient of friction. 1. Specific Gravity. In the chemical laboratory the hydrometers used are generally marked with the specific gravity direct. In the oil trade, how- ever and in general commercial practice the Baume hydrometer is used, and the following precaution is necessary. If the oil is not liquid enough to flow easily, it must be warmed until so, and then tested with the hydrometer. The latter should move easily and freely in the liquid. As all specific gravities are comparable at 60° F., it will be necessary to make correction for temperature; if the temperature of the oil is above 60° F., the readings of the hydrometer are too large; if below 60° F., the readings are too small. To convert Baume degrees into specific gravity the following table is used : °Be. Sp. gr. °B€. Sp. gr. °B€. Sp. gr. °B6. Sp. gr. 10 1. 0000 28 0.8861 46 0.7955 64 0.7216 II 0.9929 29 0.8805 47 0.7910 65 0.7179 12 0.9859 30 0.8750 48 0.7865 66 0.7143 13 0.9790 31 0.8696 49 0.7821 67 0.7107 14 0.9722 32 0.8642 50 0.7778 68 0.7071 15 0.9655 33 0.8589 51 0.7735 69 0.7035 16 0.9589 34 0.8537 52 0.7692 70 0.7000 17 0.9524 35 0.8485 53 0.7650 71 0.6965 18 0.9459 3b 0.8434 54 0.7609 72 0,6931 ^9 0.9396 37 0.8383 55 0.7568 73 0.6897 20 0.9333 38 0.8333 56 0.7527 74 21 0.9272 39 0.8284 57 0.7487 75 0.6829 22 0.9211 40 0.8235 58 0.7447 76 0.6796 23 0.9150 41 0.8187 59 0.7407 77 0.6763 24 0.9091 42 0.8140 60 0.7368 78 0.6731 25 0.9032 43 0.8092 61 0.7330 79 0.6699 26 0.8970 44 0.8046 62 0.7292 80 0.6667 27 08917 45 0.8000 63 0.7254 90 100 0.6363 0.608b 3^4 ENGINEERING CHEMISTRY For liquids lighter than water sp. gr. = q^, . 130 4- Be at 60° F. and we find that 27.2 Baume is equal to 0.8906 specific gravity.^ Fig. 67 represents a Tagliabue hydrometer for oils ; it contains a thermometer, also a scale to make the readings at 60° F. Subtract 1° Baume for every 10° F. above 60° F., and add 1° Baume for every 10° F., below 60° F. Thus, if the hydrometer, when placed in the oil, reads 26° Baume and the temperature of the oil 80° F., the correct reading will be 24.7° Baume at 60° F. The specific gravity test is an important one; by it an ad- mixture of certain oils with mineral oil is indicated. For instance, a lubricating oil of specific gravity 0.915 was found by qualitative analysis to be composed of mineral oil and menhaden oil. Knowing the kinds of oil composing the mixture, and approximation of the per cents, would be obtained as follows : Minera oil Specific gravity = 0.890 (B) Menhaden oil Specific gravity ^^ 0.927 (A) Specific gravity of mixture = 0.915 ( M ) ^ If we assume the standard temperature for this purpose to be 15.5° C. the gravity of any oil at higher or lower temperature can be calculated from the following formula: G = G' + K (T — 15.5°), in which G is the specific gravity at 15.5°, G' the specific gravity at T, T = temp, of room, and K a factor varying with the different oils as follows: Factor for Calculating Specific Gravity of Oils.^ Correction for 1° C. Cod liver oil 0.000646 Olive oil 0.000629 Rape oil ." 0.000620 Lard oil 0.000658 Peanut oil 0.000655 Cottonseed oil 0.000629 Corn oil 0.000630 Sesame oil 0.000624 2 A. E. Ivcach, "Food Inspection and Analysis," p. 372, gives the formula for ordinary use as G = G' + 0.00064 (T — 15.5° C). m Fig. 67. ENGINEERING CHEMISTRY 365 Let A -- M = C. (0.927 — 0.915 = 0.012) M — B = D. (0.915 — 0.980=0.025) ^, D . r A / o 02S\ Then ^ , .^ = per cent, of A ( ^ ) , C + D ^ V 0.037/ and — — = per cent, or Bl I C + D ^ V 0.037/ The result being Per cent. Menhaden oil 67.5 Mineral oil 32.5 A more rapid method is graphically thus; in Fig. 68 let the '- ■ 1 11 1 1 .yz'/L' )H It.^ ^-=^ 1 ■ J920 i - " " ""iiji'^^':" ""itil: :±±^-"f i 4^ ^10 zh Zt--' ! _ ^. ^ju __ __ --+--- - - ^ - 1 " ~ T~ ^^ , <-.... 1 ..... .QOd -^ -.--'■ J _ - »^l .-K^ '1 ?! ^ i -•''' J ■LI-'' '^ -i -, ,- ^.^-^^ .Hm^f .^= 6 7./ / _^- jU.lLJZ -^ ^\ 1 ?J 1 _ t^ »88fl-ti_ ^ ^ "*■ ,&^ 1 ^ "*" ! 1 1 1 ►87( it: i 1 ■ ■' ■ ■ 1 r '■ ; 1 1 1 1 1 .ftfiVl 1 . i i I'll 1 ' ! i ^1 PptI- Cpi,* \ 1 1 .850 ± __ ^iT ^^"* 1 : 111 10 zo 30 60 TO «0 lOQ 40 •'i'^ Fig. 68. abscissas represent per cents, and the ordinates the specific gravi- ties. From the point indicated (on the line AB) 0.915 the spe- cific gravity of the mixture the per cents, are read on abscissa line 67.5 for A and 32.5 per cent, for B. 366 e:ngine:ering chemistry Another instrument used for the determination of the specific gravity of oils is the Westphal balance, as improved by Williams.^ This apparatus (Fig. 69) is very accurate and should be used as a check determination of the gravity made by the hydrometer. Fig. 69. Directions for Using the Williams-Westphal Balance. To Determine the Specific Gravity of Liquids: — Hang ther- mometer plummet upon hook and after leveling the instrument, bring beam to equilibrium in air by turning the adjusting weight on threaded portion of the beam. It is desirable to verify this adjustment by immersing thermometer plummet in distilled water at 15° C. and hanging one of the largest horse-shoe weights upon the hook; this should exactly restore the beam to equilibrium. Care should be taken that the thermometer plummet be fully immersed throughout the complete swing of the beam. After wiping plummet it is immersed in the solution to be tested, this having been brought to the proper temperature, and weights added, beginning with largest until equilibrium of the beam is ^ "For Determination of the Spec. Grav. of L,iquids and Solids, both Soluble and Insoluble in Water." Dngineering chemistry 367 restored. If the solution is heavier than water one of the largest weights must hang on the hook at the end of the beam. In read- ing the weights, the large weight at the end of the beam indi- cates one, the other large weight indicates tenths, corresponding to the position it occupies on the graduated portion of the beam, the next smaller weight indicates hundredths, the next smaller thousandths and the smallest ten-thousandths. To Determine the Specific Gravity of Solids Insoluble in Water: — The plummet is first wiped dry, the pan hanger is attached at the threaded end of the beam, and water at 15° C. is added until the lower pan is immersed throughout the complete swing of the beam. The beam is now adjusted to equilibrium by turning the weight on threaded portion of the beam. The sub- iiiiiiii^ Fig. 70. Stance is now placed in the upper pan and the horse-shoe weights added as above until equilibrium is restored. This gives weight of substance in air. The substance if in the form of a fragment is now transferred to the lower pan, if in the form of a powder the upper pan with its contents is transferred to the lower posi- 368 ENGINEERING CHEMISTRY tion, and after wiping lower pan it is transferred to the upper position. Weights are now added until equilibrium is restored, this reading giving weights of substance in water. Calculations : — Weight in air ^ . _ ^^^ . , — : . ° ^^ ■ , — -. ^Specific gravity. Weight in air — Weight m water To Determine the Specific Gravity of a Substance Acted Upon by Water: — For example, Portland Cement. Select some liquid such as carbon tetra- chloride, carbon bisulphide or benzine, which has no action upon the substance. Immerse lower pan in this li'quid, and after adjusting beam to equilibrium, de- termine weight in air and weight in this liquid as above, which will give the apparent specific gravity of the substance. To Obtain True Specific Gravity : — The specific gravity of the liquid used is determined with thermom- eter plummet and the apparent specific gravity multi- plied by specific gravity of liquid used, gives the true specific gravity of substance. If oil is too thick at ordinary temperatures, for the determination of the gravity, it should be heated suffi- ciently and the modified Westphal balance (Fig. 70) used. If only small amounts of the oil are obtainable a small picnometer, or an Arseo-picnometer of Eichhorn can be used. The important feature of this instru- ment consists in a small glass bulb (attached to the spindle), which is filled with the liquid whose gravity is to be taken. Thus instead of floating the entire ap-| paratus in the test fluid, only a very small quantity of the latter is required. The glass bulb when filled with the test fluid, is closed by means of an accurately fitting glass stopper, and the instrument is then placed in a glass cylinder filled with distilled water at 17.5° C. (Fig. 71). pig" The gravity is then at once shown on the divided scale in upper portions of the spindle. ENGINEERING CHEMISTRY 369 Tabi^e of Specific Gravity of Oils Used with Minerai, OiES FOR Lubricating Purposes. Sp. Gr. Sperm oil 0.883 Olive oil 0.916 Cotton-seed oil (white) 0.925 Cotton-seed oil (brown) 0.930 Castor oil 0.960 Mineral oil 0.860 to 0.925 Dolphin oil 0.922 Neatsfoot oil 0.915 Lard oil 0.915 Tallow oil 0.903 Menhaden oil ' 0.928 Rape-seed oil 0.916 Resin oil .0.980 to 1.05 Blown oils, made by oxidation of rape-seed oil, cotton- seed oil, etc 0.930 to 0.970 Corn oil 0.922 2. The Cold Test. The degree at which an oil becomes semi-solid and refuses to flow^ freely is considered the cold test and is performed as follows : Fifty cubic centimeters of the oil are transferred to a narrow bottle (capacity lOO cc), stoppered with a rubber stopper, through which is inserted a thermometer, the bulb of which reaches an inch or more into the oil. The bottle is placed in a mixture of ice and salt, or other freez- ing compound, and retained there until the oil becomes solid. It is then removed and allowed to warm until the contents become somewhat thinner in consistence. The bottle is inclined from side to side until the oil begins to flow, when the temperature is taken. At this particular temperature the oil is neither at its normal fluidity, nor is it solid, and while this method does not correctly indicate the exact temperature of the solidifying-point, it does show the point at which the oil ceases to flow readily, the im- portant one to the oil inspector. 24 370 ENGINEJKRING CHEMISTRY In lubricating oils, to be used in railroad practice, this cold test is a vital one, and receives in the laboratories of the different railroads of the United States considerable attention. A mineral lubricating oil, non-paraffine, of good quality, does P imiiiia Fig. T2. not show any material difference in its consistency at 25° C. or 10° C, but a radical change would be indicated at 10° C. if some of the animal or vegetable oils were a component. Fig. 72 represents the glass apparatus with thermometer ar- ENGINEEiRING CHEMISTRY 37I ranged for the cold test, and is surrounded by any mixture capable of producing the required degree of cold. The following determinations of the cold test, made in my laboratory, will show the wide range in this regard between many of the oils, used in lubrication: Degrees F. Elain oil 42:8 Saponified red oil 41.0 Prime neatsfoot oil 24.8 White neatsfoot oil — Pure hoof oil 42.8 Prime lard oil 44.6 No. I lard oil 44.6 XXX lard oil 37.4 American sod oil 33.8 English sod oil 75.0 Tallow oil 79.0 Dog fish oil 19.0 Right whale oil (Pacific) 32.0 Unbleached bowhead whale oil (Pacific) 19.0 Bleached whale oil (Pacific) 8.6 Natural sperm oil (Pacific) 32.0 Bleached sperm oil (Pacific) 24.8 Herring oil (Pacific) 32.0 Natural winter sperm oil (Atlantic) 30.2 Bleached winter sperm oil (Atlantic) 24.8 Natural spring sperm oil (Atlantic) 50.0 Bleached spring sperm oil (Atlantic) 46.0 Natural winter whale oil (Atlantic) 28.0 Bleached winter whale oil (Atlantic) 23.0 Natural spring whale oil (Atlantic) 41.0 Bleached spring whale oil (Atlantic) 35.6 Prime crude menhaden oil 24.8 Brown strained menhaden oil 19.5 Light strained menhaden oil 19.5 Natural winter menhaden oil 16.O Bleached winter menhaden oil 10.4 Extra bleached winter white menhaden oil 12.0 Bank oil 24.8 Straits oil 19.5 Sea elephant oil 41.0 Black fish oil 1 7.6 3/2 ENGINEERING CHEMISTRY Degrees F. Resin oil, ist run 37.4 Resin oil, 2d run — 2.2 Resin oil, 3d run — 4.0 Castor oil — 0.4 Crude cotton-seed oil 19.4 Prime summer yellow cotton-seed oil 23.0 Off quality summer yellow cotton-seed oil 21.0 Prime quality winter cotton-seed oil 14.0 Off quality winter cotton-seed oil 17.6 Prime quality summer white cotton-seed oil 26.6 Off quality summer white cotton-seed oil 17.6 Prime quality winter white cotton-seed oil 15.8 Off quality winter white cotton-seed oil 23.0 No. I French degras oil 77.0 No. 2 French degras oil 77.0 English degras oil 64.4 Olive oil 37-4 Oleo oil 750 3. Viscosity. The test for viscosity or ''body" of lubricating oils is an im- portant one. By comparison with standards an oil may be rated as to its viscosity thus giving one of the values required for a lubricant. Engler's apparatus — probably the first used for this purpose, is made of metal (copper) and its general plan is shown in Fig. 73- This instrument is the standard for determining the viscosity of oils in Germany, and is also a standard in this country. Recommended by U. S. Bureau of Standards for use unless an- other form of viscosimeter is called for in the specification. In using this instrument the viscosity of an oil is stated in seconds required for 2CX) cc. of the oil to run into the flask, 240 cc. of the oil being placed in the viscosimeter. Water usually requiring from 50 to 53 seconds at 20° C. Heat can be applied to the water bath, the viscosity being de- termined at any temperature required up to 100° C. Higher temperatures to 360° C. can be secured by filling the outer vessel with paraffine instead of water. ENGINEERING CHEMISTRY 373 r Engler recommends that all viscosities be compared with water thus : If water requires 52 seconds for delivery of 200 cc. into Fig- Tl- — Engler Viscosimeter. the receiving flask, and the same amount of an oil under examina- tion requires 130 seconds, the ratio is determined by 130 52 2.50» the oil thus having a viscosity of 2.5 times that of water. "The American Society for Testing Materials," Report of Committee D-2, state as follows : In case it is desired to correct for specific gravity of the oil, 374 EJNGIN DURING CHEMISTRY the following formula which gives the results in specific viscosity can be used : time of efflux of oil Sp. viscosity = sp. grav. X 7-32. time of efflux of water If it is necessary to use a quantity of oil less than 240 cc. the following quantities can be employed and multiplied by the cor- responding factor :i Amount of oil put in, cc. Amount of oil run out, cc. Factor to change to 200 cc. run out and 240 cc. put in 120 100 1.65 The Saybolt Standard Universal Viscosimeter. The Saybolt Standard Universal Viscosimeter at the present time is used as the standard instrument by nearly every oil manu- facturer, dealer and user. The older types of machines are adapted to just one or two specific temperatures at which the viscosity of the oil is tested. Whereas the Saybolt Standard Universal Viscosimeter is just what the name implies and covers the following wide range: (i) Cylinder, valve and similar oils with bath at 212° and oil to be tested at 210° (2) Reduced black oils, with bath and oil at 130° (3) Neutral, spindle, paraffine and other distilled oils at. 100° (4) Or 4 at 70° It will be seen by the above that the range in temperatures covers practically every test that is necessary to determine the viscosity value of an oil. Hence the advantages of the above instruments are : First. That it saves expense due to the fact that a number of instruments are combined in one. Second. It eliminates varying conditions. Third. It lessens the possibilities of error due to the use of various instruments. ^ Gaus, Chemische Revue der Fette und H-arz-Industrie, Vol. VI, p. 221. ENGINKE^RING CHEjMISTRY 375 The instrument is so constructed that it can be adapted to any laboratory. It is equipped with an electric heating device, steam heating device and gas heating device. The bath prescribed is a paraffine pale engine oil with a flash of about 350° F. to 400° F. to be used for viscosity tests at all temperatures. This is recommended over the old method of a salt and water bath because oil and water are antagonistic. If Fig. 74. — Saybolt Standard Universal Viscosimeter. The cut shows one-half of stand jacket and of bath vessel cut away to expose inside parts. for instance, a few drops of the water by chance became mixed with the oil upon going through the viscosity tube proper, the re- sult would be erroneous to a very marked degree. However, care should be taken not to allow any of the bath to become mixed with the oil to be tested. The stirring mechanism is very convenient for the tester and enables him to obtain a uniform temperature throughout the bath. 376 EiNXINEERING CHEMISTRY Fig- 75- — Tagliabue's Improved Viscosimeter. DNGINE^ERING CHEMISTRY 377 Taking viscosimetry as a whole the Saybolt Universal Viscosi- meter is a most up-to-date, accurate and convenient instrument for present day oil testing. Tagliabue's viscosimeter (Fig. 75), is used to a very large extent by the manufacturers of lubricating oils in the United States. The following are the directions for its use : To Te;st the) Viscosity of O11.S at 212° Fahrenheit. Pour water into boiler through opening A, after unscrewing safety valve until water gauge shows that the boiler is full. See that stop-cock B is open, making direct connection between boiler and upper vessel which surrounds the receptacle in which the oil to be tested is placed: Place wire holder in set nut C and suspend thermometer so that the bulb of thermometer will be about y^ inch from the bottom of oil bath. Then after care- fully straining 80 cc. of the oil to be tested, which of course must be warmed in the case of very heavy oils, pour same into oil bath. Close stop-cocks D and E. Screw the extension F with rubber hose attached into coupling G and let the open end of hose be immersed in a vessel of water which will prevent too large a loss of steam. Place lamp or Bunsen burner under boiler; screw steel nipple marked 212 on to stop-cock H; the apparatus is now ready to use. After steam is generated wait until ther- mometer in oil bath shows a temperature of 210^ or 211°; then place the 60 cc. test glass under stop-cock H so that the stream of oil strikes the side of test glass, thereby preventing forming of air bubbles; and when the thermometer indicates its highest point open the faucet H simultaneously with the starting of the watch which is supplied with each instrument. When the running oil reaches the 60 cc. mark in the neck of the test glass the watch is instantly stopped and the number of seconds noted. Then multiply the number of seconds by 2, and the result will be the viscosity of the oil. For example: If 60 cc. of oil runs through in loij^ seconds the viscosity would be 203. It is best to repeat the test until sufficient skill is attained by practice for uniform results. 37^ • ENGINEERING CHEMISTRY It is also necessary to keep the oil well stirred before making test in order to have the oil at a uniform temperature. To Test the Viscosity oe O11.S at 70° Fahrenheit. Screw the steel nipple marked 70° on the faucet H, close stop-cock B, thereby closing communication between boiler and upper vessel. Also close stop-cock E. Fill upper vessel, through opening G, with water, as near a temperature of 70° as possible, also having the oil to be tested, at the same temperature. Hang the thermometer in position, and after stirring the oil thoroughly, blow through the rubber tube at D, thoroughly mixing the water. Should the thermometer show a lower or higher temperature than 70°, add cold or warm water until the desired temperature is obtained. Then measure carefully 90 cc. of oil to be tested, placing it in the machine, and when everything is ready, open stop-cock, and start watch at the same time, and allow 70 cc. to pass through the nipple, and as soon as the test tube is filled to the 70 cc. marked in the neck, turn off the stop-cock and stop watch at the same moment. Should it take the 70 cc. 96 seconds to run through the nipple, multiply this by 2, and you will have the viscosity of the oil, which is 192 seconds. (This multi- plication by 2 is to render readings comparable with the Saybolt viscosimeter.) "The Doolittle Torsion Viscosimeter." ^ used in the railroad laboratories of the Philadelphia and Read- ing Railroad Company, is briefly described as follows : A steel wire is suspended from a firm support and fastened to a stem which passes through a graduated horizontal disk, thus measuring accurately the torsion of the wire. The disk is ad- justed so that the index point reads exactly zero, thus showing that there is no torsion in the wire (Fig. 75a). A cylinder 2 inches long by i^ inches in diameter, having a slender stem by which to suspend it, is then immersed in the oil and fastened by a thumb-screw on the lower part of the stem to the disk. The oil is surrounded by a bath of water or paraf- ly. Am. Chem. Soc, 15, 173. EJNGINKERING CHEMISTRY 379 fine wax according to the temperature at which it is desired to take the viscosity. This temperature being obtained while the disk is resting on its supports, the wire is twisted 360° by means of the knob at the top. The disk being released, the cylinder ro- tates in the oil by virtue of the torsion of the wire. Fig. 75a. — Doolittle Viscosimeter. The action now observed is identical with that of the pendulum. If there was no resistance to be overcome, the disk would re- volve back to zero, and the momentum thus acquired would carry it to 360° in the opposite direction. What we find is that the re- sistance of the oil to the rotation of the cylinder causes the revolu- 380 E^NGINEERING CHE:mISTRY tion to fall short of 360° and that the greater the viscosity of the oil the greater will be the resistance and hence the retardation. We find this retardation to be a very delicate measure of the vis- cosity of an oil. There are a number of ways in which this viscosity may be ex- pressed, but the simplest is found to be directly in the number of degrees of retardation between the first and second complete arcs covered by the pendulum. For example, suppose we twist the wire 360° and release the disk so that rotation begins. In order to obtain an absolute reading to start from, which shall be inde- pendent of any slight error in adjustment, we ignore the fact that we have started from 360°, and take as our first reading the end of the first swing. Suppose our readings are as follows : Right, 350; left, 338; right, 328; and keeping in mind the vi- brations of the simple pendulum we will see at once that we have read two complete arcs whose difference is 2.2'^ computed as follows : 1st arc, Right 350° + Left 338° = 688° 2d arc, Left 338° + Right 328° = 666° 22° retardation. In order to secure freedom from error we take two tests — one by rotating the wire to the right, and the second to the left. If the instrument is in exact adjustment these two results will be the same, but if it is slightly out, the mean of the two readings will be the correct reading. It will also be noticed that if the exact retardation due to the oil alone is to be obtained we must subtract the factor for the resistance due to the air and the wire itself. These are readily obtained by allowing the cylinder to rotate in the air and deter- mining the retardation exactly as we have done above. This fac- tor remains constant for each instrument and is simply deducted from all results obtained. The Redwood^ Viscosimeter is on the general principle of the Engler. This is the standard viscosimeter for the English oil trade. V- Soc. Chem Ind.,5, 158. ENGlNEBiRING CHE:mISTRY 381 4. Iodine Absorption. The determination of the iodine absorption of an oil is prob- ably the most important chemical test for recognition quan- titatively in a mixture of animal or vegetable oils with mineral oils. Introduced by Hubl^ it has since maintained this position, though other chemists have introduced the bromine absorption and others of similar character. They have not attained the uni- versal confidence in the iodine process. In a mixture of two fatty oils with a mineral oil, the best results are obtained by saponifying and separating the fatty acids from the mineral oil. The iodine absorption of the mixed fatty acids is then taken, and where the nature of them has already been shown by color tests, etc., their proportion can be indicated by the following formula : 100 (I — n) Where x = the percentage of one fat, 3; = the percentage of the other, I =; iodine degree of mixture, m z=z iodine degree of fat x, n = iodine degree of fat, y. The method is as follows : Twenty-five grams of iodine and 30 grams of mercuric chlor- ide are each dissolved in 500 cc. of 95 per cent, alcohol, uniting the two solutions, and allowing to stand several hours before use. It is then standardized by tenth-normal thiosulphate sodium solution. The process of the determination of the iodine absorp- tion of an oil is as follows: o.i to 0.5 gram of the fat or oil is dissolved in 10 cc. of purest chloroform in a well-stoppered flask, and 20 cc. of the iodine solution added. The amount must be finally regulated so that after not less than 2 hours' diges- tion the mixture possesses a dark brown tint ; under any circum- stances it is necessary to have a considerable excess of iodine (at least double the amount absorbed ought to be present), and the digestion should be from 6 to 8 hours. Some potassium iodide solution is then added, and the whole diluted with 150 cc. of ^ Ding. poly. J., 253, 281. 382 ENGINEERING CHEMISTRY water, and tenth-normal thiosulphate solution delivered in till the color is nearly discharged. Starch is then added, and the titration finished in the usual way. If more than two fatty oils are present in a mixture with min- eral oil, the method of Warren^ can be used. The following determinations of the iodine absorption made in my laboratory are indicative of the variations of the absorption by the different oils : Prime lard oil No. I lard oil XXX lard oil Oleo oil • Prime neatsfoot oil Horse oil Natural bow-head whale oil Natural winter whale oil Extra bleached winter white oil Bleached spring winter white oil Crude sperm oil Prime quality winter white cotton seed oil. . • Prime quality summer white cotton seed oil • Prime quality winter yellow cotton seed oil. • Prime quality summer yellow cotton seed oil Olive oil Herring oil Dog-fish oil Porpoise head oil Resin oil, second run Resin oil, third run Rape oil Corn oil 76.4 77.2 69.8 69.9 65.1 65.6 51.6 51.6 80. 1 80.0 82.3 82.5 130.5 13'. I 121. 1 126.0 124.9 1 26. 1 1 26. 1 126.2 82.3 82.3 116. 4 114.9 110.2 110.6 II5-9 118.6 104.0 104.4 81.0 83.0 122. 1 123.8 102.7 104.7 28.9 29.1 92.1 93-4 90.4 92.2 94.0 106.8 III.O 115.0 5. Flash and Fire Test. The flash point is the degree of temperature at which ignitable volatile vapors are given off by the oil, producing a flash when brought in contact with a small flame. The fire test is a contin- uation of the flash test until the oil permanently ignites. The method used by the chemists of the Tide Water Oil Co., N. J., is as follows : The Cleveland cup is used. Fig. 76. A bath is supplied with this cup but it is not used either by this laboratory or the labora- tories of the Standard Oil Co. The cup is filled about ^ inch 1 Chem News, 62, 215 :/. Anal. Appl. Chem., 5, 215. ENGINEE^RING CHE^MISTRY 383 from the top, and a thermometer (Tagliabue's bulb immer- sion thermometer showing corrections for total immersion) is Fig. 76. — Apparatus for Determining the Flashing and Burning Points of Combustible Liquids Cleveland Cup. suspended so that the bulb is entirely immersed in the oil at the center of the cup without touching the bottom. Heat is applied 384 ENGINEERING CHEMISTRY by means of a Bunsen burner so that the temperature is raised at the rate of 10° F. a minute. As the flashing point is ap- proached, a test is made for every rise of 2° or 3° by slowly passing the test flame across the cup horizontally near the ther- mometer. The test should be made in a place free from draughts, and the test flame should be about 5 millimeters long. Any variation in these conditions, either in the size or shape of the cup, and rate of heating, or the method of testing, may lead to an appreciable error. It has been our experience, that a test flame longer than 5 millimeters gives low results, and one smaller gives high results. The temperature of the first flash is recorded as the flashing point and is reported in even 50° F. i. e. — 532° F. would be read 530° F. and 533° F. would read 535° F. Feash Point (Ceosed Cup). Foreign countries usually test high flash products by the Pensky-Martens closed cup.^ We are, therefore, obliged to use this cup for testing export lubricating oils. This cup is prac- tically of the same construction as the Abel cup (which is used for refined oils). The rate of heating is the same 10° F. per minute, as with the open-cup method. Care must be exercised in obtaining samples free from water for this test, as the presence of aqueous vapor prevents the oil from flashing. The heating and testing is continued in the same way until, on the application of the test flame, the sample takes fire which temperature is recorded as the fire test or burning point. Pensky-Martens Test. — Where great accuracy is required the Pensky-Martens tester should be employed. The method of operating is as follows : Referring to Fig. yy B is the oil container, which is placed in a metal heating vessel H, provided with a mantle L in order to protect the heating vessel from loss of heat by radiation. The oil cup B is closed by a tightly-fitting lid (show^n in plan). Through the center of the lid passes a shaft carrying the stir- ring arrangement, which is worked by means of the handle /. In another opening of the cover is fixed a thermometer. The ^ See Fig. 77. r E^NGINKERING CHEMISTRY 385 lid is perforated with several orifices, which are left open or covered, as the case may be, by a sliding cover. This can be rotated by turning the vertical spindle by means of the milled head G. By turning G, an opening of the slide can be made to Fig- 77- — Pensky-Martens Tester. coincide with an orifice in the cover, and simultaneously a very small flame, burning at the movable jet B, is tilted on to the surface of the oil. The test is performed by filling the oil into the oil cup up to a certain mark, fixing the cover, and heating the oil somewhat rapidly at first, until its temperature is about 30° C. below the 25 386 ENGINEERING CHEMISTRY expected flash-point. The temperature is then allowed to rise very slowly by making suitable use of the wire gauze shown in the figures, so that the rise of temperature within ^ minute does not exceed about 2° C. From time to time the milled head G is turned and the flame tilted into the oil cup. The temperature at which a slight explosion is produced is noted as the flash point of the oil. 6. Acidity. Acidity in oils is generally due to a partial decomposition of the oil with liberation of fatty acids. These latter act as corro- sive agents, attacking the metal bearings of machinery, forming "metallic soaps" and producing gumming and thickening of the lubricant. Properly refined mineral oils are free from acidity, but nearly all animal and vegetable oils possess it more or less. In palm oil, for instance, the free fatty acids vary from 12 to 80 per cent. In 89 samples of olive oil intended for lubricating purposes, D. Archbutt^ found from 2.2 to 25.1 per cent, of free acid (oleic, the mean being 8.05 per cent. The action of free acid on journals, bearing, etc., as a corro- sive agent, has led many engineers to include a test of free acid direct upon copper and iron. This is done by suspending weighed pieces of sheet copper and iron in the different oils, for a number of days, heating if neces- sary and determining the amount of metal dissolved by the oils. While this test may be indicative of the acidity of oils, no ratio exists between the action upon gopper and iron or even between the oils themselves in this respect, owing to the varying quantity of acid in the same oils. Oleic acid cannot be present as a constituent of a pure mineral oil ; still the acid test should be made, since poorly refined mineral oils are liable to contain small amounts of sulphuric apid left in the process of refining. The sulphuric acid is easily indicated by warming some of the oil with distilled water, adding a few drops 1 Analyst, 9, 171. Engineering chemistry 387 of hydrochloric acid (dilute) and solution of barium chloride. A white cloud of precipitate shows the presence of sulphuric acid. The following is the method for determining the acidity of oils as used in many of the railroad laboratories : Materials Required. 5^2 dozen 4-ounce sample bottles. 3 10 cc, pipettes, or i£ desired a balance weighing milligrams. 1 30 cc. burette, graduated to tenths (burette holder if desired), with pinch-cock and delivery tube. 2 ounces alcoholic solution of turmeric. 2 quarts 95 per cent, alcohol to which ^ ounce dry carbonate of soda has been added and thoroughly shaken. I quart caustic potash solution, of such strength that 31^2 cc. exactly neutralize 5 cc. of a mixture of sulphuric acid and water, which contains 49 milligrams H2SO4 per cubic centimeter. Operation. Take about 2 ounces of the clear alcohol and add a few drops of the turmeric solution, which should color the alcohol red, warm to about 150° F., then add 8.9 grams of the oil to be tested and shake thoroughly. The color of the solution changes to yellow. Fill the burette to the top of the graduation with caustic potash solution, and then run this 'solution from the burette into the bottle, a little at a time, with frequent shaking, until the color changes to red again. The red color must remain after the last thorough shaking. Now read off how many cubic centimeters and tenths of the caustic potash solution have been used, and this figure shows whether the material meets specifica- tions or not. To determine the free acid in tallow, everything is done ex- actly as above described, except that the tallow is melted before it is added to the alcohol. Ten cubic centimeters of extra lard oil, at ordinary tempera- tures, and the same amount of melted tallow at 100° F., weigh almost exactly 8.9 grams. In ordinary work, therefore, it will probably not be necessary to weigh the oil or tallow. Measure- ment with a 10 cc. pipette, will usually be sufficiently accurate, provided the pipette is warmed to about 250° F., and allowed to 388 i:nginee:ring chemistry drain, the last drops being blown out. In case of dispute, how- ever, the balance must be used. (P. R. R. method.) Lard and tallow are very liable to have considerable amounts of free acid. The specification of purchase, therefore, generally states the limits of free acid permitted. Freje Acid Test. (Amer. Society for Testing Materials. ) "About 10 grams of the oil are weighed into a 200 cc. Jena Erlenmeyer flask, 60 cc. of neutral alcohol added, the mixture warmed to about 60° C. and titrated with N/6 KOH, using phenolphthalein, the flask being frequently and thoroughly shaken. "The result in the case of a mineral oil is usually reported in percentage of sulphuric anhydride (SO3) : with an organic oil, in percentage of oleic acid. "It is suggested that it be reported, as with the saponification value, as the number of milligrams of KOH necessary to neu- tralize the acidity of i gram of oil." Specifications for Lard Oil. Must be of the best quality and made from fresh lard ; to be pur- chased and inspected by weight; the number of pounds per gallon to be determined b}^ the specific gravity of the oil at 60° F. multiplied by 8.33 pounds, the weight of a gallon (231 cubic inches) of distilled water at the same temperature. Oil will not be accepted which — I, Contains admixture of any other oil. II. Contains more acidity than the equivalent of 2 per cent, of oleic acid. III. Shows a cold test above 42° F. IV. Shows coloration when tested with nitrate of silver, as described below. V. A half pint of the oil placed in an ordinary hand lamp with- out a chimney must burn with a clear, bright flame until 90 per cent, of the oil has been consumed ; the lamp to be placed where it will not be affected by draft or air cur- rents and the wick not to be touched during the trial. I. Test of Lard Oil. — The cold test of oil is determined as follows: A couple of ounces of oil is put in a 4-ounce sample bottle and a ther- Engine;ertng chemistry 389 mometer placed in it. The oil is then frozen, a freezing mixture of ice and salt being used if necessary. When the oil has become hard, the bottle is removed from the freezing mixture and the frozen oil allowed to soften, being stirred and thoroughly mixed at the same time by means of the thermometer until the mass will run from one end of the bottle to the other. The reading of the thermometer, when this is the case, is regarded as the cold test of the oil. 2. The nitrate of silver test is as follows : Have ready a solution of nitrate of silver in alcohol and ether, made on the following formula: Nitrate of silver i gram Alcohol 200 grams Ether 40 grams After the ingredients are mixed and dissolved allow the solution to stand in the sun or in diffused light until it has become perfectly clear. It is then ready for use, and should be kept in a dim place and tightly corked. Into a 50 cc. test tube put 10 cc. of the oil to be tested (which should have been previously filtered through washed filter paper) and 5 cc. of the above solution ; shake thoroughly and heat in a vessel of boiling water 15 minutes, with occasional shaking. Satisfactory oil shows no change of color under this test. 3. For the burning test an ordinary tin hand lamp to conform to the following description will be used : Diameter of lamp at base, s% inches ; height of cylindrical portion, 2j4 inches ; height of top of burner from bottom of lamp, 2% inches. The burner will consist of two conical tubes placed side by side, each iJ/2 inches in length, 15/64 inch inside diameter at top, and 11/32 inch inside diameter at bottom. The wick will consist of a sufficient number of threads of ordinary cotton lampwicking in each tube to make a prop- erly fitting wick for lard oil. Inspection and Delivery. I. Before acceptance the oil will be inspected. Samples of each lot will be taken at random, the samples well mixed together in a clean vessel, and the sample for test taken from this mixture. Should the mixture be found to contain any impurities or adulterations, the whole delivery of oil it represents will be rejected. Maumene's Test. The rise of temperature produced when sulphuric acid is brought in contact with certain oils was first investigated by 390 EJNGIN^KRING CHEJMISTRY Maumene, and the results of his experiments pubHshed in Comptes Rendus, 35, 572. When a mixture of oils has been analyzed and the components recognized, the proportions oftentimes can be determined by this reaction; that is to say, suppose the oil under examination to show a rise of temperature of 80° C, and the oils found by analysis to be lard oil and menhaden oil, their relative proportions can be determined by the following formula : w. --■t^:- w, --'t^- Wj = proportion by weight of menhaden oil ; Wj = proportion by weight of lard oil ; Wg = weight of mixture ; t^ = temperature of menhaden oil ; /j = temperature of lard oil ; t^ = temperature of mixture. The method is as follows: Fifty grams of the oil are placed in a narrow tall beaker and 10 cc. of chemically pure sulphuric acid added drop by drop with stirring and the rise of temperature during the operation noted. Lard oil alone when treated with sulphuric acid gives a rise of temperature of 104° F. ; menhaden oil, under similar conditions, a rise of 260° F. Using these values in the above formula we obtain 54.6 per cent, lard oil and 45.4 per cent, menhaden oil. In the mixture containing a mineral oil mixed with animal, marine, or vegetable oil, the distinction would be even more pro- nounced, since the mineral oil shows but a very slight increase of temperature (generally from 35° to 41° F.). The increment of temperature would be dependent upon the other oil added to the mineral oil. Briefly stated, the rise of temperature of the following oils would be : EJNGIN EARING CHEMISTRY 391 Lard oil Tallow oil Neatsfoot oil Oleo oil •■ • Elain oil Sperm oil Whale oil Menhaden oil Dog-fish oil Cod liver oil Crude cotton-seed oil • . Rape oil Castor oil Olive oil Resin oil Mineral lubricating oil Earth nut Sea elephant Corn oil Name of observer Maumen^ °F. 104 105-109 1^3 Schaedler 215-217 136 116 107 217 156 118 109 82 i52 152 Archbutt 109 98 123 197 253-262 158 114 [ 05-1 13 I 16-140 Allen OF. 105 II3-II6 258 235 152-156 149 105-109 64-71 37-39 Stillman °F. 102.2 102 104 98 100 118 197 262 176 230 165 140 113 107 50 37 149 185 Attention is drawn to the differences in the determination in resin oil. Resin oil of the first run is white, opaque, thick liquid con- taining all of the water of the resin from which it is distilled, and it is this water that causes the rise of temperature above 10° when the oil is mixed with the sulphuric acid. Resin oils of the second and third runs are clear, limpid, dark red colored fluids, practically free from water, and when treated with sulphuric acid do not indicate more than 50° F. rise of tem- perature. From these tests it is concluded that both Schaedler and Allen tested resin oil that was a mixture of the first and second runs, or of an oil not properly separated into the different distillates. 8. Separation of Mineral Oil from a Vegetable or Animal Oil. Ten grams of the oil are weighed out in a dry weighed beaker (250 cc), and to it are added 75 cc. of an alcoholic solution of potash (60 grams of potassium hydroxide to 1,000 cc. of 95 per cent, alcohol), and the contents evaporated until all the alcohol 392 e:ngine;e;ring chemistry is driven off. In this process, if any animal or vegetable oil is present, it is formed into a soap by the potash, while the mineral oil is unacted upon. Water (75 cc.) is now added and the ma- terial w^ell stirred to insure complete solution of the soap, and then it is transferred to a separatory funnel (Fig. 78), 75 cc. of sulphuric ether added, corked, the liquid violently agitated and Fig. 78. allowed to stand for 12 hours. Two distinct liquids are now seen, the lower, the solution of the soap, the upper, the ether solu- tion (colored, if mineral oil is present, colorless, if not). The aqueous solution is drawn off in a No. 3 beaker, the ethereal solu- tion remaining in the separatory funnel. The former is placed ^NGINKEJRING CHEMISTRY 393 on a water-bath, heated for half an hour, and until all traces of ether (which is absorbed by the water in a very small amount) is gone. The solution is allowed to cool, diluted somewhat with water, and made acid with dilute sulphuric acid. Any animal or vege- table oil present will be indicated by a rise of the fatty acids to the surface of the liquid. (In this reaction the sulphuric acid de- composes the soap, uniting with the potash to form a sulphate of potash and liberating the fatty acids of the oil.) If it be desired to weigh the fatty acids, proceed as follows : Weigh carefully about 5 grams of pure white beeswax, place it in the beaker upon the suface of the oil and water, and bring the contents nearly to boiling; the melted wax and fatty acids unite; allow to cool, remove the wax, wash with water, dry be- tween folds of filter-paper, and weigh. The increase in weight of the wax over its original weight gives the weight of the fatty acids of the animal or vegetable oil in the lubricating oil.^ Another method of determining the weight of the fatty acids after saponification is given on page 397. The weight obtained must be multiplied by the factor 0.97, since the fatty acids exist in the oil as anhydrides and not as hydrates, the latter being the form in which they are weighed. Instead of weighing the animal or vegetable oil, some chemists prefer to make use of the ether solution, determining the hydro- carbon oil directly ; in which case proceed as follows : After drawing off the soap solution from the separatory funnel the ether solution is run into a weighed flask (about 250 cc.) and the ether distilled off. The residue in the flask now consists of the mineral oil and some water. It is quite difficult to get rid of all this water. Direct heating is inadmissible, since the water spurts up throwing the oil out of the flask which is lost. This can be overcome by placing a glass tube through the stopper, in shape of the letter S. Any oil ejected against the tube or cork cannot escape, but returns to the base of the flask, while the heat is gradually increased in the flask ^ Determination of Soap in lyubricating Oils, /. Soc. Chem. Ind., 1896, p. 382. 394 e;nginee:ring che;mistry and the water vaporized and passed out through the tube; three or four weighings are generally required before a constant weight is obtained. The former process is preferable, since it is per- formed much more rapidly than the latter, and also the animal Fig. 79. Fig. 80. or vegetable oil is positively shown, and generally can be identi- fied ; also many lubricating oils contain as high as 20 per cent, of hydrocarbon oil, volatile at or below 212° F. It is, of course, in EJNGINEERING CHEMISTRY 395 the ether solution, and when the water is expelled from the oil, after the ether has been driven off, a large proportion of the volatile hydrocarbon is also evaporated. If now the animal or vegetable oil is not also determined, a serious mistake would be made ; vis., reporting 20 per cent, of animal oil when it was vola- tile mineral oil. Fig. 81. Fig. 82. The fatty acids in another sample of the oil are separated and subjected to qualitative test for identification of the oil from which they are derived. These tests comprise determination of melting-point, and congealing point, iodine absorption, and Maumene's test (rise of temperature upon addition of sulphuric acid). There are several methods of determining the melting-point of the fatty acids. Where a considerable amount of the fatty acids 396 ENGINEERING CHEMISTRY is available for experiment, the apparatus in Fig. 79 can be used. The glass cylinder is filled one-half with fatty acids, the cylinder closed with a rubber stopper, through which a thermometer is in- serted, the bulb of which is covered by the fatty acids. The apparatus is supported in a beaker containing water, as shown in Fig. 80. If the fatty acids are liquid at ordinary temperatures, the water in the beaker must be cooled with ice until the fatty acids are con- gealed. The ice is removed, and the water gradually warmed until the fatty acids become melted. At this point the tempera- ture is taken and recorded. Greater delicacy in the determination of the melting-point is obtained by using a small glass tube, sealed at one end. The liquid fatty acids are placed in this tube, then congealed, the tube tied to a thermometer. Fig. 81, and both inserted in a beaker of water, as shown in Fig. 82. Another method is to cover the thermometer bulb with a layer of the solid fatty acids, about 3 millimeters thick, and immersing it in water; gradually heat the water and notice the temperature at which the fatty acids leave the thermometer bulb and ascend through the water. Table of Mei,ting-Points and Congealing-Points of Fatty Acids Fatty acids Cotton-seed oil Olive oil . . • . Rape seed oil • Castor oil Sesame oil ... Cocoanut oil . . Lard Tallow Wool fat Palm oil Corn oil Melting-point °C. Congealing point °C. 330 30-5 26.0 21.0 20.0 12.0 13.0 30.0 26.0 32.0 24.5 24.0 44.0 39-0 45-0 42.0 42.0 40.0 48.0 43-0 20.0 14.0 The oils made use of in lubrication can be reported into two groups : saponifiable and unsaponifiable. To the former belong all the fatty oils ; to the latter the mineral and resin oils. The method of Lux is made use of to determine if any fatty oils are present in a mineral oil. ENGINKERING CHEMISTRY 397 If resin oil is suspected to have been added to the mineral, it can be identified by the Liebermann-Storch reaction, or the pro- cess of E. Valenta can be used. The Liebermann-Storch reaction for detection of resin oils : One to two cubic centimeters of the oil under examination are shaken with acetic anhydride at a gentle heat; after cooling, the acetic anhydride is drawn off by means of a pipette, and tested by adding i drop of concentrated sulphuric acid. If resin oil is pres- ent, a fine violet (fugitive) color is immediately produced. This test is thoroughly reliable for the detection of resin oil in mineral oil. These three tests will indicate, qualitatively, the presence of any fatty or resin oil in a mineral oil. It is rarely, in the better class of lubricating oils, that more than one oil is added to a min- eral oil, such, for instance, as lard oil, or tallow, in which case saponification easily separates the two oils, and identification of each by special tests can be made. Twenty grams of the oil are weighed in a No. 3 beaker, 100 cc. of an alcoholic solution of potash (80 grams potassium hydroxide to i liter alcohol of 98 per cent.) are added, and heat applied with stirrifl^ until the alcohol is all driven off; add 100 cc. water, heat with agitation, cool, add 50 cc. ether, transfer to separatory funnel, stopper, shake well and allow to stand 2 hours. Draw off the soap solution. 1. Soap Solution (Containing thefatty acids of the lard and cotton-seed oils). Heat 10 minutes nearly to boiling, cool, acid- ify with dilute sulphuric acid, allow to stand a few hours; collect the separated fatty acids ; determine their weight, then test as follows : First portion; Determine the "melting point." Second portion : Determine the iodine "ab- sorption" and their rates by formula : x={l-n) X = m — fi 2. Ether Solution remaining in the sepa- ratory funnel is transferred to a flask, the ether distilled and the mineral oils weighed. When, however, the oil added to the mineral oil itself contains an adulterant, such as lard oil to which cotton-seed oil has been added, then the fatty acids separated by saponification will re- quire a more extended examination to prove the presence of both lard oil and cotton-seed oil. The preceding skeleton scheme is given to show the applica- tion of the above upon a lubricating oil that qualitative analysis has shown to contain mineral oil, lard oil and cotton-seed oil. 398 e;ngine:e;ring chemistry There are several methods for the quantitative determination of the amounts of vegetable and animal oils when mixed with each other or when the mixture is incorporated with a mineral oil. The determination of the iodine absorption is the most deli- cate and correct provided no fish blubber or olive oils are present. If the fatty acids have been separated by saponification, from a mineral oil, this iodine value can also be determined. The method of Salkowski^ depends upon the fact that vege- table oils (except olive) contain phytosterol and that animal fats (butter excepted) are free from it, containing cholestrol, the latter not being present in vegetable oils. Fifty grams of the oils are saponified with alcoholic potash; the soap solution is diluted with a liter of water and 250 cc. of ether added. When the two layers have separated, the aqueous layer is run off and the ethereal liquid filtered, and evaporated to a small bulk. To insure complete absence of unsaponified fat, it is best to saponify again with alcoholic potash and to repeat the exhaustion with ether. The ethereal layer is then washed with water and the ether evaporated in a deep basin. The residue is next dissolved in hot alcohol, the solution boiled down to i or 2 cc. and the residue allowed to cool. If phytosterol or choles- terol be present, crystals will separate out. They are dried on unglazed porcelain and their melting-points determined. The saponification value of oils is often made use of for identi- fication; but as this value varies with the age of the oil, it is ex- tremely difficult to obtain concordant results, and as the majority of oils have a saponification value of 193, excepting rape seed oil and castor oil, which are lower, it cannot be relied upon. It, however, is of no value in determining the amount of liquid waxes in the presence of oils. Gumming Test. This test gives an indication of the amount of certain changes that may be expected in a mineral oil when in use. These resinified products, resulting from use, increase the friction of the revolving or rubbing surfaces. ' Benedikt's "Oils, Fats, and Waxes," p. 255. ENGINEERING CHEMISTRY 399 It is also a measure of the amount that an oil will carbonize in a gas or gasolene engine cylinder. It is applied after the manner of the elaiden test, by thoroughly mixing together in a cordial glass 5 grams of the oil with 11 cc. of nitrosulphuric acid and keeping the mixture cooled in a pan of water at 10° to 15° C. Brownish spots or, in case of bad oil, masses, form around the edges and become red in the course of two hours. As has been shown by long experience, the oil showing the least tar or gum is the best oil; it also absorbs the least oxygen. The nitrosulphuric acid is made by saturating sulphuric acid of 1.47 specific gravity cooled to 0° C, with nitric oxide (NO). (Amer. Soc. Testing Materials.) SUI.PHUR Test.^ Proceed as follows: A portion of a sample, 0.7 to o.i gram, is burned in a calorimetric bomb containing 10 cc. of water and oxygen under a pressure of 30 atmospheres. A lower pressure sometimes gives inaccurate results. If the sample contains more than 3 per cent, of sulphur the bomb is allowed to stand in its water bath for 15 minutes after ignition of the charge. In case the sulphur content is as high as 5 per cent., oxygen under pressure of 40 atmospheres is used. With these high pressures in a Berthelot bomb of 500 to 600 cc. capacity, repeated trials have failed to show even traces of carbon monoxide or sulphur dioxide. If a smaller bomb of about 175 cc. capacity, such as the Peters or Kroker, is used, incomplete combustion from a lack of oxygen may result if too large a sample is taken. After cooling, — 15 minutes is usually enough, — the bomb is opened and its contents are washed into a beaker. If the bomb has a lead washer, 5 cc. of a saturated solution of sodium car- bonate is added, the contents are heated to the boiling point, boiled for 10 minutes, and are then filtered. This operation is necessary to decompose any lead sulphate from the washer. The united washings are then filtered, acidified with hydrochloric acid boiled to expel all carbonic acid, and the sulphuric acid content is determined in the usual way with barium chloride. ^ Allen and Robertson, Technical Paper No. 36, Bureau of Mines, p. 10. 400 ENGINEERING CHEMISTRY Gravimetric determination is preferred to volumetric, because the nitrogen contained in the air originally 4n the bomb is oxid- ized in part to nitroacids, which cause a small error if the volu- metric determination alone is used. The sulphur content of any- combustible material, from light gasolenes weighed in a tared gelatin capsule to solid bitumens and cokes, can be readily de- termined by this method. This method of burning in a bomb is accurate, practicable, and rapid, and is recommended in preference to all of the other methods there described. The calorimetric determination, if de- sired, can be made at the same time. Test for Water.^ Dilute the oil with an equal volume of benzene and whirl it vigorously in a centrifuge until the separated layer of water does not appear to increase in volume. However, as water is some- what soluble in any diluent used and also in oils, a portion of the water content will fail to appear; consequently the method in which a diluent is used can not be considered accurate. It is advisable first to agitate the diluent vigorously with water and then to separate with the centrifuge in order to saturate it with water before using. Groschuff^ states that lOO grams of benzene will dissolve 0.03 gram of water at 3° C. and 0.337 gram of water at yy^ C, whereas petroleum products (density 0.792) will dissolve from 0.0012 gram at 2° C. to 0.097 gram at 94° C. Alternate Method. — The water content may be accurately and conveniently determined during the course of an ordinary dis- tillation in the following manner : Two hundred grams of the sample are weighed into a ^- liter distilling flask and the distillation carried out in the ordi- nary manner at the rate of i drop of distillate per second. The distillation can be performed most accurately in an electric still. At temperatures between 90° and 150° C. the water distills over and can be removed from the receivers by means of a micro- ^ Allen and Jacobs, Technical Paper No. 25, Bureau of Mines, p. 5. 2 E. Groschuff, "The Solubility of Water in Benzene, Petroleum, and Paraffine Oil," Chemical Abstracts, Vol. 5, p. 2550 (Aug. 10, 191 1). DNGINKERING CHEMISTRY 401 pipette and weighed. Usually a few drops of water adhere to the condenser and fail to run into the receivers; in this event a small pellet of absorbent cotton, moistened with water, squeezed as dry as possible, and weighed, is fastened to a wire and run up into the condenser to remove these last traces of water. The in- crease in weight of the cotton pellet, figured at water, is added to the weight of the water in the receivers. With an oil containing considerable water, it is advisable to cause a slow current of dry, inert gas, such as carbonic acid, to bubble through the oil in the distilling flask to carry off the vapors of oil and water as soon as formed. The gas current will reduce bumping and overheating of the oil to a minimum. The condenser must also be kept well cooled throughout the distillation. This method is accurate to less than 0.03 per cent. GasoIve:ne Test. Dissolve 10 cc. of the oil in 90 cc. of 86° to 88° gasolene (from t sk Fig. 83. Pennsylvania crude) in the graduated tube^ shown in Fig. 83. ^ The flat tube originally proposed by Conradson cannot be obtained on the market. 26 402 ENGINEERING CHEMISTRY Allow to stand i hour at 70° F. ; not more than 5 per cent, of flocculent or tarry matter should have settled out. If the test is first applied to the oil before making the flash test and again after this test, it shows the extent to which the oil is changed upon heating. Other things being equal, the oil which is changed the least is the best oil. MiCROscopiCAE Examination. Put a few drops of the well-mixed oil on a slide and note the nature of the suspended matter — whether carbonaceous specks, flakes of parafline, which disappear on warming, or foreign mat- ter. Polarized light is a great aid in detecting parafline crystals, showing them white on a black background. The polariscope is Fig. 84. — "Gray" carbon residue flask. excellent for this same purpose, showing them when it is im- possible to see them with direct light. Carbon Residue Test. Gray's Method. — To a tared i -ounce flask of the dimensions shown in Fig. 84 add 25 cc. of the oil to be tested and weigh. Wrap the neck of the flask with asbestos paper as far down as KNGINEE^RING CHEMISTRY 403 the side arm. Stopper tightly with a good cork. Connect to a small aerial condenser by plugging the space with asbestos or glass wool. Provide a shield which will protect the flame and the flask up to the side tube. Using the flame of a good Bunsen burner, heat the flask so that the first drop of distillate will come over in approximately 5 minutes. Continue the distilla- tion at such a rate that i drop per second will fall from the end of the condenser. As the end of the distillation approaches, increase the heat just enough so that no heavy vapors are allowed to condense and drop back into the flask, continue increasing the heat until the flask is enveloped in the flame, and hold the tem- perature 5 minutes. Allow the flask to cool, remove the asbes- tos covering and cork, and burn out completely the carbon and oil in the neck as far down as the side tube, and in the side tube. Heat the bottom of the flask until no more vapors are given off. Cool and weigh. FixDD Carbon in OiIv. Residues — Petroleum, Pitch, etc. — This is done in the same manner as the determination of fixed carbon in coal, as described on page 2. Estimation oi^ Paraffins in Mine;rai, O11.S. The following method is due to Holde (after Engler and Bohm) : Ten to 20 cc. of oils poor in paraffine (Russian distillates, etc., setting below — 5° C), or 5 grams of such as are rich in that constituent (American, Scotch, or Galician oils setting at or above 0° C), are treated, at the ordinary temperature, with a mixture of 98.5 per cent, alcohol and anhydrous ether (1:1) until a clear solution is obtained. The liquid is cooled in a freez- ing mixture of ice and salt to about — 20° C, when more alcohol ether is gradually added, with thorough agitation, until no oil drops, but only solid paraffine flakes or crystals remain in sus- pension, and then, while still cooled to at least — 19° to — 21° C, the liquid is poured on to a chilled 9-centimeter filter paper, previously moistened with alcohol ether mixture, which is con- 404 e:ngineering chemistry tained in the apparatus shown in Fig. 72. The precipitate is washed with cold { — 19° to — 21° C.) alcohol ether (1:1); or for soft paraffine (2:1) at a temperature as much below — 15° C. as possible. In the case of soft paraffine, the tempera- ture should average — 18° to — 19° at the highest. In washing the precipitate it is repeatedly stirred up, and as soon as 5-10 cc. of the filtrate leaves on evaporation only a trace of fatty or paraf- fine-like residue, solid and not oily at the ordinary temperature, the washing is discontinued. If any doubt exists as to the paraffine being thoroughly freed from oil, or if the washing takes too long, the filter should be removed to another funnel, and the contents dissolved into a small flask with the least possible quantity of benzine. After evaporation of the benzine, the paraffine is redissolved in 4 to 5 cc. of warm ether, which is then mixed with twice its volume of absolute alcohol, vigorously stirred, and cooled to — 18° to — 20° to reprecipitate the paraffine, which is again filtered and washed, as already described, until free from oil. This reprecipitation is necessary for oils containing much soft paraffine, otherwise so much liquid is used in washing the precipitate that an appre- ciable quantity of paraffine is dissolved. The purified paraffine is finally dissolved into a tared flask with hot benzine or ether, which is distilled off, and the residue is heated on the steam- bath until the smell of benzine or ether has disappeared. The flask is then heated inside the water-oven for Yx hour and weighed when cold. Prolonged heating causes los^ of paraffine. The whole operation occupies from i to 2 hours. Duplicate re- sults with the same sample agree within 0.23 per cent, for hard paraffine, and 0.33 per cent, for soft paraffine. Two samples of Russian machine oil yielded 0.34 per cent, and 0.36 per cent, of paraffine respectively. An American spindle oil, fluid but thick at 2° C, and which set at 0° C, was found to contain 4. 11 per cent, of paraffine. Soap Test. The test depends upon the fact that the metaphosphates of the earthy and alkali metals and aluminum are insoluble in ENGINEERING CHEMISTRY 405 absolute alcohol. Five to lo cc. of the oil are dissolved in 5 cc. of 86° gasoline or ether, and 15 drops of a saturated solution of "stick phosphoric acid" in absolute alcohol are added, shaken and allowed to stand : the formation of a flocculent precipitate indicates the presence of soap. For the accurate determination of these soaps a known quantity of the oil must be ignited and the residue quantitatively examined. Saponification Vai,ue. This is expressed by the number of milligrams of potassium hydrate necessary to saponify i gram of the oil. From 2.5 to 10 grams of the oil, according as 65 to 20 per cent, of sa- ponifiable matter are supposed to be present, are boiled with 25 cc. N/2 alcoholic potash in a 200-cc. Jena Erlenmeyer flask. A reflux condenser is used and the boiling may require from 5 to 8 hours. The excess of alkali is titrated with N/2 HCl, using phenolphthalein. The strength of the N/2 KOH is determined by boiling 25 cc. in similar flasks alongside of those in which the oil is treated and for the same length of time. Alcohol purified with silver oxide according to Dunlap's method^ should be used as well as KOH, purified by alcohol. For heavy oils, dissolve them in 50 cc. of C. P. benzol before adding potash. Determination of Tarry Matters in PetroIvEum Products. At present the French volumetric method is exclusively used, and is very simple in its principle and practical application. According to this method the quantity of tarry matter in any petroleum product is judged by the increase in the volume of the sulphuric acid or the decrease in the volume of the tested product, which is, after being diluted with benzene, submitted to the action of sulphuric acid, which carbonizes and dissolves the tarry substances. This method is, as already mentioned, very simple and easy in practice, but is connected at the same time with a serious source of errors. Besides the principal action of carbonizing and extracting the tarry substances, sulphuric ^Journal Amcr. Chem. Soc, Vol. 28, p. 397. 4o6 ENGINEERING CHEMISTRY acid also extracts unsaturated hydrocarbons, polymerises, others, etc. Thus, for instance, a simple experience will show that even perfectly well refined oil treated with sulphuric acid de- creases in volume up to 8 per cent. But the incorrectness due to this may be obviated. Thus the quantity of unsaturated hy- drocarbons extracted by the sulphuric acid can be ascertained by determining the iodine value, and this can then be calculated to an equivalent value in sulphuric acid, according to the very simple formula, V = -^;t, where V is the volume of sulphuric acid, T the iodine value, and d the specific gravity of the sul- phuric acid. In the works, where a great number of tests are continually to be made, the operation becomes very simple, as d — the specific gravity of the sulphuric acid in use — is practically a constant, and so is the iodine value T for a whole series of products. Greases. Horace W. Gillet in the Journal of Industrial and Engineering Chemistry, June 1909, states as follows : — "Commercial greases may be divided into the following classes : "A. The tallow type: these greases are made of tallow and more or less of an alkali soap, commonly the sodium or potassium soaps of palm oil, mixed with a smaller amount of mineral oil. These were the principal types of lubricating grease 10 or 20 years ago, but to-day are less used than the greases of type B. "B. The soap-thickened mineral oil type: these are the most common journal greases of to-day, and are composed of mineral oil of various grades made solid by the addition of calcium or sodium soaps. Calcium soap is more used than sodium soap. "C. Types of A or B with the addition of a mineral lubricant — usually graphite, mica or talc. "D. The rosin-oil type: these consist of rosin oil thickened by lime, or less commonly, litharge, to which is added more or less mineral oil, either parafiine or asphalt oils being used. Engine;e:ring chemistry 407 s'i .2 w go p. . t^ 10 rO t^ ^vO 000000 ON t^OO vO ONVO CO ►-• 00000000 o' o* o" d d d d d O J-i . -• O «• ** 0000000 toooOOOOr^O vOvdot^QQ N O' . . . N 'O O il i! i d d M 0000 =1 C Q o o o • .•;: S ^ ^ P K 5 ^--^ 5 ^ S'^ ■*-* n: ::2 ea bCc^H Oco ='^urM', and the coefficient by /. /. = tan (Kent). Of the various machines used for this purpose nearly all are deficient in conducting tests under extreme pressure. However, as all the tests are relative, an idea of the value of a lubricant can be formed by a series of comparative tests upon the same instrument. An instrument for determining coefficient of friction is the Riehle, in use in many railroad laboratories in the United States, for testing lubricants. The capacity is 20,000 pounds; it deter- mines the coefficient of friction, the pressure per square inch of journal and records the temperature. 4IO ENGINEERING CHEMISTRY Method of Testing Oils on Riehle U. S. Standard Oil Testing Machine. The U. S. Standard "test bearing" of 9 square inches of pro- jected area is used. The quantity of oil in making tests (also the time of test) varies according to the will of the experimenter. The oil is applied from a sight-feed oil cup, which can be regu- lated; it is dropped on the journal in front of the "test bearing" and distributes itself along the edge of it and is carried under- neath the journal. Fig. 85. — Machine for Determining Coefficient of Friction of Oils. When a pad is used for making oil tests, it is saturated with oil and placed in the drawer underneath the journal. The oil passes from the pad to the journal and is carried to the under- side of the "test bearing." The pad (saturated with oil) must be weighed before and after the test. Whether the oil is tested with or without the pad, please note the following instructions : Observe the pressure in pounds per square inch; friction in pounds per square inch; the temperature above the temperature of the room ; the amount of oil used ; the revolutions per minute, and total revolutions of machine. e;ngine:e;ring chemistry 411 When it is desired to keep the bearing at a uniform tempera- ture water is circulated in the journal by means of the cooling apparatus provided. This is a new feature in oil testing machines, and is very much appreciated when it is desired to make uniform temperature tests. By comparison of results obtained from tests made according to the above instructions the qualities of different oils can be determined. De:scription and Operation. This machine is used to determine the lubricating properties of various oils. The load on the bearing is applied by means of a turnbuckle connection between the beam and lower lever, and is weighed on beam by large poise. The friction in pounds on the periphery of the journal is indicated by a poise on the upper or friction beam, reading by increments of i pound. The journal of the machine is mounted on four large rollers, which reduce the friction and prevent its heating, which would affect the re- sults of temperature tests. Ball thrust collar bearmgs prevent side motion of the journal, and take any thrust in this direction which would cause friction. The bearing to be tested fits in a cap to which the yoke frame is attached; this yoke frame is fitted with two knife edges equi- distant from the center of the shaft; two clevises join these knife edges with similar knife edges in the equidistant lever below, from which connection is made to the intermediate lever and the load beam. The yoke frame is thus perfectly free to rotate about the journal, and any tendency to do so will show on the friction beam. The instrument in general use, for the friction tests of lubri- cants, in Germany, is the "Mertens-Machine," (Fig. 86.) In principle it is a modification of the Thurston. It is fully described by D. Holde in his work — "Untersuchung der Mineralole und Fette," Berlin, 1909. 412 ENGINKEIRING CHEMISTRY An oil tested upon the tester may show a fine lubricant, while put under practical working upon a freight car (for instance) Fig. 86. would prove vastly inferior. This very often happens, and it has led many engineers to test each oil by a long run, with the par- ticular kind of machinery upon which it is to be used. ENGINEERING CHEMISTRY 413 Specifications for Engine Oil for the Department of Docks and Ferries, City of New York. Quality of Engine Oil. — The engine oil to be furnished under this contract shall be one of the following brand or brands equal thereto and approved by the Engineers : "Vacuum No. i Marine Engine Oil." "Kuhne-Libby Extra Marine Castor Oil." "Leonard & Ellis Valvoline Engine Oil." "New York Lubricating Oil Co.'s Diamond B Marengene." "Fisk Bros.' Luberine Engine Oil," or equal. Each can or package of oil delivered under these specifications shall be plainly marked with the marker's name and the brand. (a) Oil must be of the best quality and pass satisfactorily. (h) Specific Gravity. — Must not be less than 0.9000 at a temperature of 60° F. (c) Flashing Point. — Must not be below 420° F. (d) Fire Test. — Must not be less than 475° F. (e) Freedom from Gumming.^ — Just sufficient oil to cover the bottom will be placed in a shallow dish. This will be heated to about 250° F., then cooled slowly. When cold there must be no gummy residue in the oil or in the vessel. Oil must also pass satisfactorily such other tests for gumming as may be ordered by the engineer. (/) A common oil cup holding 2 ounces will be filled with the oil ; two threads of worsted will be used as a wick, and all the oil in the cup must feed through it ; the wick not to be touched during the trial. This test to be made at a temperature between 70° and 90° F. {g) Cold Test. — The oil must not solidif}^ at a temperature of 32° F. {h) Freedom from Acid. — A small quantity of oil rubbed on poHshed brass or copper m.ust not turn the surface of the metal green if allowed to stand for 24 hours. (i) Viscosity. — Viscosity at 70° F. must be between 800 and 850. Specifications of Cylinder Oil. (a) Must be a pure mineral, hydrocarbon oil, with a flash-point of at least 550° F. ; burning-point to be above 600° F. ; to be free from tarry or suspended matter, acid or alkali, and from mixture or adulteration with animal, vegetable or fish oils, grease, lard or tallow. Specific gravity to be not below 0.900 nor above 0.906 at a temperature of 60° F. {h) A flash-point below 550° F., or the presence of any of the above named adulterations or mixtures, or a gravity different from that specified, will be sufficient to cause the rejection of the oil. ^ This gumming is often due to the addition of rope oil. 414 ^ngine:e:ring chemistry (c) Flashing Point. — Heat a small quantity of the oil in an open vessel, not less than 12° per minute, and apply the test flame every 10°, beginning at 250° F. (d) Precipitation Test for Tarry and Suspended Matter. — Mix 5 cc. of oil with 95 cc. of 88° gasoline, and if there is any precipitation in ID minutes the oil must be rejected. This test is easiest made by putting 5 cc. of oil in a 100 cc. graduate, then filling to the mark with gasoline and thoroughly shaking. (e) Volatility. — Keep heated to 400° F, in an open vessel; it must not lose more than 5 per cent, of its weight in 2 hours. (/) To Test for Acid or Alkali. — It will be sufficient to wash a small quantity of the oil with distilled water, then drain off the water and test it with Utmus paper. Specifications for Lubricating Oil for Dynamo Engines and Other High-Speed Engines Using Forced Lubrication, Issued by the Navy Department, November 17, 1906. (Superseding Specifications 24-O-11 Issued May 25, 1906.) 1. Must be a pure mineral hydrocarbon oil, free from acidity, adul- terations, and impurities, with a flash-point (open cup) of at least 350° F. ; to be free from saponifiable substances of any character whatsoever; specific gravity to be between 0.865 and 0.875 at 60° F. ; to be purchased and inspected by weight, the number of pounds per gallon to be deter- mined by the specific gravity of the oil at 60° F. multiplied by 8.33 pounds, the weight of a gallon (231 cubic inches) of distilled water at the same temperature. 2. Viscosity (by Engler viscosimeter) as compared with distilled water (49) at 90° F. : At 90° F 300 to 320 At 150° F 105 to 115 At 225° F 65 to 75 3. Cold Test. — The oil must flow at a temperature of 32° F, 4. Freedom from Gumming. — Using a single- wick J/^-pint brass oil cup maintained at about 140° F., practically equal quantities of oil must feed through the wick in equal intervals of time for three intervals of 8 hours each. At the end of test the wick must be clean and sides of oil cup bright and clean. Inspection and Delivery, 5. Before acceptance the oil will be inspected. Samples of each lot will be taken at random, the samples well mixed together in a clean vessel, and the sample for test taken from this mixture. Should the mixture be found to contain any impurities or adulterations, the whole delivery of oil it represents will be rejected, and is to be removed by the contractor at his own expense. ENGINEERING CHEMISTRY 415 Specifications for Neatsfoot Oil Issued by the Navy Department March 24, 1908. (Superseding Specifications 24-O-12 Issued June 22, 1906.) 1. Neatsfoot oil must be free from admixtures of other oils, and must not contain more acidity than the equivalent of 2j^ per cent, of oleic acid. 2. It must have a cold test below 10° F., as determined in the follow^- ing manner: A couple of ounces of the oil will be put in a 4-ounce sample bottle and a thermometer placed in it. The oil will then be frozen, using a freezing mixture of ice and salt if necessary. When the oil has become hard the bottle will be removed from the freezing mixture and the oil allowed to soften, being stirred and thoroughly mixed at the same time by means of the thermometer until the mass will run from one end of the bottle to the other. The reading of the thermometer at this moment will be taken as the cold test of the oil. Graphite as a Lubricant. Graphite is used either alone or mixed with various oils and greases as a lubricant. Prof. W. F. Gross, of Purdue University, has made a series of extensive experiments upon- graphite as a lubricant, and in his report, states : — ''From the earlier and rather limited uses of graphite in lubri- cation, the field has gradually widened to include its use with light oils, with water, and in some cases, unmixed with other materials. It is no longer regarded merely as a material for an emergency, but now has a place in the ordinary and usual routine of the day." "Graphite does not behave like oil, but associates itself with one or other of the rubbing surfaces. It enters every crack and pit in the surfaces and fills them, and if they are ill-shaped or ir- regularly worn, the graphite fills in and overlays until a new surface of more regular outline is produced. When applied to a well-fitted journal the rubbing surfaces are coated with a layer so thin as to appear hardly more than a slight discoloration. If, on the other hand, the parts are poorly fitted, a veneering of graphite of varying thickness, which in the case of a certain experiment was found as great as Vie inch, will result. The character of this veneering is always the same, dense in structure, capable of resisting enormous pressure, continuous in service without apparent pore or crack, and presenting a superficial finish 4l6 DNGINEIERING CHEMISTRY that is wonderfully smooth and delicate to the touch." From a long and very severe series of experiments, it seems safe to con- clude that no journal is likely to be damaged from overheating or kindred causes in the presence of a supply of graphite. It is a fact worthy of all emphasis that, with solid brasses, the pres- ence of flake graphite between the rubbing surfaces makes it impossible to score or otherwise damage the surfaces of the bear- ings even though the temperature be allowed to run high. **It goes without saying that the use of graphite, in service which might be rendered without any lubrication whatever, is justified by the more perfect action and the greater durability of the parts affected. Graphite alone, in service for which it is adapted, is to be preferred to oils or greases, because of its superior cleanliness and because of the persistency with which it remains in place after being once applied. "The experiments with flake graphite as a lubricant justify the following important conclusions : (a) The addition of graphite to oil results in a lower frictional resistance of the journal than would be obtained by the use of oil alone. (b) When graphite is used with oil, the amount of oil required for a given service is reduced. (c) By the use of graphite a light or an inferior quality of oil may be employed for a given service. (d) By the use of graphite, water under favorable conditions may serve as a sufficient lubricant. (e) A small amount of graphite only is required. (/) The supply of too much graphite unduly thickens the oil and correspondingly increases its internal friction due to viscosity. (g) The benefits derived from the use of graphite persist long after its application has ceased. The supply, however, should be constant, though small, for best results." Even in the cylinders of air compressors there is sufiicient moisture to constitute a lubricating film without oil when the surfaces of the cylinders and pistons have been perfected by the presence of graphite. ENGINEERING CHEMISTRY 417 Lieut. H. C. Dinger, U. S. N. states : — Flake graphite has the peculiar properties of not being affected, either chemically or physically, by any temperature encountered in a cylinder. It is not easily carried away from the wearing surfaces, can stand any pressure, and requires only an infinitestimal clearance space be- tween surfaces by filling up all the minute cavities and irregulari- ties in the surfaces, giving in a short time, a beautiful, hard- polished surface which requires relatively little lubricant. Specifications for Lubricating Graphite Issued by the Navy Department January 17, 1907. 1. It ma}^ be of the flake or amorphous variety. Amorphous graphite must be ground fine enough to pass through a No. 20 bolting cloth. 2. Samples taken from any lot must show upon analysis at least* 85 per cent, of pure graphite. It must be free from grit, dirt, or any other deleterious substance. 3. It must be put up in air-tight rectangular tin cans with screwed tops, each containing i or 5 pounds, as may be required. 4. Each can must be marked with the name of the material, the trade- mark, if any, and the name of the manufacturer. The Calorific Power of Petroleum Oils and the Relation of Density to Calorific Power. The purpose of this paper is to put on record the calorific power of a considerable number of representative American pe- troleum oils and to point out an approximate relationship between the density and the calorific power of such oils.^ While among the homologues of a given series of hydro- carbons, decreasing proportions of hydrogen might be assumed to involve an increase in density and decrease in heat of com- bustion or calorific power, it would not necessarily follow that such a relation would obtain for the mixtures of hydrocarbons which constitute the crude petroleums or thieir commercial pro- ducts. Nor have we been able to find in the literature suffi- cient comparable data to give even an approximate expression of the quantitative relation to be expected between density and the calorific power of petroleum products. ^A. C. Shermann and A. H. Kropff, Jour. Amer. Chem. Soc, Oct., 1908, pp. 1626-31. 27 4i8 engine:e:ring chemistry Densities and Heats of Combustion Observed and Calculated. No. Specific gravity l5°/'5°. Baum6 degrees Calories per kilogram B. t. u. per pound B. t. u. cal- culated Percent- age error Description I 0.7100 67.2 11,733 21,120 20,938 —0.91 Gasolene 2 0.7175 65.1 11,327 20,389 20,854 + 2.33 " 3 0.7209 64.4 11,404 20,527 20,726 + 0.99 " 4 0.7709 51.6 11,132 20,038 20,314 4-1.28 * 5 0.7830 48.8 11,121 20,018 20,2C6 + 0.92 Kerosene 6 0.7850 48.35 11,119 20,014 20,194 40.89 California refined 7 0.7945 46.2 11,128 20,030 50,098 +0.33 West Va. 8 0.7950 46.1 11,186 20,135 20,094 —0.20 Kerosen 9 0.7964 45.8 11,242 20,236 20,082 + 0.70 * lO 0.8048 440 11,149 20,068 20,010 —0.29 Ohio crude II 0.8059 43-7 11,143 10,057 19,998 —0.29 Penna. crude 12 0.8080 43.2 1 1 ,001 19,802 19,979 + 0.88 California refined 13 0.8103 42.8 11,090 19,963 19.962 ±0.00 Kansas refined • M 0.8237 40.0 10,981 19,766 19,850 +0.42 West Va. crude 15 0.8248 39-7 11,015 19,827 19,838 +0.05 California refined i6 0.8261 39-5 11,123 20,021 19,830 —0.95 West Va. crude 17 0.8321 38.2 10,972 19-757 19,778 + O.II * i8 0.8324 38.2 10,990 19,782 19,778 —0.02 Penna. crude 19 0.8418 36.3 10,950 19.710 19,702 —0.04 Ohio crude 20 0.8421 36.25 10,997 19.795 19,698 —0.48 Indian Territory 21 0.8536 36.0 11,069 10,924 18,690 — 1. 17 -x- 22 0.8466 35-4 10,936 19.685 19,666 — 0.09 Indian Territory 23 0.8500 35.7 10,953 19,715 19,638 -0.38 California refined 24 0.8510 34.5 10,958 19.724 19,630 —0.47 Kansas crude 2S 0.8514 34.45 10,945 19,701 19,630 -0.35 * 26 0.8534 34.05 10,991 19.784 19,610 —0.86 ■x- 27 0.8580 33.2 10,772 19.389 19,578 +0.95 Kansas crude 28 0.8597 32.8 10,766 19.379 17,562 + 0.95 Illinois crude 29 0.8616 32.5 10,967 19.741 19.550 —0.95 * 30 0. 8640 32.05 10,867 19,555 19.530 —0.12 California refined 31 0.8648 31-9 10,920 19,656 19.526 —0.65 Penna. fuel oil 32 0.8660 31-65 10,864 19.555 I9,5>6 —0.19 Fuel oil 33 0.8670 31-5 10,850 19,530 19.510 — O.IO Penna. fuel oil 34 0.8690 31. r 10,852 19.534 19.494 — 0.20 Indian Territory 35 0.8708 30.8 10,919 19.654 19,482 —0.86 * 36 0.8712 30.7 10,879 19.614 19,478 —0.86 •X- 37 0.8945 30.1 10,752 19.354 19,454 +0.50 Kansas crude 38 08773 29.6 10,794 19,429 19.434 +0.03 Penna. fuel oil ■ 39 0.8800 29.0 10,804 19,447 19,410 —0.18 Kansas crude • 40 0.8807 29.0 10,797 19.435 10,410 —0.47 * 41 0.8810 28.9 10,797 19.435 19,406 —0.15 42 0.8820 88.75 10,913 19,643 19,400 — 1.22 ■X- 43 0.8828 28.7 10,694 19.249 19.306 + 0.73 Kansas crude 44 0.8833 28.5 10,819 19,474 19.390 —0.42 ¥; 45 0.8860 28.0 10,808 19,454 19.370 —0.42 Indian Territory 46 0.8862 28.0 10,762 19,372 10,370 — 01 * 47 0.8900 273 10,788 19,418 19,342 —0.39 Indian Territory 48 0.8914 27.1 10,690 19,242 19,332 + 0.45 Texas crude 49 0.8970 26.1 10,753 19,355 19.294 -0.31 50 0.9007 25.4 10,755 19,359 19,267 —0.47 ^ Enginee:ring chemistry 419 Densities and Heats of Combustion Observed and Calculated. {Co7i tinned). No. Specific gravity Baum^ degrees Calories per kilogram B. t. u. per pound B. t. .u cal- culated Percent- age error Description 51 0.9050 24.7 10,682 19,228 19.238 + 0.05 52 0.9061 24.45 10,755 19.352 19,228 —0.63 * 53 0.9076 24.4 10,605 19,089 19,226 + 0.69 Kansas crude 54 0.9087 24.1 10,712 19,282 19.213 —0.35 55 O.9II4 23.6 10,724 19.303 19.194 —0.55 Kansas crude 56 0.9137 23.2 10,571 19,028 19,178 +0.76 Texas crude 57 0.9153 22.95 10,692 19,246 19,168 —0.39 Texas crude 58 0-9155 22.9 16,560 19,008 19,166 -fo.8o Texas crude 59 0.9158 22.9 10,318 18,572 19,166 +2.58 California crude 60 0.9170 22.7 10,613 19. "03 i9.'57 -f 0.28 Fuel oil 61 0.9179 22.5 10,433 18,779 19.150 + 1.94 California crude 62 0.9182 22.5 10,547 18,985 17,149 +0.83 California crude 63 0.9336 20.0 10,600 19,080 19,048 —0.16 Texas crude 64 0.9644 15.2 10,327 18,586 18,858 + 1.42 California crude * Obtained by fractional distillation of comitiercial fuel or gas oils; Nos. 4, 25, 36, 44 and 52 were the successive fifths from one sample; Nos. 9, 21, 26, 29 and 42 from a second; Nos. 17, 35, 40, 46 and 50 from a third. It will be seen throughout the range of oils included in the table there is a general tendency toward a fairly regular decrease in calorific power as the specific gravity increases and the Baume numbers decrease. In the cases in which an approximate estimate of the calorific power is most likely to be useful, the expression of density in terms of the Baume scale and of calorific power in British ther- mal units per pound will probably be most common.^ By group- ing the samples falling within certain limits of Baume density and potting the average figures, it was found that the approxi- mate average relation between Baume density and calorific power in B. t. u. may be expressed as follows : B. t. u. =: 18,650 + 40 (Baume — lo). This formula was then applied to the data of the individual samples. In the columns headed "B. t. u. calculated" and ''per- centage error" are given for each sample the calculated British thermal units and the percentage dift'erence between the calculated and the determined values. * It should perhaps be noted that the heavier oils with lower calorific power per gram or per pound would show higher calorific power per ^ai/on than the light oils. 420 ENGINEERING CHEMISTRY It will be seen that the difference between the calorific power as determined in the bomb calorimeter and as calculated from the formula here proposed is usually small. In only Yg of the cases is the difference greater than i per cent., in only ^/g^ is it greater than 2 per cent. ; in no case is it as great as 3 per cent. The samples examined were all believed to be of fair aver- age commercial purity; the discrepancies might readily be larger in oils grossly contaminated with water or suspended matter, but the majority of such cases could probably be recognized by superficial examination. In view of the number of samples examined and the fact that about half of them were selected as representative of the pro- ducts of the principal oil-fields of the United States, while the remainder were taken at random from commercial sources, it would seem safe to infer for commercially pure samples of or- dinary American petroleum oils, varying from heavy crudes to gasoline, the calorific power may be predicted from the density with about as close an approximation to accuracy as is usually obtained in calculating fuel values from chemical analysis. If it be desired to estimate the calorific powder in terms of calories per gram, or to base the estimate upon specific gravity, or both, it need only be remembered that calories per kilogram 140 X 1.8 = B. t. u. per pound, and that specific gravity = , p, I30~r B according to the standard used in obtaining the Baume figures here given, or the following estimate may be used, which being obviously only an approximate indication, is perhaps less likely to be misleading than is a formula : A sp. gr. 0.70-0.75 indicates about 11,700-11,350 cal. per kg. A sp. gr. 0.75-0.80 indicates about 11,350-11,100 cal. per kg. A sp. gr. 0.80-0.85 indicates about 11,100-10,875 cal. per kg. A sp. gr. 0.85-0.90 indicates about 10,875-10,675 cal. per kg. A sp. gr. 0.90-0.95 indicates about 10,675-10,500 cal. per kg. Of the 63 samples here examined which fall with these limits of specific gravity, only 2 fall outside of the indicated range of calorific power by as much as 100 calories, and only 7 by as much as 50 calories. ENGINE^ERING CHE:mISTRY 421 Summary. Sixty-four samples of petroleum oils, ranging from heavy crude oil to gasoline, and representing the products of the prin- cipal oil fields of the U. S. were examined for calorific power by combination in oxygen in the Atwater-Mohler bomb calorimeter with results ranging from 10,318 to 11,733 calories per kilogram, or 18,572 to 21,120 B. t. u. per pound. In general the decrease in calorific power with increase in specific gravity was fairly regu- lar, so that the relation between the two may be expressed ap- proximately by means of a simple formula. When the calorific powers calculated from the densities by means of this formula were compared with those actually determined it was found that in Vo of the cases the difference was greater, and in Yg it was less than i per cent. ; in only ^/g^ was it greater than 2 per cent. ; in no case was it as great as 3 per cent. While it is obviously improbable that an exact quantitative relation should exist, it is believed that from the data here given the calorific power of commercially pure petroleum oils may be predicted from the density with a sufficient approach to accuracy for many practical purposes. Remarks on Lubricants and Lubrication. Lubricating oils are obtained from the crude petroleum by one of two general methods, i. e., destructive distillation and fractional distillation. Oils obtained by destructive distillation are vaporized by means of fire only. After the lighter hydrocarbons (naphtha, burning oils, gas oils, etc.) have been driven off, the heavier vapors con- taining hydrocarbons used for the manufacture of lubricating oils are carried over from the still to the condenser. In so doing the level of the crude lowers in the still and the heavier vapors come in contact with the comparatively cool sides of the still, fall back across the heated metal and become charred. Some of the lubricating qualities are destroyed in this operation. In this process petroleum coke is the final product. ^ This article was contributed by Lewis F. Lyne, Jr., General Manager of the Oil Specialties & Supply Company, 39 Cortlandt St., New York City, N. Y., who is con- sidered an authority on lubrication. 422 ENGINEERING CHEMISTRY In the process of fractional distillation the lighter hydrocarbons (naphtha, burning oils, gas oils, etc.) are driven off by means of fire only. At this point steam is introduced. The steam acts as a carrier for the heavier vapors. In the condensers, the steam, having a lower condensation temperature than the distillate, allows the heavier vapors to condense under a partial vacuum. In so doing the fractions are brought over with less charring and nearer their original state. The final product in this process is a heavy cylinder stock. Mineral lubricating oils may be divided into three general classes, i. e., paraffine oils, spindle oils and cylinder stocks. The paraffine oils are manufactured by agitating the distillate with sulphuric acid, by means of compressed air, washing with water and neutralizing with caustic soda. Spindle oils are refined by filtering the distillate under reduced pressure through Fuller's earth. Cylinder stocks are obtained, as stated heretofore as the final product of fractional distillation. They may be either filtered or unfiltered. Paraffine oil may easily be distinguished from spindle oil by means of the heat test. Place a small amount of the oil in a test tube and heat over a spirit lamp or a Bunsen burner. Note the time in which the oil discolors or turns to a darker color. The paraffine oil will discolor very quickly whereas a spindle oil will retain its color for a greater length of time as compared with the paraffine oil. Taking lubricating oils as a whole they may be sub-divided as follows: — Cylinder oils, machine or engine oils, compressor oils, dynamo oils, internal combustion engine oils, ice machine oils, turbine oils, hydraulic press oils, roll oils, cutting oils, and greases. Although not used as lubricants the following are used very ex- tensively, tempering oils and transformer oils. The conditions usually met in the lubrication of machinery are as follows : Pressure; heavy, medium or light. Speed ; high, medium or low. Temperature; high medium or low. ENGINEERING CHEMISTRY 423 The viscosity is a most important factor to be considered when selecting an oil for a specific purpose. Although an oil may be of the highest quality it may cause considerable trouble in the operation of the machine, due to the fact that the viscosity may be either too high or too low (see dynamo oil). Then again the viscosity of oils should be considered at various temperatures. The temperature at the point of application should be determined in order to secure an oil viscous enough at that temperature to insure proper lubrication. Where it is possible thermometers should be placed on all main bearings, as they are the most efficient safeguard against overheating. For instance, after a plant has had a general over- hauling, the bearings may have been tightened too much. In such a case the error may not be noticed until too late, resulting in a general shut-down. Whereas if thermometers had been placed a rise of say 20° or 30° would give warning and a heavy cylinder oil could be used for a time until normal temperature was resumed, or the bearing readjusted. Specifications may call for an oil of a certain viscosity at 70° or 100° F., but the temperature of the bearing is 150° F. In this case two oils may be submitted which have the same viscosity at 70° or 100° F. and have the same appearance, but one may be of a paraffine and the other asphaltum base. The viscosity of the paraffine base oil will be lowered to some extent upon the rise of temperature whereas in the asphaltum base oil the viscosity will be lowered to a very marked degree, thereby not retaining vis- cosity enough to insure proper lubrication. Asphaltum base oils may be generally recognized by the low gravity and cold test and high viscosity, and the rapid lowering of the viscosity upon the rise of temperature as compared with a paraffine base oil. Cyunder Oil.. Cylinder oils may be filtered or unfiltered. The former is gen- erally olive green while the latter is dark brown. Although a cyl- inder oil may be filtered it is not necessarily a superior lubricant to the unfiltered stock. When filtered the oil becomes lighter and some of the lubricating hydrocarbons are removed. In an unfil- 424 ENGINEERING CHEMISTRY tered cylinder oil all of the lubricating ingredients remain. How- ever, the unfiltered stock must not contain ^ny impurities such as tar, moisture, or dirt, which may be found in the poorer grades of unfiltered cylinder oils. Tarry matter may be found by dissolving 5 cc. of the oil in 95 cc. of a high gravity naphtha (88° B.). If a precipitate is noticeable after standing 24 hours it indicates the pres-ence of tarry matter. In some extreme cases it may be detected by its characteristic odor. If moisture is present in quantity it may be detected by dis- solving the oil in high gravity naphtha (88° B.) until the mixture is just transparent. Shake well. Globules of moisture if present will fall to the bottom of the container, while the air bubbles will rise to the top. The presence of glue due to improper barreling may be found by the irregular operation of the hydrostatic sight feed cup. The hole through which the oil is forced by the condensed steam be- comes clogged. Then again the valves groan and a surplus quan- tity of oil is necessary to overcome the difficulty. Dirt of any nature, whether filings from the steam pipes, scale from the boiler carried over by the steam, or solid foreign sub- stances of any description may be found by filtering the oil through a piece of fine cloth or filter paper. If too viscous dis- solve a quantity in some high gravity naphtha (88° B.) which will permit the filtration of the oil. The dirt will remain on the cloth or filter paper. The gravity of a cylinder oil gives no indication of the lubri- cating qualities but merely aids in determining the base (paraffine or asphaltum) of the crude from which it was made. The flash point too is of minor importance since it is deter- mined at atmospheric pressure whereas the pressure in the cyl- inder is much higher, which results in very different conditions. However, the flash point of the oil should be sufficiently high not to disintegrate at the temperature of the steam at the pressure in qviestion. The cold test is important only as regards the storage. If the EINGINEERING CHE:mISTRY 425 store room is subjected to low temperature a sufficient amount should be kept accessible for the filling of the lubricator upon short notice. In a number of instances a receptacle containing a sufficient amount of surplus oil is allowed to stand on or near the cylinder of the engine or on hot steam pipes in the engine room. Unlike the conditions in bearing lubrication, the viscosity and lubricating qualities are difficult to determine in cylinder lubri- cation. A straight cylinder oil should be used in the lubrication of steam cylinders only where superheated steam is used. Here the cylinder walls are dry and the oil adheres to them. In the case of saturated steam the cylinder walls are moist, and as a straight mineral oil will not mix with water the oil is washed off. Therefore a compound of animal and mineral oil will give the most efficient lubrication, as animal oil mixes or emulsifies with water, insuring adhesion of the compound to the cylinder walls. The moisture in the steam should govern the percentage of animal oil to be compounded, which should be kept as low as possible. Upon the application of heat the animal oil liberates fatty acids commonly known as "free fatty acids" which have a decided action on metals. This factor cannot be given too careful consideration as a compounded oil may seem to give entire satisfaction for the time being. But the inspection of the cylinder walls after say 2 or 3 years of operation, finds them "honey- combed," and in some instances the stud bolts are eaten off. A very noticeable effect of acidity in compounded cylinder oils is the necessity for frequent renewals of cylinder head gaskets, leaky piston packing and leaky joints in the exhaust pipe. The following are the specifications for steam cylinder oils used by a power corporation controlling a large number of high pow- ered generating plants : Spe)ciFications. To be a mixture of pure mineral hydrocarbon cylinder stock and acidless animal oil. Viscosity not to go below 170 Tagliabue at 212° F. and should 426 ENGINEERING CHEMISTRY show a difference of not over 5 per cent, in the viscosity due to a rise of 10° F. (This oil to have a viscosity at 212° F. equal to a sugar solution at 70° F. Solution is made of 57.1 parts by weight of best refined granulated sugar, 42.9 parts by weight of water.) Gravity to be between 24° and 26° B. at 60° F. Flash point not lower than 560° F. and burning point to be at least 50° higher. Open test: — not less than 50 cc. of oil to be heated per minute until flash point is reached. Should not test more than 2 per cent, by weight of tarry res- inous precipitate when tested by shaking 5 cc. of the oil with 95 cc. of 88° B. naphtha and filtered after 12 hours standing. Should be free from any trace of mineral acid or more than 5 per cent, free fatty acid calculated as oleic. Volatility ; should not lose more than i per cent, per hour at a temperature of 400° F. Should contain no adulteration such as soap ash, resin oil, gumming principles, grit or dirt. The following ingredients to be used in making cylinder oil for this company. 5 per cent, acidless tallow oil. 2 per cent, neatsfoot oil. 93 per cent. Pennsylvania cylinder stock. There should be no separating of stock after 10 grams have been in an open vessel 48 hours, nor should there be any tendency to coagulate at room temperature. The conditions for the above specifications vary from saturated steam at 50 pounds pressure to 200 pounds pressure with 125° of superheat. Most boiler compounds are composed chiefly of caustic soda, potash, etc. When the steam comes in contact with compounded cylinder oils, the animal oil is saponified, or in other words soap is formed which thickens and gums, leaving the mineral oil, which is washed from the cylinder walls. It might be said at this point that water is the factor most detrimental to efficient steam cylin- der lubrication, especially when carrying caustic soda or potash. It is recommended that the water from the drain cocks on the ENGINDKRING CHEMISTRY 427 cylinders be tested from time to time by inserting a piece of red litmus paper in the sample. If alkali is in excess it will turn the litmus paper blue. Steam traps are recommended where possible, as they retain the lime, magnesia and silica which come over in the steam in solid particles thereby scoring the cylinder. The groaning of valves is not always due to a poor grade of cylinder oil. In one case it was found that the slide valve had attained knife edges due to constant operation. By simply round- ing the edges by means of a fine file the difficulty was overcome. A method of comparing the oil film of various oils on cylinder walls is to rub them with a number of sheets of tissue paper. Take off enough sheets so that the stain can just be noticed on the top sheet. Then run the oil to be tested comparatively for the same length of time and repeat the operation with the tissue paper. If a thicker film remains on the cylinder wall a greater number of sheets will be stained, and this oil will give better lubrication than the original. Turbine: OiIvS. Invariably oil used for the lubrication of steam turbines should be a pure filtered mineral oil. The presence of adulter- ants or foreign matter of any kind will result in an emulsion if the steam comes in contact with the oil, and the circulatory system of lubrication will become clogged. The lighter bodied oils give the best results because of the very high rate of speed. Then again the lighter oils retain their vis- cosity to a higher degree than the heavier oils. The most important factor in turbine oil is that it should sepa- rate very readily from water. (See motor oils for emulsion test.) The same rule holds for reciprocating engines with a circu- latory system using an oil filter. Cutting O11.S. The objects of cutting oils are to facilitate a clean cutting of the surface of the metal, to overcome the chattering of the cutting tool, and to carry off the generated heat. The conditions vary in a great many ways, from the machin- 428 ENGINEERING CHEMISTRY ing of cast iron to the machining of the parts for the most ac- curate electrical instruments. A large furnace concern uses straight mineral oil on the taps for the threading of iron castings which compose the hot water heaters. Lard oil is universally looked upon as the most efficient cutting oil, but owing to the high cost of the best grade experiments were made which proved that a compound of lard oil and a good grade of mineral oil produced entirely satisfactory results. The per- centage of lard oil depends entirely upon the fineness and accu- racy of the work in question. A concern manufacturing most of the standard electrical instruments uses a compound of 30 per cent, of lard oil and 70 per cent, of a filtered mineral oil. However, for general machine work a compound of 10 to 15 per cent, of a good grade of lard oil and the remainder of medium bodied mineral oil will be found to cover the general re- quirements of cutting oils. Ice Machine Oie. The principal requirements of oil for the lubrication of ice machines are : flash point high enough to withstand the heat gen- erated when the gas is compressed ; cold test low enough that the oil will not congeal at the temperature of the expanded ammonia gas. Hence the oil must be so refined as to have minimum paraffine content. When condensed exhaust steam is used for the manufacture of ice it is essential that as little oil as possible be used in the lub- rication of the cylinder to avoid discoloration of the ice. Fatty compounds should not be used in the lubrication of the cylinder as a milky emulsion hard to eliminate will be seen in the ice if the oil gets past the piston. Transformer Oies. Oil in transformers is used as an insulating and cooling med- ium, therefore it should be a pure mineral oil obtained by the fractional distillation of petroleum, free from any adulterant. £;ngine:ering che:mistry 429 As one of the main functions of the oil is that of an insulator, the dielectric strength should be very high. The method of testing transformer oil by the Westinghouse Electric and Manufacturing Co. is as follows : (a) Use a transformer of at least i kilowatt capacity, pro- vided with suitable means for varying the voltage on oil testing apparatus. Style No. y2,/\2y (W. E. & M. Co.). (&) Carefully clean oil testing apparatus w^ith benzine or gas- oline and thoroughly drain. Use particular care to see that no moisture becomes mixed with the oil or condenses on the ap- paratus. (<:) The temperature of the oil should be between 20° C. (68° F.) and 25° C. {y]"" F.). (c?) Pour about 200 cc. of oil into the testing jar and adjust gap to 0.15 inch. (^) Apply the testing voltage, and raise rapidly and uniformly without opening the circuit until breakdown occurs. (/) Agitate oil thoroughly, reset gap and repeat test, until ten breakdowns have been obtained. Take average of ten tests as breakdown voltage of the oil. (^) Certain oil will give a minute discharge spark between the terminals at a voltage considerably lower than the true break- down voltage. Care must be taken not to mistake this discharge spark for the breakdown. The oil should be moisture free or as nearly so as possible. The separative qualities vary greatly in diiferent oils. Acid treated oils emulsify, therefore a straight filtered oil is necessary. The presence of traces of acid used in the refining of oil or of alkali is not permissible for two reasons : First, the presence of acid or alkali reduces the strength of the dielectric, and second, they have a corrosive or destructive effect on the materials of which the transformer is composed. The viscosity of transformer oil is of foremost importance since one of its main functions is that of cooling. The more viscous the oil the slower its circulation, hence the transfer of heat will be correspondingly slow. Heavy oil will not cir- 430 ENGINEERING CHEMISTRY culate freely through the oil ducts of the transformer, therefore, a high temperature gradient exists between the oil and the trans- former windings. Transformer oil should be free from deposit. The deposit is objectionable chiefly because it retards the circulation of the oil by clogging the oil ducts. The deposit is an indication that a chemical decomposition is taking place and should be rectified immediately. The flash and fire points of the oil are of minor importance as the maximum temperature does not exceed ioo° F. The loss of the oil by evaporation should be very low and if the oil disappears rapidly the cause is probably due to the oil syphon- ing out through poorly designed leads, or that the transformer box is not air tight. The color of the oil should be light for inspection of the trans- former when submerged. Then again it is sometimes necessary to make changes on the terminal board below the oil level. Dynamo Oii^. The range of sizes of dynamos of the present day calls for oils of a great variety. The oils, however, should all be filtered spindle oils free from any ingredient which tends toward gum- ming. The bearings should be at approximately room temperature and any very noticeable rise should be looked after immediately. Hot bearings may be the result of a tight belt, unequal air gaps, due to the wear of the Babbitt metal in the bearing, high viscosity of oil, etc. An instance is known where a very high grade oil was used on the bearings of a high speed motor, but invariably they over- heated. Upon close inspection it was found that the oil was so viscous that it held the oil rings stationary. The oil was drained from the reservoirs and a lower viscosity spindle oil was placed therein. The motor was started, the oil rings revolved and the bearings cooled down during operation. DNGINEEJRING CHEMISTRY 43I Compressor O11.S. The conditions on an air compressor vary to a very marked degree as compared with the steam engine. The interior of the cylinder is dry and warm, in some cases very hot, the degree of heat depending upon the pressure to which the air is compressed. Hence an unadulterated mineral oil should be used containing absolutely no fatty oil. The oil should have a high flash point due to the high degree of heat within the cylinder. Great care should be exercised on this point. Serious accidents may take place because an oil of a low flash point volatilizes and may be ignited by the high degree of heat resulting in cracked cylinders, explosions and some times total wrecks of plants. The viscosity of the oil should be as low as possible insuring maximum lubrication. The free carbon content should be at a minimum, because an oil which carbonizes rapidly will tend to clog up the pipes and prevent the valves from closing. Especial care should be exercised to keep out foreign matter, dirt, grit, and the like. The oil should also have a low cold test, as the expanding gas carries heat with it, resulting in a low temperature at that point. Hardening and Tempering OiIvS. The use of oil in the process of hardening and tempering steel is a very broad and delicate subject and the best results can only be obtained by experience combined with experiment. When the oil is used for the purpose of hardening steel, it serves as a quenching medium, whereas in tempering, it is a heated bath in which the steel is submerged to bring it up to the required temperature. By using oil lesser degrees of hardness are obtained than when water, brine, mercury or similar mediums are employed. The heat is dissipated less rapidly and the action is not so severe. In this case the result desired is the essential, taking into considera- tion the quality and shape of parts to be hardened and the de- gree of hardness required. 432 i;ngine;e:ring chemistry There still exist many differences of opinion among steel men versed in this subject regarding the proper and most efficient quenching medium and it is also true that though but one oil be used a wide range of results may be obtained by different opera- tors. It is evident therefore that the results are due chiefly to the manner in which the oil is applied, the temperature to which it is heated before quenching, and the length of time of sub- mersion. The tendency to-day is to disregard all of the old theories and "hobbies" and use a straight mineral oil of a high flash point. The flash point is one of the most important points to be con- sidered in selecting an oil for the above use. It must be high enough so that the oil will not disintegrate and volatilize at a rapid rate. In some special cases a special brand of oil such as sperm, cotton-seed or lard oil is used, but in common practice, the cost does not warrant its use, hence a straight mineral oil will give the most satisfactory and efficient results. Motor O11.S. The factor which demands the most attention in the operation of internal combustion engines is that of lubrication. Unlike the general scope of machine lubrication motor oils must withstand a very high degree of heat and with this point in view the process of manufacture must be such that oils contain no detrimental ingredients. At present there are two ways in which motor oils are manu- factured : One by destructive distiflation and treatment with sulphuric acid and an alkali ; the other by steam distillation and filtration through Fuller's earth. The former is the cheaper method of manufacture but by far the more costly in the end to the operator. The distillate is treated with sulphuric acid to throw down unstable compounds and at the same time free carbon is carried down with them. After being washed with water, an alkali is introduced to neu- tralize the acid that may remain. Thus salts of the alkali are formed, which cannot be entirely removed. Hence when subjected to a high degree of heat in the cylinders a chemical ENGINKERING CHEMISTRY 433 reaction takes place forming compounds that react on the cylin- der walls, piston and piston rings causing them to corrode and leak. Excess carbon is formed, which scores the cylinders and pistons. On the other hand the oil made by steam distillation is filtered under reduced pressure through Fuller's earth. This is entirely a mechanical process and no adulterants are used during the treatment. Hence there are no chemical reactions in the cylin- ders and carbon deposit is reduced to a minimum. Fatty compounds of any description should never be used on account of the liberation of fatty acids under heat. The base of the petroleum from which the motor oil is made should be carefully considered, that is, whether paraffine or as- phaltum. The former has been proven by extensive tests to be the more satisfactory. Heat Test. — The method of manufacture of the oil, that is whether acid treated or filtered, can easily be determined by heating the oil until vapors are given off. Retain it at this tem- perature for about 20 minutes. An acid treated oil will turn black, and upon standing for 12 hours will show a black precipitate showing that a chemical reaction has taken place and that there are foreign ingredients in the oil. A filtered oil will darken in color but will show no sediment. Emulsion Test. — Take 35 cc. of the oil and an equal amount of water (preferably distilled in a 4-ounce bottle). Shake the mixture for about ^2 hour and allow it to stand for a day. An acid treated oil will show a line of emulsion between the oil and the water. A filtered oil will separate out completely showing that there are no foreign ingredients in the oil. As all mineral motor oils are composed of about 85 per cent, carbon and 15 per cent, hydrogen a carbonless oil is impos- sible. What is meant by low carbon content of an oil is the free carbon. Even this factor never reaches zero, and can be determined by distilling (destructive distillation) a given amount of the oil in a flask and weighing after evaporating to dryness. (Rate of distillation i drop per second.) The solid matter re- maining in the flask is the percentage of carbon residue. 28 434 Engine:ering chemistry However, if the conditions are not favorable, such as leaky pis- ton rings, etc., the highest quality motor oils will leave an excess deposit of carbon. Generally the carbon residue of motor oils is higher in the heavier bodied oils. The flash point of motor oils is important only from the stand- point of disintegration. Practically all oils, disregarding quality, are destroyed when introduced into the explosion chamber of the cylinder, due to the high degree of heat therein. But the temperatures of the various parts of the engine when in opera- tion determine the loss by evaporation. The approximate aver- age temperatures of the various parts when in operation are as follows : Degrees Piston heads 300 to 1,000 Piston walls 200 to 400 Cylinder walls 180 to 350 Crank bearings 140 to 250 Oil well 90 to 200 Therefore it will be seen that an oil of a flash below 400° F. will disintegrate very rapidly and will require very frequent replenishing. Also, from the above temperatures it will be seen that the cold test is of minor importance except where the oil is fed through exterior piping. The viscosity of motor oils is the most important factor in lubrication. If the viscosity be either too high or too low the friction of the moving parts is increased, hence loss of horse- power and high cost of operation. By extensive tests it has been proven that oils between 180 and 300 seconds are regarded (Saybolt Universal Viscosimeter) as meeting the general scope of motor lubrication requirements, subdividing the above into the standard grades as regards viscos- ity they may be classified as follows : Light body for high speed light duty, where splash system of lubrication is used, 220 seconds at 70° F. (Saybolt universal viscosimeter). Medium "body, 275 to 300 seconds at 70° F, ( Saybolt universal viscosimeter). (The oil will answer the general requirements of automobile motors.) DNGINKKRING CHEMISTRY 435 The oils for heavy duty slow speed engines where force feed system of lubrication is used should have a viscosity of 400 to 450, and for extra heavy oil for air-cooled motors, such as motor- cycles, etc., the viscosity should be about 100 seconds at 212° F. Transmission oils and gear compounds should be semi-fluid. If a hard grease is used the gears cut a permanent track in the grease and in time the teeth become dry, resulting in excessive friction. The grease should be a compound of fiber grease and a heavy cylinder stock. The cheaper grades of greases are made by compounding heavy oils and paraffine wax. This is a poor lubricant. Internal combustion engines should not use as much oil as steam engines as the cylinder walls are dry. The conditions approach that of superheated steam cylinders. A steel company employing De La Vergne's oil engines as a source of motive power mix 3/2 pound of finely divided graphite with the cylinder oil and assert that it materially increases the efficiency of the engine. The Analysis of Lubricating Oils Containing Blown Rape-Seed and Blown Cotton-Seed Oils. Rape-seed oil has long been the standard oil in Europe for lubrication. Its constancy of viscosity at varying temperatures, its non-liability to acidity as compared with other seed oils, and its low cold test, unite in producing the results required of a good lubricant. It, however, is no exception to the rule that vegetable and animal oils suffer partial decomposition when sub- jected to high temperature produced by friction, with the result that fatty acids are liberated and corrosion of bearings produced. The substitution of mineral oils in varying proportions with rape-seed oil has reduced this tendency, this reduction being determined by the percentage of mineral oil present, as the latter liberates no free acids. It is a peculiar fact, however, that a mineral oil alone does not give as satisfactory results in lubrication (especially cylinder 436 ENGINEERING CHEMISTRY lubrication^) as does a mixture of mineral and vegetable or mineral and animal oils, one of the primary causes being that the viscosity of mineral oils rapidly diminishes at high tempera- tures, v^hereas the reduction of viscosity of vegetable and animal oils is very much less. If it w^ere not for this peculiarity between these two classes of oils, mineral lubricating oils could easily supplant (on the score of cheapness) all other oils used in lubrication. The admixture of oils then being required for the better class of lubricants, it follows that in England where rape-seed oil has been the standard, its use should be continued in compounded oils. The proportion of rape-seed oil added to mineral oil varies from 5 to 20 per cent. Where the mineral oil is a clear paraffine oil 20 per cent, of the seed oil is used ; where the mineral oil is a dark, heavy oil, 5 per cent, is generally added. The separation and estimation of the rape-seed oil in these mixtures presents no difficulty to the analytical chemist when no other seed oil is present, since the saponification of the seed oil, the separation of the fatty acids and recognition of the same are a part of the usual chemical work of this character. The recog- nition of the constituents of a mixed lubricating oil by analysis is a very dififerent problem from giving a formula by which the mixture can be made. This is evidenced as follows : Suppose the analysis shows — Per cent. Rape-seed oil 20 Paraffine oil 80 Paraffine oil varies in specific gravity from 0.875 to 0.921, and it is essential to include in the report of the analysis not only the amount of paraffine oil, but also the gravity, since paraffine oil of gravity 0.875 is a very different product from that of 0.921 gravity, the former selling at 7^ cents and the latter at 23 cents per gallon. This determination can be made by taking the gravity of the original mixed oil (0.912), then knowing by the analysis 1 The Railroad and Engineering Journal bA, 73-126. ENGINEERING CHEMISTRY 437 that 20 per cent, is rape-seed oil (gravity 0.918), the gravity of the 80 per cent, of paraffine oil is easily calculated. Thus : X = specific gravity of rape-seed oil (0.918), y ^= specific gravity of paraffine oil. Then i-^ + ^y = o.gi2, 0.183 + ^y ~ 0-912, 1^ = 0.729, _>^ :^ 0.910. The mixture being composed, therefore, of — Per cent. Paraffine oil (specific gravity 0.910) 80 Rape-seed oil (specific gravity 0.918) 20 The direct determination by analysis from the ether solution of the mineral oil in the mixture does not give an oil of the same specific gravity as the mineral had before it was mixed with the seed oil. This can be accounted for by the volatilization of a portion of the lighter hydrocarbons of the mineral oil when the ether is expelled during the analysis. For this reason the deter- mination of the percentage of seed oil and the calculation of the mineral oil offers less liability to failure than finding the mineral oil directly. The introduction of blown rape-seed oil instead of the normal rape-seed oil complicates the investigation and renders the use of the formula above given valueless. Rape-seed oil has a gravity of 0.915 to 0.920. Rape-seed oil blown has a gravity of from 0.930 to 0.960. Two difficulties are immediately presented : ( i ) The chemical analysis does not indicate whether the rape-seed oil is blown or not; (2) The use of the formula given without the correct gravity of the blown oil would give false results regarding the paraffine oil. To overcome this difficulty some synthetical work is required. Suppose the specific gravity of the mixed oil is 0.922 and the analysis shows 20 per cent, of rape-seed oil. It will be necessary then to produce a mixture in these proportions that will duplicate the original sample. A check upon this will be the viscosity of 438 DNGINDDRING CHEMISTRY the original sample as compared with the one to be made by formula. Thus : The original oil has a gravity of 0.922, contains (by analysis) 20 per cent, of rape-seed oil, and has a viscosity at 100° F. of 335 seconds (Pennsylvania Railroad pipette). First. — Make a mixture of paraffine oil (specific gravity 0.910) generally used in this character of lubricant, 80 per cent., and rape-seed oil (unblown), 20 per cent. The viscosity is 165 sec- onds, showing that this mixture cannot be used in place of the original oil. Second. — Make a mixture of paraffine oil (specific gravity 0.910) and rape-seed oil partially blown (specific gravity 0.930), in the same proportions as above. The resulting viscosity is 267 seconds, showing that the compound is still lacking in viscosity. Third. — Make a mixture of paraffine oil (specific gravity 0.910), 80 parts, and rape-seed oil, blown (specific gravity 0.969), 20 parts ; the viscosity is 332 seconds. This now fulfils the conditions required and the synthetical sample agrees with the original in gravity, composition, and viscosity. The use of blown rape-seed oil is being gradually replaced by blown cotton-seed oil. The latter, which has had but a limited use in lubrication, owing to its liability to acidity, has been greatly improved by this process of "blowing," which is nearly complete oxidation of the oil under comparatively high tem- perature. This largely prevents the occurrence of the acidity in the oil, and thus the main objection to its use in lubrication disappears. It is much cheaper than rape-seed oil, since it costs 30 cents per gallon, to 60 cents per gallon for the latter. The chemical re- actions of the two oils are very similar, and careful analytical work is required that the chemist be not misled. The following table of comparisons will indicate this : Specific Gravity. Cotton-seed oil 0.920 to 0.925 Rape-seed oil 0.915 to 0.920 Blown cotton-seed oil 0.930 to 0.960 Blown rape-seed oil 0.930 to 0.960 i^NGlNKERING CHEMISTRY 439 Viscosity (Pennsylvania Railroad Pipette) at 100° F. Seconds Cotton-seed oil (specific gravity 0.925) 162 Rape-seed oil (specific gravity 0.918) 210 Blown cotton-seed oil (specific gravity 0.960) 2,143 Blown rape-seed oil (specific gravity 0.960) 2,160 Heidenreich's Test. Before stirring After stirring Cotton-seed oil Faint reddish brown Brown Rape-seed oil Yellow-brown Brown Massie's Test. Cotton-seed oil ... Orange-red Rape-seed oil Orange Iodine Absorption. Cotton-seed oil 104 to 114 Blown cotton-seed oil 93 to 103 Rape-seed oil 102 to 108 Blown rape-seed oil 94 to 100 In the comparison of the tw^o oils, when not mixed with a min- eral oil, the above tests can be used. The conditions are altered, however, when either one or both are so mixed, since these tests apply only to the pure oils and not to those reduced with large percentages of mineral oil. After the separation of the seed oil from the mineral oil by saponification the identification of the seed oil depends upon the reactions of the fatty acids obtained, and a careful examination and comparison of these reactions shows that the melting points have the greatest difference and thus become a means of recognition. Thus, the fatty acids from rape-seed oil melt at 20° C, and from cotton-seed oil at 30° C. Hence, if upon analysis of a lubricating oil under above conditions, the fatty acids obtained show a melting-point of 20° C. the seed oil can be pronounced rape-seed oil. If the melting-point is between these limits, say 23° C, the seed oils are present in a mixture, the proportions of which can be determined by the following formula : 440 ENGINEERING CHEMISTRY Wi = proportion of rape-seed oil, W2 = proportion of cotton-seed oil, Ws = weight of mixture (20 per cent.), h = temperature of melting-point of fatty acids of rape-seed oil, fi = temperature of melting-point of fatty acids of cotton-seed oil, t:i =: temperature of melting-point of mixed fatty acids. Then 4 — 4 h h L —L w-i = '^4 r H — H Inserting the value : 2X — ^o w. = 20-^^ — = 14 per cent. ' 20 — 30 ^ 2X — 20 w.y = 20 -^ = 6 per cent. ^ 30 — 20 ^ Or, Per cent. Paraffine oil 80 Rape-seed oil 14 Cotton-seed oil 6 Total 100 By synthetical work upon these proportions, with comparison of viscosities of the sample submitted with the product, the result will be not only a correct analysis but a working formula can be given by which a manufacturer can duplicate the original oil. Rapid Determination of Fatty Oil Mixtures with Mineral Lubricating Oils. In the presence of large quantities of mineral oil, saponifica- tion with alcoholic potash takes a long time, since the mineral oil prevents the potash coming in contact with the fatty oil. Schreiber's method^ gives good results in a short time. Weigh 5 grams of the oil in a 200 cc. Erlenmeyer flask, add 25 to 50 cc. of half-normal alcoholic potash and sufficient benzol (CyHj.) to dissolve the oil when warmed (generally 25 cc. is enough, but with heavy cylinder oils as much as 50 cc. may be necessary; in this case it is well to add 25 cc. of neutral alcohol). Connect the ly Am. Chem. Soc, 1907, 29, 74. e;ngine:e:ring chemistry 441 flask with a 3-foot condenser and set it on the iron plate that forms the top of the steam bath, so that the steam will not strike it directly, and regulate the seat so that the condensing liquid will not be forced to the top of the condenser. In this way the con- tents of the flask can be boiled without apparently losing any of the solvent. Boil for 30 minutes. Cool, add phenolphthalein, and titrate the excess of potassium hydroxide with half-normal sul- phuric acid. On adding the acid the liquid separates into two layers and the change in color can be seen in the lower layer; the titrating, however, must be conducted slowly. Subtract the cubic centimeters of sulphuric acid from the amount used on blanks and calculate the saponification number. For all prac- tical purposes 195 may be considered as the saponification number of the fatty oils used in lubricating oils, hence if S equals the determined saponification number, 100 S divided by 195 equals the percentage of fatty oil present. OIL USED FOR ILLUMINATION. Oil used for illumination may be classified into two groups : 1. Refined products from petroleum, such as naphtha, gaso- lene, kerosene, signal oil, etc. 2. Certain refined oils of vegetable and animal origin, as colza oil, rape oil, lard oil, sperm oil, etc. Refined Products fpom Petroleum. Kerosene is the refined product from petroleum that distills over (in the refining process) after the lighter oils, naphthas, etc., have been separated, and is the principal oil in use for illumina- tion. In color it varies from standard white to water-white (colorless), and its commercial value is dependent upon its flash- ing and burning-point. In the oil trade, the burning or fire tests are classified as 110° F., 120° F., 150° F., and 300° F. The 150° F. is known as headlight oil and the 300° F. as min- eral sperm and mineral colza. The requirements for mineral oils to be used in railroad illu- mination are as follows : 442 ENGINEERING CHEMISTRY Specifications for Petroleum Burning Oils. (Conditions of Shipment and General Specifications.) This material will be purchased by weight. Barrels must be in a good condition and must have the name of the contents and the consignee's name and address on each barrel, and plainly marked with the gross and net weight which will be subject to the company's weight. When received all shipments will be promptly weighed. If not prac- ticable to empty all the barrels, lo per cent, will be emptied, and the losses of the whole shipment will be adjusted in accordance with the lo per cent, taken. Should the net weight thus obtained be less by i per cent, than the amount charged in the bill, a reduction will be made for all over I per cent. Prices should be given in cents or hundredths of a cent per pound. Shipments, one or more barrels of which are filled with oil cloudy from the presence of glue, or which contain dirt, water, or other impuri- ties, will be rejected. Two kinds of petroleum burning oils will be used, known as the 150° fire test for general use, and 300° fire test for use in passenger cars. Detaii, Specifications. 130° Fire Test Oil. This oil must conform to the following requirements : 1. It must have a flash test above 125° F, 2. It must have a fire test not below 150° F. 3. It must have a cloud test not above 0° F. 4. It must be a "water white" in color. 5. Its gravity must be between 44° and 48° B. at 60° F. 300° Fire Test Oil. This oil must conform to the following requirements : 1. It must have a flash test above 250° F. 2. It must have a fire test not below 300° F. 3. It must have a cloud test not above 32° F, 4. It must be a "standard white" in color. 5. Its gravity must be between 38° and 42° B. at 60° F. Flash Test. Fire Test. — The requirements for the flash and fire test for illuminating oils used for domestic purposes are not so rigid as for railroad practice. In fact large quantities of oil, flashing below 110° F., are used, the cheaper price being the incentive. So dangerous are these oils v^ith low^ flash-points, that many states have passed stringent law^s against their use. An oil with a fire test of 110° F. very often has a flash test of 90° F. .and ENGINKEJRING CHEJMTSTRY 443 many oils with a fire test of 120° F., flash at or below 100° F. It is the flashing-point of an oil that makes it dangerous and while the refiners of oil mark their products by the fire test the laws, as passed by many of the states, specify the flash test as the requisite. Cloud Test. — The cloud test is made as follows : Two ounces of the oil are placed in a 4-ounce sample bottle, with a thermom- eter suspended in the oil. The bottle is exposed to a freezing mixture of ice and salt and the oil stirred with the thermometer while cooling. The temperature at which the cloud forms is taken as the cloud test. The instrument that gives good satisfaction in testing illumi- Fig. 87. nating oils, not lubricating oils, for the flash and fire test is called the Wisconsin Tester (Fig. 87). It is thus described: 444 ENGINEEJRING CHEMISTRY (i) On the left side of figure is shown the instrument entire. It con- sists of a sheet-copper stand 8^ inches high, exclusive of the base, and 4^^ inches in diameter. On one side is an aperture 3^ inches high, for introducing a small spirit lamp. A, about 3 inches in height, or better, a small gas burner in place of the lamp when a supply of gas is at hand. The water bath, D, is also of copper, and is 4% inches in height and 4 inches inside diameter. The opening in the top is 2% inches in diameter. It is also provided with a j4-inch flange which supports the bath in the cylindrical stand. The capacity of the bath is about 20 fluid ounces, this quantity being indicated by a mark on the inside. C represents the copper oil holder. The lower section is 3)^ inches high and 2^ inches inside diameter. The upper part is i inch high and 3^ inches in diameter, and serves as a vapor chamber. The upper rim is provided with a small flange which serves to hold the glass cover in place. The oil holder contains about 10 fluid ounces, when filled to within J^ inch of the flange which joins the oil cup and the vapor chambers. In order to prevent reflection from the otherwise bright surface of the metal, the oil cup is blackened on the inside by forming a sulphide of copper, by means of sulphide of ammonium. The cover, C, is of glass, and is 35^ inches in diameter ; on one side is a circular opening, closed by a cork through which the thermometer, B, passes. In front of this is a second opening ^ inch deep and the same in width on the rim, through which the flashing jet is passed in testing. The substitution of a glass for a metal cover more readily enables the operator to note the exact point at which the flash occurs. A small gas jet, j4 inch" in length, furnishes the best means for igniting the vapor. Where gas cannot be had the flame from a small waxed twine answers very well. (2) The test shall be applied according to the following directions: Remove the oil cup and fill the water bath with cold water up to the mark on the inside. Replace the oil cup and pour in enough oil to fill it to within % inch of the flange joining the cup and the vapor chamber above. Care must be taken that the oil does not flow over the flange. Remove all air bubbles with a piece of dry paper. Place the glass cover on the oil cup, and so adjust the thermometer that its bulb shall be just covered by the oil. If an alcohol lamp is employed for heating the water tub, the wick should be carefully trimmed and adjusted to a small flame. A small Bunsen burner may be used in the place of the lamp. The rate of heating should be about 2° per minute, and in no case exceed 3°. As a flash torch, a small gas jet, ^ inch in length, should be employed. When gas is not at hand, employ a piece of waxed linen twice. The flame in this case, however, should be small. When the temperature of the oil has reached 85° F., the testings DNGINEERING CHE:mISTRY 445 should commence. To this end insert the torch into the opening in the cover, passing it in at such an angle as to well clear the cover, and to a distance about half way between the oil and the cover. The motion should be steady and uniform, rapid and without any pause. This should be repeated at every 2° rise of the thermometer until the temperature has reached 95°, when the lamp should be removed and the testings should be made for each degree of temperature until 100° is reached. After this the lamp may be replaced, if necessary, and the testings continued for each 2°. The appearance of a slight bluish flame shows that the flashing point has been reached. In every case note the temperature of the oil before introducing the torch. The flame of the torch must not come in contact with the oil. The water bath should be filled with cold water for each separate test, and the oil from a previous test carefully wiped from the oil cup. (3) The instrument to be used in testing oils which come under the provisions of Section 2 of the law shall consist of the cylinder, D, and the copper oil cup, C, together with a copper collar for suspending the cup in the cylinder, and an adjustable support for holding the thermometer. (4) The test for ascertaining the igniting point shall be conducted as follows : Fill the cup with the oil to be tested to within ^ inch of the flange joining the cup and the vapor chamber above. Care must be taken that the oil does not flow over the flange. Place the cup in the cylinder and adjust the thermometer so that its bulb shall be just covered by the oil. Place the lamp or gas burner under the oil cup. The rate of heating should not exceed 10° a minute below 250° F., nor exceed 5° a minute above this point. The testing flame described in the directions for ascer- taining the flashing point should be used. It should be applied to the sur- face of the oil at every 5° rise in the thermometer, till the oil ignites. There are various forms of flash and fire testing apparatus for illuminating oils, and as some are standard in certain states, a few will be described. Dire:ctions for Using the Tagliabue: Opkn Te:stkr. (Fig. 88.) The instrument should stand level. Partially fill the metal bath cup with water, leaving room for displacement by the glass oil cup, which then place in the bath. Fill the glass oil cup with the oil to be tested to within y% inch of its upper level edge. See that there is no oil on the outside of the cup, or upon its upper level edge, using filter paper to clean with in preference to cotton or woolen material. Adjust the horizontal flashing- taper-guide- 446 EJNGINEKRING CHEMISTRY ^^^^ ^jmgi ENGINEEJRING CHEMISTRY 447 wire in place. Suspend the thermometer, with the bulb of same well covered by the oil. Heat bath with small flame lamp — alco- hol, gas or other — having the flame so adjusted that it will raise the temperature of the oil not faster than 2^ per minute, without removing the lamp during the whole operation. Remove air bubbles, if any, from the surface of the oil before first trial for flash is made. At the proper trial temperatures noted below, try for flash with a small (not over ^ inch) bead of flame on the end of a piece of lighted twine, or an equivalent sized gas jet; by drawing it quickly and without pause across the guide-wire from left to right. Triai, Temperature Tabee. For oils ex- pected to have a fire test c^ Try for flash hirst at Then at 110° F. 8? F. 90° 95° ICX)° 105° 108° 1 10^ 115° 90° 95° 100° 105° 110° 113° 115' 120° 95° 100° 105° 110° 115° 118° 120' 125° 100° 105° 110° 115° 120° 123° 125' 130° 100° 105° 110° 115° 120° 125° 130= 135° 105° 110° 115° 120° 125° 130° 135' 140° 110° 115° 120° 125° 130° 135° 140' 145° 115° 120° 125° 130° 135° 140° 145' 150° 120° 125° 130° 135° 140° 145° 150' As the Tagliabue Closed Tester for Illuminating Oils (Fig. 89) is similar in construction to the Foster instrument for the same purpose, except the latter has an automatic flash extinguisher, a description of the Foster is given. The instrument consists of a copper lamp furnace containing a water bath and oil cup; the latter surmounted by a closed vapor chamber, which is pierced at two points symmetrically placed for the reception of a thermometer and a flashing lamp or taper; the apparatus being elliptical in shape, the thermometer is placed in one focus of the ellipse and the flashing taper in the other. The flashing taper consists of a small cylindrical wick holder, supported by radical arms to an annular ring, and rests upon a similar ring at the bottom of an open, shallow basin — the 44^ ENGINEERING CHEMISTRY Spaces between the radical arms giving egress to the oil vapor, w^hile the v^ack itself extends down into the body of the oil within the cup. An inverted conical thimble, resting upon the rim of the basin, prevents the dissipation of the vapor. The thermom- eter is mounted in a copper tube cut away in front to expose the scale, the bulb of the thermometer, when in position, being within the body of the oil at a definite distance below the surface. An orifice around the tube of the thermometer, definite in diameter and distance above the surface of the oil, allows a downward current of atmospheric air when the flashing taper is alight. An index is placed within the water bath and within the oil cup for maintaining uniformity in the filling of each. The heating lamp of 'the lamp furnace has its wick adjustable to facilitate uniformity in the rate of heating. Directions for Using the Foster Automatic On. Tester. (Fig. 90.) 1. Remove the thermometer with its mounting from the oil cup. 2. Lift off the oil cup containing the flashing taper, and fill the open water bath with water half full. 3. Now take out the wick holder from the oil cup, and fill this vessel with the oil to be tested — pouring in the oil at the place of the wick holder and noting the gauge mark at the thermometer hole — pour in the oil very gradually as the surface approaches the gauge mark. The gauge mark consists of a small pendant shelf, and the oil cup is properly filled when the upper surface of the oil just adheres to the lower surface of the gauge mark. Too much care cannot be taken at this point; therefore, having ceased pouring, tip the cup so that the oil flows away from the gauge; and then gradvially restoring it to the horizontal, see that the surface again adheres, and add a little more oil if it does not. 4. See that the wick of the flashing taper be adjusted to give a very small flame — a flame that does not exceed ^ inch in height. A flame that exhibits as much blue at its base as yellow at its top is right. 5. Now set the oil cup on top and into the water bath ; return the flashing taper to its place, inverting the conical thimble around e:nginke)ring chemistry 449 it, and return the thermometer to its place upon the cup; in doing this be sure that the casing of the latter is pushed down upon the cup as far as it will go. 6. Fill the lamp beneath half full of alcohol, light it and put it in its place beneath the water bath. No wnote the rate of increase in temperature as shown by the thermometer, and adjust Fig. 91 the wick to raise the temperature at the rate of 2° per minute. When the temperature has reached 90°, light the flashing taper and observe it closely. As soon as the oil under test has reached its ''flashing point" the flame of this taper will be extinguished by the "flash," and the point of attention is to note the tempera- ture at the instant the flame of the taper is extinguished. This 29 450- KNGINEKRING CHEMISTRY "flashing point" is the point of temperature at which the oil generates a vapor and indicates that this has formed an explosive mixture with atmospheric air. The instrument for determining the flash and fire test of illumi- nating oils used by the chemists of the Standard Oil Co. is the Saybolt Electric Tester, Fig. 91. Requirements oe Various States Regarding the Flash and Fire Test oe Illuminating Oils.^ state Flash pt. °F. Fire, °F. Instrument Arkansas 130° Tagliabtie (closed) Columbia, District of.. 120° Connecticut 110° 140° Tagliabue (open cup) Florida 130° Tagliabue (closed) Georgia 120° Illinois 150" Tagliabue (closed) Indiana 120° Indiana Iowa 105° "Closed Test" Kansas 110° 110° Tagliabue (closed) Kentucky 130° Louisiana 125° Tagliabue (closed) Maine 120° Tagliabue (open) Maryland 110° Tagliabue (closed) Massachusetts 110° Tagliabue (open) Michigan 120° 148° Foster (closed) Minnesota 110° Minnesota Missouri ...'. 150° Tagliabue (closed) Montana 1 10° Nebraska 100° Foster (closed) New Hampshire 100° 120° Tagliabue (closed) New Jersey 100° "Closed Tester" New Mexico 150° New York 110° Tagliabue (closed) North Carolina 100° Foster (closed) North Dakota 100° Ohio 120° Foster (closed) Pennsylvania No law or requirement Rhode Island 110° Tagliabue (closed) South Dakota 110° Foster (closed) Tennessee 120° TagHabue (open) Vermont 110° Tagliabue (closed) Wisconsin 120° "Wisconsin" (closed) ^Gill: Oil Analysis (with addition by author). EJNGINEERING CHEMISTRY 45 1 Specifications for ''Mineral Sperm Oil" (Illuminating Oil). Issued by the Navy Department, June 15, 1910. Superseding Specifications 24-O-2, issued June, 1902; April, 1905, and April 8, 1908. Must be prime white or better and free from all cloudiness, impuri- ties, or adulterations ; must not become cloudy at any temperature above 32° F. ; must be entirely free from acid; must not flash below 255° F. (open tester), 300° F. fire test, and have a specific gravity between 37° and 41° B. (0.8383 to 0.8187) at 60° F. ; Lima oil products excluded; to be purchased and inspected by weight. Inspection and Deuvery. 1. Before acceptance the oil will be inspected. Samples of each lot will be taken at random, the samples well mixed together in a clean vessel, and the sample for test taken from this mixture. Should the mixture be found to contain any impurities or adulterations, the whole delivery of oil it represents will be rejected, and is to be removed by the contractor at his own expense. 2. The quantity delivered to be determined by weight ; the number of pounds per gallon to be determined by the specific gravity of the oil at 60° F. multiplied by 8.33 pounds, the weight of a gallon (231 cubic inches) of distilled water at the same temperature. 1. Determination of the Color of Kerosenes. The grades of color of an oil are noted as standard white, prime white, superfine white and water white,^ and the instrument generally used for determination of the color in oils is the Stammer colorimeter (Fig. 92). Tube I is closed at the bottom by a transparent glass plate, is open at the top, and a project- ing lip on the side whereby the oil to be tested can be poured in or out. The tube is fastened to the stand by two screws. The measuring tube III is closed at the bottom by a colorless glass plate and is movable inside of tube I. The color glass tube II which is joined firmly to the measuring tube III is open at the bottom and at the top contains a colored glass plate, which plate can be substituted with other tinted glass plates. The movement of the joined tubes II and III is produced by enclosed ratchet work, the movement of the tubes being read on a scale on the back of the stand, and stated in millimeters. ^ In Bremen, the varieties are rated as water white, prime white, standard white, prime light straw, light straw and straw. 452 ENGINEERING CHEMISTRY Since the color of a liquid is inversely proportional to the height of the column, which is necessary to give the standard color, and since this color is here expressed by lOO, the absolute number for expressing the tone of color of any oil is obtained by dividing Fig 92. this 100 by the number of millimeters read off from the scale. For example : Millimeter scale Color 1 100.00 2 50.00 7 1429 19 5.26 The color, tone, and thickness of the standard glass is so chosen that the scale shows the following values for the ordinary brands of illuminating oils : Millimeters Standard white 50.00 Prime white 86.50 Superfine white 199-50 Water white 300.00 ENGINEERING CHEMISTRY 453 Wilson's colorimeter, largely used in England, is very similar in construction to the Stammer.^ 2. Vegetable and Animal Oils. The two principal oils of this class in use for illumination are colza and lard oil. In this country the former has never been used to any great extent, its use being confined principally to Europe, but lard oil and sperm oil, in former years, before the introduction of the petroleum products for this purpose, were largely used as illumi- nants. Except in railroad practice and then in yearly decreasing amounts their use now is very limited in this direction. In the matter of illumination, the methods made use of by the railroads are worthy of study and comparison, and it is, in a great measure, due to the investigations carried out in their interests that the great advances in this direction are due. LINSEED OILS. The Principal Chemical Tests. Specific Gravity. — The specific gravity of pure linseed oil, with- out driers, at 60° F., is 0.932, which equals 20° of the Baume hydrometer; that of boiled oil is 0.941, which equals 19° B. The addition of 25 per cent, of cotton-seed oil reduces the Baume hydrometer 1° and the addition of 10 per cent, of paraffine oil (neutral oil for instance) reduces the hydrometer ^°. Iodine Value. Hann's Method. "Weigh in a small glass capsule from 0.2 to 0.25 gram of oil. Transfer to a 350 cc. bottle having a well-ground stopper. Dis- solve the oil in 10 cc. of chloroform and add 30 cc. of Hann's solution (see below). I^et it stand with occasional shaking for I hour. Add 20 cc. of a 10 per cent, solution of potassium iodide and 150 cc. of water, and titrate with standard sodium thiosul- ^ A simple colorimeter for general purpose, as used by the Department of Agricul- ture, United States Government, and designed by Mr. Oswald Schreiner, is described in Journal American Chemical Society, September, 1905. 454 ENGINEERING CHEMISTRY phate, using starch as an indicator. Blanks must be run each time, "From the difference of the amounts of sodium thiosulphate required by the blanks and the determination, calculate the iodine number (centigrams of iodine to i gram of oil). "The iodine number of raw linseed oil varies from 175 to 193, though Gill states that a pure raw oil may run as low as 160. Boiled oil may be very much lower. "Make the Hann's solution by dissolving 13.2 grams of iodine in 1,000 cc. of glacial acetic acid, which will not reduce chromic acid, and add 3 cc. of bromine." (P. H. Wai^ker.) Specifications for Raw Linseed Oil. Issued by the Navy Department, August 2, 1915. Superseding Specifications 52-O-1, issued Aug. 15, 1912. Generai, Instructions. 1. General Specifications for Inspection of Material, issued by the Navy Department, in effect at date of opening of bids, shall form part of these specifications. QUAUTY. 2. Raw linseed oil shall be strictly pure, well-settled oil, perfectly clear and free from foots. Chemicai, Constants. 3. The oil shall show upon examination : Maximum Minimum (Percent.) (Percent.) Loss on heating ^ hour at 103° to 105° C 0.2 Specific gravity at 15.5° C 0.937 0.932 Iodine number (Hann's) .< 190.0 178.0 Saponification number 192.0 189.0 Acid number 30 Refractive index at 25° C 1.4805 1.479 Unsaponifiable matter 1.5 Physical, Characteristics. 4. The oil when flowed on a glass plate, which is held in a position inclined 30° to the vertical, shall dry practically free from tackiness in 75 hours at a temperature of 60° to 80° F. Basis oe Purchase. 5. To be purchased by the commercial gallon and inspected by weight. The number of gallons to be determined at the rate of 7^ pounds of oil to the gallon. ENGINEERING CHEMISTRY 455 Specifications for Boiled Linseed Oil. Issued by the Navy- Department, February 2, 1914. Superseding Specifications S2-O-2, issued Nov. 20, 191 1. Composition. 1. Boiled linseed oil shall be absolutely pure boiled oil of high grade, made wholly by heating pure linseed oil to over 350° F. with oxides of lead and manganese for a sufficient length of time to secure a proper combination of the constituents and be properly clarified by settling or other suitable treatment. Evidence of the presence of any matter not resulting solely from the combination of the linseed oil with the oxides of lead and manganese will be considered grounds for rejection. Chemicai, Constants. / 2. The oil shall upon examination show : Unsaponifiable matter Not more than 1.5 per cent. Lead oxide (PbO) Not less than 0.20 per cent. Manganese oxide (MnO) Not less than 0.04 per cent. Iodine number (Hanus) Not less than 178. Specific gravity at 60° F Not less than 0.938. The oil shall give no appreciable loss at 212° F. in a current of hydrogen. Physicai. Characteristics. 3. When flowed on glass and held in a vertical position, the oil shall dry practically free from tackiness in 12 hours at a temperature of 70° F. Basis of Purchase. 4. To be purchased by the commercial gallon; to be inspected by weight, and the number of gallons to be determined at the rate of T^/z pounds of oil to the gallon. FUEL OIL. Characteristics and Testing of Fuel Oil.^ From the standpoint of the petroleum trade, fuel oil in general includes all oils w^hich are not saleable for some other special purpose at a higher price than that which prevails for oils to be sold as fuel oils, to be burned under boilers. From the trade ^ Characteristics and Testing of Fuel Oil, prepared by Thomas B. Stillman, Jr., of the Babcock and Wilcox Co. Engineering staff, who has made a specialty of the appli- cation of fuel oil to steam generation. 456 e;ngine:e:ring che:mistry point of view, it also includes special distillates which are sold as Diesel oils. It does not include various distillates burned for power purposes, such as gasolene, naphtha, motor spirits, and various kerosene distillates. The actual amount of oil devoted to fuel purposes varies continually with the condition of the produc- tion of crude petroleums. During the time of flush production, such as has existed in the Oklahoma fields during the past year due to the extraordinary production in the Gushing field, a great deal of crude petroleum is sold as fuel oil on account of the neces- sity of disposing of it when no better market is available. The use of such oil is not advisable for fuel purposes, not only be- cause it contains valuable gasolene and kerosene, but these con- tents so lower the flashing point of the fuel oil as to make it open to the objections to gasolene stored in a confined space, as in a battleship, where the vapor is liable to produce explosions on con- tact with air. "Until the last twelve months, much of the production of Cali- fornia crude oil was sold for fuel purposes practically as it came from the well. Within the last year, however, the practice of topping off the valuable gasolene and kerosene in a compara- tively small proportion of California oils has so increased that not more than 25 per cent, of the fuel oil of California is now crude oil. In States other than California and Oklahoma, and to a slight extent Texas, fuel oils consist chiefly of the least valu- able distillates and some residuum. The distillates of such low value as to be sold for fuel are usually the products distilling off after kerosene, and those too heavy for burning in lamps and also too thin to be used as lubricating oils. Such oils generally have the name of gas oils, and are more valuable for use as gas oils than for fuel purposes, but the market for gas oils is easily over-supplied, and the surplus goes for fuel oil. The annual production of petroleum in the United States is about 266,000,000 barrels,^ of which at least 100,000,000 barrels is consumed as fuel. This proportion will hold fairly well for the petroleum production of the world, that is, about ^ of the world's 1 Dr. David T. Day. — Director of U. S. Bureau of Mines, 1915. ENGINEERING CHEMISTRY 457 production may be considered "fuel oil." In regions such as the East Indies, the petroleum is to a large extent too valuable for use as fuel. This is offset by Mexico and Russia, where the use of residuum and even crude oil for fuel purposes is very gen- eral, and exceeds the average proportion. There are three kinds of petroleum in use, namely, those yield- ing on distillation — 1st. Paraffin; 2nd. Asphalt ; 3rd. Olefine. To the first group belong the oils of the Appalachian range, and the middle-west of the United States. These are dark brown in color with a greenish tinge. Upon their distillation such a variety of valuable light oils are obtained that their use in the crude condition as fuel is prohibitive because of price. To the second group belong the oils found in Texas and Cali- fornia and in Mexico. These vary in color from a reddish brown to a jet black and are used very largely for fuel. The third group comprises the oils from Russia, which, like the second, are largely for fuel purposes. ^Whether or not it "pays" to use oil depends on many things. There may be reasons that makes its adoption imperative at prac- tically any cost — certain military advantages for instances — such as smoke prevention, speed of vessel, etc., but the merchant owner will be influenced 1st. By comparative cost of coal, wood or fuel delivered at the boiler ; 2nd. Relative heat value of the fuel; 3rd. Relative capacity and efficiency in steam production. This may result in being able to run a plant natural draft with oil instead of forced draft with coal, thus saving the cost of in- ^ E. H. Peabody, Member Am. Soc. M. E., Soc. N. A. M. E., in paper read before the International Engineering Congress, San Francisco, California, 1915. 458 e:ngine:e:ring che:mistry stalling and operating blowers, or in the installation of less boiler power. 4th. Expense of fitting up for oil including suitable storage provisions for the fuel. • 5th. Saving in labor due to reduction in number of firemen, elimination of coal passers, and the expense of removal of ashes. 6th. Increased life of boilers and lower maintenance charges, both in the fireroom and the engine room. When considering marine installations, the following two points : 7th. Increased bunker capacity and longer steaming radius. 8th. Time saved on voyage due to steadier steam pressure and possible time saved in fueling ship. In round numbers, as a steam producer, a pound of oil is equal to a pound and a half of coal, or approximately one ton of coal is equal to four and one half barrels of oil, or to quote another ap- proximate but handy rule, one ton of coal equals two hundred gallons of oil. Mr. Walter M. McFarland has derived a simple relationship between the relative costs of oil and coal as follows : 2A = B. Where A 1= cost of oil in cents per gallon (7.88 pounds), and the B ^ the cost of coal in dollars per ton (2,240 pounds). Thus, when the cost of coal in dollars per ton is double the cost of oil in cents per gallon the fuel costs of producing steam will be ap- proximately equal. While no general statement of cost can be used for obtaining more than an approximation for individual cases, it is hoped that what is given above may be of some assistance in showing the many advantages of oil over coal fuel and the probability of its saving money. To indicate the requirements of a good fuel oil the following specifications used in the purchase of fuel oil for the U. S. Navy may prove of interest : ENGINEERING CHEMISTRY 459 (a) Fuel oil shall be a hydrocarbon oil of best quality, free from grit, acid, or fibrous and other foreign matter likely to clog or injure the burners or valves. (b) The unit of quantity to be the barrel of 42 gallons of 231 cubic inches at a standard temperature of 60° F. For every vari- ation of temperature of 10° F. from the standard 0.4 of i per cent, shall be added or deducted from the measured or gauged quantity for correction. (c) Flash point never under 150° F. as a minimum (Abel or Pensky-Marten's closed cup), or 175° F. (Tagliabue open cup), and not lower than the temperature at which the oil has a viscos- ity of 8° Engler (water = i Engler). (Example: If an oil has a viscosity of 8° Engler when heated to 186° F. then 186° is the minimum flash point at which this oil will be accepted.) (d) Viscosity at 100° F. not greater than 200° Engler. (e) Water and sediment not over i per cent. If in excess of I per cent, the excess to be subtracted from the volume; or the oil may be rejected. Note. — If an Engler viscosimeter is not available, the Saybolt standard universal viscosimeter may be used, and 280 seconds Saybolt will be considered equivalent to 8° Engler, and 7,000 seconds Saybolt will be considered equivalent to 200° Engler. Water at 60° F. = 30 seconds Saybolt. Water and sediment will be taken by the distillation method. When oil in small lots is consigned to naval vessels or to navy yards, the centrifuge test will be used in order to obviate delay. In this test 50 cc. of oil and an equal quantity of the best commercial benzol, 50 per cent, white, will be used, and the mixture heated to 100° F. The "flash point" of fuel oil is the temperature at which it gives off inflammable gases, and is a question of irhportance in determining its availability as a fuel. In general it may be stated that the light oils have a low and the heavy oils a much higher flash point. There are, however, many exceptions to this rule ; as for example; some of the. heaviest Mexican crude oils of 11° to 12° Baume frequently have flash points of 100° or 125° F. As 460 ENGINEERING CHEMISTRY the flash point is lower the danger of ignition or explosion be- comes greater, and the utmost care should be taken in handling the oils with a low flash point to avoid this danger. On the other hand, because the flash point is high, is no justification for care- lessness in handling these fuels. With proper precautions taken, in general, the use of oil as fuel is practically as safe as the use of coal. The Baume hydrometer scale for liquids lighter than water has obtained a stronghold in the fuel oil industry, and for light oils this practice is justified by the ease with which the gravity may be determined; namely, by the simple reading of the scale on the stem of the hydrometer immersed in the liquid. But for heavy viscous oils, the very nature of the oil makes this process a slow one and liable to considerable error. It is believed by some, that for these oils it is much better to determine the weight of a known volume of the oil (as in the specific gravity bottle), and report the density in terms of the density of water at 60° F., i. e., as specific gravity. On the other hand, there are advocates of the method of heat- ing viscous oils sufficiently to make the use of the Baume hydrom- eter feasible, making the necessary corrections in temperature. The specific gravity bottle method is, however, to be preferred for accurate work, as it eliminates the possible error which may be introduced by the temperature correction, which varies for different oils. The United States Bureau of Standards has adopted the fol- lowing formula for converting readings on the Baume scale lighter than water, to terms of specific gravity. Specific gravity at 60° F. = ■ — ~ ^ & ^ 130 -f Baume The conversion table on the following page is given for handy reference : e:nginee:ring chemistry 461 Baum6 10 II 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 35 40 Specific gravity at 60° F. 1. 000 0-993 0.986 0.980 0.973 0.966 0-959 0.952 0.946 0.940 0.933 0.927 0.921 0.915 0.909 0.903 0.897 0.892 0.886 0.8S1 0.875 0.848 Weight in pounds — 60° F. Per U. S. gal. 8.337 8.280 8.222 8. 171 8.112 8.054 7.996 7-937 7.887 7.837 7-779 7.729 7.679 7.629 7.579 7-529 7-479 7-437 7-387 7-345 7-295 7.070 6.862 Per cu. ft. 62.368 61.931 61.495 61. 121 60.684 60.247 59.811 59-374 59.000 58.626 58.189 57.815 57-441 57-067 56.693 56.318 55-944 55-632 55.258 54.946 54-572 52.888 51-329 Per barrel (42 gal.) 350.17 347-72 345-32 343-17 340.70 338.27 335-82 333-35 331.25 329.15 326.71 324.61 322.51 320.42 318.31 3x6.21 314.10 312.35 310.25 308.50 306.40 296.94 288.20 The specific heat of fuel oil varies with its composition. It will be greater the richer the oil is in hydrogen, and lower in pro- portion to a greater carbon content. The following figures are reproduced from Holde's work on examination of hydrocarbon oil:^ Crude oil from Japan Pennsylvania Russia California • . . Specific heat 0.435 0.500 0.435 0.398 With the advent of the viscous crude oils of Mexico, and the increased use of the heavy distillates from other fields, coupled with the wider adoption of mechanical atomizers, the degree of the viscosity of the oil becomes a matter of considerable import- 1 Published by John Wiley & Sons, 1915. 462 ENGINEERING CHEMISTRY ance. A description of different viscosimeters and the methods of using them are explained at length in other parts of this book. The following points in regard to the value of viscosity as ap- plied to fuel oils are of importance. In handling oil through the pumps and piping of a fuel oil system it is necessary to have its viscosity reduced to at least 375° Engler, and preferably to 300°. These figures also hold for the delivery of fuel oil to steam or air atomizing burners, where the steam or air is the medium used for atomizing the oil and delivering it in a fine spray in the furnace. In the case of a mechanical atomizing burner, however, it is necessary to reduce the viscosity to 8° Engler before the oil will give a satisfactory atomization, this figure of 8° applying particularly to the better grades of oils, such as navy standard. In the case of the heavy Mexican oils a viscosity of 10° to 12° Engler is sufficiently low to produce an atomization which is the equal of that given by the better grades of oil at 8° Engler. In all cases where mechanical atomizers are used it is better to reduce the viscosity below 8° (or 12° as the case may be) as in general, the lower the viscosity, the less tendency there is for the burners to produce smoke. Certain grades of the heavier oils contain considerable sul- phur, and the question is frequently asked whether or not cor- rosion from this cause may result. Experience has demonstrated that sulphur in oil has no bad effect on boilers, except in cases of neglect, when pitting may occur under certain conditions, as with coal. Corrosion of copper heating coils has, however, been noticed in the presence of sulphur-bearing oils, and for this reason, it is the recognized practice to use steel coils. Brass and bronze fit- tings may be used however, with safety, both on pumps and on pipe lines. The function of an oil burner is to atomize or vaporize the fuel so that it may be burned like a gas. All burners may be classified under two general types : 1st. Spray burners, in which the oil is atomized by steam or compressed air; ENGINEE^RING CHEMISTRY 463 2nd. Mechanical burners, in which the oil is atomized by sub- mitting it to a high pressure and passing it through a small orifice. Spray burners are almost universally used for land practice and the simplicity of the steam atomizer and the excellent econ- omy of the better types together with the low oil pressure and temperature required makes this type a favorite for stationary plants, where loss of fresh water is not a vital consideration. In marine work or in any case where it is advisable to save feed water that otherwise would have to be added in the form of "make-up," either compressed air or mechanical means are used for atomization. Spray burners using compressed air as the atomizing agent, are in satisfactory operation in some plants, but their use is not general. The air burners require blowers, com- pressors, or other apparatus which occupy space that might be otherwise utilized, and require attention that is not necessary when steam is used. Where burners of the steam or air type are used, heating the oil is an advantage, not only in causing it to atomize more easily, but in aiding economical combustion. In the case of mechanical atomizers it is necessary to warm the oil until its viscosity is reduced to at least 8° Engler, as noted above. The temperature is, of course limited somewhat by the flash point of the oil used especially in navy work, but as heavy Mexican oils with a flash point of 125° F. have been frequently raised to 280° F. to burn them satisfactorily in mechanical burners, without serious results, there seems to be no reason why the temperature should not be carried above the flash point, if necessary, provided it is not carried high enough to cause decomposition of the oil or a carbon deposit in the supply pipe. In the case of steam atomizers if the temperature is raised to a point where an appreciable vapori- zation occurs, the oil will flow irregularly from the burner, and cause the flame to sputter. The mechanical system of burning oil is especially adapted for marine work as the quantity of steam required for putting pressure on the oil is small, and the condensed steam may be 464 ENGINEERING CHEMISTRY returned to the system. The only method by which successful mechanical atomization has been accomplished is the one in which the oil is given a whirling motion within the burner tip. This is done either by forcing the oil through a passage of helical form or by delivering it tangentially to a circular chamber from w^hich there is a central outlet. The oil is fed to these burners under a pressure that varies with the make of the burner and the rate at which each individual burner is using oil, and is usually be- tween 50 and 200 pounds to the square inch. The oil particles fly off from such a burner in straight lines in the form of a cone rather than in the form of a spiral spray, as might be supposed. Where, in the spray burners, air for combustion, is ordinarily admitted through a checker-work under the burner proper, in the mechanical burner, it is almost universally admitted around the burner and the problem of properly mixing the air with the spray of oil under these conditions is such a difficult one that very few burner manufacturers have satisfactorily solved it, they usually being well contented to obtain 10 to 11 per cent. COg in the flue gases, without CO. It is, however, possible to obtain as high as 14^ per cent. CO^ without CO by giving the air a proper "twist" as it enters around the burner, and it was with burners using this principle that the excellent results shown below were obtained. To give an idea of the importance of the composition of the flue gases in burning oil under boilers the following table is given showing the rapid loss in boiler efficiency, as the percentage of CO2 in the flue gas goes down. The presence of CO in the flue gas is another cause for lost efficiency which must be guarded against. Also the oil must be burned with practically no smoke, as a heavy oil smoke produces a tarry carbon deposit on the boiler tubes which it is difficult to remove and which prevents the proper transfer of heat to the water in the boiler. ENGINEERING CHEMISTRY 465 Boiler Efficiencies— Oil Fuel.* Showing the maximum theoretical efficiency, for a given per cent, excess air supply and flue gas temperature, based on assumptions stated below. As- sum- Assumed temperature of flue gases— Fahr. As- sum- ed ed Per 1 Per cent. 375° 400° 425° 450° 475^ 500° 550° 600° 700° 800° cent. ex- ex- cess cess sup'y Calculated boiler efficiency, Per cent. air sup'y 81.35 80.78 80.20 79.61 79.03 78.44 77.28 76.11 73-77 71.43 10 80.62 So. 04 7941 78.78 78.15 77.52 76.25 74.99 72.47 69.94 10 50 77.90 77.08 76.25 75.44 74.62 73.80 72.16 70.52 67.24 63-97 50 ICO 74.42 73-37 72.32 71.26 70.21 69.16 67.04 64-95 60.72 56.51 100 150 65.79 64.51 61.93 59-35 54.20 49.04 150 200 59.86 56.81 53.77 47.67 41.58 200 250 51.69 48.18 41.15 34-11 250 300 46.58 42.60 34.63 26.65 300 Approximate relation between per cent, CO2 and excess air supply as per assumptions below. Per cent. Co., ' 5 6 7 8 9 10 II 12 13 14 15 15-9 Per cent, excess gjj- 283 -5 155 120 93.0 72.2 55.5 41.9 30.6 21.0 12.7 5.6 1 This table was prepared by Chas. C. Moore & Co., Engineers, San Francisco, Cal. Values in above- table are conditioned on the following assump- tions : Average temperature of air for combustion entering the boiler, 80° F. ; humidity, 80 per cent. ; air per pound of oil chemically required for complete combustion, 14 pounds ; B. t. u. per pound of oil as fired, 18,500; chemical composition of the oil as follows: carbon, 86 per cent.; hydrogen, 11 per cent.; sulphur, 0.8 per cent. ; nitrogen, 0.2 per cent. ; oxygen, i.o per cent. ; water, i.o per cent. Per cent, of excess air stated is measured at boiler outlet and consequently includes leakage through boiler setting. The loss by radiation has been taken as 3 per cent. This loss varies with the size of boiler, insulation, etc., being less than 30 466 ENGINEERING CHEMISTRY indicated for larger sizes of boilers and greater for smaller sizes. An allowance of 2 per cent, for undetermined losses has been made. This quantity is subject to considerable variation and may, in exceptionally favorable instances, be as low as 0.5 per cent. The following constants are taken from Marks and Davis' steam tables : Absolute temperature, — 459.6° F. Heat of vapor- ization at atmospheric pressure, 970.4 B. t. u. Specific heat of superheated steam at atmospheric pressure, for the range of tem- perature from 212° to 700° F., 47. To give an idea of the results which may be obtained under the best conditions with steam atomizers, the following boiler tests made on a 600 horse-power Babcock & Wilcox stationary type of boiler, using Peabody steam atomizers in conjunction with a Peabody furnace, by Doctor D. S. Jacobus^, are given: Date— 1907-08 Duration of test, hours Steam pressure, by gauge Temperature of feed water Factor of evaporation Draft in furnace— inches of water . Draft at damper Temperature of flue gases °F. .... Flue gas analysis, % bv volume • • CO2 •' O CO N Oil burned per hour Water evaporated per hour, from and at 2 1 2° Evaporation from and at 212° per pound of oil Per cent, of rated capacity devel- oped B, t. u. per pound of oil EflEiciency of boiler Dec. 28 Dec. 30 Dec. 31 8 8 8 183.1 182.4 178.9 141. 8 144.2 160.7 I-I793 1. 1956 1.1526 0.03 0.2[ 0.03 0.03 0.38 0.02 401.4 492.9 378.4 1350 1268 13.90 2.91 3-43 2,12 0.06 0.29 0.22 83.53 83.60 83.76 1,436.0 2,705.0 938.0 22,052.0 38,340.0 14,234.0 15.37 14.17 15.18 105.8 183.9 68.3 17.S46 17,839 17,682 83.58 77.08 83.31 Jan. 13 8 183. 1 147.3 1. 1665 0.035 0.006 364.1 578.0 8,869.0 15.35 42.6 17.871 83.35 The heat balances^ for the above tests are as follows : 1 Formerly Professor of Experimental Engineering, Stevens Institute, and at present Advisory Engineer of the Babcock and Wilcox Co. - For method of calculating Heat Balances. See " Boiler Testing." ENGINEERING CHEMISTRY 467 Date Heat absorbed by boiler Loss due to moisture in oil ' « " " moisture from burning H. Loss due to heat in dry gases " " " CO (( a " radiation, etc. Total Dec. 28 B. t. u. Per cent. 14915.O 20.4 1225.4 1274.4 19.5 391.3 17846.0 83.58 O. II 6.86 7.14 o.ii 2.20 100.00 Dec. 30. B. t. u. Per cent. 13750.6 9.9 1277.7 2309.7 185.0 306.1 17839.0 77.08 0.06 7.16 12.95 1.04 I.71 100,00 Date Heat absorbed by boiler Loss due to moisture in oil " " " moisture from burning H. Loss due to heat in dry gases " " '* CO " •' " Radiation, etc. Total Dec. 31 B. t. u. Per cent. 14730.7 14.9 1221.4 1296.4 54.3 364.3 17682.0 83.31 0.08 6.91 7.33 0.31 2.06 Jan. 13 B. t. u. Per cent. 14895.6 16.6 I214.9 1238.0 33.6 472.3 17871.O 83.35 0.09 6.80 6.93 0.19 2.64 Under careful operating conditions, using the best type of steam atomizers approximately 2 per cent, of the steam gener- ated by the boiler will be used in atomizing the oil. If the tem- perature of the oil is reduced until its viscosity is low (10° or 15° Engler), less steam is required to produce perfect atomization, the steam consumption in actual practice, under these conditions, having been reduced to one half of one per cent, of the steam generated by the boiler. As an illustration of what may be expected when burning oil with mechanical atomizers, the following tests are given as repre- sentative of good practice: 468 ENGINEERING CHEMISTRY < 3 go o ^ CCS •o'n rj-^ >,cOO ro CjN'^-'. CO - • • • ' CO _ o ^p ^^8%!1^-S'-^ 5?^^^^^o O O I 6 COCO S ^ ^. ^-^ up 4^ ^ lOvO On <^. -. g CO o o ° £; I I ^^a^ CO lO rO 'o5 v£) _. lO lO lO ^ j^ • ^ -. ^ o fo ^ "?^ lo ^ q^ O CO ^X ^^ ^ lOOO CO '^^ O u 'OOrOf^(N(NOt f^.CO f^CO "^^ • • • . « ON t-c I I -p '-' O (s '^ CO r>» ^ ^ (M — /-^ :?^8-8'8g,Sa^S>^=o. o o Xi Xi Xi 1> 3J CJ ^'^'^'^•"^•^'~''~< ^ c- :u a- Cu cu ENGINEERING CHEMISTRY 469 ^. 1 5 >- rO VO 8 8 4; ci r^^ 1 T^ VO CO t-^ On ' Tt- 8 fn HH U A <1 K to M ON CN CO 10 1 Tf q^ Ttco 1 00 10 i-T h-T a; ^• _j ;; CO CM^ Ost^ 8 5 10^ rOvO 8 «N ^ ^ t^ ?i 10 10 CO rO 1 toco fo d pj t^ 1-1 8 r^ 1^ ^d r^ d d 10 8 •r" K^ a c: < „• Tt ' T*- 10 10 -^ 10 M ^j «■ T? h-T CN cR K CO « .." hT od" ^. J- to to 8 j; lOCO 0^ cOTJ• 8 s -^ -^ c< r-> t~^ 0; dr^c^l ^ 8 CN vd MD I-; Tt CO 8 ^ fl, ^ &, _• CO tN VO vO ^ „• Tt CO 1- (N 8 '^ ^ z^ "^ 1 r^ "^ to t^ r^ .<^vO ■M to TfCO_ 1 CO ^ 4J CO^ >->^ <^^ CO «■ 10 i-T hT 2^ m Tt >-^ hT co' S VOOO .^ rO 8 s VO ^ r^ CO 8 fj rj ro « CN q i; 00 -^ CO « t-^ *o ;- 6 t^vd d ^' 8 I- 4; J \o vd d CO 00 8 P, 2 p!, "o >> 1 to M VO - '* t^ fO 8 "_ CN CO -^ to -^ (N r^ x^ CO CO ^ rO rt; - m_ ^ r^ « '-<^ q_ P5 to "-• ►-< t-i a^ P5 ^ i-T hT m" 00 '"' '"' "^ . . - . k> . . . . t< • . . Ci • 'd • 'd • (U , * ii * s • C • 3 s : . . y a . a u o; en . rt '5 '5 '3 ►^ cr cr cr" o o o I M u 03 O O CQ K P3 I- 1/5 .2 bcfl o 0(C 0, ovo o 10 rO 1000 «- Tl- rj- On 10 rO^ «-• , "^ ^ - . rOvO^ 10 (N f^^W ^ "^ "^^ "^ "^ ^_ r^co" cf^ScT'Od^d^d^d^ "-"co" o" on o" '-"00" d\ M — p-i — hHCSi-ii-Hi-iH-iC^H-iM — CNCS— I- t^vO v5 CO ^ 00 vO o -* o :!i i 1 1 I I I I 1 1 1 1 1 1 I 1 1 1 ^3 I O O vO rocO « ' «M 1-1 (N ftfti I 1^1 I I I 1 M in be I I OnOn I ON ON C^ I do odd O M ►^ ri- 30 Tt . I d d d d ON M t^ fO cs On OnCO 00 Cn d vd cJ ro d I I VO O C 3 1-" M O HH O ►- 1 I I I I ::? I OC^C'WMM'^fOT}- ONVO rO O I-" -^ O m' >-<" f< d (S rO i-< c ~ -^ ri 10 ON Jj 8 O »OvO rOOi-;i- 3 CO O CO n t;; 'H "ifi 'c« *on CXI ^ c« 05 A» i> 3 3 S^ "tin (U4J»3q;Krj:<3=!5>3 -^ o 5 o 472 ENGINEERING CHEMISTRY Ultimate Analysis of Oils.^ In the ultimate analysis, the composition of the oil is expressed in percentages of carbon, hydrogen, nitrogen, sulphur, and oxy- gen. Unfortunately, there is no simple direct method for the de- termination of oxygen and this percentage is obtained by sub- tracting the sum of the other percentages from lOO. Hence, this method throws the algebraic summation of all the errors incident in the other determinations upon the oxygen. The determina- tion of carbon, hydrogen and nitrogen requires careful manipula- tion and a considerable degree of analytical skill, and since the errors of the oxygen determination are directly dependent upon the errors in the other determinations the accuracy of these de- terminations must be held within definite limits in order to estab- lish a degree of probable accuracy for the oxygen determination. These limits follow : carbon 0.3 per cent. ; hydrogen 0.07 per cent. ; nitrogen and sulphur 0.05 per cent. Carbon and Hydrogen Determination. Carbon and hydrogen are determined by the usual method of combustion in a current of oxygen. 0.2 gram sample of the oil is burned in a 25 burner Bunsen combustion furnace, the purifying reagents through which the oxygen is led, arranged in the order named, are sulphuric acid, potassium hydroxide, soda lime, and granular calcium chloride. The combustion tube is made of trans- parent fused silica, a little less than a meter in length and about 18 mm. internal diameter. Complete oxidation is insured by passing the products of combustion over red hot copper oxide. A layer of lead chromate following the copper oxide removes the sulphur. The absorption train is arranged as follows : The water is ab- sorbed in a 100 millimeter Schwartz U-tube filled with granular calcium chloride. The carbon dioxide is absorbed by potassium hydroxide in a Vanier combined potash bulb and drying tube. It is well to interpose a tube containing a solution of palladium chloride with a calcium chloride guard as a check against the pos- ^ Prepared by E- G. Bashore, chief chemist of the Babcock & Wilcox Company, a standard authority on the subject. ENGINEERING CHEMISTRY 473 sibility of any carbon monoxide passing over and into the ab- sorption train. Oils must of necessity be distilled over very slowly and this part of the manipulation is governed more by the experience of the operator than any hard and fast rule which can be laid down. Nitrogen Determination. Nitrogen may be determined by the Kjeldahl-Gunning method. One gram of the sample is digested with 50 cc. of concentrated sulphuric acid, 0.65 gram of metallic mercury, and 5 grams of potassium sulphate, until the carbon has been completely oxi- dized and all the nitrogen has been converted to ammonia sul- phate. After cooling, the solution is diluted to about 200 cc. with cold water. The mercury is percipitated with potassium sulphide solution (40 grams KgS per liter) and about 2 grams of granular zinc is added to prevent bumping. The solution is then made distinctly alkaline through the addition of a saturated solu- tion of sodium hydroxide, and the flask is immediately con- nected with the condenser. The ammonia from the distillation is absorbed in 10 cc. of standard sulphuric acid, i cc. of which is equivalent to 0.005 gram of nitrogen. The residual acid is titrated with standard ammonia of just half the strength of the acid (i cc. equals 0.0025 gram of nitrogen) with the use of cochineal as the indicator. The method of Will & Varrentrapp is particularly applicable for the nitrogen determination in oils. In this method the nitro- gen content of the oil is converted into ammonia by heating w4th soda-lime. The liberated ammonia is led through a standard sul- phuric solution, the excess of which is titrated with standard alkali. A glass combustion tube closed at one end contains at the sealed end a layer of oxalic acid or calcium oxalate which on heating decomposes with evolution of carbon monoxide and di- oxide. Next to this is the weighed sample of oil mixed with soda-lime and followed by a third layer of soda-lime only, the latter held in place by an asbestos plug. The tube is connected with a bulb containing a fixed amount of standard sulphuric acid 474 ENGINEERING CHEMISTRY solution of which i cc. is equivalent to 0.005 gram of nitrogen. The combustion is carried on in the usual way, proceeding to heat the tube gradually back to the oxalic acid. When the gases from the decomposition of the latter have completely driven out the ammonia, the fixed amount of standard sulphuric acid (i cc. equivalent to 0.005 gram of nitrogen) is titrated against a stand- ard solution of ammonia of just half the strength with the use of cochineal as an indicator. SuiyPHUR Determination. A quick method coupled with a fair degree of accuracy con- sists in determining the sulphur from the "acid correction" in the calorimeter bomb washings. After the acid correction has been applied, the insoluble matter is filtered off and washed with hot water. The filtrate and washings which should have a total volume of about 200 cc. are acidulated with 5 cc. dilute hydro- chloric acid and then are heated to boiling. The sulphur is pre- cipitated through the addition of 20 cc. of a hot 5 per cent, solu- tion of barium chloride. The results obtained by this method are usually somewhat low due to loss of sulphur trioxide in the gas escaping from the bomb. On the other hand, the method has the advantage of ef- fecting a material saving in time. Sulphur may be determined by the longer and more accurate Eschka method. One gram of the sample is mixed in a platinum crucible with about 2 grams of the "Eschka mixture" (i part anhydrous sodium carbonate and 2 parts calcined magnesium ox- ide). About I gram of the mixture is placed over the top to form a cover. It is necessary to have a blank on the sulphur content of the Eschka mixture. The ignition is first started with a very low flame and it is preferable to use alcohol or natural gas flame. Artificial gas often contains so much sulphur that its use may introduce an error into the determination. After the crucible has been heated very slowly and cautiously, the heat is gradually increased until the crucible and its contents become red hot. The contents of the crucible are heated with oc- casional stirring until all the black particles are burned out. e)ngine:ering chemistry 475 After cooling the contents of the crucible are transferred to a 200 cc. beaker and digested with 75 cc. of hot water for about 30 minutes. The solution is then filtered, the residue washed twice with hot water by decantation and then washed on the filter, small portions of water being used for each washing till the filtrate amounts to about 200 cc. Five cc. of bromine water is then added and the solution is made slightly acid with 5 cc. dilute hydrochloric acid. The solution is heated to boiling, and the sulphur is precipitated as barium sulphate with the addition of 20 cc. of a hot 5 per cent, solution of barium chloride. The pre- cipitate is allowed to stand at a temperature a little below boil- ing for at least 2 hours before filtering. After careful ignition to dull redness in an excess of air the crucible and precipitate are cooled and weighed. SOAP ANALYSIS. Soaps may be conveniently classified into — Toilet soaps, the finest grades of which contain no impurities or free alkali; Laundry soaps, in which tallow is present and generally an ex- cess of alkali either as sodium silicate, sodium carbonate, sodium borate, or free alkali; Commercial soaps, which may be subdivided into (a) soft soaps, potash being the base, and (b) "hydrated" soaps, soda being the base ("marine soap" being an example, formed by caustic soda and palmnut oil or cocoanut oil) ; Resin soaps, in which resin is present and an excess of alkali, with tallow, etc. ; and Medicated soaps, containing medicinal agents such as carbolic acid, tar, sulphur, etc., etc. The complete analysis of a soap often presents considerable difficulty — since many adulterants may be used in the cheaper grades, and many substances not adulterants, the use of which is permitted as colorants and for perfume. Allen states that be- 476 ENGINEERING CHEMISTRY sides the alkali and fatty acids and water requisite for the forma- tion of a soap, the following substances have been found in the different varieties — ochre, ultramarine, sodium aluminate, borax, resin, vermilion, arsenite of copper, alcohol, sugar, vaseline, cam- phor, gelatin, petroleum, naphthalene and creosote oils carbolic acid, tar, glycerine in excess, oatmeal, bran, starch, barium sul- phate, sulphur, steatite, clay. Fuller's earth, pumice stone, kie- selguhr, chalk whiting, etc. The common ''yellow soap" is formed by the combination of tallow or palm tree oil or resin with soda; "recovered grease" is also used in the cheaper grades; cotton seed oil, olive oil, hemp-seed oil, palm oil, cocoanut oil, castor oil, lard, and lard oil, are all used in the manufacture of various soaps. The scheme for soap analysis is by A. R. Leeds, Ph.D. Water. For the determination of water, the method of Lowe is often employed. From 8 to lo grams of the soap (which has been reduced to very fine shavings, and represents an average sample) are weighed out between watch glasses and heated in the air-bath, at first from 6o°-70° C. to avoid melting, then at ioo°-i05° C. to constant weight. In selecting the sample in this, as well as in all subsequent determinations, it is essential that an average speci- men be obtained, since the content of water in the different parts of the bar varies considerably. This is best effected by cutting away about one-third from the end and evenly scraping the cut surface of the remainder until a sufficient amount is obtained for analysis. If the determination of free alkali is of considerable impor- tance the soap should be dried in an atmosphere free from car- bon dioxide. The loss at 105° C. represents the water together with other volatile constituents, such as alcohol and essential oils, which may be present. U iJ 'fi g^ . .2^3 3 I' 1/5 «*H •C3T3 C-O = 3ii ;- +J O tn rt = fa t «^:=^ 5:2 -3° J5 - "5 q *- -^ . ^ "O u ^ § •BDijis am puB ajBDijis ui pauiqiuoo Bpos am aiiiiu -ja;ap pde 'xqh mi^ asodmoDag •a^^ojiis umipog •fQgoBj^ 01 a^BinoiBD pu« ajBLidins mmjBq sb qgiaAV paB a^B^idioajj 'S^Bqding umipos qSiajA JO ajBa^m aaAjis mjAv a;Ba5ix •apijomo jaAiis SB •apuoiqo ranipbs ''OO^BK oi ajBiiioiBO puB 'piOB Duniidins ibuijou miAv aiBmjx a^BUoqjBa xiimpos ^ u in "C J2 O S tj^ «-» t o <-' bco « * ■T3 C . rt-O o 3 C8 >, «3 < rt E « = ■" 0^ bcJI K O c « ^ g-'O )^ o n «* i5 ^ ^ J n-l _ a-2 r o ♦- re-- '^ 2 c «° - !5 00 ^ U 8 O o rt C -C' J .^ w w bo ■<«o •^ ^ C« iSi 3 . i Sou. ■? 3 "^ 3 ^ «> 3 K ts -2 o = ** ca S^ ~ o SIX n •>-s 3J O If a: - I o — u 1* - " •*' _ re = >.4, >< ca ,^.- C OT3 en - ^ X = o^'H •^ ^'.S « rt 4^ C I' O-c Esi^-^ o -o 2 w o - " ii o r— =. &< u rt I* a o ^ tn •a % o 0!! rt tn (A .'d W o s 1 *< s "O >, Q £ W Ui 3 Z •O o Oh X <: .ti tA ;t u u j: fi. O F Z 3 o 2 ■«J (^ a < P. 3 w "* r/) c o o < s § 41 w H ■" ^ 1^ CO a be < •j:: - o * - <« S S ^ i> a- •o re 2 ■3-- 5" ill O * li . cs — C'Z ^ H-C = 3 = 5 o- 2 as o a S = « n > - O O < ^ - a re .2 re;; o'" . i: o-^ *; ►J ciS o-o o > I "I' "5 jii c y X - - CO «55f="3 a.^ - «j s 3-^ "VO - (U - tl ^ - K >. /f StiiV o:2oS * ^ •- re . M "rt «re^>c5^^o«J 5 5'^' =v.-- xT, O 3 ■J g \r. = 3 >>re a; rt re — ' 3 o re c X . 3 a: _ o >< C > = W 3 5-= > X - x -S re o 1. >.T3 bt s:^ 'V 3 b£ > w x re n o '^ •i 1^:2 ii _x^^'o re 3 >> i; 4; C8 e:ngine:e:ring che:mistry 479 Method of K. Brown for Free Water and Alkali in Soap.^ Weigh the sample of the finely divided soap into a wide neck flask, fitted with a U-tube filled with granulated soda lime, and dry to constant weight. The loss represents the water, and the soda lime tube prevents the access of carbonic dioxide of the air and consequent conversion of the free alkali into carbonate dur- ing drying. The dried soap is then dissolved in the smallest quantity of alcohol possible and the alkali determined, as usual, by titration with standard acid using phenolphthalein as indi- cator. Unsaponified Matter. For the determination of unsaponified matter^ the soap, which has been dried in the manner indicated, is extracted in a Soxhlet extraction apparatus with petroleum ether, which, for this pur- pose, should boil below 80° C, and should leave no residue upon evaporation. After the extraction is complete, the petroleum ether is distilled off, the residue dried at 100° C. and weighed. In a boiled, well-made laundry soap, there should be no unsa- ponified matter. In addition to unsaponified fats, foreign matters are sometimes found in the petroleum ether extract, such as a soft paraffine (so called "Mineral Soap Stock"), waxes, hydrocarbon oils, phenol, etc. If waxes are found to be present, the dried soap should be extracted with boiling toluene, which dissolves the same better than petroleum ether. Total Alkali. Fatty Acids. The dried soap thus freed from unsaponified matter is next dissolved in hot water, preparatory to determining the total alkali and fatty acids. A pure soap dissolves completely in hot water, and no ordinary product should leave more than a slight residue. If the article examined is a ''scouring soap," the in- soluble residue will be found to contain quantities of fine sand '^ Angew. Chemie., i8, p. 573. 2 Allen's "Commercial Organic Analysis," Vol. II. 480 DNGINDERING CHEMISTRY and sometimes talc. The residue, if appreciable, should be washed by decantation, and eventually brought upon a filter with hot water, dried at 100° C, and weighed, after which, if desired, it can be subjected to further examination. To the aqueous solution is added an excess of half-normal sul- phuric acid setting free the fatty acids which rise to the surface. The beaker or vessel in which the precipitation was effected is next cooled with ice water. When the fatty acids^ have solidi- fied, it is best to decant the liquid, remelt with water two or three times to remove any enclosed mineral acid, again cool, filter, and wash with cold water until the washings are no longer acid, as shown by litmus. The filtrate from the insoluble fatty acids contains the total alkali now present as sulphate, the excess of sulphuric acid and any glycerol which may have been present in the soap, if saponi- fication was effected in the cold. The acid liquid may further contain a small amount of soluble fatty acids. It is first titrated with half-normal potassium hydroxide using methyl orange as indicator.- From the original amount of sulphuric acid added and the number of cubic centimeters half-normal potassium hy- droxide required to neutralize the excess of the same, the total alkali of the sample can be determined. It is calculated to Na^O. After the liquid has been rendered neutral to methyl orange (which indicates the mineral acid), phenolphthalein is added and more potassium hydroxide is run in. The number of cubic centimeters of potassium hydroxide re- quired for neutralizing to phenolphthalein corresponds to soluble in the absence of more definite knowledge as to their nature. The solution is now concentrated and tested for glycerol, which may be determined by evaporating to dryness and extracting ^ Bulletin No. 13, Pt. 4, p. 456, U. S. Department of Agriculture, Chemical Division. ^AJlen's "Commercial Organic Analysis," Vol. II, p. 260. ENGINEERING CHEMISTRY 481 with ether alcohol mixture/ or else by oxidizing to oxalic acid by means of permanganate^ (not always applicable).^ In soaps containing silicates of the alkalies (a not unusual con- stituent), the gelatinous silicic acid which separates on the addi- tion of sulphuric acid remains with the fatty acids on filtration. To separate the fatty acids from this as well as other impurities, proceed as follows : The funnel containing the filter with the fatty acids is placed in a small beaker and heated in an air-bath (Allen's method). As the filter dries, the fatty acids pass through it and collect in the beaker below, while all impurities (silicic acid, talc, etc.) re- main on the filter. Of course, it is necessary to wash the filter, which remains saturated with the fatty acids, with hot redistilled alcohol or petroleum ether, or else exhaust in an extraction appa- ratus. The alcohol or petroleum ether is distilled off and the residue treated in the same way as the main quantity of fatty acids. In determining the fatty acids in a soap, it is frequently con- venient to extract with ether in a separatory funnel.* To do this the soap solution is placed in the funnel and shaken with dilute sulphuric acid and ether. The separated acids are at once dis- solved in the ether. The aqueous solution may be drawn off be- low, the ethereal solution washed with water, the ether evapor- ated, and the residue dried at ioo° C, and weighed. Since the fatty acids exist in the soap as anhydrides and are weighed as hydrates, it is necessary to multiply the weight found by the factor 0.97, which gives the weight of fatty anhydrides. The fatty acids, after having been weighed, may be titrated with half -normal potassium hydroxide, and from these data may be ascertained what portion of the total alkali exists in combination with the fatty acids as soap. 1 Chem. Ztg , 8, 1667. - Ibid, 9, 975. ' Allen's "Commercial Organic Analysis," Vol. II, p. 290. * Chem. News, 43, 218. 31 482 ENGINEERING CHEMISTRY Free Alkali.^ To determine the per cent, of free alkali- in soap, a separate portion is weighed out and extracted with neutral alcohol in an extraction apparatus. The caustic alkali is determined in the alcoholic solution by titrating with half -normal hydrochloric acid, using phenolphthalein as indicator. If, however, the soap con- tains unsaponified fat, as is frequently the case if made by the so-called "cold-process," this method cannot be used, since in alcoholic solution unsaponified fat would be readily saponified by the free caustic alkali present. In such a case the soap must first be dried in an atmosphere free from carbon dioxide at 100° C, the unsaponified matter extracted with petroleum ether, and finally the soap dissolved in alcohol and the free alkali deter- mined in the alcoholic solution as before. Carbonated Alkali. The sodium carbonate, sodium silicate, borax, and everything insoluble in alcohol, remains behind in the extraction tube and may be dried at 100° C. and weighed. If considerable, it may be further treated, as follows : First, it should be exhausted with boiling water; one-half of this solution is then titrated with half -normal hydrochloric acid using methyl orange as indicator. The amount of acid required corresponds to carbonate, silicate, and borate. In this solution sulphates may also be determined and starch and gelatine tested for. The other half of the solution is examined qualitatively for carbonate, silicate, and borate. If there remains a considerable residue insoluble in water, it may be dried at 100° C, weighed and further examined. Resin. Resin is a very common constituent of soaps, the resinates of the alkalies having an action similar to soaps, and the cheapness of the material often suggesting a partial substitution of it for the natural fats and oils. ^ Consult "Method for the Estimation of the Total Alkali, the Free Alkali and the Carbonated Alkali in Soaps," by R. Henriques and O. Meyer, Chemical News, April 18, 1902. ^Allen's "Commercial Organic Analysis," Vol. II, p. 251. ENGINEERING CHEMISTRY 483 As a qualitative test for resin, Gottlieb's^ method is reliable and easily made. The soap is dissolved in water and heated to boiling. A strong solution of magnesium sulphate is added until the fatty acids are completely precipitated. The magnesium resinates remain in so- lution. After boiling 2 or 3 minutes, the solution is filtered and the hot filtrate acidified with dilute sulphuric acid. In the presence of resin the liquid becomes turbid, due to the separated resin acids. The boiling should be continued for Yz hour, to make sure that the turbidity is due to resin acids and not . to volatile fatty acids. One method for the quantitative determina- tion of resin in soap is that of Hiibl,^ as follows : From 0.5 to i gram of the mixture of fatty and resin acids is heated in a closed flask on the water bath with about 20 cc. of alcohol to complete solution. The acids are neutralized with al- kali, using phenolphthalein as indicator. The alcohol soap so- lution is then poured into a beaker, the flask rinsed with water, the solution diluted to 200 cc, and silver nitrate added to com- plete precipitation. The precipitate (consisting of the silver salts of the resin and fatty acids) must be protected from sunlight. It is filtered, washed with water at 100° C, and extracted in a Soxhlet tube with ether. The silver resinates dissolve in the ether, while the silver salts of the fatty acids remain behind. The ethereal solution, as it leaves the extraction tube, should be yel- low or light brown in color, but not dark brown. It is filtered, if necessary, and the filtrate shaken with hydrochloric acid in a separatory funnel. The resulting ethereal solution of the resin acids is filtered from the silver chloride, washed with water, and the filter and separator rinsed with ether, the ether distilled off, and the residue dried at 100° C. As the resin is weighed in the hydrated form, its weight must be multiplied by the factor 0.9732 to obtain the weight of the anhydride. Twitchell's method for the determination of resin in a mix- ture with fatty acids depends upon the formation (in alcoholic solution) of the ethereal salts of the latter when treated with 1 Benedikt's "Analyse der Fette u. Wachsarten," p. 121. 2 Benedikt's "Analyse der Fette u. Wachsarten," p. 125. 484 ENGINEERING CHEMISTRY hydrochloric acid, resin being unacted upon. The gravimetric method is as follows : Two or 3 grams of the mixture of fatty acids and resin are dissolved in ten times their volume of absolute alcohol and dry hydrogen chloride is passed through in a moderate stream, the flask being placed in a vessel with water to keep it cool. The gas is rapidly absorbed, and after about 45 minutes the ethereal salts separate and float on the solution. After waiting Yz hour longer, the liquid is diluted with five times its bulk of water and boiled until the acid solution is clear, the ethereal salts, with resin in solution, floating on top. To this is added some light petroleum, and the whole transferred to a separatory funnel, the flask being washed out with light petro- leum. The acid liquid is then run off, and the petroleum ether solution washed once more with water and then treated in the funnel with a solution of 0.5 gram of potassium hydroxide and 5 cc. of alcohol in 50 cc. of water. The resin is immediately saponi- fied, and the two layers separate completely. The resin soap solu- tion can then be run off, and the resin recovered, as usual, by the addition of an acid. The first stages of the volumetric method are similar to those of the gravimetric, with the exception that the contents of the flask are washed into the separating funnel with ether instead of light petroleum, and the ethereal solution is then thoroughly washed with water until all soluble acidity is re- moved; 50 cc. of neutral alcohol are then added, and the solution titrated with standard solution of sodium hydroxide. It is frequently of interest to know the origin of the fatty acids of a soap which is, however, in many cases, a problem not easily solved. The only clues are to be sought in the specific gravity, combining weight, melting- and solidifying-points and iodine number of the fatty acids. The values for the specific gravities in column III, page 485, were obtained with a Westphal and a Reiman's balance plum- met, with a thermometer of a range 95°- 101° C. Occasionally, fats, before being used in soap making, are bleached by various chemical agents, the most common of which I x" spiDB 00 d d c^ d d d d d d d d w •«»• CO d d d d d CO fO O\00 lOTj-M TfCTviOO^O lOvO UO O O 0\ hi N «0 cs d d d» d « d d d ro d -«■ d rr ■* « o - - d o .o >< S 1 SppB I > SppB AllBJ o c) vo no q o q o „ o o q q o \q q cni ^_ vq S > SppB s ^ I'^l'l 'I'i ^ '1" > 1 SptOB au'bj 1 ,, cOirONiio iii'^iirOiNi'^ irOii > ill sppn oooooo cooooooor^o oo-o -f- -(- oooooo oooooooo^do oo«o -t- + £ .to •^ "? T "1" "^ "T TT T '^TTT t "" T "^TTT (in "-"lit §8 ^ OO «D O lOOfO 00OO»OrOOOOO IDOIOO 9cg|.28J?^ Mi595>gHg^| -§oo| MDOOOOOCaOOO COOOOOvOOOOOOOOO OnOJ 00 00 00 «5 (55" d o d d d d d d ^ d d d d d d 6 6 6 6 6 6 2 o^ o d d d d o d d (g^ d d d d d d d d d d d d ^ sipuids JdUnq S.31UOM qjiM pauiuuajap 'o qOoi ^^ ;bj jo yC}iABj3 ogpads OOioOOOO lO i/^OiOOOO o voou-> ^66666 6 666666 o odd 6 ^- •0 oooi IB ;bj 3h; ui spiDB X^ib; jo lunomv ^c oooCT-oo o o o'o^o o o o o o^ o o^o o 5 e t2 : : : ; : ig : :::;:;::: ; : . ; !'5 "^'-s^- '. '. '. .'S ! "J !*.-='' ".- .l!i??2u::r'ooct^„ti.-=l'B'r- -o i siiiii-^illiiHiii|ili 300C«^0^.--™0— CS«rt on I QJ J2 ■ — " cot: o; -t " *; MO.Q =t: J CN ^■^ be i* O - u Q rt ^ S a£l3 K en ■ 00 coX> CQCL r ro h P ?. 4^ oE O c ^ = ^ & ^bfl JrT3sirr3^(«w(UDH 2 "o •3 .' 2 S 11 ■- -"I Is J! 2 to .2 <*- (LI 496 ENGINEERING CHEMISTRY PAINT ANALYSIS. Paint is a liquid preparation having a two-fold use. Primarily it acts as a protecting coating against the action of the weather, and simultaneously as a decorative agent. The liquid is usually linseed oil and turpentine and the color- ing-matter or body some solid pigment, such as finely ground red oxide of iron. It is essential in the production of a good paint that the oil used should be one that, upon drying on the surface applied, should become hard, lustrous, and somewhat elastic. Unseed oil excels all others in use for this purpose, sophisti- cation thereto only deteriorates the quality. Four qualities are essential in paint: i. Durability; 2. Work- ing qualities ; 3. Drying properties ; 4. Covering power. The following list of pigments, with their chemical composition stated, will give an idea of the great variety that can be used in paints for outside work : The list would be largely increased were other pigments included that are used for interior decorative work only. Red Pigments. — Indian red, Tuscan red (Fe^Og), vermillion (HgS), red lead (Pb304), antimony vermilion (Sb^Ss). Iron oxide, Indian red, and Tuscan red can be analyzed by the scheme for hematite. Brown Pigments. — Umbers (Fe^Og, MnOg, etc.). Van Dyke brown (Fe^Og, carbon), manganese brown (Mng04) and sepia. The composition of sepia is as follows : Per cent. Melanin 78.00 CaCOs 10.40 MgCOa 7.00 Alkaline sulphates 2.16 Organic mucus 1.84 99.40 White Pigments.— White lead (2PbC03.PbHoOo), lead sul- phate (PbS04), zinc white (ZnO), sulphide of zinc, white (ZnS), "lithopone." Also the following, added oftentimes as I ENGINEERING CHEMISTRY 497 fillers: barytes (BaS04), hlanc fixe (artificial barytes), gypsum (CaS04), strontium white (SrS04), whiting (CaCOg), china-clay (kaolin), and magnesite (MgCOg). Yellow and Orange Pigments. — Chrome yellow (2PbCr04), Chinese yellow (PbO.PbCr04), zinc chrome ZnCr04), realgar (As.Sg), "cadmium yellow" (CdS), "King's yellow" (As.S.,), yellow ochre {Vtfi^, Al^Og, SiOs etc.), and Siennas (FcoOg, H,0,Mn304). Green Pigments. — Chrome green, (CrgO), copper green (CuA), mineral green (malachite), cobalt green (ZnO, CoO, P2O5, etc.), manganese green (BaO,Mn02, etc.), emerald green ("Paris green," 3CuOAs203.Cu(C2H302)2), and Brunswick green (compounded of barytes, chrome yellow, Prussian blue, etc., also called chrome green when lead sulphate is used instead of barytes ) . Black Pigments. — Lampblack (carbon), bone-black (carbon and Ca3HP04), vegetable black, Frankfort black, coal-tar black, asphaltum black, and graphite black (C). Blue Pigments. — Ultramarine (Si02,Al203>Na20,S), Prussian blue, Chinese blue, or Brunswick blue (FcgCigNig), cobalt blue or smalts (AI2O3, CoO), Bremen blue (CuHoOa), and copper blue (CuO,C02H20). The various colored lakes, carmines, analine lakes, etc., have but a limited application in engineering chemistry. Their methods of manufacture and assay can be advantageously studied by ref- erence to "Painters' Colors, Oils and Varnishes," by George H. Hurst, F. C. S., London, 1892, pp. 249-282. The analysis of white lead pure, not ground in oil (2PbC03). Pb.HoOs), can be performed as follows: Take i gram of the white lead, transfer to a No. 3 beaker, add 25 cc. dilute nitric acid and warm gently until solution is com- plete. Dilute sulphuric acid is added in slight excess and 10 cc. of alcohol. Allow to stand 20 minutes, filter off the lead sul- phate, wash well with water, dry, ignite, and weigh as PbS04. The weight of PbS04, multiplied by 0.85258, will give the weight of (2PbC03).(PbH20,2). 32 498 ENGINEERING CHEMISTRY If the white lead is ground in oil ("Paste White Lead" con- taining about 8 per cent, of oil), it will be necessary to extract the oil before the above determinations of lead can be made. The usual solvent is petroleum ether, Thompson^ prefers benzol C. P. Twenty grams of the paint are treated with sufficient pe- troleum ether, in a Soxhlet apparatus, and the oil extracted. By evaporation of the petroleum ether in a weighed beaker the amount of oil is determined. Hurst- treats the lead paint with strong nitric acid, whereby the oil is decomposed "into an insoluble greasy matter;" this does not interfere with the determination of the lead as sulphate. If, however, the oil is to be determined also, then he recommends the use of petroleum ether. After the extraction of the oil, either by benzol or petroleum ether, the residue is dried at 102° C, and weighed portions taken therefrom for the determination of the lead, as sulphate. Instead of determining the lead as sulphate, many chemists prefer to dissolve the dried lead paint in acetic acid to a clear solution, and precipitate the lead with bichromate of potash as lead chromate. This latter is weighed as PbCrO^ in a gooch crucible and the weight found calculated to (2PbC03). (PbH,0,). Thompson recommends the absorption method for carbon di- oxide determination. The water of hydration, in the dried paint, is found by calcu- lating the amount of lead sulphate or lead chromate into lead ox- ide and subtracting this amount of lead oxide, plus the amount of carbon dioxide found from 100. The difference is the water of hydration. The water of hydration cannot be determined accu- rately in paint from which the oil has been extracted since a small amount of organic matter remains which vitiates the result. The hygroscopic water is determined by drying a sample of the paint at 98° C. to constant weight. The loss represents moisture, provided no volatile oil is present. 1 J. Soc. Chem. Ind., 15, 432. - "Painters' Colors, Oils and Varnishes," p. 44. ENGINEERING CHEMISTRY 499 Analysis of White Lead Paints.^ (Dry, not ground in oil.) 123456 PbO 86.35 85.93 83.77 84.42 86.5 86.24 CO2 10.44 11.89 15.06 14.45 11.3 11.68 H2O 2.95 2.01 i.oi 1.36 2.2 1.61 Total 99.74 99.83 99.84 100.23 loo.o 99.53 from which the composition of the white leads can be calculated to be: 123456 PbCOa 63.35 72.15 91.21 87.42 68.36 70.87 PbHiOa 36.14 27.68 8.21 12.33 31.64 28.66 Moisture 0.25 — 0.42 0.48 — — Total 99.74 99.83 99.84 100.23 100.00 99.53 No. I. English make. Made by Dutch process; of very good qualit}^ No. 2. English make. Made by Dutch process ; of very good quality. No. 3. Kerms white. Made by precipitation with carbon dioxide. It is deficient in body, although of good color. No. 4. German make. Precipitated by carbon dioxide; of good color, but deficient in body. No. 5. German make. Made by Dutch process ; a good white. No. 6. German make. Made by precipitation with carbon dioxide ; quality fair. ANALYSIS OF WHITE PIGMENTS.^ Method for Very Small Amounts of Iron. Lead Pigments. — Treat sample with HNO3 (i • i) in usual manner, dilute with H^sO, add H0SO4 to precipitate bulk of lead (not necessary to evaporate down) ; cool, filter, wash w^th i to 2 per cent, of H,2S04, make filtrate just alkaline with NH4OH, then just acid with dilute HNO3, determine iron calorimetrically by the thiocynate method, using same amounts of reagents in preparing standards. If sample contains insoluble matter, filter out and ^ "Painters' Colors, Oils and Varnishes," by G. H. Hurst, F. C. S., p. 39. ^Proceedings, Am. Soc. Test. Mats., Vol. XIV (1914). 500 ENGINEERING CHEMISTRY wash with hot water till Pb-free, and to filtrate add HoSO^ and proceed as given. The insoluble is ignited, treated with HF and HoSO^ in usual manner, brought into solution (filter out any BaS04), and added to filtrate from PbS04. If necessary, solu- tion may be made up to volume 'and aliquots taken. Other Pigments. — Treat as above, omitting the addition of H.SO^. General Method. Co/or.— -Compare with sample selected as a standard. "Vol- ume" or "apparent gravity" to be determined by the Scott vol- umeter. True specific gravity to be determined by means of a pyknometer, using C. P. benzole or by Thompson's method.^ Fineness Test. — Determine with No. 21 silk bolting cloth or a 300-mesh bronze wire screen. The Thompson classifier^ may also be used, supplemented by a microscopic study. A qualitative analysis of all pigments should first be made. Basic Carbonate of Lead. Basic cabonate of lead (white lead) should approach the com- position 2PbC03.Pb(OH)^. Moisture. — Moisture may be determined by heating 2 grams for 2 hours in a steam- jacketed oven at atmospheric pressure. Total Lead. — If pure product is being examined, dissolve i gram in 20 cc. of nitric acid ( i : i ) in a covered beaker, heating till all CO2 is expelled ; wash off cover, dilute to about 120 cc. with hot w^ater and heat till all basic salt is in solution. Filter off any insoluble, wash with hot water till Pb-free, ignite and weigh "insoluble matter." (Insoluble, if appreciable, should be ex- amined for BaS04, SiO^, and silicates.) To filtrate add 20 cc. of H2SO4 (1:1) and evaporate to fumes of SO3, cool, add about 150 cc. of water and 150 cc. of ethyl alcohol; allow to stand cold 2 hours, filter on a gooch crucible, wash with 95 per cent, ethyl alcohol, dry i hour at 110° C, and weigh PbS04; calculate to PbO or to basic carbonate. Instead of determining the lead as sulphate the nitric acid solution may be made slightly alkaline with '^Proceedings, Am. Soc. Test. Mats., Vol. XIII, p. 407 (1913). ^Proceedings, Am. Soc. Test. Mats., Vol. X, p. 601 (1910). ENGINEERING CHEMISTRY 5OI NH4OH or NaOH, then acid with acetic acid, and to to 15 cc. of a 10 per cent, solution of potassium dichromate added; heat till the yellow precipitate assumes an orange color, Let settle and filter on a gooch crucible, washing by decantation with hot water till the washings are colorless, finally transferring all of the pre- cipitate. Finally wash with 95 per cent, ethyl alcohol and then ether dry at 110° C. for i hour and weigh PbCr04. Complete Analysis. — This method with a pure white lead gives good results for CO2 and H2O, but the residue is only an ap- proximation of the true PbO. In the absence of acetic acid or other organic matter, for example, unextracted vehicle, heat i gram in a porcelain boat in a current of dry CO,2 free air, ab- sorbing the water in H,2S04 and CaClg and the CO2 in soda lime or KOH solution (specific gravit)^ 1-27). By weighing the residue of PbO in the boat all the factors for determining the total com- position are obtained. Calculate the CO2 to PbCOg and the H2O to Pb(OH)2; excess H2O is due to moisture. Acetic Acid. — Thompson's method^ is as follows : Eighteen grams of the dry white lead are placed in a 500-cc. flask, this flask being arranged for connection with a steam supply, and also with an ordinary Liebig condenser. To this white lead is added, 40 cc. of sirupy phosphoric acid, 18 grams of zinc dust, and about 50 cc. of water. The flask containing the material is heated directly and distilled down to a small bulk. Then the steam is passed into the flask until it becomes about half full of condensed water, when the steam is shut off and the original flask heated directly and distilled down to the same bulk — this operation being conducted twice. To the distillate which is received in a larger flask is added i cc. of sirupy phosphoric acid to insure a slightly acid condition. The flask is then heated and distilled using a spray trap, to a small bulk say, 20 cc. Steam is then passed through the flask until it contains about 200 cc. of condensed water, when the steam is shut off and the flask heated directly. These operations of direct distillation and steam distillation are conducted until 10 cc. of the distillate require but a drop of N/io alkali to produce a change in the presence of phenolphthalein. 1 J. Soc. Chem. Ind., Vol. 24, p. 487 (1905). 502 e:nginee:ring chemistry Then the bulk of the distillate is titrated with N/io sodium hy- droxide, and the acetic acid calculated. It will be found very convenient in this titration, which amounts in some cases to 600 to 700 cc, to titrate the distillate when it reaches 200 cc, and so continue titrating every 20 cc. as it distills over. If the white lead contains appreciable amounts of chlorine it is well to add some silver phosphate to the second distillation flask and not carry the distillation from this flask too far at any time. Carbon Dioxide. — Determined by evolution with hydrochloric acid, weighing in soda lime, KOH solution or by absorbing in Ba(OH)2 solution and titrating or weighing the BaCOg.^ Total Sulphuric Anhydride (in absence of BaS04) : — Deter- mined by dissolving in HCl and NH4CI, precipitating with NagCOg solution in excess, filtering, acidifying filtrate with HCl and pre- cipitating as BaS04. (In presence of BaS04). Determined as under basic sulphate of lead containing BaS04. Sidphur Dioxide. — Weigh 2 grams into a 250 cc. beaker, add 100 cc. of distilled water that has been freshly boiled and cooled, then 5 cc. of concentrated HCl; stir thoroughly, let stand 15 minutes, and titrate with o.oi normal iodine solution, using starch as indicator. Blank should be run on reagents and correction made. Metallic Lead. — Weigh 50 grams of the sample into a 400 cc. beaker, add a little water and add slowly 60 cc. of 40 per cent, acetic acid and after effervescence has ceased boil on hot plate. Fill the beaker with water, allow to settle, and decant clear solu- tion. Add 100 cc. of a mixture of 360 cc. of strong NH4OH, 1,080 cc. of water, 2,160 cc. of 80 per cent, acetic acid and boil until all solution is complete. Fill beaker with water allow to settle and decant clear solution. Collect residue on watch crystal, floating off everything but metallic lead. Dry and weigh. Re- sult X 2 = percentage of metallic lead in sample. Note. — If soluble barium compounds, as for example, BaCOg, 1 See J. M. Camp's method for carbon in steel in Phillips, " Methods of Analysis in Pittsburgh District; " Dudley & Voorhees' method in Scott, " White Paints and Painting Materials," p. 84; and an article by Wysor, in Chemical Etigineer, Vol. 11, p. 26. ^NGINDEIRING CHEMISTRY 503 are present, the lead and barium are separated together as sul- phates, the precipitate of BaSO^ + PbSO^ treated with hot acid ammonium acetate solution, and the lead determined in the solu- tion by the sulphate or chromate method. The BaSO^ is weighed as such. If sample contains much calcium or magnesium, deter- mine lead by chromate method, or separate the lead by a hydrogen- sulphide precipitation, dissolve PbS in hot dilute HNO3 and deter- mine lead as PbSO^ or PbCrO^. Iron, aluminum, zinc, calcium and magnesium may be determined in filtrate from PbS by usual methods. Basic Sui^phate; op I^ead. Moisture. — Heat 2 grams of the sample 2 hours in an air bath at 105° C. Insoluble Matter. — Treat i gram of sample in a 600 cc. beaker with 20 cc. of water, 20 cc. of concentrated HCl and 10 grams of NH4CI ; cover and heat about 10 minutes, then add about 400 cc. of water and boil 10 minutes. Filter and wash thoroughly with hot water. Ignite and weigh insoluble matter. If sample con- tains soluble silica, treat with HCl and HgO and evaporate to dryness, then as above with HgO, HCl and NH4CI, finally diluting and boiling. Total Soluble Sulphates {in the Absence of BaSO^). — Treat 0.5 gram of the sample with 5 cc. of water, 3 grams of NH4CI and 5 cc. of HCl saturated with bromine; digest (covered) on steam bath about 15 minutes, add 25 cc. of H2O, neutralize with dry NaaCOg and add about 2 grams more, boil 10 to 15 minutes; let settle, dilute with hot water, filter and wash with hot water ; redis- solve in HCl, reprecipitate as above and wash thoroughly with hot water. Acidify united filtrates with HCl, adding a slight excess ; boil and add slight excess of 10 per cent. BaCl^ solution. Let stand on steam bath for i hour, filter, wash with hot water, ignite and weigh BaS04. Calculate to SO3 (includes SO3 formed from SO2). Total Soluble Sulphate (in the Presence of BaSO^). — Treat i gram in a 600 cc. beaker with 10 cc. of H2O, 10 cc. of strong HCl, saturated with bromine, and 5 grams of NH4CI heat on a 504 ENGINEERING CHEMISTRY steam bath in a covered beaker for 5 minutes, add hot water to make about 400 cc, boil for 5 minutes, and filter to separate any insoluble material. (A pure pigment should be completely dis- solved.) Wash with hot water, ignite and weigh the insoluble. Remove lead with NagCOs as above, making a double precipita- tion, acidify, and to the boiling hot filtrate add slowly, with stir- ring, 20 cc. of a 10 per cent. BaCU solution; let stand for 2 hours on the steam bath, filter, wash, ignite, and weigh as BaSO^ (in- cludes SO3 formed from SO2). If sample contains much calcium this precipitate, after ignition, should be treated as under ''gypsum." Soluble Zinc Sulphate. — Boil 2 grams of the sample with 150 cc. of water and 50 cc. of alcohol for 30 minutes, filter, and wash with a mixture of alcohol and water (i : 3). Heat filtrate to boiling- and expel most of the alcohol ; then determine SO3 by usual method of precipitation with BaCls- Calculate to ZnSO^ and so SO3. Total Lead and Zinc {in the Absence of Calcium and Mag- nesium). — Insoluble matter and soluble 810,2, if present, should be removed before adding H2SO4. Dissolve i gram by boiling 15 minutes with 250 cc. of water and 20 cc. of concentrated HNO3, add 5 cc. of concentrated H2SO4, and evaporate to copious fumes of SO3 ; cool, add 250 cc. of water, let stand cold 2 hours, filter on gooch crucible, wash with i per cent. H0SO4, ignite and weigh as PbS04. Iron and aluminum, if present, should be removed before pre- cipitating zinc. If Ca and Mg are also present, see method below. To determine small amounts of Fe, Al and Mn (in absence of Ca or Mg), a large portion of sample should be treated as above, the Pb removed as PbS04, Fe and Al precipitated with NH4OH (redissolving and reprecipitating) ; ignite and weigh Al^Og -f- Feo03. This precipitate may be fused with KHSO4 and Fe de- termined volumetrically, if desired. In filtrate from Al and Fe, Mn, if present, may be determined by precipitating with NH4OH and bromine, finally weighing as Mn304. Make filtrate up to volume and determine Zn in an aliquot as ZuoPgO;, as ZnO or volumetrically with K4Fe(CN)6. ENGINEERING CHEMISTRY 505 Evaporate the filtrate to about lOO cc, cool, add 5 grams of microcosmic salt dissolved in water, then add NH4OH until the solution is just neutral to litmus paper. Add 2 drops of NH4OH and I cc. of acetic acid, stir vigorously, heat on steam bath for i hour (the precipitate should assume a crystalline character and settle well). Filter on a gooch crucible, wash with hot water, ignite at first at a low temperature and finally to redness, cool, and weigh as zinc pyrophosphate. Calculate to ZnO. Total Lead and Zinc {in the Presence of Calcium and Mag- nesium). — With a sample containing calcium or magnesium salts the lead should be precipitated as sulphide from a slightly acid (HCl) solution, the PbS dissolved in hot dilute HNO3 and the lead determined as sulphate. Filtrates from the PbS are boiled to expel H2S, a little bromine water added to oxidize iron (if present), boil to expel bromine, and then add NH4OH in slight excess. Filter and wash precipitate of Fe(OH)3 + Al(OH)3 with hot water. (If appreciable, redissolve in hot dilute HCl and reprecipitate with NH4OH, ignite and weigh Fe^Og + AIqOs-) Manganese, if present, can be precipitated by adding bromine and NH4OH and warming; filter, wash with hot water, ignite and weigh as Mn304. Unite all of the filtrates, make slightly acid with acetic acid, heat to boiling and pass H2S into the hot solution till saturated (20 to 30 minutes) ; add 5 grams of NH4CI and let stand 5 hours ; filter, wash with hydrogen sul- phide water, dissolve the ZnS in hot dilute HCl, boil off the HgS, filter out any separated sulphur and determine the zinc as ZngP^O^, as described. Calcium may be determined in the filtrate from the ZnS by expelling H2S and then adding NH4OH and ammon- ium oxalate in usual manner. Titrate with KMnO^. In the filtrate from calcium determine magnesium in usual manner by precipitating with sodium phosphate solution, finally weighing as Mg2P207. When calcium and magnesium are present zinc is best determined volumetrically by Low's^ ferrocyanide method. In the absence of iron and manganese, take the filtrate from the PbS, make alkaline with NH4OH, then just acid with HCl; add 3 cc. of concentrated HCl, dilute to 250 cc, heat and titrate just 1 "Technical MeUiodsof Ore Analysis," p. 209 (1906). 506 ENGINEERING CHEMISTRY as in standardizing the solution. When iron and manganese are present for oxidized ores as described by Low. Lead may also be determined by boiling i gram of the sample in 50 cc. of water plus 100 cc. of a mixture of 125 cc. of 80 per cent, acetic acid, 95 cc. of strong NH4OH and 100 cc. of water, diluting to about 200 cc, filtering out any insoluble, washing with above mixture and then precipitating with K^CrgOy, finally weighing as PbCr04. Zinc may be determined by boiling i gram of the sample with 30 cc. of water, 4 grams of NH4CI and 6 cc. of concentrated HCl ; dilute to 200 cc. with hot water, add 2 cc. of saturated sodium thiosulphate solution and titrate in usual man- ner with ferrocyanide. Sulphur Dioxide. — Digest 2 grams of the sample with frequent stirring in 100 cc. of freshly boiled cold water and 5 cc. of con- centrated HCl; let stand 10 to 15 minutes, add an excess of o.oi normal iodine solution and titrate back with o.oi normal sodium thiosulphate solution, using starch indicator. Report as SO^. Run blank on reagents and make corrections. Carbon Dioxide. — Determined as under basic carbonate of lead, noting precautions for sulphides, etc., under lithopone. Calculations. — Report soluble SO3 as ZnS04; deduct ZnO equivalent of the ZnS04 from total ZnO and report residue as ZnO. Deduct soluble SO3 and SO3 equivalent to SOg from total SO3 calculate remainder to PbS04; subtract PbO equivalent of PbS04 from total PbO and report remainder as PbO. Zinc-Le)ad and Leaded Zincs. Zinc-lead and leaded zincs (Ozlo white) are to be analyzed by methods given under "basic sulphate of lead." Zinc White. Moisture. — Heat 2 grams in air bath at 105° C. for 2 hours. Loss on Ignition. — Ignite i gram over Bunsen burner for 15 minutes. Soluble zinc sulphate, total sulphate, insoluble matter, CO2, lead, zinc, iron, aluminum, SOo, calcium and magnesium are to be treated as under "basic sulphate of lead." ENGINEERING CHEMISTRY 507 I^ITHOPONE. Lithopone (Ponolith, Jersey Lily White, Becton White, Charl- ton White, Orr's White) should contain about 69 to 70 per cent, of barium sulphate, the remainder being zinc sulphide with small amounts of zinc oxide and carbonate. Analysis of Pure Lithopone} Moisture. — Heat 2 grams of the sample for 2 hours at 105° C. There should be less than 0.4 per cent, of moisture. Insoluble and Total Zinc. — ^Take i gram in a 200 cc. beaker, add 10 cc. of strong hydrochloric acid, mix, and add in small portions about I gram of potassium chlorate; then heat on the water bath until about half of the liquid is evaporated. Dilute with hot water, add 5 cc. of dilute sulphuric acid (i : 10) ; boil, allow to settle, filter, wash, ignite and weigh the insoluble which will be total barium as barium sulphate together with any other insoluble. Make a qualitative examination for alumina and silica (not likely to be present). Heat the filtrate from the insoluble to boiling, add sodium carbonate solution, drop by drop, until all of the zinc is precipitated as carbonate, filter on a gooch crucible, wash, ignite and weigh as zinc oxide. Zinc Sulphide.'^ — Digest i gram with 100 cc. of i per cent, acetic acid at room temperature for ^ hour, then filter and wash ; determine the zinc in the filtrate as in the preceding analysis. The difference between the total zinc oxide and the zinc oxide soluble in acetic acid multiplied by 1. 19749 gives the zinc present as sul- phide. The zinc soluble in acetic acid may be reported as oxide, though it may be partly carbonate. This method of analysis as- sumes the absence of impurities such as salts of iron. Analysis of Lithopone in the Presence of Foreign Substances.^ Soluble Salts. — Wash 2 grams with hot water and determine the nature of the soluble salts. Moisture. — Determine on 2 grams the loss in weight at 100 to 105° C. 1 Method of P. Drawe, Zeitschriflfur angew. Chetnie, Vol. 15, p. 174 (1902). 2 Scctts Evolution Method 1 following) may te advantageously used. 3 Method of Copalle, Ann. chim. anal, appl.. Vol. 12, p. 62 {1907). 5o8 ENGINEERING CHEMISTRY Insoluble. — Oxidize i gram with nitric acid of 40° Baume (spe- cific gravity 1.38), at first cold, then hot. Then add hydrochloric acid, evaporate to very small volume, dilute with hot water, filter, ignite the precipitate which represents the barium sulphate, cor- responding to the total barium. If the insoluble exceeds 66 to 68 per cent, it is necessary to prove that the excess is not due to the addition of kaolin. Total Zinc. — Determine as oxide by precipitation as carbonate in the filtrate from the insoluble. When more than traces of iron, alumina, or lime are present, it is best to determine the zinc volumetrically. Sulphide of Zinc. — Add a slight excess of hydrochloric acid to the filtrate from the zinc carbonate and determine the sulphur by precipitation in the usual manner. This sulphur multiplied by 3.0383, or the weight of barium sulphate (BaSO^) multiplied by 0.41 741, gives the zinc sulphide. Oxide of Zinc. — Multiply the weight of the zinc sulphide by 0.83507 to obtain the zinc oxide corresponding to the sulphide. Subtract this from the total zinc' oxide and report the remainder as zinc oxide (it may be present as oxide or as carbonate). Barium Carbonate. — Digest 2 grams with boiling dilute hydro- chloric acid, dilute with hot water, filter from the insoluble and determine the barium in the filtrate by precipitation with sulphuric acid. The weight of the barium sulphate multiplied by 0.84548 gives the barium soluble in the acid calculated as carbonate. Barium Sulphate. — Subtract the barium sulphate corresponding to the carbonate from the total barium sulphate. Sulphide may be determined directly by Scott's^ evolution method, using 0.5 to i gram of pigment, mixing in evolution flask with zinc and water, running in HCl from separatory funnel and absorbing the HoS in alkaline lead-nitrate solution. Filter off the PbS, dissolve in hot dilute HNO3 and determine the lead as PbSO, or PbCrO,. Calculate to ZnS (PbS04Xo.32i7). Carbon Dioxide. — Carbon dioxide may be determined directly by evolution method by grinding i gram of sample with excess 1 Scott, " White Paints and Painting Materials," p. 257. ENGINEERING CHEMISTRY 509 of potassium dichromate (dry salt) ; transfer to flask, add 50 cc. of water and run in H2SO4 (1:1) from separatory funnel, ab- sorbing CO2 in KOH, soda lime or Ba(OH)2 solution. A tube containing KMnO^ solution or acidified CuSO^ solution may be placed in train as a precaution. Cai^cium Pigments. Whiting, Paris White, Spanish White, and Chalk. Whiting, Paris white, Spanish white, and chalk are the natural and artificial forms of calcium carbonate. Moisture. — Heat 2 grams of sample in an air-bath at 105° C. for 2 hours. Loss in wight is considered as moisture. Loss of Ignition. — Ignite i gram over blast lamp to constant weight. Complete Analysis. — Boil 2 grams of the sample in a covered vessel with 30 cc. of HCl (1:1) and a few drops of HNO3 ; wash off cover and evaporate to dryness, take up with a little HCl and about 100 cc. of hot water; boil, filter, wash with hot water, ignite and weigh insoluble matter. Insoluble should consist of silicious matter. Test insoluble for BaSO^. Heat filtrate from insoluble to boiling, having added more HCl in order to form sufficient NH4CI to hold magnesia in solution, and add NH4OH in very slight excess, heat a few minutes, filter, wash with hot water, ignite and weigh Al203+Fe203(-f-Ti02+P205). It is best to re- dissolve this precipitate in hot dilute HCl and reprecipitate with NH4OH. (If manganese is present, it may be precipitated in the united filtrates from Al and Fe by H2S and NH4OH.) Unite the filtrates and make up to a definite volume, mix and take an aliquot corresponding to 0.5 gram of sample; dilute if necessary, heat to boiling and add slowly 30 cc. of a saturated ammonium- oxalate solution, let stand on steam bath i to 2 hours; filter, re- dissolve precipitate in dilute HCl, dilute, add 10 cc. of ammonium- oxalate solution and NH4OH till alkaline, let stand on steam bath I or 2 hours ; filter, wash with hot water till free from chlorides. The precipitate may be ignited to constant weight in a platinum crucible over a Meker burner and the CaO weighed as such, or the CaO may be determined volumetrically as follows : 5IO ENGINEERING CHEMISTRY The precipitate of calcium oxalate must be washed till lo cc. of the washings plus 0.5 cc of H^SO^ heated to 70° C. do not de- colorize I drop of about N/io KMn04 solution. The beaker in which precipitation was made is placed under funnel, apex of filter is pierced with stirring rod and the precipitate washed into beaker; then pour hot dilute HgSO^ (1:4) over paper, wash with hot water, add about 30 cc. of the dilute H^gSO^ (1:4), dilute to about 250 cc, heat to 80 to 90° C, and titrate with about N/io KMnO^ (Fe value of KMnO^ X 0.50206 — CaO value). Evap- orate the united filtrates from the calcium oxalate to about 200 cc. — should any magnesium oxalate separate, dissolve it by adding a little HCl — add 5 cc. of NH4OH, heat to boiling and add 10 to 15 cc. of saturated Na^aHPO^ solution. Add a few cubic centi- meters more of NH^OH, cool in ice water with vigorous stirring. Let stand 2 to 4 hours, filter on a gooch crucible, wash with 2 per cent. NH4OH containing a little NH4NO3; ignite gently at first and finally at a bright red for 5 or 10 minutes, cool and weigh as magnesium pyrophosphate. Calculate to MgO. If magnesium is high, or for very accurate work, the NH4MgP04 should be redissolved in dilute HCl and the Mg reprecipitated as above. If MgO is very low, it may be necessary to destroy the ammonium salts in the filtrate from the calcium oxalate before precipitating the MgO. This may be effected by evaporating to dryness with excess of HNO3, taking up with HCl and water, filtering and proceeding as above. Carbon Dioxide. — Determined by evolution method as under basic carbonate of lead. Total Soluble Sulphates. — Determined as under gypsum. Alkalinity. — Alkalinity is due to free lime or possibly to sodium or potassium compounds. Boil 2 grams of the sample for 5 minutes with 100 cc. of water, filter, add phenolphthalein. If a red color develops, free lime may be assumed to be present. Titrate with N/io acid. Gypsum, Terra Alba, Plaster of Paris. Gypsum is a natural hydrated calcium sulphate, CaS04.2H20; terra alba is a fairly pure grade of raw gypsum; plaster of Paris e;ngine;e:ring chemistry 511 is a calcined or dehydrated calcium sulphate — CaS04^H20. There is also a precipitated calcium sulphate used as a basis for aniline lakes, A microscopic examination may be of importance. Moisture. — Dry 2 grams in vacuum desiccator over HgSO^ to constant weight. Combined Water and Moisture. — Heat i gram of the sample in a covered porcelain crucible on an asbestos plate for 15 minutes, then heat bottom of crucible dull red for 10 minutes over a Bun- sen burner, remove cover and heat for 30 to 40 minutes at a slightly lower temperature. Cool and weigh rapidly. Repeat to constant weight. Combined water and moisture may also be determined by heat- ing in air bath at 200° C. to constant weight. Soluble and Insoluble. — Boil 2 to 3 grams of the sample with 20 cc. of concentrated HCl, a few drops of HNO3, ^^^ about 50 cc. of water; evaporate to dryness, boil residue repeatedly with 10 per cent. HCl; filter, wash with hot water, ignite and weigh insoluble matter. Test for BaS04 — make filtrate up to 500 cc. and mix. To 200 cc. add about 2 grams of NH4CI and NH4OH till slightly alkaline, heat till only faint odor of NH4OH remains, let settle, filter, wash with hot water, ignite and weigh AloOg+Fe^Og. Heat filtrate from Al and Fe to boiling and add about 40 cc. of saturated ammonium oxalate solution, let stand on steam bath 2 hours; filter, redissolve precipitate in hot dilute HCl, add 10 cc. of ammonium oxalate solution and NH4OH till alkaline, let stand i hour on steam bath ; filter, wash with hot water till free from soluble oxalates (see test under whiting), and proceed as under whiting, titrating with KMn04. The united filtrates from the lime are evaporated and MgO deter- mined as under whiting. Soluble Sidphate. — Make 200 cc. of the filtrate from the in- soluble slightly alkaline with NH4OH, then acid with HCl, heat to boiling and add 20 cc. of hot 10 per cent. BaCl2 solution, stir well. Let stand at least i hour on steam bath; filter, wash with hot water till washings give no test for CI with AgNOg, ignite. 512 e:nginee)ring chemistry cool and weigh BavSO^. For very accurate work, the weighed BaSO^ should be purified by treating with dilute HCl, filtering, washing, igniting and again weighing. Carbon Dioxide. — Determined by evolution, weighing in soda lime, KOH bulb or as BaCOg. Quicklime and Slaked or Hydrated Lime. Quicklime (CaO), and slaked or hydrated lime (Ca(OH)2), are used in the preparation of cold water paints, for example, whitewash. These materials may be examined as under whiting. Strontia white, SrS04, and strontianite, SrCOg, occur only in small quantities and are rarely met with in paint analysis. In the usual methods^ of analysis any strontium present is weighed with the CaO or reported as BaSO^ when insoluble. Barium Pigments. Barytes or Barite. Barytes or barite is a natural sulphate of barium ; blanc fixe is precipitated barium sulphate. Being one of the cheapest white pigments, this material is seldom adulterated. It should be white, well-ground and contain not less than 95 per cent, of BaS04. A microscopic examination can be made with advantage to determine uniformity of grinding, size and angularity of par- ticles, amorphous or crystaline. Miscibility, opacity, specific gravity, volume, whiteness of color, together with microscopic study probably give more information than chemical analysis. However, the following method may be used : Moisture. — This equals the loss in weight on heating 2 grams of the sample at 105° C. for 2 hours. Loss on Ignition. — Ignite i gram of sample for 30 minutes (to constant weight). Loss may be due to organic matter, free and combined water and CO,2. Report as ''loss on ignition." Insoluble. — Boil i gram with HCl (i 13), evaporate to dry- ness, moisten with HCl, add water, boil, filter, wash with hot water, ignite in platinum crucible if previous qualitative exam- 1 For methods see Bulletin No. 422, U. S Geological Survey ; Treadwell-Hall. Analytical Chemistry," etc. DNGINEJERING CHKMISTRY 513 ination has determined the absence of lead or other easily reduced metals. Weigh insoluble and treat with HgSO^ and hydrofluoric acids in usual manner, evaporate, ignite and weigh, loss in silica ; residue should be BaS04. The residue may be fused with Na2C03, taken up with hot water, acidified with HCl, the BaS04 filtered off, washed, and ignited. If weight so obtained differs materially from that of residue from hydrofluoric acid treat- ment, examine last filtrate for Al, Fe, Ca and Mg that may have remained as residue from silicates. Alumina, Iron Oxide, etc. — Add NH^OH to the filtrate from the total insoluble, boil, filter, wash, ignite and weigh as FC2O3 + AI2O3. In filtrate determine Ca and Mg as in gypsum. Soluble Sulphate. — Boil i gram with 20 cc. of concentrated HCl, dilute to 200 cc. with hot water, boil, filter, wash, add NH4OH to filtrate till just alkaline, make just acid with HCl, boil, add 10 per cent. BaCl^ solution and weigh BaSO^ in usual manner. Calculate to CaSO^. If carbonates are present, cal- culate the remaining CaO to CaCOg. Any excess of CaO is re- ported as CaO. Carbon Dioxide. — Determine by evolution method as given under basic carbonate of lead. Barium Carbonate. — If present, it may be precipitated in first filtrate before determining Al, Fe, etc., by adding 10 per cent, ammonium sulphate solution containing a little free H2SO4, fin- ally weighing in usual manner as BaS04. Any excess of CO2 over the barium here found is calculated to CaCOg. Iron. — If in \Qry small amount, determine color imetrically as given under ''Method for Very Small Amounts of Iron." Water Soluble. — This test is sometimes applied to blanc fixe. Boil 5 grams for 15 minutes with 100 cc. of water, filter and wash. Evaporate filtrate to dryness in a weighed dish, dry 30 minutes at 105° C, cool and weigh. Test for sodium, chlorine, CaS04, etc. Witherite (BaCO^). — This may be examined by preceding methods. 33 514 e;ngine:ering chemistry SiiviCA Pigments. Silica or Silex. Silica or silex (SiOo) should be finely ground and white. Moisture. — This equals the loss in weight on heating 2 grams at 105° C. for 2 hours. Loss on Ignition. — Ignite i gram to constant weight in a plati- num crucible. Insoluble Matter. — Boil 2 grams of the sample for 30 minutes with 50 cc. of HCl (i : i), add 50 cc. of water, filter, wash, ignite and weigh insoluble matter, which should not be less than 95 per cent. This insoluble matter is treated with H2SO4 and HF in usual manner, loss being considered as silica, SiOsi the residue is fused with NaaCOg, taken up with water and HCl, evaporated to dryness, any SiO,2 (test for BaS04) filtered out and Al, Fe, Ca, and Mg determined as in gypsum. The filtrate from the in- soluble matter (that is, the soluble portion) is evaporated to dry- ness, taken up with HCl and water, SiOg filtered out, ignited and weighed as usual. In filtrate determine Al, Fe, Ca, and Mg as usual. If it is desired to determine alkalies, work on a separate portion by the method of Mr. J. Lawrence Smith as in Bulletin No. 422, U. S. Geological Survey. Iron in Small Amounts. — See "Method for Very Small Amounts of Iron." China Clay and Asbestine. Moisture. — Determined as under silica. Loss of Ignition. — Determined as under silica. Qualitative tests to prove that the materials are as represented will generally suffice. However, a complete analysis may be made as follows : Fuse I gram of the finely powdered sample in a platinum cru- cible with about 10 grams of NasCOg (requires }^ to i hour) ; cool, place in casserole, digest with hot water till mass disinte- grates; acidify with HCl, remove crucible and lid, washing thor- oughly. Evaporate to dryness on steam bath, take up with HCl and hot water, filter, wash with hot water till free from CI ; evap- ENGINEERING CHEMISTRY 515 orate filtrate to dryness and treat as before, filtering on a separate paper. Burn the two silica precipitates together in a platinum crucible, finally heating over Meker burner to constant weight; treat with H^SO^ and HF in usual manner, loss equals SiOo. If sample contains BaS04, melt from fusion should be digested in hot water till completely disintegrated, the BaCOs filtered off and washed with hot water. The BaC03 ^^id residue are dissolved in hot dilute HCl, the Ba precipitated with dilute H,2S04, and the BaS04 determined in usual manner. Filtrate from this BaS04 is added to first filtrate, acidified, evaporated for silica, etc., as described. The residue from SiOa is considered as Al^Og and Fe20o, the Al and Fe subsequently obtained being ignited in same crucible. In filtrate from SiOa make a double precipitation of Al and Fe with NH4OH (having sufficient NH4CI present to hold all MgO in solution), ignite and weigh AI2O3 -|- Fe^Og (TiOo + P^Og). This precipitate may be fused with KHSO4, dissolve in dilute H2SO4, the iron reduced (H2S followed by CO2) and titrated with KMNO4. In united filtrates from Al and Fe, manganese may be precipitated with H^S and NH4OH and weighed in usual way. Expel HoS and determine CaO and MgO as usual. Determine alkalies on a separate portion by the method of Mr. J. Lawrence Smith. Carbon Dioxide. — Determined by evolution method, weighing. Soluble Sulphates. — Boil i gram with 20 cc. of HCl (1:1) and 100 cc. of water, filter, wash. Add NH4OH till just alkaline, HCl till acid and precipitate with BaClg in usual manner. Asbestine is often treated with HCl as under silica, the soluble and insoluble portions being analyzed separately. Analysis of Red Lead.^ Approximate formula, Pb304 (probably PbO,2.2PbO). Apparent gravity and true specific gravity determined as per methods under white pigments. Fineness. — Wash 10 grams with water through No. 21 silk bolting cloth, dry and weigh residue. 1 This includes orange mineral. 5l6 ENGINEERING CHEMISTRY Moisture. — Dry 2 grams of the sample for 2 hours at 105° C. The loss in weight is considered as moisture. Organic Color. — Boil 2 grams of the sample with 25 cc. of 95 per cent, ethyl alcohol, let settle, decant off the supernatant liquid ; boil residue with water, decant as before and boil residue with very dilute NH4OH. If either the alcohol, water or NH^OH is colored, organic coloring matter is indicated. Total Lead and Insoluble Matter. — Treat i gram of the sample with 15 cc. of HNO3 (1:1) and sufficient hydrogen dioxide to dissolve all Pb02 on warming. If any insoluble matter is present add 25 cc. of water, boil, filter and wash with hot water. In- soluble contains free SiO^ and should be examined for BaS04 and silicates, if appreciable. To original solution or filtrate from insoluble, add 20 cc. of concentrated H2SO4 and evaporate to SO3 fumes; cool, add 150 cc. of water and 150 cc. of 95 per cent, ethyl alcohol, let stand cold 2 hours, filter on a gooch crucible, wash with 95 per cent, alcohol, dry at 105 to 110° C. and weigh as PbSO^. Calculate to PbO. Red lead is rarely adulterated, but should sample contain soluble barium compounds, the PbSO^ ob- tained above will contain BaSO^. In this case, digest above pre- cipitate with acid ammonium acetate solution, filter off BaSOi, wash, ignite and weigh BaS04. Calculate to BaO or BaCOg. In filtrate, determine the lead as PbSO^ or PbCr04. If sample contains significant amounts of calcium or magnesium, the HN03~H202 solution is boiled till all lead is converted into ni- trate and then lead determined as PbCr04. If Ca and Mg are to be determined, separate lead as PbS and proceed as under basic sulphate of lead in presence of these metals. Determination of Lead Peroxide {PbO 2) and True Red Lead (F^304.)— (Method of DiehP modified by Topf^— not applicable when substances are present, other than oxides of lead, that lib- erate iodine under conditions given.) Weigh I gram of finely ground sample into a 200-cc. Erlenmeyer flask, add a few drops of distilled water and rub the mixture to a smooth paste with a glass rod flattened on end. Mix in a small 2 Ding. Polyt. Jour., Vol. 246, p. 196. 3 Zeitschrifl fur analysche Chemie, Vol. 26, p. 296. ENGINEERING CHEMISTRY 517 beaker 30 grams of C. P. ''Tested Purity" crystallized sodium acetate, 2.4 grams of C. P. potassium iodide, 10 cc. of water and 10 cc. of 50 per cent, acetic acid ; stir until all is liquid, warming gently; if necessary add 2 to 3 cc. of H^^O, cool to room tempera- ture and pour into the flask containing the red lead. Rub with the glass rod until nearly all the red lead has been dissolved; add 30 cc. of water containing 5 or 6 grams of sodium acetate, and titrate at once with decinormal sodium thiosulphate, adding the latter rather slowly and keeping the liquid constantly in motion by whirling the flask. When the solution has become light yellow, rub any undissolved particles up with the rod until free iodine no longer forms, wash off rod, add the sodium thiosulphate solution until pale yellow, add starch solution and titrate until colorless, add decinormal iodine solution until blue color is just restored and subtract the amount used from the volume of the thiosulphate that has been added. Calculation. — The iodine value of the sodium thiosulphate solu- tion multiplied by 0.94193 = PbO^ ; the iodine value multiplied by 2.69973 = Pb304 ; the PbOg value multiplied by 2.86616 = Pb304. The Sodium Thiosulphate Solution (Decinormal). — Dissolve 24.83 grams of C. P. sodium thiosulphate, freshly pulverized and dried between filter paper, and dilute with water to i liter at the temperature at which the titrations are to be made. Solution best made with well-boiled HgO free from COg, or let stand 8 to 14 days before standardizing. Standardize with pure, resublimed iodine, as described in Treadwell-Hall, Analytical Chemistry, Vol. II, p. 602 (1910), and also against pure potassium iodate — the two methods of standardization should agree within o.i per cent, on iodine value. Starch Solution. — Two to 3 grams of potato starch are stirred up with 100 cc. of I per cent, salicylic acid solution, and the mix- ture is boiled till starch is practically dissolved, then diluted to I liter.i 1 Lead Peroxide. — If sample contains no appreciable amount of nitrate (nitrate has no effect on method), leach out water soluble matter as below, dry residue and determine PbO.2 as above, calculating to basis of original sample. 5l8 ENGINEERING CHEMISTRY Zinc. — If an appreciable amount, determine in filtrate from total lead as per methods under zinc white, evaporating off the alcohol. Water Soluble. — Digest lo grams of sample with 200 cc. of hot water on steam bath for i hour; filter on an 11 -centimeter S. & S. blue ribbon paper and wash with hot water till no residue is left on evaporating a few drops of the washings. Evaporate filtrate to dryness on steam bath in a weighed dish, dry 30 minutes at 105° C, cool and weigh. Take up with water and if alkaline, titrate with o.i normal acid and methyl orange; calculate to Na^COg. Another lot of water soluble matter is tested for ni- trates, nitrites, carbonates, sulphates, sodium and lead. Total Silica. — Digest 5 grams of the sample in a covered cas- serole with 5 cc. of HCl and 15 cc. of HNO3 (i • i)- Evaporate to dryness to dehydrate. Cool, treat with hot water and HNO3, boil, filter, wash with hot acid ammonium acetate solution, then dilute HCl and finally hot water. Ignite and weigh as Si02. The residue may be treated with H0SO4 and HF in cases of doubt as to purity. Carbon Dioxide. — Determined by evolution method, using dilute HCl and stannous chloride. Soluble Sulphate. — Determined as under basic sulphate of lead. Iron 0.^i(/^.-— Determined by Schaeffer's modification of Thom- son's calorimetric method ; or, in a large beaker, treat 20 grams of the sample with 20 cc. of water, 20 cc. of HNO3 (specific gravity 1.4) and 3 cc. of formaldehyde solution. Warm till all PbOg is dissolved, dilute with water, warm, filter off insoluble and wash with hot water. Ignite filter and insoluble, evaporate with H2SO4 and hydrofluoic acid. To filtrate from insoluble add 14 cc. of H2SO4 (i : i), filter off PbSO^, wash. Residue from HF and H2SO4 is dissolved in HaSO^ and added to filtrate from PbS04 ; dilute to 500 cc. and determine Fe colorimetrically in an aliquot, using same amounts of HNO3, HoSO^ and formaldehyde in comparison solution. Calculate to FcgOs. ENGINEERING CHEMISTRY 519 Specifications for White Lead Issued by the Navy Department, March 1, 1915. Generai,. 1. \A'hite lead shall be furnished, dry or in oil, as specified, and shall conform to the following requirements : QUAUTY. 2. To be as follows : (a) White Lead, Dry. — The pigment shall be pure hydrated carbonate of lead, free from all adulterants. The total acetate shall not be in excess of the equivalent of 0.15 per cent, of absolute acetate acid. (b) White Lead, in Oil. — To be of the same quality as white lead dry, and be finely ground in at least 8.50 per cent., by weight, of pure raw linseed oil in accordance with the latest issue of Navy Department Speci- fications for Raw Linseed Oil. The material shall not contain more than 0.50 per cent, of moisture. Comparison with Standard Sampee. 3. White lead, dry and in oil, shall be free from crystalline structure and be equal in whiteness, fineness, opacity or body, tinting strength, and covering quality to the standard sample of white lead, samples of which may be obtained by application to the construction officer, navy yard, New York, N. Y. Tinting Test. 4. The tinting strength required in paragraph 3 will be compared to the standard sample of dry white lead as follows : Ten grams of dry white lead will be thoroughly ground wuth 10 milligrams of dry lampblack and a sufficient weight of raw linseed oil to reduce the lead to a paste form, and compared with equal amounts of standard white lead, dry lampblack, and linseed oil ground in the same manner. Where no means are at hand for weighing in grams and milligrams, larger amounts may be used in the same proportion as indicated above. When treated as above and placed alongside of the standard sample on a glass slide, the tint of the lead under test shall not be darker than that of the standard sample. Note. — In case of samples of white lead in oil, the oil will be extracted with gasoline or some equally suitable solvent, so that the tinting test can be made on the dry pigments. Lead white, ground in oil, is a common form in the market. It visually contains about 8 per cent, of raw^ linseed oil, and has an extended use among painters, as it readily mixes v\^ith additional oil and turpentine to form liquid paint. 520 ENGINEERING CHEMISTRY CO u H o o c s cd rt I, ^ O V ^ o be > a •gtJ S S '^ S c« ^^ 4< 1-^ t 3 w a. to O uV. O ^ SB = ^ X: .^ X = es o 3 »; = « ^ a|8i L'O-TS i- I' l- ■§ a « o ^ ^ J='Si;"!•= o o 0-C /^ 4; o ♦• a.3 y O < '- be c 4> G a « O ^ •0^.5 . rt C O o ^^- a : u. iJ >- u 0> 4>-i 4^ ™ d o; 4j ^ 00 0, ^ o .— 4; < 4/ if 5 4* II 5 p. Cuic cc _ °l .^.l ^2 il .^^ 1 0- 1/ — ^15 « = 3-S - "^ 3 X — bci io-- "ENGINEERING CHEMISTRY 521 las- •53 en Cu 9-c X - ^•4 a I2I 18| m 5i5 ^ 5 C-: (Ui-< O (fl > o oS .5 f' !-•- re ii 2 ^ >» i5 "£5 ^ U U3 0< o Co I? CO fee's i (u • eJ= ■^ i-ti (V o S 3 H •- c8 u g cB rt ^ X ■« r^ OS •5 i = o ii •- ja a - ■« J2 ~ cs s *_ 11 •5£oC M C « (fl a - jjj 4;^ i M .Si « °^ £•250 bco 2 "i c<3 C S (3< S^Oo ii U) «•; (U 01 > lit: "•°^§ II c o o W*— - Si o o 3i? .5 «i s "So - o ^ u SSI •:: ^t a a O o- o £« 30. 03 >.o a: 52: ENGINEERING CHEMISTRY Defects needing correction 11 .r 3 5S a o* W •a 3 a 11 JS ^« • i .=1 w re a « t5 E..2 i| More than a slight percentage of water means an emulsified paint, and when used in such excess is a cheapener and adulterant. .a "3 3 a when dead burnt it becomes chemically inac- tive and is an important ingredient in some bright red oxides and other colors. Tone supporter. S 1 .2d •o 8 CO V a _o a 1 a X W 1 o s re So-called when below 95° F. flash. The familiar volatile known as naphtha or painters' spir- its. On eitlier paraffine or asphaltum ba.se. Boiling point i2o°-i50° F. So-called when above 95° F. flash, and on a paraffine ba.se. Evaporative, value about 35 minutes without leaving stain on paper. Penetrative value at least equal to turpen- tine. Boiling point i5o°-2io° F, r^o-called when above 95° F. .flash, and on an asphaltum base. Fully anequnte solvent properties for oil paints. Unlike benzene, tlie.se spirits possess fine flowing qualities and good flatting qualities. Better solvent for hard gums. Oxygen carrying qualities. A reasonably small peicentage of water im- proves the non-settling conditions and im- proves the brushing qualities, also tends to .satisfy the chemical activity of non-stable pigments, such as corroded lead, during pro cess of manufacture, thus decreasing chemical activity of the applied paint, and promoting durability of the job. u PI a 1 •s a "re U (Quartz silica) (Inlusorial eaith) (Decomposed silica) i *- CO a a ^^ -I Dehydrated calcium sul- phate or dead burnt cal- cium sulphate. «x y re . 2° a ?ire 'x O n CO a re a u 1; a a x" a X a re 1 X 'u •X X re a < WATER Practice (allowed by State authorities is to disregard presence of water in paint up to 1.5 per cent, ot the fluid portion, this amount being recognized as acci- dental or incidental to the raw materials or to the proc- ess of manufacture. All water beyond 1.5 per cent, in the fluid portion to be stated on label. ENGINEERING CHEMISTRY 523 SCHEME FOR THE ANALYSIS OF MIXED CHROMATE AND SULPHATE OF LEAD (LEMON CHROME) NOT GROUND IN Oil Pulverize the sample, pass through a loo-niesh sieve and mix. To i gram in a small beaker, add hvdrochloric acid and heat. Any insoluble matter (usually barytes as a gross adulteration) is to be filtered out, and washed, ignited and weighed. Lead.— One g'am is treated in a covered casserole with 25 cc. concen'rated sulphuric acid and-heated moderately until the residue is perfectly white ; cool, diliite with 50 cc. water and again cool; add 50 to 75 cc. of 04 per cent, alcohol, stir, and allow to stand i hour. Filter, wash well with alcohol, dry, ignite, and weigh as PbS04. Chromium and Sulphuric Acid (SO3). — Treat i gram with about 25 cc. concentrated hydrochloric acid, boil, dilute to 100 cc. and while hi.t add excess of ammonium by. dfoxide which precipitates the chromium and. the greater part of the lead. Boil off the excess of ammonia, filter and wash carefully with hot water. 1. Precipitate for Cr.— Dissolve in dilute HCl. nearly neutralize acid with NH4OH, precipitate Pb with HgS gas and filter into a porcelain dish. 3. Precipitate.— Pbs, reject. 4. Filtrate.— Boil off H2S and precipitate Cr with NH4OH in the usual manner. Put the moist precipitate and filter paper into i. crucible and ignite carefully. Weigh as CroOa. .2. Filtrate for SO3.— Acidify with HCl, con- centrate, add boiling solution of BaClo drop by drop and de t e r- mine SO3 as usual. Occasionally the following determinations are made : Water. — Hygroscopic. Heat ^ gram at 105° C. in an air-bath to constant weight. Volatile Matter. — Heat i gram in a porcelain crucible to low redness ; loss, less water, is volatile matter. Water Extract. — (Acetates, sulphates, bichromates, or nitrates), indicating imperfect washing in manufacture. Treat 3 grams with six successive portions of 25 cc. each, of cold distilled water, decanting and filtering each time, and evaporate the filtrate in a platinum dish to dryness on a water-bath. Specifications for Chrome (Yellow) Issued by the Navy Department, August 2, 1915. 1. The dry pigment shall contain at least 98 per cent., by weight, of normal chromate or basic chromate of lead. 2. When required in paste form, the pigment .shall be finely ground to a medium stif? paste in absolutely pure, well-settled, and perfectly clear raw linseed oil conforming to the latest issue of Navy Department Specifications for Raw Linseed Oil. Such paste shall break up readily in thinning and shall be free from grit, adulterants, and all impurities. 3. The color, shade, tone, fineness, and coloring power, determined by assaying with white zinc, shall be equal to the standard sample. 524 ENGINEJEJRING CHEjMISTRY 4. Portions of the standard sample referred to in paragraph 3 may be obtained upon appHcation to the construction officer's offices at the various navy yards. Analysis of Mixed Chromate, Sulphate, and Carbonate of Lead.^ Analysis made same as in scheme for lemon chrome; excess of lead is to be calculated to white lead, 2PbC03 + PbH^Og. Analysis of Red Chromate of Lead.^ For the lead determination take i grarn in a covered casserole, add 25 cc. concentrated nitric acid, heat to boiling, and while boiling, add }^ dozen drops, one at a time, of alcohol, by means of a pipette; boil a while longer, add water, and all of the chro- mate, if it is pure, will be found in solution. Without this alcohol treatment great difficulty will be experi- enced in getting the chromate into solution; with it it is easily accomplished. Add 25 cc. concentrated sulphuric acid, evaporate to white fumes, and complete the analysis as described. For chro- mium and sulphur trioxide determinations, boil off alcohol and proceed as previously directed. The Volumetric Determination of Red Lead. A rapid and accurate method of finding the amount of red lead (PbgOi) in a sample of red lead.^ Solutions. — I N/io iodine solution. 2 Stannous chloride solution (14.1 grams SnCl^ to 1,000 cc. H,20). 3 Starch solution. Tw^enty-five cubic centimeters stannous chloride solution are ac- curately measured into a lo-ounce Erlenmeyer flask. Forty cubic centimeters hydrochloric acid are added and the whole is raised to boiling. Boil i minute, add 100 cc. cold water, washing down the sides of the flask, and cool rapidly. Add a few cubic centi- meters of starch solution and run in from a burette sufficient 1 " Analysis of Chrome Paints," by W. I,. Brown, y. Anal. Chem.. 1, 213-215. - Known by various names, as scarlet, dark or basic chromate of lead, chrome red, Chinese red, American vermilion, and vermilion substitute. Formula: 2PbO.Cr03 or PbCr04 + PbO. =* Communicated to the author by J. H. Wainwright, Ph. B., F. C. S. e;nginee)ring chemistry 525 iodine solution to give a permanent blue color. This gives the number of cubic centimeters or iodine solution equivalent of 25 cc. stannous chloride solution. Determination. — One gram red lead is very accurately weighed, placed in an Erlenmeyer flask and moistened with water. Run in 25 cc. stannous chloride and 40 cc. hydrochloric acid. Boil until all the lead is in solution. Add 100 cc. cold water and cool rap- idly. Add a few cubic centimeters of starch solution and titrate with iodine to a permanent blue color. The difference in cubic centimeters of iodine solution used in the blank test and in the determination gives the number of cubic centimeters of iodine so- lution to which the available oxygen in the red lead is equivalent. (One cubic centimeter of N/io iodine solution = 0.8 milligram oxygen. ) The titration should be performed as rapidly as possible on account of the action of the acid upon the starch indicator. The available oxygen multiplied by 42.73 equals the percent- age of red lead. Specifications in Part for Red Lead, Dry, Issued by the Navy Department, April 1, 1915. 1, The dry pigment must be of the best quality, free from all adulter- ants, and contain at least 94 per cent, of true red lead (Pb304) — equiva- lent to 32.8 per cent, of lead peroxide (PbOa) — the balance to be prac- tically pure lead monoxide (PbO). It must contain less than o.i per cent, of metallic lead, and to be of such fineness that no more than 0.5 per cent, remains after washing with water through a No. 21 silk bolting cloth sieve. It must be of good bright color and be equal to the standard sample in freedom from vitrified particles and in other respects. 2. When mixed with pure linseed oil, pure turpentine, and Japan drier, as per standard formula, viz. : Red lead, dry pounds 20 Raw linseed oil pints 5 Petroleum spirits gills 2 Drier do 2 and applied to a srnooth vertical iron surface it must dry solidly without running, streaking or sagging. 526 ENGINEERING CHEMISTRY Analysis of Chrome Green. (Composed of yellow chromate of lead, Prussian blue and lead sulphate) To I gram of sample add 25 cc. HCl. heat to boiling several minutes, add water, allow to stand 10 minutes, then filter and wash thoroughly with hot water. 1. Residue. — Prus- sian blue (plus bary- les if present). Dry and ignite to Fe203, if barytes is not present. Weight multiplied by 2.21 equals per cent. Prussian blue. 2. Filtrate.— Nearly neutralize with NH4OH, leaving, however, the solution .slightly iicid. Pass HoS gas through till Pb is all precipitated. Filler and wash. 3. Precipitate, PbS.— Dissolve on filter with hot dilute H NO3 and boil the solution. Filter from collected S and bring filtrate of Pb- (N03)2 to small bulk, with several additions of H0SO4 Evaporate nearly to dryness, cool, idd water and alcohol, filter, wash and weigh as PbSOv 4. Filtrate. — For Cr (and Fe), boil off HoS add NH4OH, in slight excess, boil this off and wash the Cr3(OH)6 and Fe2(OH)6. as customary. Weigh precipitate as Cr^- O3 -t Fe203. After the weight is obtained, mix with one part KNO3 and three parts \a2C03;fuse in platintim crucible to clear fusion, cool, boil with water, filter and wash. 5. Residue. Dry, Ignite, and weigh as Fe203, if it is wanted as a check. 6. Filtrate, for Cr. — Make acid with HCl. Re- duce with alcohol. Preci- pitate with NH^OH in glazed porcelain dish. If the weight of Cr203 isvfry nearly the same as before, then'there has been no Fe extracted from the Prus- sian bjue by the acid treatment. 'Some varie- ties are affected bv this, others not. If the weight is less than the original, deduct it from same. I he result is Fe203. which is also to be calculated to Prussian blue and added to the other. Chrome green, in which the coloring matter is CrgOg, is seldom found in the market pure. Usually it contains from 20 per cent, to 75 per cent, of barium sulphate. Specifications for Chrome Green, Issued by the Navy Department, June 1, 1914. Composition of Pigment. 1. The pigment to contain at least 98 per cent, by weight of pure lemon chrome and Chinese blue, which mixture shall not contain more than 10 per cent, by weight of lead sulphate. Pigment Ground in On.. 2. When required in paste form the pigment shall be finely ground to a medium stiff paste in absolutely pure, well-settled, and perfectly clear raw linseed oil conforming to the latest issue of Navy Department Speci- fications for Raw Linseed Oil. Such paste shall break up readily in thinning and shall be free from grit, adulterants, and all impurities. Kngine:ering chemistry 527 Physicai, Properties. 3. The color, shade, tone, fineness, and coloring power, determined by assaying with French white zinc in oil, shall be equal to the standard sample, which may be seen at the general storekeepers' offices in the various navy yards. Packing and Marking. 4. To be delivered, unless otherwise specified, in 5-pound friction- top cans and 25-pound soldered-top tins as required, properly labeled with the name of the material, quantit}^, and the name of the manufacturer. The 25-pound cans to be provided with bails. Deliveries. 5. Material to be boxed in quantities of 100 pounds each in boxes made of sound wood % inch thick, planed on the outside, and marked with the name of the material, quantity, name of contractor, and requisi- tion or contract number under which delivery is made. Basis oe Payment. 6. Payment will be based on net weight, and net weight only should be delivered. As an example of specifications for a compound chrome paint, the following is given : Pennsylvania Railroad Company — Motive Power Department. Specifications for Cabin Car Color. The standard cabin car color is the pigment known as scarlet lead chromate. It is always purchased dry. The material desired under this specification is the basic chromate of lead (PbCr04PbO), rendered bril- liant by treatment with sulphuric acid, and as free as possible from all other substances. The theoretical composition of basic lead chromate is nearly 59.2 per cent, of the normal lead chromate, and 40.8 per cent, of lead oxide, but in the commercial article it is found that a portion of the sulphuric acid added to brighten the color remains in combination apparently with the normal lead chromate, slightly increasing the percentage of this constituent. Samples showing standard shade will be furnished on application, and shipments must not be less brilliant than sample. The compaiison of sample from shipment with the standard shade, may be either dry or by mixing both samples with oils. Shipments of cabin car color will not be accepted which — 1. Contain barytes or any other adulterant. 2. Show on analysis less than 57 per cent, or more than 60 per cent, of normal lead chromate, including the sulphuric acid combined as above stated. 528 ENGINEERING CHEMISTRY 3. Show on analysis less than 38 per cent, or more than 42 per cent. lead oxide, in addition to the lead oxide in the normal lead chromate. 4. Vary from standard shade. Specifications for Indian Red, Issued by the Navy Department, August 2, 1915. The dry pigment to contain at least 95 per cent., by weight, of oxide of iron (FeaOs) and be free from alkali, lakes, and from more than i/io of I per cent., by weight, of sulphur (S) in any form other than calcium sulphate. When required in paste form, the pigment shall be finely ground to a medium stiff paste in absolutely well-settled and perfectly clear raw linseed oil conforming to the latest issue of Navy Department Specifica- tions for Raw Linseed Oil. Such paste shall break up readily in thinning and shall be free from grit, adulterants, and all impurities. The color, shade, tone, fineness, and coloring power, determined by assaying with white zinc in oil, shall be equal to the standard sample. Portions of the standard sample referred to in paragraph 3 may be obtained upon application to the construction officer's offices at the various navy yards. Unless otherwise specified the material shall be delivered in soldered- top tins containing 5 and 10 pounds each, as required. Tins to be labeled with the name of the material, the quantity, and the name of the manu- facturer. Deliveries to be boxed in quantities of 100 pounds in boxes made of wood % inch thick, planed on the outside and marked with the name of the material, the name of the contractor, the quantity contained, and the requisition or contract number under which delivery is made. Specifications for White Zinc, Issued by the Navy Department, March 10, 1910. 1. American Process. — The dry pigment must contain at least 98 per cent., by weight, of oxide of zinc, not more than 2/10 per cent., by weight, of sulphur in any form, and be of best quality known as "XX." 2. French Process. — The dry pigment must contain at least 99 per cent., by weight, of oxide of zinc and not more than i/io per cent, by weight, of sulphur in any form, and to be of maximum whiteness as com- pared with the standard sample. 3. The requisition will state specifically which kind is desired. 4. The pigment must be of the best quality, finely ground in abso- lutely pure, well settled and perfectly clear raw Hnseed oil of the best quality only to a medium stiff paste, which will break up readily in thin- ning, and must be free from grit, adulterants, and all impurities. 5. The whiteness and fineness must be equal to the standard sample. Any indication of bluing will be sufficient cause for rejection. I^NGINEERING CHE;miSTRY 529 6. Unless otherwise specified, the material is to be delivered in fric- tion-top cans of the required size or sizes, properly labeled with the name of the material, the capacity of the can, and the name of the manufac- turer. Cans of 5 pounds capacity and over must be provided with bails ; net weight to be delivered. 7. The material is to be delivered boxed in quantities of 100 pounds each ; the boxes to be made of %-inch new pine, planed on both sides, and properly labeled with the name of the material, the name of the manufacturer, and the contract and requisition number on which the material was purchased. Specifications for Brown Zinc, Issued by the Navy Department, April 30, 1910. I. The dry pigment must contain at least 5 per cent., by weight, of the finest red lead, at least 40 per cent, of zinc oxide, and at least 20 per cent., by weight, of iron oxide and must not contain more than 5 per cent., by weight, of lime in any form. S. The pigment must be of the best quality finely ground in absolutely pure, well-settled and perfectly clear raw linseed oil of the best quality only to a medium stiff paste, which will break up readily in thinning, and must be free from grit, adulterants, and all impurities. 3. The color, shade, tone, fineness and coloring power, determined by assaying with French vv^hite zinc in oil, must be equal to the standard sample which may be seen at the general storekeepers' offices in the various navy yards. 4. Unless otherwise specified, the material is to be delivered in fric- tion-top cans of the required size or sizes, properly labeled with the name of the material, the capacity of the can, and 4he name of the manufac- turer. Cans of 5 pounds capacity and over must be provided with bails ; net weight to be delivered. Specifications for Aluminum Paint, Issued by the Navy Department, Aug^ist 2, 1915. To be of the best quality and manufacture and consist exclusively of not less than 18 per cent., by weight; of finely powdered pure aluminum in accordance with Navy Department Specifications for Powdered Alumi- num in a vehicle containing by volume 25 to 30 per cent, of pure hard varnish resins, 4 to 5 per cent, of pure raw linseed oil, 65 to 70 per cent, of pure gum spirits of turpentine conforming to the latest issue of Navy Department Specifications for Turpentine, lead-manganese driers, and be free from all adulterants or other foreign materials. 34 530 I^NGINEERING CHEMISTRY The paint shall not flash below 105° F. (open tester) and, when applied to a steel plate alongside of a standard sample, shall be equal to the standard in color, brightness, body, finish, covering properties, elasticity, and durability. When exposed to the action of oil, steam, and salt water, to be equal to the standard in all respects. Heat-resisting qualities : Iron articles dipped in the paint shall be subjected to a dull red heat. The color should not change greatly, and there should be practically no blistering or cracking off of the paint. Iron articles treated as above should show but slight cracking off of the paint when plunged while red hot into water. Portions of the standard sample referred to in paragraph 2 may be obtained upon application to the construction officer's office at the various navy yards. Fire-proof Paints, Silicate Paints, Asbestos Paints, etc. The principle of action of these paints is not to render v^ood- work or similar material fire-proof, but to retard combustion. Wood treated with a solution of zinc chloride, or w^ith a solution of sodium silicate, can be rendered nearly non-inflammable, and after such treatment and drying, paint can be applied. Instead of using the ordinary paints for this purpose, various compounds are incorporated in the paint itself to render the latter non-inflammable. Thus the preparation of Prof Abel J. Martin, of Paris, is as follows : Boracic acid, borax, soluble cream of tartar, ammonium sulphate, potassium oxalate, and glycerine mixed with glue and incorporated with a paint. It is the result obtained after many experiments in response to a prize of 1,000 francs, offered by the Society for the Advancement of National Industry in France. A committee consisting of Professors Du- mas, Palaird, and Troost, after testing the materials, consisting of painted woods and various fabrics, for 7 months, reported in favor of this preparation. The municipality of Paris made its use obligatory in all of the theatres there and it has stood the test of the last 8 years. Blue Pigments. — Ultramarine being a silicate, can be analyzed by the scheme on page 514. ENGINEE^RING CHEMISTRY 53 1 Composition op Ui^tramarine. Per cent. Si02 49.68 AI2O3 23.00 S 923 S03 2.46 NaiO 12.50 H.O 3-13 Total 100.00 Prussian Blue.^ Under the name Prussian blue are included all ferrocyanid blues such as Antwerp blue, Chinese blue, Turnbull's blue, etc. These blues are all ferric ferrocyanides, ferrous ferricyanides, or double iron potassium salts of hydro- ferrocyanic or hydro-ferri- cyanic acids. The analysis of these blues, as is generally the case with pigments, does not necessarily give results which can be used to grade samples, the strength and color tests being most important. Most text books say that Prussian blue is ferric ferrocyanide, Fe4[Fe(CN)6]3, but this substance is not known commercially. Commercial Prussian blue is a mixture of Wil- liamson's blue, KFe[Fe(CN)6], with other iron-alkali cyanids and often with aluminum-iron cyanids, altogether a most com- plex substance.^ Moisture. For the determination of moisture dry 2 grams for 2 hours at 100° C. Dry blue should contain less than 7 per cent, of moisture. Insoi.uabi,e: Impurities. Ignite I gram in a porcelain dish iat a low temperature. The ignition should be carefully carried out at a temperature just high enough to decompose the last trace of blue, but not high enough to render the iron insoluble, in hydrochloric acid. Cool, add 15 cc. of hydrochloric acid, digest for i hour on the steam bath covered with a watch glass, evapo- 1 Percy H. Walker, Bulletin Chemistry, 109. U. S. Dept. Agriculture. • Parry and Coste, The Analyst, 1906, 21, 225-230. 532 e:ngini:kring chemistry rate to a syrup, add water, boil, filter from the insoluble, wash, ignite, weigh, and determine the nature of the insoluble, probably barium sulphate. In pure Prussian blue solution should be complete. ToTAi, Iron. Decompose as in the foregoing determination, reduce, and de- termine the iron in the ordinary way. There should not be less than 30 per cent, calculated on the dry pigment. ToTAi. Nitrogen. Determine on a i-gram sample by the official Gunning method, digesting for 3 hours. ^ The percentage of Prussian blue may be obtained with sufficient accuracy for commercial purposes by multiplying the percentage of nitrogen by 4.4 and by mul- tiplying the percentage of iron by 3.03. Eight samples of pure Prussian blue examined by Parry and Coste gave the mean re- sults from which these factors are calculated. The following table shows the accuracy with which these factors give the per- centage of Prussian blue in the eight samples : Parry and Coste's Determination oe the Percentage 01 Beue in the Dry Matter oe Eight Samples. ' Prussian Factors No. I No. 2 No. 3 No. 4 No. 5 No. 6 No. 7 No. 8 Nitrogen X 4-4 Iron ^ 1 ox 94.69 94-63 100.93 101.02 107.58 109.71 94.91 97.11 98.95 94.84 99.83 98.99 100.80 102.47 102.52 100 26 Other Ddtejrminations. It is seldom worth while to make any further determinations. If desired, however, the iron and aluminum may be precipitated as hydrates by ammonium hydroxide and weighed together as oxides, and the aluminum obtained by difference after determin- ing the iron volumetrically and calculating to ferric oxide. The filtrate from the iron oxides and alumina precipitate may be made up to a definite volume and one aliquot taken for the determin- ation of sulphate and another evaporated with sulphuric acid, ignited, and weighed. Determine whether the alkali is sodium 1 U. S. Dept. Agr., Bureau of Chemistry, Bulletin No. 107, p. 7. DNGINKEJRING CHEMISTRY 533 or potassium and subtract the alkali metal corresponding to the sulphate (SO4) found. The remainder is double alkali iron ferrocyanide. Well-washed blues should be neutral in reaction. The red shade may be due to organic red. Test the solubility in alcohol, etc. Vermilion. True vermilion, or, as it is generally called, English vermilion, is sulphide of mercury. On account of its cost it is rarely used in paints and is liable to gross adulteration. It should show no bleeding on boiling with alcohol and water and no free sulphur by extraction with CSg- A small quantity mixed with 5 or 6 times its weight of dry sodium carbonate and heated in a glass tube should show globules of mercury on the cooler portion of the tube. The best test for the purity is the ash, which should not be- more than J^ of I per cent. Make the determination in a porcelain dish or crucible, using 2 grams of the sample. If it be desired to determine the amount of mercury, proceed as fol- lows : Place in the closed end of a combustion tube 45 centimeters long and 10 to 15 millimeters in diameter, a layer of 25 to 50 millimeters of roughly pulverized magnesite, then a mixture of 10 to 15 grams of the vermilion with four or five times its weight of lime, followed by 5 centimeters of lime, and plug the tube with asbestos. Draw out the end of the tube and bend it over at an angle of about 60°. Tap the tube so as to make a channel along the top, and place it in a combustion furnace with the bent neck down, resting with its end a little below some water in a small flask or beaker. Heat first the lime layer, and carry the heat back to the mixture of lime and pigment. When all the mercury has been driven off, heat the magnesite, and the evolved carbon dioxide will drive out the last of the mercury vapors. Collect the mercury in a globule, wash, dry, and weigh. Genuine vermilion is at the present time little used in paints. Organic lakes are used for most of the brilliant red, scarlet, and vermilion shades. These organic coloring matters are sometimes precipitated on red lead, orange mineral, or zinc oxide; but as a 534 ENGINEERING CHEMISTRY usual thing the base is barytes, whiting, or China clay. Parani- traniline red, a compound of diazotized paranitraniline and beta- naphthol is largely employed; but a number of colors may be used. To test for red colors in such a lake the following method from Hall may be of value, though other colors may be employed, which makes the table of only limited use. Determination of Red Coi^ors in Organic Lake Reagent Sulphuric acid, cone Hydro- chloric acid, cone. Sodium hydroxide. cone. sol. Alcohol Sodium hydroxide, cone, and alcohol Source of color Dark brown with reddi.sh under- tone becoming light yellow on diluting. Color becomes muddy." Dark reddish brown; little change on di- luting. Insoluble .... Purplish; dark brown on di- luting. Eosine Changes to yel- low; fluores- cent solution with excess of sodium hy- droxide. Changes to yel- low; reddish fluorescent so- lution with ex- cess of sodium hydroxide. Ivittle change; fluorescent .so- lution on dilut- ing. Reddish fluores- cent solution. Para-nitraniline ai?is\dhie Scarlet (2R) Purple color re turning on di- luting. Color slightly darkened; lighter on di- luting. Color changed to brownish red; light red on di- luting. Slight yellowish orange solu tion. Purple; color re turning on di- luting. Purple; red on dilut ing. Dark pur plish red lighter on diluting I^ittle change Ivittle chslnge. Darkened; lighter on di- luting. Darkened, lighter on di- luting Reddish solu- tion on dilut ing. Slight reddish solution. Color darker, reddish solu- tion on dilut- ing. It is well also to try the action of reducing and oxidizing agents such as stannous chloride, ferric chloride, etc. (See also Schultz and Julius, A Systematic Survey of the Organic Coloring Mat- ters.) Paranitranilin red is soluble m chloroform. It is also well to try the solvent action on different reds of sodium carbonate, etc. The amount of organic pigment present in such reds is generally very small, and when it can not be determined by ignition owing to the presence of lead, zinc, or carbonate, it is best determined by difference. (Percy H. Walker.) DNGINEJERING CHEMISTRY 535 Extraction of the Vehicle in Mixed Paints. Weigh from 15 to 35 grams of the thoroughly mixed pigment in a tall, narrow Erlenmeyer flask of 300 cc. capacity. Add 150 cc. of gasolene, stopper with a cork, and shake for 10 minutes with a shaking machine so constructed that the liquid is not brought in contact with the stopper; allow to stand until the pigment has settled, and decant the liquid into a beaker; some pigment will frequently go over with the gasolene, so it is well to allow this to settle and decant into a second beaker. Repeat the treatment with gasolene; after the first treatment settling is generally much more rapid. Decant the gasolene as completely as possible the second time. Then add 150 cc. of benzol (CgHe), shake and allow to settle. Decant the benzol and treat in same way with 150 cc. of ether. This method of extraction is often more satisfactory than any method using a continuous extraction apparatus ; for frequently the pigments can not be held by extrac- tion thimbles. This method of treatment will generally give an al- most complete separation of the vehicle; but in some enamel paints it is well to follow the gasolene treatment by a treatment with turpentine, and then remove the turpentine with gasolene, be- fore treating with benzol and ether. No system of extraction will remove absolutely all the vehicle, the insoluble portion being probably metallic soaps or linoxyn. After removing all of the soluble vehicle, dry the pigment, first at a low temperature in a gentle current of air, and then at 105° C, weigh, and from the loss in weight calculate the per- centage of vehicle and pigment; then from the total weight of mixed paint and the weight of clear vehicle drawn off calculate the percentage of vehicle and pigment in the original paint. Analysis of the Vehicle from Paint. Weigh 50 grams of the vehicle into a 500 cc. flask, connect with a spray trap and a vertical condenser, and pass through it a current of steam, first heating the flask in an oil bath at icx)° C. ; with the steam still passing through, raise the tempera- ture of the bath to 130°. Catch the distillate in a small weighed 536 ENGINEERING CHEMISTRY separatory funnel ; continue distillation until the funnel contains 150 cc. of water. Let the distillate stand until separated into two layers, then draw off the water, and weigh the light oils. Ex- amine as under turpentine, page 538. A slight error is caused by the solubility of turpentine in water ; this amounts to about 0.3 to 0.4 cc. for each 100 cc. of water. Cut off the steam, remove the trap, and draw air through the flask for about 15 minutes, keeping the oil bath at 130° C. The residue is now free from water and can be examined according to the following procedure for the residue from dry distillation : When sufficient vehicle is available it is well to take another portion and distill, without steam, placing the flask in an air bath. Note the temperature of the bath at which distillation takes place, and continue the distillation at a temperature of 185° C. in the air bath. This method gives somewhat lower results on volatile oils than the first method, but the distillate can be tested for water-soluble volatile liquids which would be lost by the steam distillation. Unsaponifiable matter should be determined in this residue, or in some of the original vehicle. The residue is frequently too pasty for the determination of the specific gravity, which should be made on the original vehicle. Determine the acid number of the residue of the original vehicle. Determine the percentage and the character of the ash either from the residue or the original vehicle. The iodine number is sometimes a useful index; but the constants of linseed oil which has been mixed with pigments, especially lead compounds, may be so much altered that an iodine number as low as 100 can not be taken as any proof of the presence of other fatty oils. Test for resin may be made by dissolving in carbon disulphide and adding a solution of stannic bromide or chloride in carbon di- sulphide. Use a white porcelain dish. If no water is present in either solution the presence of resin is shown by the appearance of a violet color. This test is not as delicate as the Liebermann- Storch test described under linseed oil. When much lead is present it may be lost in ashing, and for a correct determination of metals the following method is best: e:ngine:e:ring chemistry 537 Place 25 grams of the vehicle in a 500 cc. separatory funnel, dilute with 25 cc. of a mixture of equal parts of gasolene and turpentine, add 50 cc. of nitric acid (i : i), and let stand i hour, shaking every 10 minutes. Then immerse the funnel in hot water, loosen the stopper and shake gently. This drives off nearly all the gasolene. Remove from the hot water, let it separate, draw off the lower layer, and wash the upper oily layer four or five times with warm water. Add the washings to the main acid por- tion and determine the metals in the ordinary manner. If the paint is enamel paint, treat the vehicle as a varnish. VARNISH. The methods of analysis for varnish are far from satisfactory. The following method, slightly modified, is one devised by S. S. Voorhees, and while not absolutely satisfactory, is probably the best available : iNSoiyUBivE Gums. Weigh 2 grams of the varnish into a weighed 150 cc. Erlen- meyer flask, add 2 cc. of chloroform, and then 100 cc. of 88° B. gasolene; add the gasolene gradually, shaking constantly so as to avoid any precipitation, until 15 cc. are added. Allow to stand over night in a cool place. The gums will adhere to the bottom and side of the Erlenmeyer flask; decant into a weighed beaker and wash with a little 88° gasolene. Dry for 2 hours at 105° C. and weigh as insoluble gums. Soi.uBi,E Gums and Linoxyn. Evaporate the gasolene extract and dry the residue for 168 hours at 100° to 105° C, or to constant weight, and weigh. This treatment should convert all linseed oil into linoxyn. Add 15 cc. of chloroform and digest over night to dissolve the gums but not the linoxyn. Filter through a wad of absorbent cotton into a weighed beaker, evaporate the chloroform, dry for 2 hours at 105°, and weigh as soluble gums. Einoxyn is obtained by dif- ference from the first weight. 538 dnginbering chemistry Acid Vai,ue. Determine the acid value in the usual way on 10 grams of the varnish. After getting the acid value, decant the alcohol, evapo- rate, and apply the Liebermann-Storch test for resin. Ash. Determine the ash on 10 grams (in a porcelain dish). Deter- mine the reaction of ash with litmus paper; if alkaline, test for lime. It is sometimes well to determine lime, a large amount of which indicates resin. Misce;i.i.ane:ous. Volatile oils and metals are determined as in the analysis of the vehicle under mixed paints. It is not possible from such an examination as has been de- scribed to decide on the va-lue of a varnish for any particular pur- pose. An examination of the varnish film should always be made. The film is best made by flowing the varnish on glass, and films should be dried in both a horizontal and a vertical position at a uniform temperature, 38° C. Note the time of setting, the ap- pearance, the hardness and toughness of film. TURPENTINE. Directions for the; Anai^ysis o^ Turpentine. Appearance. — On receipt of samples, note and record whether the samples are free from dirt, suspended matter and water. If the samples contain water, filter through a dry filter paper into a clean dry bottle. Color. — Into a 200-millimeter colorimeter tube graduated into millimeters, place 50 cc. of the turpentine to be examined ; on the tube place a No. 2 Lovibond yellow glass ; over a second 200 milli- meter tube, place a No. i Lovibond yellow glass; add to the second tube enough of the sample of turpentine to match the color in the first tube, and record its reading in millimeters. Specific Gravity. — Determine the specific gravity by any suit- DNGIN^E^RING CHEMISTRY 539 able accurate method and report as specific gravity at 15°. 5/15°. 5 C. State the method used. Refractive Index. — Determine with a direct reading refrac- tometer at 15*^.5 C. Distillation Test. — (i) Place 200 cc. of the sample into a 300-cc. flask, 8 centimeters in diameter, with a side tube 8 centi- meters from the main bulb, and the neck extending 8 centimeters above the side tube. The neck is 2 centimeters in diameter and the side tube 5 millimeters. This flask should be fitted with a thermometer (reading from 145 to 200° C.) immersed in the vapor. The mercury bulb should be opposite the side tube of the flask and the reading 175° C. should be below the cork. The dis- tillation should be so conducted that there shall pass over about 2 drops of the distillate per second. (2) Place 100 cc. of the sample into an ordinary Engler flask. Have thermometer totally immersed in the vapor as directed in the specifications in test No. i. (3) Place 100 cc. of the sample in an ordinary Engler flask (see test No. 2) and use an ordinary long-stem thermometer. Report emergent reading and approximate length of the exposed mercury column and its approximate temperature. In all three methods of distillation note and report the initial boiling point. Note temperature at each 10 cc. of distillate and note volume of distillate at 160, 165, 170 and 175° C. If possible, note and report barometric pressure at time of making distilla- tion. Evaporation Test. — Ten cubic centimeters of the sample are placed in a glass crystallizing dish, 2^ inches in diameter and I ^ inches high, and evaporated on an open steam bath with a full head of steam for 2 hours. Cool, weigh, and report weight of residue in grams. Polymerisation. — (i) Add slowly 5 cc. of the turpentine to 25 cc. of sulphuric acid (specific gravity 1.84) contained in an or- dinary, graduated, narrow-necked Babcock flask. Shake the flask with a rotary motion to insure gradual mixing. Cool if necessary in ice water, not permitting the temperature to rise 540 DNGINKERING CHEMISTRY above 60 to 65° C. Agitate thoroughly and maintain at about 65° C. with frequent agitations for i hour. Cool, fill the flask with H2SO4, bringing the unpolymerized oil into the. graduated neck. Allow to stand i hour. Read off unpolymerized content; note and report its consistency and color, and determine its re- fractive index at 15°. 5 C. (2)^ Repeat test No. i but use sulphuric acid that is 38 N and let flasks stand 24 hours before reading the amount of unpoly- merized residue, or else centrifuge 5 minutes. Hydrochloric Acid Test. — Shake 10 cc. of the turpentine with 10 cc. of concentrated hydrochloric acid (specific gravity, 1.19). Note whether after 3 minutes standing a decided red color develops. (Test for the presence of furfural or heavy or resinous oils.) Flash Point. — (i) Support a 100 cc. nickel crucible, such as is used in determining the flash point of linseed oil, in a vessel of water at 15 to 20° C. ; the water should cover about two-thirds of the crucible. Fill the crucible to within about 2 centimeters of the top with turpentine, insert a thermometer, and heat the water bath slowly so that the temperature of the turpentine rises 1° C. per minute. Begin at 37° C. and test for the flash at each rise of o°.5 C. Report temperature at which the turpentine flashes. (2) Determine the flash point using the Tagliabue open cup. Begin testing at 30° C. and test at each degree Centigrade above that till the sample flashes. The temperature of the turpentine should not rise more rapidly than 1° C. per minute. (3) Use a closed tester such as the Pensky-Martin tester, the Abel cup, etc., following the directions for the instrument. Specifications for Spirits of Turpentine, B. & 0. R. R. Co. I. The material desired is the properly prepared distillate of pine, or pine pitch, unmixed with any other substance. a. It must be water white or prime white in color. h. Its gravity must be between 0.862 and 0.872° F. c. It must boil between 310 and 320° F., and at least 95 per cent, must distill over below 338° F. i Donk's Method ; Bulletin No. /jj or Circular No. 15, Bureau of Chemistry. ENGINEERING CHEMISTRY 54I d. Upon evaporation at 212° F., the residue must not exceed 2 per cent. e. When 6 cc. of the material are thoroughly mixed with 24 cc. of concentrated sulphuric acid in a graduated tube, kept cool while mixing and the mixture allowed to stand for ^ hour, not more than 6 per cent, must separate, as a clear layer. /. Weight to be calculated at 7 pounds per gallon. 2. Material failing to meet the above tests or found by other standard tests to be impure will be rejected. 3. All rejected material will be returned, the shipper paying freight both ways. Determination of Resin in Shellac. Standard Method i^or the Determination of Resin in Shei.i.ac. The solutions required are one of iodine monochloride contain- ing 13 grams of iodine per liter, in glacial acetic acid that has a melting point of 14.7 to 15° C. and is free from reducing impuri- ties; and another of sodium thiosulphate, made by dissolving 24.83 grams of the pure salt in a liter of water. In addition to these solutions there is required a quantity of acetic acid of the same strength as that used for making the solution of iodine monochloride. Pure chloroform and starch are also necessary. The preparation of the iodine monochloride solution presents no great difficulty, but it must be done with care and accuracy in order to obtain satisfactory results. There must be in the solu- tion no sensible excess either of iodine or more particularly of chlorine, over that required to form the monochloride. This condition is most satisfactorily attained by dissolving in the whole of the acetic acid to be used the requisite quantity of iodine, using a gentle heat to assist the solution, if it is found necessary. Set aside a small portion of this solution, while pure, and pass dry chlorine into the remainder until the halogen content of the whole solution is doubled. Ordinarily it will be found that by passing the chlorine into the main part of the solution until the character- istic color of free iodine has just been discharged, there will be a slight excess of chlorine, which is corrected by the addition of 542 ENGINEERING CHEMISTRY the requisite amount of unchlorinated portion until all free chlor- ine has been destroyed. A slight excess of iodine does little or no harm, but excess of chlorine must be avoided. Introduce 0.2 gram of ground shellac into a 250 cc. dry bottle of clear glass with a ground glass stopper, add 20 cc. of glacial acetic acid (melting point 14.7 to 15° C.) and warm the mixture gently until solution is complete (except for the wax). A pure shellac is not easily soluble; solution is quicker according to the proportion of resin present. Ten cubic centimeters of chloroform are added and the solution is cooled to 21 to 24° C. The temperature should be held well within these limits during the test. Twenty cubic centimeters of Wijs solution are added from a pipette, having a rather small delivery aperture. The bottle is closed and placed in a dark place, and the time noted. It is con- venient to keep the bottles during the test partly immersed in water which should be kept as nearly as possible between 22 and 23° c. Pure shellac will scarcely alter the color of the Wijs solution. If in small amount, resin will produce a slowly appearing red- brown color. In large amount, resin causes an immediate color- ation, increasing in intensity as time passes. After i hour 10 cc. of 10 per cent, potassium iodide water solution are added. The solution is immediately titrated, with the sodium thiosulphate solution ; 25 or 30 cc. may be run in immediately, unless the shellac is very impure, and the remainder gradually, with vigorous shak- ing. Just before the end, a little starch solution is added. The end point is sharp, as the reaction products of shellac remain dissolved in the chloroform; any color returning after ^ minute or so is disregarded. A blank determination should be run with 20 cc. Wijs solu- tion, 20 cc. of acetic acid, 10 cc. of chloroform, and 10 cc. of 10 per cent, potassium iodide solution. The blank is necessary on account of the well known effect of temperature changes on the volume, and possible loss of strength of the Wijs solution. In the case of grossly adulterated samples, or in the testing of pure resin, it is necessary to use, instead of 0.2 gram of material, a smaller amount, say 0.15 gram or even o.i gram, in order that ENGINEERING CHEMISTRY 543 the excess of iodine monochloride may not be too greatly reduced, since the excess of halogen is one of the factors in determining the amount of absorption. It is safe to say that in case less than 25 cc. of the thiosulphate solution are required, another test should be made, using a smaller amount of the shellac to be tested. In weighing shellac, some difficulty is at times experienced on account of its electrical properties. In very dry weather it may be found that the necessary handling to prepare it for weighing has electrified it, and that it may be necessary to leave it on the balance pan at rest for a few minutes before taking the final weight. No pure shellacs show a higher iodine absorption than i8. As shellac is relatively a high-priced material and as the variation between its highest and lowest figure is not great, the sub-com- mittee believes that i8 should be taken as the standard figure for shellac, determined by the method above described. As it is an accepted principle that a standard method should be so devised that its inaccuracies shall work in the direction of favoring the seller rather than of condemning too severely the article sold, the sub-committee approves the value taken by Doctor Langmuir for the iodine number of resin, namely, 228. The results of using in this method the value 18 as the iodine number of shellac and 228 as the number of resin, may be that a slightly lower percentage of resin, under some circumstances, will be found than that which is actually present. The percentage of resin is determined as follows : Iodine number of shellac = 18 Iodine number of resin = 228 Iodine number of mixture = X Percentage of resin =100 --^ ^ (228— iS)- References. "Painting Defects, Their Causes and Prevention," by Gustave W. Thomp- son, Jour. Ind. and Eng. Chem., Feb., 1915. "The Constitution of White Lead," by Edwin Euston, Jour. Ind. and Eng. Chem., March, 1914. 544 ENGINEERING CHEMISTRY THE CHEMICAL AND PHYSICAL EXAMINATION OF PAPER. This subject may be conveniently divided into eight sections: 1. Determination of the nature of the fiber; 2. Microscopical examination ; 3. Determination of free acids; 4. Determination of the nature and amount of the sizing used ; 5. Determination of the amount of ash and its analysis ; 6. Determination of the weight per cubic decimeter; 7. Determination of the thickness of the paper; 8. Determination of the absolute breaking strength. 1. Determination of the Nature of the Fiber. The introduction, in late years, of the various kinds of wood fibers in the manufacture of paper has rendered the chemical ex- amination of the same exceedingly difficult. This is more especially so where the wood fiber has been sub- jected to chemical treatment, as in the "sulphite process" or the "soda process," before being incorporated in the paper. Nearly all of the chemical reactions for the recognition of the wood fibres in paper produce certain colors with the various resins in the wood when the reagent is added. While the fiber prepared entirely by the "mechanical" process can be indicated without difficulty, even when mixed with cotton and linen in various amounts, the conditions are greatly altered when the wood fiber has been subjected to bleaching and chemical treat- ment, since the latter removes much of the resinous matters of the wood and increases the difficulty of the quantitative exami- nation. The chemical reactions of the fiber produced from the various woods used in paper-making, pine, poplar, and spruce, are iden- tical, qualitatively, with the following reagents : 1. Hydrochloric acid and phloroglucine produce a red color with "mechanical" wood pulp ; 2. Analine sulphate produces a yellow color ; 3. Naphthylamine and hydrochloric acid produce an orange yellow color ; 4. Anthracene hydrochlorate produces a red color ; ENGINEJERING CHEMISTRY 545 5. Phenol hydrochlorate produces a bluish-green color; 6. Concentrated hydrochloric acid produces a violet color; 7. Pyrrol and hydrochloric acid produce a purple-red color; 8. Pyrogallic acid and zinc chloride produce a dark violet color ; 9. Nitric and sulphuric acids produce a red color; 10. Hematoxylin solution produces a red color; 11. Alcoholic solution of cochineal produces a blue-violet. Where the wood pulp is composed entirely of "mechanical" wood fiber the above reactions are very marked and by the aid of the microscope, the varieties of wood can be determined. Wood pulp produced by the "soda" or by the "bisulphite" process gives a much weaker reaction with the chemical reagents used for identification, and in many instances where the pulp has been used many times in paper-making will give no color reac- tions sufficient for recognition. The amount of "mechanical fiber" in a mixture of "chemical fiber," linen fiber, cotton fiber and "mechanical fiber" in a paper can be determined as follows : The sample of paper is first boiled in water, then in alcohol, and afterwards digested with ether. After drying, a solution of gold chloride is added. Linen, cotton and "chemical" wood fiber have no reducing action upon the solution of gold; but the mechanical wood fiber immediately reduces gold from the solution, this action being due to the ligno-cellulose remaining in the mechanical wood fiber. One hundred grams of mechanical wood pulp, under above con- ditions will reduce 14,285 grams of gold.^ If a sample of paper be submitted for examination as to the fibers used in its manufacture, the following preliminary work is requisite: The rosin, sizing, filling, etc., in the manufactured paper must first be removed. Cut the paper into small pieces, place them in a beaker and digest with a solution of caustic soda (i part caustic soda to 30 of water), at a moderate heat for ten minutes. Pour off the liquid, replace with double the amount of distilled water, and warm ten minutes; pour off this liquid, and repeat once. Now place the paper in a solution composed 1 "Handbuch der technisch-chemische Untersuchungen," (BoUey), 6 Auf., p. 1007. 35 546 ENGINEERING CHEMISTRY of I part hydrochloric acid and 15 parts of distilled water and digest ten minutes. Wash a number of times with distilled water, until washings are no longer acid ; then dry. Suppose the sample of paper so treated to be composed of a mixture of "mechanical" chemical wood fiber, linen and cotton — a mixture to be found in many samples of good quality of writing paper. A sample of the dried paper is tested with solution of gold chloride. If no reduction of gold takes place, the indications point to the absence of mechanical wood fiber. This, however, is not absolute, since, if the paper has been made from "cuttings," "old paper stock," etc., etc., the mechanical wood pulp might have been treated quite a number of times by chemicals in the production of the finer quality of paper, and its ligno-cellulose destroyed or modified in such a way as to nullify the gold test. Generally speaking, however, the reduction of the gold clo- ride is indicative of the presence of mechanical wood fiber. ^ R. Benedikt^ gives a method for the determination of mechan- ical wood fiber in paper, dependent upon the methyl numbers of lignin contained in it. This process has been tested by W. Herzberg^ with the result that preference is given to the use of gold chloride solution. If the amount of mechanical wood fiber in a paper amounts to about 10 per cent., Gottstein* states that the fibers may be counted under the microscope, after the fibers have first been made visible by a treatment with an alcoholic phlorogiucinol solu- tion and hydrochloric acid. Fifteen per cent, or more of the mechanical wood fiber in the mixture renders the test valueless. If chemical wood fiber be present in a paper with mechanical wood fiber, no color tests for the former are positive in the pres- ence of the latter, since the mechanical wood pulps possess a greater tinctorial power. 1 "Ueber die qualitative Bestimmung des Holzschliffsim Papier," von Rich, Godeflfroy und Max Conlon, Mitteilungen aus dem R. K. technologischen Gewerbemuseum in Wien 1888. Mitteilungen aus dem Koniglischen technischen Versuchsanstalten zu Berlin (1892) p. 54. 2 Chem. Ztg., 15, 201. 3 Mitt. Konig. tech. Versuchs (1891), 44-50. < Papier-Zeitung (1884), 432. ENGINEERING CHEMISTRY 547 Should mechanical wood fiber be absent, however, a solution of resorcin can be applied to a properly prepared sample of the paper. Chemical wood fiber produces a violet color, whereas cotton and linen are without action. A solution of phenol also produces a violet color under similar conditions. 1 \ i ■'^s \ ... Fig. 93- Fig. 94. Fig- 95- Fig. 96. 548 ENGINEERING CHEMISTRY 2. Microscopical Examination. By careful manipulation of the microscope, the fibers of linen, cotton, and the various woods can be recognized. The distinction must be noticed here, however, that the fibers from paper, no matter what the source, do not have the appear- ance under the microscope that they possessed before the me- Fig. 97. Fig. 98. Fig- 99-- Fig.: 100. ENGINEERING CHEMISTRY 549 chanical and chemical treatment required in the manufacture of paper. The chemical process in paper-making is very severe upon the various fibers, since they are subjected to beating and cutting in the ''beating machine," to protracted maceration in strong alkali, to digestion in boiling water, to bleaching with chloride of lime, are loaded with various clays, and finally are sized, and often burnished. This difference between linen fibers before and after treatment is shown in Figs. 93 and 94. A comparison shows not only a radical change in the form of the fibers, but a difference in the transparency, due to removal of soluble portions of the fiber. Poplar wood fiber ( Fig. 95 ) made by chemical process, under the microscope, resembles the fibers of linen more than does any of the wood fibers. It, however, has one distinguishing char- acteristic, even among the disintegrated pulps; that is, the tan- gential fragments have among them particles bearing a grate, or screen-like appearance, as shown in Fig. 96. The coniferous woods used in paper-making show peculiarities in structure entirely different, under the microscope, from linen and cotton, the most distinctive one being the small circular "pits" or spots along the center of each fiber. A section of spruce wood, composed of 15 or more fibers, is shown in Fig. 97. After pulping and making into paper, spruce fiber has the appearance under the microscope shown in Fig. 98. It still retains the peculiar circular markings, and is readily distinguished from the linen paper fiber, Fig. 94, or from cotton fiber, Fig. 99. In Fig. 100 is shown the peculiar ''center-making" of conifer- ous fiber, as taken from a sample of writing paper sold as linen paper, but shown by both chemical and microscopical examina- tion to be composed largely of spruce fiber and linen. ^ The microscope will thus determine the differences between the various fibers used in paper-making, and, by properly ar- 1 The microphotographs used in this article are from specimens made during an in vestigation upon fibers of papers by Charles S. Schultz, past president N. Y. Microscopical Soc, and the writer, and represent the fibers magnified 200 diameters. 550 ENGINEEJRING CHEMISTRY ranged apparatus connected therewith, the percentage of each variety of fiber. According to the German official direction the sample of paper, after removal of sizing, etc., is to be steeped in a solution of 0.2 gram of iodine and 2 grams of potassium iodide in 20 cc. of water and then examined under the microscope. The fibers may be conveniently divided into three groups : 1. lyinen, hemp, and cotton; 2. Wood-cellulose ("chemical" wood-fiber), straw-cellulose and esparto ; 3. Ground wood-cellulose and jute. After treatment with the above solution, the fibers of group i are stained brown, those of group 2 are not colored, whilst the strongly lignified fibers of group 3 are colored yellow. But it has been found that this method is somewhat defective, the cel- lulose of group 2, for example, being invariably to some extent stained, whilst the members of group i are so deeply colored that it is almost impossible to distinguish their structural characters. After many experiments, the following method was found more satisfactory. The paper is placed on the. object-class of the microscope and treated with iodine solution, the unabsorbed iodine removed by means of filter-paper, and the paper covered with dilute sulphuric acid. The solution of iodine in potassium iodide should be of such a strengh that a layer of 3 cc. thickness should be of ruby- red color and quite transparent. The paper is now removed and boiled with a solution of dilute potassium hydroxide, washed thoroughly, and replaced on the object-glass. The color reactions are as follows : 1. Cotton, linen, and hemp take a violet red or wine-red color; 2. Well bleached wood-cellulose and ordinary bleached straw- cellulose are colored gray-blue or pure blue, without any tinge of red; 3. Unbleached or imperfectly bleached wood fiber absorbs very little iodine and remains colorless ; 4. Strongly-lignified fibers, such as ground wood cellulose and raw jute, are colored yellow. ENGINEERING CHEMISTRY 55 1 The numbers of each variety of fiber are now carefully counted by means of the microscope and an eye-piece micrometer ruled in squares. This chemical treatment and microscopical examina- tion is to be repeated upon at least 50 different pieces of paper from different parts of the sample, and an average taken. By this means approximate percentages of each variety of fiber in the paper can be stated.^ 3. Determination of Free Acids in the Paper. Free acids in the paper may be : 1. Chlorides, from the hypochlorites used in the bleaching, and which have not been removed by the "anti-chlor ;" 2. Sulphuric acid, from acid alums used in the sizing. Free acids are exceedingly injurious to the paper, producing gradual deterioration in the breaking strength, and also produc- ing brittleness. The amount of chlorides can be determined as follows : Take 0.5 gram of the paper, cut into small portions,, and digest with 50 cc. of boiling distilled water for two minutes, then filter. The filtrate is acidified with a few drops of nitric acid, and the amount of chlorine determined by a tenth-normal silver nitrate solution. The free sulphuric acid determination requires the determina- tion of the combined sulphuric acid in the alum, since in the titra- ion with soda solution the combined acid, as well as the free, is indicated. The combined acid is determined indirectly and then subtracted from the total acid, the difference being the free acid, thus : If the alum used is potash alum, the percentage of potash should be determined, and then the amount of sulphuric acid and alumina calculated from the formula of the alum (an- hydrous), K2AI, (SO,)^. If soda or ammonia alum be used, the determination of the soda, or ammonia, will be required. Where no clay has been used in the paper, the aluminum can be determined^ instead of the other base, and the sulphuric acid necessary to form the alum 1 J. Soc. Chem. Ind., 8, 564. 2 Basic aluminum sulphate forms an exception. Ferguson:/. Am. Chem. Soc, 16. 153, 552 ENGINEERING CHEMISTRY calculated; this latter is then deducted from the total acid. Total acid is thus determined : Two grams of the paper are cut into small pieces and digested with 2CX) cc. of boiling distilled water for three minutes, then filtered and a few drops of solution of litmus added. A solution of tenth-normal soda is gradually added from a burette, until the red color of the solution turns to blue, when the amount of alkali used is noted and calculated to sulphuric acid. From the total amount of sulphuric acid is subtracted the com- bined sulphuric acid already determined in 2 grams of paper. This latter amount is found by determination of any of the bases, alumina, potash, soda, or ammonia, and calculation of the required acid necessary to form the alum used in the paper. If aluminum sulphate, AL (804)3, be used instead of alum, then the free acid and combined acid will be the same in amount, since aluminum sulphate is an acid salt, and titration with the soda solution will give the amount directly. 4. Determination of the Nature and Amount of Sizing. A paper sized with rosin, when extracted with absolute alcohol, gives a solution which, poured into excess of water, yields a milky turbidity due to precipitated rosin.^ Another test is based on the Raspail reaction, rosin giving, with sugar solution and sulphuric acid, a violet-red color. The sugar may be omitted, as enough is formed for the reaction by the action of the sulphuric acid on the cellulose of the paper. The presence of animal size is detected by treating the aqueous extract of the paper with tannin. The following fundamental distinction between papers sized with rosin and gelatin is found to exist. In the former the rosin is distributed uniformly throughout the substance of the paper, while in the latter, whether the sizing has been performed in the pulp or sheet, it is always found exclusively on the surface of the finished product. This peculiar property of gelatin can be shown by saturating a plaster- of-Paris slab with gelatin solution colored suitably, and breaking it when dry, on which it will be found to be colored to a trifling 1 W. Hertzberg: Mitt. Konig. tech. Versuchs, 3, 107. J. Soc. Chem. Ind., 9. 99. I ENGINEERING CHEMISTRY 553 depth, the inner part being white. On these facts the following test is based : A half-sheet of paper is repeatedly crumpled and unfolded and when the surface has been thoroughly chafed, is smoothed out and written upon; if it is sized with rosin, the in- scribed characters are but little blurred; while, if animal size has been used they run freely, and are visible from the opposite side of the sheet. J^eonhardi has modified this test, removing the doubtful element introduced by the manual use of pen and ink. A pipette, of which the exit is lo centimeters above the paper, and which delivers drops weighing 0.03 gram each, is filled with a solution of ferric chloride containing 1.531 per cent, of iron. A single drop is allowed to fall and to remain on the paper for the same number of seconds that i square meter of the paper weighs in grams, when it is removed by blotting paper, and the under side of the paper brought in contact with a plug of wadding wet with a weak solution of tannin; the production of a black color proves the iron solution to have penetrated, and, therefore, shows the sizing to be of animal origin. Schuman's method for the determination of rosin in paper is as follows : Two grams of the paper are cut into fine pieces and digested below boiling fifteen minutes with a 5 per cent, solution of sodium hydroxide and filtered. The filtrate is made acid with dilute sulphuric acid, the rosin separating and rising to the surface of the liquid. This latter is filtered upon a weighed floor, dried at 100° C. to constant weight, and its weight carefully determined. Starch was used, formerly, as a sizing for paper, but in recent .years it has been largely replaced by rosin size. It can be de- tected as follows : The paper is cut into small portions and is digested with boil- ing water for fifteen minutes, then filtered. To the filtrate is added a drop of a dilute solution of iodine. A blue coloration is indicative of the presence of starch. The quantitative determination is dependent upon the conver- sion of starch into glucose by means of dilute sulphuric acid, and estimation by means of Fehling's solution. 554 ENGINEIERING CHEMISTRY From lo to 15 grams of the paper are digested with 250 cc. of distilled water, to which has been added 2 per cent, of sulphuric acid. Two or three hours' heating at 100° C. is sufficient to con- vert the starch into glucose, the exact point being determined by taking a drop of the solution and adding thereto one drop of the dilute iodine; if no blue color is shown, the conversion is com- plete. The solution is now made alkaline with soda, diluted with water to 500 cc, and two samples each of 150 cc. taken, filtered, washed well and treated with Fehling's solution,^ as usual in the determination of sugars. Sadtler states as follows regarding this test: "In carrying out the gravimetric method the Fehling's solu- tion remains in excess (indicated by the blue color of the solu- tion after boiling), while the cuprous oxide is carefully filtered off and further treated." The procedure is as follows :^ "Sixty cc. of the mixed Fehling's solution and 30 cc. of water are boiled in a beaker, and the solution containing the maltose added thereto and the mixture again boiled. It is then filtered with the aid of a filter pump, upon a Soxhlet filter (asbestos layer in a tared funnel of narrow cylindrical shape), quickly washed with hot water, and then with alcohol and ether, and dried. The asbestos filter, with the cuprous oxide, are now heated with a small flame, while a current of hydrogen is passed into the funnel, so that the precipitate is reduced to metallic copper. It is allowed to cool in the current of hydrogen, placed for a few minutes over sulphuric acid, and then weighed." 5. Determination of the Ash. Three grams of the paper are transferred to a weighed plati- num crucible and ignited until all carbonaceous matter is con- sumed. The amount of ash is indicative of the use, or not, of i Tqjlen's formula for Fehling's Solution is as follows: 34.639 grams crystallized cop- per sulphate are dissolved in 500 cc. water. 173 grams Rochelle salts and 60 grams sodium hydroxide are dissolved together in 500 cc of water. Equal volumes of these solutions are mixed when required for use. Ten cc. of this Fehling's solution corresponds to 0.0807 gram maltose— or 0.0765 gram starch. 2 Sadtler's " Industrial Organic Chemistry," p. 152. ENGINEJERING CHEiMISTRY 555 mineral filling, such as Carolina kaolin, to increase the weight of the paper. After the correct determination of the amount of the ash, it should be transferred to a 3-inch porcelain capsule, and the scheme on page 433 used for its analysis. It is always advisable to test some of the ash, before its analysis, by fusing a portion on charcoal with sodium carbonate. By this means, lead or chromium can be detected, and then prop- erly separated in the analysis of another portion of the ash. If clay in appreciable quantities, is found, it will be necessary to add 10 per cent, of its weight as water, since most clays contain from 8 to 12 per cent, of water, which, in the above instance, would have been driven off during ignition of the paper to deter- mine the per cent, of ash. If much iron be found, Prussian blue, Indian red, Venetian red, or ochre may have been used. If the color of the ash is blue, ultramarine is present; if white, silica, or a fine quality of clay, or calcium sulphate, or agalite^ may be present — the chemical analysis readily showing the one used as a filler. Ash in Commerciai, Pulps. Per cent. Sulphite 0.48 Sulphite, bleached 0.42 Soda 1.34 Soda, bleached 1.40 Straw 2.30 Straw, bleached 1.34 Ground wood (pine) 0.43 Ground wood (fir) 0.70 Ground wood (aspen) 0.44 Ground wood (lime) 0.40 Linen 0.76 Linen, bleached 0.94 Cotton 0.41 Cotton, bleached 0.76 1 A variety of talc— silicate of magnesium— in a finely powdered condition ; it has a very extensive use as paper filler. 556 ENGINEERING CHEMISTRY Ash in Fibers. Per cent. Cotton 0.12 Italian hemp 0.82 Rhea 5.63 Best Manila hemp 1.02 Sulphite fiber 0.46 Fine Flemish flax 0.70 China grass 2.87 Jute 1.32 Esparto 3.50-5.04 Soda fiber 1.00-2.50 . If the ash found is very small in amount, it will be necessary to subtract the amount of ash corresponding to the variety of liber pulp with which the paper is made, to exactly determine the amount of ash belonging to the added materials. 6. Determination of the Weight per Square Meter. It is best to use, when possible, 5 different pieces of the paper (from different packages or rolls), each piece about i square decimeter. These are placed in a drying oven and exposed to a tempera- ture of 105° C. until the weight becomes constant. The weight of the five pieces, multiplied by 20, gives the weight of i square meter of paper.^ 7. Determination of the Thickness. The thickness of paper can be accurately determined by means of any delicate micrometer screw. 8. Determination of Breaking Strength. By the strength of a paper is understood the measurement of the resistance it offers to breaking or tearing strains. This re- sistance is always greater in the direction of the length of the web of paper, as it is made on the paper-machine, than across the web. On the other hand, the amount of elongation, which is measured while determining the breaking strain, is greater in the direction across the web than parallel it.^ The tensile strength of the sheet, both across and parallel to the web, is de- termined separately, and the average values recorded. To ascer- 1 lycitfaden fiir Papier-priifung, W. Herzberg, Berlin, 1888. 2 Verhandlung des Vereines zur Beforderung des Gewerbefleisses in Preussen, 1885. ENGINEERING CHEMISTRY 557 tain the direction corresponding to the motion of the paper- machine, in any sample of machine-made paper, a circular piece is cut and placed on the surface of water, when it will be ob- served to roll up. The diameter of the disc where it is not curved indicates the direction of the length of the web. The strips of paper used for ascertaining the tensile strength and elongation are cut to the following size: i8o millimeters long by 15 milli- meters broad. Five strips, at least, are taken from different sheets and representing the length and across the web, in order to obtain good average values. These strips must be carefully cut; the edges should be smooth and run parallel. Cutting tools are provided for this purpose, consisting of an iron ruler and plates of zinc or glass. Before determining the tensile strength and elongation, careful attention must be paid to the amount of moisture in the atmos- phere. The breaking strain of paper decreases with increase of moisture in the air, while under the same influence the percen- tage amount of elongation increases. The humidity of the atmos- phere is very important when testing animal-sized paper and should on no account be overlooked. Indeed, the breaking strain values can only be compared when they are obtained in atmos- pheres of equal humidity. The percentage of atmospheric humidity chosen is 65, because it is much easier to add moisture to the atmosphere than abstract moisture from it. The former is done by boiling water in the room. The instrument in use for measuring the humidity of the air is the Koppe-Saussure's air hydrometer. Before testing, the strips of paper are placed in the room for at least two hours. The principal machines in use for determining the breaking strength of paper are: The Hartig-Reusch, the Wendler and the Chopper Apparatus, a description of the Wendler being given herewith. This ma- chine is used for ascertaining the strength and elasticity of paper. It consists in the main of four parts (Fig. loi). 1. The driver. 2. Apparatus for mounting. 3. Apparatus for transmission of power. 4. Apparatus for measuring force and stretch. 558 engine:e:ring chemistry :§ K < 5 >^ w ^ c> bO s 3 >> Z^ b Tl o ^ a> 'O •*-• a 2 rt o •• Oh > o; 10 'O ■-' a +-» CO a ^ , O 8 52; w f 5 •*-> cu HH 3 ~ rrt ffi M3 (U to ^ M rt ^ 6 •§ -§ »> (Tl vn a ,> •r^ =*, 2 a ^" Q c u ■3.2- T3g« 4; ■111* a 5 "^ a 0. j: ^- c o !* o Q u O V.' a u o X iJ a
  • V ..- . . -^ K/i"^ 1— ■ ;«33^«^«-a, -5,60-30 1 ^ ^ -n tn O — E' S" ^ 1^ « ^ c8 S,r^ cc 3 (s lU n^ 9 10 10 4 22 10 21 5 loi^ 7 3 6 9 8 9>^ 9 -lYi 10 Ij^ 10 12 23 12 22 15 7 8 6 W/2 9 9 9K2 10 ^Vx II 4^ 24 14 23 6 3 7 135^ 7 ZV2 9 1V2 10 oYz II I II I2J^ 26 24 6 IV2 8 3 7 8^ 9 13 10 iVi II 8^ 12 i,V2 27 2 25 6 12 8 8J^ 7 I3'/2 IQ 4 10 14% 12 12 12^ 28 % 26 7 8 14 8 2^ 10 II II 5% 12 8 13 4^ 29 27 7 4^ 9 3K2 8 7^ II I II 12^ 13 13 I2J4 30 8 28 7 8J4 9 9H 8 12^2 II 8 12 3^ 13 8 14 4^ 31 10 29 13 9 I3J/2 9 1^ II 14 12 10% 14 14 12^ 32 12 30 8 I 10 4 9 6'.^ 12 AM 13 i^ 14 65^ 15 55^ 33 14 31 8 bV2 10 9 9 11^ 12 II »3 8J^ 14 W/2 •15 14 35 32 8 10 10 14 10 0^ 13 2 13 1554 15 6J4 16 6 36 2 33 8 15 II 4 10 6 13 9 14 6'^ 15 14^2 16 14 37 4 34 9 2 II 10 10 II 13 15 14 J3J^ 16 6 17 6 38 6 35 9 7 II 15 11 14 6 15 5 16 14 17 14^ 39 9 36 9 II 12 4J^ n 5 14 12 15 12 17 6 18 6^ 40 II 37 9 15^ 12 10 II 10 15 3 16 3 17 1 18 15 41 13 38 10 1% 13 II 15 15 10 16 10 18 19 7 43 39 10 8 13 5 12 4 16 17 I 18 14 19 15 44 2 40 10 12 '3 II 12 9 16 6 17 7 19 4 20 8 ^\ 4 41 II 0% 14 12 14 16 13 17 4 19 II 21 46 6 42 II 5 14 5/2 13 2 17 4 18 5 20 3 21 8 47 8 43 II 9 14 II 13 7 17 10 18 12 20 10 22 48 10 44 II nV2 15 I 13 12 18 19 3 21 2 22 8 49 12 45 12 2 15 6H 14 I 18 6 19 10 21 10 23 I 50 14 46 12 6 15 Il'/2 14 6 18 13 20 I 22 I 23 9 52 47 12 10% 16 o^^ 14 II 19 3 20 8 22 9 24 I 53 4 48 12 14^ 16 6 15 19 10 20 15 23 I 24 9 54 6 49 13 3 II iiH 15 5 20 21 6 23 8 25 I 55 8 50 13 IV2 17 I 15 10 20 7^ 21 13 24 I 25 10 56 10 For 500 sheets per ream, multiply the weight by 1.041 {i. e., to every lb. add 0.66 oz.) For 516 sheets per ream, multiply the weight by 1.075 C^* to every lb. add 1.2 oz.) The mechanical tests of the paper are extended also to its thickness. For this purpose the well-known instrument called the Palmer screw gauge is employed. This gives hundredths of a millimeter. Different experiments are made with packages of 568 ENGINEERING CHEMISTRY sheets varying as to number and an average is taken of the tests expected. The weight of the paper to the square foot is also determined. This experiment is made in order to ascertain how many sheets to lOO pounds will be obtained, or rather to find what weight of paper it will require to cover a definite surface. To make such a test it sufiices to measure a sheet of paper accurately and afterward weigh it. The following experiment to which the paper is submitted serves to determine the resistance to folding, crumpling, and crushing. This is very important, since certain papers, like those used in the manufacture of bank notes, etc., pass through many hands. It is therefore necessary before printing to be exactly informed as to the resistance that the paper will present to use. Different apparatus have been devised for such a test, but they have not as yet entered into the domain of practice. The most usual method is to operate by hand. The sheet is first folded in one direction, then in a direction at right angles, and then diag- onally, and finally in a direction at right angles with the diagonal. The specimen that exhibits an aperture at the first folding is con- sidered very bad and that which resists the four foldings may be very good. In order to prove its real resistance it must be sub- mitted to new tests. The specimen is formed into a ball and is compressed. Then a second and a third ball is made of it. If the paper has resisted this test it is placed between the two hands and submitted to friction. The majority of papers do not resist such tests. It has been found that there exists a very sensible correlation between the resistance of the paper to traction and its resistance to friction. The tests are made in the Bureau that the Chamber of Com- merce posesses on Rue de Viannes, and are accessible to the public. Accurate information as to the respective qualities of the samples may be obtained by any one upon payment of a small sum of money. La Nature, 1901. References. "C. B. S. Units and Standard Paper Tests," an essay towards establish- ing a normal system of paper testing, by C. F. Cross, E. J. Bevan, Clayton Beadle and R. W. Sindall, London, 1903. e:ngine;e:ring chemistry 569 "Berichte der Papierpriifungs-Aiistalt," von Winkler, Leipzig. — Papier Zcit, 23, 1 131. "Mittheilungen aiis den Koniglichen technischen Verstichs-austalten zu Berlin," 1891-1905. "The Art of Paper Making," by Alex. Watt, 1890. "Paper Making," by Clayton Beadle, London, 1908. "Papierpriifimg," by W. Herzberg, Berlin, 1902. "A Text Book on Paper Making," by Cross and Bevan, London, 1900. "The Textile Fibers : Their Physical, Microscopical and Chemical Prop- ties," by J. M. Matthews, New York, 1905. "Paper Technology," by D. W. Sindall, London, 1905. 570 ENGINEERING CHEMISTRY WATER ANALYSIS. Scheme for Water Analysis for Scale Forming Ingredients. Evaporate i liter of the water in a weighed platinum capsule, upon a water-bath to dryness ; i transfer to a hot air-bath and heat at io5°C for thirty minutes; cool and weigh. Ignite slowly to a dull red heat untii all carbonaceous matter is consumed; cool and weigh. The loss of weight equals organic and volatile matter. Warm the contents of the capsule with lo to 15 cc. hydrochloric acid, and 25 cc. water, boil and filter through an ashless filter into 100 cc. graduated flask; wash thoroughly, bringing contents of flask to containing mark with water; mix well. <1) Residue Consists of i n s o 1 uble mineral matter— Si O2 or Si02.Al203. (CaSO*.) (3) Residue Consists of AlgO^Feg O3 dry. Ignite, and weigh as such Si02. (2) Solution. 100 cc. Divide into two portions, one of 75 cc. for bases and of 25 cc. for 6O3. 75 cc. Make alkaline with NH4OH, boil and filter (all weights obtained to be divided by 3 and multiplied by 4). AI2O, FejOs (4) Filtrate Add solution of ammonium oxalate : set aside three hours ; then filter. CaO. (5) Residue (6) Filtrate. Consists of Acidify with dilute C a C2 O4. H2SO4. Evaporate to Dry, ignite, dryness in weighed t and weigh platinum dish;ignite as CaO. to expel all ammo-ter nium salts; cool and weigh, (MgS04 + Na2S04). Dissolve in water, make alka- line with NH4OH, then acid with HCI, then alkaline with NH4OH, add sHgl.t excess of Na2HP04 solution, with con- stant stirring for two minutes, set aside fifteen minutes, fil- ter, wash with water containing Vr NH4OH. dry. ignite, and weigh as MgoPaO:. Convert this' weight to MgSO* and subtract from weight of MgS04 + NA2SO4 in (6): the difference will be NaoS04. C o n V e rt weights to MgO and NaaO. SO3. 25 CO.; Warm, add solution of b a r i u n ch 1 o r i d e and allow to settle h r e e hours; fil- wash, dry, ignite, and weigh as BaS04. Calculate to SO3 and multiply result by 4. MgO NajO. CO2 is found by combin- i n g the ch lorin e nd sul- phuric acid with the bases, thence; ad determin- i ng ho w much COo is required to convert the rest of the CaO and MgO to carbonates; a^ shown in the ex- ample giv en below. SO3 01 Concen. t r a t e 250 cc. of the water in a porcelain dish to about , i> t^T3 a; o ^ >« t/2 O t- O B a ctt o C rt _ 3 « o ""•s:'^ 3.S «5 JJ oi d' (LI I' r: »' > n <*^ 2 ^ o4;5?TO S rt ^ .ti g- be o rt V-.-y-r"i03 « HvCPbfiO ^ o cr ■ — - .^ .^ 01 » <^ D 3 wQ h-o 8 o o a :-T3.= i" n ?" «? >>0 CO. -a B'/r,--^ II Ov.- O o a w a:i2ii'^•- -S'Oj'S « a bi^s^'^'^i t 0^ P CS "^ ^- fr9 <" P P « ao ~ c a o2 21-' ^■ou rt^iSi- rt 'oJ.^Sf g O W) bc^' ^2 --^^a'^O . a-O a. II .'O •5 u « ^ > i: X o 2 « fa ■« P 5S5,boj-- . .^ r Qfi CO ™ '*P bC--cjO ^^03 , S^^ K o u a « c8 xJS'O cbU o« d i « 5 > 1
  • U2^ 300,000 300,000 200,000 200,000 300,000 200,000 200,000 20 20 20 20 20 20 20 1- P e o '-' 5 21.87 32.86 28.08 18.72 25.27 18.79 26.07 1^ ^ 3*.. D«*H h O > a; *J tn tn ii tfl » O »3.Q.EI-M 937.2 1,408.2 802.2 534.8 1,083.0 536.8 744.8 Totals 1,700,000 gall< 6,047.81b Average cost of treatment per 1,000 gallons, 1.8 cents. FEED-WATER HEATERS. Feed-water heaters as well as boiler economizers are often used as eliminators of the scale-forming materials in water. This is due to the fact that waters containing much calcium and mag- nesium carbonates when heated to the usual temperature in feed- water heaters (20o°-2io° F.), give up the excess of carbon diox- ide that holds the calcium and magnesium carbonates in solution, and the latter are precipitated and removed before the . water enters the boiler. Where calcium sulphate is a large constituent of the water, feed-water heaters using exhaust steam do not precipitate the lime salt, but if the feed-water be heated by live steam under pressure to a temperature of 240° F., then the calcium sulphate begins to precipitate. The addition of the water before it en- ters the heater of the necessary amount of sodium carbonate will precipitate the lime as carbonate, at ordinary temperatures if several hours are allowed for sedimentation, or if heat can be used the chemical action will be hastened. An example of an upright closed feed-water heater heated by exhaust steam is the "Goubert." 6l2 ENGINEERING CHEMISTRY The exhaust steam from the engine is admitted to the shell through the nozzle on one side, and spreading between the brass tubes, impinges upon them on its passage across to the outlet on the opposite side, while the aggregate area of the spaces between Fig. 113. — The Goubert closed feed-water heater, vertical type. the tubes is so much better than that of the exhaust pipe that no obstruction is offered to the flow of the stream, and absolutely no back pressure reverts upon the engine. The water condensation is removed by the drip pipe, which e;ngine;h;ring chemistry 613 should be kept always open, and it is a peculiarity of the con- struction of this heater that the oil or grease in the steam is almost entirely removed and passes off with the drip, leaving the remainder of the exhaust free from contamination and avail- able for other purposes for which live steam has ordinarily to be used. The cold feed-water enters at the bottom of the apparatus, is spread by the deflector, and, passing under the edge of the lat- ter in a thin sheet, allows the particles of mud or sediment it carries to settle, undisturbed, in the bottom of the water-chamber, there being no heat at this point, and consequently no circulation. The water then flows upward through the tubes, and being divided up in small streams becomes heated quickly ; as each tube is surrounded by steam no heat is lost by radiation before the water leaves the heater, a result that some makers of steam tube heaters have endeavored to attain by surrounding the shell with a steam jacket. The construction of the upper water-chamber, similar to that of the lower one, permits the rise of scum to the top and its subsequent removal from the surface below. A mud blow-off pipe is also provided in the bottom chamber. The Goubert Feed-Water Heater is particularly easy to clean. By lifting the top chamber the ends of all the tubes are exposed; a swab or brush may then be used to clean the tubes. This, how- ever, needs to be done but rarely, and if the surface and mud blows be open for a few seconds ever3^-day, the heater is readily kept clean and very little sediment is ever found to adhere to the interior surfaces of the tubes. By leaving the blow-off valve open at night, or when not in use, the heater can be thoroughly drained to avoid the danger of freezing in cold weather. The various forms of open feed-water heaters are worked upon the same principle of collecting the scale- forming material from the water before it enters the boilers. A sample of hard water was submitted to the writer for analysis and from the analysis to determine the best method 6i4 Engine:e:ring che:mistry of treatment for purification of the water before it entered the boilers. The analysis of the residue, dried at 212° F., of the untreated water was as follows : Untreated Water. Parts per Grains per millon. U. S. gallon Si02 12. 0.69 SO3 65. 3.77 CI 62.5 3.62 NazO 53.1 3.07 MgO 36.0 2.08 CaO 122. 7.07 AUOsFe^O. 6.6 0.38 CO2 102.4 5.98 Organic 20.4 1.18 Total 480. 27.84 Total hardness = 20.03°. T lb, 12 ozs. of lime (CaO) 80 per cent., and 10 ozs. of soda ash, (96 per cent.) were used for each 1,000 gallons of the water. After allowing the chemicals to act upon the water 12 hours in settling tanks, the water passed into the feed-water heater. Here more precipitation of the incrustating solids took place, showing that even long sedimentation under cold treatment did not suffice to reduce sludge and scale forming solids to a minimum. An analysis of this deposit being as follows : Anai.ysis oe the Deposit prom Heater. Per cent. Moisture (212° F.) 3.14 Water of hydration 15.92 Oir 10.34 Organic matter 4.15 CO. 12.01 SiO. 15.93 CaO 19.34 MgO 18.60 AloOs.FcaOs 0.48 Undetermined 0.09 Total 100.00 ^ From condensed steam returned to heater — showing that if any oil-separator was used it was not effective. ENGINEERING CHEMISTRY 615 AXAI.YS1S OF Treated Water, After Passing the Heater. Parts per Grains per million U. S. gallon SiOs 9. 0.52 SO3 64.6 3.74 CI 62.5 3.62 Na^O 136.5 7.91 MgO 16.9 0.99 CaO 19.3 I. II Al2O3.Fe.O3 2.6 0.15 CO2 59.9 3.49 Organic 9.7 0.56 Total 381. 22.09 Total hardness — 3.4°. Thus it will be seen that the amount of scale-forming material in the untreated water (20°) is reduced to 3.4° in the water as it enters the boilers — (soft water). Much better results are fre- quent. Independent of the fact that feed-water heaters are more or less removers of scale-forming materials, in the original water, they also act as fuel economizers. In the above sample the water was purified by use of chemicals and sedimentation in tanks before passing into the feed-water heaters. Improvements have been made which can best be shown by a description of one of these methods — the Sorge-Cochrane hot process feed-water softener. This process of softening of water for boiler use, may be re- garded as an extension of the Cochrane feed-water heater, in which water sprayed over a series of baffle plates is heated by immediate contact with the exhaust steam, the latter having previously been purified of oil by passing through an oil sep- arator attached to and forming a part of the heater (Fig. 115). The temperature obtained in the heater (210° F., or higher, depending upon the back pressure), has the effect of driving out air or other gases from the water, such as carbon dioxide, including carbon dioxide in combination as well as in solution. In the case of water containing salts of lime, this results in the changing of the soluble bi-carbonates to the insoluble normal 6i6 ENGINEERING CHEMISTRY Fig. 114.— Represents an interior view of a Cochrane feed-water heater. ENGINEERING CHEMISTRY 617 carbonates. These substances, therefore, precipitate and adhere to the trays over which the water flows or settle in the sedimenta- tion chamber, or are arrested by the filter in the lower part of the latter. Fig. 115.— The oil separator and drain. The hot process system extends the treatment to include the modification of sulphates, carbonates, nitrates and acids. In all cases where carbonates are present in any considerable quantity some chemical means are used for bringing about their complete transformation and removal. In large plants this reagent is usually milk of lime, and in small plants it may be either milk of lime or sodium hydrate. In a certain number of cases, sodium hydrate is the only chemical required, as the sodium carbonate resulting from the action of the hydrate with the lime and mag- nesium carbonate is about sufiicient to take care of sulphates, chlorides and nitrates that may be present. The reactions take place much more quickly in a hot solution than in a cold one, so that less storage and settling capacity are required in a hot pro- cess system than in a cold process system. The same fact is illustrated by the practice with some cold process of heating the water in order to hasten and complete the reaction. Combining the action of the water with the chemical treat- ment has two important advantages from the point of view of softening water for boiler feeding: 1st, it saves the cost of reagents for precipitating. 6l8 ENGINEERING CHEMISTRY 2nd, the same apparatus performs the duties of both softening system and open feed-water heater. The open heater may also act as a hot well for condensed returns from heating or drying systems, or for condensate from surface condensers, or in fact, for any other water about a plant suitable for boiler feeding. This is an advantage, not only from the point of view that it simplifies and reduces the cost of the apparatus, but also because the utilization of condensed steam reduces by that much the amount of raw water to be treated. Treating with one chemical instead of two, where practicable, is also a great advantage from the point of view of practical opera- tion, for while it is a comparatively complicated matter, requiring some special knowledge and skill, to analyze water for its several constituents and to proportion the feed of two reagents in accord- ance therewith, it is a comparatively simple matter to determine whether or not the feed of a single reagent has been sufficient to neutralize the substance with which it reacts. Fig. ii6 illustrates the apparatus. The water enters through the pipe marked ''Water Supply" and empties into a dis- tributing trough in the upper part of the upright section. From this trough it overflows upon alternately inclined trays, finally dropping into the settling chamber below. During this time it is surrounded on all sides and mingled with exhaust steam which has entered through the oil separator on the left. Air and gases are driven out of the water and any surplus of steam escapes through the vertical outlet pipe at the top. The chemical reagent is intoduced into the cold water supply pipe in various ways. The latest practice provides an automatic device con- trolled by the rate at which raw water enters a pump drive and the chemicals to this device from a tank where the materials are kept in suspension by a power agitator. A dilute solution is em- ployed, that is, not near the saturation point, as it does not give trouble from clogging the valves, which concentrated solutions sometimes do. The size of the settling chamber is determined more or less by experience with waters like the one which is to be treated, more time being required for the settling of some pre- cipitates than for others. The settling tank has a cylindrical ENGINEERING CHEMISTRY 619 steel plate tank, containing an inverted cone. As the treated water falls from the trays it is mingled with the reagent, passes down through the center of the cone to a filter chamber which is placed between the heater chamber and the sedimentation and reaction tank. Above the filter beds is the pump supply chamber, and in this chamber is a copper float, controlling, through a sys- tem of levers and rods, a valve in the cold water supply pipe, so that when the pumps draw water from this chamber an equiva- lent amount of water is admitted to the distribution trough for treatment. Fig. 116.— Cochrane metering heater with outside chamber for recorder float. In order to provide for a supply of water to the pumps in case the filter should become clogged up, a supplementary over- flow is used which can be seen attached to the vertical partition. The action of this is as follows : When the water level in the pump supply chamber is drawn down by the pumps, the float opens the cold water regulating valve and the water consequently flows into the settling chamber at the left. If the filter does not allow the water to pass though at a corresponding rate, it will rise in this chamber until it reaches the supplementary by-pass and overflows directly into the pump supply chamber, thus insuring for the pumps hot, treated and settled water, no matter to what extent the filter may be neglected. 620 ENGINEERIxNG CHEMISTRY Another feature of the apparatus is the overflow trap attached at the left end. Where the apparatus is working under back pressure, as in connection with an exhaust steam heating or dry- ing system, there is a trap as here shown. If, however, the ex- haust outlet from the apparatus is open to atmosphere, a water seal will answer the purpose, which is to drain the oily emulsion separated from the steam in the oil separator, to drain the waste and to dispose of the overflow from the heater itself, while pre- venting free escape of steam. In fact, it is customary to over- flow the heater for a short period each day by holding open the cold water valve until the water level in the settling chamber rises to the edge of the overflow weir, this disposing of any scum or other impurities floating on the surface. Summary ok Purifying Results Obtained in the Hot Process System. Substance in Feed Water Calcium bicarbonate, CaiHCOgJa- Calcium sulphate, CaSOi Calcium chloride, CaClj^ Calcium nitrate, Ca( N03)2 Magnesium bicarbonate, Mg(HC03)2- Magnesium sulphate, Mg(S04' Magnesium chloride, MgCla • Iron bicarbonate, Fe(HC03)2 Silica, SiOa Clay, HgAlaSijOg • •• Mineral acids Carbonic acid Hydrogen sulphide. Air Organic and Oily acids- • Trouble in Boiler Soft Scale Hard Scale Indirectly may cause corrosion Corrosion Soft scale and foaming Indirectly may cause scale Corrosion' Sludge Sludge Sludge Corrosion Corrosion Corrosion Corrosion Corrosion Remedied by Caustic soda or lime Heat and soda ash Soda ash Soda ash Caustic soda or lime Heat and soda ash Soda ash Contact with air and heat Filter Filter Soda Ash Heat Heat Heat Soda ash 1937. References—" How Should Feed Water be Heated? " by A. J. Albright, Power, July "The Paterson Oil Eliminator and Water Softener," Engineering, Dec. 21. 1906. ENGINEERING CHEMISTRY 621 Trouble Due to Water: Prevention and Cure^ Trouble INCRUSTATION Cause. Cure f Sediment, mud, clay, etc. Filtration I Readily soluble salts Blowing-off { Heating feed and precip-^ itate Caustic soda \ Bicarbonate of magnesia, j j Lime, iron Lime I ^ Magnesia Organic matter Sulphate of lime See below \ Sodium. Carbonate \ Barium chloride Corrosion f Organic matter^ Grease Chloride or sulphate \ magnesium . Sugars Acid Precip. with alum " Precip. with ferric chloride Slaked lime Sodium Carbonate of and filter and filter Sodium, Carbonate Sodium Cafbonate Dissolved carbonic and oxygen^ I Electrolytic action^ acid Slaked lime Caustic soda Heating Zinc plates Priming f Sewage ' Alkalies I Carbonate of soda y large quantities Precipitate with alum or ferric chloride and filter Heating Feed cipitate and Pre- Barium chloride 1 Compiled by Prof S M Noiton 3 Organic acids are neutralized by soda ash.— Ed. ^ Note.— Recent investigations have shown that electrolytic corrosion in aqueous solu- tions is generally conditional upon the presence of oxygen The latter is expelled from the water by heating in the Cochrane heater.— Ed. 612 Engine:ering chemistry H. De La Coux/ states : "Sodium silicate which has been proposed as a scale preventative is transformed in the presence of the carbonate of calcium in the water into silicate of calcium and falls down in a white gelatinous precipitate : — NaoSi03 + Ca(HC03), = CaSiO^ + CO, + H^O + Na^COa- With ordinary waters scale can be prevented by the addition of 600 grams of silicate of sodium solution of 35° B. per horse- power per month." Table Showing the Yearly Saving Effected by the Use of the Feed-Water Heater for Various Horse-Powers and at Different Prices of Coal Coal consump- Horse power of tion at 4 pounds per H. P. per hour Saving ofi3^ per cent. Price of coal per ton of 2,240 pounds engine Daily Yearly $1.50 $2.00 $2.50 $5.00 $3-50 $4.00 $4.50 15.00 $5-50 $6.00 Lbs. Tons Tons 50 2000 268 36.18 1 54 $ 72 $ 90 |io8 $126 M45 I163 |i8i $199 $217 60 2400 321 43-33 65 87 108 128 152 173 194 217 238 260 70 2800 375 50.62 76 lOI 126 152 177 202 227 253 278 304 80 3200 429 57.91 . 871 116 145 177 203 232 261 289 318 347 100 4000 536 72.36 108 145 187 217 253 289 325 362 398 434 120 4800 643 86.80 130! 174 217 260 304 347 390 434 477 521 160 6400 857 115.69 173! 231 289 347 404 463 520 578 635 694 200 8000 1072 144.72 217 289 362 434 506 579 651 724 796 868 250 1 0000 1340 185.90 279 372 465 558 651 744 837 929 1022 1115 300 12000 1608 226.08 339 452 565 678 791 904 1017 1 130 1243 1356 350 14000 1876 253-26 380! 506 633 760 886 1013 1139 1266 1392 1.519 400 16000 2144 289.44 4341 579 723 868 1013 1158 1302 1447 1591 1730 500 20000 2680 361.80 543 724 904 1085 1267 1447 1627 1809 1990 2170 600 24000 3216 433-30 650 867 1083 1300 1517 1733 1950 2170 2387 2600 700 28000 3752 506.20 759 1012 1265 1518 1771 2025 2278 2531 2784 3037 800 32000 4288 579-10 868 1158 1448 1737 2026 2316 2605 2895 3184 3474 900 36000 4824 651.24 977 1302 1628 1954 2279 2605 2930 3256 3581 3907 1000 40000 5360 723.60 1085 1447 1809 2170 2532 28943255 3618 3990; 4341 'Industrial Uses of Water," p 49 ENGINEERING CHEMISTRY 623 Percentage of Fuei, Saved by Heating Feed Water (Steam pressure 60 pounds) I « s ^ s (U V. 1- ill '^ s * ill Temperature of water entering boiler %%t -•Si; 120° F 140° F 160° F iSqOF 200° F 202° F 204° F 206° F 208° F 210° F 212° F 214OF 216° F 32° 1175 7.49 9.19 10.89 12.59 14.30 14.47 14.64 14.81 14.98 15.15 15.32 15.49 15.66 40" 1 1 67 6.86 «-57 10.28 12.00 13.71 I3.88I14.O5 14.22 14.40 14.57 14.74 14.91 15.08 50^ 1157 6.05 7.7« 9-51 11.24 12.97 I3.I4I3.32 13.49 13.66 13.83 14.00 14.18 14.35 60" 1147 5.23 6.97 8.72 10.46 12.21 12.38 12.55 12.73 12.90 13.08 13.25 13.43 13.60 70" 1 137 4.41 6.16 7.91 9.67 11.43 II. 61 11.78 11.96 12.14 12.31 12.49 12.66 12.84 80" 1127 3.44 5-32 7.10 8.87 10.65 10.82 11.00 II. 18 11.36 11.53 II. 71 11.89 12.07 90° 1117 2.68 4-47 6.26 8.06 9«5 10.03 10.21 10.38 10.56 10.74 10.92 II. 10 11.28 iOO" 1 107 1.80 3.61 5.42 7.23 9-03 9.21 9-39 9-57 9-75 9.93 10. II 10.29 10.47 110° 1097 0.91 2.73 4.55 6.38 8.20 «.3« 8.58 8.74 8.93 9.II 9.29 9-47 9.66 120" 1087 1.84 3.67 5-51 7.35 7.54 7.77 7.90 8.09 8.27 8.45 8.64 8.82 FUEL ECONOMIZERS. A fuel economizer generally consists of a nest of vertical iron tubes arranged in the flue leading to the chimney and utilizing the otherwise wasted heat in the gases of combustion. It is able to recover low temperature heat that would escape or has escaped, a boiler, because of the fact that the average temperature of the water within the tubes of the economizer is much lower than the temperature of the water in a boiler. This fundamental principle of heating the feed water in a separate vessel apart from the boiler and thereby saving the heat in the waste gases passing to the chimney is the distinctive invention of Mr. Edward Green, who made his first experiments upon an apparatus for this purpose m 1845.^ The maximum possible saving by an economizer is based upon the assumption that the construction of the economizer unfits it for the generation of steam, but that it may be allowed to do all the work of heating the water up to the boiler evaporating tem- perature. In a certain plant tested, the boiler pressure was 95 lbs., the corresponding temperature of evaporation being 334° F. The temperature of the feed water entering the economizer 1 The Book of the Economizer, p. 5 Kngine;e:ring che:mistry 625 was 162.5° F., so that it was theoretically possible for the econ- omizer to contribute 334. minus 162.5 =^ i7i-5 British thermal units. The number of thermal units required for the actual evaporation of water at 95 lbs., is 887.9. The total work done on the water is therefore 887.9 plus 17 1.5 — 1049.4 heat units, of which 171. 5 constitutes 16.3 per cent., or the maximum pro- portion of the work that could be done by the economizer re- ceiving water at 162.5° F., which was the actual temperature, the water in this case coming from an open heater where it had been warmed by the exhaust of the pumps and certain other auxiliary apparatus. WITH HIGH pressure: AND SUPERHEAT. With a steam pressure of 200 lbs. and a feed-water temper- ature of 60° F. the case figures out as follows : The temperature of evaporation for 200 lbs. gauge pressure is 387.5° F. The latent heat of evaporation is 839 British thermal units. The heat that can be contributed by the economizer is 333.2 British thermal units. The economizer, therefore, can contribute 28.4 per cent, of the total amount of heat required to convert the water from 60° into steam at 200 lbs. or, looking at it another way, it can contribute 39.7 per cent, as much heat as does the boiler. This is the theoretical limit, since the economizer is not expected to make steam, in fact, it is provided with a safety valve so that in case it does make steam the latter can escape. It should be noted, however, that the higher the steam pres- sure, the greater the work of the economizer may be, and, on the other hand, the lower the efficiency of the boiler will be, if it be not supplemented by an economizer. The higher the steam pressure, the less is the average difference in temperature between the gases of combustion and the contents of the boiler, therefore the slower the transmission of heat. The temperature of a boiler carrying steam at 80 lbs. is 323.6° F., while that of one carrying 200 lbs. is 387.5°. The temper- ature of the boiler determines the lower limit to which the gases may be cooled before they pass to the chimney, since no matter how much the boiler surface is extended, the gases cannot- be 40 626 ENGINEERING CHEMISTRY cooled below its temperature. The higher the steam pressure the higher is the limit and the greater must be the amount of heat lost to the chimney. By the use of the economizer the temperature of the flue gases may be reduced to almost any ex- tent, having the temperature of the cold feed water as the lower limit. As usually proportioned, the temperature of the gases leaving the economizer is from 240° to 235° F. High steam pressure and superheat unquestionably save steam. It has been shown that a single cylinder non-condensing en- gine can be run on about 16 lbs. of superheated steam per indicated horse-power and a triple condensing engine on as low as 8.75 lbs. Steam turbines have been operated with less than 10 lbs. per brake horse-power, but this good steam economy does not mean good fuel economy if the flue gases are allowed to escape at a high temperature. M. Dejace reports that with a superheat of 473° F. a saving of 16.4 per cent, of steam was realized, but the saving of coal was only 6.2 per cent. Tests by G. H. Barrus on Curtis turbines, with 150° F. superheat at the throttle, show 10 per cent, of steam saved and less than i per cent, of coal saved. The saving of coal does not pay for the in- convenience of superheat. While comparatively little heat is required merely to superheat steam, perhaps not more than one- third as much as to pre-heat the boiler feed, much heat is usually wasted in the process of superheating, since the gases from which the steam receives its heat must be hotter than the steam itself. Starting with the temperature for 200-lb. pressure steam, that is, 387.5° F., suppose we add 5CX)° superheat, this gives a tem- perature of 887° F. and the gases leaving he superheater must still be hotter. If the superheater is placed beyond the boiler or is heated by a separate furnace there will necessarily be a great loss of heat. If the superheater is located between the "passes" of the boiler, an arrangement having some disadvantages, the final temperature of the gases may be lessened to a certain extent by the addition to the heating surface represented by the super- heater, but if the steam pressure is high the gases will still be very hot. The economizer, however, will recover all the waste heat resulting from high pressure or superheat, besides recovering ENGINEJERING CHE;mISTRY 627 enough more, as in ordinary plants, to pay for itself in from two to three years. The economizer makes the fuel saving correspond to the steam saving of the engine and a superheated steam plant is incomplete without it. Where superheat is to be used a good arrangement is as follows : the boiler should be operated at a high rate so that the gases will leave at a high temperature, then fol- lowing the boiler there should be the superheater and after the gases leave the superheater they should pass through an econo- GAS ANALYSIS. Analysis of Chimney Gases for Oxygen, Carbon Dioxide, Carbon Monoxide and Nitrogen. Revised by Thomas B. Stillman, Jr., M.S. The determinations usually made are the percentages, by volume, of oxygen, carbon dioxide, carbon monoxide, and nitro- gen. The apparatus used (a modified form of the Elliott) is shown in Fig. 118, and consists of two glass tubes, ib and ah, the tube ib having a capacity of about 125 cc. and is accurately graduated from o cc. to 100 cc. in o.i cc. At d and e are three-way glass stop-cocks, connected by means of rubber tubing in the water- supply bottles, / and g. The manipulation of the apparatus is as follows : Remove the funnel cap c, and connect in its place a glass tube of small diameter, but of sufficient length to each well into the flue from which the gases are to be taken. Open the stop-cocks a and b and slowly raise g and / until both tubes are full of water including the glass tube in the flue. It is necessary in this oper- ation to be certain that no air is in the tubes and that the dis- placement by water is complete. Now gradually lower the bottle / whereby the gas is drawn into the tube ah. As soon as suf- ficient gas has been obtained for the analysis, the lower por- tion of the tube containing water 2 or 3 inches above the point J The book of the Economizer, pp. 35-37. 628 Engine;ering chemistry h, the stop-cock a is closed, the small glass tube connecting a with the flue removed, and the funnel cap c replaced. After allowing the gas to stand in the tube ah fifteen minutes to acquire the temperature of the room, and thus insure correct measure- ments, the bottle g is slowly lowered until the surface of the I ENGINEERING CHEMISTRY 629 water therein is on an exact level with o on the tube ib, the stop- cock b opened and the bottle / gradually raised until sufficient gas from ah has been transferred to bi, indicated by the volume taken reading from the mark o on the graduated tube ib to the mark 100 cc. immediately in contact with the stop-cock, b. Having thus obtained 100 cc. of the gas, the stop-cock b is closed and / is raised until all the remaining gas in ah and ab is displaced by the water. The first constituent of the gas to be determined is carbon dioxide (CO,). The gas is now trans- ferred to the tube ah by raising g and opening b, keeping a closed and / lowered. When the water reaches b the latter is closed. Fifty cc. of a solution of caustic potash are placed in the funnel cap c. (The solution is made by dissolving 280 grams of potas- sium hydrate in 1,000 cc. of distilled water.) Open the stop-cock a only partially, so that the solution of caustic potash in c may slowly drop down through the gas in the tube ah and absorb the carbon dioxide in so doing. When all the caustic potash solution in c (with the exception of 2 or 3 cc.) has passed through a, the latter is closed, thus pre- venting entrance of any air ; b is opened, / is slowly raised and g lowered. Continue the raising of / until the water in the tube ha reaches the stop-cock b and immediately close the latter. Allow the gas to stand in the tube ib five minutes before taking the reading of the volume on the tube, bearing in mind that the level of the water in g must be on a level with the water in ib to obtain equal pressure. The difference between o and the point indicated by the water in the tube ib will give the amount of car- bon dioxide absorbed from the gas by the caustic potash. Thus : cc. Original volume indicated at 0.0 After removal of carbon dioxide '. . . 12.2 or 12.2 per cent, carbon dioxide by volume. To obtain the oxygen the gas is forced from ib into ah, as before, and in c is placed 50 cc. of an alkaline solution of pyro- gallic acid. This latter solution is formed by dissolving 10 grams of pyro- 630 ENGINEERING CHEMISTRY gallic acid in 25 cc. of distilled water, placing it in c and adding 35 cc. of the caustic potash solution. This is allowed to pass slowly through a and gradually absorbs the oxygen in the gas. a is closed before all the liquid passes out of c. Repeat with the same quantity of alkaline pyrogallic solution. Transfer the gas in the usual manner to ib, and after allowing to stand five minutes, take the measurement thvis : cc. Previous reading 12.2 After absorbing oxygen 19. i Oxygen 6.9 or 6.9 per cent, by volume. Before transferring the gas to ah for the determination of the carbon monoxide, all the water in / and ah must be replaced by distilled water ;^ to do this, open the three-way cock e, open a and all the water can be caught in a large beaker at e. Wash out / and ah three times with the water, then close e in the proper manner so that the water placed in / will rise in the tube ha to a, then close a, lower /, raise g, open b, placing the gas in ah for treatment with a solution of cuprous chloride to determine the carbon monoxide. The cuprous chloride solution is made by dissolving 30 grams of cuprous oxide in 200 cc. hydrochloric acid specific gravity (1.19), and using 50 cc. as soon as the solution has reached the temperature of the room. Experience has shown that a freshly made solution acts much better in an absorbent of carbon monoxide than one that has stood several days. Fifty cubic centimeters of this solution are placed in c and allowed to drop slowly through a and absorb the carbon monoxide as it passes through the gas. This absorption should be repeated at least three times. The heat generated during this ab- sorption often causes such an increase in the volume of the gas that when the latter is transferred to the tube ib for measurement, ' The water used in the apparatus at the commencement of the gas analysis should be saturated with the gas. After determination of carbon dioxide, distilled water can be used. Con.sult foot note on page 632. ENCxINI^ERING CHEMISTRY 63I the reading may prove minus. To insure accuracy proceed as follows : The gas, after fifteen minutes, is transferred in the usual way to bi, and the water in / and ah is replaced with distilled water. The gas is now returned to ah and a solution of potassium hy- droxide is placed in c and allowed to pass through the gas in ah, absorbing all traces of hydrochloric acid gas. Repeat with this once. Return the gas to bi, allow to stand fifteen minutes, then take the reading : cc. Previous reading 19.I After using CU2CI2 solution 19.5 CO 0.4 The nitrogen is determined by subtracting the total amounts of carbon dioxide, oxygen, and carbon monoxide from loo. Thus the analysis will read : Per cent, by volume Carbon dioxide 12.2 Oxygen 6.9 Carbon monoxide 4 Nitrogen 80.5 Total loo.o In this analysis no corrections are required for the tension of the aqueous vapor, since the original gas is saturated with mois- ture, and during the analysis all measurements are made over water. ^Apparatus for Flue Gas Analysis. — The Orsat apparatus il- lustrated in Fig. 119, is the one most frequently used for analyzing flue gases. The burette A is graduated in cubic centimeters up to 100, and is surrounded by a water jacket to prevent any change in temperature from affecting the density of the gas being analyzed. For accurate work it is advisable to use four pipettes, B, C, D, B, the first containing a solution of caustic potash for the absorption of carbon dioxide, the second an alkaline solution of 1 From the 35th Edition of Steam published by the Babcock & Wilcox Company, New- York. 632 ENGINEERING CHEMISTRY pyrogallol for the absorption of oxygen, and the remaining two an acid solution of cuprous chloride for absorbing the carbon monoxide.^ Each pipette contains a number of glass tubes, to which some of the solution clings, thus facilitating the absorp- tion of the gas. In the pipettes D and B, copper wire is placed in these tubes to re-energize the solution as it becomes weakened. Fig. 119 — Orsnt apparatus. The rear half of each pipette is fitted with a rubber bag, one of which is shown at K, to protect the solution from the action of 1 The proportion.s used by the Babcock & Wilcox Co. in preparing sohitious for use the Orsat apparatus are as follows : (i) For the absorption of COo I gram of KOH (.Sticks) dissolved in 2 grams of water. (2) For the absorption of O. I gram of Pyrogallic acid dissolved in 3 grams of water. 2, grams of KOH dissolved in 2 grams of water. Mix these two solutions in the Orsat pipette in the proportion of 15 cc. of the acid to 35 cc. of KOH. (3) Mix 5 grams of water and 3 grams of Cuprous chloride. To this add 15 cc. or 13 grams of con. HCIy (sp. gr. 1.2) and place in bottle with copper gauze. Use when liquid is clear. DNGINEKRING CHEMISTRY 633 the air. The solution in each pipette should be drawn up to the mark on the capillary tube. The gas is drawn into the burette through the U-tube H, which is filled w4th spun glass, or similar material, to clean the gas. To discharge any air or gas in the apparatus, the cock G is opened to the air and the bottle F is raised until the water in the burette reaches the lOO cc. mark. The cock G is then turned so as to close the air opening and allow gas to be drawn through H, the bottle F being lowered for this purpose. The gas is drawn into the burette to a point below the zero mark, the cock G then being opened to the air and the excess gas expelled until the level of the water F and in A are at the zero mark. This operation is necessary in order to obtain the zero reading at atmospheric pressure. The apparatus should be carefully tested for leakage as well as all connections leading thereto. Simple tests can be made; for example: If after the cock G is closed, the bottle F is placed on top of the frame for a short time and again brought to the zero mark, the level of the water in A is above the zero mark, a leak is indicated. Before taking a final sample for analysis, the burette A should be filled with gas and emptied once or twice, to make sure that all the apparatus is filled with the new gas. The cock G is then closed and the cock / in the pipette B is opened and the gas driven over into B by raising the bottle F, The gas is drawn back into A by lowering F and when the solution in B has reached the mark in the capillary tube, the cock / is closed and a reading is taken on the burette, the level of the water in the bottle F being brought to the same level as the water in A. The operation is repeated until a constant reading is obtained, the number of cubic centi- meters being the percentage of CO2 in the flue gases. The gas is then driven over into the pipette C and a similar operation is carried out. The difference between the resulting reading and the first reading gives the percentage of oxygen in the flue gases. The next operation is to drive the gas into the pipette D, the 634 ENGINEERING CHEMISTRY gas being given a final wash in B, and then passed into the pipette B to neutraHze any hydrochloric acid fumes which may have been given off by the cuprous chloride solution, which, especially if it be old, may give off such fumes, thus increasing the volume of the gases and making; the reading on the burette less than the true amotmt. The process must be carried out in the order named, as the pyrogallol solution will also absorb carbon dioxide, while the cuprous chloride solution will also absorb oxygen. As the pressure of the gases in the flues is less than the atmos- pheric pressure, they will not of themselves flow through the pipe connecting the flue to the apparatus. The gas may be drawn into the pipe in the way already described for filling the apparatus, but this is a tedious method. For rapid work a riibber bulb aspirator connected to the air outlet of the cock G will enable a new supply of gas to be drawn into the pipe, the apparatus then being filled as already described. Another form of aspirator draws the gas from the flue in a constant stream, thus insuring a fresh supply for each sample. Several improvements in construction of the absorption tubes of the Orsat-Muencke gas analysis apparatus have been made. Hankee's^ absorption pipette are shown in Figs. 120 and 121. In Fig. 120 is shown a capillary glass tube fused into the pipette for better absorption, which reaches nearly to the base and through which the gas passes. A small glass vessel is placed un- der the end of the capillary tube. The gas comes out in bubbles, disperses, rises through the absorbent liquid and is again carried off through a second tube, through a Geissler stop-cock. R. Nowicki- has added to the Hankus pipette a winding glass tube shown in Fig. 121. The gas leaves at the lower end though a fine opening, rises in the winding tube, always presenting new surfaces for absorption. Fig. 122 shows the complete form of the Hahn apparatus for gas analysis. 1 Stahl and Ei'sen, 23, 1, 261, 1903. 2 Osteer. Zeitschrift. f, Betg. u. Hultemv., 1905. ENGINEERING CHEMISTRY 635 In using this apparatus the burette M, the leveHng vessel JV, (contains water), and the glass tube surrounding the combustion tube 5, is also filled with water. The absorption pipettes contain solution as follows and are filled about one-half : No. i contains a solution of potassium hydrate in water (specific gravity 1.26) ; No. 2 contains fuming sulphuric acid (HoSoO^) ; No. 3 contains a solution of alkaline pyrogallate, formed by dissolving 10 parts of pyrogallic acid in 40 parts of hot water and adding 70 parts of !tA Fig. 120. Fig. 121. potassium hydroxide solution (specific gravity 1.26) ; No. 4 con- tains an alkaline solution of ammonia-cuprous chloride. The gas tubing introducing the gas is connected at G. After raising the leveling bottle W all the air of the apparatus is dis- placed and 103 cc. to 105 cc. of the gas is measured in the burette M, and the excess passed out so that 100 cc. remain for analysis. CO2 is absorbed in pipette No. i. Illuminants, if present, are absorbed in pipette No. 2, then passed through No. i again before 636 KNGINKERING CHEMISTRY measurement. The gas is then passed 3 or 4 times through No. 3, before measuring for determination of the oxygen, and the same in using pipette No. 4 for the absorption of the CO. For the hydrogen determination a portion, or if necessary, all of the residual gas is passed over the heated palladium sponge in Fig. c. Two-thirds of the reading of the contraction in volume of the gas represents the amount of hydrogen. To determine the me- thane, the residual gas is passed in combustion tube 5, the plati- num wires therein heated to a red heat, the contained gas in the meantime being brought a number of times in contact with the hot ENGINEERING CHEMISTRY 637 platinum wires, whereby the methane is burned to COg. The half of the contraction here found upon measurement corresponds to the quantity of methane, determined by passing the gas into pipette I, to absorb the COg formed. The residual is nitrogen. The analysis made by the Orsat apparatus is volumetric ; if the analysis by weight is required, it can be found from the volu- metric analysis as follows : Multiply the percentages by volume iby either the densities or the molecular weight of each gas, and divide the products by the sum of all the products; the quotients will be the percentages by weight. For most work sufficient accuracy is secured by using the even values of the molecular weights. The even values of the molecular weights of the gases appear- ing in an analysis by an Orsat are : Carbon dioxide . ._ 44 Carbon monoxide 28 Oxygen 32 Nitrogen 28 Table i mdicates the method of converting a volumetric flue- gas analysis into an analysis by weight. TABLE I.— Conversion of a Fi^ue Gas Anai^ysis By Voi^ume To One By Weight. Gas Analysis by volume per cent. Molecular weight Volume times molecular weight Analysis by weight per cent. Carbon Dioxide CO.^ Carbon Monoxide CO Oxygen O Nitrogen N 12.2 0.4 6.9 80.5 I2-|-(2 X 16) 12 -h 16 2x16 2X 14 536.8 II. 2 220.8 2254.0 536.8 3022.8=17.7 11.2 3022.8= 0.4 220.8 3022.8= 7-3 2254.0 3022.8=74-0 Total -.. loo.o 3022.8 100. Several instruments have been devised to indicate continu- ously the COo and also to record the same, so that the fireman shall at all times be able to see what he is doing in the way of 638 ENGINEERING CHEMISTRY efficient firing, and the superintendent or chief engineer have a continuous record of what he did, and thus have a controlUng check on the fireman's work. To obtain these continuous records the usual type of circular chart driven by clockwork is almost universally employed, the recording pen being actutated by one of two general methods (a) The continuous sample method; (b) the intermittent sample method. (a) The action of the continuous sample method is based on the law governing the flow of gas through two small apertures. This law may be illustrated by a simple diagram Fig. 123 repre- Fig. 123. senting two chambers C and C^ which are in communication with each other through the aperture A. O is connected with an as- pirator D, as shown. The manometers p and q indicate the gas tension within the respective chambers. When the aspirator is set in motion, a vacuum is created in chamber C^, the gas will flow from the chamber C through aperture B to chamber C^, creating a vacimm in C which will cause gas to enter through aperture A, thus establishing a con- tinviovis flow of gas through both apertures. If a constant vacuum of say 48" of water be maintained in chamber C^ and the two apertures A and B are of the same size and are maintained at the same temperature, the manometer p ENGINEIERING CHEMISTRY 639 will show about ^ the vactium maintained in C^, due to the fact that the apertures offer equal resistance to the passage of the gas. This relation will be maintained so long as the same volume of gas flows through B that enters at A. If, however, a constituent of gas be continuously taken away or absorbed from the gas in passing through chamber C, the vacuum therein will be correspondingly increased. This increase of vacuum in C, shown by the manometer p, therefore correctly indicates the vol- ume of gas absorbed and this is utilized in practice to indicate the percentages of CO2 in the flue gas. To utilize this principle in a practical apparatus, the following condition must be fulfilled. (i) The gas must be brought to the instrument under a con- stant tension and must be drawn through the aperture with a continuous and uniform suction. (2) Both apertures must be located in a medium of constant temperatures. (3) Provision must be made that the aperture remain per- fectly clean. (4) The chamber C must be made perfectly tight so that no gas can enter, except through aperture A. (5) The CO2 to be measured must be completely absorbed after the gas passes through A, and before it passes through B. In spite of the care required to secure satisfactory results with this type of CO2 recorder it possesses one big advantage over all other types in that it gives a continuous indication of the amount of CO2 in the gases. {h) Intermittent Sample Method. In obtaining CO2 records by this method measured samples of the gas are drawn into the apparatus, the CO2 absorbed as in an Orsat or Hempel machine, the new volume of the gas measured and the result recorded on the chart. In other words, the Inter- mittent Sample Method is simply an automatic Orsat, which at definite intervals, takes a sample of the gas, automatically deter- mines the percentage of CO2 and then records it, the whole op- eration taking from 5 to 10 minutes for each sample. In this case the recording pen remains fixed during a time interval equal 640 Engine:ering chemistry to that required to make a complete analysis, — moving radially along the chart to a new position as soon as the analysis of the new gas sample has been completed. This gives a curve made up of a series of broken lines which does not truly represent a "continuous" CO2 analysis, but due to the simplicity of operation of this type, it is tinding wide use in boiler room practice, where an approximate average value of the COo is all that is desired. These recording types of CO^ machines, while originally de- signed for use with flue gases, may be adapted to a wide variety of gases in the different arts by changing the absorbing solution to receive any desired gas instead of CO^. An extremely simple method of obtaining an average value of a "continuous" flue gas sample is to take two large bottles, the upper filled with water and so connected with the lower by rub- ber tubing that the water will flow from the upper to the lower by gravity and at any desired rate by means of regulating pinch cocks. The upper bottle is connected with a sampling pipe lead- ing to the flue so that as the water runs out of the bottom of the bottle, the flue gas is drawn in at the top to replace it. This may be set so that the sample will cover several hours, at the end of which time all openings to the upper bottle are closed, and the sample taken up to the laboratory to be analyzed in either a Hempel or Orsat machine. In using this two-bottle combination, great care must be exercised to be sure that all joints are per- fectly tight, as a very slight air leak in any part will have a very decided effect on the CO2 obtained. Also the water must be thoroughly saturated so that it will not take up any of the CO^ from the gas. Although these automatic or continuous COo machines are ex- cellent for regular power plant work, when determinations are desired in boiler test work, the Orsat style of apparatus is still in favor, due to its ability to give not only the COo accurately whenever desired, but also to determine the percentage .of oxygen and carbon monoxide, a knowledge of the proportions of both of which in the flue gas is necessary for the testing engineer in working up his results. To indicate the value of the flue gas ENGINEERING CHEMISTRY 641 n 1 a I luunioo ;o |5^^§ ?? ^ punod jad ^^5^ f? ^ aniBA JB3H •sqi8 uiun ■loo+i M \o r^o 00 f** = 1 uiunioa i/^ r^ tr lO CN cj JO punod 13d. lonpo.id sno^sBO (N vO ro 10 00 10 spunod Oxjee-i^ Si.^^^ ^ ?J, 00 JO punod J3d Jiv spunod C X :^ze-£= lOf^ \0 00 (S '^ 1 uuiniODjo punod a<»d ud3oJiiis[ rxi -^ '^^ l^ t^ spunod t-^ fj h-i ^0 punod asd U38AXO (N « d X, V -a -a lU :Hg3 g -x uon diox mon diox di ater Dio UO -snqiuoo joionpojj Carbon Carbon Carbon Water Carbon and VI Snlphut «. cO ^> II c uonDBaa IB01U13110 9000++0 .ooOXotj^ ^qgpAv w- f-O 3uuuqmo3 ao'oimojv ^ = 229.5 cu. ft. of air per pound of carbon. 1 See Table 2, p. 641. 646 Engine:e:ring CHICMISTRY Actual volume for one pound carbon Per cent, cubic feet by volume Carbon dioxide 32 = 13.01 ( _ 20.91 per cent. Oxygen 16 = 7-oo ^ ^ Nitrogen. 181. 5 = 79.09 229.5 100.00 For 100 per cent, excess air volume will be as follows : 153 X 2 = 306 cu. ft. of air per pound of carbon. Actual volnme for one pound carbon Per cent, cubic feet by volume Carbon dioxide 32 = io-45 | _ ^0.91 per cent. Oxygen 32 = 10.45 ) ^ ^ Nitrogen 242 = 79-09 306 100.00 In each case the volume of oxygen which combines with the carbon is equal to (cubic feet of air X 20,91 per cent.) — 32 cubic feet. It will be seen that no matter what the excess of air supplied, the actual amount of carbon dioxide per pound of carbon re- mains the same, while the percentage by volume decreases as the excess of air increases. The actual volume of oxygen and the percentage by volume increases with the excess of air, and the percentage of oxygen is, therefore, an indication of the amount of excess air. In each case the sum of the percentages of CO^, and O is the same, 20.9. Although the volume of nitrogen in- creases with the excess of air, its percentage by volume remains the same as it undergoes no change while combustion takes place ; its percentage for any amount of air excess, therefore, will be the same after combustion as before, if cooled to the same tem- perature. It must be borne in mind that the above conditions hold only for the perfect combustion of a pound of pure carbon. Carbon monoxide (CO) produced by the imperfect combus- tion of carbon, will occupy twice the volume of oxygen entering into its composition and will increase the volume of the flue gases over that of the air supplied for combustion in the proportion of 100 + )4 the per cent, of CO 100 ENGINEEJRING CHEMISTRY 647 When pure carbon is the fuel, the sum of the percentages by volume of carbon dioxide, oxygen and ^ of the carbon monox- ide, must be in the same ratio to the nitrogen in the flue gases as is the oxygen to the nitrogen in the air supplied, that is, 20.91 to 79.09. When burning coal, however, the percentage of nitro- gen is obtained by subtracting the sum of the percentages by volume of the other gases from 100. Thus if an analysis shows 12.5 per cent. CO2, 6.5 per cent. O, and 0.6 per cent. CO, the percentage of nitrogen which ordinarily is the only other con- stituent of the gas which need be considered, is found as follows : 100 — (12.5 + 6.5 + 0.6) = 80.4 per cent. The action of the hydrogen in the volatile constituents of the fuel is to increase the apparent percentage of the nitrogen in the flue gases. This is due to the fact that the water vapor formed by the combustion of the hydrogen will condense at a temj^^ra- ture at which the analysis is made, while the nitrogen which ac- companied the oxygen with which the hydrogen originally com- bined maintains its gaseous form and passes into the sampling apparatus with the other gases. For this reason coals contain- ing high percentages of volatile matter will produce a larger quantity of water vapor, and thus increase the apparent percent- age of nitrogen. Air Required and Supplied. — When the ultimate analysis of a fuel is known, the air required for complete combustion with no excess can be found as shown in the chapter on combustion, or from the following approximate formula : Pounds of air required per pound of fuel = where C, H and O equal the percentage by weight of carbon, hydrogen and oxygen in the fuel divided by 100. When the flue gas analysis is known, the total amount of air supplied is : * This formula is equivalent to (lo) given in above. 34-56 = theoretical air required for combustion of one pound of H (see Table 2). 648 ENGINEI^RING CHEMISTRY Pounds of air supplied per pound of fuel = 3.036 (^o;L)xC. (:.) where N, CO2 and CO are the percentages by volume of nitrogen, carbon dioxide and carbon monoxide in the flue gases, and C the percentage by weight of carbon which is burned from the fuel and passes up the stack as flue gas. This percentage of C which is burned must be distinguished from the percentage of C as found by an ultimate analysis of the fuel. To find the percentage of C which is burned, deduct from the total percentage of carbon as found in the ultimate analysis, the percentage of unconsumed carbon found in the ash. This latter quantity is the dift'erence between the percentage of ash found by an analysis and that as determined by a boiler test. It is usually assumed that the entire combustible element in the ash is carbon, which assumption is practically correct. Thus if the ash in a boiler test were 16 per cent, and by analysis contained 25 per cent, of carbon, the percentage of unconsumed carbon would be 16 X -25 = 4 per cent, of the total coal burned. If the coal contained by ultimate analysis 80 per cent, of cabon the percentage burned, and of which the products of combustion pass up the chimney would be 80 — 4 ^^ 76 per cent., which is the correct figure to use in cal- culating the total amount of air supplied by formula (/^). The weight of flue gases resulting from the combustion of a pound of dry coal will be the sum of the weights of the air per pound of coal and the combustible per pound of coal, the latter being equal to one minus the percentage of ash as found in the boiler test. The weight of flue gases per pound of dry fuel may, however, be computed directly from the analyses, as shown later, and the direct computation is that ordinarily used. The ratio of the air actually supplied per pound of fuel to that theoretically required to burn it is : '"''(. CO. + CO ) '^ ' ^ ,,3, * For degree of accuracy of this formula, see Transactions, A. .S M. E , Volume XXI, 1900, page 94. ENGINEERING CHEMISTRY 649 in which the letters have the same significance as in formulae (//) and (12). The ratio of the air supplied per pound of combustible to the amount theoretically required is : N CO) (14) N — 3.782 (O - which is derived as follows : The N in the flue gas is the content of nitrogen in the whole amount of air supplied. The oxygen in the flue gas is that con- tained in the air supplied and which was not utilized in com- bustion. The oxygen was accompanied by 3.782 times its volume of nitrogen. The total amount of excess oxygen in the flue gases is (O 14 CO) ; hence N — 3.782 (O — ^/^ CO) represents the nitrogen content in the air actually required for combustion and N ^ (N — 3.782 [O — ^ CO] ) is the ratio of the air supplied to that required. This ratio minus one will be the proportion of excess air. The heat lost in the flue gases is I. Z.Z0.24W (T-0 (15) Where L = B. t. u. lost per pound of fuel, W = weight of flue gases in pounds per pound of dry coal, T =: temperature of flue gases, t = temperature of atmosphere, 0.24 = specific heat of the flue gases. The weight of flue gases, W, per pound of carbon can be com- puted directly from the flue gas analysis from the formula : ir CO, -f 80+ 7(CO + N) . 3 (CO, -h CO) where CO^, O, CO, and N are the percentages by volume as de- termined by the flue gas analysis of carbon dioxide, oxygen, car- bon monoxide and nitrogen. The weight of flue gas per pound of dry coal will be the weight determined by this formula multiplied by the percentage of carbon in the coal from an ultimate analysis. 650 ENGINKE)RING CHEMISTRY :;i::|i;;jig»i:j[j[|[i^u|iS:jp:g?:i|^ o .•=!» ^ii stn „ -C-" 4; l; nd P9 K -O != ^ >< c: s R ?^ Is s ■ -^ Tl o?i rt -1 5^ ^- hr^ Ph c "^ r "^ ^ ^^ :d ^1 H 5P&> M •>>ii ^ ^rr ►^ > n (I) „ ;^ '■« i*. ,^^ &.^ >.2; a •:3 s -^ D >, ^ s /2 >>fi >! ^ i: n so, rt V '^ -ct di ^ 'u ^ CJ b^ rt K II w X. V O. a3 * For loss per pound of coal multiply by per cent, of carbon in coal bj' ultimate analysis. ^NGIN^^RING CHEMISTRY 65 1 Fig. 124 represents graphically the loss due to heat carried away by dry chimney gases for varying percentages of CO,, and dif- ferent temperatures of exit gases. 8 8 8 8 The heat lost, due to the fact that the carbon in the fuel is 652 ENGINEE^RING CHEMISTRY not completely burned and carbon monoxide is present in the flue gases, in B. t. u. per pound of fuel burned is : L=.o,:5ox( cO^°CoJ c (17) where, as before, CO and CO, are the percentages by volume in the flue gases and C is the proportion by weight of carbon which is burned and passes up the stack. Fig. 125 represents graphically the loss due to such carbon in the fuel as is not completely burned but escapes up the stack in the form of carbon monoxide. Application of Formulae and Rules. — Pocahontas coal is burned in the furnace, a partial ultimate analysis being: Per cent. Carbon 82.1 Hydrogen 4.25 Oxygen 2.6 Sulphur 1.6 Ash 6.0 B. t. u., per pound dry 14,500 The flue gas analysis shows : Per cent. CO2 107 O 90 CO 0.0 N (by difference) 80.3 Determine: The flue gas analysis by weight (see Table i), the amount of air required for perfect combustion, the actual weight of air per pound of fuel, the weight of flue gas per pound of coal, the heat lost in the chimney gases if the temperature of these is 500° F., and the ratio of the air supplied to that theor- etically required. Solution: The theoretical weight of air required for perfect combustion, per pound of fuel, from formula (//) will be, ^ /0.821 , . 0.026 o.oi6\ W = 34 56 (— h (0.0425 ^) H ^ j = 10.88 lbs. If the amount of carbon which is burned and passes away as flue gas is 80 per cent., which would allow for 2.1 per cent, of e)ngine:e:ring chemistry 653 unburned carbon in terms of the total weight of dry fuel burned, the weight of dry gas per pound of carbon burned will be from formula (16) : W = 1 1 X 10.7 -I- 8 X 9.0 + 7 (o 4- 80.3) = 23.42 pounds 3 C10.7 + o; and the weight of flue gas per pound of coal burned will be 0.80 X 23.42 = 18.74 pounds. The heat lost in the flue gases per pound of coal burned will be from formula (6) and the value 18.74 just determined. Loss = 0.24 X" 18.74 X (500 — 60) =: 1979 B. t. u. The percentage of heat lost in the flue gases will be 1979 -^ 14,500 =13.6 per cent. The ratio of air supplied per pound of coal to that theoretically required will be 18.74 -f- 10.88 ^= 1.72 per cent. The ratio of air supplied per pound of combustible to that re- quired will be from formula (14) : 0.803 _ 0.803 — 3.782 (0.09— >^ X o) ~ ^'^^ The ratio based on combustible will be greater than the ratio based on fuel if there is unconsumed carbon in the ash. Unreliability of CO^ Readings Taken Alone. — It is generally assumed that high CO2 readings are indicative of good combus- tion and hence of high efficiency. This is true only in the sense that such high readings do indicate the small amount of excess air that usually accompanies good combustion, and for this reason high CO2 readings alone are not considered entirely re- liable. Wherever an automatic CO^ recorder is used, it should be checked from time to time and the analysis carried further with a view to ascertaining whether there is CO present. As the percentage of CO2 in these gases increases, there is a tendency toward the presence of CO, which, of course, cannot be shown by a CO2 recorder, and which is often difficult to detect with an Or- sat apparatus. The greatest care should be taken in preparing the cuprous chloride solution in making analyses and must be known to be fresh and capable of absorbing CO. In one in- 654 e:ngine:e:ring chemistry stance that came to our attention, in using an Orsat apparatus where the cuprous chloride solution was believed to be fresh, no "CO was indicated in the flue gases but on passing the same sample into a Hempel apparatus, a considerable percentage was found. It is not safe, therefore, to assume without question from a high CO^ reading that the combustion is correspondingly good, and the question of excess air alone shall be distinguished from that of good combustion. The effect of a small quantity of CO, say I per cent., present in the flue gases will have a negli- gible influence on the quantity of excess air, but the presence of such an amount would mean a loss due to the incomplete com- bustion of the carbon in the fuel of possibly 4.5 per cent, of the total heat in the fuel burned. When this is considered, the im- portance of a complete flue gas analysis is apparent. Table 4 gives the densities of various gases together with other data that will be of service in gas analysis work. TABLE 4. — Density of Gases at 32 Degrees Fahrenheit AND Atmospheric Pressure. Adapted from Smithsonian tables. Gas Chemi- cal symbol Specific gravity Air = I Weight of one cubic foot pounds Volume of one pound cubic feet Relative hydro{ Exact density, yen = i Approxi- mate N H CO, CO CH, C,He SO2 1.053 0.9673 0.0696 1. 5291 0.9672 0.5576 1.075 0.920 2.2639 1. 0000 .08922 .07829 .005621 .12269 .07807 .04470 .08379 .07254 .17862 .08071 11.208 12.773 177.90 8.151 12.809 22.371 11.935 13.785 5.59« 12.390 15.87 13.92 1. 00 21.83 13.89 7.95 14.91 12.91 31.96 16 H I 22 14 8 15 13 32 Nitrogen Hydrogen Carbon dioxide • ■ . Carbon monoxide- Methane Acetylene Sulphur dioxide - . Air ENGINEERING CHEMISTRY ANALYSIS OF ILLUMINATING GAS.^ 655 Description and Method of Operating the Standard U. G. I. Gas Analyzing Apparatus. The following pipettes and reagents are required for the analysis: (see Figs. 126, 127, 128, 129.) I ^ ik M ; 1^ 1 f 1 Fig. 126— This cut represents a complete gas analyzing apparatus, consisting of a standard U. G. I. Hempel burette, pipette stands, six pipettes, induction coil and battery. The burette as seen in Fig. 130 is similar to that described in Hempel's Gas Analysis on page 28, except that a four-way cock C replaces the three-way cock used by Hempel and the burette is bulbed, thereby shortening the same and allowing a finer graduation. ' The above article is here published through the kindness of Mr. W. H. Fulwtiler, Chemist, The United Gas Improvement Company, Philadelphia, Pa. Fig. 127.— Standard U. G. I. Hempel burette. A .special feature of this burette is the four-way stop-cock, which permits a permanent connection with the potash pipette, thus obviating the necessity of repeatedly connecting and disconnecting the pipette duri^ng the course of an analy.sis. I ENGINK^RING CHEMISTRY 657 Fig. 128.— Double-absorption cuprous chloride pipette. This is designed to replace the two double-absorption pipettes otherwise necessary in making a gas analysis. By simply turning the cock, it is possible to bring the gas in contact with the absorbent contained in either side of the pipette without disconnecting. Compactness, and ease of filling and operation are the special features of this pipette. 42 658 Knginke:ring chkmistry Fig. 129.— Tutwiler and Bond hygrometer. This instrument indicates the tempera- ture point of saturation of a gas with hydrocarbon vapor or with water vapor present in the gas at the time of testing. It is particularly valuable in connection with the mainten- ance of a uniform quality of gas, as by its use we may determine the lowest temperature to which the gas subsequently may be subjected on its way to the consumer, without injury to its heating or lighting value. EJNGINKERING CHEMISTRY 659 The capacity of the burette is about 105 cc. graduated in 1/20 cc. from 40 cc. to 102 cc. It is connected through the capillary tube D coming out from the back of the cock C with manometer tube M. The manometer is connected with the Petterson cor- rection tube R. A water jacket / surrounds the Petterson tube and burette. A potash absorption pipette K which rests on the adjustable stand S is connected permanently with the capillary tube B. A potash absorption pipette which is permanently attached to the burette as shown in the sketch. A pipette filled with strong bromine water. In order that this solution remain concentrated an excess of free bromine is kept in the pipette. A pipette for solids filled with stick phosphorous covered with water. A double U. G. I absorption pipette. This combines in one piece of apparatus the two solutions of cuprous chloride which are necessay to remove the carbon monoxide. A simple pipette filled with saturated water for storage pur- poses. A mercury explosion pipette. A U-shaped combustion tube containing about one-half gram palladium black is also required. The following is the method of procedure for an analysis of a gas containing CO2, C^Hg;,, Og, H^OH^, CgHg, and N^. Completely fill water jacket with distilled water. Turn cock C so that the interior of the burette Y communi- cates with A, open cock Z, raise leveling bulb L, which has been filled with gas saturated water, until water flows out A. Turn C so that interior of burette communicates with K, and draw over potash solution to just above cock C. Turn cock C so that Y communicates with D and, by raising and lowering L and allowing air to escape through A, fill M with water to A^. Open C to A and, by lowering L, draw in air. Close C, raise L, open C to D and admit air in M to O, and close C. 66o ENGINEERING CHEMISTRY Disconnect M momentarily at P and reconnect. The air in A is now at atmospheric pressure. Connect the tube containing gas sample with A, using glass connector similar to one used on potash pipette, being careful to displace with water all air that may be in connections. Open C to A, lower L and draw in loo cc. of gas. Close C, raise L, open C to D, and allow gas to flow into M until the water level is at O, and close Z. Take the reading on burette after allowing a minute for water to run down off the sides of the burette, add I cc. to observed reading for the i cc. gas occupying space be- tween O and cock C. Disconnect from sample tube or gas supply as the case may be. Open C to B, raise L and allow gas to flow into K, until the water from burette reaches the bulbed portion of K, being careful to draw the i cc. from manometer and to force that into the potash likewise. Turn C to D and adjust water level at N in M. Turn C to B, lower L, and draw back gas until the potash solution just reaches its previous position above C and close C. Raise L and turn C quickly through arc of i8o.° so as to allow no gas to flow back to B while turning cock so that the interior of the burette communicates with man- ometer A^. Raise L until water in M is level with O, close Z and read burette, adding i cc. to observed reading as before. The difference between this reading and the preceding gives directly the percentage of CO2 in the gas. Connect absorption pipette containing bromine to A, resting it on stand S, being careful as before to exclude all air from con- nections. Open C to A, raise L, and force gas from the burette into the pipette until water reaches the bulbed portion of the pipette, drawing the gas from the manometer tube as before, and close C. Shake the bromine pipette slightly until gas is col- ored by bromine fumes, open C, lower L, and draw gas back into burette. Close C, raise L, open C to B, and force all gas im- mediately into potash. Close C. to B, and open to D, and adjust water level. Open C to B, lower L, and draw back gas until potash assumes former position. Close C, raise L, adjust water level and read as before, the difference between this reading and the preceding gives the per- ENGlNEmaNG CHEMISTRY 66i centage of C^Hg,,. Disconnect the bromine pipette from A and connect the phosphorous pipette. Force the gas over the phosphorus as was done with the bro- mine pipette, turn C to D, raise L and adjust water level; close C. If no white fumes are given off by the gas when in the Fig, 130. pipette it is a sure indication that all of the C„H2„, compounds in the gas have not been completely removed. In this event it is necessary to again pass the gas into the bromine pipette. If fumes are given off wait a minute or two to allow them to par- tially condense, then open C to A, lower L, and draw gas back into burette. Close C, raise L, open C to D, adjust water level 662 ENGINEERING CHEMISTRY at O, and take reading. The difference between this reading and the preceding reading gives percentage oxygen present. Dis- connect phosphorus pipette and connect double absorption pipette containing cuprous chloride, being careful to have all capillaries filled with the solution. Open C to A, raise L and force all gas over one solution of cuprous chloride. Shade for two or three minutes and then draw gas back into burette until solution just passes cock on cuprous chloride pipette; turn this cock so as to connect with other solution of cuprous chloride, raise L and force gas over second solution to remove last of carbon mon- oxide, and close C. Shake for a few minutes, draw gas back into burette, and then immediately force it into the potash pipette. Adjust water level, draw gas back from potash pipette and take reading. The difference between this reading and the preceding gives percentage of carbon monoxide. It is important to notice that even with the precaution of us- ing two pipettes with freshly prepared cuprous chloride the ab- sorption of the carbonic oxide is seldom complete, usually a trace remaining unabsorbed. However, this fact introduces no error in the analysis, as this residue of carbonic oxide can be deter- mined by the combustion made to determine hydrogen. The residue of the gas mixture remaining after the absorption may consist of the following : H^ + CO -f N, -f CH, + C^H^, C3H3, etc. For all ordinary purposes it is sufficient to assume that the highest parraffin present is CgHe, as all others higher than this exist only in traces. There being no satisfactory known absorbent for any of these gases, recourse is had to the method of combustion. The analysis is accordingly continued as follows : The double absorption pipette is replaced by the storage pipette containing gas saturated water. Pass approximately 15 cc. of the residue back into the potash by opening Z, raising L and open- ing C to B. Turn C to A and pass remainder of residue into storage pipette. Close pipette with a pinch cock and disconnect. Adjust water level in M at A^. Turn C to A and by lowering L, I ENGINEERING CHEMISTRY 663 draw into the burette about 85 cc. of air. Close C, raise L and open C to D, draw the gas stored over the potash into the bur- ette, close C, raise L, turn C quickly through arc of 180° to con- nect with D, adjust water level at O, close Z and take reading. The increase over the previous reading is the amount of gas taken for the explosion. Connect mercury explosion pipette at A and pass mixture of gas and air into pipette and explode, first partly withdrawing glass connecting tube from rubber connection and placing clip on same. Adjust water level in M at A^, draw back gas from ex- plosive pipette and measure contraction resulting from the ex- plosion. Pass the gas into potash, and the resulting contraction gives the amount of carbonic acid formed during the explosion. Disconnect explosion pipette and connect phosphorus pipette. Pass gas residue over phosphorus to remove all oxygen in excess of that which was required for explosion and measure the amount of nitrogen left. This gives nitrogen introduced with gas. By subtracting amount of air used for explosion by 79.2 from this reading, one obtains nitrogen introduced with gas for explosion. This multiplied by factor obtained by dividing the amount of gas residue taken for the explosion into the whole amount of gas left after absorbing carbon monoxide, gives the total nitrogen in the original sample of gas taken for analysis. The percentage of nitrogen thus obtained should check that obtained by subtract- ing the sum of the other constituents in the gas from 100. The equations obtained from the explosion are as follows : (i) Contraction in volume = 3/2U0 + >^CO + 2CH^ + 2/2C,He. (2) CO, formed = CO -f CH, + 2C,¥[,. (3) Residual nitrogen = Ng + N^, where N^ is the nitrogen introduced with the air. An examination shows that the equations i and 2 contain 4 unknown quantities and therefore two more equations are needed for the solution. Fortunately, the method of fractional combus- tion over palladium affords the needed information. As is well known, when a mixture of hydrogen and CH.^ with oxygen or 664 KNGINEKRING CHEMISTRY air is passed over heated palladium black, the hydrogen burns to H.O, but the CH^ remains unaltered. If CO and any of the higher parraffins are also present, the CO burns, but the parraffins do not. Returning to the analysis, proceed as follows : Fill burrette to A by raising L, adjust water level at A^ and M. Draw m about 70 cc. air and measure it. Connect storage pipette and draw in about 30 cc. gas residue, and measure, the increase in volume giving the amount of gas taken for combustion. Place explosion pipette with mercury level about one-half up to capillary, on stand S, connect combustion tube to A and ex- plosion pipette, equalize pressure in combustion tube and gas burette and remeasure gas in burette. Place combustion tube in hot water by resting beaker containing water on T and pass gas mixture backward and forward over palladium until there is no future contraction, measure gas and decrease in volume gives contraction due to combustion of hydrogen and carbon monox- ide. The equations are : (4) Contraction in volume ^= 3/2H0 + ^CO. (5) CO2 former = CO. From these two equations, the value of hydrogen and CO may be readily determined. For the sake of simplicity, let us now assume that the quan- tity of gas residue was used in both the explosion and the com- bustion. We may then subtract equation (4) from (i) and (5) from (2), whence, designation the difference between the contraction due to combustion by the letter (a) and the difference in the CO2 formed by the letter ( ^ ) we find (6) 2CH, -f 2>^ C,H, = «. (7) CH, + 2C,H, =b. Ab — 2a whence (8) C^H, and (9) CH. 3 4^ — 5b 3 ENGINEEJRING CHEMISTRY 665 A very useful check on the accuracy of this determination is obtained from the following: Volume of gas taken for explosion = H2 + N. + CO + CH4 + C.He. H2 + CO are found by (4) and (5), and A^ is given by (3). Therefore, we have (10) Volume taken = (H^ + N2, + CO) = CH^ + CoHg and this value should be the same as the algebraic sum of (8) and (9) or (11) Volume taken + (H, + N, + CO) = ^^izl' This method if carefully pursued will give results that are ex- tremely accurate, and what is much to be desired, the method is very rapid. Analyses have repeatedly been made in from 30 to 35 minutes. JUNKER'S GAS CALORIMETER. The sectional drawing (Fig. 131) shows the instrument to con- sist of a combustion chamber surrounded by a water jacket, the latter filled with a great many tubes. To prevent loss by radia- tion the water jacket is surrounded by a closed air space. The whole apparatus is constructed of copper as thin as is compatible with strength. The water enters the water jacket at the bottom, and leaves it at the top, while the hot combustion gases of the flame of the gas that is on trial enter the tubes at the top and leave them at the bottom. There is, therefore, not only a very large surface of thin copper between the gases and the water, but the two move in opposite directions, during which process all the heat generated by the flame is transferred to the water, and the waste gases leave the apparatus approximately at atmos- pheric temperature. The gas to be burned is first passed through a meter, and then to insure constant pressure, through a pressure regulator. The source of heat in relation to the unit of time is thus rendered stationary, and, in order to make the absorbing quantity of heat also stationary, two overflows are provided at the calorimeter, making the head of the water and the rate of flow of the same constant. The temperatures of the water enter- ing and leaving the apparatus can be read at the respective ther- 666 e;ngine:e:ring chemistry mometers; as shown before, the quantities of heat and water passed through the apparatus are constant. As soon as the flame is Hghted the temperature of the exit thermometer will rise to a certain point and will nearly remain there. All data for ascer- taining the heat given out by the flame are therefore available. Fig. 13 Cold water inlet. Strainer. Overflow to calorimeter. Upper container. Waste overflow. and 7. Fall pipe and joint. Drain cock. Adjustment cock. Cold water thermometer. Air jacket. Perforated spreading ring. and 16. Water jacket. Baffle plates with cross slots. Lower overflow. Lower container. Hot water overflow. Heated water outlet. Gas nipple. Air supply regulator. Gas nozzle. Clamp for burner. Burner holder. Burning cap. Combustion chamber. Roof of combustion chamber. Cooling tubes. Receiver for combustion gases. Outlet for combustion gases. Throttle for combustion gases. Brass base ring. Condensed water outlet. 37 and 38. Air jacket. Test hole in air jacket. Hot water thermometer. All that is required is to measure simultaneously the quantity of gas burned and the quantity of water passed, and the differ- KNGINKERING CH£;MISTRY 667 ence in temperature between the entering and leaving water. Centigrade thermometers and 2-liter flasks are required. The meter shows o.i of a cubic foot per revolution of the large hand, the circumference being divided into 100 parts, so that 0.00 1 can be read accurately. The water supply is so regulated that the overflow is working freely, and the water-admission cock Fig. 132. is set to allow 2 liters of water to pass in about a minute and a half. The calorimeter is now ready to take the reading. The cold water, as a rule, has a sufficiently constant temperature that we note it only once: it is now 17.2° C. As soon as the large index of the meter passes zero, note the state of the meter and at the same time transfer the hot-water tube from the funnel into the measure glass, and while that is being filled note the tempera- ture of the hot water at say 10 intervals, to draw the average. 668 ENGINEERING CHEMISTRY The temperatures are 43.8°, 43.5°, 43-5°, 44-2°, 44.1°, 43-9°, 43.8°, 43.7°, 43.8°, and 43.7°, making the average 43.8°. The measure glass is now filled ; turn the gas out. Find from the readings of the meter at the beginning and the end of the experiment that there was burned 0.35 cubic foot, by means of which the temperature of the 2 liters of water was raised 26.6° C. ; viz., 43.8° — 17.2° = 26.6° C. The calculation is as follows : WT where H =1 the calorific value of i cubic foot of gas in calories, W =^ the quantity in liters of the water heated, T = the differ- ence in temperature between the two thermometers in degrees C, and G ^= the quantity in cubic feet of gas used, then ^ 2 X 26.6 , . H = ^152 calories per cubic foot or 604 (152 X 3-968) B. t. u. per cubic foot. It is mentioned before that the effect of the cooling water is such that the waste gases leave the calorimeter at about atmjos- pheric pressure. All hydrocarbons when burned form a con- siderable quantity of water, which in all industrial processes es- capes with the waste gases as steam. The latent heat of this steam is therefore not utilized when firing a stove or driving an engine with gas; in the above result, however, the latent heat is included, because in the copper tubes the steam is condensed, and its heat is transferred to the circulating water and measured with the rest. The condensed water runs down the tubes which are cut off obliquely to allow the drops to fall off easily, and is collected in the lower part of the apparatus from where it runs through the little tube into a measure glass. In condensing, steam gives off 0.6 calorie for every cubic centimeter of water formed. If therefore a graduated (cc.) cylinder be placed under the little tube the amount of water generated by burning, say i cubic foot of gas, can be directly measured. From burning i cubic foot of gas, we have collected 27.25 cc. of condensed water, and must therefore deduct 16.35 calories from the gross value found above, which gives the net calorific i:nginee:ring chemistry 669 value of the gas tested as 135.65 calories or 538 B. t. u. per cubic foot. The calorimeter is placed so that one operator can simultane- ously observe the two thermometers of the entering and escaping water, the index of the gas-meter, and the measuring glasses. No draft of air must be permitted to strike the exhaust of the spent gas. The water supply tube is connected to the nipple in the center of the upper container ; the other nipple is provided with a waste tube to carry away the overflow. This overflow must be kept running while the readings are being taken. The nipple through which the heated water leaves the calo- rimeter, is connected by an india-rubber pipe with the large meas- ure glass, and the water must be there collected without splash- ing. The smaller measure glass is placed under the tube to col- lect any condensed water. Table of Resume of Tests upon London Coal Gas. B V 3 '2 6ri a.2 i S ' £ •0 '0 2 ""b 0^ u (LI V V 3 11 3 ^0 3 e 3 3 2^- ^ ^ 2 fa J3 tr. s K^ s = S^ 8 0; V ti— V S-o ^ K H H J^ « H u hT % First Day 21.0 15.322 26.113 10.79 0.0407 .... 25.7 165.3 15-4 149.9 Second Day 22.5 12.9 - 27.68 14.78 0.0584 27.4 165.9 16.4 148.5 Third Day 17.5 13-71 28.6 14.89 0.1 103 17.5 26.43 164.8 15.86 148.94 Fourth Day 17-5 13.75 28.53 i4.7« 0.1 103 17.4 26.43 165.6 i5.«6 149.74 After the thermometers have been placed in position with their india-rubber plugs, the water supply is turned on by the cock, and the calorimeter filled with water until it begins to dis- charge. No water must at this period exude from the smaller pipe or from the test hole under the air jacket, otherwise this would prove the calorimeter to be leaking. Experiments made with this calorimeter at the Stevens Insti- tute are recorded in the Stevens Indicator, October, 1896. 6/0 KNGINKKRING CHEMISTRY The gas used was carbureted water gas, 'Xowe Process," composed as follows : Per cent, by volume CO2 -2.20 lUuminants < Q^H^ V 12 80 iCeHj O 0.00 CO 24.20 CH, 17.83 H 37-95 N 5.02 100.00 The theoretical heating value of this gas is 662 B. t. u. per cubic foot. The heating value as determined with the Junker calorimeter is 668 B. t. u. per cubic foot. MANUFACTURE OF WATER GAS. The water gas system consists in the decomposition of steam at a high temperature by incandescent carbon, thereby producing hydrogen and carbon dioxide: 2H204-C:=2H2+C02.^ In an excess of carbon, the carbon dioxide saturates itself with another carbon atom, forming carbon monoxide C02+C=2CO, making the finished product 2H2 + 2CO. In practical working the reduction of carbon dioxide to mon- oxide is never quite perfect, the unpurified gas usually contain- ing about 3 per cent, of carbon dioxide, to be extracted (as in coal gas) by lime purification. As the gas in the process of manufacture, passes from the gen- erator to the carbureters it is enriched by means of crude oil or cheaper distillates : hence the name carbureted zvater gas. The generator, carbureter, and superheater are cylindrical steel shells, thickly lined with special fire-bricks, between which and the metal are annular spaces packed with non-conducting material. The generator is usually supported on short columns, as illustrated, leaving cartage room under the hopper-shaped ash- 1 Humphrey's and Glasgows: "Carbureted Water Gas," 1895. ENGINEERING CHEMISTRY 67I pit. The grate, controlled by the several cleaning doors, is lo- cated slightly above the ash-pit, and the fire is charged with coke through the door in the extreme top. The generator is connected, both above and below the fuel-bed, with the top of the carbureter, the bottom of which leads laterally into the adjoining superheater. The carbureter and superheater, often referred to as the "fixing chambers," are filled with checker work, affording such an enormous heating surface that even the heaviest distillates can be permanently gasified at the low tem- perature necessary to the highest illuminating effect. The en- riching oil is introduced at the top of the carbureter. The oil heater is a simple and practical arrangement for pre- heating the oil on its way to the carbureter by means of the hot gas escaping from the superheater. Operation. — A fire is started in the generator, which is then deeply charged with coke and opened to the blast. The air enters in large volume below the grate and quickly kindles the fuel, while the hot products resulting from the partial combustion pass forward through the carbureter and superheater and, after part- ing with their sensible heat, escape into the stack. As soon as these generator gases have sufficiently warmed the checker-work, supplies of secondary air are admitted to the top of the carbu- reter and the bottom of the superheater, respectively, and the combustion regulated to give the requisite temperatures in the two vessels simultaneously. The generator fire being in proper condition, and the carbureter and superheater at the desired tem- peratures, the apparatus is ready for gas-making. The blasts are shut off one by one, beginning with that of the super- heater; the stack valve is closed; steam is admitted under the fuel bed, and having traversed it, passes as water-gas into the top of the carbureter. At this point the oil is introduced, and encountering the heated checker-work is vaporized and ultimately gasified in presence of the hot water-gas. This process continues until the temperature of the fire and the checker-work are suffi- ciently reduced. The oil is then shut off; next the steam; and the stack valve being opened the blasts are again admitted and the energy of the fire and the checker-work recuperated as first liNGlNEKRING CHEiMISTRY 673 described. The generator is supplied with fuel at intervals of from forty-five to sixty minutes, and cleaned usually once during each shift. The gas passes from the seal through the scrubbers and condensers and is subsequently deprived of its carbon dioxide and treated for its slight sulphur impurities in the manner com- mon to coal gas. Uncarbureted water gas has the following composition:^ Pet cent. H 4932 CH4 7.65 CO 37.97 CO2 0.14 N 4-79 O 0.13 Total 100.00 and after carbureting Per cent. H 38.05 CH4 11.85 CO 29.40 O o.io CO2 ". o.io N 3.71 Illuminants 16.79 Total 100.00 The heating power of the uncarbureted gas per cubic foot would be : Products condensed B. t. u. H 0.4932 X 348.0 = 171.63 CH4 0.0765 X 1,065.0 =: 81.47 CO 0.3797 X 34956 = 132.72 Total 385.82 and the heating power of the carburetted water gas per cubic foot would be : 1 King's Treatise on Coal Gas," Vol. Ill, p. 362. 43 674 ENGINEERING CHEMISTRY Products condensed B. t. u. H 0.3805 X 348.0 = 122.41 CHi O.I 185 X 1,065.0 = 126.20 CO 0.2940 X 349- 56 = 102.77 CO2 O N Illuminants 0.1679 X 2,000.0 = 335.80 Total 697.18 An analysis of a sample of London (Eng.) coal gas gives the following : Per cent. H 27.70 CH4 50.00 CO 6.80 C2H4 13.00 N 0.40 O CO2 O.IO Aqueous vapor 2.00 Total : 100.00 The heating power will be, per cubic foot, Products condensed B. t. u. H 0.2770 X 348.0 = 96.39 CH4 0.5000 X 1,065.0 = 532.50 CO 0.0680 X 34956 = 23.77 C2H4 0.T300 X 1,6730 = 217.49 Total 870.15 and when the products of combustion are in a state of vapor (for instance 328° F.) the heating power per cubic foot will be B. t. u. H 0.2770 X 264 = 73.12 CH4 0.5000 X 853 — 426.50 CO 0.0680 X 315 = 21.42 C2H4 0.1300 X 1,400 = 182.00 Total 703.04 ENGINEEJRING CHEMISTRY 675 There are few complete analyses of purified coal gas known,^ i. e., Heidelberg gas by R. Bunsen, Konigsburg gas by Bloch- mann, and Hannover gas by Dr. Fischer. Constituents Heidelberg gas Konigsberg gas Hannover gas Hannover gas II CeHe 1.33 0.66 0.69 0.59 CaHe 1. 21 0.72 0-37 0.64 C2H, 2.55 2.01 2. II 2.48 CH, 34.02 35.28 37-55 38.75 H 46.20 52.75 46.27 47.60 CO 8.88 4.00 II. 19 7.42 CO2 3.01 1.40 0.81 0.48 0.65 trace 0.02 N 2.15 3.18 I.OI 2.02 Total 100.00 100.00 100.00 100.00 In the Wilkinson process the water gas is made by the com- bined generator and retort process. (A full description of a com- plete plant will be found in The American Gas Light Journal, 57, 399, 401.) An analysis of a sample of Wilkinson water gas, made by the writer,^ gave as follows : Per cent. H 39.50 mu'^nal™^"'""1CsHeaverage| 6.60 CPI4 3730 CO 4.30 N 8.20 O 1.40 Impurities (HA CO2, H^S) 2.70 100.00 One cubic foot contains 681.73 B. t. u. products condensed. G. Lunge^ gives an analysis of ''Tessie du Motay" gas as follows : 1 Wagner's "Manual of Chemical Technology" (13th Edition), p. 39. " Wood's "Thermodynamics, Heat Motors, and Refrigerating Machines," 3rd. edition, pp. 260-261. a "Wassergasfabrikation in Neve York," Zeitschrift fur angewandte Chentie, 1894, pp. 137-142. 676 ENGINEERING CHEMISTRY Per cent. CO^ Illuminants 14.3 O 0.6 CO 21.1 H 28.8 CH, 25.5 N 3-1 Total 100.00 containing 754.61 B. t. u. per cubic foot, products condensed. For complete details regarding the manufacture of coal gas consult King's Treatise on Coal Gas," edited by Thomas New- bigging, London. Producer Gas Constituents Siemen's gas Anthracite producer gas Soft coal producer gas 00 23-7 8.0 2.2 4.1 62.0 27.0 12.0 1.2 2.5 57-3 100.00 27.0 12.0 2.5 2.0 56.5 H CH, CO, N Total 100.00 100.00 The heating power of the Siemen's producer gas will be 134. i B. t. u. per cubic foot, of the anthracite producer gas 153.7 ^- t- ti. per cubic foot, and of the soft coal producer gas 168. i B. t. u. per cubic foot (products of combustion condensed). Experiments made in Berlin, Germany, on the cost of power from various substances, show as follows •} H. P. DEVEI.OPED 10 H. p. 20 H. P 30 H. p. Cents Cents Cents Acetylene 5.85 5.62 5.54 Lighting gas 2.61 2.45 2.38 Power gas 2.90 2.19 i.< ' Alcohol 4.14 3.86 3 Petroleum 2.56 2.38 Benzene 3.70 3.59 Electricity 3.60 3.55 ^ Schweizerische Bauzeitung, January 26, 1901. ENGINEERING CHEMISTRY (i^^ NATURAL GAS. Natural gas from different localities has a somewhat different composition, as may be seen from the following table '} Pennsylvania Ohio Indiana Kansas Russia Carbon dioxide 0.05 0.3 0.25 0.44 0.95 Carbon monoxide ...... 0.5 0.45 0,33 .... Marsh gas 95.42 92.6 92.67 95.28 92.49 Nitrogen 4.51 3.5 3.53 3.28 2.13 Oxygen trace 0.3 0.35 ... .... Hydrogen 0.02 2.3 2.35 ... .... Hydrogen sulphide 0.2 0.15 ... .... Olefiant gas, etc 0.3 0.25 0.67 4.11 To be more specific and local in reporting the analyses, I re- fer to some of the analyses made during the past year by Prof. H. P. Cady and D. F. McFarland at the University of Kansas. Analyses of natural gas have been made from the following localities : Dexter, Eudora, lola, Moline, Fredonia, Neodesha, Lawrence, Erie, Kansas City, Bartelsville (I. T.), Bonner Springs, Par- sons, Arkansas City, Altamont, Coffeyville, Moran, Caney, Peru, Butler (Ohio), Chanute, Humboldt, New Albany, Altoona, Mound City, Buffalo, Blackwell (Okla.), Garnett, Olathe, Paola, Burlington, Augusta, Marion (Tnd.j, Morgantown (W. Va.), Elmdale, Eureka, Sheffield (Mo.), and Jennings (La.). Composition of 37 Samples of Natural Gas Pipe-lime Chanute Elmdale Olathe Dexter Caney Blackwell O2 . . . 0.24 .10 .30 trace .10 .10 .50 CO2 1.94 . . .15 00 .00 .81 .00 C2H4 5-. •• -55 .10 .00 .10 .61 CO . . .00 .00 .00 .00 .00 CH4 . 94.30 94.70 78.60 84.40 14.33 ,92.40 83.40 CzHg 0.75 .00 T.1\ .00 1.06 .00 10.31 H2 .00 .00 .00 trace .00 .2^2i He.. 0.17 .24 .56 .40 1.64 .08 .16 N2 .. 2.6od 4.96 12.13d i5.iod 82.87d 6.46d 5.i9d 4.83r 4.96r i6.04r i6.i5d 6.69r 7.35r 1 Prof. E. H. S. Bailey, Progressive Age, Aug. 15, 1907. 678 e:ngine:e:ring chemistry It will be seen that the fuel value of the gas diminishes to- ward the west and southwest; for example, Dexter, 14.33; ^l"^" dale, 78.60; Blackwell, Okla., 8340. Although there is considerable difference in the composition of natural gas from different places in this field, yet in all cases practically the only valuable fuel constituent in the marsh gas. By the combustion of each 1,000 cubic feet of this gas, as prev- iously noticed, 2,000 cubic feet of steam would be found, and when this is condensed to water we shall have 36 quarts of water, or 9 gallons. This large quantity of water causes considerable inconvenience in the use of ordinary stoves and furnaces, on ac- count of dripping down the chimney and frequently soaking through the brick and defacing the walls. For gas engines, it seems to be admitted that the gas is best which contains a low percentage of hydrogen and of inert gas. The richer the gas, the less there is to be handled and stored; the lower the percentage of hydrogen, the less will be the flame temperature of the mixture and the slower the combustion, which is said to permit of higher degrees of compression, and conse- quently greater efficiency in the engine cylinder. Natural gas appears to answer these conditions admirably. Natural gas has a calorific value of about 1,000 B. t. u. per cubic foot. Methane is the principal constituent of natural gas. Ethylene is not an abundant constituent of gases. When we attempt to use natural gas in a furnace with an air blast, our experiments show that we can readily get temperature of 1,400° C. or 2,552° F. This will readily melt brass or copper. The furnace works admirably for the cupeling of gold and silver. In the ordinary bunsen burner a temperature of about 850° C. is attained. The flame goes out very readily, because there is no hydrocarbon to transmit flame. Bunsen burners with differ- ent proportions from those ordinarily used must be constructed for use with natural gas. They must have a large tube and a longer one for the mixing of gas and air, as so much air is required for complete combustion. The foregoing statements have reference to the theoretical EJNGINKERING CHE:mISTRY 679 amount of heat that can be produced by the combustion of the fuels. So theoretically: One ton of oil produces 40,ocx),cx)0 B. t. u. One ton of bituminous coal produces 25,000,000 B. t. u. One ton of anthracite coal produces 26,000,000 B. t. u. One thousand cubic feet natural gas produces 1,000,000 B. t. u. But combustion is incomplete and there is always an increased loss in burning some parts of the fuel. We must heat and carry off four parts of nitrogen for every part of oxygen that we use.' The water, also, is changed to steam and carries off large quan- tities of heat. From some experiments made in the university laboratory some time ago, it was found that one ton of soft coal was equiv- alent in water evaporation power to about 20,000 cubic feet of gas. ACETYLENE.! Manufacture. — 7\cetylene is produced from calcium carbide by the action of water on the latter, the formula for this reaction being CaC^ + 2H0O = C2H2 -f Ca (OH)2. Sixty- four parts by weight of calcium carbide and 36 parts by weight of water give 26 parts by weight of acetylene and this formula is very closely borne out by practice in the best forms of generators and with a good quality of carbide. The action is attended by the evolution of great heat, the liberation of gas being instantaneous on the application of water, and the residue of this action being slaked lime. The preparation of this gas is accomplished in generators, which may be sub-divided into two classes, namely, (i). Water feed, (2). Carbide feed. At the inception of the acetylene industry, the first class of generator was veiy often employed and contained a chamber in which were trays containing the carbide in lumps, and superim- 1 Written and communicated to the Editor by R. E Briickner, M. E. of New York. He graduated from Stevens Institute ot Technology in iS86, and is considered a standard authority upon acetylene. 68o e;nginee:ring chemistry posed tipon it, perforated pipes or other devices for the admission of water to the carbide, The regulation of the supply of water was governed by the pressure within the generator chamber pro- duced by the gas generated. This type of generator has the following defects, which were very soon recognized and this particular type is now practically obsolete. Its first fault was what is known as "dry generation," or generation of gas with insufficient water. This produces, primarily, hot gas and, because of the insufficient water supply, it has no chance to give up any impurities which, in a correctly designed machine, may be given off to the water. Gas produced by this process contains all of the phosphureted and sulphureted hydrogen, together with the ammonia, of the carbide. The presence of air in these generators would be a distinct menace for, whereas the temperature of dissociation of pure acetylene at atmospheric pressure is 1,436° F., this temperature is materially reduced by pressure and by mixture with the gas of a per- centage of air. These troubles soon led to the abandoning of this type of generator. Carbide Feed Generators. — At the present time, the generators in common use are fitted with a hopper in which the lump carbide is stored and from which it is automatically fed into a chute en- tering the generator chamber. (Fig. 134). This chamber is filled with water and, by the rules of the National Board of Fire Underwriters, must contain one gallon of water for every pound of hopper capacity in carbide. The evolution of gas and its use from the generator automatically operates the feeding mechanism, so that only the necessary amount of carbide is fed into the gen- erator to supply the demands thrown upon it. For central station work, two methods are pursued. One is to use a generator with water motor attached to its feeding mechanism; the other, to feed by hand as the presence of a man is always necessary in a plant of any capacity. (Fig. 135.) The generator itself is fitted with grids located about a foot from the bottom of the chamber and upon these grids the carbide falls and is surrounded by water. This permits it to give up every particle of its gas and decompose completely instead of being DNGINEJERING CHEJMISTRY 68 1 Fig. 134.— General acetylene generator (not automatic). Descriptive references. — A, hand hole through which hopper is filled with carbide. B, hand hole for cleaning feed drum. C, hand hole giving access to generating chamber. £>, Carbide hopper. £, rachet wheel operating feed drum. J^, gas main from generator. G, gate valve. H, overflow pipe. /, vent pipe. /, water supply pipe. A', blow off pipe. L, blow off seal. M, blow off seal overflow. N, drain pipe. O, drain valve. P, carbide deflector. Q, flange joint connecting hopper with generating chamber. 682' e:ngine:e:ring chemistry allowed to fall into the sludge and eventually pass out of the generator with the water before its complete decomposition. Actual results on long runs in plants in this country show that the carbide delivered from the Union Carbide Company's Niagara Falls Works averages 4^/2 cubic feet per pound of carbide. This, of course, presupposes the use of a correctly designed generator. Fig. 135. — The general acetylene generator (automatic) for use of town, shop and residence lighting. Purification. — Gas generated from the best carbide and in the best generators carries with it certain impurities which nothing but a chemical cleaning will remove. The effects of these im- purities in the gas are apparent in that they form a certain haze over the flame or produce tails or fringe on the top of the flame. If glassware be superimposed above the flame, a deposit of these impurities accumulates on the glassware. The most objectionable of these impurities are phosphorated and sulphureted hydrogen. The quantities of these impurities vary with the gas produced from different carbides, but even in minute quantities, amounting to less than i per cent., both EJNGINERRING CHE:mISTRY 683 sulphureted and phosphoreted hydrogen affect the operation of burners, causing their stoppage and causing a deposit upon glass- ware. To eliminate these impurities, various reagents have been tried, the ones most generally known being puratylene, heratol, and Frankolin. These, we may say, have been arrived at by a process of elimination. Puratylene. — Puratylene is a method of using safely chloride of lime and consists of mixing chloride of lime, slaked lime, and calcium chloride into a thick paste and then dr}dng it at a tem- perature which suffices to drive off a part of the water of crys- tallization without disturbing the chloride of lime. In this man- ner a very porous mass is produced of a certain degree of hard- ness. The lime is supposed to attack the sulphureted hydrogen and also to retain any free chlorine, whereas the chloride of cal- cium absorbs any ammonia which may be present. The principal objection to this material is that it is likely to give up free chloride. It is not a very good purifier and acts upon the acetylene itself. Aside from this fact, the chlorine products are not sufficiently removed by the lime. The results of qualitative analyses are given on page 257 of Caro, Ludwig & Vogel's Handbuch for Acetylen. Heratol. — Due to the experiments of Ullmann, the use of chromic acid as a purifier, or oxidizing agent, for acetylene be- came known. Under the name of Heratol, a substance is pre- pared which consists of making a solution of ten pounds of water, three of chromic acid and one of sulphuric acid and saturating nine pounds of infusorial earth with this solution, producing a reddish mealy mass, very much resembling corn meal. This material is not sufficiently acid to attack metal and can, there- fore, be held in ordinary galvanized metal boxes. The gas is permitted to pass through these either from the bottom to the top or from the top to the bottom, and, in passing through the mass, is broken up into minute streams by the finely divided mass. The sulphureted and phosphoreted hydrogen are completely removed from the gas by this process, provided the gas is not passed through the mass at too rapid a rate of flow. Under the 684 ENGINKEJRING CHEMISTRY proper condition of flow, one pound of this mass purifies per- fectly 200 cubic feet of gas. Frankolin. — The use of this material is due to the researches of Dr. A. Frank, after whom it is named. It consists in sat- urating a mass of infusorial earth with a solution made up of one part of cuprous chloride, ten parts of 25 per cent, hydro- chloric acid, and ten parts of water. The finished mass is of a mealy consistency and is utilized the same as heratol, with the exception that it is too strongly acid to be contained in metal vessels and must, therefore, be held in earthenware, clay, cement, or enameled iron boxes. This preparation very readily attacks the impurities in the gas and completely removes the phosphoreted and sulphureted hydrogen together with the principal organic phosphorus and sulphur compounds. In fact only a very small percentage of these compounds is not removed by the cuprous chloride solution. The purification is accomplished by means of precipitation of the various compounds into copper sulphide for the sulphur compounds, and into copper phosphide for the phosphorus com- pounds, or they remain in the solution. Its most important prop- erty is that of removing silicon hydride as effectually as it does the phosphorus and sulphur compounds. It attacks the im- purities far more readily than the other purifying reagents, and, therefore permits a more rapid flow of gas during purification. Acetylene purified by this method may be tested for purity by the use of silver nitrate. A piece of filter paper saturated with a 5 per cent, solution of silver nitrate, when held in the flow from a jet, will not be discolored by the gas, whereas the raw gas im- mediately turns such paper black. Drying. — Wherever gas is to be used for compression and storage it is necessary that the moisture be removed from it be- fore entering the compressor. This is accomplished by passing the gas through quick, or unslaked, lime, and then through cal- cium chloride. Candle-power. — Acetylene is commonly known as a 50 candle- power gas, which means that when it is burned in the most ad- vantageous way in a naked flame, it delivers 50 candle-power per EJNGINEERING CHEjMISTRY 685 foot. This particular test was conducted on a Bray fish-tail burn- er. For commercial purposes, this type of burner is not practical for acetylene. The gas is too rich and, in the course of time, carbonizes the burner. Dolan's invention produced a burner with a cupped orifice. Into this cup air ducts lead. These air ducts supply air around the jet of gas and, in a measure, dilute the outer zone of this gas so that the portion of it in contact with the burner is not sufficiently rich to deposit carbon, without introducing enough air to seriously reduce the efficiency of the flame in candle-power. To this type of burner may be attributed the reduction in candle-power below that obtained from the Bray fish-tail burner, but by this form of burner the commercial use of the gas is made possible. Two jets are caused to impinge upon one another, thereby flattening out into a flat flame. With this commercial type of burner, it is not possible, as before stated, to obtain 50 candle-power to the foot, but a fair average of the various burners now manufactured would give the following results. The burners here referred to are the Von Schwarz Perfection burners. Burner Consumption Candle-power C. P. per foot I ft. 1.036 45.4 44. V^ ft. .74 30.0 40.5 /2 ft. .49 16.8 34.3 Va ft. .28 4.9 17.5 With the use of an incandescent mantle with acetylene, the candle-power may be very greatly increased per cubic foot, but, up to the present time in the United States, mantles have not been brought into general use in connection with this gas, as its intrinsic illuminating power is so high. At the present time, this industry covers a very large field. Acetylene is already in general use as a means of illuminating isolated houses, small towns, railroad cars, railroad signals, loco- motive headlights, pleasure yachts and all kinds of harbor craft, and various aids to navigation, such as buoys, beacons, stake lights and large shore lights, oxy-acetylene welding, and auto- mobile headlights. Railroad lighting is accomplished by means of compressed 686 ENGINEERING CHEMISTRY gas stored in cylinders filled with asbestos and saturated with acetone, said cylinders being held under the car and gas piped through a suitable reducing valve to the car lamps where it is burned the same as any other gas. On locomotives, the tank is usually placed under the running board on the fireman's side and the gas pipe to a headlight equipped with a parabolic reflector, the flame being located at the focus. The boat lighting by this method is much the same, the gas being piped to the various state- rooms and there burned in ordinary fixtures. For isolated houses, an automatic generator is used, as de- scribed in the earlier part of this paper, the generator being fed with carbide and its operation being governed by the use of the gas from it. Town plants are operated much in the same manner as the city gas plants, a central generator and store holder being installed and the gas piped from this store holder to the various parts of the town. A very important field of operation, which is being taken up by the governments of all the different countries, is the light- ing of navigation signals by means of this gas. Fig. 136 shows one type of buoy extensively used in Europe in connection with the absorbed, or dissolved, gas system. The gas is stored in the cylinder at the extreme bottom of the drawing, under a pres- sure of ten atmospheres, and is piped from the cylinder through the central tubular section to the tower and the lantern. The floating body consists of a steel shell made up of a bumped head at either end and a cylindrical portion, the parts being wielded together by means of the oxy-acetylene torch, which is the latest development in the acetylene field. On this particular type of buoy, there are no rivets, all the joints being made by the use of the torch so that the whole buoy is virtually one solid shell of steel. A buoy, such as is shown on Fig. 136, contains in its gas reser- voir approximately 1,100 cubic feet of acetylene and, with a flash characteristic of ^, for instance one second of light and four seconds of darkness, would run continuously without recharging or attention for a period of 275 days, and with a flash character- E^NGINKERING CHEMISTRY 687 istic of i/io would run double this time, or 550 days. Various other types of buoys are built to meet different requirements. Assuming that such a light as the one described for this buoy is to be placed as a fixed light at some inaccessible place, saving in gas and consequent length of service is, therefore, of the ut- Fig. 136. most importance. This was recognized by the invention of what is known as the sun-valve, or a device which automatically closes off the gas in the day time and relights it at night. Fig. 137 shows this apparatus. It consists of three polished, silver-plated rods supporting a frame which acts upon the central cylinder, the surface of which Fig. 137- I^NGINK^RING CHEMISTRY 689 is coated with lampblack, and its operation is based upon the fact that two surfaces, one polished and the other dead black, will absorb different amounts of light, or, in other words, the one which absorbs the light rays becomes heated to a greater extent than the one which reflects the light rays. The dark one, there- fore, expands more than the polished one. This principle was made use of in this mechanism in that the expansion and con- sequent lengthening of the black cylinder causes the closing of the vale 'J" on the seat ''g." (Fig. 137.) On a sunshiny day, this action immediately takes place and the setting of the sun brings both the polished and the black elements back to their original condition and permits the free flow of gas to the lantern where it is at once ignited from the pilot flame which burns con- tinuously. It has been found that, in latitudes resembling New York, the saving accomplished by such a device is approximately 40 per cent., or, in other words, the light just described which without the sun-valve and utilizing a flash characteristic of i/io light and 9/10 darkness, runs for 550 days, would by the use of the sun-valve run 915 days continuously. Oxy-acetylene Welding. — The exceedingly high temperatures obtainable by the use of oxygen and acetylene have been utilized in the welding of steel. This is accomplished by means of a torch, many types of which exist at the present time, in which both gases are led to a common nozzle and there mixed in the correct proportions and pressure so as to form, in burning, an arc of white flame w^hose temperature ranges from 5,600° F. to 6,300° F. The velocity of the gas in leaving the tip is such that the tip itself does not burn off or melt, but this small cone of flame can be brought down and impinged directly upon the two elements which are to be united. A Swedish iron wire filler is used and, when the edges are brought to the welding heat, this wire is fed in and melted in the same manner as solder would be, thus forming an absolute homogeneous and powerful union of the two pieces. An oxy-acetylene weld, when properly made, will run as high as 90 per cent, of the original strength of the plate, and with pure gas and the proper torch, this class of work can be done far more rapidly by this method than by any other. 44 690 ENGINEJERING CHEMISTRY These same torches are usually supplied with a cutting at- tachment which consists in first applying the heating flame and raising the temperature of that portion of the steel to be cut to a white heat. As soon as this is done, a high pressure auxiliary jet of oxygen is turned on to the white hot metal. The carbon of the steel is immediately consumed and, in a way, furnishes the fuel for the cutting. A high pressure jet of oxygen, say at 100 to 150 pounds pressure, will cut through 3 inches of steel far more rapidly than a saw and the entire cutting and welding equipment can be carried around in a wheelbarrow\ The con- venience of this system has led to its use in difficult places and for repair work on automobiles where economy of time is requi- site. VALUATION OF COAL FOR THE PRODUCTION OF GAS. Take 100 grams of the coal in small lumps, so that they may be readily introduced into a rather wide combustion tube. This is drawn out at its open end (after the coal has been put in it) so as to form a narrow tube, which is to be bent at right angles; this narrower open end is to be placed in a wider glass tube, fitted tightly into a cork fastened into the neck of a somewhat wide-mouthed bottle serving as tar vessel. The cork alluded to is perforated with another opening wherein is fixed a glass tube bent at right angles, for conveying the gas, first through a cal- cium chloride tube, next through L^iebig's potash bulbs contain- ing a solution of caustic potash, having lead oxide dissolved in it. Next follows another tube partially filled with dry caustic potash and partly with calcium chloride; from this last tube a gas- delivery tube leads to a graduated glass jar standing over a pneu- matic trough, and acting as gas-holder. Before the ignition of the tube containing the coal is proceeded with, all the portions of the apparatus are carefully weighed and next joined by means of India-rubber tubing. After the combustion is finished, which should be carefully conducted so as to prevent the bursting or blowing out of the tube, the different pieces of the apparatus are disconnected and weighed again. The combustion tube has to be weighed with the coal after it has been drawn out at its open ENGINEERING CHEMISTRY 691 end, and with the coke after the end of the combustion, when it is again cold, and for that reason care is required in managing it. We thus get the quantity of coke, tar, ammoniacal water, carbon dioxide, and hydrogen sulphide (as lead sulphide), and the gas is measured by immersing the jar in water, causing it to be at the same level inside and out. Empty the Liebig's bulbs into a beaker and separate the lead sulphide by filtration, wash well, dry and weigh. From the lead sulphide the hydrogen sulphide present is calculated. This pro- cess, devised by the late Dr. T. Richardson, of Newcastle-on- Tyne, was found by him to yield very reliable results, so as to be suitable for stating what quantity of gas a ton of coal thus anal- ized would yield. ^ Newbigging's Experimental Plant for the Determination of the Gas-Producing Qualities of Coal. The apparatus consists of the following: Retort. — Cast iron, 5 inches wide, 4.5 inches high, 2 feet, 3 inches long outside, and 0.4 inch thick. Ascension Pipe. — 2-inch wrought tube. Connections. — 1.5 inch wrought tube. Condenser. — 12 vertical, 1.5 inch wrought tubes, each 3 feet 6 inches long. Washer. — i foot long, 6 inches wide, and 6 inches deep. Purifier. — i foot 2 inches square, 12 inches deep, with 2 trays of lime. Gas-holder. — Capacity 12 cubic feet, with graduated scale at- tached. Amount of coal to be taken for each test i/iooo part of a ton, or 2.24 pounds. Care should be taken to obtain a fair aver- age sample of the coal to be operated upon. For that purpose at least 50 pounds of coal should be broken up into small pieces and thoroughly intermixed, and from this three different charges are to be taken. The retort should be at a bright heat before the introduction of the coal and maintained at that temperature during the test. If from any cause the temperature is much reduced, the 1 Crookes' " Select Methcxi in Chemical Analysis," p. 607. 692 ENGINEERING CHEMISTRY test will not be satisfactory. The time required to work off the charge of 2.24 pounds will range from forty to sixty minutes, according to the character of the coal. The illuminating power of the gas given out from each charge should be ascertained by the Bunsen photometer, no other being sufficiently trustworthy for that purpose. The average of the three is then taken, both for yield of gas and coke and for the illuminating power of the ENGINEERING CHEMISTRY 693 gas, and this fairly represents the capabilities of the coal. The further conditions to be observed are that the holder be emptied of air or of the previous charge of gas, and that the condenser be drained of its contents. The test charge may be continued until the v^hole of the gas is expelled, or otherwise, depending on circumstances. In comparing two coals, an equal production from both may be obtained, and the comparative illuminating power then ascertained. The coke and "breeze" should be carefully drawn from the retort into a water-tight receptacle made of sheet iron closed by a lid. This is then placed in a bucket or other vessel of cold water, and when sufficiently cooled, the coke is weighed. For ascertaining the quantity of tar and ammoniacal liquor produced, drain the yield of three charges from the condenser and washer and measure, this in a graduated liquid measure. The number of fluid minims in a gallon (English) is 76,800. Then: Pounds. Pounds per ton. ^ ^, , r ^ r The total num- ^ (The weight of ^ nl^mrortar ' ber of minims I 6.75 three charges : 2240 : ^ ZlTc^uol ob- ] ' ] ^'il'^li'l^T \ [ of coal J 'j^ ^^^^^^ J from^a^tonof J and this amount divided by 76,800 gives the gallons of tar and liquor produced per ton. A good variety of gas coal should pro- duce from 2,240 pounds of coal 12,000 cubic feet of gas, illumi- nating power 20 sperm candles. Newcastle coal on an average produces 12,700 cubic feet of gas per ton of coal, illuminating power of 15 sperm candles. As an example of the method of determining the value of a rich cannel coal for production of gas and coke, the results of a working test upon 1,196 pounds of coal, made by the writer, are given herewith. The analysis of the coal gave the following percentages : Per cent. Moisture at 103° C. (>4 hour) 1.31 Volatile and combustible matter (ignition, 7 mmutes) . . 57-99 Fixed carbon 28.36 Sulphur (KNO3 -f NaaCOs fusion) 2.54 Ash 979 Total 99.99 694 ENGINEERING CHEMISTRY The testing plant was especially designed and arranged for trials of this character, having a capacity for each charge of 250 pounds of coal. The coke produced was as follows * Pounds Pounds Pounds Pounds Pounds Pounds Coal 224 94 68 224 98 224 97 150 67 224 98 Or>V*» rjmfl npffl . or the coal produced 43.3 per cent, of coke. Theoretically, from the analysis of the coal, the amount of coke that could be pro- duced would be 44.2 per cent., a difference of 0.9 per cent. The coke, as shown by analysis, was composed as follows : • Per cent. Moisture 1.30 Volatile and combustible matter 2.24 Fixed carbon 79-82 Sulphur 1.83 Ash 14.81 Total 100.00 The amount of coal gas produced from the 1,196 pounds of cannel coal was as follows : Coal Pounds 150 224 224 150 224 224 Gas produced Cubic feet 789.7 i,in.o 1,156.0 759-9 1 ,090,0 1,160.0 6,066.6 Equivalent per lone ton (2,240 lbs.) I '.793 II, 1 10 11,560 11,349 10.900 11,600 68,302 or at the rate of 11,384 cubic feet per ton of 2,240 pounds, the gas having a candle-power of 36. ENGINEERING CHEMISTRY 695 MANUFACTURE OF OIL GAS. Oil gas is usually formed by vaporization of mineral oil at high temperatures. Two processes are in use: the "Pintsch" and the **Keith/' the former probably representing 90 per cent, of the production of oil gas. In the manufacture of Pintsch oil gas, in the United States, "mineral seal" oil is often used. This oil is a petroleum product having a specific gravity of about 0.840, flashing-point 266° F., and fire test 311° F. Several analyses of this oil, by the author, give carbon 83.30 per cent., hydrogen 13.20 -per cent., the remainder being oxygen, nitrogen, etc., and the analysis of the gas therefrom gave: per cent. CO 0.5 CH. 57.7 H 34 r Benzene vapor, CgHg '\ Illuminants } Propylene CgHg [ 38. i ( Ethylene CjH^ ) The heating power would indicate 1,390 B. t. u. per cubic foot, products condensed. W. Ivison Macadam, F. C. S., tabulates the results of a series of his tests upon the Pintsch and Keith oil gas as follows : Paraffin Oii, Into Gas Specific gravity of the oil Weight of I gallon of the oil Number of gallons per ton of oil Flashing-point Burning-point Gas from i gallon of oil Gas from i ton of oil Candle-power of gas Illuminating value of i cubic foot in grains of sperm Illuminating value of i ton in lbs. of sperm Average of trials Average of trials with Keith's with Pintsch's apparatus. Apparatus. 0.875 0.877 8.758 lbs. 8.779 lbs. 255.76 255-15 289° F. 295° F. 347° F. 354° F. 84.93 cu. ft. 90.03 cu. ft. 21,720 cu. ft. 24,757 cu. ft. 61.38 candles. 60.82 candles 1.473 grains 1,459 grains 4,570 lbs. 5,160 lbs. 696 ENGINKDRING CHEMISTRY The Manufacture of Pintsch Gas, Its Distribution and Uses. By R. Vi.iii inches are also used. In case either light is changed or moved for any reason, it may easily be put back in place by placing it centrally over the line indicated on the table. To facilitate the adjustment, 2 plumb-bobs are hung over each of the lines at the ends of the table, so it is easy to see whether the flames are properly centered in one direction. In the other direc- tion they are centered by sighting along the bar. The bar is placed at right angles to the two lines laid out on the table and cen- trally between them. It is laid out in inverse squares so that **i" is in the center. If the length of the bar is y and the dis- ( y xY tance from the candle is x, the candle-power is ^ — —. The mark that indicates 4 candle-power is twice as far from the light to be measured as it is from the candle, 9 is three times as far, etc. The bar should be made so that it may be raised or lowered at pleasure, and be planed to a thin edge on top so that no light will be reflected from it on the disk. On the bar is a sight-box in which a paper disk is placed at right angles to and centrally over 'the bar. There are several kinds of disks used, but the one most commonly preferred in this country is made by taking a piece of white sized paper of medium thickness, and cutting out of the center a many- pointed star about i^ inches in diameter outside the points. This paper with the star cut from the center is then placed between two pieces of tissue paper and the three held together either by placing between pieces of glass or else by being fastened with thin starch water. At the back of the sight- box are two mirrors, so placed that the observer may stand in front of the bar and see both sides of the disk. On the front of ENGINEERING CHEMISTRY 707 7o8 DNGlNEiDRING CHEMISTRY the sight-box a hood is so placed as partially to screen the eyes of the observer from the lights. At one end of the bar is the light to be tested. This is con- nected to a pipe sealed in mercury, so that it may be moved back and forth or raised and lowered at pleasure. It is usually ar- ranged with a micrometer cock so that the rate of flow may be regulated as closely as may be necessary. At the other end of the bar is a candle balance. The balance is usually arranged for two candles and all readings are multiplied by 2. This balance is so constructed that the position of the candles may be adjusted vertically or horizontally. This end bar is so arranged that the candle balance may be removed and a standard burner put in its place. The standard burner commonly used is a Sugg Argand burner, size D. This is covered with a thin sheet metal chimney, ij4, inches diameter. This chimney has an opening on one side, ^^/go inch high and i}^ inches wide. On the opposite side the chimney is cut away to prevent light being reflected through the slot in front. Th^ stand- ard burner, like the one through which the gas is tested, is so arranged that it may be adjusted in all directions. A meter to measure the gas is necessary. As gas is burned at the rate of 5 feet an hour when being tested, the meter is so geared that one of the hands makes a complete revolution each time a twelfth of a foot of gas passes. A clock is attached to the meter with a large second, hand, so when the meter hand mentioned and the second hand move together, gas is passing at the rate of 5 feet an hour. In addition to these hands are one indicating feet and one minutes. Some meters are furnished with a third set of hands reading feet and hundreds. The meter has a thermometer to show the temperature of the gas and a universal level so that it may be properly leveled. On the side is a glass gauge and a mark indicating the height of the water, which should always be constant. The pipe connections to the meter are so arranged that opening a cock will allow the gas to pass around instead of through it. This permits the op- erator to start or stop the meter at pleasure without interfering with the light. e:ngine:kring chemistry 709 A pressure gauge connected with the various parts of the ap- paratus enables the operator to ascertain the pressure of the gas at different points. One of these connections is to the pipe a short distance below the test burner. This gives the pressure near the point of ignition. The pressure is read in inches and fractions of an inch of water. A gas governor is connected before the inlet to the meter, which reduces the pressure to about i^ inches of water. Beyond the meter is a smaller governor which reduces the pressure to about 0.9 of an inch and prevents alteration of the flow of gas due to the irregularities in the meter. Black screens are arranged to screen the eye of the observer from the light. These are sometimes fixed and at others set on the bar. The latter arrangement is preferable, as they may be moved to suit different positions of the sight-box. For testing gas of not over 18 candle-power the Standard I^on- don Argand burner is used. For higher candle-power gas the ordinary sawed lava tip is best. The latter is commonly known as the batwing burner. The photometer should be set up in a small, light-proof room with dead black walls. The latter can be hung with black vel- vet or painted with glue and lampblack. Great care should be taken to insure proper ventilation without draft. The tempera- ture of the room should be kept as near 60° F. as possible, and the air should not be allowed to become vitiated by the products of combustion. The table should be set so that readings may be taken from both sides of the bar. Manner of Using the Photometer. When one starts to use a new photometer, or one with which the experimenter has not previously worked, the instrument should be carefully verified. First make sure that the lines defining the distance between the lights are the proper distance apart and parallel, and that the bar is perpendicular to and midway between them. Next" see that the bar is level. The disk must be at right angles to the bar, and the small pointer under the sight-box in line with the disk. 7IO e:n GIN BERING CHEMISTRY The two mirrors should be made of the best plate glass and well silvered. They should be kept clean. The disk should exactly bisect the angle made by the mirrors. The bar should be veri- fied so that the operator may be sure that it is properly divided, and the meter should be tested with a meter prover. In test- ing the meter be sure that the temperature of the room, of the water in the meter, and of the water in the prover are the same. The pressure gauge should be verified by a U-shaped water gauge. The knife edges of the candle balance should be clean and sharp, and the lever should be free to move without rubbing. The weight for the candle balance should be weighed on an analytical balance to be sure that it is correct. For testing coal gas no choice is allowed in the burner, but when water gas or any high grade gas is to be tested it is neces- sary to get a burner suited to the gas. The most suitable burner can be quickly determined by experiment, and the greatest efifi- ciency is usually obtained with a burner of such size that the gas is almost on the point of smoking. When the photometer light is burned continually, as is usually the case in gas works, the tip on the fiat-flame burner should be changed at intervals of two or three weeks. Care should be taken that the tip is smooth. Any tips that are chipped on top or rough in the slot should not be used. In preparing for a test, the burner and candles should be placed in their proper positions and at such a height that the center of the flames will be on a level with the center of the disk. The height of the candle flame is taken when the candle end of the balance is down. The gas should be burned long enough to be sure that the apparatus is cleaned out and that fresh gas is being burned. Before starting it is necessary to control the pressure under the burner so that it will not vary during the test. The governor on the outlet of the meter will do this if it is in order. If the pres- sure varies, the governor must be cleaned before starting the test. During the test the pressure gauge must be shut off, as in case there is change of pressure it will store or give out enough gas to vitiate the result. The meter should be level and the water at the proper height. EINGINKERING CHEMISTRY 7II The wicks of the candles should never be touched. The candles are lighted and allowed to burn until the wick curls over to the edge of the flame and burns away as the candle is con- sumed. The end of the wick should glow. No test should be started until th^ wicks are bent over and the ends are glowing. The candles should always be burned eight or ten minutes before starting a test. A common practice which gives good results is to allow the candles to burn eight or ten minutes and then extin- guish them for two or three minutes. The candles are then re- lighted and allowed to burn about two minutes before starting the test. They are commonly placed in the holders in such a way that the ends of the wicks are as far away from each other as possible. When the apparatus has been brought to the proper condition for testing the flow of gas is adjusted to as near 5 feet an hour as possible, and the meter is allowed to run until the twelfth-of- a-foot hand points to o, when it is by-passed. The clock is stopped at o. The candles are counterbalanced by the sliding weight on the balance lever until the weight almost carries the lever down. In a few seconds the candles burn sufficiently to allow the bal- ance to fall, and at that instant the meter and clock should be started. As soon after as possible the 40-grain weight should be dropped into the scale pan, which brings the candles down again. The operator should always move about the room deliberately so as to avoid as far as possible creating currents of air. The candle flames must be still before beginning to take readings. A reading should be taken every minute for ten minutes. When the screen is apparently illuminated equally on both sides it should be moved a little to the right and to the left, and in each case the illumination on that side should increase. Five readings should be taken on one side of the bar and the sight- box turned around and five taken from the opposite side. In case the bar is accessible from only one side, the readings should be made with one eye and the screen turned in the sight-box after half have been completed. This will eliminate the errors due to possible dift"erences in eyes and in the sides of the screen. The last reading should be taken during the first half of the 712 DNGINEEJRING CHEMISTRY tenth minute and the times noted when the candle balance falls, and when the gas hand completes its tenth revolution. The tem- perature of the gas and the reading of the barometer should also be noted. After this the candles may be extinguished. They should be blown out and the ends of the wicks extinguished with a piece of sperm. The wicks should never be touched with any- thing else. If the candle balance falls in less than nine and one-half or more than ten and one-half minutes, or, if the gas hand takes less than nine and one-half or more than ten and one-half min- utes to make lo revolutions, the test should be discarded. Long practice has shown that withiji these limits the light given by the candles vary approximately with the consumption of sperm and that given by the burner approximately with the gas consumed. If the candles take x seconds to burn 40 grains and the gas hand y seconds to make 10 revolutions, the average reading V 600 V • multiplied by 2 should be multiplied by -f — X or - — 600 X X This will give the candle-power of the gas uncorrected for temperature and pressure. The standard pressure is 30 inches of mercury, and the stand- ard temperature is 60° F. To correct the pressure multiply by 30 and divide by the barometric reading. In correcting for tem- perature the gas is assumed to be a perfect gas saturated with water-vapor. The following is the formula for correction for pressure and temperature : 17.64 (/; — «) 1 71 =■ 460 1 Numerous inquiries having been received for the derivation of this formula, it is given as follows : A gas expands or contracts 1/490 of its volume at 32° F. per a change of 1° F. 492 — 32 = 460 = temp, of 32° F. on absolute scale. 460 + 60 = 520 = temp, of 60° F. on absolute scale. The volumes of a given quantity of gas are to each other as the distances from the absolute zero. If 60° F. = 520° absolute is taken as the standard temp., the correclion for a dry gas for temp, is : V X ^^ ft. [v = vol. at temp. P F. = (460 + P) abs] 460 {Continued on page ■/ 1 J ) e:nginekring chemistry 713 n ^ the number by which the observed volume is to be multi- pHed to reduce it to 30 inches and 60° F. ; h = the height of the barometer in inches ; t =^ the temperature Fahrenheit; a = the tension of aqueous vapor at ^°. The table on page 714 will facilitate corrections for various pressures and temperatures. Inasmuch as a flame is not perfectly transparent, a test made with it at right angles to the bar does not give the mean of the light that is emitted horizontally. The richer the gas the greater is the difference between the candle-power measured on the flat and on the edge of the flame. A gas that gives 25 candles meas- urement flat will not give over 19.5 candles measured on the edge. When the flame is at ah angle of 10° with the bar it gives almost as much light as when it is measured at 90° F. The best photometers are made so that the burner may be turned on its axis and the light measured at all angles. When it is desired to measure the light emitted by a burner at various altitudes, mirrors are used to reflect the light to the disk, as the latter is kept vertical and in the same horizontal plane as the standard burner. In such cases it is necessary to test very carefully the amount of light absorbed by the mirroi:s at all angles. There is a popular impression that photometrical work is not accurate and therefore not to be depended upon, but, if care is taken by the operator in his work, and all the apparatus is prop- erly adjusted, the error will be less than i per cent. By taking the average of a series of measurements the error can be reduced to a point where it is inappreciable. { Conini ued from page ji2. ) If 30" is taken as the standard pressure, the pressure correction for a dry gas is z/ + ' . The correction for temp, and pressure of a dry gas is 5- X 4- =17/3 '' 460 4- / ' 30 460 -t- / The pressure in the meter which balances that of the atmosphere (A) is due in small part to the water vapor (a). That due to the gas is {h — a). At the standard pressure and temp, h = 30", a = 0.5179", {h — a) = 30 — 0.5179 = 29.4821. In order to make the factor ^iVi i—^ -) reduce to I (one) at the standard conditions when ^(30) is reduced to 20.4821 > 400 + t ' ^ ^ to constant, i-/]A, must be raised to 17.64. & CO 0^a^C7^0 O O O - ►- ^ «N (N n rOrOrO-r-^-^toiOiOvOvOvO t^r^r^OOOO 6 d d d d d d d d d d d 6 d d d d d d d 6 6 6 6 6 6 6 6 6 6 6 i OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO °s f^Q "^r^O -^t^O "^r-^O rOr^O rot^O ror^O fO^ O rovO O -''-'OnOnOnOnOnOnOnOnO\OnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOh' °R t>.i-4 Ttt-^Hi Ti-t^M rfr^i-H Ttr^o "«tt^O "^t^O -rj-r^o rj-r^o -^ k Q ^tt^ O « M <-i C» CS (N mrOf^-rf^-rt^^ "~-N0 vO vO t^ t^ t^QO COCO ONO^ONO Q ON On On On On On OnOnOnOnOnOnOnOnOnOnOn On "On OnOnOnOnOnOnOnOn On O O 66666666 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6'^-^ -' 1 CtNO ONMNO OncsnO OnhvO OnMnO Onn*0 OnMnO On rOvO On rOvO On ro^O (y\ rO M HH ►_ (N M . t^ r^CO coco OnOnOnQ O Q <-i OnOnOnOnO>OnOnOnOnOnOnOn O*- OnOnOnOnOnOnOnOnOnOnOnOnOnOnQ O O 6666666666 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 ^ ^ ■^ m 1 tx. 1-1 -^00 ►-• -^00 1-1 -^00 - "^00 1-1 ■*00 1-1 ■^CO >-• lOCO •- u-jQO >- lOOO 1-1 lOOO HH f>< (S M rO'Ti'^'^rfTtlOiO lOvO vO NO t^ 1^ I^OO 00 COONOnOnQQQ'--i-i On OnOnOnOnOnOnOnOnOnOnOnOnOnOnOnO^OnOnOnOnOnOnOnOnO O d 6 6 6 6 6 6 666666666666 6 6 6 6 6 6 <^ ^ - •-; ►.; m i M NO On -i lOOO " toCO ►- »000 ^ « -' ^ -' i-' •-' -; •-; i W NO ON r* NO On rOvO On OnO On rOvO O rONO O rOr^O rOr^O Tfr^O Ttt^n- Tf rOrOrOrtrJ-rtiOiO lOvO nO NO 1^ f^CO COCOONONO^OQQ«*^i-i^'<**MrOfO OnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnO O O O O O O O O d d d 6666666666 6 6 6 6 6 6 6 «* «-" m m m ^ m -.' m «' « % t^ M Tj- rx « TtCO -1 Tj-CO "^ »O00 »- lOCO ^ m % M IOONM voOnNvO 0^ rONO On (N no O rOvO O fOr^O rft^O tTI^^m Ttl^^- tJ- Tf rl" ■rf »0 lO iOnO vO nQ I^ t^ r^CO COONONONOQQ»-ii-ii-irNMC-i lOOO cs vooncs ioOnmvO o^ cono on OnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnO O O O O O O O O O O O O 6666666666 6 6 6 6 6 6 ^ « « ^ >-:^^J,^J,,^^^^ m i 1-1 irjaO M lO 30 -• >/5oO - t^ t^CO COCO OnOnO^O O O ^ — 1-1 M fN M r'JrOrO'^'^'^'O »0>0 OnOnOnOnOnOnOnOnOnOnOnOnOnOOOOOOOOOOOOOOOOOO d d d d d d d 6 d d d d d i-«' i-! i-' i-<' -<■ «' ^ i-' « •-<* -" i-J ^' -' -<' -<■ « i-<* % O -^ t^ t-i -^00 '- lOOO CS lO On M NO ON rONO On rOvO O rO t-^ O •* t^ i- -^OO - ID 'oN^'oN o^ ON ON on^'on o^o^oIS 8 8 o o o ?o'o^o"o'o'o^o o"o "o 6 6 6 6 6 6 6 6 6 6 6 6 -' -* ►-' « -•* >-^ ^ ^ hI i-.' •-■ ^ ^ «" ►-: k,* ^ ^ Ji V. 1 O M M rO Tl- lONO t^oq ON M (N rO-^w^ND t^CO On O --i M rO -rf ionO t^CO 0> O 00 OO' OO" 00 00 00 CO 00 CO 00* tJN On Ox Ch ON On On On On <> d O d 6 6 6 O 6 6 i- NMC«MMWMNMWP<> r-< "^ ei O _e.,- TO a oS^ °'J^n 3T3 as « i> '-' c X -Mi s'c t a«';;;So CI-- U'S ^ 5 4^ .V3 O a 1 (LI 5| o a 3*^ °«ai2 o «-2 5 Ua>< §ia-s «.2i3 C3 CS 3 o CJC O •^ o aj: •o o 25 >4^« c •=! wo 5 -*-'.: ^ I a.S § = § -ccwal^o ,« 3-2 3 >.?t; ' a; 'S a •^^ CO CQ O I •3 o CO o a :3 o d 5 a> CO d U. ci §. n _ ^d ••o 5d l5 ^3 qa 3^- O — o (LI n (D si C8 Oh S5 1| .2° '77 eo+ e <" 1/ n go j:Sfl^ « >, .o O •- " K o «; fe I < ^.9 p^ •- 1 " D o w hT •0 o II 1 8:{?"R35'Svt?S % 1 ^OOOJ^COC g; 1^ 'o ir; ea £^«' % "^ .21.-2 c n [« 181 ^£ a o.t: d ;*' O «5 fea;, 2|lt rt « rt 2 ^•2-35 ja (u-j- o rt 0.-0 o rt u, 3 ct »- O-r rt « a .2.2 >.: rt to S §&ta. 3^ + O « u i/J'- (LI ^ax §a^ - « > (11 i> ( £^ o ^ *- .'O'O'^ 'O ^ o a^la o ^ W-a ^^ o> 1 .■Hi ■B^ ,x a.i^ 3 s c o §^ 3.C «a ^■n •2! Bo .tgu pS" ?iii III ^S| E-Ss 3 '*^ 'u-'H'^ l-l aSs .2** m Oo; . ^u-c c « a 8-= 3 >,> o §^fl --Qti • c; .^a 2^2 o ^ i«<-^ m o •- sis X & 5^3 -^ a; (J- T3 C o c _ _ §^ (LI O "^ —
  • « C u * C 3 bc-'o oil bo.5 a Si >,2 2 ^r, .2-0 aH t; ^?u£'§ w3 ^-•z;x*" ui; 0.3 ■"HP cfl'^ii ° - o 5 ^--^3 "^-o c:S a L ^ G a^ wot! T§Hi-2ls w-a aii'^.2 « JiC! ■<'utCl T) ^ ftp :^.2 S ^ — O (LI «.ii=Oua U V J- rt|.0 o curt ij (Li5 .£-3.2 .2.5P'3 - 5M . Q'o!^ ^ "^ -a;j: • ii ■»-• (« «> !» 3 c; ^^^ > OJ"" .ti -k-i rt PUS TJ-d (S rt >.^a C8 (lT^ 2-^ "■^ 2g5^5 2 55H"g rt a S-OQ O 2.-e' (U o "> O O "5 o &2 Ciii-.t: O O C 3 <«'0 O rt— •S4'Srt-i3i3^«rt^ ■5« ^a bC^C O t- "- bo W y .: -'5 ^ Ji ."2 •- ^ o ^ Sn-aO^ a « " c w '^ o 5y|fl.^^?--o:S^j ^fe.a-siJlo.^r.^^'' \^B^^. O 3 O .2 3.C w O _ "^ 'cS ^
  • * I.S3 oO wTii S^x .y fc'^s •2 ell S a ° Gt..^ P. "^M _, re •== C 3 c = 5-^=1 '^ ^ ? o 1. ^, V !«.-^ "^"Js ^ o ^ ^.t S ^ S « ^ c li «i"" ■^-.ti 1^^; ?.t; o rt S •6 S.^ 3 (U u t3 V- ■<-> «g^ o o'So o c o O ui art > C °-" te ^73 O lU _ _ _^ o c ifl -> £ S ^^ . 2-^ '^>'E tl ^-^ J- 5 o ° H flti ■= S bcf to fe 4; -art 2^ V Ji'U bi)° "5T3 a* o-P c rt*^P^rt. "o !r: o li >^ !.>•;- = "S'^sg.y 2fl^rt g .-o Bda6^%an Ot3^ ■^3 3 • (U C3 w ^a'-^u CO > c?'^5 t c O (U - t« O. I/} ^-^ ,^ 1) flCsa" 2 If o pj: ^ 3 S*rt*^ 3 a I rt Pi: c (u rtu ^ i .ii 3 3 ►* O «-SCpW2 lu ^ a; a « •-' .2 "!^ .5 i! •- ° 3 ■*-• S in C '^ 3 o 3 2 4; St^ yj-S M O 11 3^ >^ O X O 3 .- o Oi- O c« bCt/: ^ W g.^ 3 CS O «boy "" (U o O o >>'§ iig ^•'H ^s«i •"-•SrtOs 3 2'"l >.^Xi^fecortI^ o ■" ii 3 rt -5 rt a 5 ii«ao o&>-2 w o-;: 2 o - o n 5 3u--^C0^3ol- e'^>>o-^1^a^ S^rtoSi;a^ 2^.5 o."^ >^^o-^ o^fcs?i2 2-r':^- •O gc3 tl rt - 3„'- -I C53 O C3 o ^:^ OJ3 O CO o - xa^i ^ ■^ J-tBT3 S.2 1-2 8 ' ji ti'C 3-0 a^5^s PI'* •^'033° « cT, i! ^ - o ^' 3 -n o ".^-iSg 3 O !/) ? rt U5 ' 1; U 4, "" bccc « a a ^ ^- J' hf «^ (3 (U 3-« °3^^n '^s::;a^ ^ll_a_ •-0 fJ 1/3 "4 *-• ■»-' §Brtrtoi«e^- 2 "^X -c ^ 3 (U P g ^. ri eg ci'O j;i; 0< lU II i ^, lU •- i ^' * •" d ^- '^'^ a - •^•ria.;:^ c 96o 1.85 Gold Au 197.2 1065 ^9-^o . 0.0314 Helium He 4 >— 271.3 -267 1.98 (H = i) Hydrogen H 1.008 —256.5 -252.5 0.0694 (air=i) 3.4000 Indium In 115 155 Red heal 7.12 0.0336 i ENGINEERING CHEMISTRY ■33 Melting-, Boiling-Point, Specific Gravity and Specific Heat of the Elements with International Atomic VJe\ghts—{Confijiued) I Elements. Atomic weight. Melting- point, °C Boiling- point, °C Specific gravity Specific heat Iodine I 126.97 114. 2 184.35 4-94 0.0541 Iridium Ir 193.0 1950 •• 22.42 0.0323 Iron Fe 55.9 1804 7.68 O.I 138 Krypton Kr 81.8 -169 — 151-7 2 81 (air=i) Lanthanum La 138.9 810 6.15 0.0448 Lead Pb 206.9 327 [400-1600 11.34 0.0310 Lithium Li 7.03 186 — 230 —184 1.105 (air==i) 0.2175 Palladium Pd 106.5 I541 IT.80 0.0582 Phosphorus P 31.0 44.2 290 1-83 0.1699 Platinum Pt 194.8 1780 21 48 0.0324 Potassium K 39.15 62.5 757-5 0.87 0. 1660 Praseodymium Pr 140.5 940 6.47 . ... Radium Ra 225 • . . Rhodium Rh 103.0 2000 12.10 0.0580 Rubidium Rb 85.5 38.5 696 1.52 Ruthenium Ru I01.7 >i95o 8.60 0.06 II Samarium Sm 150-3 7-78 Scandium Sc 44.1 .... Selenium Se 79.2 170-180 690 4.26 00953 Silicon Si 28.4 1200 3500 2.49 0.1730 Silver Ag 107.93 955 2050 10.53 0.0557 Sodium Ma 23-05 97.6 877-5 0.97 0.2734 Strontium Sr 87.6 900 burns 2.54 Sulphur S 3206 ii5-'i9 444-6 2.04 0.1630 Tantalum Ta 18-. 2250 10.70 Tellurium Te 127.6 446 1390 6.20 0.0475 Terbium TW 160 . . . . . . Thallium Tl 204.1 301.7 1600-1800 n 85 0.0326 Thorium Th 232.5 11.00 0.0278 Thulium Tm 171 .... Tin Sn 119.0 232 1450- 1600 7-29 0559 Titanium Ti 48.1 3000 3 54 O.I 135 Tungsten W 184 1700 18.77 0.0356 Uranium U 238.5 800 18.68 0.0270 Vanadium V 51.2 1680 5.87 O.II53 Xenon Xe 128 — 140 — 109. T 4.42 (air^i) Ytterbium Yb 1730 Yttrium Yt 89.0 ... 3-80 Zinc Zn 654 419 91S 7 H 0.0935 Zirconium Zt 90.6 1500 4.15 0.0660 734 Engine:e;ring chemistry Conversion Tables Found vSought. Factor. Found. Sought. Factor. AI2O3 A\, 0.53398 Mg2P20, 2Mg 0.21883 NH.Cl NH3 0.31882 MiijOa 2Mn 0.69695 PtCle(NH,)2 2NH3 0.07692 MngO, 3Mn 0.72084 PtCle(NH,)2 N, 0.06329 MnS Mn O.63211 Pt 2NH3 O.17518 Hg HgO 1.07984 (NHJ^SO, 2NH3 0.25815 HgS Hg 0.86208 Sb,03 Sb., 0.83366 MoS Mo 0.49992 Sb^Oj Sb2 0.75046 NiO Ni 0.78524 Sb^Ss Sbg 0.71438 NiSO^ Ni 0.37849 AS2O3 AS2 0.75757 (NHJ^PtCle 2N 0.06329 AS2O5 AS2 0.65217 PbSO, Pb 0.68292 AS2S3 AS2 0.60928 Pt 2N O.14414 BaSO^ BaO 0.65654 Pdl, Pd 0.29448 BaSO, Ba 0.58790 Mg2P2G, 2P 0.27852 BiaOj 2Bi 0.80654 Mg2P,G, P2O5 0.63756 KBFI4 B 0.08683 U,P,0„ P2O5 O.I9817 AgBr Br 0.42556 CNH,)2PtCle Pt 0.43911 CdS Cd 0.77712 K2SO4 K2 0.44898 CdSO^ Cd 0.53786 K2SO4 K2O 0.54075 CaO Ca 0.71428 KaPtClfi K2O 0.19404 CaS04 CaO O.41158 AgCl Ag 0.75275 CO2 C 0.27278 Si02 Si 0.47020 CaCO, CO.. 0.44002 SiFl, Si 0.57878 BaC03 CO.; 0.22332 Na2S0, Na2 0.32435 AgCl CI 0.24725 Na2S04 Na20 0.43674 Qt,0, Cr, 0.68483 NaCl Na 0.39408 CrjOs 2CVO2 I. 31520 BaSOi S 0.13755 CoO Co 0.78696 BaSO, SO3 0.34346 CuO Cu 0.79858 SrSO, Sr 47674 Cu,S CU2 0.79827 Tl^PtClfi 2TI 0.50046 CaFl^ FI2 0.48088 Sn02 Sn 0.78681 BaSiFL 6F1 0.40783 XiO, Ti 0,60065 Agl I 0.54031 U3O, 3U 0.84873 Fe^O, Fe, 0.70000 Vd205 2Vd 0.56145 Fe,03 2FeO 0.89999 W0O3 Wo 0.79310 Li^COs Li2 0.18944 ZnO Zn 0.80338 MgO Mg 0.60375 Zr03^ Zr 0.73913 'Improvements in Methods of Chemical Calculations." Consult J. Anal. Chem., i, 402. ENGINEERING CHEMISTRY 735 Specific Gravity of Gases and Vapors. Gas or vapor Formula Molecular weight Specific gravity (air= I) Weight of one liter in grams at o° C and 769 m.m Acetone Acetylene Air Aldehyde Ammonia Amylic alcohol Arsenous anhydride Arsine Benzene Bromine Butane • Carbon disulphide Carbon dioxide Carbon monoxide Carbon oxychloride Carbon oxysulphide Chlorine cyanide Chloroform Cyanogen Ethane Ethet Ether acetic Ethylic alcohol Ethylene • Hydrobromic acid Hydrochloric acid Hydrocyanic acid Hpdrofluoric acid • Hydrogen sulphide (sulphur eted hydrogen ) Hydriodic acid Methane Methylic alcohol Nitric oxide Nitrous oxide Phosphine (phosphureted hydrogen Phosphorus Phosphorus pentachloride • • Phosphorus trichloride Propane • Selenium hydride Silicon chloride Silicon fluoride Steam Sulphur . . . • . Sulphuric acid Sulphuric anhydride Sulphurous anhydride Tellurium Tellurium hydride CgHeO Q,H, C,H,0 NH3 C5H12O AS^Og AsHg CeHe Bra C4H10 CS2 CO2 CO COC12 COS CNCl CHCI3 (CN), C,H« C4H10O C.HgOa C^HeO C,H, HBr HCl HCN HF H2S HI CH, CH4O NO N2O PH3 P4 PCI5 PCI3 SeH., SiCl^ SiF, H2O S H,SO, S63 SO2 Te^ TeH., 580 26 o 440 17.0 88.0 198.0 78.0 78.0 160.0 58.0 76 o 440 28.0 99.0 60 o 615 119 5 52 o 300 74.0 88.0 46.0 280 81.0 36.5 27.0 20 o 34.0 128.0 16.0 32.0 30 o 440 340 124.0 208.5 137.5 44.0 81.0 169.5 104 o I8.0 64.0 98.0 80.0 64.0 256,0 130.0 2 0025 9200 1 0000 1 5320 0.5960 3.1470 3.8500 2.6950 2.7700 5 3933 2.0041 2 6450 1 5290 0.9674 34163 2 0748 2 1244 4 2150 I 8064 I 0366 2.5650 3 0670 1.6133 0.9674 2.7310 1.2474 0.9456 06930 I 1921 4.4330 0.5560 I. 1200 I 0390 I 5269 I 1850 4-3550 3.6500 4.7420 1.5204 2.7846 5.9390 3.6000 0.6235 2.2000 2.1500 2.7630 2.234 8.9160 4.5276 2.5896 1. 1650 1.29378 1.9811 0.7707 4.0696 79105 3.4851 3.5821 6.8697 2.5914 3.4204 1.9662 I 2510 4.4174 2.6828 2.7473 4.4507 2.3360 1.3404 3.3170 3.9662 2.0862 1. 2510 3-5316 1.6131 1.2228 8960 1.5416 5.7456 0.7155 1.4483 1.3436 1 9745 1.5350 5.6318 4.7201 6.1299 1.9660 3.6011 7.6208 4.6554 0.8063 2 8430 2.7803 3.5730 2.8680 11.5310 5.8550 736 ENGlNEJIvRING CHEMISTRY Equivai^knt of Degrees Baume and Specific Gravity at 6o= FOR Liquids Heavier than Water. U5 ■ F. Degrees Baum^ = I45 — sp. grav. Degrees Specific Degrees Specific 1 Degrees specific Degrees Specific Baum6 gravity. Ban me. gravity. Baum^. gravity. Baura^. gravity. CO I.OOOO 18.0 I.1417 36.0 1.3308 54.0 1.5934 0.5 1.0007 18.5 1. 1462 36.5 1.3364 54-5 6022 I.O 1.0069 19.0 1. 1509 37.0 1.3426 55.0 611I 1.5 I.OI05 19.5 I.J555 37.5 1.3488 55.5 6201 2.0 1. 0140 20.0 I. 1600 38.0 I.3551 56.0 6292 2-5 I.0175 20.5 1. 1647 38.5 I.3615 56.5 6384 30 I.02II 21.0 I. 1694 39.0 1.3679 57.0 6477 3-5 1.0247 21.5 I.1774 39.5 1.3744 57.5 6571 4.0 1.0284 22.0 I.1789 40.0 1. 3810 58.0 6667 4-5 1.0320 22.5 1. 1837 40.5 1.3876 58.5 5763 5-0 1.0357 23.0 I. 1885 41.0 I 3942 590 6860 5-5 1.0394 23.5 I. 1934 41-5 1. 4010 59-5 6959 6.0 1.0432 24.0 1.1983 42.0 1 .4078 60,0 7059 6.5 1.0469 24.5 1.2033 42.5 1. 4146 60.5 7160 7.0 1.0507 25.0 1.2083 43.0 1. 4216 61.0 7262 7-5 1.0545 25.5 1. 2134 43-5 1.4286 61.5 7365 8.0 1.0584 26.0 1.2185 44.0 1.4356 62.0 7470 8.5 1.0623 26.5 1.2236 44-5 1.4428 62.5 7576 9.0 1.0662 27.0 1.2288 450 1.4500 63.0 7683 9-5 1. 0701 27.5 1.2340 45-5 1.4573 63.5 7791 10. 1. 0741 28.0 1.2393 46.0 1.4646 64.0 7901 10.5 I. 078 I 28.5 1.2446 46.5 I.4721 645 8012 II.O I.082I 29.0 1.2500 47.0 1.4796 65.0 8125 II. 5 I.0861 29.5 1.2554 • 47-5 1.4872 65.5 8239 12.0 1.0902 30.0 1 . 2609 48.0 1.4948 66.0 8354 12.5 1.0943 30.5 1.2664 48.5 1.5026 665 8471 13.0 1.0985 31.0 1. 2719 49.0 I. 5104 67.0 8590 13-5 I. 1027 31.5 1.2775 49-5 I.5183 67.5 8710 14.0 I. 1069 32.0 1.2832 50.0 1.5263 68.0 8831 14-5 I. nil 32.5 1.2889 50.5 1.5344 685 8954 15.0 I.II54 33.0 1.2946 5I.O 1.5426 69.0 9079 15.5 I.II97 33-5 1.3004 51.5 1.5508 695 9205 16.0 I. 1240 34.0 1.3063 52.0 1.5591 70.0 9333 16.5 1. 1284 34.5 I. 3122 52.5 1.5676 17.0 1. 1328 35.0 1. 3182 53.0 1.5761 17-5 1. 1373 35-5 1.3242 53-5 1.5847 ENGINEERING CHEMISTRY 737 Equivai^ent of Degrees Baume and Specific Gravity at 60° F FOR Liquids Lighter than Water Baume Weight in pounds-6o° F Specific gravity I^iquids lighter than water Per U. S. gal. Per cu ft Per barrel 42 gals lO 1. 000 8-337 62.368 350.2 II 0993 8.280 61.93 347 7 12 0.986 8.222 61.50 345-3 13 0.980 8.171 61.12 343-2 14 0973 8.H2 60.68 340.7 15 0.966 8.054 60.25 338.3 16 0-959 7.996 59-81 335.8 17 0.952 7.937 59-37 333.4 18 0.946 7.887 59.00 331.3 19 0940 7.837 58.63 329.2 20 0-933 7.779 58.19 326.7 21 0.927 7 729 57.S2 324.6 22 0.921 7.679 57.44 322.5 23 0.915 7.629 57.07 320.4 ■■^4 0.909 7.579 5669 318.3 25 0903 7.529 56.32 316.2 26 0.897 7.479 55.94 314.1 27 0.892 7-437 55.63 312.4 28 0.886 7-387 55.26 310.3 29 0.881 7.345 54.95 308.5 30 0.875 7.295 54.57 306.4 35 0.848 7.070 52.89 296.9 40 0.823 6.862 51.33 288.2 Determination of Phosphorus Pentoxide in Calcium Phosphate. Weigh 0.5 gram of finely pulverized calcium phosphate, trans- fer to a 6-inch porcelain evaporating dish, add 20 cc. nitric acid, 10 cc. hydrochloric- acid, and evaporate nearly to dryness. Allow to cool, add 25 cc nitric acid, 75 cc. water, boil, and filter into a one-fourth liter fl^sk Wash with water until reaction is no longer acid, and make solution and washings up to the containing mark by the addition of more water Temperature of solution =: 15-5° C. Mix well and take duplicate samples, each of 25 cc, transfer to No, 3 beakers, and treat as follows : Concentrate by evaporation to about 15 cc. Cool somewhat, and add carefully ammonium hydroxide until the solution is alkaline, then make reaction slightly acid with nitric acid. 47 738 e:ngine:kring chemistry Add 50 cc. of standard ammonium molybdate solution/ with stirring, and then some more ammonium hydroxide, but not enough of the latter to render the liquid alkaline. Add 20 cc, ammonium molybdate solution, and set aside over night. Filter, test filtrate with a few drops of ammonium molybdate solution, to be certain that all of the phosphoric acid is precipi- tated, and wash precipitate well on the filter with water contain- ing one-eighth its volume of ammonium molybdate solution. The filtrate and washings are neglected. Fifteen cc. ammonium hydroxide are poured upon the filter dissolving the precipitate; if this it not enough use more am- monia until the precipitate dissolves and the solution formed is caught in a No. 2 beaker. The filter-paper, free from the pre- cipitate is washed thoroughly with hot water, and the filtrate and washings made acid with hydrochloric acid. This produces a precipitation of the yellow ammonium phosphomolybdate. Am- monium hydroxide is added in quantity just sufficient to dissolve this and to form a colorless solution. Thirty cc. of a standard magnesia mixture^ solution are now added gradually with constant stirring for three minutes, and the beaker with the precipitated ammonium magnesium phosphate set aside for thirt}^ minutes. Filter upon an ashless filter, wash with water containing one- eighth in volume of ammonium hydroxide, dry, ignite at first with a gentle heat, finally at a red heat, in a porcelain crucible to constant weight, and weigh as magnesium pyrophosphate. After ignition this precipitate should be white or light gray in color. Example : Grams. Crucible -f- MgaPzOi 15-5567 Crucible 15.5210 JMgsPzOi 0.0357 1 This solution is composed of 50 grams MO3, which are dissolved in 200 cc. NH4HO— 200 cc. HoO, then pour slowly into 1500 cc. HNO3 (sp. gr. 1.2) with constant stirring. - This solution is composed of 100 grams MgS04, 100 grams NH4CI dissolved in 800 cc. HgO, 400 cc. NH4HO (sp. gr. 0.96) is added thereto and thoroughly mixed. ENGINEERING CHEMISTRY 739 Then, Mg.P.Oi : P2O., : : 0.0357 : -r X — 0.00283 gram, from which percentage is readily calculated. Reference. For a method for complete analysis of phosphates and superphosphates consult "Principles and Practice of Agricultural Analysis," by H. W. Wiley, Vol. II, ist edition, pp. 101-141. Determination of Iron in Hematite (SnCl2 Method).^ The following solutions are employed : (i) Potassium bichromate, 4.9 grams dissolved in i liter of water; i cc. = 0.005 gram Fe. (2) Stannous chloride, 100 grams dissolved in i liter of hydro- chloric acid (500 cc. strong acid, 500 cc. water). (3) Mercuric chloride, 50 grams dissolved in i liter of water. (4) Potassium ferricyanide, a piece one-fourth the size of a pea in 40 cc. water. Weigh 0.500 gram of the ore into a No. 2 beaker, moisten with water, add 30 cc. strong hydrochloric acid, cover with a watch- glass, and w^arm gently until solution of the iron is complete and the residue appears white; boil, add stannous chloride from a pipette till the liquid becomes colorless, boil a few minutes, trans- fer solution to a No. 5 beaker, dilute with water to 300 cc, add 35 cc. mercuric chloride, and stir .well. The potassium bichromate solution is now added until 4 drops fail to develop a blue color with ferricyanide indicator in one-half minute. Burette-reading on half-gram samples of ore, gives the per cent, of ore when i cc. potassium bichromates 0.005 gram Fe. Precautions. — Avoid a large excess of stannous chloride : one or two drops more than is required to destroy the yellow color of the iron solution is sufficient. In adding mercuric chloride solu- tion pour all in at once. If added slowly, metallic mercury is precipitated and the operation spoiled. The stannous chloride solution must be added to a concen- trated boiling solution which is strongly acid with HCl in order 1 J. M. Wilson, Chemist to Junction Iron and Steel Co., Steubenville, Ohio, "Methods of Iron Analysis," p. 16. 740 e:ngine;e;ring chemistry that the reduction may take place rapidly. The excess of SnCU is then oxidized by adding mercuric chloride, which is too weak an oxidizing agent to oxidize any of the ferrous iron. iVn excess of stannous chloride easily reduces the mercurous chloride formed to the condition of metallic mercury. This should never be allowed to happen in an iron titration because the metallic mercury is easily oxidized by the K2Cr207 and would cause errors. Therefore it is particularly important in the reduction by SnClg, to avoid excess of SnCl2 of more than 3 or 4 drops. The disappearance of the color of the FeClg in the solution affords a ready means of observing when sufficient SnCls has been added. The reduction by stannous chloride is very rapid and accurate but requires care. Determination of Iron by Titration with Solution of Potassium Bichromate. A. Whe:re the: Iron Solution is in the Ferrous Condition. Take 1.5 grams of crystallized ammonium ferrous sulphate, transfer to a No. 3 beaker, and dissolve in 100 cc. of cold water; add 10 cc. dilute sulphuric acid. Make a solution of potassium bichromate by dissolving 14.761 grams of the "C. P." salt in 6*00 cc. water in a graduated liter flask and adding water to the containing mark. Mix well. Each cubic centimeter is equivalent to 0.0168 gram of iron. If potassium bichromate is added to a solution of ferrous salt in the presence of a strong acid, the ferrous salt is converted into ferric; thus, 6FeSO, + ICCr^O, + yU.O = 3Fe2(SO,)3 + K,SO, + Cr(SO,)3 +7 HA With 29.522 grams of potassium bichromate dissolved in 2 liters of water, 33.6 grams of iron may be changed from a ferrous to a ferric salt (295.22 being the molecular weight of K2Cr207 and 336 being 6 times the atomic weight of iron). One cc. of the bi- chromate solution corresponds to 0.0168 gram Fe (33.6 -^ 2000 cc. = 0.0168 gram Fe). e;ngine:e;ring che;mistry 741 Fill a 50 cc. burette with some of this solution, and drop the bichromate slowly into the beaker containing the iron solution until a drop of the latter placed upon a white porcelain slab and brought in contact with a drop of a very dilute solution of potas- sium ferricyanide, recently made, no longer produces a blue or greenish coloration, showing the ferrous salt to be all oxidized to ferric salt. Note the number of cubic centimeters of the bi- chromate solution required to do this, and calculate the per- centage of iron in the ammonium ferrous sulphate. Example : Ammonium ferrous sulphate taken i-503 gram. 12.78 cc. bichromatic solution required to oxidize. T cc. = 0.0168 gram iron. 12.78 cc, = 0.2147 gram iron. _, 0.2147 X 100 _ ^ . Then = 14.28 per cent. iron. 1.503 Theoretical percentage: (NHJaSO^FeSO^ + 6H2O : Fe : : 100 : x X = 14.28 per cent. B. Wh^rk th^ Iron Soi^utign Exists in th^ FERRIC State:. As the use of bichromate requires the iron to be in the ferrous condition so as to be oxidized by the bichromate, the ferric salt is reduced to ferrous as follows : Take 1.5 grams of ferric sulphate, transfer to a 200 cc. flask, dissolve in 50 cc. water, add 10 cc. H,2S04, and a few pieces of granulated zinc. All the zinc must be dissolved and the solu- tion colorless before it can be titrated with the bichromate. If the solution is not colorless, more zinc and sulphuric acid must be added. It is essential in this process, that all the ferric salt be re- duced to ferrous, otherwise the number of cubic centimeters of the bichromate used would give too low a result for the percent- age of iron. To keep the iron solution in the flask from oxidizing while it is being reduced by the hydrogen, from the reaction of zinc and sulphuric acid, several methods are available. 742 ENGINEERING CHEMISTRY 1. The method described by Fresenius, in which carbon dioxide is passed through the flask during reduction. 2. The stopper of the flask is arranged to allow escape of the hydrogen generated by the dissolving of the zinc by the sul- phuric acid, but prevents inlet of air. The stopper is of rubber (one perforation), through which passes a glass tube. At the upper end of the glass tube a piece of rubber tube (closed at b with a glass rod) is adjusted and at a an opening is made in the rubber tube, which allows the exit of gas, but which closes and prevents the entrance of air — the so-called Bunsen valve (Fig. A). The reduction of the ion solution by stannous chloride is to be preferred to the reduction by means of zinc. See page 27. Example : Ferric sulphate taken i .520 grams. 18.01 cc. bichromatic solution required to oxidize. „, 18.01 X 0.0168 X 100 ^ • • r • -. 1 . Then, = 1990 per cent, iron in ferric sulphate. Theoretical percentages: FegC 304)3 + 9H2O : Feg : : 100 : x X = 19.92 per cent, iron in ferric sulphate Determination of Iron by Means of a Solution of Potassium Permanganate. Weigh 1.585 grams crystallized KsMugOg (C.P.), transfer to a half-liter graduated flask, dissolve in 400 cc. water, and add water to the containing mark. After solution and thorough admixture, I cc. of this liquid will correspond to 0.0056 gram Fe. If we add a solution of ferrous salt containing an excess of sulphuric acid and permanganate of potassium, the former salt is converted into a ferric salt by the oxidizing action of the latter; thus: loFeSO, -f 8H,SO, -4- K.Mn.Og =: 5Fe,(SO js + K,SO, -f 2MnSO, -i- 8H,0. ENGINEERING CHEMISTRY 743 Dissolve 1.5 grams ammonium ferrous sulphate in 75 cc. water, in a No. 3 beaker, and add 10 cc. dilute sulphuric acid. The permanganate solution is added from a burette until the liquid in the beaker maintains a faint permanent pink color. Thus : Example : Ammonium ferrous sulphate taken, 1.542 grams. Amount of K2Mn20s solution required = 39.32 cc. I cc. K2Mn208 solution = 0.0056 gram Fe. 39.32 cc. K2Mn20s solution = 0.22019 gram Fe, or 14.28 per cent. Fe in ammonium ferrous sulphate. INDEX Abrasion cylinder, 269. Abrasion test for road material, 268. Absorption o£ water by rock, 272. Absorption pipette, Hankee's, 634. Absorption power of blotting paper, 563. Absorption power of building stones, 257. Absorption test for brick, 268. Accelerated test for cement, 225. Acetic acid in pigment, 501. Acetylene buoys, 686. Acetylene, candle-power, 684. Acetylene, drying, 684. Acetylene, manufacture, 679. Acetylene, purification, 682. Acid, free, in paper, 551. Acid, free, in water, 581, Acid resisting metal, 169. Acid sludge, 325. Acid value of varnish, 538. Acidity, lubricating oils, 386. Agalite, 555- Ajax metal, 165. Albertite, 274. Alizarene, 534. Alkali in soap, 479, 482. Alkalimetric method for phos- phorus, 142. Alkalinity in pigments, 510. Alkali resisting alloy, 170. Alloys, 160. Aluminum bourbonz, 167. Aluminum bronze, 167. Aluminum bronze, analysis, 170, 237. Aluminum oxide, m cement, 196. Aluminum oxide, in clay, 251. Aluminum oxide, in cylinder de- posits, 728. Aluminum oxide, in fire sand, 251. Aluminum oxide, in iron ores, ^2. Aluminum oxide, in kaolin, 251. Aluminum oxide, in paper, 558. Aluminum oxide, in pigments, 509, 5^3, 515. , Aluminum oxide, m slag, 88. Aluminum oxide, in stone, 251. Aluminum oxide in water, 570, 576, 583. Aluminum oxide in Welsbach mantles, 730. Aluminum paint, specifications, 529. Aluminum silicate, 521. Aluminum silver, 170. American cements, 202, American Foundry Association, cast iron analysis, no. American Foundry Association, coke analysis, 53. American Society for Testing Material, cement specifica- tions, 201. American Society of Civil Engi- neers, cement testing, 207. Ammonium nitrate in dynamite, 731. Animal oil, 453. Animal oil in cylinder deposits, . 728. , . Animal sizing, 552. Anthracene, 325. Anthracite coal, specifications, 20. Anti-friction metal, 165. Antimony in alloys, 167. Antimony in brass or bronze, 182. Antimony, rapid determination, .178. Antimony vermillion, 496. Apparent specific gravity of rock, 272. Araeo picnometer, 368. Argentine, 165. Arsenic bronze, 167. Arsenic in Paris green, 494. Arsenic in pyrites, 78. Artificial bitumens, 275. Asbestine, 514, 521. Asbestos paints, 530. Ash in coal, 2, 15, 51. Ash in coke, 54. Ash in paper, 554. Ash in varnish, 538. Ashbury metal, 165. Asphalt, definition, 273, 343. Asphalt, ductility test, 301, 344. Asphalte, 274. :46 INDEX Asphaltene, 326. Asphalt, flash and fire test, 311. Asphalt, float or fluidity, 298. Asphaltic cement, 346. Asphaltic petroleum, 327, 344, Asphalt, melting point, 313. Asphalt, oil, 336, 457.^ Asphalt pavement mixture, 279. Asphalt pavement specifications, 340. Asphalt, penetration test, 295, 343. Asphalt, preliminary treatment, 307. Asphalt rock, 339. Asphalts, 326. Asphalt, specific gravity, 271. Asphaltum, 274. Asphaltum black, 497. Asphaltum spirits, 522. Asphalt, various analyses, 276. Asphalt, volatilization test, 283, 309, 345- Asphalt v^'earing surface, 342. Atomic vv^eights of elements, 732. Atomizers, steam, 465. Atwater-Mahler Calorimeter, 421. B B alloy, P. R. R., 165. B. t. u. values in coal, 22. B. t. u. determination by calorim- eter, 2"]. Babbitt metal, 165. Bacteriological examination of water, 600. Ball method for cement consist- ency, 230. Barium carbonate in pigments, 508, 513. Barium pigments, 512. Barium sulphate in Paris green, 495. Barium sulphate m pigments, 508, 521. Barytes, 497, 5I2, 521. Basic carbonate of lead, 500, 520. Basic sulphate of lead, 502, 520. Baume gravity, 327. Baume gravity tables, 736, 'JZI- Bell metal, 160. Benedict nickel pipe, 186. Benzene, 522. Benzol in bitumen, 274, 328. Benzols, 353. Bismuthate method for mangan- ese, 137. Bitumen, artificial, 275, 325. Bitumen briquettes, 301. Bitumen, definition, 328. Bitumen, insoluble in parafiin naphtha, 289. Bitumen, insoluble in carbon tetrachloride, 291. Bitumen, native, 335. Bitumen, solubility in carbon di- sulphide, 285. Bituminous aggregates, extraction, Bituminous binders, consistency, 299. Bituminous coal, specifications, 22. Bituminous material, table, 274. Bituminous road material, classi- fication, 280. Bitumens, distillation test, 322. Bitumens, melting point, 313. Black pigments, 497. Blanc fixe, 497. Blast furnace as power plant, 98. Blast furnace charges, graphic method, 95. Blotting paper, absorption quality, 563. Blown petroleum, 329. Blue pigments, 497. Boat for steel analysis, 127. Boghead cannel coal, composition, 16. Boiler compounds, 607. Boiler efficiencies — oil fuel, 465. Boiler troubles, 621. Boiling point of elements, 732. Bone black, 497. Bone fat, 407. Brass, 160. Brass analysis, 178. Brass, specifications, 185. Brazing metal, specifications, 190 Breaking point, oil, tar and pitch, 352. Breaking strength of paper, 556. Bremen blue, 497, Brick, absorptive power, 257. Brick, crushing strength, 254. Brick, microscopical examination, 263. Brick testing, 260. Brick testing machines, 261. Briquettes, cement, 261. INDEX 747 Briquette molds, 216. Bristol pyrometer, 723. Britannia metal, 165, 168. Bronze, analysis, 178. Bronze, specifications, 193, Brown pigments, 496. Brown, zinc, specifications, 529. Brunswick blue, 497. Brunswick green, 497. Building stone, 250. Bunsen combustion furnace, 472. Bunsen photometer, 706. Bureau of Highways concrete specifications, 246. Bureau of Standards sieve speci- fications, 235. Burners for fuel oil, 462. Cabin car color, specifications, 527. Cadmium yellow, 497. Calcium carbonate, 521. Calcium carbonate in Paris green, 495. Calcium oxide in cement, 196, 238. Calcium oxide in clay, 251. Calcium oxide in cylinder de- posits, 728. Calcium oxide in fire sand, 251. Calcium oxide in iron ores, 73. Calcium oxide in kaolin, 251. Calcium oxide in limestone, 66. Calcium oxide in paper, 538, Calcium oxide in pigments, 509, 515. Calcium oxide in slag, 88. Calcium oxide in stone, 251. Calcium oxide in water, 570, 576, 583. Calcium oxide in Welsbach man- tles, 730. Calcium phosphate, analysis, ']2>'J' Calcium pigment, 509. Calcium sulphate, 521, 522. Calcium sulphate in paper, 558. Calcium sulphate in water, 576. Calcium sulphide in cement, 200. Calorific power of blast furnace gas, 104. Calorific value of various oils, 471. Calorimeter, Atwater-Mahler, 421. Calorimeter, cooling correction, 30. Calorimeter, Emerson, 34. Calorimeter, Junker's, 665. Calorimeter, oxygen, 27. Calorimeter, Parr, 2>2>' CameHa metal, 165. Camp's agitator, 141. Candle-power, acetylene, 684. Candle-power computer, 718. Candles for photometry, 711. Car equipment, for Pintsch gas, 703. Car sampling, iron ores, 80. Carbene, 329. Carbide feed generators, 679. CarboHc oil, 353. Carbon, anneaHng, or temper, 115. Carbon bisulphide, 329. Carbon dioxide in cylinder de- posits, 728. Carbon dioxide in flue gas, 627. Carbon dioxide in gypsum, 512. Carbon dioxide in illuminating gas, 659. Carbon dioxide in iron ores, 74. Carbon dioxide in limestone, 68. Carbon dioxide in pigments, 502, 506, 508, 510, 513, 515, 518. Carbon dioxide in water, 570, 576, 598. Carbon, fixed, in oils, 332. Carbon, free, in oils, 333. Carbon, graphitic, 116. Carbon in asphalt, 292. Carbon in cast iron, i, 12, 117. Carbon in coal, 2, 51. Carbon in coke, 54. Carbon in oil, 403. Carbon in steel, 126. Carbon metal, 165. Carbon monoxide in flue gas, 627. Carbon monoxide in illuminating gas, 661. Carbon residue in oil, 402. Carbon tetrachloride, 330. Carbureted water gas, 670. Cardboard tester, Schooper, 561. Cargo sampling, in iron ores, 81. Castile soap, specifications, 489. Cast iron, no. Catalyzers, 127. Cellulose fiber in paper, 550. Cement, analysis, 208. Cement, asphaltic, 346. Cement, briquettes, 216. Cement, compressive strength, 222. Cement, consistency, 212, 228. 748 INDEX Cement, constancy of volume, 203, 223. Cement, examination of, 195, 236. Cement, fineness, 203, 211, 228. Cement, methods of testing, 207. Cement, mixing, 231. Cement, Portland, scheme for analysis, 197. Cement, sampling, 207. Cement, soundness, 231. Cement, specifications, 201, 207. Cement, specific gravity, 201, 209, 228. Cement, tensile strength, 219, 233. Cement, time of setting, 215, 231. Centrifuge extractor, 316. Cerium oxide in Welsbach man- tles, 730. Chalk, 509. Chamber presses, 603. Chert, 271. China clay, 497, 514. Chinese blue, 497. Chinese yellow, 496. Chlorides in paper, 551. Chlorine in water, 570, 576, 589. Chrome green, 497, 526. Chrome yellow, 497, 523. Chromium in iron ore, 75. Chromium in pigment, 523, 526. Chromium in steel, 151. Clay, 250. Cleveland cup, 383. Clinker in coal, 52. Coal, anthracite, specifications, 20. Coal, bituminous, specifications, 22. Coal, classification of size, 17. Coal, penalization, 24. Coal, proximate analysis, i. Coal, sampling, 6. Coal, schedule of proposals, 49. Coal tar, 330, 333. Coal tar black, 497. Coal, tests for slate, 17. Coal, value as fuel, 50. Coal, value for gas production, 690. Cobalt blue, 497. Cobalt green, 497. Coke, 53. Coke, analysis, 53. Coke, by-product, 693. Coke, composition, 53. Coke, compression test, 61. Coke oven tars, 330. Coke, physical tests, 57. Coke, specifications, 62. Cold test for oils, 369. Color of water, 598. Colorimeter, Stammer, 451. Colorimeter, Wilson, 453. Colorimeter, Wolff's, 592. Colorimetric method for carbon in steel, 134. Color test for kerosene, 451. Combustion apparatus for steel, Combustion method for illuminat- ing gas, 662. Commercial soaps, 475. Compression of Pintsch gas, 701. Compression strength for cement, 22.2. Compression test of concrete, 245. Compressor oil, 431. Concrete, 242. Concrete, oil-mixed, 247. Concrete pavement specifications, 246. Condenser tubes, specifications, 188. Cone sampling, 83. Consistency of bituminous bind- ers, 299. Consistency, normal of cement, 212, 229. Constancy of volume, cement, 203, 223. Conversion tables, per cent, ele- ments and radicals in com- pounds, 734. Copper analysis, 91, Copper blue, 497. Copper green, 497. Copper in brass, 161, 179. Copper in copper slags, 162. Copper in ores, 92. Copper in Paris green, 495. Copper in pyrites, 78. Copper in steel, 148. Copper oxide in cylinder deposits, 728. Copper pipe, specifications, 191. Copper, rolled, specifications, 191. Copper specifications, 185. Corrosion, cause and cure, 621. Cotton fiber in paper, 550. INDEX 749 Cracked oils, 331. Creosote oils, 354, 356. Cross breaking, paving brick, 264. Crusher run stone, 271. Crushing of paving brick, 265. Crushing strength of building stones, 254. Cupro, magnesium, 168. Cuprous chloride pipette, 657. Cut back products, 331. Cutting oils, 427. Cyanides, analysis, 729, Cyanogen in cyanides, 729. Cylinder deposits, analysis, 728. Cylinder oil, 423. Cylinder oil specifications, 413. .425. Cylinder stock, 422. Dead oils, 331. Degras oil, 409. Dehydrated tar, 332. Delta metal, 163. Density and calorifific power in oils, 417. Deoxidation of brass, 163. Deoxidized bronze, 165. Department of Docks and Ferries, New York, oil specifications, 413. Derveaux water purifier, 600. Destructive distillation, 332. Didymium oxide, in Welsbach mantles, 730. Direction of fiber in paper, 566. Distillation, creosote oil, 357. Distillation, light oil, 352. Distillation test for bitumens, 322. Distillation test, turpentine, 539. Dolomite, 70. Doolittle viscosimeter, 381. Dow form of briquette mold, 306. Drying acetylene, 684. Drying oil in creosote oil, 356. Ductility of asphalt, 301, 344. Ductility of asphalt cement, 347. Dulin rotarex, 320, Dust, in paving material, 271. Dynamite, analysis, 731. Dynamo oil, 430. Dynamo oil, specifications, 414. Eggertz color test, 136. Eichorn, araeo picnometer, 368. Elements controlling properties of cast iron, 116. Elements, table, 732. Elliott gas analysis apparatus, 628. Elongation resistance of paper, 564. Emerald green, 497. Emerson calorimeter, 34. Emulsions, 332. Engine oil, specifications, 413. Engler viscosimeter, 299, 372, 462. Eosene, 534. Eschka Fresenius method for sul- phur in coal, 2. Eschka method for sulphur in oil, 474. Ether, petrolic, 337. European cements, 202. Eutectic, Guthrie's, 167. Evaporation test, turpentine, 539. Evolution titration method for sulphur in steel, 144. Explosion pipette, 663. Extraction, bitumenous aggre- gates, 315. Face sampling, 84. Factice, 275. Factor for specific gravity in oil, 364. Fats and fatty acids, examina- tion, 485. Fatty acids in soap, 479. Fatty acids, melting point, 395. Feed water heaters, 611. Feed water heater, table of sav- ing, 622. Fenton white metal, 165. Ferric oxide in cement, 196, 237. Ferric oxide in clay, 251. Ferric oxide in fire sand, 251. Ferric oxide in iron ores, 'JQ.. Ferric oxide in kaolin, 251. Ferric oxide in limestone, 66. Ferric oxide in paper, 558. Ferric oxide in pigments, 509, 512, 515, 518. Ferric oxide in stone, 251. Ferric oxide in water, 570, 576, 583. Ferro aluminum, 167, Ferro aluminum, analysis, 173. Ferrous oxide in cement, 200. 750 INDDX Ferrous oxide in slag, 88. Ferro-tungsten, 167. Fery radiation pyrometer, 725. Fiber of paper, 544. Filter for steel analysis, 128. Filter presses, 603. Filtration of water, 600. Fineness, cement, 203, 211, 229. Fineness, pigments, 500, 515. Fire clays, composition, 253. Fireproof paints, 530. Fire sand, 250. Plash and fire test for oils, 382. Flash test for asphalt, 311. Flash test for asphalt residuum, 346. Flash test for turpentine, 540. Float test for asphalt, 298. Flour, in paving material, 271. Flue gas analysis, 627. Flue gas, conversion table, 637. Fluidity test, for asphalt, 298. Flux, in asphalts and bitumens, 332, 345. Flux, paraffin, 279. Fluxed asphaltic cements, 348. Ford Williams' method, mangan- ese ores, 90. Foster flash and fire tester, 448. Foundry chemistry, 115. Frame presses, 604. Frankfort black, 497. Frankolin, 684. Freezing test for stone, 258. Friction coefficient for oils, 409. Fuel economizers, 623, Fuel oil, 455. Fuel, testing, 43. Furfural, in turpentine, 540. Gangue, in pyrites, 78. Gas, acetylene, 679. Gas analysis, 627. Gas, calorimetry, 665. Gas, chimney or flue, analysis, 627. Gas from blast furnaces, loi. Gas house coal tar, 333. Gas, illuminating, analysis, 655. Gas, natural, composition, 677. Gas, oil, manufacture, 695. Gas, Pintsch, 696. Gas, producer, composition, 676. Gas producing quality of coal, 691. Gas, variation in volume, table, 714. Gas, water, 670. Gases, density, 654. Gases, specific gravity, 735. Gases, various illuminating, com- position, 675. Gasolene test for oils, 401. Generators, acetylene, 679. German silver, 167. Gilsonite, 274, 333. Glycerine in fats and soaps, 486. Glycerine soaps, 490. Glycerine specifications, 492, Goubert feed water heater, 612. Grain, 116. Graduation of sand in concrete, 246. Grading mineral aggregate in bi- tumens, 319. Grahamite, 274, 333. Granite, absorptive power, 257. Granite, crushing strength, 254, 257. Granite, in road material, 271 Granitoid, 271. Graphic method for blast furnace charges, 95. Graphite, as lubricant, 415. Graphite black, 497. Graphite, in cast iron, 114. Graphite, specifications, 417. Gravimetric method for nickel in steel, 149. Greases, 406. Greases, various, composition, 407. Green fuel economizer, 623. Green pigments, 497. Guide gibs, 190. Gumming test for oils, 398. Gums in varnish, 537. Gun metal, 160. Guthrie's eutectic, 167. Gypsum, 497. Gypsum, analysis, 511. Hahn, gas analysis apparatus, 636. Hankee's absorption pipette, 634. Hann's method for iodine value, 453. Harcourt pentone lamp, 716, Hardening and tempering oils, 431- Hardness of water, 584. IND^X 751 Hardware metal, 168. Hartig Reusch apparatus, 557, Heating asphalt cement, 347. Heat loss in chimney gases, 650. Heidenreich's test, 439. Hematite, 71. Hemp fiber in paper, 550. Hempel burette, U. G. I. modi- fication, 656. Heraeus quartz glass thermom- eter, 719. Heratol, 683. High carbon tars, 334. Hogarth flask, 58. Horse-power developed, table of, 676. Howard and Morse, apparatus for consistency of binders, 300. Huble, iodine absorption, 381. Hydraulic bronze, 168. Hydraulic compression machine for cement and concrete, 243. Hydraulic metal, 169. Hydrogen in illuminating gas, 662. Hydrometer, Koppe-Saussure's, 557- Hydrometer, Sohmer, 281. H)^droxide, in cyanides, 729. Hygrometer, Tutwiler and Bond, 658. Ice machine oils, 428. Ignition loss in pigments, 506, 509, ■512, 514. Illuminating gas, 655. Impact test, concrete, 248. Impact test, paving brick, 266. Impact tester, 270. Incrustation, cause and cure, 621. Indian red, 496. Indian red, specifications, 528. Insoluble iron ores, 75. Insoluble matter in pigments, 503, 507, 508, 512, 514, 531. Interpretation of cement tests, 224. Iodine absorption, oils, 381. Iodine value, linseed oil, 453. Iron determinations, 738, 739, 740, 741. Iron in brass or bronze, 183. Iron in cylinder deposits, 728. Iron in pigment, 512, 532. Iron in pyrites, 78. Iron in tin plate, 158. Iron ores, analysis, 71. Iron ores, composition, ^T. Iron ores, sampling, 80. Iron oxide paint, 496. Isolite, 703. journal boxes, 190. Junker's gas calorimeter, 665. Jute fiber in paper, 550. Kaolin, 250, 497. Kayserzinn, 170. Keith gas, 695. Kennicott process for water soft- ening, 609. Kerosene, 441. King's yellow, 497. Koppe-Saussure's air hydrometer, 557. Lamp black, 497. Lanthanum oxide in Welsbach mantles, 730. Lard oil, specifications, 388. Laundry soaps, 475. Lead analysis, 92. Lead chromate in Paris green, 495. Lead covered sheets, 154. Lead in brass or bronze, 179. Lead in lead covered sheets, 157. Lead in ores, 93, Lead in pigments, 500, 502, 504, 5 16, 523. Lead in pyrites, 78. Lead in tin plate, 157. Lead peroxide in pigments, 516. Lead, pig, specifications, 194, Lead sulphate, in Paris green, 495- Lead sulphate in pigment, 526. Lead sulphate paint, 496. Le Chatelier, specific gravity ap- paratus, 210. Le Chatelier, pyrometer, ^22. Leeds scheme for soap analysis, 477- Lemon chrome, analysis, 523, Liebermann-Storch reaction, 397. Lignites, 14. 752 INDEX Lime, hydrated, in paint, 512. Lime in Portland cement, 199, 200. Limestone, 65. Limestone, absorptive power, 257. Limestone, analysis, 66. Limestone, crushing strength, 254. Limonite, 71. Limpid point, 356. Linen fiber in paper, 550. Linoxyn, in varnish, 537. Liquids, specific gravity and de- grees Baume, 736. Lithopone, 496, 507, 520. Locomotive water, 606. London coal gas, calorimetric tests, 669. Low carbon tars, 334. Lowe process gas, composition, 670. Lowe water gas apparatus, 672. Lubricants, 421. Lubricating oils, examination, 362. Lye, concentrated, specifications, 492. Macadam, toughness test, 269. Magnesite, 497. Magnesium as deoxidizer, 163. Magnesium oxide in cement, 198, 238. Magnesium oxide in clay, 251. Magnesium oxide in cylinder de- posits, 728. Magnesium oxide in fire sand, 251. Magnesium oxide in iron ores, 73. Magnesium oxide in kaoHn, 251. Magnesium oxide in limestone, 66. Magnesium oxide in paper, 558. Magnesium oxide in pigments, 510, 515. , Magnesium oxide m pyrites, 78. Magnesium oxide in slag, 88. Magnesium oxide in stone, 250. Magnesium oxide in water, 570, 576, 583. . . , ^ , Magnesium oxide in Welsbach mantles, 730. Magnesium silicate, 521. Magnesium silicate in paper, 558. Magnetite, 71. Magnolia metal, 165. Malachite, 497. Maltha, 274, 334. Malthene, 334. Manganese bronze, 167, 189. Manganese brown, 496. Manganese green, 497. Manganese in cast iron, ill. Manganese in pyrites, 78. Manganese in steel, 137. Manganese ores, Ford Williams method, 90. Manganese oxide in clay, 251. Manganese oxide in fire sand, 251. Manganese oxide in iron ores, 73. Manganese oxide in kaolin, 251. Manganese oxide in slag, 88. Manganese oxide in stone, 251. Manganese resistance metal, 169. Manganin, 169. Manganous oxide in cement, 200. Manheim gold, 163. Mantles, Welsbach, analysis, 730. Marble, absorptive power, 257. Marble, crushing strength, 254. Marsh gas, 274. Massie's test, 439. Matrix, 271. Maumene's test, 389. Mechanical burners, 463. Medicated soaps, 475. Melanin, 496. Melting point in bitumens, 113. Melting point of elements, 732. Melting point of fatty acids, 395. Melting point of oils, tars and pitches, 350. Mercury sulphide, 533. Merten's machine for friction of oils, 411. Methane, 662. Aleyer tube, 131. Microscopical examination of brick, 263. Microscopical examination of oil, 402. Alicroscopical examination of paper, 548. Mineral aggregate, in bitumens, 319- Mineral green, 497. Mineral matter in water, 583. Mineral oil in cylinder deposits, 728. Mineral rubber, 335. Mixed lead pigment, analysis, 524. Mixing asphaltic cement, 346. IND^X 753 Mixing cement, 231. Mixtures, asphalt, 345. Moisture in coal, i, 51. Moisture in coke, 54. Moisture in gypsum, 511. Moisture in iron ores, 71. Moisture in pigments, 500, 506, 507, 509, 512, 514, S16, 531. Molds for bitumen briquettes, 306. Molds, for cement briquettes, 217. Molybdate magnesia method for phosphorus, 140. Monel metal, 194. Mortar, absorptive power, 257. Mosaic gold, 163. Motor oils, 432. Mucus, organic, 496. Muntz's metal, 160, 191. Naphthas, 335, 522. Naphthalene, 335, 353, 356. Naphthalene, in bitumens, 274. Native bitumens, 335. Natural cement, 195. Natural cement, constancy of vol- ume, 204. Natural cement, fineness, 204. Natural cement, specifications, 204. Natural cement, tensile strength, 204. Natural cement, time of setting, 204. Natural gas, composition, 677. Neatsfoot oil, specifications, 414. Needle metal, 169. Nessler reagent, 589. Newbigging's plant for gas pro- ducing quaUty of coal, 691. New York State tester for flash point, 312. Nickel's apparatus for paper fiber, 566. Nickel in pyrites, 78. Nickel in steel, 149. Nitrates in dynamite, 731. Nitrates in water, 594. Nitrites in water, 595. Nitrogen in flue gas, 627. Nitrogen in illuminating gas, 662. Nitrogen in oil, 473. Nitrogen in Prussian blue, 532. Nitroglycerine in dynamite, 731. 48 Nitrosulphuric method for silicon in steel, 146. Nomenclature, standard, for pig- ments, 520. Non-bituminous road materials, 271. Non-fluid oils, 408. Oil asphalts, 336. Oil, calorific power, 417. Oil, carbolic, 353. Oil containing blown rape seed and blown cotton seed oil, 435. Oil, drying, in creosote oil, 356. Oil for wood block pavement, 359. Oil, fuel, 455. Oil gas, 695. Oil in cylinder deposits, 728. Oil, lard, specifications, 388. Oil, linseed, 453. Oil, mineral sperm, specifications, 451- Oil mixed concrete, 247. Oil, nitrogen determination, 473. Oil pitches, 336. Oil, sulphur in, 474. Oil tars, 336. Oils, animal, 453, Oils, breaking point, 352. Oils, burning, cloud test, 443. Oils, burning, flash and fire test, 442. Oils, compressor, 431. Oils, cracked, 331. Oils, creosote, 354, 356. .Oils, cutting, 428. Oils, cylinder, 423. Oils, dead, 331. Oils, dynamo, 430. Oils, hardening and tempering, 431. Oils, ice machine, 428. Oils, illuminating, 441. Oils, lubricating, acidity, 386. Oils, libricating, carbon residue, 402. Oils, lubricating, coefficient of friction, 409. Oils, lubricating, cold test, 369. Oils, lubricating, emulsion test, 433. Oils, lubricating, examination, 362. 754 IND^X Oils, lubricating, fatty oil mix- tures, 440. Oils, lubricating, fixed carbon, 403. Oils, lubricating, flash and fire , test, 382. Oils, lubricating, gasolene test, 401. Oils, lubricating, gumming test, 398. Oils, lubricating, heat test, 433. Oils, lubricating, iodine absorp- tion, 382. Oils, lubricating, Maumene's test, 389. Oils, lubricating, microscopical ex- amination, 402. Oils, lubricating, paraffin deter- mination, 403. Oils, lubricating, soap test, 404. Oils, lubricating, specific gravity, 363. Oils, lubricating, sulphur test, 399. Oils, lubricating, water test, 400. Oils, melting point, 351. Oils, motor, 432. Oils, paraffin, 422. Oils, rosin, detection, 397. Oils, specific gravity, 350, 355. Oils, spindle, 422. Oils, testing, 349. Oils, transformer, 428. Oils, turbine, 427. Oils, vegetable, 453. Oils, ultimate analysis, 472. Olefiants in illuminating gas, 660. Olefine, 457. Olsen impact tester, 270. Orange pigments, 497. Organic color, in pigments, 516. Organic lakes, 534. Organic matter in limestone, 66. Organic matter in water, 583, 596. Orsat gas apparatus, 632. Orthoanisodine, 534. Oven, N. Y. Testing Laboratory, 284. Oxidation method for sulphur in steel, 144. Oxyacetylene welding, 689. Oxygen, dissolved in water, 598. Oxygen, for calorimeter combus- tion, 28. Oxygen in flue gas, 627. Oxygen required for gas combus- tion, 641. Ozlo white, 506. Ozocerite, 274. Packfong, 167. Paper, acids in, 551. Paper, analysis, 558. Paper, ash, 554. Paper, breaking strength, 556. Paper, chemical and physical ex- amination, 544. Paper, direction of fiber, 566. Paper, elongation resistance, 564. Paper, microscopical examination, 548. Paper, nature of fiber, 544. Paper, Paris Chamber of Com- merce tests, 562. Paper, sizing, 552. Paper, thickness, 556, 567. Paper, weight, 556, 567. Paraffine, 336, 457- Paraffine flux, 279. Paraffine in bitumens, 274. Paraffine in dynamite, 731. Paraffine in illuminating gas, 662. Paraffine in oils, 403. Paraffine naphthas, 336. Paraffine petroleums, 336. Paraffine scale in asphalt, 293, 336. Paraffine spirits, 522. Parallel system of sampling, 80. Para-nitraniline, 534. Paris green, analysis, 494. Paris white, 509. Parr calorimeter, 33. Parsons, white metal, 165. Paste, cleaning and polishing, 493. Pats, cement, 224. Pattern metal, 169. Paving brick, testing, 264. Penalization of coal, 24. Penetration test for asphalt, 295, 343, 344, 345. Penetration test for asphalt ce- ment, 346, 347, 348. Penetrometer, 296. Penn. anthracite, analysis, 16. Pennsylvania R. R. cabin car color, 527. Pensky Martin closed cup, 384. Peroxide fusion method for sul- phur, 5. IND^X 755 Persulphate method for mangan- ese in steel, 139. Petrolene, 337. Petroleum burning oils, specifica- tions, 442. Petroleum products, tarry matter in, 405. Petroleums, 337. Petrolic ether, 337. Pewter, 165. Phono electric wire, 168. Phosphor bronze, 165, 193. Phosphorus in cast iron, ill. Phosphorus in coal, 6. Phosphorus in coke, 56. Phosphorus in steel, 40. Phosphorus pentoxide in calcium phosphate, 737. Phosphorus pentoxide in cement, 200. Phosphorus pentoxide in iron ores, ^2. Phosphorous pentoxide in lime- stone, 66. Phosphorus pentoxide in slag, 88. Photometer, Bunsen, 706. Photometer, standard bar, 715. Photometry, 705. Physical characteristics of pig- ments, 520. Physical tests on coke, 57. Physical tests on quicklime, 241. Pig iron, 114. Pig iron composition, 115. Pig iron sampling, 125. Pig iron, specifications, 124. Pig lead, specifications, 194. Pinchbeck, 163. Pintsch gas, 696. Pintsch gas apparatus, 697. Pintsch gas compression, 701. Pintsch gas purification, 700. Pintsch hydrocarbon, 698. Pintsch tar, 698. Pipe, brass, specifications, 185. Pitch, breaking point, 352. Pitches, 337. Pitch, melting point, 350. Pitch, specific gravity, 350. Pitch, testing, 349. Plaster of Paris, analysis, 510. Platinoid, 169. Polymerization, of turpentine, 539. Porter Clark process, 604. Portland cement, 195. Portland cement, constancy of volume, 205. Portland cement, definition, 205. Portland cement, fineness, 205.^ Portland cement, specific gravity, 205. Portland cement, sulphuric acid and magnesia in, 205. Portland cement, tensile strength, 205. Portland cement, time of setting, 205. Potash, in cement, 200, 239. Potash, in clay, fire sand, kaolin and stone, 252. Potash, in water, 576. Potash specifications, 491. Potassium, in cyanides, 729. Potassium nitrate in dynamite, 731. Powder from blast furnace gas, ig8. Priming, cause and cure, 621. Producer gas, composition, 676. Prussion blue, 497. Prussian blue, analysis, 531. Prussian blue in • chrome green, 526. Puratylene, 683. Purification of acetylene, 682. Purification of Pintsch gas, 700. Pyrites, composition of various^ 79- Pyrites, scheme for analysis, 78. Pyro bitumens, 275, 337. Pyrogenetic, 338. Pyrometers, electrical resistance type, 719. Pyrometers, optical and radition type, 725- Pyrometers, thermo electric, 722. Pyrometry, 718. Pyroxylene in dynamite, 731. Qualitative tests for alloys, 174. Quantitative tests for alloys, 175- Quickhme in paint, 512. Quicklime, sampling, 240. Quicklime, specifications, 240. Quickhme, testing, 240. Rattler, for brick, 265. 756 IND^X Reagents for steel analysis, 128. Realgar, 497. Red chromate of lead, analysis, 524. Red lead, 496. Red lead, analysis, 515. Red lead, specifications, 525. Red lead, volumetric determina- tion, 524. Red pigments, 496. Reduced oils, 338. Reduced petroleums, 338. Redwood viscosimeter, 380. Refined tar, 338. Refractive index of turpentine, 539. Requirements, state, for flash and fire test of oils, 450. Residual oils, 338. Residual tar, 338. Residuum, petroleum, 280. Residuum, semi-asphaltic, 346. Residuums, asphaltic, 345. Residuums, paraffine, 346, Resin, in dynamite, 731. Resin, in shellac, 541. Resin, in soap, 482. Resin soaps, 475. Resin specifications, 492. Retorts for Pintsch gas, 699. Riehle cement tester, 221. Riehle friction apparatus for oils, 410. Road material, abrasion test, 268. Road material, bituminous, exam- ination, 280. Rock asphalt, 339. Rope net system of sampling iron ores, 81, Rose metal, 165. Rosin oils, detection, 397. Rosin, sizing, 552. Rosine, 167. Rotarex, Dulin, 320. Round sampling, 84. Rubble, 271. S Salt water soap, specifications, 489. Samples, asphaltic cement, 348. Sampling cement, 207. Sampling coal, 6. Sampling coke, 53. Sampling iron ores, 80. Samphng lead covered sheets, 154. Sampling pig iron, 125. Sampling quicklime, 241. Sampling tin plate, 154. Sand, for concrete, tests, 245. Sand, standard, 215. Sandsaone, absorptive power, 257. Sandstone, crushing strength, 254. Saponification value, 405. Saybolt flash and fire tester, 450, Saybolt viscosimeter, 374. Scale forming ingredients in water, 583. Scarlet, 534. Schedule, coal proposals. Treas- ury Department, 49. Schooper, cardboard tester, 561. vSchwartz U-tube, 472. Seger cones, 726. Semi-asphaltic petroleum, 339. Separation of mineral from vege- table and animal oil, 391. Sepia, 496. Shellac analysis, 541. vShot metal, 165. Sienna, 497. Sieve shaker, 321. Sieve specifications, 235. Silex, 514. Silica, in building stone, 250. Silica, in cement, 196, 236. Silica, in clay, 250. Silica, in cylinder deposits, 728. Silica, in iron ores, 71. Silica, in kaolin, 250. Silica, in limestone, 66. Silica, in paper, 558. Silica, in pigment, 515, 518. Silica, in slag, 88, Silica, in water, 570, 576, 583. Silica pigments, 514, 522, 530. Silicon bronze, 167. Silicon in cast iron, no, 121. vSilicon in steel, 146. Sizing, in paper, 552. Slag, blast furnace, analysis, 88. Slag, broken, mechanical analysis, Slags, blast furnace, composition, 89. Slate, in coal, 17. Soap analysis, 475. Soap analysis scheme, Leeds, 477. Soap analysis scheme, Wright and Thompson, 478. INDEX 757 Soap, Castile, 489. Soap, mineral, 491. Soap powder, 490. Soap, salt water, 489. Soap, specifications, 488. Soap test, for oils, 404. Soaps, glycerine, 490. Soaps, transparent, 489. Soaps, various, composition, 487. Society of Chemical Industry, cement analysis, 236. Soda or sodium oxide in cement, 200, 239. Soda or sodium oxide in clay, 252. Soda or sodium oxide in fire sand, 252. Soda or sodium oxide in kaolin, 252. Soda or sodium oxide in paper, 558. Soda or sodium oxide in stone, 252. Soda or sodium oxide in water, 576. Soda, specifications, 491. Sodium in cyanide, 729. Sodium potassium cyanide, analy- sis, 729. Sodium nitrate in dynamite, 731. Soft bearing metal, 170. Sohmer hydrometer, 281. Soil, in road material, 272. Solder, 160. Solder, soft, 165. Soluble salts in gypsum, 511. Soluble salts in pigments, 507. Sorge Cochrane hot process feed water softner, 615. Soundness, of cement, 231. Spanish white, 509. Spathic iron ores, 71. Spawl, in road material, 272. Specific gravity, asphalt residuum, 346. . Specific gravity, asphalts, 281, 327. Specific gravity, cement, 201, 209. Specific gravity, coal and coke, 57. Specific gravity, creosote oil, 357. Specific gravity, elements, 732. Specific gravity, fuel oil, 460. Specific gravity, gases and vapors, 735- Specific gravity, linseed oil, 453. Specific gravity, liquids, 736, ^n. Specific gravity, lubricating oil, 363. Specific gravity, oil, tar and pitch, 350- Specific gravity, oils used with mineral oils, table, 369. Specific heat, of elements, 732. Specifications, aluminum paint, 529. Specifications, asphalt pavement, 340. Specifications, brass, bronze and copper, 185. Specifications, brass castings, 185. Specifications, brass pipe, 185. Specifications, brazing metal, 190. Specifications, brown zinc, 529. Specifications, burning oils, 442. Specifications, cabin car color, 527. Specifications, cement, 201, 226. Specifications, chrome green, 526. Specifications, chrome yellow, 523. Specifications, cleaning and polish- ing paste, 493. Specifications, coal, 20, 46. Specifications, coke, 62. Specifications, concentrated lye, 492. Specifications, concrete pavement and curb foundations, 246. Specifications, condenser tubes, 188. Specifications, copper pipe, 191. Specifications, cylinder oil, 413, 425. Specifications, dynamo oil, 414. Specifications, engine oil, 413. Specifications, fuel oil, 459. Specifications, glycerine, 492. Specifications, guide gibs, 190. vSpecifications, Indian red, 528. Specifications, journal boxes, 190. Specifications, lard oil, 388. Specifications, linseed oil, 453. Specifications, manganese bronze, 189. Specifications, mineral sperm oil, 451. Specifications, Monel metal, 194. Specifications, neatsfoot oil, 415. Specifications, pig iron, 124. Specifications, pig lead, 194. Specifications, potash, 491. 7S8 INDEX Specifications, quicklime, 240. Specifications, red lead, 525. Specifications, rolled bronze, 193. Specifications, rolled copper, 191. Specifications, sieves, 235. Specifications, soap, z|88. Specifications, soap powder, 490. Specifications, soda, 491. Specifications, steel, 126. Specifications, tin plate, 159. Specifications, turpentine, 540. Specifications, white lead, 519. Specifications, white metal, 189. Specifications, white zinc, 528. Specifications, wood block pave- ment, 358. Speculum metal, 163. Spray burners, 462. Stammer colorimeter, 453. Standardization of calorimeter, 28. Starch in paper, 553. Steel analysis, 126. Sterline, 168. Sterro, 163. Stone, 250. Stone, broken, mechanical analy- sis, 273. Stone chips, in road material, 272. Storage of cement test pieces, 219. Strontianite, 512. Strontium oxide in cement, 200. Strontium white, 497, 512. Sugg-Argand burner, 708. Sulphates, in g>'psum, 511. vSulphates, in paper, 551. Sulphates, in pigments, 503, 510, 513, 515, 518, 523. Sulphates, in water, 570, 576, 583. Sulphonation test, 357. Sulphur dioxide, in pigments, 502, 506. Sulphur in calorimeter washings. Sulphur in cast iron, no. Sulphur in cement, 239. vSulphur in coal, 4, 52. Sulphur in coke, 55. Sulphur in dynamite, 731. Sulphur in fuel oil, 462. Sulphur in oil, 474. Sulphur in pyrites, 78. Sulphur in slag, 88. Sulphur in steel, 144. Sulphur test for oils, 399. Sulphuric acid method for silicon in steel, 147. Sulphuric anhydride in cement, 198, 239. Sulphuric anhydride in clay, 252. Sulphuric anhydride in fire sand, 252. Sulphuric anhydride in iron ores, 72. Sulphuric anhydride in kaolin, ^52. Sulphuric anhydride in limestone, 66. Sulphuric anhydride in stone, 252. Sun valve, 687. Tagliabue, flash and fire tester, 445- Tagliabue freezing apparatus, 258. Tagliabue viscosimeter, 381. Tailings, in road material, 272. Talcose, 521. Tar acids in creosote oil, 356. Tar, breaking point, 352. Tar, melting point, 350. Tar pitches, 339. Tarry matter in petroleum prod- ucts, 405. Tar, specific gravity, 350. Tars, 339. Tar testing, 349. Tearing length of paper, 561. Tempered lead, 170. Tensile strength, cement, 219, 233. Terra alba, analysis, 510. Testing cement, methods, 207. Testing concrete, methods, 242. Testing oil, tar and pitch, 349. Thermometer, Beckman, 28. Thermometer, Heraeus quartz glass, 719. Thickness of paper, 556, 567. Thorium oxide, in Welsbach mantles, 730. Timber, for wood block pave- ment, 360. Time of setting, cement, 215, 231. Tin in alloys, 167. Tin in brass or bronze, 161, 181. Tin in tin plate, 155. Tin plate analysis, 153. Tin plate, composition, 154. Tin plate, sampling, 154. Tin plate, specifications, 159. INDEX 759 Tinting test, 519. Titanic acid in cement, 200. Titanic oxide in clay, fire sand and stone, 252. Titanium in iron ore, 76. Tobin bronze, 165. Toilet soaps, 475. Total solids, in water, 583, 596. Toughness test, for macadam, 269. Transformer oils, 428. Trap rock, crushing strength, 254. Trolley wheel bronze, 169. True red lead, in pigments, 516. Turbine oils, 427. Turbine wheel mixtures, 167. Turpentine, analysis, 538. Turpentine, specifications, 540. Tuscan red, 496. Tutwiler and Bond hygrometer, 659. Type metal, 165. Ultimate analysis of oils, 472, Ultramarine, 497, 531. Umbers, 496. Union Pacific R. R. Report on water supply, 611. United Gas Improvement Com- pany, analysis illuminating gas, 655. United Gas Improvement Com- pany, candle-power computer, 718. Lnited Gas Improvement Com- pany, standard bar photom- eter, 716. U. S. Steel Corporation method of sampling, 83. Unsaponified matter in soap, 479. Van Dyke brown, 496. Vanier combustion train for steel analysis, 133. Vanier potash bulb, 472. Variation in coal analysis due to size, 16. Varnish analysis, 337. Vegetable black, 497. Vegetable oils, 453. Vehicle in mixed paints, 535. Vermillion, 496, 533. Viscat apparatus for cement con- sistency, 213. Victor metal, 169. Viscosimeter, Doolittle, 378. Viscosimeter, Engler, 299, 372, 462. Viscosimeter, Redwood, 380. Viscosimeter, Saybolt, 374. Viscosimeter, Tagliabue, 376. Viscosity of oils, 372, 423, 462. Volatile and combustible matter in coal, i, 51. Volatile matter in coal, alternate method, 14, Volatile matter in coal, muffle method, 14. Volatile matter in coke, 54. Volatile matter in pigment, 523. Volatilization test, asphalt, 283, 309, 345. Volatilization test, asphaltic ce- rnent, 347. Volatilization test, asphalt re- siduum, 346. Volumetric method for nickel in steel, 149. W Water, acid in, 579. Water analysis certificate, 599. Water analysis for scale forming ingredients, 570. Water, bacteriological examina- tion, 600. Water, boiler, complete analysis, 576- Water, boiler, rapid analysis, 582. Water, boiler, sample analyses, 581. Water, color, 598. Water extract, in pigments, 523. Water feed generators, 679. Water, filtration, 573, 599, 600. Water for locomotives, 606. Water gas manufacture, 670. Water gas tars, 339. Water, hardness, 584. Water in gypsum, 511. Water in iron ores, 74, Water in mortars, 214. Water in pigment, 522, 523. Water in soap, 476, 479. Water of hydration in clay, fire sand and stone, 253. Water, sanitary analysis, 588. 760 INDEX Water test for oils, 400. Wearing surface, asphalt, 342. Weight of paper, 556, 567. Welding, oxy-acetylene, 689. Welsbach mantles, analysis, 730. Welsh coal, composition, 17. Wendler apparatus for paper test- ing, 557. White brass, 170. White lead, 496. White lead, analysis, 497. White lead, specifications, 518. White metal, 165. White metal, analysis, 172. White metal, specifications, 189. White pigments, 497. White pigment, analysis, 499. White zinc, specifications, 528. Whiting, 497, 509. Wijs solution, 542. Williams-Westphal balance, 366. Wilson colorimeter, 453. Wire, firing, 28. Wisconsin flash and fire tester, 443. Witherite, 513. Wolff's calorimeter, 592. Wood block pavement specifica- tions, 358. Wood fiber in paper, 546. Wood pulp, in dynamite, 731. Wool grease, 408. Wright and Thompson, scheme for soap analysis, 478. Yellow ochre, 497. Yellow pigment, 497. Yttrium oxide, in mantles, 730. Welsbach Zinc analysis, 93. Zinc chrome, 497. Zinc in brass or bronze, 161, 183. Zinc in ores, 94. Zinc, in pigments, 504, 507, 508. 516, 518. Zinc, in pyrites, 78. Zinc-lead, 506. Zinc-lead white, 520. Zinc oxide, 520, Zinc oxide in cylinder deposits 728. Zinc oxide in Paris green, 495. Zinc oxide in pigments, 508. Zinc sulphate in pigments, 504. Zinc sulphide, white, 496. Zinc sulphide, in pigments, 507, 508. Zinc white, 496. Zinc white, analysis, 506. Zirconium oxide, in Welsbach lamps, 730. SCIENTIFIC BOOKS PUBLISHED BY The Chemical Publishing Company, Easton, Penna. ARNDT-KATZ— A Popular Treatise on the Colloids in the Indus- trial Arts. Translated from the Second Enlarged Edition. i2mo. Pages VI +73 $0.75 ARNOLD— The Motor and the Dynamo. 8vo. Pages VI + 178. 166 Figures $1.50 BENEDICT — Elementary Organic Analysis. Small Svo. Pages VI 4- 82. 15 Illustrations $1.00 BERGEY— Handbook of Practical Hygiene. Small 8vo. Pages 164.. $1.50 BILTZ — Practical Methods for Determining Molecular Weights. (Translated by Jones.) Small 8vo. Pages VIII + 245. 44 Illustrations $2.00 BOLTON— History of the Thermometer. i2mo. Pages 96. 6 Illus- trations $1.00 BURGESS — Soil Bacteriology Laboratory Manual. i2mo. Pages VIII -f 123. 3 Illustrations $1.00 CAMERON— The Soil Solution, or the Nutrient Medium for Plant Growth. Svo. Pages VI -f 136. 3 Illustrations . $1.25 COLBY— Reinforced Concrete in Europe. 8vo. Pages X + 260... $3.50 DOLT— Chemical French. Svo. Pages VIII + 39S $3.00 EMERY — Elementary Chemistry. i2mo. Pages XIV + 666. 191 Illustrations $1.50 ENGELHARDT— The Electrolysis of Water. Svo. Pages X + 140. 90 Illustrations •. $1.25 FRAPS — Principles of Agricultural Chemistry. Svo. 2nd Edition. Tages VI + 501. 94 Illustrations $4.50 GILMAN — A Laboratory Outline for Determinations in Quantita- tive Chemical Analysis. Pages 88 $0.90 GUILD — The Mineralogy of Arizona. Small i2mo. Pages 104. Illustrated $i.oo HALLIGAN — Elementary Treatise on Stock Feeds and Feeding. 8vo. Pages VI + 302. 24 Figures $2.50 HALLIGAN— Fertility and Fertilizer Hints. 8vo. Pages VIII + 156. 12 Figures $1.25 HALLIGAN— Soil Fertility and Fertilizers. 8vo. Pages X + 398. 23 Figures $3.50 HARDY — Infinitesimals and Limits. Small i2mo. Paper. Pages 22. 6 Figures $0.20 HART — Chemistry for Beginners. Small i2mo. Vol. I. Inorganic. Pages VIII + 214. 55 Illustrations, 2 Plates $1.00 HART — Chemistry for Beginners. Small i2mo. Vol. II. Pages IV + 98. II Illustrations $0.50 HART— Second Year Chemistry. Small i2mo. Pages 165. 31 Illustrations $1.25 HART, R. N. — Leavening Agents. 8vo. Pages IV + 90. 13 Illus- trations $1.00 HART, R. N.— Welding. 8vo. Pages XVI + 182. 93 Illustrations .. $2.50 HEESS— Practical Methods for the Iron and Steel Works Chemist. 8vo. Pages 60 $1.00 HILL — A Brief Laboratory Guide for Qualitative Analysis. i2mo. Pages VI + 80 $1.00 HINDS— Qualitative Chemical Analysis. 8vo. Pages VIII + 266.. $2.00 HOWE — Inorganic Chemistry for Schools and Colleges. 8vo. Pages VIII + 422 $3.00 JONES— The Freezing Point, Boiling Point and Conductivity Meth- ods. Pages VIII + 76. 2nd Edition, completely revised $1.00 KRAYER— The Use and Care of a Balance. Small i2mo. Pages IV + 42. 18 Illustrations $0.75 LANDOLT— The Optical Rotating Power of Organic Substances and Its Practical Applications. 8vo. Pages XXI -\- 751. 83 Illus- trations $7-50 LEAVENWORTH— Inorganic Qualitative Chemical Analysis. 8vo. Pages VI + 153 $1-50 LE BLANC — The Production of Chromium and Its Compounds by the Aid of the Electric Current. 8vo. Pages 122 $1.25 LOCKHART— American Lubricants. 8vo. Pages X + 236. Illus- trated $2.00 MASON— Notes on Qualitative Analysis. Small i2mo. Pages 58 $0.80 MEADE — Chemists' Pocket Manual. i2mo. 3rd Edition. Pages IV + 530. 43 Figures $3.50 MEADE— Portland Cement. 2d Edition. 8vo. Pages X + 512. 169 Illustrations $4.50 MOELLER-KRAUSE— Practical Handbook for Beet-Sugar Chemists. 8vo. Pages VIII -\- 132. 19 Illustrations $1.25 MOISSAN— The Electric Furnace. 8vo. Pages X -f 305. 41 Illus- trations $2.50 NIKAIDO— Beet-Sugar Making and Its Chemical Control. 8vo. Pages XII -f 354. 65 Illustrations $3.00 NISSENSON— The Arrangement of Electrolytic Laboratories. 8vo. Pages 81. 52 Illustrations $1.25 NOYES — Organic Chemistry for the Laboratory. 3d Edition, revised. 8vo. Pages XII -\- 292. 41 Illustrations $2.00 NOYES AND MULLIKEN— Laboratory Experiments on Class Reac- tions and Identification of Organic Substances. 8vo. Pages 81 . . $0.50 PARSONS— The Chemistry and Literature of Beryllium. 8vo. Pages VI -i- 180 $2.00 PFANHAUSER— Production of Metallic Objects Electrolytically. 8vo. Pages 162. 100 Illustrations $1.25 PHILLIPS— Chemical German. 2d Edition. 8vo. Pages VIII -\- 252 $2.00 PHILLIPS— Methods for the Analysis of Ores, Pig Iron and Steel. 2d Edition. 8vo. Pages VIII + 170. 3 Illustrations $1.00 PRANKE — Cyanamid (Manufacture, Chemistry and Uses). 8vo. Pages VI + 112. 8 Figures $1.25 SE6ER — Collected Writings of Herman August Seger. Papers on Manufacture of Pottery. 2 Vols. Large 8vo per vol. $7.50 or $15.00 per set. STILLMAN— Engineering Chemistry. 5th Edition. 8vo. Pages VIII 4- 760. 150 Illustrations $5.00 STILLMAN— Examination of Lubricating Oils. 8vo. Pages IV + 125. 35 Illustrations $i-25 TOWER— The Conductivity of Liquids. 8vo. Pages 82. 20 Illus- trations $1.50 VENABLE— The Development of the Periodic Law. Small i2mo. Pages VIII + 321. Illustrated $2.50 VENABLE— The Study of the Atom. i2mo. Pages VI + 290 $2.00 VULTE — Household Chemistry. i2mo. 2nd Edition. Pages VI + 243 $150 VULTE AND VANDERBILT— Food Industries— An Elementary Xext-book on the Production and Manufacture of Staple Foods. 2d Edition. Bvo. Pages X + 327. 81 Illustrations $2.00 WILEY — Principles and Practice of Agricultural Analysis. Vol. I — Soils. Pages XII + 636. 92 Illustrations $4.00 WILEY — Principles and Practice of Agricultural Analysis. Vol. II — Fertilizers and Insecticides. Pages 684. 40 Illustrations. 7 Plates $4.50 WILEY — Principles and Practice of Agricultural Analysis. Vol. Ill — Agricultural Products. Pages XVI -|- 846. 127 Illustrations $6.00 WYSOR — Analysis of Metallurgical and Engineering Materials — a Systematic Arrangement of Laboratory Methods. Size 8j^ x loj^. Pages 82. Illustrated. Blank Pages for Notes $2.00 WYSOR— Metallurgy— a Condensed Treatise for the Use of College Students and Any Desiring a General Knowledge of the Sub- ject. 2d Edition, revised and enlarged. 8vo. Pages XIV -f 391. 104 Illustrations $3.00 ZIE6EL — Brief Course in Metallurgical Analysis. Pages VI -j- 72... $1.00 THIS BOOK IS DUE ON THE LAST DATE STAMPED BEIiOW AN INITIAL FINE OF 25 CENTS WILL BE ASSESSED FOR FAILURE TO RETURN THIS BOOK ON THE DATE DUE. THE PENALTY WILL INCREASE TO SO CENTS ON THE FOURTH DAY AND TO $1.00 ON THE SEVENTH DAY OVERDUE. LIBRARY, COIiLEGE OF AGRIOULT UKIVKRSITY OF OATjIFOENj- Book Slip-10m-9,'4c Stillman \^'\ Engine e: 'ing chemis try "^* iiii 7 m m^4':. \^^ ^' UNIVERSITY OF CALIFORNIA LIBRARY