The Comparative Value of Different Specimens of Iodine for Use in Chemical Measurements DISSERTATION PRESENTED IN PARTIAL FULFILLMENT OF THE REQUIRE- MENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN THE GRADUATE SCHOOL OF THE OHIO STATE UNIVERSITY BY SAMUEL MORRIS THE OHIO STATE UNIVERSITY ^- 1921 The Comparative Value of Different Specimens of Iodine for Use in Chemical Measurements DISSERTATION PRESENTED IN PARTIAL FULFILLMENT OF THE REQUIRE- MENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN THE GRADUATE SCHOOL OF THE OHIO STATE UNIVERSITY BY SAMUEL MORRIS THE OHIO STATE UNIVERSITY 1921 THE COMPARATIVE VALUES OF DIFFERENT SPECIMENS OF IODINE FOR USE IN CHEMICAL MEASUREMENTS. Introduction. This paper 1 gives the results of experiments in which iodine as ordinarily prepared as a standard in volumetric analysis was compared with iodine of atomic weight purity, the medium of comparison being a sodium thio- sulfate solution. Papers of a similar sort that were found are by Gross 2 and Meineke. 3 Gross' paper is very brief and contains none of the original data. Meineke, on the other hand, prepared a number of specimens of iodine by different methods and through the medium of a thiosulfate solution compared them with an iodine purified by the method of Stas. His technique, however, 1 This is the third of a series of investigations at the Ohio State Laboratory in which some fundamental point in analytical chemistry is studied by comparing the substance or operation in question with a standard substance prepared and measured with as near an approach to "atomic weight" accuracy as may be. Other similar investigations are in progress and it is hoped in time to develop new methods of preparing standard sub- stances and new apparatus for increasing the accuracy of the important analytical measurements. The other papers are in J. Am. Chem. Soc., 36, 2360 (1914), and 40, 1664 (1918). 1 Gross, J. Am. Chem. Soc., 25, 987 (1903). 8 Meineke, Chem. Ztg., 16, 1126. 1219, and 1230 (1892). 492858 was that of an accurate analytical procedure, rather than of a precise measurement, and therefore it seemed worth while to supplement his results by the scheme followed in this paper. Meineke's papers are rich in information about the purification of iodine and contain a wealth of experimental data. Preparation of Materials. Water. "Conductivity" water was used throughout. Starch Indicator. A sterilized starch paste made by the method given in Tread- well's "Quantitative Analysis" was employed. Potassium Iodide. One of the so-called analyzed brands was used. Tests showed that no iodine was set free from its solutions under the conditions of this work. Hydrogen Sulfide. The gas was generated from ferrous sulfide with c. P. sulfuric acid. Before use it was washed 5 times with water. Potassium Permanganate. The c. p. salt was recrystallized twice in platinum. It gave no reaction for chlorine. The Various Specimens of Iodine. "ATOMIC WEIGHT" IODINE. One of the general methods of Baxter 4 was faithfully followed, excepting as noted below, the cycle of purification processes being carried through three times. The exceptions were (a) a modification of the apparatus for distilling iodine from a liquid mixture and (6) the apparatus for the final drying of the iodine at the conclusion of the wet processes of purification. The distillation apparatus is illustrated in Fig. 1. The essential feature is the condensation tube, b, about 19 mm. in diameter and long enough to project 8 or 10 cm. into the GOOcc. Erlenmeyer flask, a. It is cooled by a stream of tap water, passing through it as shown in the drawing and is held in position by a bulb blown near the top. This bulb is given an umbrella shape which serves for closing the neck of the flask against loss of iodine and also for protecting it against falling dust or other contamination. By heating the mixture in the flask almost to boiling the iodine distils and solidifies on the condensa- tion tube. There is no contamination due to bumping with con- sequent projection of liquid onto the iodine. A slight danger of splashing from falling drops of water that had condensed on the cold tube is eliminated by placing a narrow glass plate as shown at e. The drops strike this and glide down gently. This ap- Fig. 1. paratus, which is analogous to one described in Treadwell's "Quantitative Analysis," has proved of great value in the distillation of iodine. Drying and sublimation of the iodine was effected in an apparatus made from a Hempel desiccator modified by inserting supports for carrying a number of watch glasses, and by grinding in a condensation tube which projected through the tubulure at the top. Phosphorus pentoxide was placed on the watch glasses and in the circular tray of the desiccator. Iodine to be dried is placed in a dish in the bottom of the apparatus, which is moderately warmed. The temperature of the interior at a point about midway between the top and bottom was 40 . Under these conditions iodine vapor ascends slowly over and around the phosphorus pentoxide and solidifies on the condensation tube, which is cooled with tap water. Like the dis- tillation apparatus, this drying tower was found to have many advantages. Its use saved time because it needed no attention while in operation. As a final note, it should be stated that all transfers of the "atomic weight" iodine 4 Baxter, J. Am. Chem. Soc., 26, 1579 (1904). from one vessel to another that occurred in the last of the series of purification operations were made in a dust-proof box equipped with rubber sleeves. IODINE 2. Commercial iodine was ground with potassium iodide and sublimed, and the sublimation repeated without the use of potassium iodide. The final product was then ground in a mortar and placed over calcium chloride. IODINE 3. This was commercial "resublimed" iodine just as received from the laboratory store-room. IODINE 4. Iodine recovered from titration residues and from residues from the purification of iodine by the Baxter method was the starting point of this preparation. When free iodine was present in these residues it was reduced by hydrogen sulfide and the resulting hydriodic acid neutralized by sodium hydroxide. The mixture was then evaporated in the open air, treated with manganese dioxide and sulfuric acid and the iodine distilled. This iodine was then twice dissolved in potassium iodide solution and redistilled. It was dried in a vacuum desiccator over calcium chloride for one week. IODINE 5. This was the same as Iodine 4, except that it was dried in the phosphorus pentoxide apparatus. IODINE 6. This was prepared in all respects like the "atomic weight" iodine, excepting that it was not sublimed in the phosphorus pentoxide apparatus. The product from the last distillation from potassium iodide solution was, instead, placed over sul- furic acid where it remained for 93 days before it was used. Comparison of the Various Specimens of Iodine. Balance and Weights. A long-armed Troemner balance and a set of gold-plated brass weights with the fractional parts of platinum were used. These weights had been calibrated by the Bureau of Standards. All weighings were made at night in order to avoid the vibrations of the building. The method of tares was used throughout and it is believed that the weighings are defined to within 0.02 to 0.03 mg. Since there is no reason to suppose that the various specimens of iodine differed materially in density, weights were not reduced to a vacuum, but they were corrected for the different densities of brass and platinum, so that the values as finally given are those that would have been obtained had all of the weights been of platinum. Preparation of the Iodine for Weighing. 1. In case the iodine was to be fused it was placed in a porcelain boat in the heating tube of a Richards bottling apparatus. 6 After passing dry air over the sample for 20 to 30 minutes, the part of the tube occupied by the boat was cautiously heated until the iodine just melted. With care the iodine could be maintained in the liquid state for one or two minutes without undue loss by volatilization. The usual procedure was to keep it fused for one minute and then push the boat into a cooler part of the tube towards the bottling chamber. The bottom of the boat being then in contact with cold glass cooled while the upper part remained warm. Iodine that had sublimed on the edges of the boat volatilized again, so that in the end the only iodine visible was; the hard, glassy looking cake in the bottom of the boat. During the whole operation the current of dry air was maintained and continued until every- thing was completely cool. The boat was then bottled and after lying in the balance case for at least an hour was weighed. The weighing bottle and empty boat had previously been put through the same process at 8 Richards and Willard, J. Am. Chem. Soc., 32, 25 (1910). 6 least twice or until constant weight was reached. In the weighing of all samples a weighing bottle and boat as nearly as possible like those con- taining the iodine were used as a counterpoise. The air passing through this bottling apparatus was first filtered through a 2-liter bottle filled with cotton wool. It next passed through an all-glass series of purifying and drying towers, the first one of which contained glass beads over which silver nitrate solution trickled, and the second was filled with sodium carbonate that had previously been fused and the fusion cake broken into pieces; next followed 4 towers of beads over which cone, sulfuric acid was kept flowing, and finally a tube filled with alternate layers of glass wool and phosphorus pentoxide. 2. When the sample of iodine was not to be fused, it was transferred from its container to the weighing bottle as quickly as possible and with cautions against contamination from dust. Sodium Thiosulfate Solution. The solution of sodium thiosulfate used throughout the work was a residuum of 5 or 6 liters from a larger lot that had been made up in December, 1918, and standardized in January, 1919, by titrating it from a volume buret against resublimed iodine prepared as described above for Iodine 2. Its value then was 0.1003 N. During the interval of 2 years the solution had stood in an uncolored glass bottle exposed to the light but protected by a soda-lime tube against acid fumes and carbon dioxide. When it assumed a rdle in this investigation it was placed in a dark room. The first hope that a thiosulfate solution as old as this had become constant was unfounded, for a few months' work showed a progression in the results that pointed to a gradual decrease in its iodine value. Accordingly, in the following table the final thiosulfate values of the various specimens of iodine are calculated to the value which the thio- sulfate solution had on April 10, 1921. The correction factor was calculated from the average thiosulfate solution values of 0.40000 g. of "atomic weight" iodine as found in Groups I and II, and from the corresponding values of Iodine 2, as found in Groups III, IV, and V in the table. This gives three comparable time intervals, which coincide well with the whole work, and from the data the average daily change in this thiosulfate solution in terms of weight of solution to react with 0.40000 g. of iodine was found to be 0.000787 g. This amount multiplied by the time in days between the date of an experiment and April 10, was added or subtracted, as the date demanded. The authors are, of course, aware that such a procedure is not to be recommended in precise measurements, but in their case it became a necessity in order to save the results. The error introduced does not destroy the value of the results as a contribution to analytical chemistry Solution and Titration of the Iodine. Two titration flasks were employed similar in all respects excepting that one of them was closed with a rubber stopper and the other one was entirely of glass as illustrated by Fig. 2. The ground-in stopper carried 2 tubes, a, for introducing the sodium thiosulfate solution and b, for the escape of displaced air. Loss of iodine vapor during a titration was prevented by having Tube a, dip below the surface of the liquid and by potassium iodide solution in the bends of Tube b. The introduction of the weighed portions of iodine into this titrating flask was accomplished by first loosening but not removing the stopper of the weighing bottle and then putting the bottle into the flask. Next with the flask in an inclined position the stopper of the weighing bottle was removed with a glass hook which was quickly withdrawn and the flask closed. Two cc. of saturated potassium iodide solution was then added through Tube a, followed by a little water. By properly rotating and inclining the titrating flask the boat was brought out of the weighing bottle so that all surfaces, that is, of the weighing bottle, its stopper, which had been left in the flask, and of the boat, could be bathed in the strong potassium iodide solution. When the iodine was dissolved, the volume was made up to about 150 cc. with water and was ready for titration. This was done according to the following routine. Thiosulfate solution was added in approxi- mately 5g. portions with intermediate mixing by shak- ing the flask. When all but a trace of iodine had been re- duced, starch indicator was added and the titration finished. In order to reach a uniform end-point 4 comparison flasks of the same size and color of glass as the titrating flask were prepared and charged with solutions in every way like that in the titration flask at the end of a titration excepting in respect to the content of iodine. Comparison flask No. 1, contained no iodine and Nos. 2, 3 and 4 had amounts of iodine equivalent respectively, to 0.006 g., 0.012 g., and 0.018 g. of the thiosulfate solution. These varying amounts of iodine gave with starch a set of color standards by which it was easy to follow the approach of the end-point in the solution being titrated. In any event each end-point was tested by titrating back with an iodine solution. It was found that a more uniform end-point could be obtained with the light from an ordinary tungsten bulb than with the varying light of the sun. The bulb was supported a few centimeters above the desk top with a sheet of white paper above it to reflect the light downwards upon other white paper on the desk. Upon this white surface on the desk were placed the color standard flasks and from time to time the titrating flask for compari- son. The observations were made from a point above the light bulb. In this way the end-points were reproducible within 0.006 g. of the thio- 8 sulfate solution. Another reason for preferring artificial light to that of the sun is the action of the latter in liberating iodine. Concentrated solutions of potassium iodide became distinctly yellow in 10 minutes in direct sun- light. TABLE I ANALYTICAL RESULTS Iodine G. Iodine G. Thio- sulfate solution G. Thio- Average Average Purity of sulfate corrected iodine solution for change calculated in thio- to 0.40000 sulfate iodine solution G. G. G. % Group I, April 10, 1921, .39206 32 ,3748 33 .030 "Atomic Weight" Iodine, .45868 37 .8628 33 .019 not fused. .48325 39 .8879 33 .014 33.030 33.030 100.000 .49642 41 .0163 33 .050 57185 47 .2322 33 .038 Group II, April 29, 1921, .54560 45 .0777 33.049 "Atomic Weight" Iodine, .33932 28 ,0087 33 .018 fused. .40043 33 .0710 33 .035 33.045 33.030 100.000 .57602 47 .6233 33 .071 ,49461 40 .8684 33 .051 Group III, Jan. 18, 1921, .37134 30 .6151 32 .978 Iodine 2, resublimed once .46267 38 .1505 32 .983 from KI and once alone; .55278 45 .6171 33.009 ground and kept over .46183 38 .0386 32 .946 32.973 33.038 100.024 CaCl 2 . .42732 35 .1963 32 .946 .41060 33 .8484 32 .975 .44523 36 .7008 32 .972 Group IV, Feb. 16, 1921, .54446 44 .9246 33 .005 Iodine 2, as described .52526 43 .3212 32 .991 32.997 33.039 100.018 above. .46477 38 .3382 32 .995 .46002 37 .9478 32 .995 Group V, April 20, 1921, .49583 40 .9530 33 .038 Iodine 2, as described .49537 40 .9209 33 .043 33.044 33.036 100.018 above. .51626 42 .6567 33 .051 ,46010 38 0061 33 042 Group VI, Feb. 17, 1921, .26374 21 .7366 32 .968 Iodine 3, product just as 0. 36245 29. 9089 33. 008 32.995 33.036 100.018 received from store-room, .34615 28 .5655 33 .009 except that it was fused before weighing. Group VII, Feb. 23, 1921, .35231 29 .0509 32 .984 Iodine 3, as above, but not .40486 33 .4021 33 .002 fused before weighing. .38186 31 .4888 32 .984 32.994 33.030 100.000 .43344 35 .7709 33 .011 .41145 33 .9214 32 .977 .46869 38 .6702 33 .003 g Group VIII, April 21, 1921, Iodine 3, as above, but not fused. Group IX, March 17, 1921, Iodine 4, recovered from residues; dried over CaCU for one week but not sub- limed. Group X, March 22, 1921, Iodine 5, same as No. 4, excepting that it was sub- limed in PjOj apparatus. Group XI. Feb. 14, 1921, Iodine 6, "Atomic Weight" Iodine; not sub- limed but kept 93 days over H 2 SO 4 . Group XII, May 5, 1921, "Atomic Weight" Iodine after 7 hours exposure in current of air saturated with water vapor. 0.45876 0.56198 0.49452 0.50921 37.8893 46.4199 40.8364 42.0799 33.036 33.041 33.031 33 .055 0.45363 37.3096 32.899 0.28391 23.3198 32.854 .36848 30 .4221 33 .024 .39392 32 .5201 33 .022 .46061 38 .0158 33 .014 .38952 32 .1678 33 .033 .53442 43 .9857 32 .921 .40481 33 .3249 32 .929 .42985 35 .3811 32 .924 42995 35 .4024 32 .936 n .42324 34 .9834 33 .062 .50817 41 .9983 33 .058 .54587 45 .1022 33 .050 .49967 41 .2612 33 .031 .43242 35 .7428 33 .063 33.041 33.032 100.006 32.877 32.889 99.57 33.023 33.038 100.024 32.928 32.971 99.82 33.053 33.033 100.009 Discussion of Results. With the exception of Nos. 4 and 6, the specimens of iodine examined agreed within 0.02% with the "atomic weight" iodine. The deviation from the "atomic weight" iodine, except as noted, was always above 100%, which would seem to indicate such impurities as chlorine and bromine. This, however, is unlikely, since the unpurified iodine, No. 3, has a smaller positive error than the others. It would also seem from the nature of its preparation that Iodine 5 would hardly be contaminated that way. The cause is seemingly in a systematic error that has escaped detection. Evidence of water in No. 4 is found in the fact that it rose in value when put through the phosphorus pentoxide drying tower. The low value of No. 6 could be accounted for on the basis of a statement by Treadwell, 6 that iodine takes up sulfuric acid when dried over it. This statement, however, can hardly be accepted in the light of evidence to the contrary in the literature. Baxter 7 dried some of his atomic weight iodine over sulfuric acid and makes no mention of it as a possible impurity that was removed in the subsequent sublimation of the iodine. Meineke 8 reports several experiments in which iodine was exposed for days over ' Treadwell, "Quantitative Analysis," John Wiley and Sons Co., 3rd Ed., p. 646. 7 Baxter, /. Am. Chem. Soc., 32, 1594 (1910). 8 Meineke, Ref. 3, pp. 1126, 792. 10 sulfuric acid without evidence that any of the acid was taken up by the iodine. It is the belief of the authors that Iodine 6 still contained water in spite of its long exposure over sulfuric acid. An inspection of Nos. 4 and 6 shows that they had one point in common, namely, distillation from water and condensation on a cold surface on which water was condensing at the same time in contact with the iodine. This condensation in the wet way was the last purification process, except the drying (?) over a desiccating agent to which Nos. 4 and 6 were subjected. No. 4, undoubtedly con- tained water (see above). Lack of time and material made it impossible to test No. 6 directly, and until the point can be investigated experiment- ally, the hypothesis is offered that iodine when it solidifies in the presence of liquid water entrains some of the water in a way that withstands months of exposure over a desiccating agent. The complete removal of water from iodine demands dry sublimation in the presence of a desiccating agent. Is Iodine Hygroscopic? Meineke 9 exposed dry iodine for 5 days to the air of his laboratory and reports that the change observed could be ascribed to experimental errors. Powdered iodine exposed under a bell jar with a dish containing water took up 0.09% of water in 48 hours. Iodine crystals when subjected to the same conditions took up onlyO .05% of water in 5 days. An exposure of 10 days over sulfuric acid brought these last two specimens back to 100.00%. Meineke's iodines had all been sublimed in the dry way and therefore the situation is not comparable with that of No. 6, above. In this investigation, the following experiment was made to test the hygroscopic nature of iodine. About 5 g. of the "atomic weight" iodine was spread in a thin layer in a glass tube and a current of air previously bubbled through water was passed over it for 7 hours at the rate of about 9 liters an hour. The results are given in Group XII, of the table, and show that if any water was taken up, the amount was well within the experi- mental errors. This experiment and those by Meineke show that iodine is not hygroscopic to a degree that would affect analytical work. A Possible Contribution to the Data of Adsorption. One of Meineke's experiments shows that when iodine is finely divided and exposed to air saturated with water vapor, a small amount of water is taken up. It is reasonable to suppose that this water is adsorbed on the surface of the solid iodine, and the question then arises as to what effect the relatively high vapor pressure of the iodine has on such adsorption. If adsorption is a surface phenomenon, then a surface like that of an iodine crystal which is constantly sending off molecules and therefore constantly presenting a fresh surface must introduce complications. The rate of adsorption under Ref. 3, p. 1126. 11 such conditions would be slow and if later such a solid were placed in an atmosphere free from water vapor the portion adsorbed would soon be given up because the surface holding it would have disappeared. One would therefore expect solids with high vapor pressures to be slightly or not at all hygroscopic. The Use of a Rubber Stopper in the Titrating Flask. The titrations of Group VII were made in a flask closed with a rubber stopper, and those of the strictly comparable Table VIII were carried out in the all-glass apparatus. The manipulation was the same in both cases. The agreement between these two sets of results is almost perfect, and therefore the use of a rubber stopper may be permitted in analytical operations. Summary. Some new forms of apparatus for the preparation and handling of pure iodine are described. Various specimens of purified iodine were compared with a highly purified iodine. The agreement was found to be close; 0.024% was the maximum deviation. Experiments are described which show that a rubber stopper may be used for closing the flask in which iodine is titrated. The usual method, as given in text-books of analytical chemistry, of drying iodine by exposing it in a desiccator with a drying agent, is ques- tioned if the iodine had previously solidified in the presence of liquid water. The hygroscopic properties of iodine are discussed. A suggestion is made that solids like iodine that have a measurable vapor pressure at room temperature possess peculiar adsorptive properties due to the fact that a fresh surface is being continually exposed. The author wishes to express his thanks to Professor C. W. Foulk for his help and untiring interest throughout this work. AUTOBIOGRAPHY. I, Samuel Morris, was born near Bloomingburg, Ohio, October 20, 1879. I received my secondary school education in the public schools of Fayette County Ohio, graduating from Bloomingburg High School in 1898. My undergraduate and graduate work was all done at the Ohio State Uni- versity, receiving the Degree of Bachelor of Arts in 1905, the Degree of Master of Arts in 1908 and the Degree of Doctor of Philosophy in 1921. YD 04904 JL/M& UNIVERSITY OF CALIFORNIA LIBRARY