/ Issued NovemU U. S. DEPARTMENT OF AGRICULTURE, OFFICE OF PUBLIC ROADS— CIRCULAR No. %. LOGAN WALLER PAGE, l>ni! UC SOUTHERN REGIONAL LIBRARY FACILITY AA 001 217 987 NAPHTHALENE IN ROAD TARS. I. -THE EFFECT OF NAPHTHALENE UPON THE CONSISTENCY OF REFINED TARS. MY PRfiVOST hubbard, Mtc /loads, ASP CLIFTON \. DRAPER, Uni. - .agton h. l-KMN I IN LI A R \ ,ETTER OF TRANSMITTAL. U. S. Department of Agriculture, Office of Public Roads, Washington, D. C, August 2, 1911. Sir: I have the honor to transmit herewith the manuscript of a circular, by Mr. Prevost Hubbard and Mr. Clifton X. Draper, of this office, on the Effect of Naphthalene upon the Consistency of Refined Tars. This publication is one of a series of papers which will be prepared under the general subject, Naphthalene in Road Tars. A systematic investigation of the effect of various constituents upon the quality and adaptability of tars used as road binders would seem to be of the greatest importance, first, because of the present lack of data bearing upon this subject, and second, because the utilization of tars in road treatment and construction seems destined to effect a large proportion of the road work in this country. I therefore respectfully request that this paper be issued as Circular 96 of this office. Respectfully, Logan Waller Page, Director. Hon. James Wilson, Secretary of Agriculture. 2 SRLF URL THE EFFECT OF NAPHTHALENE UPON THE CONSISTENCY OF REFINED TARS. Naphthalene, C ,H 8 , often occurs in coal tar in Larger quantities than any other one hydrocarbon and for this reason it is natural to suppose that it exerts an appreciable influence upon the quality and applicability of coal tars used as road binders. In this country the dse of tars in road treatment and construction has advanced rapidly during the past few years and upon the market to-day are to be found a number of tar preparati< ns intended for such use, I" preparations i rule made by distilling nil' the Lighter and more volatile constituents from both crude coal tar and crude water . tar or mixtures i !' the two. The residual products, which vary in consistency according to th< distillation, are sold for road purposes. Man; i - and many failure- have re- sulted from the use of coal tar and wat< r road binders, and often no clew has been obtainable as to the reason for such dive results. In view of this fact, a systematic .study of the principal chemical constituents of road-tar preparations and their effect upon the quality of such preparations is now being made in the Labora- tories of the Office of Public Roads, United Stun- Department of Agriculture. In the present paper, which is the first of a series upon naphtha- lene, no attempt will be made to review the chemistry of t ! i i — sub- stance, -ince this discussion is reserved for a later publication which will cover the entire subject as presented in the individual pap A few remark- concerning certain physical properties of naphtha- lene may. however, not be amiss at this point. Naphthalene, in the pur exists in white crystallim of thin rhomboidal scales melting at ''■' C. and having i boiling point of 218 ('. !i specific gravity :it 16 ( '. is L.1517. It ha very characteristic odor, commonly familiar in moth balls, and i- ex- tremely volatile, considering it- high molecular weight, 90 much 90 that in the manufacture of coal gas it is only partially deposited in the condensers, while the remainder is carried into the purifiers 10481 Clr. 06 11 4 NAPHTHALENE IN ROAD TABS. and in(<> the gas mains, where it deposits in i he bends of pipes during cold weather and often causes them to become clogged. Naphtha- lene volatilizes far below its boiling point and, in crude tars, dis- tills to a considerable extent with aqueous yapor and also with the light tar oils, which accounts for its occurrence in the first fractions. Even at ordinary temperatures, it volatilizes slowly and gives off a penetrating tarry odor. Tt is slightly soluble in hoi water, but dissolves easily in alcohol, ether, fatty and essential oils, acetic acid, and especially in the phenols and lighter tar distillates. No entirely satisfactory method for the quantitative determination of naphtha- lene in tar has, as far as the authors are aware, been devised up to the present time, although methods have been put forth by vari- ous investigators for the determination of naphthalene in illuminat- ing gas. It is a rather generally accepted theory that, in the manufacture of coal gas, the formation of naphthalene is due to certain complex reactions which take place only at the high temperatures at which the retort is maintained in modern gas-house practice. According to Cooper, 1 "A few years ago — when lower temperatures were in vogue, and when lower makes per ton of coal carbonized were reg- istered — naphthalene was an almost unknown quantity ; but now that the stress of competition demands the highest possible output of gas, it seems also to bring in its train the formation of large quan- tities of naphthalene." It is certain, however, that other factors, such as the size, shape, and inclination of the retort with relation to the volume of the charge of coal, and the varying conditions to which the evolved gases are subjected before and during their passage from the retort, have a great deal to do with the quantity of naphthalene formed. Tars used as road binders are for the most part soft pitches obtained by distilling from 15 to 30 per cent of the lighter products from the crude material. The soft pitch contains a large propor- tion of the naphthalene which may have been present in the original tar, and it is the study of the effect of naphthalene upon this grade of material that is the subject of this paper. As consistency is one of the most important properties of a road tar, it was thought well first to investigate the effect of varying percentages of naphthalene upon a given tar. From its very nature naphthalene itself can not be considered as a binding material. For some time past it has been known, however. to serve as a flux for the binding constituents of tars, in spite of the fact that it is a crystalline solid. Thus it is possible to combine, by heating together, a hard tar pitch and a quantity of naphthalene with the formation of a soft product. It was thought, therefore, that i Journal of Gas Lighting, Vol. CX, May 24. 1010, p. 498. E( 1 UPON T1IK CONSISTENT 5 ABS. 5 a comparison of its fluxing value with thai of certain less volatile tar distillates would prove most instructive. For this purpose a rep- resentative coal tar was taken and distilled to 238 < '. in order t» remove all the naphthalene. This uaphthalene-free tar pitch, v liich was used for the following tests, was so hard that it was difficult t<> make an impression on ii with the finger. An arbitrary measure of the consistency of this pitch was then obtained by means of the New York Testing Laboratory float-test apparatus, which is customarily used for this purpose in the routine examination of tar prodl in the Office of Public Road-.' The iloat apparatus, briefly described, consists of an aluminum float or cup, o\ inches in diameter, into which is screwed a -mall brass conical collar or mold. This collar i- ; inch high, with an internal diameter at the top of :: inch and at the bottom of { inch. In making the tesl the brass collar is placed with the small end down on a brass plate which has been previously amalgamated with mercury by first rubbing it with a dilute solution of mercuric chlorid nitrate and then with mercury. A small quantity of the material to be tested is heated in a metal spoon unt il quite fluid, with care thai it suffers no appreciable loss by volatilization and that it i- kepi free from air bubbles. It is then poured into the collar in a thin stream until slightly more than level with the top. The surplus may be removed after the material ha- cooled to room temperature, by menus of a spatula blade which has been slightly heated. The collar and plate are then placed in a tin cup containing ice water maintained at 5° C. and left in this hath for ai least 1"> minute-. Meanwhile another cup is tilled about three fourth- full of water, placed on a tripod, and the water heated to any temperature at which it i- desired to make (he test. This temperature is accurately maintained, and is at no time throughout the entire tesl allowed to vary more tha le- half a degree centigrade from the temperature -elected. After the material to he tested ha- been kepi in the ice water for at lea-t IS minutes, the collar and content- are removed from the plate and screwed into the aluminum Iloat. which i- then immediately floated in the wanned hath. A- (he plug of hitiuuinoii- material becomes warm and fluid it is gradually forced upward and out of (he collar until water gains entrance to (he Iloat and causes it to sink. The time in seconds between placing (he apparatus on the water and when the water breaks through the hilumeii i- determined by means of a -top watch and is taken a- a iiiea-iuv of the consistency of (he material under examination. Results SO expressed are only relative, hut (hey are comparable to within a good degree of pre- cision. The temperature at which the water bath is maintained i- purely an arbitrary matter and. for routine work in the office, 32 ( '. 'Bulletin 38, r. 8. Dept of kgrlcull ire Ofll f Public Roads, pp m 16. 6 NAPHTHALENE IX ROAD TABS. and 50° C. have been selected as a matter of convenience when testing the comparatively soft and comparatively hard road tars. For the work described in this circular 55° C. and 80° C. were adopted as most convenient temperatures. The former was selected after making a scries of float tests upon the prepared coal tar pitch at temperatures varying from 40° C. to 00° C. In these tests it found that at 10° C. and 45° C. the tar had not softened sufficiently to complete the test at the end of 1 hour; at 50° (\ a test of 24 minutes and 18 seconds was obtained; at 55° C, 9 minutes and 47 seconds, and at 60° C., 3 minutes and 45 seconds. The duration of the test at 55° C. seemed most satisfactory as allowing sufficient leeway for comparative results without consuming - too much time. It was decided to select 80° C. as a second working temperature, as it was the nearest convenient one to the melting point of naphthalene. It was thought that, by making consistency tests upon the naphthalene tar-pitch mixtures, later described, at both 55° C. and 80° C, the results would show not only how much the mixtures softened at the higher temperature, but what differences, if any, might be due to naphthalene below and above its melting point. When the matter of temperatures had been decided upon and the consistency of the tar pitch had been determined at these tempera- tures, naphthalene mixtures were made as follows : Approximately 30- gram samples were accurately weighed into a series of small tin dishes, and varying percentages of c. p. flake naphthalene were added, starting w T ith 0.5 per cent of naphthalene and gradually increasing this amount to the extent of 30 per cent. In all, 13 samples were prepared. Before adding the naphthalene, each weighed sample of tar was warmed in the tin dish until fairly fluid, after which the naphthalene was added and the mixture well stirred. It was found better to add the naphthalene a little at a time during the process of mixing rather than to add the entire amount at one time, as by so doing there was less danger of losing some of the naphthalene by volatilization and a uniform mixture was more readily obtained. As the naphthalene w T as added, an increase in the fluidity of the mixtures was readily seen. Where 20 per cent of naphthalene was used, the mixture was found to be fluid at normal temperatures, whereas the original pitch was semisolid. The naphthalene com- bined with the tar with surprising readiness and, upon cooling. no traces of the naphthalene could be seen in the resulting mixture, except where high percentages were added. It was found, however, that between 15 and 20 per cent of naphthalene caused the tar to become saturated. The 15 per cent mixture was homogeneous when cold, but in the 20 per cent mixture a small amount of naphthalene crystals separated out upon cooling. This fact indicates that the tar pitch forms a sort of eutectic with naphthalene similar to certain metallic alloys. Above the approximate eutectic of 20 per cent the ECT UPON THE I BEFINED I \l:s. 7 naphthalene crystallized out more and mon percentage was increased. A mixture of 30 per cent of naphthalene solidified. This was due to the crystallization of the i naphthalem consistency of this mixture at 55 ('. was found to be higher than the preceding mixtures, owing to the separation of the naphthalene, which does not melt at this temperature. At 80 C. the consistency 10' 8 10 12 14 16 18 20 22 24 26 28 30 of this mixture ^;' s oil; that Is, it became very Quid, as would naturally be expected. The maximum increase of fluidity <>t" the tar was obtained with 20 per cent of naphthalene, which is prac- tically the eutectic point. These facts will be noticed «>n the curves A and A' i fig. l I, which show the consistency of the mixtures plotted against the varying percentages of naphthalene. 8 NAPHTHALENE I X ROAD TARS. Curve A shows the decided effecl which naphthalene has upon the tar :il 55° C, and A' the effecl ai 80° ( '. A.-, the temperature of the water bath during the floal tesi is raised, the curve A will approach A' and a scries of curves might be obtained, lying between A and A'. for water-bath temperatures between 55° C. and 80° ( '. The sensitiveness of the tar pitch to small increments of naphtha- lene is shown by the sharp drop in the first section of curve A, start- ing with the addition of 0.5 per cent of naphthalene, which increases the fluidity 1 minute 12 seconds. The fluidity increases very markedly until 7 per cent of naphthalene is reached, when the curve becomes more gradual, while increasing percentages of naphthalene have a much smaller effect. The curve drops, however, until 20 per cent of naphthalene is reached and then rises gradually to 25 per cent of naphthalene, owing to the separation of naphthalene, as previously explained. At 30 per cent of naphthalene a sharp rise is noted and the curve, if continued further, would rise almost perpendicularly, because the naphthalene is so much in excess that an accurate meas- urement of the consistency would be impossible. For reference to the percentage increase in fluidity owing to increasing increments of naphthalene, the data given in Table I calculated from the curves will show statistically what the curves represent graphically. Table I. — Percentage increase in fluidity of coal-tar pitch owing to additions of naphthalene and naphthalene-free tar distillate. Due to naphtha- Due to distillate. Per cent of lene. naphthalene or distillate. At 55° C At 80" C. At 55° C At 80° C Per cent. Per cent. Per cent. Per cent. 0.5 12.2 6.7 3.1 3.8 1.0 26.9 11.5 19.1 5.8 1.5 44.5 19.2 29.6 14.4 2.0 .->! 1. S 27.9 37.1 17.3 2.5 59.3 32.7 44.6 20.2 3.0 04.1 36.5 49.9 25.0 5.0 76.5 51.0 64.2 39.4 7.0 83.3 63.5 75.1 49.0 10.0 91.5 75.0 82.3 61.5 15.0 94.4 83.7 89.6 74.0 20.0 97.3 90.4 94.5 S3. 7 25.0 96.9 89.4 96.6 89.4 30.0 88.1 75.0 100.0 100.0 It will be observed that 7 per cent of naphthalene increases the fluidity of the tar 83.3 per cent at 55° C, while 20 per cent increases it 97.3 per cent, showing that the additional 13 per cent of naphtha- lene makes an increase of only 14 per cent in fluidity. This fact is shown by the gradual slope of the curve A between these points as contrasted with the steep slope up to 7 per cent. For comparison with the consistency results obtained with naphthalene, a portion of the original coal-tar pitch was saved for I.l' I ECT CPON I'll J". CONSISTENCY "I I 9 use in mixtures with a naphthalene-free tar distillate. Duplicate tests were carried out in the same way as were the naphthalene tests, except that equivalent percentages of a naphthalene-free distillate were employed. This distillate was a mixture of oils taken from the distillation of several water-gas tars and showed the following com- position upon fractional distillation in an Engler flask: Table II. — Fractional distillation of tar distillate. rature Per ■ in °C. ted. tiled. 1 I P 1.0 1.8 71.0 7.5 < ■ 1 l'|i ' l» ' 1 i' : 37.0 i Dry. The specifi: gravity of this distillate was L.006 at 26 C. and. upon cooling to —10° C., no solids crystallized out of solution. Thirteen samples were weighed out in the sj manner as for the naphthalene tests and the percentages of distillate added as before. Consistency results were obtained upon tin' mixtures of tar and distillate, according to the float apparatus, and at the same temperatures as in the naphthalene tests 55 ('. and B0* ('. The procedure was exactly duplicated in both ca that the results could be studied under exactly the same conditions. Tn figure 1 the results of this second series of tests are shown by curves B and B', and the corresponding tabulated data are shown in Table I as compared with the results from naphthalene. The effect exerted by naphthalene upon the fluidity of tar is, from these result-, a very important one in coal tars used for road con- struction. All coal tars contain naphthalene, but the amount in road-tar preparations varies from to LO per rem and higher. The degree of fluidity which a tar maintains under service condition- i- a most important consideration. The fact that a tar i- originally of proper consistency lot- certain work is no reason for supposing that this consistency will he maintained under service condition-. Tf the fluidity of the tar is due principally to nonvolatile tar oil-, it will undoubtedly maintain it- original consistency for a Ion period than it' the fluidity is due mainly to high naphthalene content. Therefore, in two tars id' the same consistency or degree of fluidity, one with a small naphthalene contenl and the other with a high naphthalene content, it would l>e natural to predict from tie' fore going data that upon exposure the low naphthalene tat- would main 10 NAPHTHALENE IX ROAD TARS. lain its original consistency for a longer time, if other things are equal. Owing to the volatility of naphthalene, the high naphthalene tar would undoubtedly lose more or less naphthalene, and hence its fluidity might be expected to decrease rapidly. This subject is now under investigation and will be covered in a later paper. 2 4 6 S 10 12 14 16 18 20 22 P£/?C£Arr or /vAP#rH4L£W£ : Fig. 2. 24 26 28 30 In addition to refined coal tars, refined water-gas tars and also refined mixtures of coal tar and water-gas tar are used to a consid- erable extent in the treatment and construction of roads. For this reason it was thought advisable to determine what effect naphthalene might have upon the consistency of a water-gas tar pitch. Water- gas tar as produced in the modern manufacture of carburetted water EFFECT UPON THE CONSIS i OF REFINED TABS. 11 gas is similar in many way-; to it- sister product, coal tar. Ii however, marked characteristics in which it differs from coal tar, namely, a lower specific gravity, a tower free-carbon contest, and a relatively .small amount of naphthalene. A representative sample of water-gas tai was distilled until the residual pitch had as nearly as possible the same consistency at 5S I . as the coal-tar pitch in the previous experiments. This water-tar pitch had a consistency of 9 minutes 19 seconds. I i divided into two portions, one for naphthalene mixtures and one for distillate mix- tures. As in the previous work, samples wer< weighed out ? varying percentages of naphthalene and a duplicate set of samples with an equivalent amount of the same naphthalene-free tar distil- late as u^ed in the coal-tar tests. The consistency was obtained in both sets of tests with the float apparatus, and the water bath was maintained at 55 ('. and 80 c ('.. respectively. The results from the naphthalene series arc showu in figure •_' by curves C and (". and those of the corresponding distillate l> and I It will he noted that these curves follow \< rv closely those shown in figure 1 and are of the same general type. The relation between the consistencies obtained with naphthalene and the naphthalene-free distillate is practically the same as for coal tar. and th live influence of the increase in fluidity i- similar, a- shown in Table 1 1 1. Indeed, this similarity i- remarkably close when it i- realized that the coal-tar pitch contained L8.34 per cent of fr< u, while the water-gas tar pitch contained only 1.77 p< r cent of free carbon. Table III. — Percentage of increase in fluidity of water-gas tar pitch, <>\