. i <*M ; V r *^-: * -w I THE TERNARY SYSTEM : BENZENE, ACETIC ACID, AND WATER BY A. T. LINCOLN It was shown by Bancroft , 1 about ten years ago, that the condition of equilibrium in a large number of cases of physical reactions could be represented by the Law of Mass Action ; that the exponential factors are not necessarily integers and in most cases are not ; and that, as in the case of chemical reactions, they are independent of the temperature. He showed that in the case of two non-miscible liquids and a consolute liquid, the equilibria can be represented by the Mass Caw Equation, and that there are only two sets of equilibria over the whole range of concentration, and these are represented by two different equations. He has shown the application of the mass law to a large number of other cases of physical reactions, such as, to two partially miscible liquids and a consolute liquid, to the pre- cipitation of a salt by a liquid, to the precipitation of a liquid by a salt, and to the precipitation of one salt by another. In none of these latter cases, however, has the relation between the facts and theory been worked out as yet with a very high degree of accuracy. In the case of one ternary system, benzene, water, and alco- hol, the writer 2 has shown that the Mass Caw Equation does represent the conditions of equilibria with a very high de- gree of accuracy, and that the Caw of Mass Action is applicable to this physical reaction, also, that the exponential factors are independent of the temperature as in the case of chemical re- actions. Previously, Waddell 3 studied the system, benzene, acetic acid, and water. He concludes from his experiments, that the Caw of Mass Action does not apply to this physical re- action, that the equilibria are not represented by simple expo- 1 Proc. Am. Acad. 30, 324 (1894). 2 Jour. Phys. Chem. 4, 161 (1900). 3 Ibid., 3, 233 (1898). 249 ' 7 > i\ Benzene , Acetic Acid and Water nential formulae, and that at higher temperatures the deviation from the Mass Law Equation is very much more pronounced than at lower temperatures. In view of the fact that my work had shown such a marked approximation of the experimental results to the values required by the theory in the case of a sys- tem analogous to this one with which Waddell worked, it seemed worth while to repeat his work and to ascertain if his conclu- sions are correct. With this in view the work was undertaken, and the results are given below. The thiophene-free benzene employed was fractionated and that portion coming over at 79.5 0 C under a pressure of 755.8 mm was collected and then recrystallized twice. The distilled water of the laboratory was treated with barium hydroxide in contact with which it remained for several days, when it was siphoned off and distilled. This distillate was collected by means of a block tin condenser and only the middle portion of the distillate was collected. A sample of the acetic acid was fractionated a number of times and the portion coming over between 115 0 - 11 6. 5 0 was collected and recrystallized many times. This sam- ple had a melting-point of 14.6°, and according to Allen, 1 repre- sents a purity of 98.7 percent. On titration with a N/40 barium hydroxide solution it was found to be 98.6 percent pure. In all the following work a correction was made for the water con- tained in acetic acid of this purity. All of the flasks and bottles used were thoroughly cleaned and steamed. The bottles in which the benzene, water and acetic acid, as well as the standard solution of barium hydroxide, were stored, were connected with accurately calibrated burettes, which were so arranged that the air which entered the bottles passed through drying vessels containing sulphuric acid or potassium hydroxide. They were, also, so connected that when the burettes were emptied, the air which took the place of the liquid came from the storage bottles. By these precautions the solutions were thoroughly protected during the series of experi- ments. 1 Commercial Organic Analysis, I, p. 387. ,13840 250 A. T Lincoln The determinations were made in 50 cc flasks (or in- 200 cc flasks) which had been thoroughly cleaned and steamed. Into one of the flasks were introduced 5 cc of acetic acid (or 100 cc) and to this was added a definite quantity of benzene, and then enough water was introduced to produce clouding at room tem- perature. The mixture was then warmed up a few degrees above the temperature at which the determination was to be made. When the contents of the flask became homogeneous the flask was transferred to a bath which was kept at the desired temperature. After remaining in the bath long enough to ac- quire the temperature of the same, if clouding did not result, the flask was removed, and a few drops of water added from the burette, and the flask warmed until the contents became homo- geneous, when it was returned to the bath and allowed to remain with occasional shaking until it had acquired the temperature of the bath. If clouding did not result, this process was continued until it was found that one drop of water caused the second liquid layer to appear. In order to. ascertain this point the flask had to be removed from the bath in which it was kept. So in order to make the observation and at the same time prevent the clouding resulting from cooling the walls of the flasks by con- tact with the air, the flasks were placed in a beaker in the bath and the beaker containing the flask and water from the bath was removed and the observation made. The bath employed was an ordinary Ostwald thermostat provided with a water turbine for stirring and no difficulty was experienced in keeping the temperature constant to within a few hundredths of a degree. In the manner just described the data given in the follow- ing tables were collected. In Table I are the data for the equilib- rium determinations at 25 0 C and the calculated values for the amount of water that should have been found and also the values of the constant are given in the last column. The head- ings of the other columns are self-explanatory. In Table II are given the data for the determinations made at 35 0 C. The headings of the columns are self-explanatory Benzene , Acetic Acid and Water 251 except the column marked 3, which contains the values found in order to produce the same degree of clouding. We shall refer to this subsequently. TabIvE I. Temperature 25 0 x — cc benzene, y = cc water per 5 cc acetic acid Formula n log x -|- log y = log C Tog 0 = 0.2875; # =0.6136 x benzene y water log C Found Calculated I 10.06 0-45 0-445 0.2829 10.06 o-45 0.445 0.2829 2 8.04 0.57 0.54 0.31 14 8.04 o.55 0.54 0.2959 8.04 o.55 0.54 0.2959 3 6.03 0.64 0.64 0.2850 6.03 0.62 0.64 0.2710 4 5-03 0.72 0.72 0.2878 5.03 0.69 0.72 0.2694 5.03 0.70 0.72 0.2756 5 3.02 0.99 0.98 0.2903 3.02 0.97 0.98 0.2814 6 2.51 1. 12 1. 10 0.2947 2.51 1. 12 1. 10 0.2947 7 2.01 I.29 1.26 0.2966 2.01 I.27 1.26 0.2898 8 I *5 I 1.47 1.50 0.2771 1. 5i 1.49 1.50 0.2830 9 1. 01 i-93 i-93 0.2882 1. 01 1.87 i-93 0.2745 10 0.80 2.23 2.22 0.2889 0.80 2.20 2.22 0.2830 252 A. T. Lincoln Table I. — ( Continued ). Formula n! log x -f- log_y = log C' = 0.244 ; n — 0.9166 x found y found x calc. y calc. log cy 1 1 O.60 2.81 0.608 2.80 0.2438 O.60 2.80 0-599 2.80 0.2453 12 0.50 3.26 0.510 3 - 3 i 0.2373 O.50 3-25 0.511 3 - 3 i 0.2360 13 0-35 4-55 0.320 4-59 0.2401 0-35 4-53 0.321 4-59 0.2382 H O.23 6.82 0.228 6.75 0.2487 0.22 6.82 7.00 0.2310 15 0.17 9-53 0.158 8.91 0.2738 0.16 9-53 9.42 0.2496 One of the greatest difficulties to be contended with in the experimental work was obtaining the point of equilibrium. It was difficult to detect the appearance of the second liquid layer as it did not manifest itself in the same manner over the whole range of concentration. Over one portion there was first a very slight opalescence which, upon further addition of water, in- creased until a decided cloudiness resulted, and finally the second liquid layer was very apparent. Over the other portion of the concentrations where the water was in excess of the benzene, the second liquid layer appeared as fine clear globules which floated on the surface, thus indicating that it was the benzene layer that was separating out. Owing to these two different appear- ances of the second liquid layer, it was somewhat difficult to determine the true point of equilibrium. It was no doubt this difficulty which presented itself to Waddell when he determined the equilibrium of the system ben- zene, acetic acid and water, for he states that he took as the end- point, i. e., as the point of equilibrium, the same degree of cloud- ing. That one cannot use the same degree of clouding as the end-point for the establishment of equilibrium is very apparent Benzene , Acetic Acid and Water 253 Table II. Temperature 35 0 C x = cc benzene, y = cc water, per 100 cc acetic acid Formula n log x -f- log y — log C = Mean = 0.810. n = 0.610 (1) x found (2) y found ( 3 ) y calc. log C 18. IO 1. 16 I. II 0.8322 18. IO 1. 12 — 1. 11 0.8169 2 16.09 1.22 1.26 1.23 0.8061 16.09 I. 21 1.26 1.23 0.8025 3 10.06 I.56 1.79 1.58 0.8051 IO.06 i *54 1. 81 1.58 o -7995 4 6.03 2.18 2.30 2. l6 0.8148 6.03 2.17 2.30 2. l6 0.8128 5 4.02 2.77 — 2.76 0.8113 4.02 2.78 — 2.76 0.8129 6 3.OI 3-27 — 3-30 b.8067 3.OI 3-25 — 3*30 0. 8040 Formula n' log x — log y — log C' = Mean = 0.842. n = 0.92 x calc. y calc. log C' 7 I. OO 7.01 1. 01 6.95 0.8457 I. OO 7.00 I. Ol 6-95 0.8451 8 0.65 10. 10 0.666 10.33 0.8322 0.66 10. 10 0.666 10. 19 0.8383 9 0.48 13.64 0.480 13-65 0.8416 0.47 13.64 0.480 13.92 0.8331 from the fact that in one portion of the series of concentrations there is a decided clouding, while at the other end of the series there is no clouding, but the separation of the second liquid layer as clear transparent globules. In that portion of the con- centrations where decided clouding does take place, the same de- 254 A. T. Lincoln gree of clouding does not represent the points of equilibrium. For example, in one experiment, No. 3 at 30°, it requires 1.56 cc of water to produce a decided opalescence and 1.79 cc to pro- duce a decided clouding, while in another experiment opales- cence was produced by the addition of 2.18 cc, while the same degree of clouding as in the preceding experiment was produced by 2.30 cc of water. The calculated value in the first case was 1.58 cc and in the second 2.16 ccof water. In Table II, column 3, are the values of the quantities of water that were required to produce the same degree of clouding in these various cases. By comparison with the corresponding values in column 2 it will be observed how much more was required than that just necessary to produce the decided opalescence which we took as the indication of the appearance of the second liquid layer, that is, as the point of equilibrium. From this I think we are justi- fied in concluding that Waddell’s results are wrong. He was not working with a system in equilibrium, but had an excess of one of the components, and for that reason we could not expect the results to conform to the law of Mass Action. If we apply the law of Mass Action to the data given in the tables above wherein we let x = cc of benzene, y — cc of water, and 2 = cc of acetic acid, then our equation takes the form x a y p — 2 a + p . Since the acetic acid was kept constant and we have x a y p = C 1 , and if we define a //3 = n , our equation then takes the form x n y — C, or expressed logarithmically, we have n log x + log y = log C. Now this is in the form of the equa- tion of a straight line wherein we have the logarithm of the quantities in place of the quantities themselves. Hence, if we plot the logarithm of the quantities of benzene and water used, the resulting curve should be a straight line with the slope n. From this curve the value of n can be determined, which in the case of the determinations at 25 0 C given in Table I are plotted on Fig. 1. It will be readily seen that we have two distinct curves and that one curve does not represent the condition of equilibrium over the whole range of concentration ; but confirms Bancroft’s statements that for two non-miscible liquids and a consolute liquid there are two sets of equilibria, and we have Benzene , Acetic Acid and Water 255 two curves corresponding to these two sets of equilibria. Fur- ther, the different end-points seem to correspond with these two sets, for over the greater part of one we get the second liquid layer appearing as fine transparent globules and over a consider- able part of the other as an opalescence. Having determined the values for n and n\ we then have the distinct curves of dif- ferent slopes. If now we substitute the value of n and n ' in their respective equations we obtain the values for log C and log C 1 as given in the last column. If the mean value be substi- tuted in our formula and we solve for the value of the amount of water that should have been added, we obtain the values given in column y calc , which agrees fairly well with those found experimentally and given in column 2. Under x calc are given the calculated values for benzene, assuming the value of y known. 93 iao 10.3 1 1.0 Fig. 1 If this physical reaction follows the Mass Law the expo- nential factors should be independent of the temperature, i. e., the values of n should be the same at whatever temperatures the equilibrium is established. Waddell states that at 35 0 C the deviation of this equilibrium from the Mass Law is even more 256 Benzene , Acetic Acid and Water pronounced than at 25 0 C. A series of determinations was made at 35 0 C and the results are given in Table II, and the plotted results are represented in Fig. 1. It will be observed that there are two curves corresponding to the two equilibria and further the value of n (0.61) at 35 0 is very nearly the same as n (0.6136) at 25 0 , while the values for n’ (0.92 against 0.9166) are almost exactly the .same at both temperatures. Hence it seems that we are justified in concluding that the temperature does not affect the value of the exponential factor and that this physical reaction between benzene, acetic acid and water con- forms to the Mass Law Equation. In this paper we have shown : 1. That Waddell was wrong in selecting the same degree of clouding as the point of equilibrium in the system benzene, acetic acid and water. 2. That equilibrium in the system benzene, acetic acid and water can be represented by the Mass Law Equation. 3. That for the range of temperature from 25 0 to 35 0 the exponential factor is constant, as in the case of chemical reac- tions. I wish to express my gratitude to Mr. J. V. Mapes, who did a part of the experimental work herein presented. University of Illinois , Feb., 1904.