Quantitative Determination of Benzene In the Presence of Cyclohexane RICH4RD L. BISHOP
AND
E. LOUISE WALLACE, Eastman Kodak Company, Rochester, N. Y.
A""""
qNT psper (1) by Corson and Brady reported the development of a method for the determination of small amounts of benzene in cyclohexane by measuring the heat of reaction of benzene with a nitrating acid. The paper brought out two useful points: the composition of the nitratin8 acid producing the best results, and the variation in reaction heat due to changes in the initial or starting temperature of the reaction. By making a number of changes, both in the method and in the design of the reaction apparatus, it has been possible to develop a more rapid and practical procedure.
readings a minute or two apart. I n addition, an error was introduced when the room temperature changed between any two readings, because both the stirrer and the thermometer, when removed, changed to room temperature and were then replaced in the vacuum flask. Then, too, each time one or the other was removed some of the solvent was lost. It was discovered that the stirrer paddle, which consisted of a 7-mm. glass rod with 19 mm. of one end flattened to a width of 13 mm. and turnine a t 3000 r. D. m.. was not a verv efficient stirring device for d twaphase mixture, such as cyclohexane and nitrating acid. This Door mixinr tended to len&hen the time of campletereactianbetween the benzene and the nidtrating acid. Therefore, with a 10- to 20-minute period in which to measure the temperature rise, any change in the room temperature and subsequent change in the rate of rad& tion of heat from the vacuum flask introduced a considerable error. To overcome these difficulties, it was necessary to construct the apparatus shown in Figures 1 and 2, and t o adjust the amount of nitrating acid used, to ensure thorough mixing. The results which follow point out the distinct advantages of this redesigned mction heat apparatus. REAGENTS. Nitrating Acid, Mix 4 n . .., .r Y"l"lllt.J "I 69 t o 70 pt:r cent nitric acid (densit]i 1.41) with 1 volume of 95.5 per'cent sulfuric acid (dc:nsity 1.82). Cyclohexa ne, density at 200 c. 0.; '781, freezing point 6.0" C Benzene,dtensityat2OO C. 0.8790, frreeaingpoint5.5" C.
Apparatus and Reagents W-hen using the method described by Corson and Brady
( I ) , which required removal of the stirring device and insertion of a thermometer whenever a temperature reading was necessary, some difficulty was encountered in manipulation of the apparatus. This was especially noticeable in experimental work when it was necessary to take tenperature
General Procedure The pro1:edure found most satisfa ctory for routine work &!nd used, with slight modifications, in . 1 1 the experim'ems uescnwea :" Z"ll"..... in this papei,..,n.uLLuw,. I
nr
Weigh 50 grams of sample into a 120-ml. (4-ounce) tared bottle (small neck, plastic tap) and place this and the bottle of nitrating acid in a constant-temuerature bath a t 68' F. Pipet 50 ml. of nitrating acid in the vacuum flask and place the flask on the
FIGURE 1. EQUIPMENTFOR REACTION-HEAT TEST
563
ASSEMBLY OB REACTION-HEAT APPARATUS HIGURE Z.
X 10 inoh,
... ,. .
564
INDUSTRIAL AND ENGINEERING CHEMISTRY
Vol. 15, No. 9
lessened the precision of the results. On the other hand, any change which increased the mixing effect beyond that of the apparatus described in this paper, made almoqt no change in the results, except to decrease slightly the time of complete reaction. In fact, results m+erevery satisfactory when 100 nil. of nitrating acid and 50 grams of sample u-ere used in a larger vacuum flask (6.25 em., 2.5 inches in inside diameter) than the one described in this paper.
25
20
6 Y
EFFECT OF STRENGTH OF XITRATISG ACID.
ti
The effect of acid concentration \vas tested IO with acids of the composition shonn in Table 11. Figure 5 shows that No. 2 acid niixture is very satisfactory. It also has the advantage of being the one most easily pre5 pared from ordinary laboratory c. p. acids and has been chosen therefore for use in the general procedure. A slight variation 9 I I I I I I I I in the acid strength will not cause any ap0 .5 IO i5 20 25 30 3 5 40 4 5 5 preciable error in the results. Figure 6 ~llusP E R CENT BENZENE B Y WEIGHT trates the variation in AT tviththe time of HEATDERIVED FROM KXOWN hIIXTURES OF BEXZEKE FIQURE 3. REACTION i,eaction for a 2 per cent benzene in cycloAND CYCLOHEXASE hexane niixture when nitrat,ing acids of AT. Difference between initial temperature and temperature after 3 minutes of stirring reaction mixture differeiit strengths are used. A similar Nitrating acid, same as Xo. 2, Table I1 variation was found to hold true in 0.65. 3.0, and 4.0 per cent benzene-cyclohesane mixtures. EFFECTOF CERTAIN IMPURITIES I N THE SBMPLE. I t K R S TABLE I. REACTION-HE.4T TESTDAT.I (FIGURE 3) found necessary to determine benzene in cyclohexane which Weight contained a small amount of acetone and methyl alcohol. Percentage of Initial Room Benzene Temperature AT Temperature The reaction between these impurities and the nitratiiig acid F. ' F. F. caused a considerable rise in temperature. This difficulty 0.0 69.90 -0.60 74 was overcome by extracting bhe sample with water, and 0.5 71.01 2.20 77 drying with calcium chloride. Any appreciable amount of 1.5 , 71.25 7.80 76 2 . 0 7 0 . 3 2 10.80 76 n-ater in the sample would no doubt cause a rise in ternpew,2.5 70.72 13.50 3.0 70.75 16.40 76 ture. However, a mixture of cyclohexane and benzene 4.0 71.22 22.20 76 contains only 0.01 per cent of moisture when saturated :tt 5.0 71.10 27.60 76 room temperature. O
stirring device as shown in Figure 2. After 3 minutes of standing, start the motor, and after approximately 1 minute record the "initial" temperature which will be about 70" F. Stop the motor. Pour the sample from the 4ounce bottle into the vacuum flask and start the motor. Take readings after 1, 2, 3, and 5 minutes of stirring. The difference between the temperature a t 3 minutes and the initial temperature is a measure of the benzene conteut of the sample which is determined from Figure 3.
Discussion EFFECTOF STIRRINGPERIOD.The nitration is complete after 2 minutes of stirring, as shown in Figure 4.
Any slope to the curve after the 2-minute period is a result of radiation from the V B C U U ~flask, since the initial temperature of 70" F. is below the room temperature, which is 74" to 76" F. EFFECT OF IMPROPER NIXING.The most important point in this discussion is the efficiency in the mixing of the reacting substances. It was found that any change in the design of the apparatus, the quantity of reacting substances, or the speed of stirring, which tended to lessen the mixing effect, greatly increased the reaction time and
24
27
20 18 I6 LL
e
14
12 10 8 6 4
2 0
TIME
OF
REACTION IN MINUTES
FOR VARIOCS FIGURE 4. VARIATIONIN AT WITH TIMEOF REACTION MIXTURESOF BENZEKE IN CYCLOHEXANE
Initial temperature approximately 70' F. Nitrating acid, saAe as No. 2, Tablc I1 Stirring motor was stopped after firet 10-minute period
565
ANALYTICAL EDITION
September 15. 1943
,2-
IO -
/
/
9-
8-
7-
w %-
6-
5-
4-
3-
2-
I I
I
I
I
2 3 PER CENT BENZENE ET' W E I G H T
1-
4
DATA FOR KNOWK MIXTURES FIGURE 5. REACTION-HEAT OF BENZENE IN CYCLOHEXANE USINGACIDSFROM TABLE I1 Difference between initial temperature and temperature after 3 minutes of stirring reaction mixture
AT.
Results The accuracy of the results obtained by the described method is indicated in Table 111. These results were obtained on knowns made from mixtures of cyclohexane from
I
I
2
I
TIME
I
3
I
4
OF RCACTION IN MINUTES
IS AT WITH TIMEOF REACTION FOR FIGURE 6. VARIATION FOURSTREXGTHS OF NITRATING ACID
various sources, and benzene. Different batches of nitrating acid Rere used in the course of the experimentation. In each case the acids were made up to approximately the same strength in accordance with the method outlined under reagents.
Summary OF NITRATING ACIDS TABLE11. COMPOSITION (FIGURES 5 ASD 6)
Acid
KO.
1
. '
;
4
HnSOa
"Os %
" 0 2
%
%
42.94 41.32 40.58 39.77
0.16 0.01 0.01 0.01
38.35 38.06 37.28 36.54
Hi0
% 18.5.5 20.61 22.13 23.68
TABLE111. ANALYSIS OF KNOWNS IT 3-Minute minus initial Temperature O F. 7.82 7.75 10.63 10.87 10.48 10.40 10.82 10.98 10.79 16.30 16.35 21.78 22.08
Benzene Preaent --Weight
Benzene Found percentage
-
Did erence 0.00 -0.03 0.00
70.04 -0.03 -0.05 t0.02 -0.05 10.02 0.00 0.00 -0.0c5 70.02
The method described in this paper was found to be very Lvorkable and rapid, giving a high degree of precision and accuracy. A similar method has been developed t o determine benzene in ethylene dichloride and in propylene dichloride, following the same general procedure and using the same strength of acid. Horrever, it was necessary to draw a separate curve of AT against per cent benzene for each solvent, as the heat of the nitration reaction in the presence of different types of solvents was not the same.
Acknowledgment The authors are indebted to F. J. Hopkinson of the Tndustrial Laboratory for many helpful suggestions and the equipment used in this work, and t o S. C. Pool and the staif of the Chemical Plant Laboratory of the Eastman Kodak Company for the analyses of the nitrating acids.
Literature Cited (1) Corson, B. B.,and Brady, L. J., IND.ESG.C'HEV., ANLL.ED., 14. 531-3 (1942).
