March 15, 1943
ANALYTICAL EDITION
analyses may, so far as the absorption tubes are concerned, be interrupted without affecting the results. Thus it is not always necessary to have a continuous period of several hours in order to perform carbon and hydrogen analyses. This is of some importance to students whose time is likely to be interrupted. (While such interruptions do not affect the tube weights, some combustion tube fillings, especially those containing lead dioxide, must be conditioned immediately before running an analysis.) The relative constancy of the tubes over a period up to 20 hours indicates the definite absence of any trend in the weight of the tubes. It also indicates that the Pregl tubes may safely be left open to the air for several hours without protecting them with rubber caps. A great many analyses of research compounds prepared in this laboratory have been run with the absorption system described. The carbon values have agreed regularly within 0.3 per cent relative error and the hydrogen within 1 per cent relative error. However, the high hydrogen values usually associated with high temperatures and humidities (4) are still obtained with this system of tares, just as with the lead shot tares, and the proper correction must be determined using a known compound. The high hydrogen values are therefore probably not a direct result of high humidity. This system of tare requires the use of extra weights from the set up to 300 to 400 mg. to tare the continuous gain in weight of the absorption tubes. The use of so many weights as a tare has not produced a noticeable error in the results. The elimination of wiping apparently makes the use of Pyrex absorption tubes satisfactory. The omission of the
225
waiting period shortens the total time for tne analysis by a t least 10 minutes.
Summary The behavior of a carbon and hydrogen absorption system, in which a third Pregl-type tube is used both as a control and tare, is described. If the tubes are not wiped, they are constant in weight as soon as they can be weighed. Thus they may be weighed without the usual waiting period. The tubes remain fairly constant in weight up to periods of 1 hour and show only slight variations over much longer periods. Hence, a rigid weighing program is not required. The omission of wiping makes possible the use of Pregl tubes made of Pyrex. The use of a third tube as tare and control does not eliminate the high hydrogen values usually obtained at high atmospheric humidities with other absorption systems. The use of extra weights to tare the gain in weight of the absorption tubes does not introduce a noticeable error.
Literature Cited (1) Clark, R. O.,and Stillson, G. H., IND.ENG.CHEW,ANAL.E D . , 12. 494 (19401. \----. (2) Fri’edrich, A,, Mikrochemie, 19, 23 (1935). (3! Niederl and Niederl, “Micromethods of Quantitative Organic Analysis”, 2nd ed., New York, John Wiley & Sons, 1942. (4) Power, F.W., IND.ENG.CHEM.,ANAL.ED., 11, 660 (1939). (5) Pregl, F., “Die quantitativen organische Mikroanalyse”, 3rd ed., p. 48, Berlin, Julius Springer, 1930. (6) Royer, G. L., Norton, A. R., and Sundberg, 0. E . , IND. ENG. CHEM..ANAL.ED., 12, 689 (1940).
Microdetermination of Hydroxyl Content of Organic Compounds Acetic Anhydride-Pyridine Mixture as Reagent JACK W. PETERSEN, KENNETH W. HEDBERG, AND BERT E. CHRISTENSEN Oregon State College, Corvallis, Ore.
T
HE simplest procedures for the determination of the hydroxyl content of organic compounds are those based on esterification. As yet little attention has been given to their application on a micro scale, a field in which they would be especially useful. Several macro- and semimicromethods (1, 2, 4, 6 , 7 ) employing both acetic anhydride and acetyl chloride have been described in the literature. Extensive esterification experiments with acetyl chloride hare been reported recently from this laboratory (1). Attempts to use this reagent on a micro scale, however, have led to several other limitations besides those already mentioned (1). Attention was therefore directed to a study of the esterifications with acetic anhydride-pyridine mixture. Both Peterson and West (4) and Verley and Bolsing ( 7 ) have published methods based on the use of this reagent. Stodola (6) has reduced the procedure to a micro scale. Since extensive testing of this mixture has not been previously reported, the behavior of a large number of typical alcohols and phenols treated with acetic anhydride-pyridine solutions was studied. As a result of this work a simple microchemical technique based on the use of a hermetically sealed tube has been developed in this laboratory which gives fairly satis-
I
factory results for the microdetermination of the hydroxyl content of organic compounds.
Reagents c . P. acetic anhydride, redistilled and acetate-free, kept in wellstoppered (screw cap) bottle; c. P. pyridine, redistilled and water-free; and 0.04 N sodium hydroxide, carbonate-free.
Apparatus The reaction vessel consists of a melting point tube, 3 mm. in diameter and 6 cm. in length, made from a soft-glass test tube. Three medicine droppers, for the delivery of alcohol, acetic anhydride, and pyridine, respectively, are made by drawing one end of a 6-mm. soft-glass tubing to a fine capillary and equipping the other end with a rubber policeman. Glass plungers, 1.0 mm. X 0.5 cm., are made from soft-glass rod. A microcentrifuge. Analytical Procedure Introduce 2 t,o 10 mg. of the compound into a weighed reaction tube by means of the dropper, or, in the case of solids, employ the technique described by Niederl for filling Rast tubes ( 3 ) . Centrifuge and again reweigh the tube. Using the same technique, add a proximately 20 to 25 mg. (4to 5 drops) of pure acetic anhydride {om the second dropper, recentrifuge, and weigh
Vol. 15, No. 3
INDUSTRIAL AND ENGINEERING CHEMISTRY
226
TABLE I. ACETYLATION WITH ACETICANHYDRIDE-PYRIDINE MIXTURES No. of Determinations
OH Theory
lo
%
y-Phenyl-n-propyl Butyl Isoamyl Allyl Cyclohexanol Octanol-2 Hexanol Benzyl see-Butyl Cinnamyl Diethylcarbinol 2-Ethylbutanol
2 2 2 2 2 2 2 2 1 2 2 2
12.4 22.9 19.3 29.3 17.0 13.1 16.6 15.7 22.9 12.7 19.3 16.6
12.3 22.7 19.25 29.15 16.90 12.95 15.45 15.85 22.4 12.25 18.4 15.9
4 5 12
Polyhydric alcohols Ethylene glycol Propylene glycol Diethylene glycol Mannitol Sorbitol
4 6 4 2 2
54.6 44.7 32.1 56.1 56.1
54.0 41.0 30.1 56.15 54.7
20 29 4 1 0
Substituted alcohols Ethylene chlorohydrin 1,3-Dichloropropanol-2
4
2
21.1 13.1
20.0 12.8
3 0
Terpenes Borneol Menthol Geraniol Eugenol Isoeugenol Citronellol
2 2 2 2 4 4
11.0 11.0 11.0 10.4 10.4 10.9
11.1 10.9 9.9 10.55 9.75 9.45
0
5 7 28
Miscellaneous Benzoin Cholesterol4
*.
8.0 4.4
8.11 4.5
0 11
Primary and Secondary Alcohols
2'
OH Found (Average)
No. of
Average Deviation Parts p e r
Primary and Secondary Alcohols
Determmations
%
1000
0 5 8 2 6 4 15 10
..
9
0
OH
Theory
OH
Found (Average)
%
Average Deviation Parta per 1000
Phenols Phenol o-Cresol p-Cresol m-Cresol a-Naphthol
!E::%1
Hydroquinone Orcinol Catechol 2-Hydroxy-l,4-dimethylbenzene Pyrogallol Substituted phenols Vanillin p-Hydroxy-benzaldehyde Gallic acid Thvmol Guhiacol o-Chlorophenol p-Chlorophenol Methyl salicylate Salicylic acid m-Hydroxybenzoic p-Hydroxybenaoic 4-Hydroxy-1,2-dimethyl benzene m-Nitrophenol p-Nitrophenol o-Nitrophenol p-Aminophenol 2-Hydroxy-1 ,I-dimethylbenzene
2 3
13.9 40.5
13.8 40.0
8 17
2
11.2
10.85
4
2 2 2 2 2 2 2 2 2 2
13.9 27.1 11.0 13.7 13.2 13.2 11.2 12.3 12.3 12.2
13.65 27.3 11.15 13.65 13.1 13.0 11.35 12.35 12.2 12.25
4 4 4
2 2 2 2 2
13.9 12.2 12.2 12.2 15.6
14.15 12.1 11.35 11.45 15.05
4
2
13.9
13.8
7
4 2
39.8 45.5
39.9 45.55
5 3
4
0
0
4
4
9
4
9
4
2 1
Hydrolyzed in usual way, then 10 ml. of 95% ethanol added t o dissolve ester before titration. b Ran 48 hours.
Sugars Sucroseb Xylose
again. In order to ensure the quantitative conversion of the alcohol to the ester, a ratio of at least 2 moles of anhydride per equivalent of hydroxyl should be maintained. Add 4 to 6 drops of pure pyridine and again centrifuge. The amount of pyridine does not appear to be critical except in a few cases involvin solubility. Insert a small glass rod in the tube, seal, then shale well to ensure complete mixing, and set aside for 24 hours. At the same time run a blank to determine the volume of standard base required to neutralize the acid derived from 1m . of acetic anhydride. Place &e reaction tube in a 50-ml. Erlenmeyer flask, add 5 ml. of water, and then break the tube by means of a stout stirring rod. Titrate released acid with 0.04 N sodium hydroxide. The per cent hydroxyl can then be calculated by means of the formula:
poor results a t higher temperatures, possibly because of decomposition. The authors' experience indicates that the ease of acetylation varies with the individual compounds. The ratio of acetic anhydride to alcohol has been found to be somewhat critical and should not fall below 100 mole per cent excess per equivalent of hydroxyl. Since the hydroxyl is determined by difference, it is not a good policy to employ too large an excess for accurate work. Care should be exercised in the selection of acetic anhydride. It is essential that pure redistilled anhydride (from a good fractionating column) be used to obtain the best results. I n some of the initial work erratic results were traced to the presence of a considerable quantity of acetic acid in the acetic anhydride. As soon as this was remedied the results were nearer the theoretical values and were much more reproducible. I n later experiments the acetic anhydride was redistilled in a 7-plate column. The titer of this anhydride agreed perfectly with the theoretical value. The experiments with sugars were performed merely to ascertain the possibilities of a micromethod in this field. As indicated, the results are satisfactory and are in agreement with the data of Peterson and West (4).
'4
%(OH)
=
(m. e. of anhydride used - m. e. of acid found) x 1700 mg. of sample
where m. e. of acid found = ml. X normality, m. e. of anhydride used = mg. of anhydride x ratio X normality, and ratio = ml. of base required to neutralize acid derived from 1 mg. of anhydride.
Results and Discussion The results obtained with this procedure are given in Table I. I n most cases they are the average of duplicate determinations. This method is not applicable to the determination of tertiary alcohols. Ethyl citrate, t-amyl, and t-butyl alcohol gave very low results which confirm the observations of others. Although all experiments were extended over a period of 24 hours a t room temperature, these conditions are not to be considered as well-established optima. Since the reaction mixture is hermetically sealed, experiments can be conducted with equal ease a t elevated temperatures. Several compounds have been reported to acetylate in 15 minutes a t 100" C. in acetic anhydride and pyridine. In this laboratory borneol has been quantitatively acetylated in 35 minutes a t 100" C. On the other hand a number of alcohols give
Literature Cited (1) Christensen, B. E., Pennington, L., and Dimick, K. P., IND.
ENG.CHEM.,ANAL.ED.,13, 821 (1941). (2) Freed, M., and Wynne, A. M., Ibid., 10,456(1938). (3) Niederl and Niederl, "Organic Quantitative Microanalysis", p. 218, New York, John Wiley & Sons, 1942. (4) Peterson, V. L.,and West, E. S., J . Biol. Chem., 74, 379 (1927). (5) . . Smith, D. M.,and Bryant, W, M. D., J. Am. Chem. SOC.,57, 61 (1935). (6) Stodola, F. H., Mikrochemie, 21, 180 (1937). (7) Verley, A., and Bolsing, F., Ber., 34, 3354 (1901). PUBLISHED with the approval of the Monographs Publications Committee, Oregon State College, as Research Paper No. 73, School of Science, Department of Chemistry.
