Conversion of 2, 3-Butylene Glycol to 1, 3-Butadiene by Pyrolysis of

Pilot-Plant Conversion of 2,3-Butylene Glycol Diacetate to 1,3-Butadiene. Industrial & Engineering Chemistry. Schniepp, Dunning, Geller, Morell, Lathr...
0 downloads 0 Views 1MB Size
8n

INDUSTRIAL A N D ENGINEERING CHEMISTRY

Septemhr, 1945

vacuum distillation of the unneutraliaed prodiwt. By operating at a sufficiently reduced pressure so that the diacetate boiled at a temperature lower than that at which the sulfuric acid esters decomposed, it was possible to remove practically all of the diacetate-acetic acid mixture by distillation. The residual mixture of diacetate and the sulfuric acid esters can be recycled as the esterification catalyst, since several experiments have shown that this material functions fully as well as fresh sulfuric acid. This method has the following advantages over the neutralization method: It requires no filtration equipment; there are no d i 5 solved solids to foul the distillation equipment; when the esterification is carried to a high degree of completion, the distillate contains practically no glycol monoacetate. Fractionation readily separates the acetic acid from the diacetate, or the mixture may be used directly for pyrolysis (4). Recycling of the catalyst. residue also serves to prevent the loss of incompletely acetylated glycol. The distillation method does, however, require a vacuum of at least 28 inches to accomplish the separation of diacetate from the sulfuric acid esters without loss due t o charring. This continuous esterification process was applied to the p r e p aration of ethylene glycol diacetate, methyl Cellosolve acetate, diethylene glycol diacetate, and glycerol triacetate. Excellent results were obtained in every case, and application of the method to the manufacture of a wide variety of industrially important acetates is indicated. A modification of the method has recently been applied to the manufacture of 2,3-butylene glycol diacetate in a column of commercial design by the staff of Joseph E. S~JJgrrun & Sons, Inc., who cooperated with this laboratory throughout the butylene glycol-butadiene investigation (9). It was found possible to reach and maintain the desired temperature

in the reaction zone by supplying heat only to the column calandria. When a sufficiently high concentration of diacetate waa produced in the reaction zone and when the temperature necee sary for rapid esterification was maintained, a glycol to &ace tate conversion of QWQ% resulted during a 2-hour residence time in the column. Figure 4 is a flow diagram of the proposed process for acetylation of 2,bbutylene glycol, developed cooperatively with the Seagram organization. ACKNOWLEDCM ENT

The authors wish to acknowledge the contributions of H. R. Hay, wlio conducted the laboratory and pilot-plant batch esterifications; F. W. Rustenbach, who constructed the glass reaotion column; and H. H. Geller, who conducted all of the analytical work connected with the continuous studies. The authors ah30 acknowledge the subRtantia1 contributions of the research and engineering staff of Joseph E. Seagram & Sons, Inc., particularly H. W. Grubb, who modified and adapted this process to largescale pilot-plant equipment of standard commercial design, and of the research st& of Merck & Company, Inc., who collaborated in the early development work. LITERATURE CITED

(1) Atwood, K., private communication. (2) Callaham, J. R., Chem.& Met. Efno., 51, 94 (1944). (3) Keyes, D. B., IND. ENQ.CREW, 24, 1096 (1932). (4) Morell, 8. A,, Geller, H. H., and Lsthrap, E. C., Zbid.. 37. 877 (1946).

(5) Othmer, D. F., Ibid., 27, 250 (1935). (6) Schniepp, L. E., Dunning, J. W., Geller, H. H., Morell, S.A.. and Lathrop, E. C., Ibid., 37,884 (1945).

CONVERSION of 2,3=BUTYLENE GLYCOL to 1,3=BUTADIENE by PYROLYSIS of DIACETATE s. A.

MORELL', H. H. OELLER, AND E.

c.LATHROP

NORTHERN REQIONAL RPsEARCH LABORATORY. U. S. DEPARTMENT OP AORICULTURE. PEORIA, ILL.

T h e conversion of 2,3-butylene glycol to butadiene by pyrolysis of Its diacetate has been studied over a wide range of temperatures and contact times. Yields of 82% of butadiene (purity, 99%) were obtained on one-pass pyrolyses In the temperature range 676' to 600' C. An additional 6% was secured by Isolation and pyrolysis of the Intermediates left in the pyrolysis liquors. Acetic acid recoveries of 99% were obtained under optimum oondltlons. Methyl vinyl carblnyl and brotyl acetates were Identifled as the intermediates of the reaction. The main by-products were methyl ethyl ketone, methyl ethyl ketone enol-acetate (2-acretoxy-2-butene), and methyl acetyl acetone.

NE of the important problem chosen for study at the inception of this laboratory in 1940 was the development of a commercial process for the production of 2,3-butylene glycol by fermentation of carbohydrate materials ($1). It was understood that when suEiciently high yields of 2,Sbutylene glycol were obtained to make the procese of commercial interest and

0

M n t address. Pabst Brewing Companjr. Milwaukee, WLS.

when the chemical was available, the development of a process for 1,bbutadiene would be undertaken. Work was actively started in January, 1942, at which time the importance of producing butadiene had become a major problem of national defense. The possibility of direct catalytic dehydration by the methodR in commercial use in Germany for converting 1,bbutylene glycol t6 1,bbutadiene (6)was given first consideration as the basis of a process. A study of approximately seventy dehydration catalysts under a variety of conditions indicated that this route was impractical. Highest one-paas yields of butadiene reached 20% of theory, the main product in every case being methyl ethyl ketone, which did not yield butadiene under any of the conditions tried. The predominate tendency to form ketones under dehydration conditions is characteristic of compounds having hydroxyl groups on adjacent carbon atoms: CHaCHOHCHOHCHs 2,3-Butylene glycol (I)

catalytic dehydration

CHaCH&OCH, Methyl ethyl ketone (11)

Van Pelt and Wibaut (14)studied the behavior of a number of acetates under pyrolysis conditions and found that the olefina or

INDUSTRIAL AND ENGINEERING CHEMISTRY

878

diolefins obtained were those normally expected on dehydration of the original alcohol or diol. Smith et al. (19) also showed that the pyrolysis of esters offers better control of this type of reaction. On the basis of evidence available, the pyrolysis of the esters of 2,3-butylene glycol was undertaken. The diacetate was chosen because of the commercial availability of acetic acid, acetic anhydride, and ketene: CHaCH-CH-CH?

I

t

A

CHaCOO OCOCHs -2CHaCOOH 2,3-€3utylene glycol diacetate (111)

CHS=CH--CH-CkI> 1,3-Rut,adiene(IV)

The production of butadiene by this method was first disclosed in a patent granted in 1938 to Hill and Isaacs (7'). I n general, their claims have been confirmed. The yields cited by Hill and Isaacs are based upon several recycles of the pyrolysis liquors; the present investigation has shown that this procedure did not lead to optimum results. The accumulation of a nonbutadieneyielding intermediate, methyl ethyl ketone enol-acetate, was not recognized by Hill and Isaacs, which probably led them to believe that the second stage of the butadiene reaction did not occur so readily as the first. Obviously their yield calculations, based on ester converted, took no account of this intermediate. By August, 19422, laboratory results were sufficiently promising to warrant pilot-plant investigations, which are described in another publication (16). Both laboratory and pilot-plant studies were conducted simultaneously, and the results led to a more complete understanding of the reactions. The pyrolysis of 2,a-butylene glycol diacetate to butadiene appears to be a noncatalyeed, homogeneous, gas-phase reaction involving two stages: (a) elimination of one molecule of acetic acid to form a mixture of unsaturated acetates (V, VI, and VII), and ( b ) loss of another molecule of acetic acid from V and VI1 t o form butadiene (IV). The over-all course of the reactions-Le. conversion to butadiene, composition and yield of unsaturated acetate intermediates, and formation of by-products-is determined essentially by temperature and contact time. The quality of butadiene is exceptionally high, 99+ yo pure under all conditions of operation. Optimum butadiene production is obtained under conditions which give a maximum. first-pass conversion; this is due to the fact that methyl ethyl ketone enol-acetate (VI), which is formed to a greater or lesser extent under all conditions, does not yield butadiene. The diacetate pyrolysis reaction proceeds essentially through the hlVC (methyl vinyl carbinol) acetate intermediate (V) ; crotyl acetate (VII) probably arises from V by an allylic rearrangement. Both of these intermediates may be pyrolyzed to butadiene in even higher yields than the diacetate: CH&H-CH=CHz

CHsC=CHCHa

dCOCHa ACqCH, Methyl ethyl ketone enolMethyl wnyl carbinol acetate (V) acetate (VI) CH,CH=CHCHzOCOCHa Crotyl acetate (VII)

* In view of the critical rubber shortage and the urgency for finding new source8 of butadiene, a fourth Industrial Conference [No. Regional Research Lab., Rep:. rra6-204 (1942)l waa convened at the Northern Regional Research Laboratory on August 3, 1942, to disclose the data obtained and d&cum the commercial possibilities of the reaction. As a result of this conferenae, a cooperative research program waB organised under the general sponsorship of the Office o! Rubber Director, War Production Board. The following agencies collaborated: Columbia Brewing Company (Doane Agricultural Service), Commercial Solvents Corporation, Heyden Chemical Corporation, Iowa State College, Lucidol Corporation, Merck and Company, Inc., National Research Council of Canada, Perrrtsylvania Sugar Company, Polytechnic Institute of Brooklyn, Schenley Research Institute, Inc., Joseph E. Seagram & Sons,Inc., and Univenity of Wisconsin. The present communication presents only results of work conducted at Northern Regional Research Laboratory.

VOl. 37, No. 9

The following brief description of the pyrolysis procedure shows the nature and extent of by-product formation, as well as acetic acid recovery: The diacetate was fed at a controlled rate through a heated reaction tube. The exit gmes entered a fractionating column where the butadiene and other gases were separated from the condensed liquid (called "pyrolysis liquor") which was continuously collected from the bottom of the column. The gas stream was scrubbed with water and dried, and the butadiene condensed in traps cooled by dry ice. The volume of gas not condensed in the traps (vent gas) was measured by a wet test meter; it comprised only a small fraction (