The Hydroisomerization of Ethylcyclohexane-~~CL4 HERMAN PINES
ALFRED W. SHAW
Ipatieff High Pressure and Catalytic Laboratory, Department of Chemistry, Northwestern University, Evanston, Illinois
I. INTRODUCTION The hydroisomerization reaction is not only of industrial importance but is also of theoretical interest. The catalysts reported for this reaction consist of a hydrogenation component, such as nickel, platinum, etc., deposited on acidic supports, such as silica-alumina (1) or platinum on alumina containing halogen ( 2 ) .A detailed study of the hydroisomerization reaction as a function of catalyst composition and experimental conditions has been reported ( I ) . Ciapetta (Sa) studied the hydroisomerization of ethylcyclohexane over nickel on silica-alumina catalyst and reported that isomerization was the primary reaction and that the isomers consisted of dimethylcyclohexanes and of trimethylcyclopentanes. The dimethylcyclohexanes were stated to be composed of the 1,l- and 1,2-dimethylcyclohexanes, the latter predominating, and possibly of small amounts of 1,3and 1,4-dimethylcyclohexanes. Ciapetta (Sa) suggested that the formation of dimethylcyclohexanes proceeded through a direct isomerization of ethylcyclohexane by the rearrangement of the ethyl group. From the observation made that lower reaction temperatures were required for the hydroisomerization of alkenes as compared with the corresponding alkanes, Ciapetta (Sb) concluded that the initial and rate controlling steps in the isomerization reaction of saturated hydrocarbons was the dissociation of a carbon-hydrogen bond on the surface of the hydrogenation component. A systematic study of the mechanism of the hydroisomerization reaction has been undertaken in our laboratories in order to elucidate the various steps involved in such reactions. Ethylcyclohexane has been chosen as a model compound for the study of cycloalkanes, inasmuch as it may form a variety of isomers, of which the alkylcyclohexanes can be readily investigated. In order to determine the path through which such reactions may proceed, ethylcyclohexane having a labeled carbon atom in the a position of the side chain has been hydroisomerized. It was reasoned that if the hydroisomerization to dimethylcyclohexanes proceeds by rearrangement 569
HERMAN PINES AND ALFRED W. SHAW
of the side chain only, then the radioactivity should remain on the side chain. If the hydroisomerization reaction proceeds by a carbonium ion mechanism, as all the facts indicate, then it would be expected, in view of the previous work (4, b), that the reaction would proceed by ring contractions and expansions. In this case the radioactivity would be distributed in the ring and side chain. 11. METHODSAND PROCEDURES The ethylcyclohexane-ar-Cl4was synthesized by the acetylation of benzene with acetic a~id-1-C’~. The carbonyl labeled acetophenone produced . was reduced catalytically to ethylcyclohexane-a-C14. The catalyst used for the reaction was prepared according to the procedure of Ciapetta and Hunter ( 1 ) and consisted of 5 wt. % of nickel deposited on silica-alumina cracking catalyst. The reaction was performed in a flow system under 25 atm. of pressure, at 360°, hourly liquid space velocity of 1.0, and the molar-hydrogen-to-hydrocarbon ratio of 4: 1. The method used for the investigation of the hydroisomerate is shown in Fig. 1. The nongeminal alkylcyclohexanes were selectively dehydrogenated to the corresponding aromatic hydrocarbons under conditions where 1,1dimethylcyclohexane was not dehydrogenated. No attempt was made to determine the radioactivity distribution or the concentration of 1,I-dimethylcyclohexane, which according to the study reported by Ciapetta was undoubtedly present in the reaction mixture. In order to purify and degrade this compound, it would have required a level of radioactivity much higher than was used in these experiments. The validity of the procedure used was determined by the dehydrogenation and oxidation of the ethylcyclohexane-a-C14according to the outline in Fig. 1.Neither isomerization nor the presence of radioactivity in the ring was found. The composition of the alkylcyclohexanes produced were determined by means of the isotope dilution technique. The radioactivity determinations were made on “infinitely thick” barium carbonate plates (6),corrected for coincidence and background using a Tracerlab TGC-2, G.M. tube, and a Nuclear Instrument and Chemical Corporation Scaling Unit Model 162. The individual counting rates were known to within an accuracy of 3 %. 111. RESULTSAND DISCUSSION The composition of the alkylcyclohexanes in the hydroisomerate and the distribution of the radioactivity between the ring and side chain are summarized in Tables I and 11. The experimental results show that while the radioactivity is approaching statistical distribution, the relative concentration of the various alkylcyclohexanes are approaching their thermo-
Solid Barium Terephthalate
HCI .1 Crude Isophthalic Acid
Crude Terephthalic Acid en
LIQUID REACTION PRODUCT
I Sat’d. Hydrocarbons Selective Dehydrogenation
Aromatics a n 0 4
CUO Quinoline I
Crude Tereand Isophthalic Acid
Crude Beneoic and o-Phthalic Acid ’
Basicified Concentration HCI Filtration
J Solid Crude Benzoic Acid
Liquid Product Chromatographic Separation
O H: : :
CHC13 Soluble I Evaporate I
CUO Quinoline .L
J Solution Crude o-Phthalic Acid .L
J Ether Layer I Evaporate I& -
Residue SOClZ CC14 J J Solid Solution Phthalic Anhydride
1HC1 Schmidt Reaction
F]1 1 J
FIG. 1. Procedure used for the analysis of hydroisomerate. Radioactivity determinations were made on the enclosed compounds.
HERMAN P I N E S AND ALFRED W. SHAW
TABLE I The Radioactivity Distribution Between the Ring and Side Chain of the Alkylcyclohexanes" Expt. 1 Compound
Numbers are rounded off to the nearest whole per cent of the total radioactivity. Difference between recovered radioactivity and 100% recovery. c Determined by difference.
dynamic equilibrium values (Table 111). A small amount of the ethylcyclohexane recovered has also undergone a skeletal isomerization, since some of the radioactivity was found to be in the ring. Assuming therefore, that the recovered reacted ethylcyclohexane also underwent a statistical distribution of the isotope, it was possible to calculate the amount of ethylcyclohexane produced in the reaction. On this assumption and from the analytical data obtained, it was calculated that about 44% of the ethylcy~lohexane-a-C~~ underwent reaction. It was observed that only 85% of the hydroisomerate underwent dehydrogenation t o the corresponding aromatic hydrocarbons. The remaining
HYDROISOMERIZATION OF ETIIYLCYCLOHEXANE-~U-C'~
TABLE I1 The Composition of the Alkylcyclohexanes i n the Liquid Product"
Determined by the isotope dilution technique. Recovered unreacted ethyl~yclohexane-cu-C1~. c Recovered reacted ethylcyclohexane.
TABLE I11 Comparison of Experimental and the Theoretical Thermodynamic Equilibrium Concentrations of the Alkylcyclohexanes at 560' Compound
taken from the data of Kilpatrick et al. (7), and converted to a 1,l-dimethyl cyclohexane free basis.
HERMAN PINES AND ALFRED W. SHAW
15% was composed, according to the index of refraction, of alkylcyclopentanes and of 1,I-dimethylcyclohexane.If it is assumed that the latter is also present in amounts corresponding to its equilibrium concentration, namely, 7.1 %, relative to the other alkylcyclohexanesproduced, it can then be concluded that the ratio of alkylcyclohexanes to alkylcyclopentanes produced was also approaching an equilibrium composition (7). The experimental data indicate that the reaction proceeds through a carbonium ion mechanism involving repeated ring contractions and expansions. It is apparent that once the carbonium ion is produced, it remains on the catalyst long enough to permit scrambling of the carbon atoms as indicated by the almost complete iostope and chemical equilibration. A similar scrambling of the carbon atoms was observed when propanel-C13(8)and n-butane-l-Cl3 (9) were contacted in the presence of aluminum bromide promoted with water. In the study however, of the liquid-phase isomerization of 2-methylbutane-l-C14over an aluminum bromide catalyst (10) and in the case of the isomerization of t-butyl and t-amyl chloride ( l l ) ,such a deep seated scrambling of the carbon atoms was not found.
Received: March 2, 1956
REFERENCES 1 . Ciapetta,
F. G., and Hunter, J. B., Ind. Eng. Chem. 46, 147, 158 (1953).
2. Donaldson, G. R., Pasik, L. F., and Haensel, V., Ind. Eng. Chem. 47, 731 (1955). Ja. Ciapetta, F. G., Ind. Ens. Chem. 46, 159 (1953).
db. Ciapetta, F. G., Ind. Eng. Chem. 46, 162 (1953). 4. Pines, H., Olberg, R. C., and Ipatieff, V. N., J. Am. Chem. Soe. 74,4872 (1952). 6. Pines, H., and Myerholtz, R. W., J. Am. Chem. SOC.77,5372 (1955). 6. Calvin, M., Heidelberger, C., Reid, J. C., Tolbert, B . M., and Yankwich, P. E., “Isotopic Carbon,’’ p. 29. Wiley, New York, 1949. 7. Kilpatrick, J. E., Werner, H. G., Beckett, C. W., Pitzer, K . S., and Rossini, F. D., J. Research Natl. Bur. Standards 39, 523 (1947). 8. Beeck, O . , Otvos, J. W., Stevenson, D. P., and Wagner, C. D . , J . Chem. Phys. 16, 255 (1948). 9. Otvos, J. W., Stevenson, D. P., Wagner, C. D . , and Beeck, O., J. Chem. Phys. 16, 745 (1948). 10. Roberts, J. D. , and Coraor, G. R., J . Am. Chem. Soe. 74,3586 (1952). 11. Roberts, J. D., McMahon, R . E., and Hine, J. S., J. Am. Chem. SOC.7 2 , 4237 (1950).