CHEMISTRY
O. V. BRAGIN, A. L. LIBERMAN, G. K. GUR’YANOVA,
Submitted 1963-01-01 | RussiaRxiv: ru-196301.71968 | Translated from Russian

Abstract

Full Text

CHEMISTRY

O. V. BRAGIN, A. L. LIBERMAN, G. K. GUR’YANOVA,
Academician B. A. KAZANSKII

HYDROGENOLYSIS AND MUTUAL INTERCONVERSIONS

OF CIS- AND TRANS-1,2-DIMETHYLCYCLOPENTANES

IN THE PRESENCE OF RHODIUM, OSMIUM,

IRIDIUM, AND PALLADIUM CATALYSTS

Recently, we established (\(^{1}\)) that, in the presence of platinized carbon, the stereoisomeric 1,2-dimethylcyclopentanes are converted into one another over a wide temperature range, forming an equilibrium mixture of isomers:

\[ \begin{array}{c} \text{cis-1,2-dimethylcyclopentane} \;\rightleftharpoons\; \text{trans-1,2-dimethylcyclopentane} \end{array} \]

At the same time it was shown that, in both directions, these transformations proceed at a high rate, substantially exceeding the rate of hydrogenolysis of the cyclopentane ring.

Having established this fact for a platinum catalyst, we naturally turned to other metals of Group VIII of the periodic system. It seemed important to determine whether the above-mentioned configurational isomerization could occur on the surface of rhodium, osmium, iridium, and palladium catalysts. It was also of interest to compare—if isomerization takes place in the above cases—how rapidly this reaction proceeds on different metals and at what temperatures the mixtures obtained are at equilibrium. From the work of Kemball and co-workers (\(^{2}\)), devoted to the mechanism of deuterium exchange of cyclopentane homologs and published in 1962, it followed that palladium and rhodium, like platinum, promote the isomerization of stereoisomeric cyclopentanes. With regard to osmium and iridium, the question remained open. The second group of questions that interested us concerned the hydrogenolysis reaction of the five-membered ring in the presence of the catalysts listed above. It is known (\(^{3}\)) that, on the surface of platinized carbon, hydrogenolysis of cyclopentane and its homologs proceeds readily with the formation of paraffinic hydrocarbons; conversely, on palladized carbon this reaction does not occur at all (\(^{4}\)). There is no information in the literature on the possibility of hydrogenolysis on the surface of rhodium, iridium, and osmium.

On the basis of the foregoing, in the present work we investigated, under comparable conditions, the possibility and extent of configurational isomerization for stereoisomeric 1,2-dimethylcyclopentanes in the presence of Rh, Ir, Os, and Pd deposited on activated carbon, as well as the hydrogenolysis of cyclopentane hydrocarbons in the presence of the same catalysts.

The results of these investigations, summarized in Table 1, show that cis- and trans-1,2-dimethylcyclopentanes are mutually converted into one another in the presence of all four catalysts over a broad temperature range (150–280°). It should be noted, however, that their isomerizing activity differs (Fig. 1). Thus, Rh/C is almost as active in this reaction as the previously studied Pt/C (\(^{1}\)). Conversely, among the five catalysts discussed, Ir/C is the least active catalyst.

isomerization. In this same series, Os/C and Pd/C occupy an intermediate position. It should be especially noted that, on the catalysts studied in this work, the state of equilibrium is reached from the trans form of 1,2-dimethylcyclopentane substantially faster than from the cis form (see, for example, experiments Nos. 32 and 35 on Pd/C, Nos. 42 and 45 on Ir/C, Nos. 50 and 55 on Os/C, and Nos. 64 and 68 on Rh/C).

Table 1

Dependence of the composition of trans- and cis-1,2-dimethylcyclopentane catalyzates on the nature of the catalyst and the temperature of the experiments

Experiment No. Catalyst Temperature, °C $n_D^{20}$ of catalyzate trans-isomer, % cis-isomer, % Hydrogenolysis products $C_7$, % Hydrogenolysis products $C_6$, % Ratio (cis):(trans) in catalyzate Calculated equilibrium constant $K_p$
trans-1,2-dimethylcyclopentane
33 Pd/C 150 1.4125 97.5 2.5 no no 0.03 0.121
32 Pd/C 200 1.4135 89.0 11.0 » » 0.13 0.150
34 Pd/C 255 1.4140 81.1 18.9 » » 0.23 0.180
38 Pd/C 255 1.4137 83.6 16.4 » » 0.20 0.180
43 Ir/C 150 1.4129 96.1 3.9 » » 0.04 0.121
42 Ir/C 200 1.4130 89.7 8.2 2.1 » 0.09 0.150
44 Ir/C 255 1.4116 80.7 12.7 6.6 » 0.16 0.180
48 Ir/C 255 1.4117 78.9 12.9 8.2 » 0.16 0.180
49 Ir/C 280 1.4106 74.4 13.9 10.5 1.2 0.19 0.195
51 Os/C 150 1.4115 93.4 3.5 3.1 no 0.04 0.121
50 Os/C 200 1.4113 82.6 10.1 6.1 1.1 0.12 0.150
53 Os/C 255 1.4108 75.5 14.3 8.2 2.0 0.19 0.180
58 Os/C 255 1.4097 76.8 11.6 9.3 2.3 0.15 0.180
59 Os/C 290 1.4107 79.4 12.9 6.3 1.4 0.16 0.201
61 Rh/C 150 1.4120 100 no no no 0.121
68 Rh/C 150 1.4129 91.2 8.8 traces » 0.10 0.121
60 Rh/C 200 1.4059 61.6 10.8 27.5 » 0.18 0.150
62 Rh/C 255 1.4013 43.6 8.8 41.4 6.2 0.20 0.180
66 Rh/C 255 1.4008 40.2 8.5 43.3 8.0 0.21 0.180
67 Rh/C 280 1.4012 43.0 8.5 36.3 12.1 0.20 0.195
cis-1,2-dimethylcyclopentane
36 Pd/C 150 1.4206 16.9 83.1 no no 4.92 0.121
35 Pd/C 200 1.4161 63.1 36.9 » » 0.59 0.150
37 Pd/C 255 1.4138 80.5 19.5 » » 0.24 0.180
46 Ir/C 150 1.4211 14.6 85.4 » » 5.85 0.121
45 Ir/C 200 1.4180 36.6 61.6 1.8 » 1.68 0.150
47 Ir/C 255 1.4160 53.9 37.9 6.1 » 0.70 0.180
49 Ir/C 290 1.4117 63.2 24.9 10.3 1.0 0.39 0.201
55 Os/C 150 1.4178 21.9 72.8 4.9 0.4 3.32 0.121
54 Os/C 200 1.4130 46.5 40.9 10.7 1.9 0.88 0.150
56 Os/C 255 1.4121 58.9 31.5 7.2 2.4 0.53 0.180
57 Os/C 280 1.4133 58.9 34.0 5.2 1.9 0.58 0.195
64 Rh/C 150 1.4142 74.9 25.1 no no 0.34 0.121
63 Rh/C 200 1.4102 71.6 14.6 13.8 » 0.20 0.150
65 Rh/C 255 1.4018 42.2 9.4 40.0 7.4 0.22 0.180
69 Rh/C 280 1.4021 43.2 10.4 36.7 9.7 0.24 0.195

This is probably determined by the fact that the thermodynamic equilibrium of these two stereoisomeric forms in the temperature range studied is strongly shifted toward trans-1,2-dimethylcyclopentane.

The isomerization of the stereoisomers on each of the catalysts was carried out in a flow system in a stream of hydrogen. Simultaneously with isomerization, hydrogenolysis of 1,2-dimethylcyclopentane also occurred on the iridium, osmium, and rhodium catalysts, with formation of the corresponding paraffinic hydrocarbons of composition $C_7$ and comparatively small amounts of paraffins of lower molecular weight. On Pd/C, in agreement with the data of Kazanskii and Areshidze$^{(4)}$, hydrogenolysis of 1,2-dimethylcyclopentane does not occur. The rates of hydrogenolysis of this hydrocarbon on Ir/C, Os/C, Rh/C, and the previously studied Pt/C differ markedly. In the range 150–280°, the most active hydrogenolysis catalyst proved to be Rh/C. The Ir/C and Os/C catalysts are substantially less active in this reaction. Thus, at 255°, trans-1,2-dimethylcyclopentane underwent hydrogenolysis on Rh/C, Os/C, and Ir/C by 51, 11, and 8%, respectively (experiments Nos. 66, 58, and 48). It should be noted that in this reaction also each of the catalysts studied possesses a certain specificity. For example, on Rh/C the reaction proceeds with a high temperature coefficient (Fig. 2); moreover, on this catalyst (as also on Os/C) the secondary conversion of the formed isoheptanes into hydrocarbons of composition $C_6$ and even $C_5$ proceeds to an appreciable extent. The appearance of isohexanes in the catalyzate may be explained by the following scheme:

reaction scheme: conversion of dimethylcyclopentane through heptane isomers to methylcyclopentane and lower hydrocarbons, with formation of $n$-$C_7H_{16}$ and $i$-$C_6H_{14}$

All the hydrocarbons shown in this scheme were identified in the catalyzates by gas–liquid chromatography. Among the hydrocarbons of composition \( \mathrm{C}_6 \), 2,3-dimethylbutane has the highest content. It is interesting to note that on Os/C the yield of hydrogenolysis products passes through a maximum with increasing temperature* (see Fig. 2). The reason for this phenomenon is not yet clear, but in any case it is not connected with poisoning of the catalyst, which was checked by special experiments. On Ir/C the hydrogenolysis of 1,2-dimethylcyclopentane in the temperature range studied proceeds to approximately the same extent as on Os/C. However, cracking of the isoheptanes formed on Ir/C proceeds to a quite insignificant extent (experiments Nos. 48 and 49), which makes this catalyst, in its selective action, closer to Pt/C, on which we did not observe any appreciable cracking.

Figure 1

Fig. 1. Change in the ratio of the cis and trans isomers of 1,2-dimethylcyclopentane as a function of temperature on various Group VIII metals

It is interesting that in the catalyzates of 1,2-dimethylcyclopentane on all catalysts, including Pt/C, along with the hydrogenolysis products, methylcyclopentane was detected, the amount of which increased with the temperature of the experiments. Apparently, this fact indicates the possibility of some additional strain along the \( \mathrm{C}_{\text{ring}}—\mathrm{C}_{\text{alkyl}} \) bond arising upon adsorption of the initial molecule on the catalyst surface, as a result of which this bond undergoes hydrogenolysis. This seems all the more probable since the formation of cyclopentane during the hydrogenolysis of methylcyclopentane on Pt/C is practically not observed \((^3)\).

Figure 2

Fig. 2. Dependence of the yield of products of hydrogenolysis of trans-1,2-dimethylcyclopentane on temperature on various Group VIII metals

The results obtained, summarized in Table 1 and presented in Fig. 2, demonstrate a substantial difference in the behavior of the catalysts studied with respect to the hydrogenolysis reaction of the pentamethylene ring. Nevertheless, the ratio of the rates of the isomerization and hydrogenolysis reactions, previously found for Pt/C, is preserved here as well: the rate of isomerization in all the cases studied substantially exceeds the rate of hydrogenolysis.

The presence in the catalyzates obtained from 1,2-dimethylcyclopentanes on Rh/C and Os/C of relatively large amounts of isohexanes, as well as methylcyclopentane, shows that on these catalysts hydrogenolysis of any C—C bonds is readily accomplished, and not only in the five-membered ring, as on Pt/C. Consequently, the mechanism of the reaction under discussion on the first two catalysts must differ from the mechanism of hydrogenolysis of cyclopentanes on platinum, which is possible only according to the sextet–doublet scheme \((^5)\).

* A similar picture is observed on the surface of the same catalyst also in the hydrogenolysis of cyclopentane.

Obviously, hydrogenolysis of cyclopentanes on Os/C and Rh/C can also proceed according to the usual doublet scheme.

As for the mechanism of configurational isomerization, we still do not have sufficient data for any definite conclusions. However, it seems to us that this reaction can most probably proceed either by a dissociative mechanism through a “semi-dehydrogenated” state, or through a transition complex on the catalyst surface by an associative mechanism formally similar to the Walden \(S_\mathrm{N}2\) inversion. In accordance with the latter mechanism, the molecule of the dialkylcyclopentane undergoes edgewise adsorption and reacts with atomic hydrogen, also adsorbed on the surface, for example:

reaction scheme: three cyclopentane structures on a catalyst surface with arrows between them

The considerations presented above concerning the mechanisms of the reactions studied in this work deserve further discussion.

Experimental Part

Starting hydrocarbons. The synthesis and constants of the stereoisomeric 1,2-dimethylcyclopentanes used in the work were described by us earlier \((^1)\).

Catalysts. The Ir/C, Os/C, Rh/C, and Pd/C catalysts were prepared by the method of N. D. Zelinskii and M. B. Turova-Polyak \((^6)\), by reduction of the corresponding salts with formalin; they contained 20% metal by weight.

Experimental procedure. The experiments were carried out in a flow system over 5-ml portions of catalyst. The hydrocarbon was passed over the catalyst in a stream of hydrogen (hydrogen flow rate at the outlet of the system \(\sim 1\) liter/hour) at an approximately constant volumetric rate \((0.2\ \mathrm{hr}^{-1})\). The duration of each experiment was 1.5 hours; the catalyzate collected during the first hour was discarded in order to eliminate the influence of hydrogen adsorbed on the catalyst and of catalyzate residues from the preceding experiment; the catalyzate collected during the last 30 min was analyzed. The catalyzates were analyzed by gas–liquid chromatography on the SKB chromatograph of the Institute of Organic Chemistry, Academy of Sciences of the USSR, in a stream of helium at \(80^\circ\) on a four-meter column packed with diatomaceous brick coated with 20% PFMS-4 silicone oil. Detailed separation of the paraffin portion of the catalyzate was carried out on a capillary chromatograph with a flame-ionization detector, using a copper capillary 50 m long; dioctyl sebacate was used as the liquid phase. The results of all experiments are summarized in Table 1.

The authors express their gratitude to T. V. Vasina for preparing several catalyst samples.

Institute of Organic Chemistry named after N. D. Zelinskii
Academy of Sciences of the USSR

Received
12 VII 1963

References

  1. A. L. Liberman, O. V. Bragin, G. K. Gur’yanova, B. A. Kazanskii, DAN, 148, 591 (1963).
  2. F. G. Gault, J. J. Rooney, C. Kemball, J. Catalysis, 1, 255 (1962).
  3. B. A. Kazanskii, Usp. khim., 17, 641 (1948).
  4. B. A. Kazanskii, Kh. I. Arshidze, Soobshcheniya AN GruzSSR, 3, 885 (1948).
  5. A. L. Liberman, Kinetika i kataliz, 4, No. 6 (1963).
  6. N. D. Zelinskii, M. B. Turova-Polyak, in: Collected Works of Academician N. D. Zelinskii, Publishing House of the Academy of Sciences of the USSR, 1955, vol. 3, pp. 271, 375.

Submission history

CHEMISTRY