Chemistry

A. L. LIBERMAN, T. V. VASINA, and Academician B. A. KAZANSKII

ON THE SPATIAL CONFIGURATION OF STEREOISOMERIC 1,4-DIISOPROPYLCYCLOHEXANES

In 1956 we showed \((^{1})\) that, among the stereoisomers of 1,4-diisopropylcyclohexane, the lower-boiling isomer has a higher refractive index and specific gravity. Meanwhile, for all other stereoisomeric 1,4-dialkylcyclanes known at that time, higher values of these constants characterized, on the contrary, the higher-boiling isomers. A preliminary \((^{1})\), and subsequently a more detailed \((^{2})\), study of the Raman spectra of stereoisomeric 1,4-diisopropylcyclohexanes and comparison of them with the spectra of other cis–trans-1,4-dialkylcyclohexanes \((^{3})\) (including 1,4-dimethylcyclohexane, for which the spatial configuration of the stereoisomers has been proved with sufficient reliability \((^{4})\)) showed that in our case the lower-boiling isomer is the cis form. Thus, we encountered a peculiar deviation from the Auwers–Skita rule, which states that the cis forms of disubstituted cyclanes should boil higher than the trans forms. As is known, a certain reconsideration of this rule in recent years has concerned only series of 1,3-dialkylcyclopentanes \((^{5})\) and 1,3-dialkylcyclohexanes \((^{4})\), and no violations of the rule in the series of 1,4-disubstituted cyclohexanes had been observed before our work. Only in 1958 did Eliel and Haber \((^{6})\) observe a violation of the rule exactly like that for 1,4-diisopropylcyclohexanes, but using oxygen-containing derivatives as examples: 2-, 3-, and 4-methylcyclohexanols.

In view of the fundamental importance of such facts of violation of the Auwers–Skita rule, we considered it important to confirm it in the case of 1,4-diisopropylcyclohexanes, for greater reliability, by a synthetic route, in addition to the spectroscopic evidence mentioned above. The present work is devoted to solving this problem.

As starting substances we took the dimethyl esters of cis- and trans-hexahydroterephthalic acids, the spatial configuration of which, as well as that of the acids themselves, is beyond doubt \((^{7,8})\). The intention was to convert each of the stereoisomers separately into the corresponding form of 1,4-diisopropylcyclohexane according to the following scheme:

\[ \begin{array}{ccccc} \begin{array}{c} \mathrm{COOCH_3}\\ |\\ \hexagon\\ |\\ \mathrm{COOCH_3} \end{array} & \xrightarrow{\;4\mathrm{CH_3MgCl}\;} & \begin{array}{c} \mathrm{OH}\\ |\\ \mathrm{H_3C{-}C{-}CH_3}\\ |\\ \hexagon\\ |\\ \mathrm{H_3C{-}C{-}CH_3}\\ |\\ \mathrm{OH} \end{array} & \xrightarrow{\;2\mathrm{HCl}\;} & \begin{array}{c} \mathrm{Cl}\\ |\\ \mathrm{H_3C{-}C{-}CH_3}\\ |\\ \hexagon\\ |\\ \mathrm{H_3C{-}C{-}CH_3}\\ |\\ \mathrm{Cl} \end{array} & \xrightarrow{\;4\mathrm{H}\;} & \begin{array}{c} \mathrm{H}\\ |\\ \mathrm{H_3C{-}C{-}CH_3}\\ |\\ \hexagon\\ |\\ \mathrm{H_3C{-}C{-}CH_3}\\ |\\ \mathrm{H} \end{array} \end{array} \]

If the reactions proceeded stereospecifically, comparison of the constants of the isomeric hydrocarbons obtained by this scheme and those previously isolated by precise ...

rectification (1), would have given a definitive answer concerning their spatial configuration. However, it has so far been possible to carry out this synthesis only for the trans form. In the case of the cis form, all stages of the synthesis proceeded nonstereospecifically, and the corresponding cis dichloride could not even be isolated in individual form.

A detailed description of the syntheses and properties of the stereoisomeric diols, the trans dichloride, and certain other substances obtained for the first time will be given later; here only brief information is presented. The starting dimethyl esters of cis- and trans-cyclohexane-1,4-dicarboxylic acids were obtained by hydrogenation of dimethyl terephthalate, separation by freezing out, and then careful purification of the trans form (by repeated recrystallization, m.p. 70.0–70.5°) and of the cis form—by crystallization from ether at low temperature and distillation in vacuum on an efficient column (b.p. 126.9–127.2°/7 mm; freezing point 9.6°; \(n_D^{20}\) 1.4590; \(d_4^{20}\) 1.1111).

It is curious to note that in such distillation of a mixture of the dimethyl esters of stereoisomeric hexahydroterephthalic acids, the pure ester of the cis acid passes over in the first fractions, whereas the trans ester crystallizes out from the last fractions and is also present in considerable amount in the residue. Thus, in this case the trans form, contrary to the Auwers–Skita rule, boils higher than the cis form, just as was observed in the case of 1,4-diisopropylcyclohexane ().

1,4-Bis-(\(\alpha\)-hydroxyisopropyl)-cyclohexanes were obtained by the interaction of the esters of the stereoisomeric acids with a large excess of methylmagnesium chloride or bromide (cis form: m.p. 100.5–100.7°; trans form: m.p. 159.3–159.8°). The corresponding dichlorides were prepared by saturating solutions of the diols in methanol with dry hydrogen chloride. The trans isomer was obtained in individual form (m.p. 134.3–135.4°); however, from the cis diol a mixture of cis- and trans-dichlorides was formed which could not be separated.

Replacement of chlorine by hydrogen proved to be a very difficult task. Thus, at room temperature the chlorine in trans-1,4-bis-(\(\alpha\)-chloroisopropyl)-cyclohexane is not hydrogenated by hydrogen either at atmospheric or at elevated (150 atm.) pressure in the presence of Raney nickel, platinum, and palladium, including palladium deposited on calcium carbonate, which, according to the literature data (9), causes elimination of chlorine even from the benzene nucleus. When the temperature is raised to 50–70° in the presence of these catalysts, chlorine is not eliminated, but is split off as HCl with formation of diolefins, which are then hydrogenated, giving a mixture of stereoisomers (\(n_D^{20}\) 1.4511). With magnesium, the dichloride does not react even in the presence of a large excess of ethyl bromide. Thus, the usual methods for replacing halogen by hydrogen proved in our case to be unsuitable.

In 1949 G. P. Men’shikov, in studying the structure of the alkaloid heliotrine, successfully eliminated an atom of chlorine introduced by him by the action of an aqueous solution of chromous chloride (10). This procedure would have been inapplicable in our case because of the complete insolubility of the dichloride under study in water; however, we succeeded in finding another suitable solvent—ethyl acetate. In it, not only the dichloride and \(CrCl_2\) dissolve sufficiently well, but also \(CrCl_3\), from which we prepared \(CrCl_2\) directly in ethyl acetate solution. In this solvent we were able to carry out the desired reaction at room temperature. The constants of the trans-1,4-diisopropylcyclohexane obtained in this way are given in Table 1. Comparison of them with the constants of this hydrocarbon from our previous work (1) shows that, according to the combination-scattering spectra, the configurations of the stereoisomers had been determined correctly.

The fact that the synthetic preparation differs somewhat in its constants from that obtained by distillation (1) and having a high degree of purity is explained by the presence in the former of a small impurity of the cis isomer.

(7–10%). Since the sharp melting points of the diol and the dichloride suggest that they could not have contained such a large amount of the other form, it should be assumed that the stereoisomeric impurity was formed, probably, in the course of chlorine elimination. In all likelihood, during this reaction, under conditions of local overheating, partial elimination of hydrogen chloride occurs. Then hydrogen chloride, the concentration of which under the reaction conditions is rather high, can again add to the double bond that has formed, with the concomitant formation of cis- and trans-forms. As a result of further elimination of chlorine, one of these forms gives a stereoisomeric impurity, for example:

Table 1

Comparison of constants of stereoisomeric 1,4-diisopropylcyclohexanes

cis-form                trans-form

Nevertheless, this side reaction evidently proceeds only to a slight extent, and the elimination of chlorine leads predominantly to the trans-isomer. Thus, the conclusion we made earlier—that the lower-boiling 1,4-diisopropylcyclohexane has the cis structure, and the higher-boiling one the trans structure—has received additional confirmation.

Experimental Part

Solution of CrCl₃ in ethyl acetate. Into a porcelain dish were placed 30 g of recrystallized chromic chloride and 180 ml of concentrated hydrochloric acid, and the mixture was heated on a water bath until the evolution of chlorine ceased; hydrochloric acid was added as it was consumed. The reaction mixture was then evaporated until a viscous sticky mass was obtained; after cooling, this mass was transferred to a flask with 350 ml of dry ethyl acetate, os-

freed from ethyl alcohol. Chromic chloride dissolved* with the formation of a dark-brown solution. As required, this solution was filtered from insoluble KCl and saturated with dry hydrogen chloride; in this process the color of the solution changed to dark crimson.

Reduction of trans-1,4-bis-(α-chloroisopropyl)cyclohexane. Into a round-bottom four-necked flask of 1 l capacity, fitted with a mercury seal, stirrer, dropping funnel, and also inlet and outlet tubes for hydrogen, closed with Tishchenko bottles containing sulfuric acid, 63 g of zinc dust was placed. The air was then displaced from the flask by hydrogen, and from the dropping funnel 165 ml of a solution of CrCl₃ in ethyl acetate, saturated with dry HCl, was cautiously added. When all the solution had been added and the reaction had ended (the color of the solution became pure blue), the reaction mixture was cooled to room temperature and a solution of 5.0 g of trans-1,4-bis-(α-chloroisopropyl)cyclohexane in 200 ml of ethyl acetate was added to it. Stirring was carried out continuously for 75 h, during which time 350 ml of a saturated solution of dry HCl in ethyl acetate was gradually added in portions of 10–20 ml in 18 additions; during this time the zinc dissolved almost completely. The reaction mixture was then poured into a threefold volume of water, and from a Favorskii flask the entire distillate passing over up to 100° was distilled off. The distillate was salted out with potash. Ethyl acetate was distilled off from the dried upper layer from a small Favorskii flask, into which the solution was continuously fed during the distillation from a dropping funnel. The residue (4.8 g) was combined with a similar residue (3.1 g) from a parallel experiment with 4 g of dichloride. The combined residue was treated with two 1 ml portions of conc. sulfuric acid (3 ml of substance was absorbed), washed with water, dried, and chromatographed on silica gel. A fraction (2.45 g) was thereby obtained, the properties of which are given in Table 1. The next fraction already gave a positive reaction for unsaturated compounds and halogen.

In the combination-scattering spectrum of the diisopropylcyclohexane fraction obtained in this way,** all the lines characteristic of the higher-boiling isomer, previously recognized as the trans form, were present; however, according to these spectral data, its fraction also contained 7–10% of the cis isomer. The content of the cis-form impurity, on the basis of the freezing temperature and the previously found (¹) cryoscopic constant of the trans form (0.043 mole fraction per degree), was 9%; assuming additivity of the refractive indices and specific volumes, the content of this impurity was 8% by both constants. The yield of the fraction was 69% of theoretical, calculated on the dichloride.

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

Received
25 I 1960

CITED LITERATURE

1. A. L. Liberman, T. V. Lapshina, B. A. Kazanskii, DAN, 107, 93 (1956).
2. V. T. Aleksanyan, Kh. E. Sterin et al., in: Studies in Experimental and Theoretical Physics. In Memory of Grigorii Samuilovich Landsberg, Publishing House of the Academy of Sciences of the USSR, 1959, p. 43.
3. P. A. Bazhulin, A. I. Sokolovskaya et al., Izv. AN SSSR, OKhN, 1956, 1130.
4. G. A. Haggis, L. N. Owen, J. Chem. Soc., 1953, 408.
5. S. F. Birch, R. A. Dean, J. Chem. Soc., 1953, 2457.
6. E. L. Eliel, R. G. Haber, J. Org. Chem., 23, 2041 (1958).
7. W. H. Mills, G. H. Keats, J. Chem. Soc., 1935, 1373.
8. R. Malachowski, J. Jankiewiczówna, Ber., 67, 1783 (1934).
9. M. Busch, H. Stowe, Ber., 49, 1063 (1916).
10. G. P. Menshikov, ZhOKh, 19, 1702 (1949).

* If evaporation was carried out until a solid dry residue was obtained, dissolution proceeded poorly and required several days.
** The spectral analysis was carried out by V. T. Aleksanyan and Kh. E. Sterin (Commission on Spectroscopy, Department of Physicomathematical Sciences, Academy of Sciences of the USSR), to whom we take this opportunity to express our gratitude.