On the Structural Causes of Optical Activity
Abstract Generated abstract
This paper reexamines proposed structural sources of optical activity in organic compounds from the standpoint of the author’s mesostructure theory. It argues that several cases usually attributed to molecular asymmetry, including allenes, alicyclic allene analogues, spirans, unusual acids, and compounds of nitrogen, phosphorus, arsenic, and sulfur, can instead be interpreted through ordinary asymmetric atoms produced or clarified by mesovalent interactions, while optical activity in allene derivatives is deemed impossible if double bonds are not rigid. The paper concludes that most alleged categories of molecular asymmetry should be reduced to asymmetric atomic centers, leaving complex inorganic compounds, and uncertainly some diphenyl derivatives, as the principal remaining cases of molecular asymmetry.
Full Text
Prof. I. I. Kotyukov.
On the Structural Causes of Optical Activity.
(Theory of the Mesostructure of Organic Compounds III).
At the present time there exist two structural causes for the occurrence of optical activity—the presence of an asymmetric atom in the molecule and the asymmetry of the molecule as a whole. By an asymmetric atom we mean an atom all of whose valence units are substituted by radicals differing in composition, structure, and configuration. Such atoms up to now have been C, N, P, S, Se, Si, As, and Sn. An asymmetric molecule is one which, when represented by a tetrahedral model, cannot be brought into coincidence with its mirror image by any rotation. Molecular asymmetry is at present accepted for the following substances.
1) The optically active Werner complexes and those of his school. 2) Spiranes. 3) Allenes. 4) Inositol. 5) Perkin’s alicyclic preparations and their analogues and, finally, 6) diphenyl derivatives.
As regards allene derivatives, the possibility of their optical activity was predicted by Van ’t Hoff as early as 1875. If the double bond in organic compounds is regarded as rigid, then in substances of the general formula
\[ \begin{array}{ccccc} R_1 & & & & R_3 \\ & \backslash & & / & \\ & & C = C = C & & \\ & / & & \backslash & \\ R_2 & & & & R_4 \end{array} \]
the presence of mirror isomers is possible owing to the different relative disposition of the groups \(R_1, R_2, R_3\), and \(R_4\) in the event of their fixation in space by the rigid double bond. However, all attempts to prepare optically active allenes have hitherto met with unfailing failure. Thus, despite carefully arranged experiments on the resolution of diphenylnaphthyl-allene-carboxylic acid, in which almost all the usual methods were tried, Lapworth1 nevertheless did not obtain antipodes.
From the point of view of the theory of mesostructure, such work is doomed in advance to failure, for if the rigidity of the double bond is denied, then free rotation takes place around it, as in malic acid, and therefore the radicals cannot fix their positions in space. Optical isomerism in allene derivatives is impossible.
The failures in resolving allene derivatives prompted Perkin, following Markwald’s idea, to prepare and resolve 1-methyl-cyclohexylidene-acetic acid,
\[ \begin{array}{ccccc} & CH_3 & & CH_2 & \!-\! CH_2 \\ & \backslash & / & & \backslash \\ & & C & & C : CH \cdot COOH \\ & / & \backslash & & / \\ H & & CH_2 & \!-\! CH_2 \end{array} \]
The above-named investigators hoped, by introducing a ring into the allene molecule in place of one double bond, to obtain, so to speak,
its analogue and to verify van ’t Hoff’s prediction. Their expectations in this case were crowned with complete success—the preparation showed optical activity1. Since there was no asymmetric carbon atom here, a new term was proposed for this case—“centrosymmetry.” Subsequently, a series of analogues was prepared for this substance, likewise obtained in optically active form.
However, the theory of mesostructure explains the optical activity of these substances not at all by analogy with van ’t Hoff’s allene scheme. The double bond here is not rigid; rotation can occur about it, and, were it not for special conditions, there would be no optical activity here. These special conditions here are the presence of carboxyl oxygen with a residue and of unsaturated hydrogen atoms in the ring. That the latter indeed possess residual valences is evident, among other things, from the isomerism of hexahydrobenzophenone oxime2.
A mesobond arises between these atoms, as a result of which two ordinary asymmetric carbon atoms appear in the molecule.
\[ \begin{array}{ccccccccc} & \mathrm{CH_3} & & \mathrm{CH_2} & - & \mathrm{CH_2} & & \mathrm{H} & \\ & \diagdown & \diagup & & & \diagdown & & \diagup & \\ \mathrm{H} & - & \mathrm{C} & & & \mathrm{C} & : & \mathrm{C} & \diagdown \\ & \diagup & \diagdown & & & \diagup & & & \mathrm{C{-}OH} \\ & & \mathrm{CH_2} & - & \mathrm{CH_2} & & & \diagup\!\!\diagup & \\ & & & & & & \mathrm{H}\cdots\mathrm{O} & & \end{array} \]
and this substance receives the following meso-formula:
\[ \begin{array}{ccccccc} & & \mathrm{CH_2} & & \mathrm{CH} & & \mathrm{OH} \\ & \diagup & & \diagdown & & \diagdown & \diagup \\ \mathrm{CH_2} & & & & \mathrm{C} & & \mathrm{C} \\ | & & & & | & & \Vert \\ \mathrm{CH_3\!\cdot\!CH} & & & & \mathrm{CH} & & \mathrm{O} \\ & \diagdown & & \diagup & & \diagdown\!\!\cdots & \vdots \\ & & \mathrm{CH_2} & & & \mathrm{H} & \end{array} \]
But if we account for the optical activity of such substances not by a specific asymmetry, but by an ordinary asymmetric atom, then it is natural to apply this method also to analogous spirans. From this point of view, Radulescu’s symmetrical spirans3 have two identical ordinary asymmetric carbon atoms.
\[ \begin{array}{ccccccccc} & & & \mathrm{H}\cdots\cdots\cdots\mathrm{O} & & & & & \\ & & \diagup & & \Vert & & & & \\ \mathrm{HO_3S} & \lozenge & {-} & \mathrm{CH} & & \mathrm{C.NH} & {-} & \lozenge & \mathrm{SO_3H} \\ & & & \diagdown & & & \diagdown & & \\ & & \mathrm{NH.C} & {-} & \mathrm{C} & {-} & \mathrm{CH} & & \\ & & \Vert & & & & \diagup & & \\ & & \mathrm{O} & \cdots & \mathrm{H} & \cdots & & & \end{array} \]
An analogous structure is also found in the corresponding amine.
Using the same method, one can explain a number of curiosities from the field of optical activity and bring them under the usual interpretation of this question.
Thus, A. Werner points to the existence of a third isomer of malic acid, obtained from Crassulaceen1. If one excludes the possibility of the manifestation of mesovalency in the hydroxyl oxygen, then even here two meso-formulas may be possible:
\[ \ce{HOOC . CHOH . CH2 . COOH} \qquad\qquad \ce{HOOC . CHOH . CH2 . COOH} \]
\[ \begin{array}{c} \ce{OH}\\[-2pt] \ce{OH\quad C--CH \qquad H}\\[-2pt] \ce{\ \ \ \Vert \qquad \backslash \quad / \quad \cdots \ O}\\[-2pt] \ce{\ \ \ O \cdots H \qquad C \qquad C}\\[-2pt] \ce{\qquad\qquad\quad \backslash \quad / \quad \Vert}\\[-2pt] \ce{\qquad\qquad\qquad C \qquad O}\\[-2pt] \ce{\qquad\qquad\qquad |}\\[-2pt] \ce{\qquad\qquad\qquad OH} \end{array} \qquad\qquad \begin{array}{c} \ce{OH}\\[-2pt] \ce{\qquad\qquad C}\\[-2pt] \ce{H \qquad \quad \backslash \quad / \quad \Vert}\\[-2pt] \ce{\cdots O \qquad CH \qquad C \qquad O}\\[-2pt] \ce{\ \ \ \Vert \qquad | \qquad \backslash \quad \cdots}\\[-2pt] \ce{C \qquad C \qquad H}\\[-2pt] \ce{| \qquad |}\\[-2pt] \ce{OH \qquad OH} \end{array} \]
Each of these substances has, strictly speaking, two asymmetric carbon atoms, and therefore broad mesoisomerism is possible here.
In 1901, Noyes2 accidentally obtained an optically active acid in which there was no asymmetric carbon atom. He then doubted the structure and carried out a repeated, careful investigation of the structure. The latter was confirmed. Since there was nothing with which to explain the activity, this case was forgotten and was found by us while compiling an index of optically active substances.
As is evident from the meso-formula placed below, Noyes’s acid has an ordinary asymmetric carbon atom:
\[ \begin{array}{c} \ce{CH3 . C == C . COOH}\\[-2pt] \ce{\quad | \qquad | \quad \cdots}\\[-2pt] \ce{(CH3)2C \qquad CH2}\\[-2pt] \ce{\quad \backslash \qquad /}\\[-2pt] \ce{\qquad CH2} \end{array} \qquad\qquad \begin{array}{c} \ce{CH3 \qquad OH}\\[-2pt] \ce{\cdot C \qquad C}\\[-2pt] \ce{\quad \backslash \quad / \quad \Vert}\\[-2pt] \ce{(CH3)2 C \qquad C \qquad O}\\[-2pt] \ce{\quad \backslash \qquad | \qquad \cdots}\\[-2pt] \ce{\qquad CH2 \qquad CH \cdots H} \end{array} \]
Nor is there anything special in the much-discussed preparations of Meisenheimer, which possess optical activity with identical “radicals”3. The latter are two identical bonds with oxygen.
\[ \begin{array}{c} \ce{R1}\\[-2pt] \ce{\quad \backslash}\\[-2pt] \ce{R2 -> N=O}\\[-2pt] \ce{\quad /}\\[-2pt] \ce{R3} \end{array} \qquad \text{and} \qquad \begin{array}{c} \ce{R1}\\[-2pt] \ce{\quad \backslash}\\[-2pt] \ce{R2 -> P=O}\\[-2pt] \ce{\quad /}\\[-2pt] \ce{R3} \end{array} \]
We regard the double bond as something integral, and therefore here the usual tetrahedron applies. To this category also belongs the recently prepared arsine sulfide1.
\[ \mathrm{HOOC \cdot C_6H_4 \cdot As} \begin{array}{c} \nearrow \mathrm{CH_3}\\[-0.4em] \searrow \mathrm{C_2H_5}\\[-0.2em] \big\Vert\\[-0.2em] \mathrm{S} \end{array} \]
Of great interest are the sulfur compounds that have only just been obtained2.
\[ \begin{array}{c} \mathrm{NH_2 \cdot C_6H_4}\\ \mathrm{CH_3 \cdot C_6H_4} \end{array} \!>\!\mathrm{S{=}O} \qquad \mathrm{CH_3 \cdot C_6H_4 \cdot SO_2 \cdot N - S} \!<\! \begin{array}{c} \mathrm{C_6H_4 \cdot COOH}\\ \mathrm{CH_3} \end{array} \]
The cause of optical activity here undoubtedly lies in the sulfur atom. However, according to the classical structures, it not only has “identical radicals,” but is, in addition, trivalent. It is quite impossible to imagine the structure of two antipodes with a trivalent asymmetric atom without some additional hypotheses. The author of the preparation supposes that the fourth valence of sulfur exists as such and is directed in space while remaining free. Such a very untimely interpretation becomes entirely unnecessary if one assumes that sulfur is linked by a mesovalence with one or with two ortho hydrogens of the nucleus. It then becomes tetravalent, and before us is the usual case of optical activity. If all the radicals are different, or if both ortho hydrogens of the nucleus are replaced by $\mathrm{CH_3}$, then there is no doubt that it will not be possible to resolve these substances into antipodes.
Summarizing all that has been said above, we see that of the six categories of molecular asymmetry outlined by us only two remain: the asymmetry of complex compounds and of diphenyl derivatives. The isolated case of inosite cannot be regarded as a category. As for the asymmetry of complex compounds, it apparently admits no doubt, although with time a somewhat different, more detailed interpretation of their structure is possible, in view of their still very great complexity. The cause of the optical activity of diphenyl derivatives remains unclear for the present, since there is still little factual material here. In some cases it is possible to outline definite meso-formulas; in others this is very difficult. Investigators in this field find optical activity in some derivatives quite easily, while in others it does not occur despite careful special searches. Both the one and the other at present arouse sincere astonishment even among the experimenters themselves. Thus, at the present time, the most important, almost exclusive, cause of optical activity should be considered to be the asymmetric atom, in the overwhelming majority of cases tetravalent. Molecular asymmetry remains only for inorganic complex compounds.