Page:1902 Encyclopædia Britannica - Volume 26 - AUS-CHI.pdf/776

This page needs to be proofread.



was to be excluded. But, apart from the fact that a hypothesis which requires so much helping out is of little practical use from the chemist’s point of view, it seems necessary to admit that whatever may be the nature of chemical affinity, the same forces are active throughout— that what sometimes is termed residual affinity is of the nature of ordinary affinity, and that the difference is in quantity rather than in quality. Structure and Isomerism. The study of structure is correlative with that of valency; in fact they are strictly interdependent studies, our views on valency being founded on the success with which we are able to apply them in deducing satisfactory structural formulae. Of late years chemists have devoted their attention almost exclusively to the investigation of carbon compounds, to such an extent as even to provoke the complaint that they are guilty of unduly neglecting the study of other elements. But even if true there would be slight reason to deplore this. In consequence of the extraordinary activity displayed by workers in organic chemistry, our knowledge of carbon compounds has reached a high degree of'development, and materials are fast being accumulated from which it will be possible ultimately to deduce precise conclusions as to the manner in which properties are dependent on structure ; since it is now clear that the properties of the carbon compounds are primarily functions of their structure, and only in a minor degree dependent on their composition. The number of carbon compounds already known is legion, and there are still infinite possibilities before us. The technique of the subject a so Difficulty ^ highly developed, and new combinations of deter- are readily effected; it is easy, therefore, as a mining rule, to test any hypothesis that may be framed, structure. anq ultimately to secure evidence which is sufficient either to prove or disprove its accuracy. But in the case of all other elements a far less plastic material is dealt with, and the field of investigation's a comparatively narrow one; it is therefore less easy to avoid a dogmatic construction of the evidence. The difficulty of dealing even with such simple compounds as the chlorides has already been referred to. To deal with the oxides is still more difficult. Indeed, there can be little doubt that our knowledge of the structure of inorganic compounds generally is far less definite than is often supposed to be the case, and that much is taught in the text-books for which there is no justification. To take two examples, the alkalies and acids. Caustic soda and potash are now invariably represented by the formulae iSTaOH, KOH, but there is no evidence which can be called evidence that these are correct: they are but an expression of a prevailing fashion, being used because the view is held that the relationship of the alkalies to water is—or ought to be— that represented by such formulae, and because of the existence of alcohols. It is, at least, as probable that they are but hydrates of the oxides, e.g., Na2O.OH.2. Again, sulphuric acid is always represented as S02(0H)2; the arguments on which this formula rests, however, are equally applicable to the old Berzelian formula, S03.0H2. The statement commonly made that by the action of phosphorus pentachloride sulphuric acid is first converted into SOgHCl and then into S02C12, and that thereby proof is obtained of the existence of two OH groups in the acid, to begin with, is half wrong, as the second product is pyrosulphuryl chloride, S205C12, not sulphur^l chloride, S02C12. Moreover, the production of the compound SO3HCI cannot be held to prove that the acid contains the hydroxyl group; as this compound is readily obtained by combining sulphuric anhydride with hydrogen chloride, its production may be a consequence of the removal of

“ water ” from the acid by the .action of the pentachloride, and the consequent union of the residual oxide with hydrogen chloride formed by the interaction of water and phosphorus pentachloride. It is not impossible that SOgHCl is a molecular compound, and not an acid chloride, as the corresponding compound formed by combining ethyl chloride with sulphuric anhydride behaves as a mild form of the latter, i.e., as a sulphonating agent, not as an acid chloride. Lastly, it may be pointed out that the properties of sulphuryl chloride, S02C12, are altogether peculiar, and not those of an acid chloride; as a rule it acts as a chlorinating agent, consequently its conversion into sulphuric acid by water affords no proof of the constitution of the acid. It may well be that water is initially “chlorinated” by it, and that in consequence the S02 group becomes oxidized to S03, which then unites with water, forming sulphuric acid. It is unnecessary to pursue such arguments farther. It is sufficient to point out that in many cases our views of structure are of the nature of preconceived opinions, being based on a too narrow interpretation of the facts. It is time that students were made more fully aware of the precise value of the formulae they are in the habit of using, as it is essential that the chemist should preserve a free and balanced judgment. The dogmatic treatment too frequently accorded to our subject is absolutely subversive of true scientific method and serves to retard progress. Passing to carbon compounds, the greatest advance to lie noted in our knowledge of structure is that due to the introduction of geometrical conceptions by van’t Hoff and Le Bel independently in 1874, which has given rise to the now well-known doctrines of stereoisomerism, i.e., isomerism due to differences in the arrangement of the atoms in space. In this doctrine it is assumed that the molecule is a stable system of material points, and that the radicles associated with a carbon atom occupy fixed positions ; it is further supposed that they are situated in the manner represented by taking the tetrahedron as the model of the carbon atom and grouping the four radicles with which it is associated in all saturated compounds at the four corners of the model—of which the carbon atom may be supposed to occupy the centre of gravity. On this assumption, a compound CIt'4 can exist in only one form, so long as any two of the radicles are identical, but when all four are unlike two forms are possible. The models of these two forms are two tetrahedra which cannot be superposed but bear to each other the relation of right to left hand; the two forms are therefore said to be. enantiomorphously related, and as neither planes nor centre of symmetry can exist in such forms, the carbon atom which is their central feature is termed an asymmetric carbon atom. Of greatest importance is the explanation this doctrine affords of cases of isomerism such as that met with in the two tartaric acids, the separation of which from racemic acid was Pasteur’s great achievement and the one which first gave him fame. Pasteur, indeed, in discovering the relationship of these acids and their possible structure, clearly foreshadowed the hypothesis (cf. Frankland, Pasteur Memorial Lecture, Trans. Chem. Soc. 1897, p. 691). It is now known that in all cases in which carbon compounds manifest the power in solution of rotating the plane of polarized light—as do the two tartaric acids—they contain one or more asymmetric carbon atoms. On the other hand, it has been shown that in the case of a large number of compounds represented by a formula containing an asymmetric carbon atom, but initially produced in an optically-inactive form, it is possible, by methods such as were introduced by Pasteur, to resolve the inactive product into two isomerides of equal but opposite optical