The fundamental researches on photochemical equivalence were those of E. Warburg. They occupy in the modern era of photochemistry somewhat the position of the classical photo chemical investigations by Bunsen and Roscoe in preceding years. Warburg introduced a convenient term, the quantum efficiency: this is the ratio of the effective photochemical equivalent c/;• to the ideal photochemical equivalent P. Both 4) and P are expressed in terms of the number of gram-molecules absorbing and decom posed by one gram-calorie of radiation of frequency v. When Einstein's law holds 0/ P = I. Tabulated results of several reac tions show that relatively few reactions confirm the Einstein rela tion, among which are some for which the efficiency is actually less than unity, but which give this value when extrapolated back to the start of the reaction. This procedure is justifiable when a strongly absorbing product is formed. On the other hand there are many reactions whose efficiency (4/P) is much greater than unity, even to hundreds and thousands of molecules per quantum.
Mechanism of Photochemical Reactions.—Although it is true that certain substances are more obviously photosensitive than others, a useful classification of photochemical reactions is not possible in terms either of chemical structures or of types of chemical reaction. This is particularly evident in the field of organic chemistry. The great variety of substances and types of reaction is illustrated in the following table: To obtain a classification it is therefore necessary to search more deeply into the fundamental mechanisms of photochemical reactions. Every advance in knowledge of these will displace a previous classification or system but will supply a better working hypothesis. Einstein's proof of his principle implied that the primary photochemical process is a dissociation. of a molecule, e.g., into atoms. Thus in the combination of hydrogen, H2, and chlorine, in light to form hydrogen chloride, HC1, it was assumed that the primary reaction was C12 -I- hv-=-C1-1- Cl. Now the photochemical efficiency of the H2,C12 reaction is very great—it can amount to thousands of molecules of HC1 per quantum absorbed. This was explained by the conception of secondary reaction chains (see subsequent section). Primary decomposition of molecules also implies a wave-length limit to a given photochemical reaction; because if the quantum, hv, absorbed be of less energy than the heat of formation of the molecule, it should be impossible to decompose it directly, and therefore, according to Einstein's principle, the relation of photochemical yield 4 to the wave-length would have the form shown in fig. I.
Now cases are certainly known where the wave-lengths active in photochemical change are below the value corresponding to the heat of formation. In fact, the existence of definite photo
chemical spectrum thresholds has scarcely been demonstrated.
A difficulty of exactly the opposite kind for the primary de composition of the molecule developed from the investigations of physicists on the energy required to make gas molecules fluoresce. It was shown that at low pressures a molecule of gas can take up many times its heat of formation without decom posing. For example, iodine, can absorb and re-emit as resonance-radiation an amount of energy five times its heat of formation On the modern theory of atomic and molecular struc ture as related to the emission and absorption of light, named, after its principal founder, the Bohr theory (see Mom), these results are explained as follows. The total internal energy of a molecule, however acquired (and understanding by a molecule a multinuclear structure) will consist of : Energy of rotation + Energy of oscillation + Energy of electron displacements.
The whole energy, and the distribution of this energy, is governed by quantum rules, i.e., it cannot vary continuously, but can only assume a series of finitely differing values, or quantum states. The gain of internal energy of an excited molecule consists pri marily in the exaltation of the electron system to higher quantum states, when the oscillation and rotation energies are also altered by the coupling of their periods with those of the electron sys tem. The rotational and oscillation energies are mainly respon sible for absorption and emission of infra-red rays. At present there is but slender evidence for direct intervention of these in chemical reactions. A primary chemical decomposition, i.e., one not involving collision with chemically indifferent molecules, might occur by such increase of rotational energy that the cen trifugal force exceeded the chemical affinity. Similarly for an excessive increase of oscillation of the parts. Apparently the con stitution of molecules is such that this cannot be effected either by direct infra-red absorption or by absorption in the bands cor responding to electronic displacements, but only at frequencies beyond the band spectrum. (A typical molecular band spectrum is illustrated in the article BAND SPECTRA, q.v.) The theory of such spectra is not yet entirely adequate to complete prediction of the distribution of energy in all cases. But certain conclusions are definite. The change in rotational energy on excitation is small enough to be negligible, so that it is quantum-coupled in crease of oscillation energy that is critical for dissociation.