222 REPORTS ON THE STATE OP SCIENCE. — 1920. 



Absorption Spectra of Organic Compounds. — Report of Committee 

 (Sir J. J. DoBBiE, Chairman; Professor E. E. C. Baly, Secretary; 

 and Dr. A. W. Stewart). Drawn up by the Secretary. 



Various theories have been advanced from time to time to explain the absorption 

 bands exhibited by organic compounds, and it would seem advisable at this time 

 to deal with these and to state the position that has been reached in this branch 

 of scientific investigation. There is no doubt that the pioneer in this field of 

 work was the late Sir Walter Noel Hartley. He was the first to undertake a 

 detailed investigation on scientific lines of the absorption exerted by organic 

 compounds in the visible and ultra-violet regions of the spectrum. He was the 

 first to recognise the fact that isolated measurements of the absorption spectrum 

 of a substance in solution are valueless, and he devised the method whereby com- 

 plete records of the absorption could be obtained. Hartley's method consisted 

 in measuring the oscillation frequencies of the light for which complete absorption 

 is shown by definite thicknesses of a solution of known strength of the sub- 

 stance. The observations were repeated with the same thicknesses of more and 

 more dilute solutions until no measurable absorption was observed. By plotting 

 the oscillation frequencies against the thicknesses expressed as equivalent thick- 

 nessee of some selected concentration an absorption curve was obtained, called by 

 Hartley a molecular curve of absorption. 



At the present time this method of observation has been displaced by the 

 quantitative measurement of the light absorbed. The absorptive power exhibited 

 by a given substance for light of a given frequency is expressed in terms of the 

 molecular extinction coefficient, log lo/I^dc, where lo/I is the ratio of the 

 intensities of the incident and emergent light as observed with a layer d cms 

 thick of a solution containing c gram molecules of the absorbing substance 

 dissolved in a litre of some diactinic solvent. 



Reference may, be made to the use of a solution of the substance under 

 examination. In general it may be said that the absorptive power exerted 

 by compounds is large, with the result that it is necessary to use very thin layers 

 for purposes of observation. This is impossible of realisation with solid sub- 

 stances, and indeed with many liquids the thickness required is so small that 

 without very accurate and expensive apparatiis the necessary thin layers cannot 

 be obtained. By common consent, therefore, solutions of known strength in 

 diactinic solvents are employed. It must be remembered, however, that the 

 mfluence of a solvent on the absorptive power of a compound is often very 

 marked, and due allowance must be made for this effect. The question of the 

 influence of a solvent will be discussed later. 



The region of the spectrum dealt with by Hartley extended from the red 

 end to the limit of the ultra-violet as set by a quartz spectrograph working in 

 air, that is to say, between the limits of wave-length 6000 and 2100 Angstroms. 

 He showed in the first place that substances can in general be divided into two 

 classes, namely, those which exhibit selective absorption, i.e., absorption bands 

 between the above spectral limits, and those which exhibit only general absorp- 

 tion. It is not necessary here to detail the whole of Hartley's work, but one 

 important fact was established, namely, that, providing no disturbing factor 

 intervenes, the absorption curves shown by compounds of similar constitution are 

 themselves similar. This fact was made use of in determining the constitution 

 of a few substances with reference to which the chemical arguments at the time 

 were at fault. It was shown for instance that phloroglucinol is a true trihydroxy- 

 benzene and not ketonic since its absorption curve is very similar to that of its 



CO 



trimethyl ether.* Similarly the constitution of isatin C^IlX /CO, 



' See references, p. 243, 



