244 Walter Gordy 



its rate of decay. Also, the very fact that electrons could achieve such freedom 

 witliin an organic solid might itself be classed as desirable information. For- 

 tunately, however, the electron resonance signals are often rich with information 

 about the environment of the unpaired electrons. Our problem is to decode 

 their messages. There are at least tliree important sources of information 

 in these resonances. The first is the hyperfine structure arising from inter- 

 actions of the electron spin moment with magnetic moments of various nuclei 

 around or near the unpaired electron. The second is the small residual spin- 

 orbit interaction which in some instances makes the g factor slightly anisotropic 

 and different from the free spin value of 2.0023. The third is the information 

 which can be obtained from the line widths and shapes. The most important 

 of these sources is the nuclear hyperfine structure. 



Most instruments used for detection of electron spin resonance plot the 

 intensity of absorption at a particular frequency as a function of d.c. magnetic 

 field. The appearance of the plot depends upon the instrument as well as 

 upon the actual, intrinsic shape of the resonance. I shall not discuss possible 

 variations in the actual line-shapes, but shall here assume that the resonances 

 have gaussian shape when the intensity of absorption at a constant frequency 

 is plotted as a function of d.c. magnetic field. A high-fidelity receiver and 

 recorder (or cathode ray scope) would reproduce closely the actual shape of 

 the resonance curve, as shown in Fig. 1(a). The high-fidelity systems are not, 

 however, the most sensitive systems. The most sensitive methods of detection 

 employ modulation of the resonance relative to the observation frequency. 

 A narrow-band amplifier is tuned either to the modulation frequency or to a 

 higher harmonic of tliis frequency. If one uses a frequency modulation which 

 is very small as compared to the width of the resonance and a phase-sensitive 

 amplifier tuned to the modulation frequency, a curve like that in Fig. 1(b) is 

 obtained. This curve represents the first derivative of the actual line-shape. 

 If one uses such a receiver and tunes to the second harmonic of the modulation 

 frequency, a curve hke that in Fig. 1(c) is obtained. This curve represents 

 the second derivative of the actual fine-shape. Both the first and second 

 derivative curves are commonly employed in display of electron spin resonances. 

 In interpretation of the curves it is desirable to know what method of detection 

 has been employed, especially when there are structural components incom- 

 pletely resolved. In the illustrations which follow we shall sometimes use first 

 and sometimes second derivative displays. 



This brief description of the appearance of the signals and the simplified 

 theory of the structure of the resonance given below will, I hope, make it possible 

 for you, whether you are a biologist, chemist, physicist, or hybrid, to share 

 with us some of the fun of trying to decode the complex microwave messages 

 which we have been receiving from biological substances. You will be able, 

 I hope, to decide for yourself what is definitely proved by the resonances, 

 what is strongly suggested but not proved, and what is merely hinted. 



1 . Nuclear Hyperfine Structure 



The hydrogen nucleus, with a relatively large magnetic moment, 2.79 nm, 

 and nuclear spin of |, is abundant in all organic matter. The only other nucleus 

 with a non-zero spin abundantly found in biochemicals is N^^ (/ = 1 and 



