MICROWAVE PARAMAGNETIC RESONANCE ABSORPTION 483 



For a high degree of saturation 



7'TiT2CQo(l - r-)P,„ » 1 



and substituting for 



Ron 

 r 



r = 



2 



- 1 



Ron 



+ 1 



we get : 



Pinyi'TiT2C ^ ^ 



The above relation shows that under saturated conditions the Q of the 

 cavity does not enter and one might as well not use one or use a very 

 much overcoupled cavity. This is one of the reasons why in microwave 

 gas spectroscopy,* where lines are easier saturated a cavity is not used. 

 (The more important reason is that in most cases one sweeps the fre- 

 quency of the source, so that a cavity is difficult to use.) 



Equation (75) also shows that the signal and also signal to noise goes 

 down with increasing RF power. The above argument does not hold for 

 the out-of-phase (dispersion) signal, in particular it breaks down com- 

 pletely for signals observed under fast adiabatic passage conditions.^ 

 For the latter case one wants as high an RF field as possible. 



IX. ACKNOWLEDGEMENT 



I profited greatly from discussions with various members of the 

 resonance group at the University of California in particular with 

 Profs. A. F. Kip, and A. M. Portis and at Bell Telephone Laboratories 

 with Drs. R. C. Fletcher and S. Geschwind. I would like especially to 

 thank E. Gere for his expert help in the construction of the equipment 

 and to Prof. C. P. Slichter and Dr. R. H. Silsbee for helpful criticism of 

 the manuscript. 



REFERENCES 



1. See for example F. Bloch, Phys. Rev., 70, p. 460, 1946. N. Bloembergen, E. M. 

 Purcell, R. V. Pound, Phys. Rev., 73, p. 679, 1948. 



* In microwave gas spectroscopy the fractional power loss per unit length a is 

 used. Its relation to the susceptibility is a = 8, 2x"/^a where Xc? is the guide 

 wavelength. 



