Passage Effects in Paramagnetic Resonance Experiments

1067 1070 1072 1078 1081 1081 1083 1091 1096 1103 1107 1111 1112 Shapes of electron spin resonance ( E S R ) a n d nuclear magnetic resonance signals, observed under various conditions, have previously been predicted theoretically a n d confirmed experimentally for a few cases.1"5 T h e purpose of this work is as follows: i. To understand a n d predict the shapes of observed traces, as an a i m in itself. ii. To use the shapes of the traces to gain some understanding a b o u t the local fields in the sample, m a i n l y to understand to w h a t extent the fields can be considered static, a n d to w h a t extent they must be considered as d y n a m i c ( n a m e l y , responsible for forbidden lines, etc.). iii. To determine various conditions under which the resulting traces m a y cause misleading determinations of line shapes, relaxation times, etc. iv. To distinguish true " p h y s i c a l " effects from spurious "passage effects" (like splitting of lines). T h e work was performed on single crystals of phosphorus-doped silicon, containing a b o u t 1.7 X 10lfi donors per c m , at temperatures of 10°, 4.2° a n d 1.2°K. T h e E S R line d u e to the donors was observed at a b o u t 9300 mc, e m p l o y i n g a standard superheterodyne microwave spectrometer, 6 under various experimental conditions. (Mainly various rates and periods of the m o d u l a t i n g magnetic field, a n d various microwave power levels.) Various traces were predicted theoretically a n d observed experimentally (or vice versa).