382 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 60 



number trapped before they reach it. These trapped holes establish a 

 space charge which reduces the electric field in the region where the 

 electron hole pairs are produced. Eventually equilibrium is estab- 

 lished when the number of holes which reach the cathode is exactly 

 equal to the number of electrons leaving the crystal. The counting 

 rate is now uniform but at a lower level than initially. Now, what 

 happens if we use infrared light on the diamond? It is found that 

 a polarized diamond at once shows an improvement ; the comiting rate 

 rises until it is restored to its original value. The same effects are 

 produced as light of shorter and shorter wavelength, i.e., of increas- 

 ing energy, is used, until at about 6,500 A a new phenomenon pre- 

 sents itself. Instead of improving the counting rate, light of this 

 energy, or greater, reduces the counting rate. This is the main evi- 

 dence for two levels of impm-ities, one of which is essentially a hole 

 trap, and the other an electron trap. 



Further information is obtained by irradiating the diamond 

 through the negative electrode. With this arrangement it is the holes 

 which can reach the cathode because of its proximity, while the elec- 

 trons are trapped as they move toward the anode. The counting rate, 

 the establishment of the polarizing field, and the eventual equilibrium 

 counting rate is found to be much the same as before. It is thus con- 

 cluded that the density of electron traps and of hole traps is much 

 the same. 



Much of the so-called progress in science consists of gathering a few 

 scraps of information and on the basis of these to give one's imagina- 

 tion full rein and so develop a model — a picture — of what is taking 

 place and which will explain the observations. On the basis of this 

 physical model, predictions can usually be made about what may hap- 

 pen imder a difi'erent set of experimental conditions. We try these 

 new conditions and sometimes receive confirmation of the hypotheses 

 and sometimes see that they must be modified. It is even more fas- 

 cinating than putting together the pieces of a jigsaw puzzle or solving 

 a crossword. 



Omitting the many wrong turnings we made, the following band 

 picture is suggested as best satisfying the results. Near the bottom of 

 the band there are acceptor energy levels, aU of which are occupied 

 by electrons. This set of energy levels is probably between 0.8 and 1.2 

 eV above the valence band. Higher up are a set of donor levels, the 

 greater majority of which have lost their electrons to the acceptors. 

 The energy of this band is about 2.0 eV above the valence band or it 

 might be at 2.0 eV below the conduction band. This point remains to 

 be clarified and the two cases cannot be distinguished at present. In 

 semiconductor parlance, it would appear that the counting diamonds 

 are p-type when they are activated. 'Wlien the beta irradiation enters 



