p. G. KLEMENS 



displacement cascade an appreciable fraction of the bonds is broken, some 

 by the displacements, some by the heat liberated in the displacement spike, 

 and when the material settles down again it goes into the metastable vitreous 

 state. 



Similar thermal conductivity studies have been undertaken by Berman, 

 Foster and Rosenberg-^ on neutron-irradiated sapphire and diamond, and 

 it seems that these results should be interpreted in a similar way. At Oak 

 Ridge, Cohen (private communication) has studied potassium chloride 

 crystals after gamma and neutron irradiation. After gamma irradiation 

 only single-point defects appeared: this is reasonable for the Compton 

 electrons could only produce single displacements. After neutron irradiation, 

 more complicated imperfections appear, but the detailed interpretation is 

 still uncertain. The effect of neutron irradiation on the thermal conductivity 

 of fused silica was also studied by her — see below. 



That large-scale defects in irradiated quartz should be vitreous inclusions 

 is confirmed by the fact that very intense neutron irradiation turns quartz 

 into an amorphous material 29.3o,3i There is, however, a small difference 

 in density between fused silica and the irradiation-produced glass, the 

 latter being about 2 per cent denser. This suggests a less open and, therefore, 

 less disordered structure, so that the irradiation-produced glass should have 

 a higher thermal conductivity, particularly at low temperatures^^, a pre- 

 diction recently confirmed by Cohen. 



Vitrification due to heavy neutron bombardment does not occur in all 

 insulators^^, but presumably only in glass-forming substances, or possibly 

 also in those which just fail to satisfy the usual criterion of vitrificability^*. 

 Certain radio-active minerals occur in an amorphous state, the metamict 

 state, due to radiation damage by alpha particles produced within them^ 

 and Primak pointed out the essential similarity between these two pheno- 

 mena. Metamictization probably occurs only as a result of heavy particle 

 or neutron bombardment, when displacement clusters or spikes are formed. 



Wittels and SherrilP^ observed density changes in quartz on neutron 

 bombardment. Very curiously, the rate of density change increases quite 

 markedly before decreasing again on approaching saturation, instead of 

 gradually decreasing with dosage, as would be expected for simple satura- 

 tion. Klemens^'^ suggested that this is due to the vitreous inclusions: each 

 inclusion will tend to expand, but will be constrained by the surrounding 

 crystal, so that the system is under internal stress. After further irradiation, 

 when the density of inclusions is higher, they will interact elastically, stresses 

 will be relieved by plastic flow, and the delayed expansion will occur. On 

 this model any change of lattice parameters would be essentially unrelated 

 to changes in hydrostatic densities, and the accelerated change of the latter 

 should not be accompanied by a similarly accelerated change of the former; 

 however. X-ray fines should be broadened and skewed because of the 

 inhomogeneous strain. 



Further extensive measurements by Wittels^'^ did not confirm this theory, 

 for although there was appreciable broadening and skewing of the lines, and 

 there were good indications of the presence of vitreous inclusions from diffuse 

 rings in Laue photographs, there was also a real shift of the lines, and the 

 change in lattice spacings not only paralleled the change in hydrostatic 



279 



