366 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1960 



atomic radius, some of the boron atoms might diffuse into the holes 

 in the diamond. 



I doubt whether we were successful in these attempts, but we did 

 find that using carbon instead of boron led to an interesting effect. 

 After being heated in this way it was found that when a voltage was 

 applied to the crystal, it glowed with a bright green color. This is a 

 quite new electroluminescent effect which diamonds do not normally 

 show. The precise mechanism involved in the production of light is 

 not clear, and with all our other interests we have had to put this 

 aside for a future, less busy time. However, the effect itself is very 

 beautiful. 



Several properties of the diamond and indeed of all solids are 

 mainly controlled by the form of the lattice and its modes of vibra- 

 tion. The specific heat is one of the best known in this respect. There 

 are other properties, such as electrical conductivity, which are thought 

 of as being mainly controlled by the electrons in the crystal. A di- 

 vision such as this is more apparent than real, for of course there is 

 always interaction between the lattice and the electrons, and the one 

 affects the other. With this reservation, let us consider two properties. 

 The first is thermal conductivity. 



Tliose who have handled large diamonds will have been struck 

 by how cold a handful of them feel. Wool and felt are warm to the 

 touch, metals are cold and diamonds are colder still. These feelings 

 of warmth and of cold are because of the thermal conductivity of the 

 material. A metal such as copper is a very good conductor of heat, 

 but the thermal conductivity of diamond is four times as good at 

 room temperature. Copper has a thermal conductivity of 4.1 watts/ 

 cm. deg. K. and diamond 16 watts/cm. deg. K. If different parts of a 

 solid are at different temperatures, then heat may be transferred 

 from the hot to the cold parts by two processes. The transfer may 

 be due to free electrons moving through the lattice, rather like con- 

 vection currents in gases; or it may be due to vibrations or waves 

 passing down the lattice, rather like heat transfer by radiation in a 

 gas. The waves traveling through the lattice are referred to as 

 phonons, and the whole structure behaves, as Kathleen Lonsdale has 

 remarked, like granadilla pips wobbling in a soft jelly. In the case of 

 copper there are large numbers of electrons freely available in tlie 

 crystal to participate in the conduction of heat, but for the diamond, 

 none of the electrons are free and we are left with the lattice vibra- 

 tions as the mode of heat transfer. 



In a gas the thermal conductivity is given by the expression 

 s=CuL/Z, where G is the heat capacity of unit volume, u is the aver- 

 age particle velocity and L is tlie mean free path of a particle be- 

 tween collisions. The idea of a mean free path for particles has been 



