PYROLYTIC FILM RESISTORS: CARBON AND BOROCARBON 289 



found to be independent of thickness above 6 X 10~^ cm, but for thinner 

 films it is probable that the true hardness is greater than that observed, and 

 the hardness shown in Fig. 1 1 for 950 deg C is probably low for this reason. 

 Through this observed dependence of hardness on the conditions of pyrolysis, 

 it appears that the hardness of pyrolytic carbon is correlated with the 

 extent to which its crystals are preferentially oriented. 



It has been shown previously that the hardness of pyrolytic carbon is a 

 function of crystal size^^, but these measurements were made on dendritic 

 growths of carbon, in which the crystals are randomly oriented. It is prob- 

 ably significant that the hardness according to these earlier measurements 

 increased with decrease in crystal size and reached a maximum value for 

 the crystal size at which, according to X-ray data" •^®, the lattice expansion 

 along the c-axis begins to manifest itself. This may be an indication that the 

 anisotropy in hardness of graphite crystals" is accentuated as the inter- 

 planar spacing increases, thus facilitating shear parallel to the basal plane. 

 Were it not for the expansion of the lattice along the c-axis, it is not im- 

 probable that the hardness would increase monotonically with decrease in 

 crystal size, since slip would be confined to progressively smaller and more 

 perfect domains. 



The hardnesses of several specially selected specimens of crystal graphite 

 were determined by the rocking pendulum method^'', employing a 90° 

 diamond prism. The apparatus was insensitive for measurement of hard- 

 nesses greater than 7 on Moh's scale, but the hardness of clean basal surfaces 

 of graphite was found to lie between 6.5 and 7. The pendulum method does 

 not eliminate purely elastic effects which may be appreciable in view of 

 the large compressibility along the c-axis. For this reason the true hardness 

 of the basal plane may considerably exceed this figure, which is, however, 

 in agreement with published values". In view of the interatomic contraction 

 in the base plane of pyrolytic carbon crystals it is probable, also, that the 

 hardness of their basal planes exceeds that of the basal plane of macrocrystal 

 graphite. 



With the prism edge oriented on the side of a relatively perfect graphite 

 crystal so as to produce shear parallel to the base plane, the observed hard- 

 ness was 0.5 on Moh's scale. Values of hardness on this scale from 1.0 to 

 1.5 were obtained for polycrystal graphite, these values being in agreement 

 with other measurements. 



In view of the pronounced anisotropy in the hardness of graphite and the 

 probably greater anisotropy of individual crystal packets of pyrolytic carbon, 

 the apparent relationship between scratch hardness and the degree of pre- 

 ferred crystal orientation is that to be expected, since preferential orienta- 

 tion of the type observed exposes the hardest surfaces of the crystals to the 

 scratching tool. 



