A study of the transformation of diamond to graphite

Abstract

Diamond fragments with {111} surfaces, thin enough for examination by transmission electron microscopy, have been prepared. The fragments have been heated under nonoxidizing conditions in the temperature range of 1500 to 1900 $^\circ$C. Subsequent examination by transmission electron microscopy and selected area electron diffraction has shown that under some conditions very slight graphitization of the diamond can be detected after heating for 45 min at 1500 $^\circ$C. Higher temperatures produce an increase in the rate of graphitization until at 1900 $^\circ$C almost all the areas suitable for examination have been converted to graphite after heating for 5 to 10 min. At the initial stages of graphitization the c axis of the graphite crystallites is oriented perpendicular to the underlying diamond surface with the a axes lying parallel to the $\langle110\rangle$ directions in the diamond surface. As the amount of graphitization increases the c axis is mainly oriented in a cone around the $\langle111\rangle$ direction which is perpendicular to the diamond surface. Heavy graphitization produces a wide variation in the orientations of the c axis with a tendency for it to be parallel to all four $\langle111\rangle$ directions of the diamond. The graphite has a crystallite size which varies between 100 and 150 $\overset{\circ}{\mathrm A}$ in linear dimensions. Evaporated carbon films which are almost amorphous have been heated in the temperature range of 1500 to 1800 $^\circ$C. Grain growth occurs to a crystallite size of 100 to 130 $\overset{\circ}{\mathrm A}$ with the c axis of the graphite crystallites oriented perpendicular to the film surface. Carbon films have also been evaporated on thin diamond fragments and heated to 1500 $^\circ$C. Grain growth of the carbon film again occurs with the c axis of the crystallites oriented perpendicular to the diamond substrate. It is suggested that during the graphitization of diamond, grain growth occurs which produces the graphite crystallites that are examined after the diamonds have been heated. During the growth of the crystallites the nature of the underlying diamond surface determines the orientations of the crystallites. Because the structures of diamond and graphite are so dissimilar it is considered very unlikely that a direct phase transformation occurs in the graphitization process. The diamond in the neighbourhood of regions where graphitization has occurred is highly strained and evidence is presented for the nucleation and glissile movement of dislocations in diamond at 1750 $^\circ$C.