C yclone fouling in fluid cokers can lead to untimely shutdowns for oil sand upgrading operators. This problem was described extensively by Richardson (1 997). The cyclones, located above the fluid bed, are responsible for separating solid coke particles from the vapour stream before it enters the scrubbing section of the coker. Excessive buildups of coke have been noted to form in various parts of each cyclone, as shown in Figure 1. These coke formations lead to an increased pressure drop through the coker and a reduction in the overall efficiency of the cyclones. This problem worsens with the passage of time until, eventually, the coker must be shut down so that the coke formations in the cyclones can be removed. Richardson (1997) discussed a number of mechanisms which may be responsible for the production of these coke structures. It is possible that the vapour phase rising from the fluid coke bed may continue to thermally crack, forming smaller molecular species and coke. Since coke i s a byproduct of thermal cracking, the formations observed in the cyclones would be initially formed by reactions of the vapour phase. Deposition of the coke onto surfaces within the cyclones would initiate the surface formation mechanism. The growth of the coke surface may continue as a result of interaction of vapour phase free radicals and the coke deposited on the walls of the cyclones. The purpose of this study was to determine if coke could be produced from the vapour phase derived from the thermal cracking of Athabasca bitumen and to investigate the effects of reaction time and temperature on the vapour phase coke yield.Coke is composed mostly of carbon (more than 95 percent by mass) combined with small amounts of hydrogen (Speight, 1991). Much research has been performed in the field of gas phase carbon formation.