Oil recovery from the Athabasca tar sand was studied experimentally, employing a number of solvents with or without steam injection.

In the miscible type experiments, carbon tetrachloride, toluene, benzene, and naphtha employed as the displacing fluids, at flood advance rates of 2 to 42 ft/day. Highest recovery of 93.3% of the in-place bitumen was obtained in the case of carbon tetrachloride, while in the case of naphtha it was 37.2%. at a rate of 42 ft/day. Decrease in rate led to increase in recovery. For example, the recoveries increased to 72.8 and 75.2% at rates of 12 and 2 ft/day, in the case of naphtha.

A modification of the process tested consisted in recycling the produced solvent-tar mixture, with a view to improving the economics of the process. Experimental results showed that such a process can be highly efficient. For example, recycling two pore volumes of naphtha increased recovery to as high as 84.2%.

A number of experiments involved steam injection as well as injection of solvent slugs followed by steam. In the case of the more volatile solvents, the recovery was less than that for steam alone (e.g. 50.8% vs 73.4% for a 10% carbon tetrachloride slug). The reasons for such behavior are discussed, and conditions under which prior solvent injection would be effective are outlined.