Production of titanium, zirconium, and potentially aluminum is based on carbochlorination of the respective oxides with subsequent reduction of the product chlorides to metal. Selection of the carbon reagent for the carbochlorination reaction is based on availability, reactivity, purity, and price. Petroleum coke, metallurgical coke, and pyrolyzed petroleum have been used commonly. These reductants, along with carbon monoxide, have been studied at the Bureau of Mines for the carbochlorination of domestic kaolin to produce aluminum chloride. Reactivity has been the major concern, but perusal of pertinent literature indicated that chlorinated carbon byproducts should also be considered. ~ : Exploratory experiments were undertaken to determine if chlorinated carbon compounds may be byproducts of metallurgically important carbochlorination reactions and reactions between chlorine and various forms of carbon. Twelve tests were made in two groups. Group 1 experiments were carried out in a 50-mm-diameter externally heated quartz tube with ancillary, all glass, airand ice-cooled condensers. All of the glassware was washed, rinsed with distilled water, rinsed with pesticide-grade hexane, and heated to 500 ~ before use to remove organic compounds. Teflon tape was used in the ground glass joints that connected the pieces of equipment. The bed materials were mixed and pelletized to facilitate uniform gas flow through the bed. Pellet beds were baked out at 700 ~ under flowing highpurity nitrogen prior to chlorination. During bed bake-out the condensing system was not attached to the reactor. The reaction was carried out by adjusting the reactor temperature and introducing chlorine (commercial grade) along with the nitrogen at 1 1/min each. For Test 1, carbon monoxide was substituted for the nitrogen during the reaction time. After the allotted reaction time, the glassware was cooled under a flow of nitrogen; it was then disassembled, and the chloride product condensers were rinsed with distilled water and with hexane to extract the organic compounds. The water and hexane rinses were combined and partitioned in a 500 ml separatory funnel. The aqueous fraction was extracted three times with 50 ml portions of hexane. The total volume of hexane was dried with anhydrous sodium sulfate and evaporated to 1 ml in high-purity nitrogen at room temperature. Blank tests were made without a bed of material in place. The second group of four tests was made to investigate the effect of carbon types that would not be used as commercial reductants but could indicate the effect of carbon type on the chlorinated products. These tests were carried out in a much smaller quartz and glass apparatus because of the limited quantities of glassy carbon and diamonds* available.