NOx reduction experiments were conducted by direct injection of urea into a diesel fueled, combustion-driven flow reactor which simulated a single engine cylinder ($966cm^3$). NOx reduction tests were carried out over a wide range of air/fuel ratios (A/F=20-40) using an initial NOx level of 530ppm, and for normalized stoichiometric ratios of reductant to NOx (NSR) of 1.5 to 4.0. The results show that effective NOx reduction with urea occurred over an injection temperature range of 1100 to 1350K. NOx reduction increased with increasing NSR values, and about a 40%-60% reduction of NOx was achieved with NSR=1.5-4.0. Most of the NOx reduction occurred within the cylinder and head section (residence time <40msec), since temperatures in the exhaust pipe were too low for additional NOx reduction. Relatively low NOx reduction is believed to be due to the existence of higher levels of CO and unburned hydrocarbons (UHC)inside the cylinder, and large temperature drops along the reactor. Injection of secondary combustible additives (diesel fuel/$C_2H_6$) into the exhaust pipe promoted further substantial NOx reduction (5%-30%) without shifting the temperature windows. Diesel fuel was found to enhance NOx reduction more than $C_2H_6$, and finally practical implications are further discussed.