The study investigated the impact of ethanol blends on criteria emissions (THC, NMHC, CO, NOx), greenhouse gas (CO2), and a suite of unregulated pollutants in a fleet of gasoline-powered light-duty vehicles. The vehicles ranged in model year from 1984 to 2007 and included one Flexible Fuel Vehicle (FFV). Emission and fuel consumption measurements were performed in duplicate or triplicate over the Federal Test Procedure (FTP) driving cycle using a chassis dynamometer for four fuels in each of seven vehicles. The test fuels included a CARB phase 2 certification fuel with 11% MTBE content, a CARB phase 3 certification fuel with a 5.7% ethanol content, and E10, E20, E50, and E85 fuels. In most cases, THC and NMHC emissions were lower with the ethanol blends, while the use of E85 resulted in increases of THC and NMHC for the FFV. CO emissions were lower with ethanol blends for all vehicles and significantly decreased for earlier model vehicles. Results for NOx emissions were mixed, with some older vehicles showing increases with increasing ethanol level, while other vehicles showed either no impact or a slight, but not statistically significant, decrease. CO2 emissions did not show any significant trends. Fuel economy showed decreasing trends with increasing ethanol content in later model vehicles. There was also a consistent trend of increasing acetaldehyde emissions with increasing ethanol level, but other carbonyls did not show strong trends. The use of E85 resulted in significantly higher formaldehyde and acetaldehyde emissions than the specification fuels or other ethanol blends. BTEX and 1,3-butadiene emissions were lower with ethanol blends compared to the CARB 2 fuel, and were almost undetectable from the E85 fuel. The largest contribution to total carbonyls and other toxics was during the cold-start phase of FTP. (C) 2011 Elsevier Ltd. All rights reserved.