This review discusses the problems of sulfur reduction in highway and non-road fuels and presents an overview of new approaches and emerging technologies for ultra-deep desulfurization of refinery streams for ultra-clean (ultra-low-sulfur) gasoline, diesel fuels and jet fuels. The issues of gasoline and diesel deep desulfurization are becoming more serious because the crude oils refined in the US are getting higher in sulfur contents and heavier in density, while the regulated sulfur limits are becoming lower and lower. Current gasoline desulfurization problem is dominated by the issues of sulfur removal from FCC naphtha, which contributes about 35% of gasoline pool but over 90% of sulfur in gasoline. Deep reduction of gasoline sulfur (from 330 to 30 ppm) must be made without decreasing octane number or losing gasoline yield. The problem is complicated by the high olefins contents of FCC naphtha which contributes to octane number enhancement but can be saturated under HDS conditions. Deep reduction of diesel sulfur (from 500 to < 15 ppm sulfur) is dictated largely by 4,6-dimethyldibenzothiophene, which represents the least reactive sulfur compounds that have substitutions on both 4- and 6-positions. The deep HDS problem of diesel streams is exacerbated by the inhibiting effects of co-existing polyaromatics and nitrogen compounds in the feed as well as H2S in the product. The approaches to deep desulfurization include catalysts and process developments for hydrodesulfurization (HDS), and adsorbents or reagents and methods for non-HDS-type processing schemes. The needs for dearomatization of diesel and jet fuels are also discussed along with some approaches. Overall, new and more effective approaches and continuing catalysis and processing research are needed for producing affordable ultra-clean (ultra-low-sulfur and low-aromatics) transportation fuels and non-road fuels, because meeting the new government sulfur regulations in 2006-2010 (15 ppm sulfur in highway diesel fuels by 2006 and non-road diesel fuels by 20 10; 30 ppm sulfur in gasoline by 2006) is only a milestone. Desulfurization research should also take into consideration of the fuel-cell fuel processing needs, which will have a more stringent requirement on desulfurization (e.g., < 1 ppm sulfur) than IC engines. The society at large is stepping on the road to zero sulfur fuel, so researchers should begin with the end in mind and try to develop long-term solutions. (C) 2003 Elsevier B.V. All rights reserved.