Hydrogen peroxide is a clean oxidizing agent that is useful for converting organic compounds into value-added products (bulk and fine chemicals), as well as for industrial and municipal waste-water treatment, and water disinfection. However, because of the high cost of its production by the standard anthraquinone process, [1] hydrogen peroxide cannot be used for the production of bulk organic chemicals or for water treatments. Moreover, the anthraquinone method is not a green process. Hence, it is of great practical importance to develop an environmentally friendly process based on the direct oxidation of hydrogen to hydrogen peroxide. Although the formation of hydrogen peroxide in the palladium-catalyzed liquid-phase oxidation of hydrogen has been known since 1914, and several patents have been issued since then, [2±11] this process could not be put into practice. This is mostly because of its highly hazardous nature (the explosive limits of hydrogen/oxygen gas mixtures are very wide and are further widened with increasing pressure), and/or poor stirred reaction mixtures were degased under vacuum, and purged three times with argon. Stirring was stopped, and the solutions were allowed to react at room temperature under an anaerobic atmosphere. After 48 h, a 250-mL aliquot was removed, diluted with CH 3 CN (750 mL), and sonicated for 7 min to precipitate the CA enzyme. The suspension was centrifuged on an ultrafree-CL-Biomax membrane (PBCC 5000 UFC4 BCC25). This treatment was also applied when no CA was present. The filtrate was lyophilized and redissolved in H 2 O/CH 3 CN (1:1, 200 mL). The solution was analyzed by reversed-phase HPLC with detection at 230 nm on a Waters 2690 instrument equipped with a Merck RP-Select B reversedphase column (5 mm, 250 Â 4 mm, flow rate: 1 mL min À1 ). A ternary solvent gradient (solvent A: 0.1 % trifluoroacetic acid in H 2 O; solvent B: 0.08 % trifluoroacetic acid in CH 3 CN; solvent C: isopropanol) was optimized so that most of the compounds used in this study have different retention times: C: constant at 2 %; B: 0 % during 3 min, then increased to 80 % over 79 min.The assay described above was optimized to limit side reactions such as disulfide formation, alkyl chloride hydrolysis, and trialkyl sulfonium formations. Some of these side products have been identified on the chromatograms and are mentioned below. The products 3 a ± e were synthesized and characterized separately to validate their assignments on the chromatograms.The following retention times and absorption coefficients were measured: 3 a (10 min, e 230 12 000 cm À1 m À1 ), 3 b (20 min, e 230 16 000 cm À1 m À1 ), 3 c (33 min, e 230 15 000 cm À1 m À1 ), 3 d (57 min, e 230 23 000 cm À1 m À1 ), 3 e (57 min, e 230 23 000 cm À1 m À1 ), 2 d (40 min), 2 e (40 min), 4-hydroxymethylbenzoic acid (14 min), 3-hydroxymethyl benzoic acid (14 min), benzoic acid (22 min), a,a-tosylamide disulfide (61 min, e 230 23 000 cm À1 m À1 ).[1] For reviews on molecular imprinting, see: K. Mosbach, O. Ramström,