Xylose can be selectively converted to xylitol in water, with an optimized yield of 98%, in the presence of a simple silica supported metallic cobalt-Co/SiO2-catalyst. This catalyst displays initial outstanding catalytic properties in proper solvent, the best one being pure water. Recyclability studies show a moderate deactivation of the catalyst, while selectivity to xylitol remains almost unchanged after 4 cycles, confirming the promising character of this catalyst formulation for the xylitol production process. Xylose is a product issued from the hydrolysis of beechwood hemicelluloses. Different routes exists for the valorisation of xylose, amongst the selective hydrogenation to xylitol (scheme 1). Xylitol is an extensively used molecule in food, cosmetics, and pharmaceutical industry[1-4]. Xylitol has then been identified as one of the top-twelve value added chemicals that can be obtained from biomass [5]. As an example, xylitol is a valuable water soluble sweetener due to its atypical properties amongst, compared to sucrose, a higher sweetness strength and a lower energy capacity. Today, xylitol is widely incorporated in daily consumed products (chewing gums, chocolates, toothpastes, etc.) [6]. With an estimated market of 340 M$, xylitol is becoming the most popular "natural" sweeteners [7]. Industrial production of xylitol by selective catalytic hydrogenation appears as an economically viable approach. Scheme 1. Hydrogenation of xylose to xylitol. Xylitol can be produced by catalytic hydrogenation in the presence of RANEY® nickel. Reaction is performed in liquid phase, under high hydrogen pressure and temperature in batch-type reactors [8]. Selectivity above 90%, at high conversion degree can be obtained at 80-130°C under 40-70 bar of H2 (5 wt.% catalyst of the xylose mass present at 40-60 wt.% in water) in the presence of RANEY® nickel. Despite the high selectivity and activity reached using this catalyst, with economic advantages related to the use of noble metal free catalysts for the reaction, the intrinsic properties of the nickel, amongst toxicity and limited stability,[8-9] makes the research of efficient catalyst substitutes. Indeed, the use of xylitol for food applications imply the absence of any trace of nickel in final product, nickel being recognized as a carcinogen compound. Most of recent studies were devoted to the study of supported noble metals, such as Ru, Rh and Pd, in monometallic or bimetallic catalytic systems, with the possible adding of dopants as Sn [8-15] Ru appeared as the most efficient. Hernandez-Mejia et al.[16] reported that xylose can be selectively converted to xylitol in the presence of rutile phase TiO2 supported Ru (1 wt.%), despite the low surface area displayed by the rutile support. Authors then reported xylose yields higher than 90%, for a reaction performed in water at <140°C under 20 bar of H2 and at a catalyst to xylose ratio (wt./wt.) of 1. Compared with rutile, better dispersion of Ru over anatase does not allowed to reach comparable activity, and conversion was not selective to xylitol. As recently reviewed by Delgado Arcano et al., nickel and ruthenium catalysts then remains the most widely studied phases for the xylose to xylitol reaction, while titania is an adapted support [15] Replacement of noble metal in catalytic formulation is however preferable due to limited resources, fluctuating costs, and consumption by high technology markets. Alternative catalysts, based on earth abundant transition metals such as cobalt, and in a lower extent copper, displaying lower toxicity than nickel are presenting satisfying activity and stability in water under neutral to basic conditions for some hydrogenation reactions [17]. Then, for some hydrogenation of biomass derived molecules, cobalt formulations are considered as a viable formulations for hydrogenation reactions. Metallic cobalt indeed present interesting activity for the C=O bonds, as already observed for the selective conversion of cinnamaldehyde to cinnamyl alcohol,[18-20] furfural to furfuryl alcohol,[17,21] HMF hydroconversion,[17,22] etc.