Recently the use of lignocellulosic feedstock as source of fermentable sugars has been extensively investigated in order to develop biorefinery processes to produce liquid biofuels and bio-commodities. In particular, second generation biofuels have been proposed as the result of the conversion of sugars from residual biomass such as agriculture and food industry by-products. An example of liquid biofuel produced through sugar-based biorefinery is bio-butanol [1]. The possibility to produce butanol from low cost feedstocks, such as lignocellulosic food wastes, has been recently explored [2]. This process, as well as all the second generation biorefinery processes, asks for physical or chemical pretreatments of the lignocellulose (e.g., acid hydrolysis, alkaline wet oxidation and steam explosion) in order to mitigate the hindrance effect of lignin that limits the accessibility of carbohydrates to the hydrolytic enzymes. Despite the progress made in developing innovative biomass delignification processes [3], some drawbacks still exist that need to be faced. Further efforts are therefore needed in order to identify biomass pretreatments that are able to break up the lignocellulosic structure, to decrease the cellulose crystallinity, and at the same time to provide a high hemicellulose recovery while limiting the formation of inhibitory compounds [4]. Moreover, most of the pretreatment processes requires large water, solvents or energy consumption and are typically implemented at large scale as first operation in the biorefinery sites. More versatile processes that might be locally applied for distributed production and collection of lignocellulosic feedstock are required in order to develop 'intermediate' carbon vectors in the form of stable and ready-to-use resources. To this aim, the investigation of mild thermochemical processes such as torrefaction can offer wide opportunities. Torrefaction is a mild thermo-chemical treatment where biomass is heated in an inert environment up to a temperature ranging between 200 and 300 °C. It is characterized by low particle heating rate (< 50 °C/min) and a relatively long reactor residence time (5 -120 min) [5]. Even though, so far, torrefaction has been mostly adopted as a biomass pretreatment technology for thermochemical conversion pathways (i.e. combustion, pyrolysis and gasification), there are evidence in the pertinent literature [6] that torrefied feedstocks can be enzymatically hydrolyzed for lignocellulosic bioethanol production. Brachi et al. (2017) [7] have also found that torrefaction promote the conversion of crystalline cellulose into an amorphous state, the latter being more susceptible to the enzymatic hydrolysis than the former. Among the other effects, torrefaction improves the grindability of fibrous materials [5], which may reduce the energy demand for grinding the feedstock before hydrolysis, and turned out to be a suitable pretreatment to improve the fuel properties of lignin residues resulting from enzymatic hydrolysis [8]. In the light of the abovementioned benefits, the potential of combining biomass torrefaction with enzymatic hydrolysis for lignocellulosic butanol and isopropanol production has been preliminary investigated in the present paper by using coffee silverskin as a feedstock. In more details, thermogravimetric runs and lab-scale fixed bed torrefaction tests have been carried out to identify the process temperature, which allows minimizing the content of lignin in the torrefied solid, while preserving the original amount of cellulose and hemicellulose. The basic idea is to exploit the different thermal stability and reactivity that hemicellulose, cellulose and lignin exhibit because of their different composition and structure.