Green procedures for the selective aqueous-phase hydrogenation of biomass-derived levulinic acid to γ-valerolactone : innovative design for catalytic recycle and regeneration

This Thesis work was aimed at studying the catalytic upgrading of an important biomass derived platform chemical such as levulinic acid (LA). In particular, the hydrogenation/dehydration of levulinic acid (LA) to γ-valerolactone (GVL) was considered with a major focus on the implementation of original methods for the recovery and recycle of catalysts, both heterogeneous and homogeneous ones. Research activities were carried out within a cotutelle agreement between the University of Ca’ Foscari Venezia (Italy) and The University of Sydney (Australia). In Venice, a liquid triphase system made by an aqueous phase, an organic phase, and an ionic liquid was designed and applied for the conversion of LA to GVL. It was demonstrated that, operating at 100−150°C and 35 atm of H2, in the presence of either Ru/C or RuCl3 as catalysts, the use of the triphase system designed to match the investigated reaction allowed: i) to obtain up to quantitative conversions and 100% selectivity toward the desired product; ii) to recover the product by simple phase separation; and iii) to preserve the catalyst activity for in situ recycles without any loss of metal. Overall, the study proved the concept that a multiphasic catalytic system could remarkably improve the global sustainability of the investigated hydrogenation reaction, where a key step was the catalyst segregation in an IL phase and its recycle. In Sydney, the behavior of iron as a hydrogenation promoter of model organic compounds was explored. In particular, the recovery and recycle of iron was examined through electrochemical methods including both cyclic voltammetry and controlled potential electrolysis (CPE). The deposition of Fe0 was carried out in acidic aqueous solutions (pH 3) of FeSO4. Under such conditions, at 25 - 50°C, the aqueous phase hydrogenation of both cyclohexanone and LA in the presence of Fe0 and diluted acid was performed selectively to produce the expected products (cyclohexanol and GVL, respectively) on a practical laboratory scale (0.4 mmol, ~50 mg). Though, conversions must be further optimized. Based on the gravimetric analyses of deposits of Fe0, the measurement of the net Q charge involved in CPE experiments, and the GC-MS analysis of reaction mixture, preliminary results suggested that Fe0 acted as promoter/catalyst for the hydrogenation of the cyclohexanone and of the levulinic acid. Globally, although the investigation was far from being exhaustive, it was proved that the hydrogenation of carbonyl derivatives including the target of this study (i.e. LA), was feasible in aqueous solutions through an iron-based clean procedure.