Sustainable design of biomass valorization processes for the production of gamma valerolactone integrating wind energy and green hydrogen

Abstract

The transition to net zero emissions in the chemical and process industries relies on the integrated use of renewable raw materials (e.g. biomass) and energy sources. The objective of this work is to propose a methodology aimed at the sustainable integration of renewables within the framework of green chemistry processes, where lignocellulosic biomass waste is valorized to produce biochemicals. The novelty is in the sustainability analysis extended to the supply systems, which allows for an optimized integration of multiple renewable energy sources to meet energy and raw materials demands. A sustainability comparison among different configurations for the process and its supply systems was made using a Multiple-Criteria Decision Analysis that allows for a ranking based on economic, environmental, and safety impacts. A case study was developed to describe the potentiality of the proposed methodology, based on a waste lignocellulosic biomass valorization process for the production of gamma valerolactone. Several configurations were developed and compared. The best Levelized Cost of Product (i.e. 2.08 €/kg_GVL) was obtained when lignocellulosic biomass was used to also cover the heat demand of the process, the production of electricity occurs on-site using wind turbines, and green hydrogen was produced and stored on-site to manage peak-shavings and valley fillings. From the environmental standpoint, maximum impacts (2 kg_CO2eq/kg_GVL) result when conventional sources of energy were considered, i.e. natural gas for thermal supply and grid electricity. Looking at the safety aspects, hydrogen production and storage sections account for approximately 10% of the overall hazard profile of the process, while the remaining contribution comes from the gamma valerolactone production section. The robustness of the results was demonstrated using uncertainty and sensitivity assessment on the input parameters, based on a Monte-Carlo approach.