Optimal design of experiments for the identification of kinetic models of 5-hydroxymethylfurfural hydrogenation

The increasing issues about the use of fossil raw materials advocate for an expanding utilization of biomass-based fuels and chemicals. Among the bio-based furan compounds, the 5-hydroxymethylfurfural (HMF) has received considerable attention in the chemical industry since it can be hydrogenated to 2,5-dimethylfuran (DMF) that is a valuable alternative fuel. The identification of a suitable kinetic model, where all the non-measurable kinetic parameters can be reliably estimated, is crucial to pursue the process optimization. In this work, the kinetic models currently available in literature for the HMF hydrogenation process are investigated to underline their strengths and weaknesses using a sensitivity-based identifiability analysis. The application of identifiability analysis techniques allows to define a set of fully identifiable kinetic models to be used for statistically reliable predictions. Furthermore, the use of design of experiments techniques leads to the characterization of design space regions that maximize the quality of the statistics related to the different estimates.