(562b) Computational Study on the Catalytic Reductive Depolymerization Mechanism of a ??O?4 Lignin Dimer Model Compound in Subcritical Methanol

Lignin is one of the three primary components of lignocellulosic biomass, which is generally composed of three phenylpropane units, i.e., p-hydroxyphenyl (H-type), guaiacyl (G-type), and syringyl (S-type) units which are interconnected by several ether and C–C linkages. Traditionally, lignin has been treated as a waste product in the pulp and paper industry, where it is burned to supply energy and recover pulping chemicals in the operation of paper mills. To utilize lignin more efficiently, reductive catalytic fractionation (RCF) has emerged as a promising way to achieve the conversion of lignin to value-added chemicals. This approach combines solvolytic extraction of lignin from holocellulose (cellulose and hemicellulose) and reductive stabilization of reactive intermediates, leading directly to a depolymerized lignin oil. Conventional RCF was carried out in batch reactors where biomass, a reducing catalyst, and a hydrogen donor are combined in polar protic solvents. However, it has been reported that the physical mixing of the biomass and catalyst complicates kinetics studies. Also, lignin extraction is solely dependent on the solvolytic conditions and does not require a redox-active catalyst. On the other hand, reductive stabilization of the unsaturated fragments is exclusively governed by the redox-active catalyst. Therefore, several reactor configurations have been applied to physically separate the solvolytic lignin extraction-depolymerization from the catalytic hydrogenolysis-stabilization step for RCF processes, including conducting solvolysis and hydrogenolysis as separate batch reactions or in tandem flow-through reactors. Polar-protic solvents such as methanol are usually used at high temperatures (up to 250 °C) in combination with a heterogeneous catalyst (e.g., Pt/C, Ru/C, or Pd/C), to obtain high-yield lignin oil that is rich in aromatic monomers with saturated side chains.

Computational studies can provide a mechanistic understanding of the reaction pathways and the effect of solvation in RCF. In our work, density functional theory (DFT) calculations were used to develop a mechanistic understanding of the underlying chemistry of two major processes in the RCF approach, solvolysis and stabilization. To better understand solvolysis, guaiacylglycerol-beta-guaiacyl ether (GGE) was employed as the model compound of lignin to understand the relative reactivity and chemical pathways that a β-O-4 bond will undergo in subcritical methanol, producing coniferyl alcohol and guaiacol. A key objective is to investigate the role of subcritical methanol in this reaction, which acts as a hydrogen donor, solvent, and reactant simultaneously. Both microsolvation and implicit solvation approaches have been used in this case to explore the effects of subcritical methanol. Concerted retro–ene and Maccoll elimination reactions (which are suggested to be the dominant pathway for typical pyrolysis temperatures between 500 and 600 °C) have been studied to evaluate the possibility of the solvent promoting these reactions in subcritical conditions (~230 °C). Our calculations show that the depolymerization of the GGE molecule through a retro-ene fragmentation reaction is kinetically and thermodynamically more favorable compared to Maccoll elimination reactions. The reactions start with the intramolecular hydrogen transfer, which is also the rate determining step. The free energy barriers of this process are 245 kJ/mol in the gas phase and 212 kJ/mol with the inclusion of implicit solvent. Even though the implicit solvent approach reduces the reaction barrier by 33 kJ/mol, the barrier remains high for a reaction to occur at a temperature of 500 K. However, a potential depolymerization mechanism is found using the microsolvation approach. We found that the reaction proceeds through a quinone intermediate, which is assisted by intermolecular hydrogen transfer from phenolic group of GGE molecule and two explicit methanol molecules. Then, one of these methanol molecules donates two hydrogens to activate the β-O-4 bond cleavage of the GGE molecule. This reaction pathway agrees well with a number of experimental observations including the retention of α and β hydrogen, the formation of formaldehyde in the reaction mixtures, and the significance of the phenolic group in the lignin model compound. It can be concluded from this result that the methanol molecule plays a vital role as hydrogen donor and acceptor that facilitate the depolymerization of lignin. Moreover, the calculations suggest that, in some cases, the implicit solvent might not be sufficient for modeling the reaction in the liquid phase and one or more explicit solvent molecules might have to be included.

In addition to the solvolysis reaction, we have also studied catalytic hydrodeoxygenation/hydrogenation reactions of coniferyl alcohol over the Pd(111), Pt(111), and Ru(0001) catalyst surfaces, a process called stabilization. Coniferyl alcohol represents the reaction product from solvolysis. In this case, the solvent effect on the rate-determining step has been taken into account using our hybrid QM/MM method, called Explicit Solvation for Metal surfaces (eSMS). Since it captures the true molecular nature of the solvent molecules, eSMS can reliably predict solvation effects on metal surfaces. To conclude, by studying solvolysis and stabilization of a model reactant such as GGE, we hope to be able to provide a comprehensive picture of the lignin conversion process during RCF with the goal of optimization both the reaction conditions and the catalyst.