(307c) Comparison of Thermal- and Electro-Catalytic Conversion of Biomass-Derived Oxygenates

Hydrogenation and hydrogenolysis reaction studies have been carried out traditionally under high temperature and pressure in the presence of hydrogen gas fed into the reactor from an external source. Recently, electrochemical hydrogenation and hydrogenolysis has attracted a lot of attention since reactions can be carried in mild conditions (room temperature and ambient pressure) using water as the hydrogen source. Electrocatalysis can be combined with renewable energy resources (e.g. biomass and solar cells) to yield a sustainable path to producing valuable chemicals. An additional advantage of electrocatalysis is the added degree of freedom (i.e. electrode potential) that can be controlled to tune reaction pathways [1].

Despite the key advantages of electrocatalysis in the upgrading of biomass-derived oxygenates, it remains to be explored the differences in selectivities between thermal- and electro-catalytic hydrogenation and hydrogenolysis of polar oxygenated compounds. Here we report selective conversion of biomass-derived small oxygenates (C5 and C6 sugars) in both thermal- and electro-catalysis. Specifically, by combining experimental and first-principles density functional theory (DFT) calculations, we show that the reaction selectivities are distinct under these two conditions. That is, hydrogenation of furfural to form furfuryl alcohol is the major product in an aqueous-phase thermal reaction while 2-methyfuran is the primary and major product in electrocatalysis under more negative potentials. We further show that this difference in selectivities results from varied fractional charges transferred at the transition states along two reaction pathways.

Materials and Methods

Electrochemical reactions were carried out with an Agilent B2902A Precision Source/Measure Unit (SMU). Data was collected vs. Ag/AgCl reference. Copper foil (thickness 0.05 mm, Carolina) was used as the working electrode. Prior to each experiment, the surface was mechanically polished until no discoloration was visible then chemically polished using 0.1 M HCl solution to remove any oxides. Platinum foil was used as the counter electrode. The proton exchange membrane separating the two chambers of the elechtrochemical cell was immersed in 0.5 M H₂SO₄ for 24 hours prior to experiment. The liquid products from the anode and cathode chambers were extracted using Diethylether with a 1:1 ratio. 5 mL of pure acetonitrile was used to trap volatile products. The liquid product was filtered and analyzed by gas chromatography.

The plane-wave density functional theory calculations were carried out using Vienna Ab initio simulation package (VASP). The Perdew-Burke-Ernzerhof generalized gradient approximation exchange-correlation potential (PBE-GGA) was used. The model system is a supercell built based on four (4x4) repeated slabs of Cu and Ni (111). Explicit water molecules were added to model the Cu/water interface. The method is similar to our previous study of the Pd/water interface [2]. Reaction energies of all elementary steps were calculated using the model of computational hydrogen electrode. Electrochemical barriers at different electrode potentials were calculated based on fractional charges at the transition states.

Results and Discussion

The evolution of product distribution shown in Figure 1a as a function of time suggests that this reaction is of the form A → B → C → D, with B representing the intermediate furfuryl alcohol which undergoes a sequential ring rearrangement reaction in the presence of water and hydrogen at higher temperatures to form 2-cyclopenten-1-one and cyclopentanone. Products evolution as a function of electrode potentials in Figure 1b shows that, when a (H⁺, e^-) pair drives the reaction, hydrodeoxygenation of furfural to form 2-methylfuran is the preferred route [3]. We find in DFT calculations that, in a neutral aqueous solution, the hydrogenation of the carbon in the aldehyde group followed by the oxygen is the path of minimum energy to produce furfuryl alcohol (Figure 1c). However, in electrocatalysis, the dehydration path of furfural through the double hydrogenation of the oxygen (proton-assisted) followed by the hydrogenation of the carbon to make 2-methylfuran is preferred. The electrode potential controls the free energy of the dehydration step. A bias of -0.55V can be applied to alter the reaction path from hydrogenation (furfuryl alcohol route) to dehydration (2-methylfuran route).

Figure1 a) Product distribution in an aqueous-phase thermal catalytic reaction of furfural on 10% Cu/SiO₂ (P_H2 = 300 psi), b) Product distribution in a liquid-phase electro-catalytic reaction on a Cu foil (0.5 M H₂SO₄), c) DFT-calculated free energies in liquid water, d) Comparison of reaction pathways of furfural in thermal- and electro-catalytic reactions.

References

1. Z. W.Seh, J.Kibsgaard, C. F. Dickens, I. Chorkendorff, J. K. Nørskov, T. F. Jaramillo. Science 355, eaad4998 (2017).
2. Zhao Z, Bababrik R, Wang B, Resasco DE. Nature Catalysis. 2019Jan;2(5):431–6.
3. Bababrik, R., Santhanaraj, D., Resasco, D.E. J Appl Electrochem 51, 19–26 (2021).