(632d) Hydrodeoxygenation of Phenolics at Solvent-Metal Interfaces: Enabling New Catalytic Pathways By Modifying the Reactive Hydrogen Species

Authors: 
Hensley, A., University of Toronto
Shangguan, J., University of Toronto
Bray, J., Washington State University
Gradiski, M., University of Toronto
Morris, R., University of Toronto
McEwen, J. S., Washington State University
Chin, Y. H., University of Toronto

Critical to the production of
carbon neutral biofuels and fine chemicals is the activation of strong polar
bonds―such as C-O bonds within biomass-based compounds―through selective
catalytic hydrodeoxygenation (HDO). Activation of such bonds can be markedly
affected by the chemical identity of solvents,[1,2] with polar protic solvents
such as water potentially ionizing hydrogen adatoms (H*) on transition metal
surfaces, forming protons (H+) and opening up new catalytic routes.[3]
Thus, we need to determine the chemical identity and the charge of reactive
hydrogen species (H* versus H+) when in contact with solvent molecules
and their respective roles on the elementary reaction pathways involved in
aromatic ring hydrogenation and C-O cleavage reactions.

Here, we establish the catalytic
requirements for breaking the strong aromaticity in phenolics (guaiacol,
phenol, catechol, etc.) and cleaving the C-O bonds using guaiacol as a model
compound, which contains the functional groups (aromaticity, C=C, C-O) found in
lignin monomers. As such, we anticipate that determining the elementary
reaction pathways for the HDO of guaiacol at the H2O-Ru interface will
give key insights into the broader effects of hydrogen charge and solvents on
the activation of strong polar bonds. To determine the underlying reaction
pathways, we present an integrated, collaborative effort between experiment and
theory that involves computational modeling, kinetic assessments, isotopic studies,
and nuclear magnetic resonance (NMR) characterization measurements. Significantly,
ongoing theoretical and experimental studies show that these elementary
pathways and solvent effects are not only applicable to guaiacol HDO at the H2O-Ru
interface but are also generalizable to the HDO of other phenolics.

At the H2O-Ru interface,
guaiacol adsorbs to the Ru surface through its aromatic ring and then rapidly
loses its hydroxyl hydrogen atom to the metal surface (Figure 1A), forming a surface
guaiaoxy species as the most abundant surface intermediate (MASI). Through this
interaction, the non-oxygen bonded ring carbons acquire a higher electron
density (-0.2 electrons). Kinetic experiments show that the C-OCH3
bond is preferentially cleaved after two hydrogen addition steps (i.e. TOR ~ [H*]2)
and H/D isotopic experiments combined with 13C and 1H NMR
analysis have further confirmed this by showing that H/D scrambling occurs at
specific carbon positions on the aromatic ring prior to C-O bond cleavage.[4] As
such, partial hydrogenation of the aromatic ring is required for breaking the
strong aromaticity of the MASI so as to weaken the strong C-O bonds. A comparison of the elementary barriers
for the first hydrogen addition step on the MASI with H* versus H+
shows that the attack of the negatively charged ring carbons by near surface H+
decreases the barrier for C-H formation by 0.58 eV. After breaking the ring
aromaticity with H+, the barriers for subsequent H* addition onto
ring carbons decrease by 0.36 eV and the C-O cleavage barrier decreases by 0.57
eV. This is consistent with the experimental results for C-O bond cleavage which
is ~30 times faster in water as compared to cyclohexane (Figure 1B).

The accelerated activation of the
strong polar C-O bond via the cooperative effect of H* and H+ at the
H2O-Ru interface presents a path forward for the design of
solvent-metal interfaces with high reactivity for C-O bond cleavage via tuning
the proton availability (e.g. surface work function, acidic surface sites) and
solvent stabilization of charged transition states (e.g. dielectric constant,
hydrogen bond strength). Phenol HDO at a H2O-Fe interface
demonstrates a similar solvent mediated proton/electron transfer effect in
breaking ring aromaticity and thereby lowering the barrier for C-O bond
cleavage.[1] Furthermore, the design of catalytic interfaces for the cleavage
of strong polar bonds in phenolics can be accelerated by Brønsted-Evans-Polanyi
(BEP) relations for elementary reactions on phenolics that capture both vapor
and liquid phase reactions (Figure 1C), and our work here provides the
ingredients necessary to efficiently implement such future research strategies.

 

[1]
Hensley, A.J.R.; Wang, Y.; Mei, D.; McEwen, J.-S. ACS Catal. 2018,
8, 2200-2208.

[2]
Hibbitts, D.D.; Loveless, B.T.; Neurock, M.; Iglesia, E. Angew. Chem. Int.
Ed.
2013, 52, 12273-12278.

[3]
P.S. Rice, Y. Mao, C. Guo, P. Hu, Phys. Chem. Chem. Phys. 2019, 21,
5932-5940.

[4]
J. Shangguan, N. Pfriem, Y.-C. Chin, J. Catal. 2019, 370, 186-199.

Figure 1. Activating C-O bonds at solvent-metal
interfaces via catalytic HDO. (A) Guaiacol―model biomass-based
compound―and H2 adsorb onto Ru in the presence of water and
guaiacol then converts to the most abundant surface intermediate (MASI).
Solvent stabilization of H+ versus H* changes the most abundant
reactive intermediate. The silver, black, red, white, and purple spheres in the
inserts represent Ru, C, O, H, and H+, respectively. (B) Solvent effects
result in a 30-fold increase in the guaiacol C-O cleavage turnover rate in
water versus cyclohexane (423 K, 50 mg Ru/C, 1 bar H2, 0.04 M guaiacol).
(C) BEP relation for C-H cleavage in phenolics on transition metal surfaces
shows that solvent-metal interfacial reactions can be predicted via vapor phase
relations.