Mapping reactivities of aromatic models with a lignin disassembly catalyst. Steps toward controlling product selectivity

Article abstract of DOI:10.1039/c5cy01555c

Copper-doped porous metal oxides catalyze the one-pot disassembly of biomass-derived lignin via C-O bond hydrogenolysis and hydrodeoxygenation in supercritical methanol. This catalytic system cleanly converts lignin as well as lignocellulose composites, such as sawdust, to organic liquids with little or no formation of intractable tars or chars. However, this catalyst based on Earth-abundant components also catalyzes less desirable aromatic ring hydrogenations and various methylations that contribute to the diversity of products. In this context, we undertook a quantitative experimental and computational evaluation of model reactions relevant to the reductive disassembly of lignin by this catalyst system in order to determine quantitatively the rates of desirable and less desirable chemical steps that define the overall product selectivities. Global fitting analysis methods were used to map the temporal evolution of key intermediates and products and to elucidate networks that provide guidelines regarding the eventual fates of reactive intermediates in this catalysis system. Phenolic compounds display multiple reaction pathways, but substrates such as benzene, toluene, and alkyl- and alkoxy-substituted aromatics are considerably more stable under these conditions. These results indicate that modifying this catalytic system in a way that controls and channels the reactivity of phenolic intermediates should improve selectivity toward producing valuable aromatic chemicals from biomass-derived lignin. To this end we demonstrate that the O-methylating agent dimethyl carbonate can intercept the phenol intermediate formed from hydrogenolysis of the model compound benzyl phenyl ether. Trapping the phenol as anisole thus gave much higher selectivity towards aromatic products.

Full text of DOI:10.1039/c5cy01555c

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Mapping reactivities of aromatic models with a
lignin disassembly catalyst. Steps toward
controlling product selectivity†
Cite this: DOI: 10.1039/c5cy01555c
Christopher M. Bernt,‡a Giovanni Bottari,‡b Jacob A. Barrett,a Susannah L. Scott,a
b
a
*
*
Katalin Barta and Peter C. Ford
Copper-doped porous metal oxides catalyze the one-pot disassembly of biomass-derived lignin via C–O
bond hydrogenolysis and hydrodeoxygenation in supercritical methanol. This catalytic system cleanly con-
verts lignin as well as lignocellulose composites, such as sawdust, to organic liquids with little or no forma-
tion of intractable tars or chars. However, this catalyst based on Earth-abundant components also catalyzes
less desirable aromatic ring hydrogenations and various methylations that contribute to the diversity of
products. In this context, we undertook a quantitative experimental and computational evaluation of model
reactions relevant to the reductive disassembly of lignin by this catalyst system in order to determine quan-
titatively the rates of desirable and less desirable chemical steps that define the overall product selectivities.
Global fitting analysis methods were used to map the temporal evolution of key intermediates and prod-
ucts and to elucidate networks that provide guidelines regarding the eventual fates of reactive intermedi-
ates in this catalysis system. Phenolic compounds display multiple reaction pathways, but substrates such
as benzene, toluene, and alkyl- and alkoxy-substituted aromatics are considerably more stable under these
conditions. These results indicate that modifying this catalytic system in a way that controls and channels
the reactivity of phenolic intermediates should improve selectivity toward producing valuable aromatic
chemicals from biomass-derived lignin. To this end we demonstrate that the O-methylating agent dimethyl
carbonate can intercept the phenol intermediate formed from hydrogenolysis of the model compound
benzyl phenyl ether. Trapping the phenol as anisole thus gave much higher selectivity towards aromatic
products.
Received 15th September 2015,
Accepted 25th November 2015
DOI: 10.1039/c5cy01555c
www.rsc.org/catalysis
by free radical initiated polymerization of aromatic mono-
lignols (principally p-coumaryl, coniferyl and sinapyl alcohols)
Introduction
Lignocellulose, the principal component of woody biomass, is
a non-edible and sustainable feedstock that has the potential
in ratios that depend upon the source.8 The principal link-
ages are aryl–ether bonds, although the monolignols are occa-
sionally crosslinked by C–C bonds. Notably, lignin has the
highest relative carbon content of the major lignocellulose com-
ponents and represents a greatly underdeveloped renewable
to be
a renewable resource for producing commodity
chemicals and fuels.1–7 However, practical applications
require an integrated approach to maximize atom and
electron economy in extracting the energy and material
contained therein. Lignocellulose is about 70–80% carbohy-
drates, with the balance largely being lignin. The latter is a
complex, heterogeneous polymer (Fig. 1) formed biologically
a Department of Chemistry and Biochemistry and the Center for the Sustainable
Use of Renewable Feedstocks, University of California, Santa Barbara, Santa
Barbara, California 93106-9510, USA. E-mail: ford@chem.ucsb.edu
b Stratingh Institute for Chemistry, University of Groningen, Nijenborgh 4, 9747
AG Groningen, The Netherlands. E-mail: k.barta@rug.nl
† Electronic supplementary information (ESI) available: Description of the algo-
rithms used to convert quantitative GC data to product amounts and tables
summarizing analysis of the temporal evolution of products from reactions of
individual substrates over the Cu-doped PMO catalyst. See DOI: 10.1039/
c5cy01555c
Fig.
1 A representative structure for a lignin fragment based on
‡ These authors contributed equally to this work.
structural data by Sannigrahi et al.7 and Stewart et al.9
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
resource for producing aromatic chemicals.1–3 However, bio-
refineries for “2nd generation” ethanol fuel production typically
focus on the carbohydrates, and the lignin fraction, owing to its
chemical recalcitrance and heterogeneity, is mostly burned to
produce low-grade heat.1 This is also the case with the produc-
tion of paper goods from lignocellulose. Thus, designing selec-
tive pathways to convert lignin to chemicals and/or liquid fuels
should add considerable value to biomass conversion schemes.
It is estimated1,10 that 62 Mt residual lignin will be gener-
ated annually by 2022 from US-mandated ethanol production
alone. This is on the same scale as BTX (benzene, toluene
and xylenes) production from fossilized carbon.11 While BTX
could be a drop-in target for biomass conversion, lignin dis-
assembly will also lead to functionalized aromatics that have
the potential to serve as the building blocks for new mate-
rials.12 Effective utilization of lignin remains a major intellec-
tual challenge1–3 that is drawing increasing interest from the
chemical community.13–20
catalyst in supercritical ethanol.30 However, it is clear that
improved selectivity to generate product streams composed
of a limited number of aromatic chemicals would signifi-
cantly enhance the value of lignin as a renewable feedstock.
In these contexts, we describe here a reactivity study of
various small molecules (Fig. 2) that have structures/func-
tional groups representative of the key reactive intermediates
anticipated during lignin disassembly by Cu-PMO catalysts in
sc-MeOH. Global kinetics analysis of the resulting reaction
pathways allows mapping the reaction networks for the
expected intermediates. Such networks can provide valuable
insight into the underlying disassembly mechanisms and
side-reactions that strongly influence the selectivity of this
catalytic process as well as guidelines for intercepting the
responsible intermediates, hence improving selectivity. Note
that Gates and coworkers31,32 have used a different numerical
method to generate analogous reaction networks for the plat-
inum catalysed reactions of H2 with vaporized models for
pyrolysis-derived bio-oils.6,33–35
Our laboratories previously demonstrated the clean reduc-
tive disassembly of organosolv lignin to organic liquids with-
out formation of intractable chars or tars.21 This process is
heterogeneously catalysed by a copper-doped porous metal
Experimental section
Materials
oxide (Cu20PMO) that is prepared by calcining
a 3 : 1
Mg2+ : Al3+ hydrotalcite in which 20% of the Mg2+ had been
replaced by Cu2+. The reaction is carried out in a batch reac-
tor in super-critical methanol (sc-MeOH). The same catalytic
system also liquefies cellulose and, remarkably, disassembles
lignocellulosic materials such as sawdust or wood chips to
organic liquids with little or no char formation.22,23 A rela-
tively complex mixture of aliphatic alcohols and ethers is
obtained from the carbohydrate fraction, while a similarly
complex mixture of propylcyclohexanol derivatives is gener-
ated from the lignin.21,22 The latter products are attributed to
the hydrogenolysis of various phenyl ethers, with competing
and subsequent hydrodeoxygenation (HDO) and hydrogena-
tion of alkenes and aromatic rings. The methanol medium
provides the reducing equivalents21–24 via alcohol
reforming25 and the water-gas shift reaction,26 both catalysed
by the Cu-PMO to generate a gas phase consisting of H2 and
CO plus some CO2 and small amounts of methane. Less
desirable side-reactions in this medium are ring methyla-
tions. These processes and the composition of the gas phase
products were described in greater detail previously.21–24
The importance of the copper in these catalysts was dem-
onstrated in our earlier studies with pine sawdust.22 While
this substrate was readily converted to organic liquids by the
Cu20PMO in sc-MeOH, the products are largely char and
unreacted biomass under similar conditions with a Mg/Al
hydrotalcite-derived PMO catalyst not containing copper.
Analogous results were found when no catalyst was added.
Cu-doped PMO catalysts were subsequently shown to be
effective in the H2 hydrogenation of candlenut lignin to aromatic
products27 and for the reduction of 5-hydroxymethylfurfural
to tetrahydrofuran derivatives.28,29 A related study recently
Methanol was purchased from Fischer and dried over molec-
ular sieves. Model compounds used in reactivity studies were
purchased from Sigma and used without further purification.
Copper-doped porous metal oxide catalysts were prepared
Fig. 2 Aromatic model compounds studied representing phenyl ether
linkages and types of functional groups anticipated as intermediates in
reductive lignin disassembly.
demonstrated disassembly of organosolv lignin using
a
copper-doped magnesium–aluminum mixed metal oxide
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
from 3/1 Mg2+/Al3+ hydrotalcites with 20 mol% of the Mg2+
replaced by Cu2+. Prior to use, the catalysts were calcined at
460 °C for 24 h.21,22,24 Un-doped PMO catalysts used in con-
trol studies were prepared by calcination of 3/1 Mg2+/Al3+
hydrotalcite purchased from Sigma.
of 3.0 mL min−1. The column temperature program started
with a hold at 60 °C for 2 min followed by a 25 °C min−1
ramp up to 200 °C where the temperature was held for 4
min. The corresponding MS program had an ion source tem-
perature at 250 °C, interface temperature at 230 °C and
recorded from 2.0 to 11.6 min during the GC program.
Reactivity studies
Quantitative measurements
Catalysis studies were conducted in high pressure, stainless
steel mini-autoclave batch reactors consisting of 3/4 inch
Swagelok unions capped by two 3/4 inch Swagelok plugs (see
Fig. 4 of ref. 23). The total volume of these mini-reactors is
10 mL. Typically, catalyst samples (50 or 100 mg), the organic
substrate (1 mmol) and the internal reference decane (20 μL)
plus dry methanol (3 mL) were sealed in the mini-reactor
using a torque wrench. For reactions with dimethyl carbon-
ate, 1 mL of the methanol was replaced with DMC. Catalysis
runs were conducted by placing a set of these mini-reactors
(typically 4 to 8) with identical mixtures into a heating block
mounted in a furnace set at the chosen temperature (typically
310 °C).
GC-FID measurements were conducted on an Agilent/HP
6890N (G1530N) gas chromatograph equipped with a flame
ionization detector. This GC-FID instrument was operated
with two different columns. When used with a 30 m × 0.25
mm Agilent DB-1 + DG column, with a 0.25 μm dimethyl-
polysiloxane lining and a guard column, 0.5–2 μL samples
were injected at a temperature of 225 °C in split mode with a
200 : 1 ratio. With this column, the GC program was run in
pressure flow control mode at 104.8 kPa, 276 mL min−1 total
flow, 1.4 mL min−1 column flow, a linear velocity of 32 cm s−1
and a purge flow of 3.0 mL min−1. The temperature program
started with a hold at 50 °C for 2 min followed by a 25 °C
min−1 ramp up to 200 °C where the temperature was held for
2 min. When used with a 30 m × 0.25 mm Agilent DB-5 col-
umn with a 0.25 μm (5%-phenyl)-methylpolysiloxane lining,
0.5–2 μL samples were injected at a temperature of 250 °C in
split mode with a 20 : 1 ratio. With this column, the GC pro-
gram was run in pressure flow control mode at 68.1 kPa, 27.6
mL min−1 total flow, and a 0.70 mL min−1 column flow. The
FID detector was set at 250 °C with 30 mL min−1 H2 and 350
mL min−1 air flow. The temperature program started with a
hold at 45 °C for 6 min followed by a 4 °C min−1 ramp up to
175 °C where the temperature was held for 10 min.
Further quantitative measurements were performed on a
Hewlett Packard 5890 GC-MS-FID. The GC is equipped with a
60 × 0.25 mm i.d. and 0.25 μm Restek RTX-1701 film capil-
lary column and a 1 : 1 split ratio to the MSD and FID was
set. Temperature of injector and detector were set at 250 °C
and 285 °C, respectively. The temperature program starts
from 40 °C (10 min) and is then increased up to 250 °C with
a heating rate of 10 °C min−1.
The data reported for individual products were drawn
from these quantitative GC-MS or GC-FID experiments and
converted to μmoles using stand curves or algorithms set
forth by Scanlon and Willis as described in the ESI.†36
After a designated time interval (0.5–18 h), a reactor was
removed from the oven and quenched by rapid cooling in a
room temperature water bath or ice bath. The built-up gases
(which analyze as a mixture of H2, CO, CO2 and some meth-
ane)22 were then released, and the remaining contents of the
vessel were transferred to a 15 mL, disposable centrifuge tube
(Warning: the high pressures usually produced during a
mini-reactor experiment require extreme caution in han-
dling). The reactor was then rinsed with MeOH (4 mL), and
combined with the original contents. The tube was
centrifuged at 6500 rpm for 15 min to separate the liquid
and solid fractions. The liquid contents were collected from
the tube, and the remaining solid was suspended in another
3 mL volume of MeOH, which was centrifuged to separate liq-
uid and solid. The additional MeOH was intended as a wash
to extract residually adsorbed substrate and soluble products
(the earlier studies with organosolv lignin and lignocellulose
composites as substrates21,22 showed negligible or no gain in
the mass of the recovered catalyst after such treatment indi-
cating little organic residue, including chars, on the
Cu20PMO). The two liquid fractions were combined for analy-
sis by GC-MS and/or GC-FID techniques. In this way, the tem-
poral evolution of each component was determined, with
each time point representing a semi-independent experiment.
Product analysis was performed primarily by gas chroma-
tography with flame ionization or mass spectral detection.
GC-MS measurements were taken on a Shimadzu GC-2010
gas chromatography coupled to a Shimadzu GCMS-QP2010
mass spectrometer. Samples of 1 μL volume were injected at
a temperature of 225 °C in split mode with a 200 : 1 ratio.
This instrument was equipped with a 30 m × 0.25 mm
Agilent DB-1 column with a 0.25 μm dimethylpolysiloxane
lining. The GC program was run in pressure flow control
mode at 40 kPa, 162.3 mL min−1 total flow, 0.79 mL min−1
column flow, a linear velocity of 32.5 cm s−1 and a purge flow
Global fitting and kinetics analysis
The initial design of reaction networks was based on the
most parsimonious application of previously observed PMO
catalysed reactivities: hydrogenation, hydrodeoxygenation,
C–O bond hydrogenolysis and methylation. Reactions were
assumed to be first order with respect to the substrate reac-
tant at each step. Refinements to the model were based on
the results of global fitting to the proposed models. Global
fitting was performed using DynaFit version 4.05.087 soft-
ware on a desktop computer (BioKin Ltd.).37
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Results and discussion
The fits described in this study are sufficient to explain the
defining trends in the observed data. A more detailed discus-
sion on the descriptive statistics of these fits can be found in
the ESI.†
Once the aryl–ether bonds are cleaved by hydrogenolysis or
solvolysis during lignin disassembly, a series of molecular
intermediates are generated.21 Expected cleavage products
are phenolics with various substitution patterns on the aro-
matic ring including methoxy and oxidized aliphatic groups.
In order to channel these intermediates into desirable and
selective product streams, it is necessary to evaluate their
reactivities under the conditions relevant to catalytic depoly-
merization. The simple models representative of those moie-
ties (Fig. 2) include mono-oxygenated (phenol and anisoles)
and di-oxygenated (guaiacol and veratrol) aromatics that are
common lignin motifs and disassembly intermediates, sev-
eral other typical aromatic types (toluene, benzyl alcohol, sty-
rene and acetophenone), as well as products of aromatic
hydrogenation (cyclohexanol and cyclohexanone). Several di-
aromatic model compounds were also investigated in order
to probe the rates of the sequential processes following aro-
matic ether hydrogenolysis.
The results of these kinetics studies are used to develop
quantitative reaction networks that show how product distri-
butions are defined by the competing steps catalysed under
the conditions of lignin disassembly. Reactions were carried
out in methanol (3 mL), which becomes supercritical at reac-
tion temperatures >240 °C in the 10 mL mini-reactors. In
most cases, these batch reactors contained substrate (1.0
mmol), Cu20PMO catalyst (100 mg) and MeOH (3.0 mL). Dec-
ane (20 μL) was also added as an internal standard for quan-
tification. A typical run involved 7–8 of these mini-reactors
with identical quantities of catalyst, substrate and solvent
that were heated together in an oven at the defined tempera-
ture (typically, 310 °C). At different time intervals, reactors
quenched to room temperature in a water bath, and the
products were analysed using GC-MS fragmentation patterns
for identification and GC-FID and GC-MS integrated areas for
quantification.
Reactions of aromatic alcohols phenol and guaiacol
The time course of major products from phenol over an 18 h
period is shown in Fig. 3 and summarized in ESI† Table S-1.
After just 1 h at 310 °C, roughly half the phenol was con-
sumed. After 3 h, this increased to 92%. The major products
at this time were methylcyclohexanol (25%), cyclohexanol
(28%), anisole (9%) and 2-methylphenol (7%). Lesser prod-
ucts included dimethylphenols (3%), dimethylcyclohexanols
(3%), 2-methylanisole (2%), 2-methylcyclohexanone (1%),
cyclohexanone (1%), methoxycyclohexane, methylcyclo-hex-
ane, cyclohexane and others. For reaction times ≥9 h, the
major product(s) were methylated cyclohexanols (MCHs),
while anisole was the most abundant aromatic product. These
products suggest that phenol is depleted via three primary reac-
tion channels: (i) reduction to cyclohexanol, (ii) methylation
of the aromatic ring to give cresols (methylphenols) and (iii)
methylation at oxygen to form anisole. Subsequent reactions
convert the primary product cyclohexanol to MCHs
(Scheme 1). Ring methylation is a common reaction for phe-
nols,38 and was shown to be a major pathway in earlier stud-
ies with the model compound dihydrobenzofuran (DHBF)
under comparable conditions.24
The reactivity of phenol under analogous conditions but
with a PMO catalyst derived by calcining Mg/Al (3/1) hydro-
talcite without copper was nearly two orders of magnitude
slower. After 18 h, GC analysis showed that 68% of the phe-
nol was still present, the remainder being anisole (12%), cre-
sol (2%) some unidentified lower retention time (RT) species
Under this catalytic system the solvent contributes to the
product mass by both methylation and hydrogenation so it is
more appropriate in this case to consider the molar material
balance of substrate derived products. In the present study
the material balance over the course of catalysis was greater
than 90% for most substrates, including the less reactive
dimer diphenyl ether. The experiments with lower molar bal-
ance were those with phenolic substrates and those that lead
to phenolics as major intermediates. In those cases, the prod-
uct spread gave numerous small peaks that could be identi-
fied in most cases but were too small to quantify reliably, so
lower molar balances are largely attributed to product prolif-
eration. Details for specific substrates are included in the
ESI.† In the kinetics analysis, unaccounted products were
treated as undefined product sinks. The temporal data thus
obtained and summarized in ESI† tables were evaluated by
global kinetics analysis using the DynaFit software37 to give
reactivity networks describing the interconnected catalytic
pathways of different substrates and expected intermediates.
Fig. 3 Temporal evolution of products formed from phenol (1 mmol)
in a batch reactor containing Cu20PMO in sc-MeOH (T = 310 °C).
Inset: Sum of the other products not shown in parent graph (see ESI†).
Material balance 78% after 6 h and 75% after 18 h.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
Scheme 1 Reaction network for phenol: (1 mmol substrate, 100 mg
Cu20PMO, 3 mL MeOH, 310 °C). Observed rate constants are shown
for each step, in units of h−1. Dashed lines indicate conversion to
unidentified compounds. Asterisks indicate rate constants fixed based
on independent investigation of reaction products.
Scheme 2 Reaction network for guaiacol. See Scheme 1 for reaction
details.
thought to be monomers (<1%) and higher RT products that
were dimeric (5%) (see ESI† Table S-1b). These products are
consistent with the findings of Crocella et al.38
Global kinetics analysis of the temporal profiles for phe-
nol reactions resulted in the reactivity network in Scheme 1.
As will be evident below, these results show that phenol is
particularly susceptible to aromatic hydrogenation under the
catalytic conditions.
Lercher et al.40 Scheme 2 illustrates the key steps for the sub-
strate guaiacol, deduced by the kinetics analysis.
Alkoxy aromatics
As shown in Scheme 1, there are two likely routes to methyl-
cyclohexanol from primary phenolic products: methylation of
cyclohexanol and hydrogenation of cresol. To verify the rates of
these secondary reactions, separate sets of experiments were
carried out for cyclohexanol and for cresol. The results of the
cyclohexanol experiment are described below, and the detailed
results with cresol can be found in the ESI† (SI Table S-2). In
the case of cresol, more than half (51%) was consumed after
four hours. Of the total mixture after 4 h, dimethylphenol
(xylenol) was the major product (33%) followed by methyl-
cyclohexanol (6%) and methylmethoxybenzene (5%). These
results indicate that ring methylation significantly slows the
hydrogenation of the phenolic derivative. Thus, the principal
products from the long-term reaction of phenol under the
described conditions are MCHs formed by methylation of the
cyclohexanolIJs) resulting from phenol hydrogenation.
In analogy to phenol, guaiacol (2-methoxyphenol) displayed
more complicated reaction sequences under these conditions.
The most important initial steps were methylation of the phe-
nolic oxygen to give veratrol (1,2-dimethoxybenzene), and
hydrogenolysis of the Caryl–OCH3 bond to give phenol. After 3 h,
guaiacol was largely (92%) consumed, with the major products
being veratrol (18%), cyclohexanol (15%), cyclopentylmethanol
(14%) and phenol (11%). Lesser products that are likely derived
from those formed initially are cyclohexanediols (11%), methyl-
cyclohexanols (5%), 2-methoxy-4-methylphenol (methyl anisole)
(5%), dimethylcyclohexanol (4%), dimethylphenols (2%),
2-methoxycyclo-hexanone (3%), anisole (2%) as well as others
(see ESI† Table S-3). Although not observed as an intermediate
by GC-MS analysis, catechol is a likely intermediate species,
given the appearance of both 1,2-cyclohexanediol and cyclo-
pentylmethanol in the final reaction mixture consistent with
mechanisms proposed by Deutsch and Shanks,39 as well as by
Both anisole (methoxybenzene) and ethoxybenzene proved to
be dramatically less reactive than phenol or guaiacol under
typical catalysis conditions. Notably, ring hydrogenations
were observed to a much smaller extent than with the pheno-
lics, and the predominant pathway was aromatic ring HDO.
For anisole, only 5% of the substrate was consumed in the
first 3 h. Notably, the primary product (by far) was benzene,
formed by Caryl–OMe bond hydrogenolysis. The others (cyclo-
hexanol, 2-methyl-cyclohexanol and methylcyclohexane)
apparently result from CH3–O bond hydrogenolysis followed
by reactions of the intermediate phenol. At longer reaction
times (6 h), a modest amount of the ring-hydrogenated
methoxy-cyclohexane (∼2%) was also observed (see ESI†
Table S-4).
Similarly, ethoxybenzene showed preferential hydro-
genolysis of the Caryl–OEt bond, affording the HDO product
benzene as the major product, and a low rate of ring hydroge-
nation (3% after 6 h). After 6 h, only 12% of this substrate
was consumed. The products were benzene (6%), ethoxy-
cyclohexane (3%), cyclohexanol (1%), 2-methyl-cyclohexanol
(0.7%) and several unknowns (see ESI† Table S-5).
Veratrol (1,2-dimethoxybenzene) proved to be considerably
more reactive than anisole; however, the principal pathway is
aromatic HDO to give anisole and not ring hydrogenation.
After 3 h, 52% was consumed; after 12 h, this rose to 97%.
The predominant product was anisole (40% after 3 h), appar-
ently formed via Caryl–OMe bond hydrogenolysis. The anisole
yield peaked at ∼12 h then diminished, very likely owing to
slow hydrogenolysis to benzene and hydrogenation to
methoxycyclohexane. Other products after 3 h included cyclo-
hexanol (3%), 2-methylcyclohexanol (3%), benzene (2%),
dimethyl-cyclohexanol (1%), methoxycyclohexane (1%) and
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
cyclohexane (0.9%) (see ESI† Table S-6). Direct hydrogenation
to 1,2-dimethoxycyclohexane was not observed.
DPE proved to be more reactive than anisole or ethoxy-
benzene, but much less so than phenol, guaiacol or veratrol.
After 3 h, 23% was consumed. The first hydrogenolysis step
of DPE would give equal amounts of benzene and phenol.
Accordingly, benzene was 48% and phenol 4% of the break-
down products. The other products, cyclohexanol (28%),
2-methylcyclohexanol (15%) and much smaller quantities of
anisole (2%), cyclohexane (1%) and dimethylcyclohexanol
(1%), can largely be attributed to subsequent reactions of the
phenol intermediate (Scheme 3).
Simpler aromatics and those with pendant functional groups
The oxygen-free aromatics benzene and toluene displayed
very little reactivity under the conditions effective for catalytic
lignin disassembly. For example, when toluene was the sub-
strate, 99% remained after 3 h at 310 °C. Methylcyclohexane
(1%) was the only product detected. Benzene was similarly
unreactive (2% conversion to cyclohexane after 3 h).
BPE proved much more reactive. With only 50 mg catalyst,
this substrate was about 85% consumed after only 1 h, and
none was left after 3 h. Carbon–oxygen bond hydrogenolysis of
BPE should lead either to benzene plus benzyl alcohol or to tol-
uene plus phenol, depending upon which C–O bond is cleaved.
The data summarized in Fig. 4 (see ESI† Table S-7) show
that toluene is formed immediately in nearly stoichiometric
quantities, while the other principal (initial) product is phenol.
Thus, the predominant first step is Cbenzyl–O hydrogenolysis. As
expected, the phenol is consumed by subsequent reactions
(Scheme 4) giving (mostly) non-aromatic products.
(1)
In contrast, benzyl alcohol proved to be quite reactive (full
conversion after 3 h), but primarily toward HDO, as illus-
trated in eqn (1). Toluene was the principal product (>98%),
but small amounts of methylcyclohexane and cyclo-
hexylmethanol were also found. While traces of an
unidentified product were also seen, ring methylation clearly
plays a very minor role and hydrodeoxygenation of the ben-
zylic OH is the main pathway, in line with the earlier
findings.39
As with phenol, the reactivity of benzyl phenyl ether under
analogous conditions over 3 : 1 Mg/Al PMO proved to be con-
siderably less than over Cu20PMO (see ESI† Table S-7b).
Styrene is also very reactive, but solely toward vinyl hydro-
genation to give ethyl benzene selectively (>99% after 3 h).
Traces (<1%) of ethylcyclohexane were found, but no ring
methylation products were detected.
Scheme 3 Observed pathway for diphenyl ether (DPE).
(2)
Under analogous conditions, acetophenone (acetylbenzene)
also gave full conversion after 3 h; the primary pathway was
the formation of ethylbenzene (82%) (eqn (2)), which likely
took place via ketone hydrogenation to 1-phenylethanol,
followed either by direct hydrogenolysis of the C–OH bond or
by dehydration to styrene, then hydrogenation. Another signifi-
cant product was n-propylbenzene (16%), presumably the
result of acetyl group methylation (perhaps via the enol isomer)
followed by HDO. Interestingly, sec-butylbenzene was also iden-
tified as a minor (0.5%) product.§
Model compounds with two aromatic rings
Three such compounds were investigated: diphenyl ether
(DPE), benzyl phenyl ether (BPE) and 2-phenoxy-1-
phenylethane-1-one (1).
Fig. 4 Temporal evolution of products during Cu20PMO-catalyzed
reactions of benzyl phenyl ether (ca. 1 mmol). Total material balance
was 82% after 18 h.
§ Various pathways can be envisioned for HDO depending on the substrate,
including direct hydrogenolysis of C–OR bonds and dehydration followed by
hydrogenation of the resulting un-saturated species.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
Subsequent hydrogenolysis of 2 would first give 1-phenyl
ethanol plus phenol, and rapid HDO of the former would
give ethylbenzene. Notably, ethylbenzene (600 mmol) repre-
sents the most plentiful product from 1 under these condi-
tions (ESI† Table S-8), but it is notable that both pathway (a)
and (b) predict this product given the expected reactivities of
the respectively proposed 1-phenylethanol and acetophenone
intermediates. The substantial quantities of 2-methylcyclo-
hexanol (350 μmol) and cyclohexanol (250 μmol) products
point toward phenol as a reactive intermediate.
Scheme 4 Observed reaction pathways for benzyl phenyl ether.
Notably, the second most plentiful product after ethyl-
benzene is propylbenzene (300 μmol, ESI† Table S-8), and this
can best be explained by pathways (b) or (c). In pathway (b),
the acetophenone intermediate might undergo methylation
and subsequent HDO of the formed propyl-phenyl ketone.
Indeed, this product was also observed when acetophenone
was used as substrate. Alternatively, it appears that 1 might
undergo facile methylation of the methylene adjacent to the
carbonyl (pathway c) to give compound 3, in analogy to the
reaction proposed for acetophenone. Once formed, 3 could
undergo steps analogous to pathways (a) or (b) to give propyl-
benzene plus phenol-derived products.
Pathway (d), the scission of the phenyl ether bond to give
benzene, would not be not expected given the reactivity pat-
tern seen for benzylphenyl ether. This suggestion was con-
firmed by the observation of very little benzene in the product
mixture (Table S-8†); thus, (d) must play at most a minor role.
These reactions also indicated a very different product pro-
gression. Ether cleavage was an order of magnitude slower,
with 41% of the BPE unreacted after 6 h. Thus, the clean
ether hydrogenolysis, which is a hallmark of Cu20PMO, is
indeed dependent upon the copper dopant. Surprisingly
some toluene is formed, but unlike the case with CuPMO,
this did not appear concurrently with the loss of the BPE, but
later, indicating that is a secondary product. Longer RT prod-
ucts shown by GC-MS to be other dinuclear species were also
formed. A possible mechanism could be analogous to the
one described by Pelzer et al.,41 where hydrolysis (made pos-
sible by residual water) converts BPE to phenol and benzyl
alcohol. The poor material balance at longer reaction times
and the higher molecular weight products are consistent with
the pathways proposed by Pelzer et al. Given that the
undoped PMO is ineffective in lignin disassembly, the origins
of these secondary reactions were not pursued further. Since
the reactions with the undoped PMO are considerably slower,
they would not have a significant effect on the global fitting
kinetics analyses of the primary reactions seen with
Cu20PMO.
Aliphatic alcohols and ketones
The above di-aromatic model compounds as well as oxygen-
ated simple aromatics have provided ring hydrogenation
products, mainly cyclohexanol, which originates from phenol
hydrogenation. Thus, the reactivities of such derivatives were
further investigated to probe the origin of major products
such as the methyl-cyclohexanols.
Cyclohexanol, the expected result of phenol hydrogena-
tion, proved to be quite reactive. Analysis after 1 h reaction
under the standard conditions found that 12% of that sub-
strate was consumed; after 3 h, this increased to 36%. Meth-
ylated cyclohexanol was by far the dominant product initially
formed (Fig. 5) (ESI† Table S-9).
On turning to 2-phenoxy-1-phenylethane-1-one (1), the
above studies offer some expectations regarding reaction path-
ways for this more complex substrate. Based on the results for
acetophenone and of benzyl phenyl ether, one would expect
the two fastest reactions for 1 (1000 μmol) to be carbonyl
group hydrogenation to give 2-phenoxy-1-phenylethanol (2)
(pathway (a) in Scheme 5) and direct hydrogenolysis of the
CH2–O bond to give acetophenone and phenol (pathway b).
2-Methylcyclohexanol was somewhat less reactive than
cyclohexanol with only 19% being consumed after 3 h. The
principal products were dimethylcyclohexanols (13%)
2-methylcyclohexanone (3%), methylcyclohexane (1%),
2-methyl-1-methoxycyclohexane (0.6%) and unknowns (<2%).
(3)
The formation of cyclohexanones as minor products in
several cases suggests that the cyclohexanols undergo revers-
ible dehydrogenation to the ketone analogs (eqn (3)),
followed by methylation at the β-carbon. If so, cyclohexanone
itself should be very reactive toward ring methylation as well
Scheme 5 Likely steps in disasembly of 2-phenoxy-1-phenylethan-1-
one (1).
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 5 Temporal evolution of products formed from cyclohexanol (ca.
1 mmol) in a batch reactor containing Cu20PMO in sc-MeOH (T = 310
°C). Total material balance was greater than 99% after 6 hours.
as to (reversible) ketone hydrogenation. This proved to be the
case when cyclohexanone was used directly as the substrate.
After 3 h, the starting material was mostly consumed, and
the product mixture contained 2-methylcyclohexanol (55%),
cyclohexanol (22%) and dimethylcyclohexanols (13%) and the
lesser products, 2-methylcyclohexanone (2%), methyl-
cyclohexane (2%) and cyclohexane (1%).
Scheme 6 Reaction network for simple mono-aromatic lignin model
compounds over Cu20PMO.
Not surprisingly,
a
mixture of cis- and trans-1,2-
cyclohexanediol gave an even larger collection of products Of
particular interest was the ring contraction product cyclo-
pentylmethanol (12%), which is possibly the result of a
pinacol rearrangement.39,40 The same product was seen when
guaiacol was the substrate (see above).
end-products of phenolic hydrogenation, the relatively fast
methylations of the latter (via a cyclohexanone intermediate)
have a major impact on product distributions.
Reaction of BPE in methanol/dimethyl carbonate
Overview of kinetics results
Comparisons across the reactivity network provide valuable
insight into improving the catalytic selectivity. Since phenolic
derivatives are inevitable outcomes of reductive lignin disas-
sembly, one must address the facile ring hydrogenation and
methylation of these species in order to enhance the yield of
aromatics. Possible scenarios might include designing a
HDO co-catalyst to be very active toward phenols,34 modifying
the catalyst composition to suppress phenolic hydrogenation,
or chemically trapping such intermediates. The third
Scheme 6 is a comprehensive reactivity network compiled
from the most significant individual networks described
above, while Table 1 compares calculated rate constants in
terms of reaction and substrate types (a more detailed net-
work is given in ESI† Fig. S-1). The k's determined for individ-
ual steps and measured for different initial substrates are sta-
tistically self-consistent. Not only does this allow for
comparisons across the whole network, but it is also an indi-
cator that any competitive effects to the observed kinetics
under these conditions must be minimal. Such global quanti-
fication thus allows one to identify key pathways leading
toward or away from the desirable aromatic products when
this catalytic system is employed for lignin disassembly.
For example, hydrodeoxygenation via hydrogenolysis is
quite fast with benzylic alcohols, but much slower with the
cyclohexanols or phenol. For phenol, ring hydrogenation and
methylation are the dominant pathways. In contrast, these
are suppressed for anisole and ethoxybenzene to the point
where slow HDO to the even less reactive benzene is the most
significant reaction. Ring methylation of phenols contributes
to product proliferation. However, since cyclohexanols are
Table 1 Overview of kobs trends observed with 1 mmol substrate, 100
mg Cu20PMO, and 3 mL MeOH, at 310 °C
Qualitative rate
Fastest
kobs (h−1
>1.0
)
Reactions
Benzyl phenyl ether hydrogenolysis
Styrene hydrogenation
Benzylic alcohol hydrodeoxygenation
Phenolic hydrogenation
Cyclohexanol methylation
Phenolic methylation
Fast
0.2–0.75
0.05–0.1
<0.05
Slow
Cresol methylation
Slowest
Other aromatic hydrogenation
Aryl phenoxy hydrogenolysis
Phenolic hydrodeoxygenation
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
Fig. 7 Changes in product distribution upon using dimethyl carbonate
as a cosolvent vs. using MeOH only. Products after 6 h reaction.
Fig. 6 Temporal product evolution for the reactions of benzyl phenyl
ether in 2 : 1 MeOH : dimethyl-carbonate over Cu20PMO (300 °C).
Material balance was 93% after 18 hours.
described in Fig. 4, we deemed it important to compare two
systems containing the same quantities of MeOH. Fig. 7 illus-
trates the product yields determined by GC-FID for the reac-
tions of BPE over Cu20PMO after side-by-side reaction for 12 h
under identical conditions, one with only 2 mL of MeOH as
solvent the other with 3 mL of 2: 1 MeOH : DMC. Clearly the
aromatics yield proved dramatically higher in the latter case
owing largely to the interception of the phenolic intermediates.
In another control experiment, the reaction of BPE with the
Mg/Al (3/1) PMO was also carried out in 2 : 1 MeOH : DMC
under the standard conditions. The result was analogous to
the reaction in MeOH with the exception that the yield of
anisole was substantially greater and that of phenol much less.
approach was investigated by using dimethyl carbonate
(DMC), a known O-methylating agent that is activated by Mg/
Al metal oxides,42–44 as a cosolvent.
The reaction with BPE as the substrate was run under oth-
erwise typical conditions in a solution prepared with 2 mL
MeOH and 1 mL DMC. After 1 h, the benzyl phenyl ether was
89% consumed, with the concurrent production of toluene,
phenol and anisole (Fig. 6). Also found were traces of cresols
and aliphatic compounds previously shown to be derived
from phenol. By 6 h, the BPE was completely converted with
the primary products being toluene (900 μmol) and anisole
(490 μmol); thus, O-methylation of phenol channelled the
products away from ring hydrogenation.
Summary
The reactivity network in Scheme 7 was drawn from the
temporal evolution of products shown in Fig. 6. Thus, it is
clear that the percentage of phenol-derived aromatics, mostly
in the form of anisole, is dramatically higher than observed
in methanol alone (Fig. 4) under comparable conditions.
Since the mini-reactor experiments summarized in Fig. 6
have less methanol (2 mL) than was used to gather the data
The model systems quantitatively elucidated here clearly dem-
onstrate four key substrate reactions catalysed by Cu20PMO
under conditions where clean lignin and lignocellulose disas-
sembly has been demonstrated.21–23 Two of these are generally
desirable, namely hydrogenolysis of aryl ether bonds and
hydrodeoxygenation. The other two, aromatic hydrogenation
and methylations of aromatic rings and of aliphatic carbons
in positions adjacent to alcohol or ketone functionalities, lead
to a proliferation of products, hence lower selectivity. Phenolic
compounds, especially phenol, are susceptible to aromatic
ring hydrogenation and methylation, thus are precursors to
cyclohexanols that undergo subsequent methylation. In con-
trast, simpler aromatics such as benzene and toluene are rela-
tively unreactive under analogous conditions as is anisole,
which primarily undergoes slow HDO to benzene.
In this context, we have demonstrated that intercepting
phenolic intermediates by alkylating the aromatic –OH dra-
matically reduces product proliferation and aromaticity loss.
While this was accomplished using DMC as a cosolvent,
these observations clearly point toward strategies that will
enhance selectivity by addressing phenolic reactivity. Given
the complexity of lignin as a substrate, any such strategy will
inevitably give multiple products; however, narrowing the
Scheme 7 Reactivity network for benzyl phenyl ether with Cu20PMO
in 2 : 1 MeOH : DMC (300 °C).
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
distribution will improve the eventual success of biological or
chemical funnelling, hence the valorization of lignin.1–4,45
The essential role of copper in the methanol reforming
that is the source of the reducing equivalents in the present
system has been discussed previously.21–23 However, the role
of the supporting PMO is less well understood. Lewis acids
have been shown to activate the aryl ether linkages to hydro-
genolysis;46,47 however, it is not clear how these would
impact the other favourable and less favourable pathways of
this catalytic system. For example, previous studies have
demonstrated that mixed Mg/Al oxides catalyze phenol meth-
ylation.38 Alternatively, methylation of the phenol oxygen,
perhaps via catalysis by the acidic and basic sites at the sup-
port surface, would strongly affect overall reaction selectivity.
Ongoing studies in our respective laboratories are addressing
how the Cu-doped PMOs transform under the reaction condi-
tions as well as how modifications of the support define reac-
tivity and selectivity.
7 P. Sannigrahi, A. J. Ragauskas and G. A. Tuskan, Poplar as a
Feedstock for Biofuels:
A
Review of Compositional
Characteristics, Biofuels, Bioprod. Biorefin., 2010, 4, 209–226.
8 W. Boerjan, J. Ralph and M. Baucher, Lignin Biosynthesis,
Annu. Rev. Plant Biol., 2003, 54, 519–546.
9 J. J. Stewart, T. Akiyama, C. Chapple, J. Ralph and S. D.
Mansfield, The Effects on Lignin Structure of Overexpression
of Feru-late 5-Hydroxylase in Hybrid Poplar, Plant Physiol.,
2009, 150, 622–635.
10 R. Lal, Soils and Sustainable Agriculture: a Review, Agron.
Sustainable Dev., 2008, 28, 15–22.
11 Global Aromatic Chemicals (Benzene, Toluene & Xylene)
Market 2014. http://www.researchandmarkets.com/research/
6xr4ck/global_aromatic, (accessed Jan 29, 2015).
12 A. G. Pemba, M. Rostagno, A. L. Tanner and S. A. Miller,
Cyclic and Spirocyclic Polyacetal Ethers from Lignin-Based
Aro-matics, Polym. Chem., 2014, 5, 3124–3221.
13 B. Sedai, C. Díaz-Urrutia, R. T. Baker, R. Wu, L. A. Silks and
S. K. Hanson, Aerobic Oxidation of β-1 Lignin Model Com-
pounds with Copper and Oxovanadium Catalysts, ACS Catal.,
2013, B, 3111–3122.
Acknowledgements
14 S. Van den Bosch, W. Schutyser, R. Vanholme, T. Driessen,
S.-F. Koelewijn, T. Renders, B. De Meester, W. J. J. Huijgen,
W. Dehaen, C. M. Courtin, B. Lagrain, W. Boerjan and B. F.
Sels, Reductive Lignocellulose Fractionation into Soluble Lignin-
Derived Phenolic Monomers and Dimers and Processable Car-
bohydrate Pulps, Energy Environ. Sci., 2015, 8, 1748–1763.
15 R. Ma, W. Hao, X. Ma, Y. Tian and Y. D. Li, Catalytic
Ethanolysis of Kraft Lignin into High-Value Small-Molecular
Chemicals over a Nanostructured alpha-Molybdenum Car-
bide Catalyst, Angew. Chem., Int. Ed., 2014, 53, 7310–7315.
16 J. Zhang, Y. Chen and M. A. Brook, Reductive Degradation of
Lignin and Model Compounds by Hydrosilanes, ACS
Sustainable Chem. Eng., 2014, 2, 1983–1991.
17 A. K. Deepa and P. L. Dhepe, Lignin Depolymerization into
Aromatic Monomers over Solid Acid Catalysts, ACS Catal.,
2015, 5, 365–379.
18 A. Rahimi, A. Ulbrich, J. J. Coon and S. S. Stahl, Formic-acid-
induced depolymerization of oxidized lignin to aromatics,
Nature, 2014, 515, 249–252.
19 Y. Jiang, Z. Li, X. Tang, Y. Sun, X. Zeng, S. Liu and L. Lin,
Depolymerization of Cellulolytic Enzyme Lignin for the
Production of Monomeric Phenols over Raney Ni and Acidic
Zeolite Catalysts, Energy Fuels, 2015, 29, 1662–1668.
20 M. R. Sturgeon, M. H. O'Brien, P. N. Ciesielski, R. Katahira, J. S.
Kruger, S. C. Chmely, J. Hamlin, K. Lawrence, G. B. Hunsinger,
T. D. Foust, R. M. Baldwin, M. J. Biddy and G. T. Beckham,
Lignin depolymerisation by nickel supported layered-double
hydroxide catalysts, Green Chem., 2014, 16, 824–835.
We thank the Center for the Sustainable Use of Renewable
Feedstocks (CenSURF), a NSF Center for Chemical Innovation
(NSF CHE-1240194) and the People Programme (Marie Curie
Actions) of the European Union's Seventh Framework
Programme FP7/2007–2013/ under REA grant agreement no.
622724 [Asymm.Fe.Sus.Cat] for support of this research. We
also thank Megan Chui for catalyst preparation.
Notes and references
1 A. J. Ragauskas, G. T. Beckham, M. J. Biddy, R. Chandra, F.
Chen, M. F. Davis, B. H. Davison, R. A. Dixon, P. Gilna, M.
Kel-ler, P. Langan, A. K. Naskar, J. N. Saddler, T. J.
Tschaplinski, G. A. Tuskan and C. E. Wyman, Lignin
Valorization: Improving Lignin Processing in the Biorefinery,
Science, 2014, 344, 709.
2 C. O. Tuck, E. Perez, I. T. Horvath, R. A. Sheldon and M.
Poliakoff, Valorization of Biomass: Deriving More Value
from Waste, Science, 2012, 337, 695–699.
3 J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius and B. M.
Weck-huysen, The Catalytic Valorization of Lignin for the
Produc-tion of Renewable Chemicals, Chem. Rev., 2010, 110,
3552–3599.
4 P. Azadi, O. R. Inderwildi, R. Farnood and D. A. King, Liquid
fuels, hydrogen and chemicals from lignin: A critical review,
Renewable Sustainable Energy Rev., 2013, 21, 506–523.
5 Y.-C. Lin and G. W. Huber, The critical role of
heterogeneous
conversion, Energy Environ. Sci., 2009, 2, 68–80.
6 T. Werpy and G. Petersen, Top Value Added Chemicals from
Biomass, Volume II Results of Screening for Potential
catalysis
in
lignocellulosic
biomass
21 K. Barta, T. D. Matson, M. L. Fettig, S. L. Scott, A. V. Iretskii
and P. C. Ford, Catalytic Disassembly of an Organosolv
Lignin via Hydrogen Transfer from Supercritical Methanol,
Green Chem., 2010, 12, 1640.
–
Candidates from Biorefinery Lignin, Pacific Northwest
National Laboratory and the National Renewable Energy
Laboratory, 2007.
22 T. D. Matson, K. Barta, A. V. Iretskii and P. C. Ford, One-Pot
Catalytic Conversion of Cellulose and of Woody Biomass Sol-
ids to Liquid Fuels, J. Am. Chem. Soc., 2011, 133, 14090–14097.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
23 K. Barta and P. C. Ford, Catalytic Conversion of Non-food
Woody Biomass Solids to Organic Liquids, Acc. Chem. Res.,
2014, 47, 1503–1512.
24 G. S. Macala, T. D. Matson, C. L. Johnson, R. S. Lewis, A. V.
Iretskii and P. C. Ford, Hydrogen Transfer from Supercritical
Methanol Over a Solid Base Catalyst: a Model for Lignin
Depolymerization, ChemSusChem, 2009, 2, 215–217.
25 S. Lee, Methanol Synthesis Technology, CRC Press, Boca
Raton, Florida, 1990.
26 A. A. Gokhale, J. A. Dumesic and M. Mavrikakis, On the
Mechanism of Low-Temperature Water Gas Shift Reaction
on Copper, J. Am. Chem. Soc., 2008, 130, 1402–1414.
27 K. Barta, G. R. Warner, E. S. Beach and P. T. Anastas,
Depolymerization of Organosolv Lignin to Aromatic
Compounds Over Cu-Doped Porous Metal Oxides, Green
Chem., 2014, 16, 191–196.
28 T. S. Hansen, K. Barta, P. T. Anastas, P. C. Ford and A.
Riisager, One-pot reduction of 5-hydroxymethylfurfural via
hydrogen transfer from supercritical methanol, Green Chem.,
2012, 14, 2457–2461.
Effective Carbon Number Concept, J. Chromatogr. Sci.,
1985, 23, 333–340.
37 P. Kuzmic, Program DYNAFIT for the Analysis of Enzyme
Kinetic Data: Application to HIV Proteinase, Anal. Biochem.,
1996, 237, 260–273.
38 V. Crocella, G. Cerrato, G. Magnacca, C. Morterra, F. Cavani,
L. Maselli and S. Passeri, Gas-phase Phenol Methylation of
Mg/Me/O (Me
=
Al, Cr, Fe) Catalysts: Mechanistic
Implications Due to Different Acid-base and Dehydro-
genating Properties, Dalton Trans., 2010, 39, 8527–8537.
39 K. L. Deutsch and B. H. Shanks, Hydrodeoxygenation of
Lignin Model Compounds Over a Copper Chromite Catalyst,
Appl. Catal., A, 2012, 447, 144–150.
40 W. Song, Y. Liu, E. Baráth, C. Zhao and J. A. Lercher,
Synergistic effects of Ni and acid sites for hydrogenation and
C-O bond cleavage of substituted phenols, Green Chem.,
2015, 17, 1204–1218.
41 A. W. Pelzer, M. R. Sturgeon, A. J. Yanez, G. Chupka, M. H.
O'Brien, R. Katahira, R. D. Cortright, L. Woods, G. T.
Beckham and L. J. Broadbelt, Acidolysis of the α-O-4 Aryl-
Ether Bonds in Lignin Model Compounds: A Modeling and
Experimental Study, ACS Sustain. Chem. Eng., 2015, 3,
1339–1347.
42 R. Luque, J. M. Campelo, T. D. Conesa, D. Luna, J. M.
Marinas and A. A. Romero, Catechol O-Methylation with
Dimethyl Carbonate Over Different Acid-base Catalysts, New
J. Chem., 2006, 30, 1228–1234.
29 A. J. Kumalaputri, G. Bottari, P. M. Erne, H. J. Heeres and K.
Barta, Tunable and Selective Conversion of 5-HMF to 2,
5-Furandimethanol and 2, 5-Dimethylfuran over Copper-Doped
Porous Metal Oxides, ChemSusChem, 2014, 7, 2266–2275.
30 X. Huang, T. I. Koranyi, M. D. Boot and E. J. M. Hensen,
Catalytic Depolymerization of Lignin in Supercritical
Ethanol, ChemSusChem, 2014, 7, 2276–2288.
31 M. J. Girgis and B. C. Gates, Reactivities, Reaction Networks,
and Kinetics in High-Pressure Catalytic Hydroprocessing,
Ind. Eng. Chem. Res., 1991, 30, 2021–2058.
32 R. C. Runnebaum, T. Nimmanwudipong, D. E. Block and
B. C. Gates, Catalytic conversion of compounds
representative of lignin-derived bio-oils: a reaction network
for guaiacol, anisole, 4-methylanisole, and cyclohexanone
conversion catalysed by Pt/gamma-Al2O3, Catal. Sci. Technol.,
2012, 2, 113–118.
33 Q. Bu, H. Lei, A. H. Zacher, L. S. Ren, J. Liang, Y. Wei, Y.
Liu, J. Tang, Q. Zhang and R. Ruan, A review of catalytic
hydrodeoxygenation of lignin-derived phenols from biomass
pyrolysis, Bioresour. Technol., 2012, 124, 470–477.
43 M. B. Talawar, T. M. Jyothi, P. D. Sawant, T. Raja and B. S.
Rao, Calcined Mg-Al Hydrotalcite as an Efficient Catalyst for
the Synthesis of Guaiacol, Green Chem., 2000, 2, 266–268.
44 J. N. G. Stanley, M. Selva, A. F. Masters, T. Maschmeyer and
A. Perosa, Reactions of p-coumaryl alcohol model com-
pounds with dimethyl carbonate, Towards the upgrading of
lignin building blocks, Green Chem., 2013, 15, 3195–3204.
45 J. G. Linger, D. R. Vardon, M. T. Guarnieri, E. M. Karp, G. B.
Hunsinger, M. A. Franden, C. W. Johnson, G. Chupka, T. J.
Strathmann, P. T. Pienkos and G. T. Beckham, Lignin
valorization through integrated biological funneling and
chemical catalysis, Proc. Natl. Acad. Sci. U. S. A., 2014, 111,
12013–12018.
34 X. Wang and R. Rinaldi, Route for Lignin and Bio-Oil Con-
version: Dehydroxylation of Phenols into Arenes by Catalytic
Tandem Reactions, Angew. Chem., Int. Ed., 2013, 52,
11499–11503.
46 A. C. Atesin, N. A. Ray, P. C. Stair and T. J. Marks, Ether C-O
Bond Hydrogenolyiss Using a Tandem Lanthanide Tri-flate/
Supported Palladium Nanoparticle Catalyst System, J. Am.
Chem. Soc., 2012, 134, 14682–14685.
35 X. Wang and R. Rinaldi, Exploiting H-transfer reactions with
RANEY (R) Ni for upgrade of phenolic and aromatic bio-
refinery feeds under unusual, low-severity conditions, Energy
Environ. Sci., 2012, 5, 8244–8260.
36 J. T. Scanlon and D. E. Willis, Calculation of Flame
Ionization Detector Relative Response Factors Using the
47 T. H. Parsell, B. C. Owen, T. M. Jarell, C. L. Marcum, L. J.
Haupert, L. M. Amundson, K. I. Kenttamaa, F. Ribeiro, J. T.
Miller
and
M.
M.
Abu-Omar,
Cleavage
and
Hydrodeoxygenation (HDO) of C-O Bonds Relevant to Lignin
Conversion Using Pd/Zn Synergistic Catalysis, Chem. Sci.,
2013, 4, 806–813.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.