Catalytic cleavage of the Β-O-4 aryl ether bonds of lignin model compounds by Ru/C catalyst

Article abstract of DOI:10.1016/j.apcata.2019.05.034

Lignin is a potential renewable feedstock for aromatic compounds. Lignin glues cellulose and hemicellulose together in a rigid structure that protects plants from weather, insects, and disease. This rigidity also poses a barrier to cleavage of lignin into

Full text of DOI:10.1016/j.apcata.2019.05.034

Accepted Manuscript
Title: Catalytic cleavage of the ␤-O-4 aryl ether bonds of
lignin model compounds by Ru/C catalyst
Authors: Md Anwar Hossain, Thanh Khoa Phung,
Mohammad Shahinur Rahaman, Sarttrawut Tulaphol, Jacek B.
Jasinski, Noppadon Sathitsuksanoh
PII:
S0926-860X(19)30247-9
DOI:
Reference:
https://doi.org/10.1016/j.apcata.2019.05.034
APCATA 17100
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
30 November 2018
23 May 2019
29 May 2019
Please cite this article as: Hossain MA, Phung TK, Rahaman MS, Tulaphol S,
Jasinski JB, Sathitsuksanoh N, Catalytic cleavage of the ␤-O-4 aryl ether bonds of
lignin model compounds by Ru/C catalyst, Applied Catalysis A, General (2019),
https://doi.org/10.1016/j.apcata.2019.05.034
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Catalytic cleavage of the β-O-4 aryl ether bonds of lignin model compounds by
Ru/C catalyst
1
,
1,2,3,
1
Md Anwar Hossain , Thanh Khoa Phung
, Mohammad Shahinur Rahaman ,
4
1,2
1,
Sarttrawut Tulaphol , Jacek B. Jasinski , and Noppadon Sathitsuksanoh *
1Department of Chemical Engineering, University of Louisville, Louisville, KY 40292
2Conn Center for Renewable Energy Research, University of Louisville, Louisville, KY
4
0292
3International University, Vietnam National University – HCMC, Quarter 6, Linh Trung
Ward, Thu Duc District, Ho Chi Minh City, Vietnam
4Department of Chemistry, King Mongkut’s University of Technology Thonburi,
Bangkok, Thailand 10140

These authors contributed equally to this work
*
Corresponding author: N.sathitsuksanoh@louisville.edu
Graphical abstract
1

Highlights
. The aliphatic hydroxyl groups (C -OH) of lignin model compound serve as the
1

internal hydrogen source (self-hydrogen) in Ru-catalyzed hydrogenolysis of the
β-O-4 aryl ether bonds.
2
3
. The hydrogenolysis activity of C
β
-O bonds of lignin increases as a function of the
Ru content of the catalyst and reaches 112 mmol/g Ru-h.
. The self-hydrogen transfer hydrogenolysis of the C
alternative and efficient strategy for lignin conversion without a requirement for
external, high-pressure H
β
-O bonds provides an
2
.
2

Abstract
Lignin is a potential renewable feedstock for aromatic compounds. Lignin glues
cellulose and hemicellulose together in a rigid structure that protects plants from
weather, insects, and disease. This rigidity also poses a barrier to cleavage of lignin into
aromatic compounds. Typically, lignin is depolymerized by metal-catalyzed
hydrogenolysis of its β-O-4 aryl ether (C
pressure. Here, we show that the abundant aliphatic hydroxyl groups (C
structure, can serve as the hydrogen source in Ru-catalyzed hydrogenolysis of the C
bonds. We pretreated the Ru/C catalyst under reducing and oxidizing conditions to
generate various Ru/RuO ratios. Then we investigated the effects of Ru and RuO on
hydrogenolysis of the C -O bonds of lignin model compounds. We used X-ray diffraction
XRD) and H Temperature-programmed reduction (TPR) to determine changes of grain
β
-O) bonds; this process requires high H
-OH) in lignin
-O
2

β
2
2
β
(
2
size and Ru content of the Ru/C catalysts, respectively. Our results revealed that Ru/C
catalyzed hydrogenolysis of β-O-4 aryl ether bonds with internal hydrogen (self-
hydrogen) as a hydrogen source. The elimination of external H in lignin hydrogenolysis
2
is an efficient approach for lignin conversion to valuable aromatic chemicals.
Keywords: lignin; ruthenium; self-hydrogen; hydrogenolysis; hydrogen transfer.
3

1
. Introduction
Lignin is a by-product of paper and pulp manufacturing and biorefineries [1-3]. Lignin
has aromatic backbones, making it an ideal renewable feedstock of aromatic
compounds for a range of applications, including automotive brakes, wood panel
products, surfactants, phenolic resins, phenolic foams, bio-dispersants polyurethane
foams, and epoxy resins [4-12]. Despite this great potential for high-value exploitation,
only ~5% of lignin by-products have been used in mostly low-value commercial
applications, such as the production of low-grade fuels for heat and power and as
concrete additives [4, 8]. One reason for this restricted usage is that the complex
chemical structure of lignin makes it difficult to release its aromatic monomers.
Typically, lignin is composed of three major phenolic monomers (monolignols): (1) p-
coumaryl alcohol, (2) coniferyl alcohol, and (3) sinapyl alcohol. These monolignols are
polymerized into lignin in the form of phenylpropanoid units (a phenyl (C
)), p-hydroxyphenyl (H), guaiacyl (G), and syringyl (S) units. These phenylpropanoid
units are joined to form C -O (β-O-4 aryl ether) and C-C (β-5 and β-β) bonds (Figure 1
adapted from Ref. [13])). Lignin contains ~50-65% β-O-4 aryl ether bonds and
6
) + a propane
(C
3
β
(
abundant aliphatic and aromatic hydroxyl groups [14]. These linkages and functional
groups make lignin a stiff and rigid structure, giving plants strength and protecting them
from external disturbances such as insects, disease, and weather. This rigid structure of
lignin is another reason why it is difficult to break lignin into aromatic monomers.
4

To release renewable aromatic monomers, investigators have used hydrogenolysis by
metal catalysts, such as Ru/C, Ni/C, Pd/C, Pd/Al , CuMgO , NiRu, and
CuCr ·CuO, to break the abundant C -O bonds of lignin [15-26]. Typically, a high H
pressure (≥ 10 bar) is required for hydrogenolysis [22, 27, 28]; however, the high H
2
O
3
x
2
O
4
β
2
2
pressure causes undesired side reactions of over-hydrogenating the aromatic rings,
cracking, and coke formation [27]. Moreover, the hydrogen sources are not naturally
available and renewable, making H -mediated hydrogenolysis uneconomic on a large
2
scale [29, 30]. To minimize the effect of side reactions, the industry requires hydrogen-
lean or hydrogen-free catalytic systems.
Oxophilic metals, such as ruthenium (Ru), have partially filled d-bands. The oxophilicity
of Ru enables strong interaction with oxygen atoms in the adsorbates, resulting in the
direct cleavage of C-O bonds [31-33]. Previous studies have shown that reducible RuO
2
catalysts have Lewis acid sites that facilitate the hydrodeoxygenation of furanics in the
liquid phase [34, 35]. Supported Ru catalysts have been applied to the hydrogenolysis
of lignin and its model compounds [36, 37], the effect of Ru and RuO
hydrogenolysis and hydrodeoxygenation of lignin is not well understood. The partial
oxidation of Ru to RuRuO creates a bifunctional catalyst containing: (1) Ru metal sites,
catalyzing hydrogenolysis/hydrogenation; and (2) RuO Lewis acid sites, facilitating
hydrogenolysis [34]. We hypothesized that the aliphatic -OH groups (C -OH) of lignin
could serve as internal hydrogen donors; the released hydrogen would break lignin’s
2
on the
2
2

C
β
-O bonds without further aromatic ring saturation. We further expected that this
5

hydrogenolysis reaction would occur with a Ru/C catalyst having an optimal Ru/RuO
composition.
2
To test these conjectures, we synthesized Ru/C catalysts with various ratios of Ru/RuO
2
and assessed their activities in hydrogenolysis of the C
compounds: 2-phenoxy-1-phenylethanol and 2-phenyl ethyl phenyl ether. The aliphatic
-OH group of 2-phenoxy-1-phenylethanol enabled the hydrogenolysis of the C -O
bonds in the absence of H . An increase in Ru content enhanced hydrogenolysis
activity. The elimination of the H requirement in lignin hydrogenolysis provides a simple
yet efficient approach for lignin conversion to aromatic chemicals.
β
-O bond in two lignin model
C

β
2
2
2
. Experimental
2
.1. Materials
All reagents were used as received. Their manufacturers, purity, and CAS numbers are
shown in Table S1.
2
.2. Catalysts pretreatment and characterization
To tailor the catalyst composition, the commercial Ru/C catalyst (parental Ru/C) was
treated in various conditions before catalytic testing. For reduction, the fresh Ru/C
catalyst was reduced in H flow of 40 cc/min at 250°C for 3h, followed by the 12h
2
passivation in air, to form the Ru/C-Red catalyst. For oxidation, the fresh Ru/C catalyst
6

was oxidized in pure O flow of 40 cc/min at 200°C for 3h to create the Ru/C-Ox
2
catalyst. The commercial Ru/C and commercial pre-reduced Ru/C catalysts were also
used as references. Descriptions of these Ru/C catalysts and their pretreatment
conditions are summarized in Table S2.
X-ray diffraction (XRD) was performed with a Bruker D8 Discover diffractometer
(
Billerica, MA, USA) using CuK
α
radiation in the 2θ range from 20° to 60° with 1
Temperature-Programmed Reduction (H -TPR)
second/step (0.02 increment). H
2
2
experiments were performed with a Micromeritics ChemiSorb 2720 equipped with a
thermal conductivity detector (TCD) (Norcross, GA, USA). About 20-40 mg of sample
was pretreated at 250°C for 1h under He flow to remove adsorbed water. Then, the
sample was cooled to room temperature under He flow. TPR profiles were recorded by
heating the samples from room temperature to 800°C at a heating rate of 10°C/min in
the 10.01% H
2
/Ar at a flow rate of 40 cc/min. The RuO
2
content in the Ru/C catalysts
was calculated by the following reaction:
0
4+
0
RuO + 2H → Ru + 2H O (Ru → Ru )
2
2
2
H
2
consumption and mole of metallic Ru and RuO
2
were calculated using the known H
2
(vol.%) as a calibrant as follows:
∑
Areaoutlet peaks
molecalibrant(mol)
H consumption (mol H /g total Ru)=
×
x
2
2
Areacalibrant
masscatalyst(g)
1
Ru loading (wt.%)
7

1
moleRuO2(mol) = mole = mole
O
2
H consumption
2
2
molemetallic Ru (mol) = moletotal Ru-moleRuO2
moleRuO2
mole +mole
RuO in total Ru (mol.%) =
×100
2
Ru
RuO2
Energy Dispersive X-ray Spectroscopy (EDS) was conducted on the spent Ru/C
catalyst to assess the change in the Ru content after the reaction. EDS was performed
on the Thermo ScientificTM FEI Nova600 FEG Scanning Electron Microscope (SEM)
equipped with EDS (Hillsboro, OR, USA).
2
.3. Hydrogenolysis of lignin model compounds
All reactions were performed in a 25 mL autoclave reactor (Parr Instrument, Moline, IL,
USA). The reactant concentrations were 1 wt.% of 2-phenoxy-1-phenylethanol or 2-
phenyl ethyl phenyl ether in ethanol. The catalyst loading was 20 wt.% (~18.9 mg
catalyst) with respect to the reactant (94.68 mg reactant in 12 mL ethanol solution).
Prior to the reaction, the reactor was purged three times with N
2
to remove O . The
2
reactor was then pressurized to 8 bar under N at room temperature (for consistency).
2
The hydrogenolysis reaction was performed at 280°C for 4h with a stirring rate of 500
rpm. The reaction was stopped by quenching in a cold water bath. The reaction sample
was centrifuged to remove any residual solids, then diluted with ethanol prior to the
product analysis. Dodecane was used as an internal standard.
8

The reactants and products were identified and quantified by the Agilent 7890B GC
Agilent Technologies, Santa Clara, CA, USA) equipped with Mass spectrometry (MS)
(
and Flame Ionization Detectors (FID). An HP-5MS column (30mx0.25mmx0.25µm,
Agilent Technologies, Santa Clara, CA, USA) was used for product separation with the
following temperature program: injection temperature 275°C and FID detector
temperature 300°C; split ratio 1:50. The temperature program started at 45°C and
increased at 10°C/min to 250°C, then held for 20 min. Reactant conversion, product
selectivity, and specific activity were calculated using the pre-determined response
factors with dodecane as an internal standard. The calculations are as follows:
mole of reactant reacted
Conversion (%) =
×100%
×100%
initial mole of reactant
mole of product generated
mole of feed reacted
Selectivity (%) =
mole of feed reacted
weight of Ru x time
Specific activity (mmol/(g *h)) =
Ru
2
.4. Catalyst stability evaluation
The catalyst stability was examined by conducting catalyst recycling experiments for
four times. After the reaction, the spent catalyst was recovered by filtration and reused
without washing/drying in the next experiment. The reactants and products from each
recycle run were quantified by GC-MS/FID. The Ru content of the spent catalyst from
each recycle run was analyzed by H
2
-TPR. In separate experiments, the spent catalyst
9

was sampled with the reaction products after each recycle run into the U-shaped
reactors directly for H -TPR to minimize the catalyst oxidation in air.
2
3
. Results and Discussion
To examine the effect of metallic Ru and RuO
bonds, we varied the amount of Ru and RuO by treating the parental Ru/C in various
conditions (Table S2). The commercial pre-reduced Ru/C served as the control. We
used H -TPR and XRD to determine the amount of RuO in these four catalysts and
identify their phases, respectively. We correlated changes in Ru content with the
2
on the hydrogenolysis of the C -O
β
2
2
2
hydrogenolysis activities of the catalysts toward the C
compounds.
β
-O bonds of the lignin model
3
.1. Catalysts pretreatment and their reducibility
The relative Ru and RuO
2
content in Ru/C catalyst play an important role in
-TPR technique to determine the
as a stoichiometric reductant, we
hydrogenolysis of the C
amount of Ru and RuO
β
-O bonds. We first used the H
in the four catalysts. Using H
2
2
2
found that, as the catalyst temperature increased from 50 to 350°C, the reduction
profiles of Ru/C catalysts showed two major reduction peaks. The first reduction peak
appeared at <100°C and the second reduction peak occurred at >100°C (Figure 2). The
presence of two reduction peaks suggested that ruthenium took different oxidation
states while reducing from Ru4+ (RuO
) to Ru (metallic Ru). The H
0
2
2
-TPR profile of the
1
0

pre-reduced-Ru/C showed a small reduction peak at 74°C. This reduction peak was
observed previously in Ru/CeO and Ru/C catalysts [38]. The commercial Ru/C catalyst
parent) had two observable reduction peaks at <100°C and >200°C with a shoulder
peak at 246°C. The first reduction peak at low temperature appeared to be broad
doublets. Other investigators have observed doublets with Ru supported on Al , ZrO
CeO , and carbon [32, 38, 39]. Doublets are hypothesized to form because of (1) the
strong interaction between Ru species and the support [38, 40], and (2) the reduction of
2
(
2
O
3
2
,
2
4+
2+
Ru to Ru , suggesting the formation of easily reducible surface species. The second
reduction peak (>200°C) could be assigned to the reduction of Ru2+ to Ru or the
+
reduction of Ru2+ to Ru (metallic Ru) [41-46]. After reducing the parental Ru/C in H
0
2
,
the second reduction peak at 205°C and the shoulder peak at 246°C disappeared
Figure S1) and the H -TPR profile became similar to that of pre-reduced Ru/C. This
(
2
result suggested that this H
-TPR condition was sufficient to completely reduce the Ru4+
2
0
to Ru . Then, we allowed this catalyst to passivate in the air for 12 h to generate partial
RuO
2
phase. We used the term Ru/C-Red to represent this sample. The Ru/C-Red had
-TPR profile to that of Ru/C. However, its first reduction peak was rather
a similar H
2
broad and the second reduction peak was shifted to a higher temperature (217°C)
compared with that of the parental Ru/C (205°C). Next, we oxidized the parental Ru/C
using O
second reduction peak from 205°C to 135°C, indicating that there were structural
changes in RuO from oxidation. The large reduction peak at 135°C on the Ru/C-Ox
resulted from the formation of larger RuO particle sizes, decreasing the interaction
2
flow at 200°C for 3h to obtain the Ru/C-Ox sample. We observed a shift of the
2
2
1
1

between supports and RuO
2
[47-51]. We also used these H -TPR profiles to determine
2
the Ru and RuO content in all catalysts.
2
The Ru content in the Ru/C catalysts, calculated based on the H consumption, was in
2
the following order: pre-reduced Ru/C (95.3%) > Ru/C-Red (85.5%) > Ru/C (82.0%) >
Ru/C-Ox (74.7%) (Table S3). The parental Ru/C catalyst had ~82% Ru (metallic) and
~
18% RuO
generate RuO
the RuO content of the Ru/C-Ox increased ~28% compared to that of parental Ru/C.
2
. Ru/C-Red, reduced Ru/C catalyst by H
2
and passivated in the air to
2
, had a 22% decrease in RuO content. After oxidizing the parental Ru/C,
2
2
For Ru/C-Ox, the carbon support was oxidized during the pretreatment, causing a
relative increase in Ru content per total gram of Ru/C-Ox. We used thermogravimetric
analysis of the Ru/C catalysts, phase identification by XRD spectra, and an estimated
amount of RuO
2
2
and Ru content by H -TPR to determine the mass percentage of Ru,
RuO , and C (Table S4). These results indicated that after oxidation, the Ru/C-Ox lost
2
carbon support by ~22 wt.%. These Ru/C catalysts were used in the hydrogenolysis of
the lignin model compounds.
2
H -TPR results confirmed the reducibility of the Ru/C catalysts. They demonstrated that
we could tune the oxidation state of Ru/C under various pretreatment conditions [52].
Our results also illustrated that ruthenium was easily passivated, as shown in the
formation of RuO
2
under the ambient condition. The use of commercial Ru/C catalyst
needs to be with cautions because it can be passivated, affecting the Ru and RuO
2
1
2

contents. The catalyst pretreatment conditions affected the grain size and
dispersion/agglomeration of Ru catalysts on supports. For example, the shift of the
reduction peak to the lower reduction temperature of the Ru/C-Ox (compared to that of
parental Ru/C) suggested an increase in the grain size of the RuO
grain size of catalyst also affects its catalytic activity [53]. To identify changes in Ru and
RuO phases and grain size of catalysts, we applied the XRD technique on all catalysts.
2
. The change in
2
3
.2. Identification of Ru and RuO
2
phases and determination of catalyst grain
sizes
XRD spectra of Ru/C catalysts revealed the Ru and RuO
dispersion on the carbon support (Figure 3). The parental Ru/C and pre-reduced Ru/C
catalysts had broad XRD spectra, suggesting that (1) the metallic Ru and RuO were
highly dispersed on the carbon, and (2) their grain sizes were small [26]. After the
reduction in H and passivation, the Ru/C-Red illustrated three XRD peaks, associated
2
phases and their degrees of
2
2
o
o
o
with the presence of metallic Ru at 2 of ~39 for Ru(100), 42 for Ru(002), and 44 for
Ru(101) [46, 54]. The crystallite sizes of the Ru catalysts were calculated from the
Debye-Scherrer equation and the broadening of the main peaks [55]. The mean
crystallite size was ~4.4 nm for Ru(100) and Ru(101). The peak Ru(002) had low
intensity, and we did not calculate its crystallite size. In the case of Ru/C-Ox, we
observed sharp XRD peaks associated with the presence of Ru and RuO . The three
2
XRD diffraction peaks of Ru (100), (002), and (101) became more pronounced
compared with the diffraction peaks from the Ru/C post-reduction. This increase in peak
intensity resulted from (1) the lower content of the carbon support due to the oxidation
1
3

of carbon and (2) the sintering of the Ru, which formed larger Ru aggregates. Our
calculation showed that the crystallite size of the Ru was ~19-25 nm. Three XRD peaks
of RuO
2
emerged, corresponding to RuO
2
(110), RuO
2
(101), and RuO
2
(211) at 2θ of 28,
peaks to
and/or (2) the sintering of the RuO
particles. The crystallite sizes of RuO were determined to be
6-10 nm. The formation of the larger RuO particles observed by XRD corroborated
the shift of the reduction temperature to lower temperature in the H -TPR profile.
o
3
5, and 54 respectively [44-46]. We attributed the emergence of these RuO
2
(1) the oxidation of the metallic Ru, forming RuO
2
2
particles into larger RuO
2
2
~
2
2
3
.3. Catalytic activity of parental Ru/C catalyst in hydrogenolysis of the β-O-4 aryl
ether bond
We tuned the Ru/C catalyst’s oxidation states under various oxidation and reduction
conditions, generating four catalysts, Ru/C, pre-reduced Ru/C, Ru/C-Red and Ru/C-Ox.
Then we characterized these catalysts by H
probed for their hydrogenolysis activity on lignin model compounds. To assess the
activity of the C -hydroxyl (C -OH) group on the hydrogenolysis of the C -O bond, we
first tested the parental Ru/C on 2-phenethyl phenyl ether (1a) and 2-phenoxy-1-
phenylethanol (1b) under N . The parental Ru/C was not active for hydrogenolysis of 2-
2
-TPR to determine the Ru content and


β
2
phenethyl phenyl ether (1a), whereas it promoted 52.5% conversion on 2-phenoxy-1-
phenylethanol (1b) (Table 1). Ethylbenzene (2) and phenol (4) were major reaction
products, confirming the occurrence of the hydrogenolysis reaction. Typically,
hydrogenolysis of the C
β
-O bond requires high H
2
pressure and metal catalysts,
1
4

including NiMo sulfide, Ni, and Pd [22, 27, 28, 56]. However, our results showed that the
OH group at the C position (C -OH) enabled cleavage of the C -O bond by Ru/C at
80°C in the absence of H
-
α

β
2
2
.
Zhang et al. observed similar products for hydrogenolysis of 2-phenoxy-1-phenylethanol
on NiMo sulfide catalysts, but H and alcohol were needed [57]. Based on our identified
reaction products (Table 1), we proposed a reaction pathway (Figure 4) wherein Ru/C
catalyzed the hydrogenolysis of C -O bond of 2-phenoxy-1-phenylethanol (1b) by
activation of C -OH. The activation of C -OH resulted in hydrogen transfer and the
hydrogenolysis of C -O bond and yielded phenol (4) and phenylethanone (5). If the
Ru/C catalyst only catalyzed hydrogenolysis of the C -O bond of 2-phenoxy-1-
2
β


β
β
phenylethanol (1b), we could only observe phenol (4) and phenylethanone (5). In our
case, the presence of 2-phenethyl phenyl ether (1a), ethylbenzene (2), styrene (3), and
p-ethylacetophenone (6) indicated that Ru also catalyzed side reactions. The presence
of styrene (3) and ethylbenzene (2) suggested that 1-phenylethanone (5) underwent
hydrodeoxygenation to form styrene as an intermediate. The styrene was then
hydrogenated to ethylbenzene.
The formation of p-ethylacetophenone (6) occurred by the alkylation of phenylethanone
(
5) and ethanol. We did not observe this product when the reaction was run in dioxane.
Moreover, in dioxane as a solvent, the selectivity toward p-ethylacetophenone (6) was
low (<4%). Ru/C also catalytically cleaved C -OH of 2-phenoxy-1-phenylethanol,

1
5

forming 2-phenethyl phenyl ether (1a) by the undesired deoxygenation pathway [58].
Cao et al. used PdCl , Pd/C, and Ru/C under 70 bar CO and found a high yield of 2-
2
2
phenethyl phenyl ether (1a) [59]. In that study, the formation of 2-phenethyl phenyl ether
+, derived from the
(
1a) resulted from the formation of the better leaving group, –OH
2
hydroxyl group at the C
59].

-OH position with the second hydrogen derived from ethanol
[
Although our result suggested that Ru/C activated the C

-OH to release its hydrogen for
hydrogenolysis of the C -O bond in an ethanol solvent, we could not rule out the
β
possibility that hydrogen was derived from ethanol. Ethanol is a polar protic solvent,
known to donate hydrogen under the reaction condition we employed. Thus, to
decouple the contribution of hydrogen from C -OH and ethanol, we ran a similar

experiment using a non-hydrogen donor solvent, the aprotic polar solvent 1,4-dioxane.
Interestingly, we obtained 52.7% conversion of 2-phenoxy-1-phenylethanol (1b), similar
to the yield in the ethanol solvent (Entry 3, Table 1). However, using dioxane as a
solvent, we found a lower selectivity toward aromatic hydrocarbon products (Σ(2,3) =
2
0.6%) compared with selectivity in ethanol (32.2%). A previous study on
hydrogenolysis of 2-phenethyl phenyl ether (1a) on Pd/C showed a 30% yield of
ethylbenzene (2) in isopropanol, but ethylbenzene is not observed in dioxane and
toluene [60]. These results demonstrated that Ru activated C
to form “hydrogen pool” for hydrogenation to cleave C -O bond [60]. Moreover, ethanol
promoted the hydrodeoxygenation reaction as shown by a higher selectivity toward

-OH, releasing hydrogen
β
1
6

aromatic hydrocarbon products compared with selectivity in dioxane. In addition,
ethanol can be obtained from renewable resources. For these reasons, we used ethanol
in the remaining studies. Next, we wanted to compare the effect of Ru content in the
Ru/C catalyst in hydrogenolysis of C
various pretreatment conditions.
β
-O bond. We tuned the Ru content by applying
3
.4. Catalytic activity of Ru and RuO catalysts in hydrogenolysis of the β-O-4 aryl
2
ether bond
To assess the effects of Ru and RuO
2
in hydrogenolysis of the C -O bond, we evaluated
β
Ru/C catalysts under various pretreatment conditions to obtain various Ru content
Table S5). For each Ru/C catalyst, we calculated the catalytic activity for reactant
(
conversion per (total) gram of Ru per unit time. The reaction using the pre-reduced
Ru/C had the highest hydrogenolysis activity of 112 mmol reactant/gRu*h, whereas the
Ru/C-Ox had the lowest hydrogenolysis activity of 4 mmol reactant/gRu*h. One reason
for this difference is that the Ru/C-Ox had the highest RuO
crystallite sizes of Ru and RuO (~19-25 nm for Ru and ~6-10 nm for RuO
aggregation and/or sintering of the small Ru and RuO particles. The large Ru crystallite
size lowered the surface of the active sites, resulting in lower catalytic activity [53].
Moreover, RuO has a weak oxygen surface bonding to bridge oxygen atoms on the
RuO surface [61]. Conversely, metallic Ru has a strong adsorption interaction with
2
content (25%) and large
2
2
) from
2
2
2
oxygen and –OH group [62-64], which promotes the hydrodeoxygenation of 2-phenethyl
phenyl ether (1a) and ethylbenzene (2). The increasing trend of the catalytic activity with
1
7

increasing metallic Ru suggested that the metallic Ru was the active phase for the
hydrogenolysis of C
β
-O bonds. Moreover, metallic Ru was the active site that activated

C -OH, releasing this “self-hydrogen” from 2-phenoxy-1-phenylethanol (1b).
The surface of RuO
air during storage/handling. Our H
under the ambient condition, changing the Ru and RuO
activity. The use of alcohol as a solvent is beneficial because alcohol can reduce the
RuO in-situ, maintaining the catalyst in the active form and enhancing the catalytic
activity over time [65]. We hypothesized that, during the reaction, RuO would be
2
on the Ru/C catalyst exists upon oxidation of Ru when exposed to
2
-TPR results also showed that Ru was passivated
2
contents and affecting catalytic
2
2
reduced in-situ when ethanol was the solvent [66, 67], enhancing the catalytic activity.
To test this hypothesis, we performed the hydrogenolysis of the 2-phenethyl phenyl
ethanol (1b) in ethanol over 12h. We determined the Ru content of the spent catalyst by
sampling the spent catalyst with the reaction products after each reaction into the U-
shaped reactors directly for H
air. We observed an increase in reactant conversion over time and reached 99.7% after
2h (Table S6). Moreover, the Ru content of the spent catalyst increased as a function
2
-TPR. By doing so, we minimized the catalyst oxidation in
1
of time and reached 3.5 % at 12h (Figure S2). The selectivity toward aromatic product
yields (Σ(2,3)) remained ~31-35% regardless of reaction time. This increase in reactant
conversion was correlated with an increase in metallic Ru from the in-situ RuO
2
reduction in ethanol.
3
.5. Catalyst stability and its recyclability
1
8

Metallic Ru was an active phase for hydrogenolysis of 1b and the Ru/C catalysts were
reduced in-situ in the presence of ethanol. To assess the stability of the Ru/C catalyst,
we recycled our catalyst four times and determined the reactant conversion and product
selectivity. We used H -TPR to determine changes in the Ru content of the spent Ru/C
2
catalysts after each recycle. We observed a slight increase in metallic Ru content after
recycles (Figure S3). The reactant conversion increased in the second recycle from
5
2.5% (fresh Ru/C) to 73.1% (after second recycle) (Table S7) and progressively
decreased to 55.5% after four recycles. With these results, we hypothesized that the
catalyst deactivated because of the Ru leaching out of the carbon support into the
solution. Thus, we performed an elemental analysis of the spent catalyst by EDS. Our
EDS results showed ~50 wt.% decrease in Ru content in the spent catalyst after four
recycles. These results suggested that the Ru was not stable in ethanol under the
investigated condition and leached out in the solution. Previous studies have shown a
similar leaching behavior of Ru into the reaction solution [68-70]. The stability of
supported Ru depends on many factors, such as Ru precursors, types of supports,
reaction
solvents,
and
interactions
between
catalyst
with
reactants/intermediates/products [71]. Further investigation is needed to identify the root
cause of Ru leaching and to improve the stability of the supported Ru for hydrogenolysis
of lignin model compounds.
4
. Conclusions
Lignin is a potential renewable aromatic feedstock. Rigidity and cross-linked lignin
polymers make lignin difficult to be cleaved, releasing monoaromatic compounds. Lignin
consists of ~50-65% β-O-4 aryl ether (C -O) bonds and abundant aliphatic and aromatic
β
1
9

hydroxyl groups. We have demonstrated that the Ru catalyst catalyzed the
hydrogenolysis of C -O bonds of a lignin model compound using the internal hydrogen
source from C -OH. These results provide an alternative and efficient strategy for lignin
conversion without a requirement for external, high-pressure H . We pretreated Ru
catalysts in various conditions and identified the metallic Ru as the active phase for
hydrogenolysis of C -O bonds. X-ray diffraction (XRD) and H Temperature-
programmed reduction (TPR) measurements supported that the Ru was the active site
for hydrogenolysis of C -O bonds. The use of alcohol as a reaction solvent enabled the
in-situ reduction of RuO from Ru/C, increasing catalytic activity over time. These results
β

2
β
2
β
2
have potential application in lignin conversion to aromatic chemicals from pulp and
paper manufacturing and from biorefineries.
Acknowledgment
This work was supported in part by the National Science Foundation under Cooperative
Agreement No. 1355438. This work was performed in part at the Conn Center for
Renewable Energy Research at the University of Louisville, which belongs to the
National Science Foundation NNCI KY Manufacturing and Nano Integration Node,
supported by ECCS-1542174. The authors thank Dr. Howard Fried for his comments
and suggestions on the manuscript.
2
0

References
[
[
[
[
[
[
[
[
[
[
[
[
1] A. Duval, F. Vilaplana, C. Crestini, M. Lawoko, Holzforschung 70 (2015) 11-20.
2] J. Hu, Q. Zhang, D.-J. Lee, Bioresour. Technol. 247 (2018) 1181-1183.
3] S. Sathawong, W. Sridach, K. Techato, J. Environ. Chem. Eng. 6 (2018) 5879-5888.
4] L. Hu, H. Pan, Y. Zhou, M. Zhang, Bioresour. 6 (2011) 3515−3525.
5] S.-H. Lee, Y. Teramoto, N. Shiraishi, J. Appl. Polym. Sci. 84 (2002) 468-472.
6] J.H. Lora, W.G. Glasser, J. Polym. Environ. 10 (2002) 39-48.
7] R. Gosselink, E. De Jong, B. Guran, A. Abächerli, Ind. Crops Prod. 20 (2004) 121-129.
8] C. Li, X. Zhao, A. Wang, G.W. Huber, T. Zhang, Chem. Rev. 115 (2015) 11559-11624.
9] Y. Zheng, D. Chen, X. Zhu, J. Anal. Appl. Pyrol. 104 (2013) 514-520.
10] C. Li, H. Bai, G. Shi, Chem. Soc. Rev. 38 (2009) 2397-2409.
11] C.A. Kelley, B.T. Hammer, R.B. Coffin, Environ. Sci. Technol. 31 (1997) 2469-2472.
12] C.H. Christensen, J. Rass‐Hansen, C.C. Marsden, E. Taarning, K. Egeblad,
ChemSusChem 1 (2008) 283-289.
13] J. Ralph, G. Brunow, W. Boerjan, Lignins. Encyclopedia of life science, Chichester: Wiley,
007.
[
2
[
[
14] S. Dutta, K.C.W. Wu, B. Saha, Catal. Sci. Technol. 4 (2014) 3785-3799.
15] J. Zakzeski, P.C.A. Bruijnincx, A.L. Jongerius, B.M. Weckhuysen, Chem. Rev. 110 (2010)
3
[
552-3599.
16] E.M. Anderson, R. Katahira, M. Reed, M.G. Resch, E.M. Karp, G.T. Beckham, Y. Román-
Leshkov, ACS Sustain. Chem. Eng. 4 (2016) 6940-6950.
[
[
[
17] K.L. Deutsch, B.H. Shanks, Appl. Catal. A 447-448 (2012) 144-150.
18] X. Huang, T.I. Korányi, M.D. Boot, E.J.M. Hensen, Green Chem. 17 (2015) 4941-4950.
19] X. Huang, O.M. Morales Gonzalez, J. Zhu, T.I. Korányi, M.D. Boot, E.J.M. Hensen, Green
Chem. 19 (2017) 175-187.
[
[
20] K.H. Kim, B.A. Simmons, S. Singh, Green Chem. 19 (2017) 215-224.
21] S. Van den Bosch, W. Schutyser, S.F. Koelewijn, T. Renders, C.M. Courtin, B.F. Sels,
Chem Commun. 51 (2015) 13158-13161.
22] J. Zhang, J. Teo, X. Chen, H. Asakura, T. Tanaka, K. Teramura, N. Yan, ACS Catal. 4
2014) 1574-1583.
23] J.M. Nichols, L.M. Bishop, R.G. Bergman, J.A. Ellman, J. Am. Chem. Soc. 132 (2010)
2554-12555.
24] T. Nimmanwudipong, R.C. Runnebaum, D.E. Block, B.C. Gates, Energy Fuels 25 (2011)
417-3427.
25] T. Nimmanwudipong, C. Aydin, J. Lu, R.C. Runnebaum, K.C. Brodwater, N.D. Browning,
D.E. Block, B.C. Gates, Catal. Lett. 142 (2012) 1190-1196.
[
(
[
1
[
3
[
[
[
[
26] X. Li, T. Guo, Q. Xia, X. Liu, Y. Wang, ACS Sustain. Chem. Eng. 6 (2018) 4390–4399.
27] Q. Song, J. Cai, J. Zhang, W. Yu, F. Wang, J. Xu, Chin. J. Catal. 34 (2013) 651-658.
28] C. Zhang, J. Lu, X. Zhang, K. MacArthur, M. Heggen, H. Li, F. Wang, Green Chem. 18
(
2016) 6545-6555.
[
[
[
[
29] A. Wang, H. Song, Biores. Technol. 268 (2018) 505-513.
30] A. Wang, D. Austin, H. Song, Fuel 246 (2019) 443-453.
31] H. Liu, E. Iglesia, J. Phys. Chem. B 109 (2005) 2155-2163.
32] S. Iqbal, S.A. Kondrat, D.R. Jones, D.C. Schoenmakers, J.K. Edwards, L. Lu, B.R. Yeo,
P.P. Wells, E.K. Gibson, D.J. Morgan, ACS Catal. 5 (2015) 5047-5059.
[
[
33] N. Duong, Q. Tan, D.E. Resasco, C.R. Chim. 21 (2018) 155-163.
34] M.J. Gilkey, P. Panagiotopoulou, A.V. Mironenko, G.R. Jenness, D.G. Vlachos, B. Xu, ACS
Catal. 5 (2015) 3988-3994.
35] A.V. Mironenko, D.G. Vlachos, JACS 138 (2016) 8104-8113.
[
2
1

[
36] M. Verziu, A. Tirsoaga, B. Cojocaru, C. Bucur, B. Tudora, A. Richel, M. Aguedo, A.
Samikannu, J.P. Mikkola, Molecular Catalysis 450 (2018) 65-76.
[
[
[
[
37] H. Li, G. Song, ACS Catalysis (2019) 4054-4064.
38] S. Aouad, E. Saab, E. Abi-Aad, A. Aboukaïs, Kinet. Catal. 48 (2007) 835-840.
39] W. Li, H. Liu, E. Iglesia, J. Phys. Chem. B 110 (2006) 23337-23342.
40] S. Hosokawa, H. Kanai, K. Utani, Y.-i. Taniguchi, Y. Saito, S. Imamura, Appl. Catal. B 45
(
[
3
2003) 181-187.
41] P. Betancourt, A. Rives, R. Hubaut, C.E. Scott, J. Goldwasser, Appl. Catal. A 170 (1998)
07-314.
[
[
[
42] G.R. Tauszik, G. Leofanti, S. Galvagno, J. Mol. Catal. 25 (1984) 357-366.
43] J. Zheng, S. Meyer, K. Köhler, Appl. Catal. A 505 (2015) 44-51.
44] T.L. Stuchinskaya, M. Musawir, E.F. Kozhevnikova, I.V. Kozhevnikov, J. Catal. 231 (2005)
4
[
1-47.
45] C. Wang, C. Tian, Y. Guo, Z. Zhang, W. Hua, W. Zhan, Y. Guo, L. Wang, G. Lu, J. Hazard.
Mater. 342 (2018) 290-296.
46] X. Huo, D.J. Van Hoomissen, J. Liu, S. Vyas, T.J. Strathmann, Appl. Catal. B Environ. 211
2017) 188-198.
47] Ž. Petrović, M. Ristić, M. Marciuš, B. Sepiol, H. Peterlik, M. Ivanda, S. Musić, Ceram. Int. 41
2015) 7811-7815.
[
(
[
(
[
[
48] M. Hiratani, Y. Matsui, K. Imagawa, S. Kimura, Thin Solid Films 366 (2000) 102-106.
49] M. Wang, W. Weng, H. Zheng, X. Yi, C. Huang, H. Wan, J. Nat. Gas Chem. 18 (2009) 300-
305.
[
[
[
50] J. Chen, J. Li, Y. Zhao, Y. Zhang, J. Hong, J. Nat. Gas Chem. 21 (2012) 673-679.
51] J. Kang, S. Zhang, Q. Zhang, Y. Wang, Angew. Chem. Int. Ed. 48 (2009) 2565-2568.
52] M. Verziu, A. Tirsoaga, B. Cojocaru, C. Bucur, B. Tudora, A. Richel, M. Aguedo, A.
Samikannu, J.P. Mikkola, Mol. Catal. 450 (2018) 65-76.
53] L. Dong, L.-L. Yin, Q. Xia, X. Liu, X.-Q. Gong, Y. Wang, Catalysis Science & Technology 8
2018) 735-745.
54] S. Ren, F. Huang, J. Zheng, S. Chen, H. Zhang, Int. J. Hydrogen Energy 42 (2017) 5105-
113.
55] B.D. Cullity, S.R. Stock, Elements of X-Ray Diffraction, 3rd ed., Pearson Education Limited,
Harlow, 2001.
56] L.-P. Xiao, S. Wang, H. Li, Z. Li, Z.-J. Shi, L. Xiao, R.-C. Sun, Y. Fang, G. Song, ACS Catal.
(2017) 7535-7542.
57] C. Zhang, J. Lu, X. Zhang, K. MacArthur, M. Heggen, H. Li, F. Wang, Green Chem. 18
2016) 6545-6555.
58] R. Barton, M. Carrier, C. Segura, J. Fierro, S. Park, H. Lamb, N. Escalona, S. Peretti, Appl.
Catal. A 562 (2018) 294-309.
[
(
[
5
[
[
7
[
(
[
[
[
[
[
[
59] Y. Cao, R. Ma, N. Wang, M.-Y. Wang, X.-D. Li, L.-N. He, J. CO2 Util 24 (2018) 328-333.
60] X. Zhou, J. Mitra, T.B. Rauchfuss, ChemSusChem 7 (2014) 1623-1626.
61] S. Wendt, M. Knapp, H. Over, JACS 126 (2004) 1537-1541.
62] A. Bjelić, M. Grilc, M. Huš, B. Likozar, Chem. Eng. J. 359 (2019) 305-320.
63] J.K. Nørskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, J.R. Kitchin, T. Bligaard, H.
Jonsson, J. Phys. Chem. B 108 (2004) 17886-17892.
[
[
64] D. Liu, Y.M. López-De Jesús, J.R. Monnier, C.T. Williams, J. Catal. 269 (2010) 376-387.
65] J. Jae, W. Zheng, A.M. Karim, W. Guo, R.F. Lobo, D.G. Vlachos, ChemCatChem 6 (2014)
848-856.
[
[
[
66] Y. Zhang, J. Yu, H. Niu, H. Liu, J Colloid Interface Sci 313 (2007) 503-510.
67] C. Cheng, D. Shen, S. Gu, K.H. Luo, Catal. Sci. Technol. 8 (2018) 6275-6296.
68] J. Garcia, H. Gomes, P. Serp, P. Kalck, J. Figueiredo, J. Faria, Carbon 44 (2006) 2384-
2391.
2
2

[
[
[
69] F. Michalek, D. Mädge, J. Rühe, W. Bannwarth, Eur. J. Org. Chem. 2006 (2006) 577-581.
70] M. Besson, P. Gallezot, Catal. Today 81 (2003) 547-559.
71] N. Grosjean, C. Descorme, M. Besson, Appl. Catal. B 97 (2010) 276-283.
2
3

Figure 1. An example of a truncated (poplar) lignin structure. This lignin model depicts
β-O-4 aryl ether bonds as the abundant linkages.
Figure 2. H -TPR profiles of Ru/C catalyst from different pretreatment conditions.
2
2
4

Figure 3. XRD patterns of Ru/C catalyst from different pretreatment conditions. Note:
metallic Ru (○) and RuO (*)
2
2
5

Figure 4. The proposed reaction pathway of hydrogenolysis of 2-phenoxy-1-
phenylethanol over Ru/C catalyst with C -OH as the hydrogen source.

2
6

Table 1. Conversion and product selectivity from hydrogenolysis of 2-phenethyl phenyl ether (1a) and 2-phenoxy-1-
phenylethanol (1b) over Ru/C catalyst.
Ru/C
o
2
80 C, N
2
Ru/C
o
+
+
+
+
+
2
80 C, N
2
1
a
Selectivity (%)
Entry
Feed
Conversion (%)
2
3
-
4
5
6
-
1a
-
1
2
3
1a
1b
-
-
-
-
52.5
52.7
25.4
13.2
6.8
7.4
34.6
33.6
8.2
3.7
-
20.1
13.6
1b*
21.9
2
7

Reaction condition: 280°C, 4 h, 8 bar N , 1 wt.% reactant/ethanol, 20 wt.% catalyst loading. * 1,4-
2
dioxane was used as a solvent.
2
8