Demethoxylation of hydrogenated derivatives of guaiacol without external hydrogen over platinum catalyst

Article abstract of DOI:10.1016/j.mcat.2019.03.019

Selective deoxygenation of 2-methoxycyclohexanone, one of the hydrogenated by-products in guaiacol hydrodeoxygenation, to phenol, cyclohexanone and cyclohexanol was investigated over carbon supported noble metal catalysts without external H₂. Pt/C exhibited the best performance and the yield of target products reached 48% in water solvent at 493 K. This system can be applied to demethoxylation of 2-methoxycyclohexanol (49% yield). Demethoxylation of guaiacol is also possible under 0.1 MPa of H₂ (46% yield). The yield of the target demethoxylation products was strongly dependent on the catalyst amount; too much catalyst decreased the yield due to the over-reaction, while the reaction stopped before total conversion of intermediates when the catalyst amount was too small. Fresh Pt/C catalyst has activity in hydrodeoxygenation of the target products and the reusability test showed deactivation of Pt/C during reaction, suggesting that deactivation at appropriate reaction progress controlled by catalyst amount is a key to good yield of the target products. In contrast to other noble metal catalysts, Pt/C has activity in both dehydrogenation of cyclohexane ring and hydrogenolysis of C–O bond, both of which contributed to the conversion of 2-methoxycyclohexanone to target demethoxylation products, according to the reactions of cyclohexanone and cyclohexanol as model substrates.

Full text of DOI:10.1016/j.mcat.2019.03.019

MolecularCatalysis471(2019)60–70
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Demethoxylation of hydrogenated derivatives of guaiacol without external
hydrogen over platinum catalyst
⁎
⁎
Akari Miyagawaa, Yoshinao Nakagawaa,b, , Masazumi Tamuraa,b, Keiichi Tomishigea,b,
a Department of Applied Chemistry, School of Engineering, Tohoku University, 6-6-07 Aoba, Aramaki, Aoba-ku, Sendai, 980-8579, Japan
b Research Center for Rare Metal and Green Innovation, Tohoku University, 468-1 Aoba, Aramaki, Aoba-ku, Sendai, 980-0845, Japan
A R T I C L E I N F O
A B S T R A C T
Keywords:
Selective deoxygenation of 2-methoxycyclohexanone, one of the hydrogenated by-products in guaiacol hydro-
deoxygenation, to phenol, cyclohexanone and cyclohexanol was investigated over carbon supported noble metal
catalysts without external H2. Pt/C exhibited the best performance and the yield of target products reached 48%
in water solvent at 493 K. This system can be applied to demethoxylation of 2-methoxycyclohexanol (49% yield).
Demethoxylation of guaiacol is also possible under 0.1 MPa of H2 (46% yield). The yield of the target de-
methoxylation products was strongly dependent on the catalyst amount; too much catalyst decreased the yield
due to the over-reaction, while the reaction stopped before total conversion of intermediates when the catalyst
amount was too small. Fresh Pt/C catalyst has activity in hydrodeoxygenation of the target products and the
reusability test showed deactivation of Pt/C during reaction, suggesting that deactivation at appropriate reaction
progress controlled by catalyst amount is a key to good yield of the target products. In contrast to other noble
metal catalysts, Pt/C has activity in both dehydrogenation of cyclohexane ring and hydrogenolysis of CeO bond,
both of which contributed to the conversion of 2-methoxycyclohexanone to target demethoxylation products,
according to the reactions of cyclohexanone and cyclohexanol as model substrates.
Platinum
Dehydrogenation
Demethoxylation
Lignin
1. Introduction
next benzene (cutting at (3)) [6–8]. While phenol is more useful than
others as a chemical feedstock because of the key compound for resin
Amid the concern that petroleum, which is indispensable in modern
society, will be depleted in the near future, biomass is attracting at-
tention as a substitute for petroleum because biomass can be used not
only for energy but also for production of chemicals [1–4]. Lig-
nocellulose is inedible biomass, and one of its main components, lignin,
possesses aromatic rings, so lignin is expected as renewable raw ma-
terial of cyclic compounds [1,2]. Fast pyrolysis is a representative lignin
conversion method and quickly produces liquid compounds (bio-oil) at
high yield [1–3]. However, bio-oil contains more oxygen than petro-
leum. Therefore, researches on hydrodeoxygenation of bio-oil have
been actively conducted using external H2 [2,2,3,4]. In such researches,
guaiacol (2-methoxyphenol) is widely used for substrate as a re-
presentative compound of bio-oil components [5–19] since guaiacol has
three types of CeO bonds existing much in bio-oil: (1)
production, selective production of phenol is not easy; although ca-
techol tends to be produced easily from guaiacol, selective conversion
from catechol to phenol is difficult because catechol hydrodeoxygena-
tion to phenol and phenol hydrodeoxygenation to benzene dissociate
similar type of CeO bond (3). Severe reaction conditions (typically
T > 523 K, PH2 > 4 MPa [5–7]) and sulfur contamination (in the case of
sulfide catalysts) are another problems [2,7]. Many researchers have
tried to develop the new catalytic system for selective production of
phenol and its hydrogenated products (cyclohexanone and cyclohex-
anol used for nylon production as KA oil) from guaiacol [8,6–19]. Re-
cently, our group reported that Ru/C + MgO [8] and Ru-MnOx/C [9]
are effective demethoxylation catalysts, giving cyclohexanol in 81%
yield at lower temperature and H2 pressure (433 K, 1.5 MPa, Fig. 1). In
this system, we obtained methanol as C1 product derived from the
methoxy group in 86% yield, which is more useful than methane. There
is one problem in this system that the ring-hydrogenation products,
mainly 2-methoxycyclohexanol, are formed as by-products; hydro-
genation tends to proceed easily at lower reaction temperature because
C
methyl−OCaromatic, (2) Caromatic−OCmethyl and (3) Caromatic −OH; order
of bond energy: (1) < (2) < (3), Fig. 1) [3,4]. Typical hydro-
deoxygenation systems such as NiMoS and CoMoS systems convert
guaiacol to catechol (cutting at (1)) and phenol (cutting at (2)) first, and
⁎ Corresponding authors at: Department of Applied Chemistry, School of Engineering, Tohoku University, 6-6-07 Aoba, Aramaki, Aoba-ku, Sendai, 980-8579,
Japan.
E-mail addresses: yoshinao@erec.che.tohoku.ac.jp (Y. Nakagawa), tomi@erec.che.tohoku.ac.jp (K. Tomishige).
https://doi.org/10.1016/j.mcat.2019.03.019
Received 2 February 2019; Received in revised form 13 March 2019; Accepted 15 March 2019
2468-8231/©2019ElsevierB.V.Allrightsreserved.

A. Miyagawa, et al.
MolecularCatalysis471(2019)60–70
Fig. 1. Guaiacol deoxygenation using external H2 and target reaction of this work.
of exothermic reaction. The demand of 2-methoxycyclohexanol is much
lower than that of phenol or the hydrogenated products (cyclohexanone
and cyclohexanol). On the other hand, 2-methoxycyclohexanol can be
produced in high yield (> 99%) with a suitable catalyst by guaiacol
hydrogenation (Fig. 1) [20,21]. Hydrogenation of guaiacol to 2-meth-
oxycyclohexanone is also possible [22,23]. Moreover, production of
cyclohexanones from cyclohexanols has been established [24–28].
Therefore, conversion of 2-methoxycyclohexanol or 2-methox-
ycyclohexanone to phenol or the hydrogenated products is desirable to
be developed; however, these reactions have not been focused as the
main target reactions in the literature.
Table 1
Dispersion of noble metal catalysts in this study.
Catalyst
Dispersion /%
Particle size (XRD) /nm
CO adsorption
XRD
Pt/C
15
24
7.5
35
17
14
6.8
6.7
8.0
17
Rh/C
Pd/C
Ru/C
a
a
–
–
a
XRD peaks due to Ru metal were not observed.
In this paper, we investigated the conversion of hydrogenated de-
rivatives of guaiacol (2-methoxycylohexanone and 2-methox-
ycyclohexanol, Fig. 1). The target products in this work are phenol,
cyclohexanone and cyclohexanol. Since the conversion of 2-methox-
ycyclohexanol or 2-methoxycyclohexanone to phenol does not consume
H2, rather produces H2, in reaction formulas (Eqs. (1) and (2)), we
carried out the reaction without expensive external H2. We explored
these reactions with various noble metal catalysts. We found that Pt/C
catalyst can convert 2-methoxycyclohexanone and 2-methox-
ycyclohexanol to target products. The catalytic behavior of each noble
metal was also investigated in detail.
After sealing the reactor, the inner gas was purged by flushing five
times with 3 MPa of high purity Ar (99.9999%) to remove residual H2
used in the catalyst reduction completely. The autoclave was then he-
ated to the reaction temperature for about 30 min and the temperature
was monitored using a thermocouple inserted in the autoclave. When
the temperature reached the target one, the time was defined as “0 h”.
During this experiment, stirring rate was fixed at 300 rpm. After an
appropriate reaction time, the reactor was cooled down and the gases
were collected in a gas bag. The autoclave contents were diluted with 2-
propanol (FUJIFILM Wako Pure Chemical Corporation, > 99.7%) con-
taining 2-methoxyethanol (FUJIFILM Wako Pure Chemical Corpora-
tion, > 99%) as an internal standard. The mixed solution was trans-
ferred to a vial, while the catalyst was separated by centrifugation and
filtration. Analyses were conducted with GC and GC–MS. The used
columns were MS-13X for H2, Porapak N for methane, carbon dioxide
and carbon monoxide, Rtx®-1 PONA for other gas components and
catechol and benzene in liquid, and TC-WAX for other liquid compo-
nents. Conversions, selectivity and carbon balance (C.B.) were calcu-
lated on carbon basis (Eqs. (3)–(5)). Amount of “others” included the
products that cannot be detected by GC, and the amount was calculated
by the difference between the carbon amounts of introduced substrate
and the total carbon amounts of the detected products.
(1)
(2)
2. Experimental section
2.1. Catalysts and activity tests
Commercially available carbon-supported noble metal catalysts
(FUJIFILM Wako Pure Chemical Corporation, Pt, Rh, Pd and Ru: 5 wt%)
were used. The dispersion of metal particle on the catalysts was de-
termined by CO adsorption and XRD, and they are shown in Table 1.
The XRD patterns are shown in Fig. S1. Catalysts were reduced in H2
flow (> 99.99% H2, 30 mL min−1) at 473 K for 30 min in a glass tube
just before use. Activity tests were performed in a 190 mL stainless steel
autoclave. The reduced catalyst (0.1 g), 2-methoxycyclohexaone
(Tokyo Chemical Industry Co., Ltd., 5 mmol, > 95%), water (20 mL)
and a magnetic spinner were introduced into an autoclave under N2
atmosphere in order to avoid exposing the pre-reduced catalyst to air.
Con version(%)
Amount of detected 2- methoxycyclohexanone[mol-C]
Amount of introduced 2- methoxycyclohexanone[mol-C]
=
1
(3)
(4)
× 100
Amount of each product[mol-C]
Amount of converted substrate[mol-C]
Selectively(%) =
× 100
61

A. Miyagawa, et al.
MolecularCatalysis471(2019)60–70
Table 2
Reaction of 2-methoxycyclohexanone over various catalysts.
Catalyst
Blank
Pt/C
Rh/C
Pd/C
Ru/C
Reaction time /h
Conversion /%
1
3
24
1
3
24
1
3
24
1
3
24
1
3
24
14
23
75
97
90
0.8
0.5
2.8
2.6
99
82
1.1
1.0
3.3
3.2
100
81
69
95
0.5
0.2
1.9
1.6
91
98
0.7
0.3
2.5
2.3
98
50
96
2.1
1.8
0.6
0.3
82
99
2.1
2.0
1.1
1.2
97
49
75
99
Carbon balance/%
104 98
93
88
96
97
94
80
H2 amount /mmol (observed) 0.0
0.0
0.0
1.0
0.9
0.2
0.1
2.9
3.2
1.3
1.5
4.1
3.8
1.3
1.0
4.0
3.3
1.3
0.7
2.0
2.0
0.1
−0.2
1.4
1.1
0.4
−0.1
2.0
1.8
3.1
3.7
3.1
1.9
(calculated)a
0.0
C1 amount /mmol (observed) 0.5
(calculated)b
Selectivity /%
Demethoxylation
phenol
0.4
< 1 < 1 < 1
16
10
5
14
14
10
13
19
16
16
4
18
4
9
8
6
1
1
3
1
6
8
5
1
1
5
1
2
2
6
cyclohexanone
cyclohexanol
Hydrogenation
2-methoxycyclohexanolc
Dehydrogenation
guaiacol
0
0
0
0
< 1
2
4
0
< 1
1
< 1
< 1
0
0
0
0
1 (0.4) 28 (0.4) 11 (0.4) 2 (0.4) 34 (0.5) 40 (0.5) 20 (0.4) 26 (0.6) 33 (0.6) 35 (0.4) 32 (0.4) 38 (0.4) 34 (0.4)
0
3
2
1
< 1
< 1
11
8
6
< 1
49
39
18
0
< 1
26
0
5
Demethylation
2-hydroxycyclohexanone
1,2-cyclohexanediolc
catechol
84
0
78
0
74
2
5
2
9
0
0
5
2
39
1 (0.0) 3 (0.1)
5 (0.3)
5
1 (0.2) 4 (0.3)
5 (0.3)
3
4 (0.1)
0
4 (0.4)
5
5 (0.1)
3
4 (0.2)
0
11 (0.4) 9 (0.7)
0
0
0
8
1
1
1
0
0
Deoxygenation
cyclohexane
benzene
0
0
0
0
0
0
1
4
2
6
5
5
0
1
< 1
5
2
0
< 1
< 1
< 1
1
0
2
1
17
< 1
< 1
< 1
C1
methanol
16
0
15
0
11
0
3
3
1
8
7
9
3
4
5
8
0
1
9
0
6
1
CH4
3
3
5
< 1
< 1
1
< 1
3
< 1
< 1
10
0
0
< 1
2
< 1
1
CO2
0
< 1 < 1
2
4
6
< 1
12
3
< 1
5
2
9
0
7
Others
0
7
12
0
12
0
21
< 1
26
4
6
20
5
11
4
31
3
(cyclopentanoned)
0
0
3
3
Reaction conditions: 2-methoxycyclohexanone 5 mmol, water 20 mL, M/C 0.1 g, PAr 1 MPa, 493 K.
a
Calculated formation amount of H2: (phenol) – (cyclohexanone) – (cyclohexanol) × 2 – (2-methoxycyclohexanol) + (guaiacol) × 2 + (catechol) × 2 – (1,2-
cyclohexanediol) – (cyclohexane) × 3 + (cyclopentanone) × 2 + (cyclopentanol) – (n-hexane) × 4 + (anisole) – (methoxycyclohexane) × 2 – (methane) + ((CO2
production amount) – (cyclopentanone) – (cyclopentanol)) × 3.
b
Calculated formation amount of C1: (Demethoxylation products) + (Total deoxygenation products) + (Demethylation products) + (Decarboxylation products
(cyclopentanone + cyclopentanol)).
c
d
The cis/(cis+trans) is shown in the parentheses.
C5 products are included in “Others”.
2.2. Characterization
Carbon balance(C.B.)(%)
Amount of all detected compounds[mol-C]
Amount of introduced 2-methoxycyclohexanone[mol-C]
=
× 100
Catalysts used for conversion of 2-methoxycyclohexanone were re-
covered as the samples of the characterization measurements by the
following method. First, we conducted the activity tests under standard
reaction conditions (2-methoxycyclohexanone 5 mmol, catalyst 0.1 g,
water 20 mL, PAr 1 MPa, 493 K, 24 h). Next, we collected all the reaction
mixture using 2-propanol as a dilution solvent. The catalysts were
collected by centrifugation and heated at 383 K for 12 h.
(5)
The method of catalyst reuse was as follows. First, we conducted
normal activity test with fresh catalyst. Next, we collected all the re-
action mixture using 2-propanol as a dilution solvent. The used catalyst
was recovered by centrifugation and washed with 30 mL of water (3
times). These recovery and washing processes were operated in glove
bag to avoid exposure of the catalyst to air. About 90% of the catalyst
was recovered in this method and the obtained catalysts were used for
next reaction.
X-ray diffraction (XRD) patterns were recorded by
a Rigaku
MiniFlex 600 diffractometer. Cu Kα (λ = 0.154 nm, 40 kV, 15 mA) ra-
diation was used as an X-ray source. The samples were reduced under
H2 flow (30 mL min−1) for 30 min at 473 K and passivated by 1%O2/He
flow (10 mL min−1) for 30 min at room temperature before measure-
ment for fresh catalysts.
Reactions of other substrates were carried out by using guaiacol
(Tokyo Chemical Industry Co., Ltd., > 98%), 2-hydroxycyclohexanone
dimer (Sigma-Aldrich, > 97%), trans-1,2-cyclohexanediol (Tokyo
Chemical Industry Co., Ltd., > 99%), catechol (Tokyo Chemical
Industry Co., Ltd., > 99%), phenol (FUJIFILM Wako Pure Chemical
Corporation, > 99%), cyclohexanone (Tokyo Chemical Industry Co.,
Ltd., > 99%), cyclohexanol (FUJIFILM Wako Pure Chemical
Transmission electron microscope (TEM) images were taken with
HITACHI HD-2700. Fresh catalysts after reduction under H2 (30 mL
min−1) at 473 K were used as samples for the TEM observation.
Supersonic waves dispersed the sample in ethanol. The samples were
placed on Cu microgrids for TEM under air atmosphere. Average par-
ticle size was calculated by Σnidi3/Σnidi2 (di: particle size, ni: number of
particles with size di).
Corporation, > 98%), and
methanol
(Kanto
Chemical
Co.
Inc., > 99.8%). 2-Methoxycyclohexanol was obtained by hydrogena-
tion of guaiacol (reaction conditions: guaiacol 5 g, tetrahydrofuran
(FUJIFILM Wako Pure Chemical Corporation, > 99.5%) 20 mL, Pd/C
1 g, 438 K, 72 h) and tetrahydrofuran used as a solvent was removed by
distillation. The purity of the obtained 2-methoxycyclohexanol was
98% and the cis/(cis + trans) ratio was 0.33 on GC area basis.
Inductively-coupled plasma atomic emission spectrometry (ICP-
AES, ThermoFisher iCAP6500) was carried out to measure the Pt
amount in the solution after reaction and filtration.
The amount of CO chemisorption was measured by MicrotracBEL
BELCAT II by a pulse adsorption method. The catalyst (0.05 g) in the
measurement cell was reduced under flowing H2 (30 mL min−1) at
62

A. Miyagawa, et al.
MolecularCatalysis471(2019)60–70
473 K for 30 min. After cooling to room temperature, 10% CO/He was
introduced as pulse gas at 308 K. In the calculation of the dispersion of
noble metals, one surface metal atom was assumed to adsorb one CO
molecule (CO/M = 1).
(12)
(13)
TG-DTA was carried out with Rigaku Thermo Plus EVO-II instru-
ment under N2 flow (100 mL min−1) using 0.01 g of sample at a heating
rate of 10 K min−1. Fresh catalysts were measured after heating at
473 K for 0.5 h under N2 flow (30 mL−1) to remove water contained in
the catalysts.
2-Methoxycyclohexanol, the hydrogenation product (Eq. (10)), was
produced with 28% selectivity at 1 h and the amount was decreased
after 1 h. Demethylation products (2-hydroxycyclohexanone (Eq.
(6)),1,2-cyclohexanediol (Eq. (11)) and catechol (Eq. (8))) were also
detected at first and then decreased. Considering that hydrolysis of 2-
methoxycyclohexanone to 2-hydroxycyclohexanone and methanol (Eq.
(6)) can proceed without a catalyst as described above, most de-
methylation products were probably formed via hydrolysis. Amounts of
guaiacol (dehydrogenation product, Eq. (7)) and total deoxygenation
products (benzene (Eq. (12)) and cyclohexane (Eq. (13)) were not large.
Among C1 products, methanol was produced by hydrolysis (Eq. (6));
however, hydrogenolysis of ReOCH3 bond (demethoxylation) can also
give methanol (Eq. (14)). Hydrogenolysis of ReOCH3 bond is a pre-
ferable reaction because it gives target demethoxylation products.
Methane can be formed by hydrogenolysis of ROeCH3 bond (Eqs. (15)
and (16)) or hydrocracking of hydrocarbons. Carbon dioxide can be
produced by aqueous phase reforming (APR) of methanol (Eq. (9)) or
decarbonylation + water gas shift (Eq. (17)) [32]. Water gas shift re-
action is another source of H2.
3. Results and discussion
3.1. Optimization of catalyst and reaction conditions
The reactions of 2-methoxycyclohexanone were carried out at 493 K
in water solvent under Ar atmosphere (Table 2). Without catalyst, 2-
methoxycyclohexanone was converted slowly (14% conversion at 1 h),
and 2-hydroxycyclohexanone was mainly produced. As a C1 product,
methanol was produced almost stoichiometrically with 2-hydro-
xycyclohexanone, which means that 2-hydroxycyclohexanone was
produced by hydrolysis of 2-methoxycyclohexanone (Eq. (6)). The
target products (demethoxylation products: phenol, cyclohexanone and
cyclohexanol) were hardly produced.
(6)
In the presence of noble metal catalysts, conversion was increased,
and the selectivity was also changed. Formation of H2 was observed,
and possible formation route of H2 included dehydrogenation of cy-
clohexane ring (Eqs. (7) and (8)) and aqueous phase reforming (APR) of
methanol (eq. (9)).
(14)
(15)
(16)
(7)
(8)
After all, there may be many reaction routes in this 2-methox-
ycyclohexanone conversion; later we discuss the reaction scheme in
detail. Here, the production amount of H2 and sum of C1 was calculated
(17)
based on Eqs. (6)–(17), and the values were compared with the actually
observed ones (Table 2). The observed amount almost agreed with the
calculated one for both H2 and C1 in most runs. Next to Pt/C, Rh/C
promoted 2-methoxycyclohexanone conversion and production of
target demethoxylation products production. Pd/C and Ru/C gave
much smaller amount of target demethoxylation products. When Pd/C
was used, the main products were guaiacol (Eq. (7)) and 2-methox-
ycyclohexanol (Eq. (10)). Guaiacol/2-methoxycyclohexanol ratio was
high at 1 h, and then decreased at longer reaction time. This trend in-
dicates that Pd/C promoted dehydrogenation of 2-methoxycyclohex-
anone to guaiacol (Eq. (7)) and the hydrogenation of another 2-meth-
oxycyclohexanone with produced H2 (Eq. (10)). In the case of Ru/C,
guaiacol was hardly produced, and the main products were 2-hydro-
xycyclohexanone (Eq. (6)) and 2-methoxycyclohexanol (Eq. (10)). 2-
Hydroxycyclohexanone was probably produced by non-catalyzed hy-
drolysis (Eq. (6)). Ru/C seems to have little activity in dehydrogenation
of 2-methoxycyclohexanone (Eq. (7)) but have activity in hydrogena-
tion (Eqs. (10) and (11)) and APR (Eq. (9)). Regarding the production
amount of H2 and sum of C1 at long reaction time (24 h) over Ru/C, the
observed C1 amount was larger than the calculated one and the ob-
served H2 amount was smaller. This difference is probably due to the
hydrocracking. Anyway, Pt/C showed the highest selectivity to target
(9)
When Pt/C was used for the reaction, 2-methoxycyclohexanone was
quickly reacted and conversion reached 97% at 1 h. The target de-
methoxylation products were observed significantly, and the total yield
reached 48% (phenol 13%, cyclohexanone 19%, cyclohexanol 16%) at
24 h. Other identified products were 2-methoxycyclohexanol (a hy-
drogenation product, Eq. (10)), guaiacol (a dehydrogenation product,
eq. 7) 2-hydroxycyclohexanone and its hydrogenation/dehydrogena-
tion products (demethylation products, Eqs. (6), (11) and (8)), benzene
and cyclohexane (total deoxygenation products, Eqs. (12) and (13)) and
C1 products (methanol, methane and carbon dioxide). Anisol, meth-
oxycyclohexane, C2-C5 hydrocarbons, n-hexane and cyclopentanone
were detected a little and they were included in others. However, most
of “others” were the undetected products by GC.
(10)
(11)
63

A. Miyagawa, et al.
MolecularCatalysis471(2019)60–70
Fig. 2. Effect of reaction temperature on the reaction of 2-methoxycyclohexanone over Pt/C.
Reaction conditions: 2-methoxycyclohexanone 5 mmol, water 20 mL, Pt/C 0.1 g, PAr 1 MPa. Detailed data are shown in Table S1.
demethoxylation products, and we selected Pt/C as catalyst in the fol-
lowing study.
was used, further deoxygenation of target demethoxylation products to
cyclic hydrocarbons (i.e. over-reaction) significantly proceeded and
decreased the yield of target products. In short, the reaction stopped
before total conversion of intermediates with too small amount of
catalyst, and too much catalyst promotes over-reaction of target pro-
ducts. Deactivation of the catalyst seems to be a key factor for this
phenomenon: smaller the amount of catalyst is, faster deactivation
occurs and reaction stops. Deactivation was investigated in the next
section. Anyway, 0.1 g of Pt/C was found to be the best catalyst amount
for this reaction. To summarize this section, the best catalyst is Pt/C and
the best conditions are 493 K, water solvent and 0.1 g of catalyst (with
5 mmol 2-methoxycyclohexanone substrate.)
The effect of reaction temperature was investigated in the range of
463–513 K using Pt/C catalyst (Fig. 2). The detailed data are shown in
Table S1. At low temperature such as 463 K, the main product is 2-
methoxycyclohexanol, where H2 can be supplied by the formation of
guaiacol or phenol. At 483 or 493 K, the yield of target products become
maximum (ca. 50%) at 24 h and the yield of 2-methoxycyclohexanol is
rather small. At higher temperatures (503 and 513 K), the formation of
total deoxygenation products was significant and the yield of the target
demethoxylation products was lower than that at 493 K. The yield of
“others” including undetected products by GC increased with higher
reaction temperature and longer reaction time. Produced catechol and
phenol, highly dehydrogenated compounds, may polymerize at high
temperature [8,29–31]. As a result, we selected 493 K as the reaction
temperature.
3.2. Performance of Pt/C
Fig. 4 shows the time course of reaction of 2-methoxycyclohexanone
over 0.1 g of Pt/C at 493 K in water solvent. The detailed data are
shown in Table S4. Conversion at 0 h means the reaction during the
heating from room temperature to the target one which took about
30 min. The conversion increased rapidly to 99% and selectivity of
target demethoxylation products also increased rapidly during 3 h.
From 3 h to 24 h, the selectivity increased gradually, and highest yield
of 48% (phenol 13%, cyclohexanone 19%, cyclohexanol 16%) was
obtained at 24 h. Deoxygenation of target products and methanol to
hydrocarbons proceeded slowly after 24 h.
The effect of solvents was tested by using water, ethanol, n-dode-
cane and toluene as well as none (solventless condition) (Table S2).
Water solvent is by far better solvent to obtain target demethoxylation
products. In the case of ethanol solvent, large amount of C1 products
and H2 were produced from ethanol and the main identified product
from substrate was 2-methoxycyclohexanol, probably because of the
large supply of H2 from the solvent. In the cases of n-dodecane solvent,
toluene solvent and solventless condition, the main products were 2-
methoxycyclohexanol and guaiacol, and the produced H2 was smaller
than the case of water solvent. As mentioned before, H2 was supplied to
the system via dehydrogenation (Eqs. (7) and (8)) and APR of methanol
(Eq. (9)) mainly, and APR of methanol only proceeds in water solvent.
Because H2 was needed for target demethoxylation, the smaller amount
of H2 in the non-water solvent systems except ethanol can limit the
formation of the target demethoxylation products. Carbon monoxide
which is detected in non-water solvent, especially ethanol, is another
factor for suppressing the formation of the target products because
carbon monoxide easily covers to the Pt metal surface. In water solvent,
however, active Pt site can be regenerated by the conversion of ad-
sorbed carbon monoxide and water to carbon dioxide and H2, i.e. water
gas shift reaction.
Next, Pt/C catalyst was applied to related substrates (2-methox-
ycyclohexanol and guaiacol) and the results are shown in Table 3 and
Table S5. In the case of guaiacol, small amount of H2 was added to the
gas phase because it is difficult to supply H2, which is needed for hy-
drogenolysis, by dehydrogenation of guaiacol. The target demethox-
ylation products were obtained from both substrates in similar total
yields: phenol 1%, cyclohexanone 15% and cyclohexanol 33% from 2-
methoxycyclohexanol (Entry 4); phenol 41%, cyclohexanone 4% and
cyclohexanol 1% from guaiacol (Entry 6). Pt/C is effective to de-
methoxylation of both guaiacol and hydrogenated guaiacol. The highest
yield of target demethoxylation products from 2-methoxycyclohexanol
was obtained at shorter time than the case of 2-methoxycyclohexanone
(Entries 2 and 4). This trend suggests that 2-methoxycyclohexanol is an
intermediate of demethoxylation. The reaction route will be discussed
in detail in later section.
Fig. 3 shows the reaction results using different amount of Pt/C
(0.02-0.20 g). The detailed data are shown in Table S3. When we used
smaller amount of Pt/C, 2-methoxycyclohexanone reacted very slowly,
and the yield of the target demethoxylation products did not increase
even in 48 h reaction. On the other hand, when a large amount of Pt/C
The reuse experiments were conducted to check the stability of Pt/C
(Fig. 5 and Table S6). The used Pt/C was collected, washed with water
64

A. Miyagawa, et al.
MolecularCatalysis471(2019)60–70
Fig. 3. Effect of catalyst amount on the reaction of 2-methoxycyclohexanone over Pt/C.
Reaction conditions: 2-methoxycyclohexanone 5 mmol, water 20 mL, Pt/C 0.02-0.2 g, PAr 1 MPa, 493 K. Detailed data are shown in Table S3.
and used for next reaction without exposure to air. The used catalyst
showed good conversion, but selectivity of target demethoxylation
products much decreased at first reuse, and further slightly decreased in
subsequent reuses. In this reuse method, some amount of catalyst was
lost during the recovery process, and 0.07 g catalyst remained after
third reuse from initial amount of 0.1 g. About 10% of catalyst can be
lost each use. Here, the performance of the reused catalyst is compared
with the data for the effect of catalyst amount (Fig. 3). The performance
in first reuse (ca. 20% yield of demethoxylation products) was similar to
that of 0.05 g Pt/C, and that in third reuse (12% yield) was comparable
to that of 0.02 g Pt/C. The comparison indicates that the catalyst was
deactivated during the reaction.
Pt is negligible.
The amount of CO chemisorption was measured to count the
number of active surface metal atoms (Table 4). Pt dispersion of Pt/C
after reduction was 12% while dispersion of Pt/C after reaction was 2%.
On the basis that the particle size of Pt was almost constant during the
reaction, the deactivation can be due to the coverage of Pt surface with
some carbonaceous species such as polymeric by-products. Highly de-
hydrogenated compounds such as catechol, phenol and diketones can
polymerize at high temperature [8,29–31]. The deposition of organic
substance on the catalyst was investigated with TG-DTA. Although it is
difficult to measure the amount of deposit accurately because the ac-
tivated carbon support loses some weight by thermal decomposition,
the weight loss of the used Pt/C was clearly larger than that of fresh Pt/
C above 500 K (Fig. S3).
The deactivation behavior and selectivity change were investigated
in detail. Highly deactivated Pt/C gave much amount of 2-methox-
ycyclohexanol (Fig. 5). 2-Methoxycyclohexanol is the main product in
very short reaction time under the standard reaction conditions using
fresh catalyst (Fig. 4). In the reactions with a smaller amount of Pt/C
(Fig. 3), the main product was also 2-methoxycyclohexanol and the
reaction almost stopped at around 3 h. In the case of large Pt/C amount,
cyclic hydrocarbons can be produced at 3 h. These data indicate that
deoxygenation activity of Pt/C was lost during use (deactivation). Even
in the standard run (0.1 g of Pt/C) the deactivation in deoxygenation
proceeded to some extent. Moderate deactivation seems to be vital to
obtain target demethoxylation products.
These characterization results can make the reason for the deacti-
vation of Pt/C clear. The results of XRD and TEM indicate that the size
of Pt particle size was unchanged during the reaction. So, aggregation is
not the reason of the deactivation. ICP-AES showed leaching did not
occur. On the other hand, the results of CO adsorption and TG-DTA
show that organic compounds are deposited on the surface of used Pt/
C. Carbon supported platinum catalyst is difficult to regenerate by
calcination because of combustibility of carbon support. Catalysts with
higher resistance to the deactivation and regeneration ability should be
developed in future.
3.3. Characterization
3.4. Formation route of the target products
Pt/C catalyst was characterized by XRD, TEM and CO adsorption.
XRD patterns of Pt/C after reduction, after 24 h reaction and after third
reuse are shown in Fig. 6. The sharp peaks at 39.8°, 46.3° and 67.5° can
be assigned to Pt°. Pt particle size was estimated to be 6.7, 7.2 and
7.1 nm for the catalysts after reduction (a), after 24 h reaction (b) and
after third reuse (c), respectively, by using Scherrer’s equation. These
values were similar, which indicates that the deactivation of Pt/C was
not due to the change of particle size. The almost constant particle size
during catalytic use was also confirmed by TEM analysis (Figs. 7 and
S2): the average Pt particle size was 6.4 nm after reduction (a) and
6.7 nm after reaction (b). These values agreed with the XRD data.
ICP-AES showed that the leached amount of Pt to liquid phase
during the 24 h reaction was < 0.1%, which indicates that leaching of
To clarify the reaction route from 2-methoxycyclohexanone to
target products (Fig. 8), we carried out the reaction tests of various
related substrates (Table 5 (C7 compounds) and 6 (C6 compounds)). In
order to discuss the selectivity in detail, the reaction time was set to be
very short (0 h; only the heating to 493 K). The heating time was about
30 min for all runs. In 2-methoxycyclohexanone reaction, H2 was absent
at first but it was produced over time to about 1.3 mmol. Then, another
type of initial gas conditions was applied to these experiments in order
to evaluate the actual reactivity: a little amount of H2 (0.1 MPa = ca.
7 mmol) in 0.9 MPa of Ar, in addition to Ar atmosphere (1.0 MPa) as
usual.
When 2-methoxycyclohexanone was used as a substrate, the main
products under Ar atmosphere were 2-methoxycyclohexanol, 2-
65

A. Miyagawa, et al.
MolecularCatalysis471(2019)60–70
Fig. 5. Reuse of Pt/C in 2-methoxycyclohexanone conversion.
Reaction conditions: 2-methoxycyclohexanone 5 mmol, water 20 mL, Pt/C 0.1 g
(fresh), PAr 1 MPa. 493 K, 24 h. Detailed data are shown are Table S6.
hydroxycyclohexanone and guaiacol (selectivities were 31%, 25% and
18%, respectively) (Table 5, Entry 1). 2-Hydroxycyclohexanone can be
produced in two routes: (1) hydrogenolysis of C–O (Eq. (18)) or (2)
hydrolysis (Eq. (6)).
(18)
In the former route, methane is produced as C1 product while me-
thanol is produced in the latter route. In fact, methanol was detected
mainly and methane was hardly detected, which indicates that hydro-
lysis was the main route in 2-hydroxycyclohexanone production. Saying
more about 2-hydroxycyclohexanone, it was produced under blank
conditions along with methanol (Table 2). From these data, we can say
that the first step of 2-methoxycyclohexanone conversion can be de-
hydrogenation and hydrolysis, and the produced H2 by dehydrogena-
tion was quickly consumed in hydrogenation.
Fig. 4. Time course of reaction of 2-methoxycyclohexanone over Pt/C. (a)
cyclic products, (b) C1 products and H2.
Reaction conditions: 2-methoxycyclohexanone 5 mmol, water 20 mL, Pt/C
0.1 g, PAr 1 MPa, 493 K. Detailed data are shown in Table S4.
Table 3
Reaction of guaiacol derivatives over Pt/C.
Entry
Substrate
Time /h
C.B.c /%
Conv. /%
Selectivityd /%
H2 amount /mmol
CH3OH
CH4
CO2
1
1
86
81
90
79
80
78
97
100
40
97
94
99
30
48
38
51
43
46
27 (0.4)
3
–
–
6
12 (0.1)
3 (0.2)
10 (0.0)
0
4
3
1
2
1
4
2
3
5
3
3
6
7
2
6
6
7
1
2
0.8
1.3
2.2e
0.0e
0.7
0.4
2
24
1
2 (0.4)
- (0.4)
- (0.5)
1 (0.4)
0
< 1
10
4
3a
4a
5b
6b
1
0
–
–
9
< 1
12
5
1
1
0
17 (0.6)
6 (-)
3
11
493 K. Full list of product compounds is shown in Table S4.
a
b
c
d
e
Isomeric mixture (cis/(cis+trans) = 0.33) was used as substrates.
Reaction conditions: substrate 5 mmol, water 20 mL, Pt/C 0.1 g, PAr 1.0 MPa or Ptotal 1.0 MPa (PH2 0.1 MPa (7 mmol) + PAr 0.9 MPa).
Carbon balance.
The cis/(cis+trans) is shown in the parenthese.
Included initial H2.
66

A. Miyagawa, et al.
MolecularCatalysis471(2019)60–70
Table 4
Characterization results of used Pt/C.
State
CO adsorption
Dispersion /%
Particle size /nm
amount
/cm3 gcat
−1
CO adsorption XRD XRD
TEM
After reduction 0.81
15
17
16
6.7
7.2
6.4
6.7
After reactiona
a
0.11
2.0
Reaction conditions: 2-methoxycyclohexanone 5 mmol, water 20 mL, Pt/C
0.1 g, PAr 1 MPa, 493 K, 24 h (shown in Table 2).
Fig. 6. XRD patterns of Pt/C catalysts.
(a) After reduction, (b) after reaction (Table 2, Pt/C, 24 h reaction) and (c) after
reuse for 3 time (Fig. 5).
Fig. 8. Proposed reaction pathways of demethoxylation of 2-methox-
ycyclohexanone over Pt/C.
(19)
Another route to C6 products is hydrolysis of 2-methoxycyclohex-
anone to 2-hydroxycyclohexanone which proceeds slowly in homo-
genous aqueous phase as discussed in Section 3.1.
Next, we investigated the route to target products (phenol, cyclo-
hexanone and cyclohexanol) via 2-hydroxycyclohexanone. When 2-
hydroxycyclohexanone was used as a substrate (Table 6, Entries 1 and
2), it was mainly converted to 1,2-cyclohexanediol and catechol which
are hydrogenation (Eq. (11)) and dehydrogenation products (Eq. (8)),
respectively. 1,2-Cyclohexanediol was converted to 2-hydro-
xycyclohexanone (a dehydrogenation product) and cyclohexanol (a
deoxygenation product) (Table 6, Entry 3 and 4). Catechol was con-
verted to 1,2-cyclohexanediol (a hydrogenation product) and phenol (a
deoxygenation product) under H2 containing conditions (Entry 6).
From these data, the reaction routes from 2-methoxycyclohexanone to
target products (phenol, cyclohexanone and cyclohexanol) include: 2-
methoxycyclohexanone → 2-hydroxycyclohexanone → 1,2-cyclohex-
Fig. 7. TEM images of Pt/C catalysts.
(a) After reduction, (b) after reaction (Table 2, Pt/C, 24 h reaction). The his-
tograms are shown in Fig. S2.
2-Methoxycyclohexanol conversion was slow and main products
were guaiacol and 2-methoxycyclohexanone under Ar atmosphere
(Table 5, Entry 3). Under H2 containing conditions, little products were
detected (Table 5, Entry 4). Only dehydrogenation seems to proceed
from 2-methoxycyclohexanol. When guaiacol was used as a substrate,
phenol and 2-methoxycyclohexanol were mainly produced under H2
containing conditions (Table 5, Entry 6). Without initial H2, the con-
version of guaiacol was low (Table 5, Entry 5). From these data, one
route from 2-methoxycyclohexanone to C6 products can be 2-methox-
ycyclohexanone → guaiacol → phenol (Eq. (19)).
anediol
→ cyclohexanol (route B; hydrolysis + hydrogenation +
deoxygenation, Eq. (20)) and 2-methoxycyclohexanone → 2-hydro-
xycyclohexanone
→ catechol → phenol (route C; hydrolysis +
dehydrogenation + deoxygenation, Eq. (21)), as well as 2-methox-
ycyclohexanone → guaiacol → phenol route (route A; dehydrogena-
tion + demethoxylation, Eq. (19)) as shown above (Fig. 8).
67

A. Miyagawa, et al.
MolecularCatalysis471(2019)60–70
Table 5
Reaction of C7 intermediates over Pt/C.
Entry
Substrate
Gas condition
Conv.
Selectivityb /%
H2 amount
C. B.d
CH3OH
CH4
CO2
1
2
Ar
H2+Ar
Ar
15
20
4
31 (0.5)
53 (0.6)
- (0.2)
- (0.3)
0
18
7
–
–
25
17
1
4
3
5
3
6
5
0
2
0
1
< 1
< 1
1
8
6
1
0
0
9
< 1
< 1
2
1
0
1
1
0
0
0.1
102
100
110
105
96
6.2c
0.4
3a
4a
5
17
0
60
48
0
5
5
H2+Ar
Ar
< 1
4
0
0
30
0
< 1
0
1
8.4c
0.0
–
0
0
0
6
H2+Ar
12
30 (0.6)
–
2
0
37
6
2
2
5.9c
95
Reaction conditions: substrate 5 mmol, water 20 mL, Pt/C 0.02 g, PAr 1.0 MPa or Ptotal 1.0 MPa (PH2 0.1 MPa (7 mmol) + PAr 0.9 MPa), 493 K, 0 h.
a
b
c
d
Isomeric mixture (cis/(cis+trans) = 0.3) was used as a substrate.
The cis/(cis+trans) is shown in the parentheses.
Including initial H2 (7 mmol).
Carbon balance.
decarbonylation from C6 or C7 compounds (Eq. (17)). In fact, cyclo-
pentanone, which is reported to be decarbonylation product of guaiacol
[32–34], was detected a little (Table 2); however, the yield of cyclo-
pentanone and its related compounds was much lower than the carbon
dioxide yield in mol. The other route is aqueous phase reforming (APR)
of methanol (Eq. (9)). In APR reaction, methanol and water are con-
verted to carbon dioxide and H2, and this reaction has been indeed
reported to be catalyzed by Pt at 493 K [35–41]. In our system, water
solvent was used in this 2-methoxycyclohexanone conversion. Besides,
the time course data (Fig. 4 (b)) show that the decrease of methanol can
be correlated with increase of carbon dioxide and H2. Table 8 shows
reactions using methanol as substrate. Carbon dioxide was produced
much, in fact, especially in the case without H2 (Entries 2 and 4).
Therefore, the formation of carbon dioxide was mainly due to APR of
methanol and the contribution of decarbonylation of C6 and C7 com-
pounds was small.
(20)
(21)
Table 7 shows the results of reactions of target products. In the case
of phenol, the main reactions are hydrogenation to cyclohexanol and
cyclohexanone, and in addition, C–O cleavage to benzene also pro-
ceeded (Table 7, Entry 1, Eq. (12)). Cyclohexanone gave phenol and
cyclohexanol, and C–O cleavage products were hardly produced. Cy-
clohexanol gave C–O cleavage product (cyclohexane, Eq. (13)) as well
as cyclohexanone when H2 was present (Table 7, Entry 5). These data
show that cyclic hydrocarbons such as benzene and cyclohexane are
produced from phenol and cyclohexanol and not directly from cyclo-
hexanone.
For the formation of methane, there are three routes: demethylation
(Eqs. (15) and (18)), hydrogenolysis of methanol (Eq. (16)) and
cracking. As discussed above, the main reaction routes from 2-meth-
oxycyclohexanone to demethoxylation products do not involve de-
methylation. Cracking produces methane from C6 ring moiety of the
substrate, and when cracking is involved significantly, observed C1
amount will shift toward larger than converted methoxy group amount.
As shown in Table 2, observed C1 amount was similar to that calculated
by converted methoxy group over Pt/C catalyst, indicating that
Now we discuss the formation of each C1 product. All the reaction
routes (A (Eq. (19)), B (Eq. (20)) and C (Eq. (21)) produce methanol as
C1 product. However methane and carbon dioxide were actually also
produced in the reaction of 2-methoxycyclohexanone (Fig. 4(b)). The
amount of methanol increased at first, but decreased gradually.
Meanwhile, yields of carbon dioxide and methane remained to increase
with time.
There are two formation route of carbon dioxide. One is
Table 6
Reaction of C6 intermediates over Pt/C.
Entry
Substrate
Gas Conditions
Conv. /%
Selectivityb /%
H2 amount /mmol
C. B.d /%
CH3OH
CO2
1
Ar
69
91
41
40
4
–
55 (0.4)
32
3
9
3
< 1
0
0
1
1
0
0
0
0.0
97
2
H2+Ar
Ar
–
88 (0.4)
- (0.2)
2
4
2
< 1
< 1
< 1
0
4.6c
1.1
84
3a
4a
5
61
27
0
6
9
13
14
0
8
97
7H2+Ar
Ar
- (0.3)
0
0
53
0
7.5c
0.0
97
37 (1.0)
31 (0.4)
–
0
101
102
6
H2+Ar
30
10
–
37
10
8
< 1
6.7c
Reaction conditions: substrate 5 mmol, water 20 mL, Pt/C 0.02 g, PAr 1.0 MPa or Ptotal 1.0 MPa (PH2 0.1 MPa + PAr 0.9 MPa), 493 K, 0 h.
a
b
c
d
trans-1,2-Cyclohexanodiol was used as a substrate.
The cis/(cis+trans) is shown in the parentheses.
Including initial H2 (7 mmol).
Carbon balance.
68

A. Miyagawa, et al.
MolecularCatalysis471(2019)60–70
Table 7
Reaction of target products over Pt/C.
Entry
Substrate
Gas condition
Conv. /%
70
Selectivity /%
H2 amount /mmol
C.B.b /%
1
H2+Ar
–
33
44
12
3
4.6a
97
2
3
4
5
Ar
52
65
14
16
41
9
–
55
84
–
2
2
0.4
88
H2+Ar
Ar
–
3
4
6.1a
0.5
90
2
84
52
< 1
0
2
108
93
H2+Ar
0
–
42
9.7a
Reaction conditions : substrate 5 mmol, water 20 mL, Pt/C 0.02 g, PAr 1.0 MPa or Ptotal 1.0 MPa (PH2 0.1 MPa + PAr 0.9 MPa), 493 K, 0 h.
a
b
Including initial H2 (7 mmol).
Carbon balance.
Table 8
Reaction of methanol over Pt/C.
Entry Gas condition Time /h Conv. /% Yield /%
H2 amount C.B.b /%
/mmol
CH4 CO2
1
2
3
4
H2+Ar
Ar
1
25
30
80
85
0
0
2
2
6
6.5a
4.6
81
91
74
68
1
20
51
51
H2+Ar
Ar
24
24
15.5a
9.1
Reaction conditions : methanol 5 mmol, water 20 mL, Pt/C 0.1 g, PAr 1.0 MPa or
Ptotal 1.0 MPa (PH2 0.1 MPa + PAr 0.9 MPa), 493 K, 1 or 24 h.
a
b
Including initial H2 (7 mmol).
Carbon balance.
cracking can be a minor route. As for hydrogenolysis of methanol in
methane formation (Eq. (16)), methane was rarely observed in me-
thanol conversion even in long time reaction as shown in Table 8. This
indicates that hydrogenolysis of free methanol hardly proceeds. The
involvement of methanation of carbon dioxide can be also ruled out
since the selectivity to methane from methanol under H2 + Ar was very
small at long reaction time and that to carbon dioxide was high
(Table 8, Entry 3). Then, the reaction results of related substrates
(Tables 3 and 5) are closely re-considered. High selectivity to methane
was observed when guaiacol was the substrate at longer reaction time
(1 and 3 h; Table 3, Entries 5 and 6). On the other hand, selectivity to
methane was much lower at short reaction time (0 h; Table 5, Entries 5
and 6). Considering that the reaction temperature for 0 h reaction was
practically lower than longer reaction time runs, methane can be
formed with demethylation of guaiacol at high reaction temperature;
i.e. hydrogenolysis of adsorbed methanol produced by demethoxylation
can proceed before the desorption [42].
Fig. 9. Reaction of cyclohexanone under Ar atmosphere.
Reaction conditions: cyclohexanone 5 mmol, water 20 mL, M/C 0.1 g, PAr
1 MPa. Detailed data are shown are Table S7. aCarbon balance.
CeO bond was investigated by the reaction of cyclohexanone to phenol
(under Ar atmosphere) and that of cyclohexanol to cyclohexane (under
H2 + Ar atmosphere).
As shown in Fig. 9 (and Table S7), cyclohexanone was converted to
phenol over Pt/C, Rh/C and Pd/C but hardly converted over Ru/C. This
trend clearly shows that Ru/C had little activity in dehydrogenation of
cyclohexane ring. Pt/C catalyzed cyclohexanone conversion well, but
the produced H2 and phenol reacted additionally to give cyclohexanol
(hydrogenation) or benzene (hydrogenolysis). Formation of benzene
(and cyclohexane) was smaller over Rh/C and was negligible over Pd/
C.
3.5. Comparison with other noble metal catalysts in reactivity of related
substrates
When cyclohexanol was used as substrate under H2 + Ar atmo-
sphere (Fig. 10 and Table S8), cyclohexane was produced over Pt/C,
Rh/C and Ru/C by C–O hydrogenolysis, but Pd/C rarely catalyzes hy-
drogenolysis of cyclohexanol to cyclohexane, which indicates clearly
that Pd/C had less activity for hydrogenolysis of CeO bond. The order
of the ability of C–O cleavage is Pt/C > Rh/C > Ru/C > Pd/C. Ru/C
gave some amount of C2-C5 hydrocarbons (cracking). The activity of
Ru/C in cracking was also evident in the rise of C1 balance to C1
products in the conversion of 2-methoxycyclohexanone (Table 2). Ru°
species was reported to have high catalytic activity for cracking
[43,44]. We have reported the ability of (noble metal)/C catalysts in
C–O cleavage of glycerol at 393 K: Ru/C and Rh/C showed significant
activity; however, Pt/C and Pd/C showed very low activity [45]. The
trend is due to the much high temperature in the present case (493 K).
After all, Pt/C and Rh/C have activity in dehydrogenation of
Model reactions over other noble metal catalysts were conducted in
order to know the difference among noble metals in the performance
for 2-methoxycyclohexanone conversion more clearly. In 2-methox-
ycyclohexanone conversion, dehydrogenation of cyclohexane ring and
demethoxylation are revealed to be the key reactions. From the result of
noble metals screening (Section 3.1, Table 2), Pt/C and Rh/C are good
for demethoxylation, Pd/C is only active for dehydrogenation/hydro-
genation (less active for C–O cleavage) and Ru/C has poor activity for
dehydrogenation of cyclohexane ring. These data, however, were
complex because many compounds were produced and the character of
each noble metal was not clear. Here, the catalytic ability of each noble
metal in dehydrogenation of cyclohexane ring and hydrogenolysis of
69

A. Miyagawa, et al.
MolecularCatalysis471(2019)60–70
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2019.03.019.
References
[1] G.W. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044–4098.
[2] W. Mu, H. Ben, A. Ragauskas, Y. Deng, Bioenerg. Res. 6 (2013) 1183–1204.
[3] H. Wang, J. Male, Y. Wang, ACS Catal. 3 (2013) 1047–1070.
[4] L. Yang, K. Seshan, Y. Li, Catal. Today 298 (2017) 276–297.
[5] V.N. Bui, D. Laurenti, P. Afanasiev, C. Geantet, Appl. Catal. B 101 (2011) 239–245.
[6] I.D. Mora, E. Méndez, L.J. Duarte, S.A. Giraldo, Appl. Catal. A Gen. 474 (2014)
59–68.
[7] E. Laurent, B. Delmon, Appl. Catal. A Gen. 109 (1994) 77–96.
[8] Y. Nakagawa, M. Ishikawa, M. Tamura, K. Tomishige, Green Chem. 16 (2014)
2197–2203.
[9] M. Ishikawa, M. Tamura, Y. Nakagawa, K. Tomishige, Appl. Catal. B 182 (2016)
193–203.
[10] G.-Y. Xu, J.-H. Guo, Y.-C. Qu, Y. Zhang, Y. Fu, Q.-X. Guo, Green Chem. 18 (2016)
5510–5517.
Fig. 10. Reaction of cyclohexanol under H2 + Ar atmosphere.
Reaction conditions: cyclohexanol 5 mmol, water 20 mL, M/C 0.1 g, PH2
0.1 MPa (7 mmol) + PAr 0.9 MPa. aIncluded initial H2. Detailed data are shown
are Table S8. aCarbon balance.
[11] M. Kim, J.-M. Ha, K.-Y. Lee, J. Jae, Catal. Commun. 86 (2016) 113–118.
[12] M.Á. González-Borja, D.E. Resasco, Energy Fuels 25 (2011) 4155–4162.
[13] J. Sun, A.M. Karim, H. Zhang, L. Kovarik, X.S. Li, A.J. Hensley, J.-S. McEwen,
Y. Wang, J. Catal. 306 (2013) 47–57.
[14] S. Liu, H. Wang, K.J. Smith, C.S. Kim, Energy Fuels 31 (2017) 6378–6388.
[15] X. Liu, W. Jia, G. Xu, Y. Zhang, Y. Fu, ACS Sustain. Chem. Eng. 5 (2017) 8594–8601.
[16] M. Zhou, Y. Wang, Y. Wang, G. Xiao, J. Energy Chem. 24 (2015) 425–431.
[17] J. Long, S. Shu, Q. Wu, Z. Yuan, T. Wang, Y. Xu, X. Zhang, Q. Zhang, L. Ma, Energy
Conv. Manag. 105 (2015) 570–577.
cyclohexane ring and hydrogenolysis of CeO bond. High activity of
both reactions can be connected to good yield of target products
(phenol, cyclohexanone and cyclohexanol) from 2-methoxycyclohex-
anone. Ru/C has C–O hydrogenolysis activity, but the poor activity in
dehydrogenation of cyclohexane ring inhibits the formation of target
products. Pd/C has less activity for hydrogenolysis of CeO bond.
[18] S.-K. Wu, P.-C. Lai, Y.-C. Lin, H.-P. Wan, H.-T. Lee, Y.-H. Chang, ACS Sustain. Chem.
Eng. 1 (2013) 349–358.
[19] J. Feng, Z. Yang, C.-Y. Hse, Q. Su, K. Wang, J. Jiang, J. Xu, Renew. Energy 105
(2017) 140–148.
[20] A.K. Deepa, P.L. Dhepe, ChemPlusChem 79 (2014) 1573–1583.
[21] X. Wang, S. Zhu, S. Wang, J. Wang, W. Fan, Y. Lv, Appl. Catal. A Gen. 568 (2018)
231–241.
[22] G. Xu, J. Guo, Y. Zhang, Y. Fu, J. Chen, L. Ma, Q. Guo, ChemCatChem 7 (2015)
2485–2492.
4. Conclusions
[23] C. Zhao, J. He, A.A. Lemonidou, X. Li, J.A. Lercher, J. Catal. 280 (2011) 8–16.
[24] K. Kon, S.M.A.H. Siddiki, K. Shimizu, J. Catal. 304 (2013) 63–71.
[25] W.-H. Kim, I.S. Park, J. Park, Org. Lett. 8 (2006) 2543–2545.
[26] T. Mitsudome, Y. Mikami, H. Funai, T. Mizugaki, K. Jitsukawa, K. Kaneda, Angew.
Chem. Int. Ed. 47 (2008) 138–141.
2-Methoxycyclohexanone conversion to phenol, cyclohexanone and
cyclohexanol without external H2 was achieved by using Pt/C catalyst
(0.1 g) and water solvent at 493 K with 5 mmol of 2-methox-
ycyclohexanone. The yield of target demethoxylation products reached
48% (phenol: 13%, cyclohexanone: 19% and cyclohexanol: 16%). Pt/C
can facilitate both dehydrogenation of cyclohexane ring to aromatic
ring and hydrogenolysis of CeO bond well, both of which are important
for this reaction. 2-Methoxycyclohexanol can be also used as a substrate
over Pt/C, giving 49% yield of target products. Demethoxylation of
guaiacol with a small amount of H2 (0.1 MPa at room temperature) to
target products over Pt/C also proceeds (46% yield). Produced target
products are kept and are not converted to benzene or cyclohexane via
over-hydrodeoxygenation in long time reaction under the appropriate
conditions. This suppression of over-hydrodeoxygenation is related to
deactivation of catalyst; too much amount of catalyst leads to formation
of benzene and cyclohexane. Three routes exist in this reaction (Fig. 8):
(A) dehydrogenation to guaiacol + demethoxylation (Eq. 19), (B) hy-
drolysis to 2-hydroxycyclohexanone + hydrogenation to 1,2-cyclohex-
[27] K. Shimizu, K. Kon, M. Seto, K. Shimura, H. Yamazaki, J.N. Kondo, Green Chem. 15
(2013) 418–424.
[28] T. Mitsudome, Y. Mikami, K. Ebata, T. Mizugaki, K. Jitsukawa, K. Kaneda, Chem.
Commun. (Camb.) 44 (2008) 4804–4806.
[29] J. Chang, T. Danuthai, S. Dewiyanti, C. Wang, A. Borgna, ChemCatChem 5 (2013)
3041–3049.
[30] C. Zhao, D.M. Camaioni, J.A. Lercher, J. Catal. 288 (2012) 92–103.
[31] D. Gao, C. Schweitzer, H. Tae, A. Varma, Ind. Eng. Chem. Res. 53 (2014)
18658–18667.
[32] D. Gao, Y. Xiao, A. Varma, Ind. Eng. Chem. Res. 54 (2015) 10638–10644.
[33] T. Nimmanwudipong, R.C. Runnebaum, K. Brodwater, J. Heelan, D.E. Block,
B.C. Gates, Energy Fuel 28 (2014) 1090–1096.
[34] 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.
[35] A.V. Kirilin, A.V. Tokarev, L.M. Kustov, T. Salmi, J.-P. Mikkola, D.Y. Murzin, Appl.
Catal. A Gen. 435–436 (2012) 172–180.
[36] D.L. King, L. Zhang, G. Xia, A.M. Karim, D.J. Heldebrant, X. Wang, T. Peterson,
Y. Wang, Appl. Catal. B 99 (2010) 206–213.
[37] R.R. Davda, J.W. Shabaker, G.W. Huber, R.D. Cortright, J.A. Dumesic, Appl. Catal.
B 56 (2005) 171–186.
[38] J. Zakzeski, B.M. Weckhuysen, ChemSusChem 4 (2011) 369–378.
[39] R.R. Davda, J.W. Shabaker, G.W. Huber, R.D. Cortright, J.A. Dumesic, Appl. Catal.
B 43 (2003) 13–26.
anediol + deoxygenation
(Eq.
20)
and
(C)
hydrolysis +
dehydrogenation to catechol + deoxygenation (Eq. (21)). H2 was sup-
plied by both dehydrogenation of cyclohexane ring and aqueous phase
reforming of produced methanol.
[40] R.D. Cortright, R.R. Davda, J.A. Dumesic, Nature 418 (2002) 29.
[41] S. Liu, M. Tamura, Z. Shen, Y. Zhang, Y. Nakagawa, K. Tomishige, Catal. Today 303
(2018) 106–116.
[42] R.N. Olcese, M. Bettahar, D. Petitjean, B. Malaman, F. Giovanella, A. Dufour, Appl.
Catal. B 115-116 (2012) 63–73.
Acknowledgements
[43] Y. Nakagawa, S. Oya, Y. Nakaji, D. Kanno, M. Tamura, K. Tomishige,
ChemSusChem 10 (2017) 189–198.
[44] Y. Nakaji, Y. Nakagawa, M. Tamura, K. Tomishige, Fuel Proc. Technol. 176 (2018)
249–258.
This work is supported by JSPS KAKENHI grant number 18H05247.
We thank Technical Division of School of Engineering, Tohoku
University for TEM measurement.
[45] I. Furikado, T. Miyazawa, S. Koso, A. Shimao, K. Kunimori, K. Tomishige, Green
Chem. 9 (2007) 582–588.
70