Tunable selectivity of phenol hydrogenation to cyclohexane or cyclohexanol by a solvent-driven effect over a bifunctional Pd/NaY catalyst

Article abstract of DOI:10.1039/d0cy02188a

Hydrogenation of phenol is an important strategy to produce cyclohexane or cyclohexanol as both of them are raw materials for the synthesis of nylon-6 and nylon-66. Herein, we report a novel method for the selective hydrogenation of phenol to cyclohexane or cyclohexanol over a bifunctional Pd/NaY catalyst by regulating the solvent polarity. It was found that solvent polarity has a strong influence on the hydrogenation reaction mechanism. Under the identical conditions, 100% selectivity to cyclohexane could be obtained when reacting inn-octane (nonpolar solvent), while 92.3% selectivity to cyclohexanol was achieved in EtOH (polar solvent). The polarity of the solvent not only affects the competitive adsorption capacity but also the adsorption manner of phenol over the acid sites and the Pd nanoparticles in the Pd/NaY catalyst. DFT calculations show that different solvents have an almost negligible effect on the reaction energy barriers but highly affect the hydration reaction of cyclohexanol if the trace amount of water formed could not be timely removed from the catalytic system. This solvent-driven catalysis exhibits good recyclability, showing great promise for industrial applications. These findings not only provide new insights into the hydrogenation mechanism of phenolics, but also might help to develop facile strategies for the selective conversion of other phenolics into desired products.

Full text of DOI:10.1039/d0cy02188a

Catalysis
Science &
Technology
View Article Online
View Journal | View Issue
PAPER
Tunable selectivity of phenol hydrogenation to
cyclohexane or cyclohexanol by a solvent-driven
effect over a bifunctional Pd/NaY catalyst†
Cite this: Catal. Sci. Technol., 2021,
11, 1881
a
a
a
a
a
Heng Xia, Hongzi Tan, * Hongyou Cui, * Feng Song, Yuan Zhang,
a
b
c
c
Rongrong Zhao, Zhe-Ning Chen, * Weiming Yi and Zhihe Li
Hydrogenation of phenol is an important strategy to produce cyclohexane or cyclohexanol as both of
them are raw materials for the synthesis of nylon-6 and nylon-66. Herein, we report a novel method for
the selective hydrogenation of phenol to cyclohexane or cyclohexanol over a bifunctional Pd/NaY catalyst
by regulating the solvent polarity. It was found that solvent polarity has a strong influence on the hydroge-
nation reaction mechanism. Under the identical conditions, 100% selectivity to cyclohexane could be
obtained when reacting in n-octane (nonpolar solvent), while 92.3% selectivity to cyclohexanol was
achieved in EtOH (polar solvent). The polarity of the solvent not only affects the competitive adsorption ca-
pacity but also the adsorption manner of phenol over the acid sites and the Pd nanoparticles in the Pd/NaY
catalyst. DFT calculations show that different solvents have an almost negligible effect on the reaction en-
ergy barriers but highly affect the hydration reaction of cyclohexanol if the trace amount of water formed
could not be timely removed from the catalytic system. This solvent-driven catalysis exhibits good recycla-
bility, showing great promise for industrial applications. These findings not only provide new insights into
the hydrogenation mechanism of phenolics, but also might help to develop facile strategies for the selec-
tive conversion of other phenolics into desired products.
Received 12th November 2020,
Accepted 20th December 2020
DOI: 10.1039/d0cy02188a
rsc.li/catalysis
9
,10
ings.
Besides, phenol can be obtained from biomass and
1
. Introduction
1
1
coal refineries inexpensively.
Both cyclohexane and cyclohexanol are important chemical
intermediates as they are widely used for the production of
Recently, it was reported that hydrogenation of phenol
could be efficiently catalyzed by Pd-based catalysts, including
1–3
12–17
caprolactam for nylon-6 and adipic acid for nylon-66.
On
Pd/C, Pd/MgO, Pd/SiO
2
, Pd/Al
2
O
3
and Pd/MOFs.
The sur-
the industrial scale, cyclohexane is currently produced by hy-
drogenation of benzene, while cyclohexanol is derived from
face acid properties of supports also have an important influ-
1
8,19
ence on the catalytic performance.
As reported by J. W.
4
–6
oxidation of cyclohexane. Although cyclohexane can be syn-
thesized through hydrogenation of benzene, this process is
less economical, and oxidation of cyclohexane to cyclohexanol
is far from satisfactory because of the harsh reaction condi-
Zhong et al., the acid sites on catalysts have the ability to en-
hance the chemisorption of phenol and promote the conver-
2
0
sion rate. P. Yan et al. found that the hydrogenation and de-
hydration reactions of guaiacol could be effectively catalyzed
by metal sites and acid sites in Ni/Beta catalysts, respectively,
7
,8
tions. An attractive production route for cyclohexane and
cyclohexanol is selective hydrogenation or hydro-
deoxygenation of phenol through a one-step process, because
such a catalytic system offers atom efficiency and energy sav-
2
1
affording cyclohexane as the major product. There was a lin-
ear relationship observed between deoxygenation activity and
2
1
concentration of acid sites. Moreover, a combination of
metals (Pd, Pt, and Ni) and liquid (H PO ) or solid acid cata-
3
4
2
−
lysts (Nafion, SO
4 2
/ZrO , HZSM-5 and HBEA) demonstrated
a
School of Chemistry & Chemical Engineering, Shandong University of Technology,
high selectivity to cycloalkanes in the conversion of phenolic
Zibo, Shandong 255049, P. R. China. E-mail: hztan@sdut.edu.cn, cuihy@sdut.edu.cn
1
8,22–26
b
compounds.
Over Pd-based catalysts supported on dif-
State Key Laboratory of Structural Chemistry, Fujian Institute of Research on
Structure of Matter, Chinese Academy of Sciences, Fuzhou, Fujian 350002, P. R.
China. E-mail: znchen@fjirsm.ac.cn
ferent oxides, whether hydrodeoxygenation or hydrogenation
could take place highly depends on the presence or absence
of acid sites. Although a lot of effort has been devoted to de-
c
School of Agricultural Engineering and Food Science, Shandong University of
27
Technology, Zibo, Shandong 255049, P. R. China
veloping high-performance catalysts, the catalytic mechanism
of hydrodeoxygenation or hydrogenation of phenol still needs
to be further clarified at the molecular level.
†
Electronic supplementary information (ESI) available: Materials, catalyst prepa-
ration, catalyst characterization, catalyst evaluation, and computational methods
and models. See DOI: 10.1039/d0cy02188a
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 1881–1887 | 1881

View Article Online
Paper
Catalysis Science & Technology
Besides catalysts, the properties of used solvents are likely
to affect the reaction pathway of phenol, as well as the reac-
2
8
tivity. It is well known that solvents have impact on the H
solubility in the liquid phase and thus on the reaction rate of
2
2
9
phenol. The competitive adsorption behavior of solvent and
reactant molecules over the catalyst surface is probably al-
3
0
tered by different solvents. For example, selective hydroge-
nation of nitrobenzene to aniline or cyclohexylamine could
be switched by using different solvents in the presence of a
Pd/MIL-101 catalyst, because the solvent polarity is able to
regulate the linkage between intermediates and the organic
3
1
ligand in MIL-101. To better understand the catalytic mech-
anism of phenol conversion in different solvents, it is worth-
while to clarify the interactions of solvent molecules with the
active sites on the catalyst, and the competitive adsorption
behavior with the reactant, intermediates and products.
In this work, we reported a solvent-driven effect on the re-
action pathway of phenol hydrogenation over a bifunctional
Pd/NaY catalyst. The catalytic performance over different cat-
alysts and in six solvents with different polarities was investi-
Fig. 1 (a) PXRD patterns of the Pd/NaY catalyst and NaY zeolite; (b) Pd
3d XPS spectra of the Pd/NaY catalyst; (c) normalized XANES spectra at
the Pd K-edge of Pd foil, PdO powder, and the Pd/NaY catalyst; (d)
Fourier transform spectra of the experimental EXAFS spectroscopy of
Pd foil, PdO powder and the Pd/NaY catalyst.
2
gated. The catalysts were characterized by N adsorption–de-
sorption, inductively coupled plasma (ICP) spectroscopy,
powder X-ray diffraction (PXRD), X-ray photoelectron spectro-
scopy (XPS), NH temperature programmed desorption (NH -
3 3
TPD), and X-ray absorption near-edge structure (XANES) and
extended X-ray absorption fine structure (EXAFS) spectro-
scopy. Based on the kinetic experimental data, adsorption of
phenol in different solvents and density functional theory
(Fig. 1a). For the NaY zeolite, the diffraction signals at 6.3°,
10.1°, 11.9°, 15.8°, 18.7°, 20.5°, 23.7°, 27.2°, 31.4° and 37.9°
were identified as the (111), (220), (311), (331), (511), (440),
(533), (642), (555) and (666) lattice planes, respectively.
The newly emerged diffraction signals at 40.2° and 46.7° of
3
2,33
the Pd/NaY sample can be assigned to the {111} and {200} lat-
3
4
tice planes of Pd nanoparticles, indicating the successful
deposition of Pd species on the NaY zeolite. XPS was
performed to characterize the redox species of deposited Pd.
As shown in Fig. 1b, the peaks at 340.3 eV and 335.1 eV are
(DFT) calculations, the effect of solvent polarity on the reac-
tion pathway was analyzed and discussed.
3
4
attributed to Pd 3d3/2 and Pd 3d5/2 of zerovalent Pd, reveal-
ing that the deposited Pd species in Pd/NaY was in the metal-
lic state. Moreover, the white line peak of the Pd/NaY catalyst
in the XANES spectra (Fig. 1c) is nearly identical to that of Pd
foil, and only the Pd–Pd bond signal appeared in the EXAFS
spectrum of the Pd/NaY catalyst (Fig. 1d), both of which fur-
ther unveil that the deposited Pd species in the Pd/NaY cata-
lyst was in the form of nanoparticles, in accordance with the
results of PXRD and XPS. In addition, the TEM and HRTEM
images (Fig. 2) of the Pd/NaY catalyst exhibited that the Pd
nanoparticles with the mainly exposed {111} crystal facets
were uniformly dispersed on the support NaY zeolite.
2
. Results and discussion
The textural structure information of the bifunctional Pd/NaY
catalyst and its support NaY zeolite is summarized in
Table 1. The Pd content in Pd/NaY measured by ICP was 1.98
wt%, which agrees very well with the designed value of 2.0
wt%. The N adsorption–desorption isotherms at a tempera-
2
ture of 77 K were also obtained (Fig. S2†). Both the Pd/NaY
catalyst and NaY zeolite exhibited similar type IV isotherms
with BET areas of 576.83 m
m g for the NaY zeolite. As compared with the NaY zeolite
2
−1
g
for Pd/NaY and 672.42
2
−1
3
−1
(pore size: 3.93 nm, pore volume: 0.33 cm g ), the slight de-
3
−1
crease in pore size (3.86 nm) and pore volume (0.28 cm g )
of the Pd/NaY catalyst exactly demonstrates that the depos-
ited Pd species were uniformly loaded into the cavities of the
NaY zeolite.
PXRD was employed to characterize the structure and crys-
tallinity of the prepared Pd/NaY catalyst and the support
To analyze the surface acid properties of Pd/NaY and Pd/SiO
as a counterpart, NH -TPD was conducted. As shown in Fig. 3,
Pd/NaY presented a strong NH desorption peak in the tempera-
ture range of 100–300 °C, while the Pd/SiO catalyst did not
show any NH desorption in the whole tested temperature
range. The strong NH desorption peak of the Pd/NaY catalyst
2
3
3
2
3
3
Table 1 Textural structure information on the Pd/NaY catalyst and NaY zeolite
2
−1
3
−1
Samples
Designed Pd content (%)
Measured Pd content (%)
BET area (m g
)
Pore size (nm)
Pore volume (cm g
)
Pd/NaY
NaY zeolite
2.0
—
1.98
—
576.83
672.42
3.86
3.93
0.28
0.33
1882 | Catal. Sci. Technol., 2021, 11, 1881–1887
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
ideal counterpart for comparing with Pd/NaY to judge the cata-
lytic role of acid sites in the conversion of phenol.
The catalytic activity in the conversion of phenol was eval-
uated by using Pd/NaY, Pd/SiO and NaY zeolite as catalysts,
2
respectively (Table 2). Six solvents with different polarities
were examined including water (H O), N,N-dimethyl-
2
formamide (DMF), methanol (MeOH), ethanol (EtOH),
n-butanol (n-BuOH) and n-octane, whose polarities can be es-
Fig. 2 TEM (a and b) and HRTEM (c) images of the Pd/NaY catalyst.
timated by their dielectric constant order: H
2
O (80.4) > DMF
(
38.3) > MeOH (33.6) > EtOH (24.3) > n-BuOH (17.8) >
n-octane (2.1). Under the catalysis of the Pd/NaY catalyst, dis-
tinct differences in the phenol conversion and product selec-
tivity were observed when using the solvents with different
polarities. In strong polar solvents like H O (entry 1), DMF
2
(
entry 2) and MeOH (entry 3), no conversion of phenol was
observed. It has been reported that the nonreactivity of phe-
nol in MeOH is due to the MeOH adsorption on the Pd sur-
face, as well as to the interaction between MeOH and the re-
3
6
actant molecules. The polarity of H O is much stronger
2
than that of MeOH, leading to a similar result. In addition to
the strong polarity of DMF, the alkalescent amide groups in
DMF might be also responsible for the nonreactivity of phe-
nol, as it could further strengthen the interaction between
the solvent and acid sites on the catalyst and the interaction
between the solvent and phenol, leading to the absolute
chemical inertness of phenol in DMF. In lower polarity EtOH
Fig. 3 NH
3
-TPD curves of the Pd/NaY catalyst and its counterpart Pd/
SiO catalyst.
2
can be attributed to the surface acid sites in the NaY zeolite. It
has been reported that the framework of the NaY zeolite is com-
posed of AlO (Al ) and SiO (Si ) units, in which the frame-
(entry 4) compared to H O, MeOH and DMF, 78.2% phenol
2
3+
4+
4
4
conversion was obtained, with 7.7% selectivity to cyclohexa-
+
work charge is balanced by monovalent cations (such as Na
none and 92.3% selectivity to cyclohexanol via the benzene-
ring hydrogenation reaction. In the much less polar n-BuOH
(entry 5), 62.4% selectivity to cyclohexanone and 37.6%
+
35
and H ), resulting in the large number of acid sites formed.
The absence of acid sites in the Pd/SiO catalyst makes it an
2
Table 2 Catalytic results of the conversion of phenol
Selectivity (%)
Ring-hydrogenation
Hydrodeoxygenation
Conversion
(%)
Entry
Catalyst
Reactant
Solvent
εa
b
1
Pd/NaY
Pd/NaY
Pd/NaY
Pd/NaY
Pd/NaY
Pd/NaY
NaY
Phenol
Phenol
Phenol
Phenol
Phenol
Phenol
Phenol
Phenol
Phenol
Phenol
Cyclohexanol
Cyclohexanol
Benzene
Benzene
H
DMF
2
O
80.4
38.3
33.6
24.3
17.8
2.1
24.3
2.1
24.3
2.1
0
0
0
0
0
0
7.7
62.4
0
0
0
0
69.3
0
0
0
0
0
0
92.3
37.6
0
0
0
0
0
0
0
b
2
b
3
MeOH
EtOH
n-BuOH
n-octane
EtOH
n-octane
EtOH
n-octane
EtOH
b
4
78.2
68.7
100.0
0
0
0
54.2
12.0
100.0
15.2
100
b
5
0
b
6
100.0
0
0
b
7
b
8
NaY
b
9
Pd/SiO
Pd/SiO
2
0
0
b
10
11
12
13
14
2
25.4
—
—
0
5.2
100.0
100.0
100.0
100.0
b
b
c
Pd/NaY
Pd/NaY
Pd/NaY
Pd/NaY
24.3
2.1
24.3
2.1
n-octane
EtOH
n-octane
c
0
0
a
c
b
Dielectric constant of the different solvents. Reaction conditions: 0.1 g catalyst, 1.0 g reactants, 70 mL solvent, 5.0 MPa H
Reaction conditions: 0.1 g catalyst, 1.0 g benzene, 70 mL solvent, 5.0 MPa H , 235 °C, 2.0 h.
2
, 235 °C, 4.0 h.
2
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 1881–1887 | 1883

View Article Online
Paper
Catalysis Science & Technology
selectivity to cyclohexanol were obtained. When reacting in
nonpolar n-octane (entry 6), phenol was completely trans-
formed into cyclohexane with a selectivity of 100%, which in-
dicates that the hydrodeoxygenation reaction occurred simul-
taneously. These results demonstrate that the reaction
products can be regulated by the solvent effect.
In order to shed light on the catalytic role of metallic sites
and acid sites on the bifunctional Pd/NaY catalyst, the catalytic
activities of the NaY zeolite (with acid sites but no metallic sites)
Fig. 4 The variation tendencies of product concentrations over the
Pd/NaY catalyst in (a) n-octane and (b) EtOH as a function of time. Re-
action conditions: 0.1 g catalyst, 1.0 g phenol, 70 mL solvent, 5.0 MPa
and Pd/SiO (with metallic sites but no acid sites) were evalu-
2
2
H , 235 °C.
ated either in EtOH or in n-octane. Under the catalysis of the
NaY zeolite, no phenol was transformed no matter which one of
the reaction solvents was used (entry 7 & 8), proving that with-
out metallic sites, the hydrogenation reaction could not take
place as the hydrogenation of benzene-ring is a prerequisite for
phenol conversion. Interestingly, in the solvent EtOH (entry 9),
shown in Scheme 1. The high concentrations of cyclohexanone
and cyclohexanol suggest that after the keto-tautomerization of
phenol to cyclohexa-2,4-dien-1-one, consecutive hydrogenation
of CC bonds to cyclohexanone is the first step in path A,
followed by hydrogenation of cyclohexanone to cyclohexanol
and further hydrodeoxygenation to cyclohexane. The process of
keto-tautomerization with cyclohexa-2,4-dien-1-one participating
phenol could not be hydrogenated when using Pd/SiO as the
2
catalyst, despite demonstrating an excellent catalytic perfor-
37
mance for many hydrogenation reactions. When replacing
EtOH with n-octane as the solvent (entry 10), however, 54.2%
phenol conversion was achieved. These results clearly prove that
the solvents play very important roles in the hydrogenation of
phenol. In terms of product selectivity, in the reaction medium
27
in phenol hydrogenation has been adequately demonstrated.
The concentration of cyclohexane increased slowly in the initial
time, and increased rapidly with the consumption of cyclo-
hexanol and benzene. A little amount of benzene could also be
formed with a tendency to increase gradually at the early stage
to a maximum and then declining due to the further hydrogena-
tion to cyclohexane (Fig. S3a†). Such a time course tendency in-
dicates that the reaction pathway of benzene hydrogenation to
cyclohexane (path B) is also tenable but it is not the main reac-
tion pathway. In path B, the CO bond in cyclohexa-2,4-dien-1-
one was first hydrogenated to produce cyclohexadienols,
followed by dehydration of cyclohexadienols to benzene, and
further benzene was hydrogenated to cyclohexane.
2
n-octane, Pd/SiO afforded predominantly ring-hydrogenation
products, cyclohexanone (69.3%) and cyclohexanol (25.4%), but
the hydrodeoxygenation product (cyclohexane, 5.2%) was very
low, indicating that the dehydration of cyclohexanol was
prohibited due to the absence of acid sites.
To further verify the role of acid sites in the hydro-
deoxygenation process, cyclohexanol was also used as the
substrate and subjected to hydrogenation. In the EtOH sol-
vent (entry 11), the conversion to cyclohexanol was only
1
2.0% in spite of the abundant acid sites on Pd/NaY, but it
was enhanced to 100% in the n-octane solvent (entry 12). As
we know, hydrogenation of cyclohexanol undergoes dehydra-
tion first and then hydrogenation, and cyclohexane is the
unique product. Thus, the selectivity to cyclohexane should
be 100% no matter what the conversion rate is. These results
demonstrate that not only the acid sites but also the solvent
polarity play a very crucial role in determining the reaction
mechanism. Under the same reaction conditions, the hydro-
genation of benzene in EtOH resulted in merely a cyclohex-
ane yield of 15.2% (entry 13), but a cyclohexane yield of
When reacting in polar EtOH (Fig. 4b and S3b†), however,
the major products were cyclohexanone and cyclohexanol; al-
most no cyclohexane was formed. Such a case reveals that
the hydrodeoxygenation of phenol is hard to take place in po-
lar EtOH. The hydrogenation rate of phenol to cyclohexanone
and cyclohexanol seemed lower than that when reacting in
n-octane. Besides, no benzene was detected through the en-
tire reaction course. These results provide evidence that
hydrodeoxygenation reactions were prohibited by EtOH.
Combining with the poor conversion rate of benzene in
the EtOH solvent (entry 13, Table 2), we conclude that path A
is the only reaction pathway over the Pd/NaY catalyst in
1
00% was achieved in n-octane (entry 14). The much lower re-
activities of cyclohexanol and benzene in EtOH as compared
to those in n-octane reveal that the polarity of the solvent has
a significant influence on the reaction rate. This can also be
used to explain why no cyclohexane was formed in the hydro-
genation of phenol when reacting in EtOH.
In an attempt to obtain more information on the reaction
pathway, the kinetic study of phenol conversion over the Pd/
NaY catalyst was carried out in n-octane and EtOH, respectively
(Fig. 4). When reacting in nonpolar n-octane, as shown in
Fig. 4a, cyclohexanone and cyclohexanol are the major interme-
diates as their concentrations presented a maximum with the
reaction time. There are two pathways for phenol conversion as
Scheme 1 The proposed reaction pathway for phenol conversion.
1884 | Catal. Sci. Technol., 2021, 11, 1881–1887
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
EtOH, but the dehydration reaction of cyclohexanol is
impeded.
In order to clarify which reaction pathway is mainly re-
sponsible for the formation of cyclohexane when reacting in
nonpolar n-octane, cyclohexanol dehydration (path A) or ben-
zene hydrogenation (path B), phenol conversion was exam-
ined under different H pressures using n-octane as the sol-
2
Fig. 6 The recyclability of the Pd/NaY catalyst for phenol conversion
in (a) n-octane and (b) EtOH. Reaction conditions: 0.1 g catalyst, 1.0 g
vent (Fig. 5). Under high H pressure (5.0 MPa and 3.0 MPa),
2
which allows sufficient hydrogenation, cyclohexane was the
unique product with a phenol conversion of 100%. Under 1.0
MPa H2 pressure, the selectivities to the reaction products
were 49.4% for cyclohexane, 44.0% for cyclohexanone and
phenol, 70 mL solvent, 5.0 MPa H
h in EtOH. Phenol conversion (blue) and selectivity to cyclohexane
2
, 235 °C, 20 min in n-octane and 4.0
(
(
pink), cyclohexanone (orange), cyclohexanol (green) and benzene
gray).
6
.5% for benzene. By further decreasing the H pressure to
2
0
.2 MPa, the selectivity to benzene increased to 15.6% accom-
panied with 2.6% selectivity to cyclohexane and 81.8% selec-
tivity to cyclohexanone. These results revealed that both path
A and path B are the coexistent reaction pathways when
reacting in n-octane, but path A is predominant.
Adsorption of phenol over the Pd/NaY catalyst in the two
solvents was conducted to investigate the solvent effect on
the interaction with the catalyst surface. Considering the dif-
ficulty in the measurement at high temperature, the adsorp-
tion experiments were carried out at room temperature
(Fig. 7). With the increase of phenol concentration in both
solvents, the adsorption capacity of phenol increased gradu-
The recyclability of the Pd/NaY catalyst was estimated by
successively reacting for five runs. After each run, the catalyst
was centrifuged and used directly for the next run. As shown
in Fig. 6a, the conversion of phenol and the selectivity to
cyclohexane could be preserved at about 95% and 71% after
five-run tests with n-octane as the solvent. When artificially
controlling the phenol conversion to be less than 100% in
n-octane during catalytic stability study, the dominant prod-
uct cyclohexane would be accompanied with the intermediate
products of benzene, cyclohexanol and cyclohexanone, which
further undergo hydrogenation or hydrodeoxygenation to
form cyclohexane. In EtOH (Fig. 6b), however, the conversion
of phenol was about 78% with 92% selectivity to cyclo-
hexanol. No decrease in the catalytic activity was observed.
The Pd content in the spent catalyst (1.96 wt%) is nearly
identical with that in the fresh one by ICP measurement, in-
dicating that no metal leaching occurred during the reaction.
Moreover, XPS, TEM and PXRD characterization showed that
the spent Pd/NaY catalyst was almost the same as the fresh
one (Fig. S4–S6†), suggesting that Pd/NaY is a robust catalyst
for the hydrogenation of phenol.
−
1
ally. At a phenol concentration of 0.25 g L , the adsorption
capacity in EtOH was equivalent to that in n-octane due to
the sufficient adsorption sites on Pd/NaY available. By further
increasing the phenol concentration, the adsorption capacity
of phenol in n-octane was much higher than that in EtOH,
suggesting that the polar solvent molecules are likely to com-
petitively occupy the adsorption sites (acid sites) with phenol.
As a consequence, the dehydration of phenol was greatly
suppressed.
For the conversion of phenolic compounds, the dehydra-
tion reaction is critical, which determines the selectivity to-
3
8,39
wards alcohol or alkane species.
The dehydration mecha-
nism of cyclohexanol was investigated by DFT calculations.
The possible reaction pathways for the dehydration of cyclo-
hexanol both in n-octane and in EtOH were considered
(Fig. 8), but the results show that solvents have an almost
negligible effect on the reaction energy barrier. Our calcula-
tions revealed that in the absence of zeolite, the direct dehy-
dration of cyclohexanol seems infeasible with high free
Fig. 5 The product distribution for phenol conversion under different
Fig. 7 The adsorption quantity of phenol over the Pd/NaY catalyst in
the solvents EtOH and n-octane with different phenol concentrations.
Reaction conditions: 0.2 g catalyst, 40 mL solvent, 25 °C, 20 h.
2
H pressures. Reaction conditions: 0.1 g catalyst, 1.0 g phenol, 70 mL
n-octane, 235 °C, 4.0 h.
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 1881–1887 | 1885

View Article Online
Paper
Catalysis Science & Technology
the interaction between polar solvent molecules and the Pd
surface decreases the reactivity of the reactants because of
the competitive adsorption with phenol. Besides, the compet-
itive adsorption by the polar solvent molecules makes the hy-
droxy group of phenol inaccessible to the acid sites, leading
to “nonplanar” adsorption on the Pd/NaY catalyst, which ef-
fectively intercepts the process of cyclohexanol dehydration.
More importantly, cyclohexanol, the dehydrated products of
cyclohexene and water are dissolved easily in the polar EtOH
solvent, leading to the very little formation of hydrocarbon
products. Meanwhile, in nonpolar n-octane without competi-
tive adsorption, the hydroxy group of phenol could access the
acid sites with “planar” adsorption on the Pd/NaY catalyst,
resulting in the occurrence of cyclohexanol dehydration.
Moreover, cyclohexanol and cyclohexene are soluble but wa-
ter is insoluble in nonpolar n-octane. The produced water is
thus timely expelled and turned into the vapor phase, pro-
moting the dehydration reaction of cyclohexanol.
3
. Conclusions
Fig. 8 DFT calculations of the reaction pathways for the dehydration
of cyclohexanol without or with the catalysis of the NaY zeolite in the
solvents (a) n-octane and (b) EtOH (values are relative Gibbs free ener-
In summary, we have demonstrated that the Pd/NaY catalyst,
featuring encapsulated Pd NPs in the pores of the NaY zeolite
with abundant acid sites, could be used as a bifunctional cat-
alyst for efficient hydrogenation of phenol. Under the identi-
cal conditions, 100% selectivity to cyclohexane could be
achieved when reacting in nonpolar n-octane, while 92.3% se-
lectivity to cyclohexanol was achieved in polar EtOH. The po-
larity of the solvent has a decisive influence on the hydroge-
nation reaction mechanism, which not only affects the
competitive adsorption capacity but also the adsorption man-
ner of phenol on the acid sites and the Pd nanoparticles on
the Pd/NaY catalyst. DFT calculations further show that dif-
ferent solvents have an almost negligible effect on the reac-
tion energy barriers but highly affect the hydration reaction
of cyclohexanol if the trace amount of water formed could
not be timely removed from the catalytic system. This
solvent-driven effect provides a facile strategy for the selective
conversion of other phenolics into desired alcohol or alkane
products.
−
1
gies ΔG in kcal mol ).
Fig. 9 The proposed catalytic mechanism of the solvent-driven selec-
tivity for the phenol conversion over the bifunctional Pd/NaY catalyst.
−
1
energy barriers of 70.2 kcal·mol (TS1) in n-octane and 69.6
−
1
Conflicts of interest
kcal mol (TS1′) in EtOH. Brønsted acids are essential to cat-
4
0
alyze the dehydration of cyclohexanol, and acidic zeolites
are effective catalysts for dehydration reactions. As shown in
Fig. 8, with the aid of the zeolite, the dehydration of cyclo-
hexanol becomes kinetically favorable under the catalysis of
the Brønsted acid sites of the zeolite, in which the free energy
There are no conflicts to declare.
Acknowledgements
The project was supported by the National Key R&D Program
of China (2019YFD1100602) and the National Natural Science
Foundation of China (21978158 and 21973094).
−
1
barriers are reduced to 26.7 kcal mol (TS2) in n-octane and
−
1
2
2.3 kcal mol (TS2′) in EtOH. The DFT calculation results,
however, could not rationalize the observed different selectiv-
ities in different solvents as the reaction mechanisms in
n-octane and EtOH are similar.
Notes and references
It is worth noting that the dehydration of cyclohexanol to
cyclohexene is thermodynamically unfavourable, leading to a
very limited yield for hydrocarbons when the reactants and
products are in a homogeneous phase. As shown in Fig. 9,
1 Y. Wang, J. Yao, H. Li, D. Su and M. Antonietti, Chem. Soc.
Rev., 2011, 133, 2362–2365.
2 L. Zhou, J. Xu, H. Miao, F. Wang and X. Li, Appl. Catal., A,
2005, 292, 223–228.
1886 | Catal. Sci. Technol., 2021, 11, 1881–1887
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
3
4
5
6
7
8
S. Hu, M. Xue, H. Chen and J. Shen, Chem. Eng. J.,
010, 162, 371–379.
A. K. Suresh, M. M. Sharma and T. Sridhar, Ind. Eng. Chem.
Res., 2000, 39, 3958–3997.
22 G. Li, J. Han, H. Wang, X. Zhu and Q. Ge, ACS Catal.,
2015, 5, 2009–2016.
23 M. V. Bykova, D. Y. Ermakov, S. A. Khromova, A. A. Smirnov,
M. Y. Lebedev and V. A. Yakovlev, Catal. Today, 2014, 220-
222, 21–31.
24 H. Yun, Y. Lee, S. Son, M. Chung, M. Jang, E. J. Shin, S.
Jung, W. Kwak, W. J. Jeong and H. G. Ahn, J. Nanosci.
Nanotechnol., 2017, 17, 2776–2779.
25 Z. Luo, Z. Zheng, Y. Wang, G. Sun, H. Jiang and C. Zhao,
Green Chem., 2016, 18, 5845–5858.
2
S.-C. Hu and Y.-W. Chen, Ind. Eng. Chem. Res., 1997, 36,
5
153–5159.
T. Takahashi, T. Kai, H. Kimura and A. Inoue, Mater. Trans.,
001, 42, 1599–1602.
2
C. M. Cirtiu, A. F. Dunlop-Brière and A. Moores, Green
Chem., 2011, 13, 288–291.
S. E. Lyubimov, M. V. Sokolovskaya, A. A. Korlyukov, O. P.
Parenago and V. A. Davankov, J. Iran. Chem. Soc., 2020, 17,
26 L. Wang, P. Ye, F. Yuan, S. Li and Z. Ye, Int. J. Hydrogen
Energy, 2015, 40, 14790–14797.
1
283–1287.
A. K. Talukdar and K. G. Bhattacharyya, Appl. Catal., A,
993, 96, 229–239.
27 P. M. de Souza, R. C. Rabelo-Neto, L. E. P. Borges, G. Jacobs,
B. H. Davis, T. Sooknoi, D. E. Resasco and F. B. Noronha,
ACS Catal., 2015, 5, 1318–1329.
9
1
1
1
1
1
1
1
1
1
1
1
2
2
0 C. Chen, P. Liu, M. Zhou, B. K. Sharma and J. Jiang,
Energies, 2020, 13, 846–858.
1 M. Tymchyshyn and C.-B. Xu, Bioresour. Technol., 2010, 101,
28 F. Li, B. Cao, W. Zhu, H. Song, K. Wang and C. Li, Catalysts,
2017, 7, 145–155.
29 I. Urukova, J. Vorholz and G. Maurer, J. Phys. Chem. B,
2005, 109, 12154–12159.
30 J. He, C. Zhao and J. A. Lercher, J. Catal., 2014, 309,
362–375.
31 X. Chen, K. Shen, D. Ding, J. Chen, T. Fan, R. Wu and Y. Li,
ACS Catal., 2018, 8, 10641–10648.
32 W. Fu, L. Zhang, T. Tang, Q. Ke, S. Wang, J. Hu, G. Fang, J.
Li and F. S. Xiao, J. Am. Chem. Soc., 2011, 133, 15346–15349.
33 M. Aimen Isa, M. Hanif Halim, T. L. Chew and Y. F. Yeong,
IOP Conf. Ser.: Mater. Sci. Eng., 2020, 736, 052019.
34 R. Wojcieszak, M. J. Genet, P. Eloy, P. Ruiz and E. M.
Gaigneaux, J. Phys. Chem. C, 2010, 114, 16677–16684.
35 A. Maghfirah, M. M. Ilmi, A. T. N. Fajar and G. T. M. Kadja,
Mater. Today Chem., 2020, 17, 2468–5194.
36 P. C. D. Mendes, R. Costa-Amaral, J. F. Gomes and J. L. F. Da
Silva, Phys. Chem. Chem. Phys., 2019, 21, 8434–8444.
37 X. Li, L. Cheng and X. Wang, Res. Chem. Intermed., 2018, 45,
1249–1262.
2
483–2490.
2 K. V. R. Chary, D. Naresh, V. Vishwanathan, M. Sadakane
and W. Ueda, Catal. Commun., 2007, 8, 471–477.
3 P. Claus, H. Berndt, C. Mohr, J. Radnik, E.-J. Shin and M. A.
Keane, J. Catal., 2000, 192, 88–97.
4 I. Witońska, A. Królak and S. Karski, J. Mol. Catal. A: Chem.,
2
010, 331, 21–28.
5 D. Zhang, Y. Guan, E. J. M. Hensen, L. Chen and Y. Wang,
Catal. Commun., 2013, 41, 47–51.
6 Y.-Z. Xiang, L.-N. Kong, C.-S. Lu, L. Ma and X.-N. Li, React.
Kinet., Mech. Catal., 2010, 100, 227–235.
7 Y. Xu, Z. Peng, Y. Yu, D. Wang, J. Liu, Q. Zhang and C.
Wang, New J. Chem., 2020, 44, 5088–5096.
8 C. Zhao, D. M. Camaioni and J. A. Lercher, J. Catal.,
2
012, 288, 92–103.
9 S. Velu, M. P. Kapoor, S. Inagaki and K. Suzuki, Appl. Catal.,
A, 2003, 245, 317–331.
0 J. Zhong, J. Chen and L. Chen, Catal. Sci. Technol., 2014, 4,
38 L. Nie, P. M. de Souza, F. B. Noronha, W. An, T. Sooknoi and
D. E. Resasco, J. Mol. Catal. A: Chem., 2014, 388–389, 47–55.
39 X. Zhu, L. Nie, L. L. Lobban, R. G. Mallinson and D. E.
Resasco, Energy Fuels, 2014, 28, 4104–4111.
3
555–3569.
1 P. Yan, M. M.-J. Li, E. Kennedy, A. Adesina, G. Zhao, A.
Setiawan and M. Stockenhuber, Catal. Sci. Technol., 2020, 10,
8
10–825.
40 A. Clearfield and D. S. Thakur, J. Catal., 1980, 65, 185–194.
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 1881–1887 | 1887