Selective phenol hydrogenation under mild condition over Pd catalysts supported on Al2O3 and SiO2

Article abstract of DOI:10.1007/s11164-018-3687-3

Cyclohexanone (CHONE) is the key intermediate in the manufacture of nylon-6 and nylon-66. Selective hydrogenation of phenol into CHONE was investigated over Pd/SiO₂ and Pd/Al₂O₃. The results show that the yield of CHONE reaches 98% or more over Pd/Al₂O₃ and Pd/SiO₂ at 333?K under atmospheric pressure in cyclohexane solvent. High activity of Pd/Al₂O₃ is promoted by Lewis acidity, and phenol can be converted 100% within 300?min. The hydrogenation of CHONE occurs until the conversion of phenol approaches completion. Pd/SiO₂ with smaller Pd nano-particles presents higher selectivity. For polar solvent, such as ethanol and dichloromethane, the activity of Pd catalysts decreases greatly. Auxiliary experiments verify that phenol adsorbs on Pd catalysts via the formation of π–c with an aromatic ring. Increased hydrogen pressure not only promotes significantly the rates of hydrogenation, but also increases the selectivity for CHONE, especially over Pd/SiO₂-1 catalyst.

Full text of DOI:10.1007/s11164-018-3687-3

Research on Chemical Intermediates
https://doi.org/10.1007/s11164-018-3687-3
Selective phenol hydrogenation under mild condition
over Pd catalysts supported on Al O and SiO
2
3
2
1
1
1
Xinzheng Li · Ling Cheng · Xingyi Wang
Received: 20 October 2018 / Accepted: 20 November 2018
©
Springer Nature B.V. 2018
Abstract
Cyclohexanone (CHONE) is the key intermediate in the manufacture of nylon-6 and
nylon-66. Selective hydrogenation of phenol into CHONE was investigated over Pd/
SiO and Pd/Al O . The results show that the yield of CHONE reaches 98% or more
2
2
3
over Pd/Al O and Pd/SiO at 333 K under atmospheric pressure in cyclohexane
2
3
2
solvent. High activity of Pd/Al O is promoted by Lewis acidity, and phenol can be
2
3
converted 100% within 300 min. The hydrogenation of CHONE occurs until the con-
version of phenol approaches completion. Pd/SiO with smaller Pd nano-particles
2
presents higher selectivity. For polar solvent, such as ethanol and dichloromethane,
the activity of Pd catalysts decreases greatly. Auxiliary experiments verify that phe-
nol adsorbs on Pd catalysts via the formation of π–c with an aromatic ring. Increased
hydrogen pressure not only promotes signiꢀcantly the rates of hydrogenation, but
also increases the selectivity for CHONE, especially over Pd/SiO -1 catalyst.
2
Keywords Catalytic hydrogenation · Solvent eꢁect · Pd · Phenol · Cyclohexanone
Introduction
Cyclohexanone (CHONE) as an intermediate is applied widely for the manufac-
ture of nylon-66 and nylon-6 [1]. Generally, CHONE is industrially synthesized
through the hydrogenation of phenol and oxidation of cyclohexane [2, 3]. The oxida-
tion pathway required high temperature and thus produced a considerable amount
of cyclohexanol (CHANOL) and organic acids that would make puriꢀcation diꢂ-
cult. With the hydrogenation pathway, phenol was hydrogenated to CHANOL, and
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s1116
4
-018-3687-3) contains supplementary material, which is available to authorized users.
*
Xingyi Wang
wangxy@ecust.edu.cn
1
Laboratory for Advanced Materials, Research Institute of Industrial Catalysis, East China
University of Science and Technology, Shanghai 200237, China
Vol.:(012
1
3456789)
3

X. Li et al.
then CHANOL was dehydrogenated at high temperature to CHONE. On the other
hand, the one-step selective hydrogenation of phenol to CHONE is advantageous
from an eꢂciency standpoint, and the reaction can be conducted in either the gas
phase or liquid phase [4–6]. Generally, the hydrogenation of phenol is carried out
in the vapor phase over supported Pd catalysts [4, 6–16]. Vapor-phase hydrogena-
tion could be carried out at high temperatures, during which Pd catalysts are usually
deactivated due to coking. For liquid-phase hydrogenation over Pd catalysts, high
temperature and long reaction time were necessary to obtain high phenol conver-
sions [17–20]. However, one-step selective hydrogenation to produce CHONE is
still a preferred industrial pathway in terms of eꢂciency [1, 21], and it has been
desirable for developing new technique. Of recent studies about one-step selective
hydrogenation, many researchers have contributed to this area, and multiple cata-
lysts have been screened for this purpose, such as Pd/C, Pd/Al O , Pd/NaY zeo-
2
3
lite [22], Raney Ni [23], Pd-based ꢀlm/chamber [24, 25], amorphous alloy [20] and
bimetal catalysts [4]. However, the attainment of high selectivity (>95%) at elevated
conversion (>80%) with a satisfactory rate is a great challenge [12, 26], because the
CHONE can be further hydrogenated to CHANOL under the reaction conditions [6,
2
5]. Some reports focused on the supports such as mesoporous hollow silica sphere
(
MHSS) [27, 28], TiO and TiO -carbon materials [29]. The Pd–Au bimetal catalyst
2 2
supported on MHSS [30] showed a good activity for phenol hydrogenation at 50 °C
with conversion of 97.5% and the selectivity for CHONE of 96% within 45 min.
Dual supported Pd–Lewis acid catalyst shows the highest selectivity (>99.9%).
Additionally, the eꢁect of reaction conditions, such as hydrogen pressure and
temperature, has been also investigated. Although various solvents, such as CH Cl
2
2
[
22], scCO [31] and CH CH OH [30], were investigated in phenol hydrogenation,
2 3 2
the eꢁect of solvents has not been studied in detail. However, up to now, the pro-
duction eꢂciency of CHONE through the hydrogenation of phenol over Pd catalysts
is still low, because either of long reaction time or of high hydrogen pressure. In this
work, we prepared Pd/Al O and Pd/SiO catalysts with impregnation followed by
2
3
2
reduction with NaBH and tested their behaviors during one-step phenol hydrogena-
4
tion to CHONE in terms of the size of Pd nano-particles, adsorption conꢀguration,
solvents and hydrogen pressure.
Experimental
Catalyst preparation
γ-Al O (99.9%, Ling Feng, Lt ed) materials and SiO (99.9%, Ling Feng Lt ed)
2
3
2
were impregnated with a given concentration of H PdCl (59.5% Pd, Heraeus Lt ed)
2
4
aqueous solution, and placed in air for 12 h at room temperature after an ultrasonic
treatment for 10 min and then dried at 353 K in air for 6 h. The dried solids were
reduced by NaBH (99.9%, Guo Yao Lt ed) solution (pH=12, nNaBH :nPd=10:1)
4
4
at room temperature followed by an ultrasonic treatment for 30 min. Finally, the mix-
tures were ꢀltered and dried at 333 K for 12 h. Based on X-ray ꢃuorescence analy-
ses, the Pd loading of as-synthesized Pd catalysts is listed in Table 1. According to
1
3

Selective phenol hydrogenation under mild condition over…
Table 1 The physical properties of Pd catalysts
Catalysta
Pd loading wt%
b
Surface
D (nm) Pore vol-
d
d
e
(nm)
NH uptake
Pd
3
c
2
−1
area (m /g)
ume (ml/g)
(μ mol g
)
Pd/Al O -1
0.59
0.25
0.53
0.30
192
195
292
296
6.586
6.585
6.593
6.592
0.31
0.34
0.36
0.39
13
5
10
294
290
ND
2
3
Pd/Al O -0.5
2
3
f
Pd/SiO -1
2
Pd/SiO -0.5
5
ND
2
a
“
-1” and “-0.5”, the normal percentage content of Pd
b
c
d
e
f
Actual content determined by ICP
Area based on BET model
Pore size, calculated by BJH method from the desorption branch
Mean particle size of catalysts, determined by TEM
Not detected
normal content of Pd, 1 wt% and 0.5 wt% loadings of Pd on Al O are labeled as Pd/
2
3
Al O -1 and Pd/Al O -0.5, respectively, and 1 wt% and 0.5 wt% loadings of Pd on
2
3
2
3
SiO are labeled as Pd/SiO -1 and Pd/SiO -0.5, respectively.
2
2
2
Catalyst characterization
The characterization of Pd catalysts by N adsorption–desorption isotherms, X-ray
2
diꢁraction (XRD), transmission electron microscopy (TEM), X-ray photoelectron
spectroscopy (XPS), NH temperature-programmed desorption (NH -TPD) and
3
3
inductively coupled plasma-atomic emission spectrometry technique (ICP-AES)
is similar to the methods reported [32]. Pd loading on supports was determined by
X-ray ꢃuorescence (XRF) using an XRF-1800 instrument. Compositional analy-
sis for the used catalyst in diꢁerent proportions of dichloromethane (DCM) and
cyclohexane was performed with an energy-dispersive X-ray spectroscopy (EDAX)
Si(Li) energy dispersive spectroscopy (EDS) detector.
Activity test
In a typical reaction procedure, 1 wt% Pd catalyst (4.0 g) or 0.5 wt% Pd catalyst
(
8.0 g) was added to a stirred solution composed of 1 g of phenol and 50 ml of
solvent in a modiꢀed 100-ml three-necked ꢃask with a jacket connected to a cool-
ing recirculator (Scientz DL-3020) at 333 K (2 mol% Pd relative to phenol). The
ꢃask containing the reaction mixture was purged six times with hydrogen in order
to remove air completely before being stirred vigorously under ordinary hydro-
gen atmosphere. A typical reaction was carried out in an autoclave batch reactor
(
DaLian TongChan, 100 ml). After loading the reactant, catalyst and solvent, the
reactor was ꢃushed six times with H . At 333 K and a stirring speed of 750 rpm, H
2
2
was charged from 3 to 11 bar. The products and reactants during the hydrogenation
of phenol were analyzed by gas chromatography (GC, GC2060 with a PEG-20M
1
3

X. Li et al.
column) equipped with a ꢃame ionization detector (FID). For the hydrogenation of
t-butyl-phenol, the products and reactants were analyzed by GC-mass spectrometry
(
GC-MS).
Kinetic study of phenol hydrogenation
Time of ꢃights (TOFs) in the diꢁerent reaction systems were calculated based on
the reaction network of phenol hydrogenation over Pd catalysts [22], as follows: (1)
phenol hydrogenation to CHONE and (2) CHONE hydrogenation to CHANOL. For
reaction measurements, quantities of reactants and catalysts were as follows: (a)
phenol (4.0 g), 1 wt% Pd catalyst (0.5 g) in 50 ml of cyclohexane; (b) phenol (2.0 g)
and 1 wt% Pd catalyst (0.5 g) in 50 ml of ethanol; (c) phenol (1.0 g) and 1 wt% Pd
catalyst (0.5 g) in 50 ml of DCM; (d) phenol (4.0 g) and 0.5 wt% Pd catalyst (1.0 g)
in 50 ml of cyclohexane; (e) phenol (2.0 g) and 0.5 wt% Pd catalyst (1.0 g) in 50 ml
of ethanol; (f) phenol (1.0 g) and 0.5 wt% Pd catalyst (1.0 g) in 50 ml of DCM; (g)
CHONE (0.5 g) and 1 wt% Pd catalyst (0.5 g) in 50 ml of solvent. After reaction, the
workup was as described in this section.
Result and discussion
Characterization of Pd catalysts
XRD analyses (not shown) of Pd/Al O show that there appear to be diꢁraction
2
3
peaks of γ-Al O , while the face-centered cubic Pd cannot be identiꢀed, due to its
2
3
diꢁraction peaks overlapping with ones from γ-Al O . BET surface area of Pd/Al O
2
3
2
3
2
is calculated to be 192 m /g (Table 1) from N adsorption isotherms. The average
2
size of Pd particles is estimated to be 5 nm and 13 nm by TEM (Fig. 1) for 0.5% and
1
% loading, respectively, indicating the dispersion of Pd on Al O can be promoted
2 3
by the decrease in Pd loading. NH -TPD tests show that Pd/Al O catalysts with
3
2
3
−
1
diꢁerent Pd loading possess the acid amount of ca. 290 μ mol g (Table 1), which,
as known, are ascribed to the Lewis acidity of Al O . Pd/SiO catalysts possess
2
3
2
2
2
70 m /g of surface area, and 5-nm and 10-nm average sizes of metallic Pd particles
with face-centered cubic Pd lattices for 0.5% and 1% loading, respectively (Table 1).
No acidity is detected on Pd/SiO catalysts. XPS results [see Supporting Information
2
(
SI)] show that the binding energies of Pd 3d are in the range of 335.0–335.6 eV
5
/2
for all Pd catalysts under study, conꢀrming that the Pd species exist in a metallic
2
+
state, indicating that all almost Pd precursor was reduced by NaBH .
4
The hydrogenation of phenol
Figure 2 shows the concentration of phenol, CHONE and CHANOL as a func-
tion of time during phenol hydrogenation over Pd/Al O and Pd/SiO catalysts
2
3
2
at 333 K under atmospheric H pressure. The result from Fig. 2a shows that the
2
activity of Pd/Al O is dependent on Pd loading to a lesser extent. The increase of
2
3
1
3

Selective phenol hydrogenation under mild condition over…
6
0
60
40
20
0
d=13.0 nm
d=12.0 nm
40
20
0
4
8
12
16
20
4
8
12
16
20
d (nm)
d (nm)
6
0
60
40
20
0
d=5.0 nm
d=5.0 nm
40
20
0
0
2
4
6
8
10
12
0
2
4
6
8
10
12
d (nm)
d (nm)
Fig. 1 TEM of Pd/Al O -1 (a), Pd/SiO -1 (b), Pd/Al O -0.5 (c) and Pd/SiO -0.5 (d)
2
3
2
2
3
2
Pd loading from 0.5 to 1% increases the time needed for 80% conversion from 88
to 123 min. However, at 99% conversion of phenol, this time diꢁerence decreases.
The formation of CHONE is observed once phenol converts, while CHANOL
is produced signiꢀcantly until the conversion reaches higher value, 93% for Pd/
Al O -1 and 99% for Pd/Al O -0.5, indicating that phenol molecules probably
2
3
2
3
compete with CHONE for adsorbing sites with a favorable situation. Moreover,
the CHANOL production rate is quicker over Pd/Al O -0.5, due probably to a
2
3
larger interface between Pd and Al O . As known, CHONE molecules can adsorb
2
3
Lewis acid sites with carbonyl groups. Therefore, CHONE can be activated on
Al O and then attacked by active H species on Pd. As a result, there is a maxi-
2
3
mum on the curve of CHONE concentration. For Pd/SiO catalysts, the rates for
2
phenol conversion is lower than Pd/Al O catalysts, and it needs 200 and 283 min
2
3
for 80% conversion over the catalysts with 0.5% and 1% Pd loading, respectively
(
Fig. 2b). And phenol is completely converted almost after 500 min. However, it
is worth noting that CHONE is hydrogenated into CHANOL at a very low speed
and the selectivity of CHONE can be maintained at a high level with the increase
in reaction time (98 and 95% after 100% conversion of phenol over Pd/SiO -1
2
and Pd/SiO -0.5 catalysts, respectively). As mentioned above, Lewis acid sites
2
1
3

X. Li et al.
Fig. 2 The phenol hydrogena-
tion over Pd/Al O (a) and Pd/
100
Pd/Al O -0.5
2
3
2
3
Phenol
CHONE
CHANOL
SiO (b) catalysts with various
2
8
6
4
2
0
0
0
0
0
Pd loadings at 333 K under
atmospheric H pressure
2
Pd/Al O -1
2
3
Phenol
CHONE
CHANOL
0
50
100
150
200
250
300
350
400
Time / min
(
A)
100
8
6
4
2
0
0
0
0
0
Pd/SiO -0.5
2
Phenol
CHONE
CHANOL
P
d/SiO -1
2
Phenol
CHONE
CHANOL
0
100
200
300
Time / min
400
500
(
B)
are important for activation of CHONE. Because there are almost no acid sites on
Pd/SiO catalysts (Table 1), it can be expected that the activity of Pd/SiO cata-
2
2
lysts for the hydrogenation of CHONE to CHANOL is much lower than Pd/Al O
2
3
catalysts.
For kinetic consideration, the rates were estimated based on the data obtained
under the condition described in the experimental section. The results (Fig. 2a)
shows that over Pd/Al O -1 catalyst, the dependence of reaction rate on the con-
2
3
centration of phenol is of ꢀrst order within the experiment, while the linear rela-
tion of conversion with the concentration of phenol over Pd/Al O -0.5 catalyst is
2
3
observed. This phenomenon indicates that more active H species can be produced
on Pd/Al O -1 catalyst so that the rate is dependent on the adsorption of phenol.
2
3
For Pd/SiO catalysts, the order of reaction on the concentration of phenol is basi-
2
cally ꢀrst. Over Pd/Al O -1 and Pd/SiO -1 catalysts, the initial rates of phenol
2
3
2
−
1
hydrogenation are 5.52 and 2.59 mmol h , respectively, and TOFs can be estimated
based on the mole number of translated phenol per mole of Pd atoms to be 24.8
1
3

Selective phenol hydrogenation under mild condition over…
a
mol h−1) of hydrogenation reaction over Pd catalysts
−1
Table 2 TOF data (mol
substrate
b
Pd
Solvent
Pd/Al O
Pd/SiOc
2
3
2
Tphenol-0.5
Tphenol-1
TCHONE-1
Tphenol-0.5
Tphenol-1
TCHONE-1
Chlorhexidine
Ethanol
DCM
76
8
0.003
24.8
1.8
0.004
6.8
0.1
0.002
12.9
1.4
0.004
13.2
1.4
0.006
0.2
0.017
0.005
a
The hydrogenation at 333 K and 1 bar PH2
b, c
“
-1” and “-0.5”, the normal percentage content of Pd, respectively
−
1
−1
and 13.2 mol mol
h
(Table 2), respectively. Taking the solubility of H hav-
2
Pd
ing no signiꢀcant change within the range of 283–333 K, the temperature depend-
ence of the initial reaction rate can be used to generate apparent activation ener-
gies. The associated Ea value (with 95% conꢀdence limits, see SI) for phenol
hydrogenation over Pd/Al O -1 and Pd/SiO -1 catalysts are estimated to be 27.0 and
2
3
2
−1
3
4.5 kJ/mol , respectively. Obviously, Pd/Al O -1 with Lewis acidity is favorable
2
3
for phenol hydrogenation [22] and thus more active than Pd/SiO -1. The rates of
2
CHONE hydrogenation were measured under the same conditions as mentioned
−
1
above. Pd/Al O -1 catalyst presents 1.5 mmol h , higher than Pd/SiO -1 catalyst
2
3
2
−
1
−1 −1
(
0.034 mmol h ). The TOF over Pd/Al O -1 is 6.78 mol mol h , much higher
2 3 Pd
−
1
−1
than that over Pd/SiO -1 (0.17 mol mol h ; Table 2). These data verify that low
2
2
Pd
selectivity over Pd/Al O catalysts with Lewis acidity for CHONE can be ascribed
3
to high activity for CHONE hydrogenation.
Eꢀect of hydrogen gas pressure
The eꢁect of hydrogen pressure (P ) on the hydrogenation over Pd catalysts was
H2
investigated in the range of 3–11 bar and the results are shown in Fig. 3a–d. It
can be seen that the increase in P results in the increase in the rate of each
H2
step during hydrogenation process of phenol → CHONE → CHANOL to diꢁerent
extents, depending on Pd loadings and supports. For Pd/Al O -1 catalyst, it needs
2
3
3
00, 70, 45 and 25 min to complete hydrogenation of phenol at 3, 6, 9 or 11 bar,
respectively (Fig. 3b). The relation of corresponding TOF (calculated by the ini-
tial rates) with P is shown in Fig. 4. TOF is directly proportional to P within
H2
H2
−1 −1
−
1
−1
1
9
–9 bar. TOF increases from 25 mol mol
h
at 1 bar to 249 mol mol
h
Pd
at
Pd
−
1
−1
bar. However, TOF drops to 171 mol mol h at 11 bar, which is probably
Pd
due to excessively occupying active sites by hydrogen. For Pd/Al O -0.5 cata-
2
3
lyst, the increase in P almost does not promote phenol hydrogenation (Fig. 3a),
H2
while CHONE → CHANOL is speeded greatly. This phenomenon indicates that
CHONE can be activated on Pd/Al O -0.5 catalyst with a large interface between
2
3
Pd and Al O to react with active H. The increases in the amount of active H spe-
2
3
cies through increasing P will promotes more eꢁectively the hydrogenation of
H2
CHONE. For Pd/SiO catalysts, the activity diꢁerence obtained by changing P
2
H2
is large, from 1 to 3 bar (the time need for 100% conversion of phenol decreases
1
3

X. Li et al.
1
00
100
80
60
40
20
0
8
6
4
2
0
0
0
0
0
Pd/Al O -1
2
3
Pd/Al O -0.5
2
9
3
3
6
9
11 bar
6
bar
Phenol
Phenol
CHONE
CHANOL
CHONE
CHANOL
0
5
10 15 20 25 30 35 40 45
0
10
20
30
40
50
60
70
80
Time / min
Time / min
(A)
(B)
1
00
100
8
6
4
2
0
0
0
0
0
80
60
40
20
0
Pd/SiO-0.5
2
9
Pd/SiO -1
2
3
6
bar
3
6
9
bar
Phenol
CHONE
CHANOL
Phenol
CHONE
CHANOL
0
10 20 30 40 50 60 70 80 90 100 110
Time / min
0
10 20 30 40 50 60 70 80 90 100 110
Time / min
(
D)
(C)
Fig. 3 Phenol hydrogenation over Pd/Al O -0.5 (a), Pd/Al O -1 (b), Pd/SiO -0.5 (c) and Pd/SiO -1 (d)
2
3
2
3
2
2
catalysts at 333 K under diꢁerent H pressure
2
Fig. 4 TOFs of phenol hydro-
genation under diꢁerent pres-
sure over Pd catalysts
Pd/Al O -1
2
2
1
1
40
00
60
20
2
3
Pd/SiO -1
2
8
4
0
0
0
0
2
4
6
8
10
12
H Pressure / bar
2
from 500 to 105 min), while it is smaller from 3 to 9 bar (from 105 to 90 min
−
1
−1
for 100% conversion; Fig. 3c, d). TOF increases from 13 mol mol
h
at 1 bar
Pd
to 76 mol mol−
1
h
−1
at 3 bar P (Fig. 4). This phenomenon is explained by
P
H2
Pd
H2
1
3

Selective phenol hydrogenation under mild condition over…
the change in controlling step. At 1 bar, the amount of active H species is low,
and the rate over Pd/SiO catalysts is dependent on the concentrations of phenol
2
and active H species. When P increases up to 3 bar or higher, the amount of
H2
active H species increases rapidly to a signiꢀcantly high value which is enough
to meet the reaction. Then, the absorption of phenol probably becomes the rate-
controlling step so that the conversion of phenol almost does not increase with
the further increase in P . Moreover, the diꢁerence in the activity between Pd/
H2
Al O and Pd/SiO catalysts becomes small with increase in P . The formation
2
3
2
H2
of active H species on the surface of Pd catalyst is promoted by raising P , lead-
H2
ing to the decrease in electronic density around Pd. Thus, the adsorption of phe-
nol on the Pd atom of Pd/SiO catalysts become strong, which is favorable for
2
speeding the rate of the controlling step.
Figure 5 shows the highest yields of CHONE during phenol hydrogenation as a
function of P . The rates of CHONE hydrogenation were measured at 1 bar H and
H2
2
−
1
−1
3
33 K. The TOF over Pd/Al O -1 is 6.78 mol mol h , much higher than that over
2
h
3
Pd
−
1
−1
Pd/SiO -1, 0.17 mol mol
(Table 2), and so the highest yield of CHONE over
2
2
Pd
Pd/SiO -1 (97.0%) is higher that of Pd/Al O -1 (91.5%). As seen in Fig. 3 with the
2
3
increase in P , the hydrogenation of CHONE over Pd/Al O catalysts occurs sig-
H2
2
3
niꢀcantly when approaching 100% conversion of phenol, while Pd/SiO -1 and Pd/
2
SiO -0.5 catalysts present a signiꢀcant yield of CHANOL at ca. 95% and 98% con-
2
version, respectively. Therefore, the high yield of CHONE is dependent of H active
species and CHONE activation in the reaction system. The Pd–Au bimetal catalyst
supported on MHSS [30] showed a good activity for phenol hydrogenation and high
selectivity of CHONE. The conversion of phenol was 97.5% and the selectivity of
CHONE was higher than 96% within 45 min at 50 °C. Han revealed the Lewis acids
as a promoting agent can improve the selectivity for CHONE through strong adsorp-
tion of CHONE on Lewis acid sites [22]. Under optimizing conditions in this work,
the highest yields of CHONE over Pd/SiO and Pd/Al O are 97% and 98%, respec-
2
2
3
tively. It is noted that the hydrogenation of CHONE over Pd/SiO -0.5 is not pro-
2
moted obviously by the increase in P and, so, high selectivity for CHONE can be
H2
maintained with the increase in reaction time.
Fig. 5 The highest cyclohex-
anone yield as a function of
pressure during the hydrogena-
tion of phenol
100
9
9
8
8
8
6
2
8
4
0
Pd/Al O -1
2
3
Pd/Pd/SiO -1
2
Pd/Al O -0.5
2
3
Pd/Pd/SiO-0.5
2
0
2
4
6
8
10
12
H Pressure / bar
2
1
3

X. Li et al.
Eꢀect of solvent
In order to investigate the eꢁect of solvent properties on phenol hydrogenation over
Pd catalysts, the solvents selected in this work must possess similar hydrogen gas
solubility so that it can guaranteed that these experiments were not aꢁected by the
diꢁerence in H mass transfer. Additionally, a group of solvents with a large diꢁer-
ence in solvent properties are selected in terms of the values of dielectric constant
(
ε), dipole moment (μ), hydrogen donor capability (α) and solvent polar empiri-
cal parameters (ET (30)). The selected solvents with similar hydrogen gas solubil-
ity as in the following three categories: non-protic non-polar solvent: cyclohexane
[
ε=2.02, μ=0.00, α=0, ET (30)=30.9]; protic solvents: ethanol [ε=24.6, μ=1.69,
α=0.86, ET (30)=51.9]; and non-protic polar solvents: DCM [ε=9.08, μ=1.14,
α=0.13, ET (30)=40.7] must be considered.
The eꢁect of solvent on the activity for phenol hydrogenation and the selectivity
for CHONE was investigated at 333 K and 1 bar P and the results are shown in
H2
Fig. 6. TOFs of phenol or CHONE hydrogenation are calculated based on the origi-
nal rates and listed in Table 2. It can be seen that a great diꢁerence in the activity of
Pd/Al O and Pd/SiO catalysts for phenol hydrogenation in cyclohexane, ethanol
2
3
2
and dichloromethane is observed. Within the experimental time range, the reaction
in DCM almost cannot be carried out, and the conversion in ethanol reaches low
values. For example, over Pd/Al O -1 or Pd/SiO -1 catalyst, TOFs in ethanol and
2
Pd
3
2
−
1
−1
−1 −1
DCM are 1.8 and 0.004 mol mol h , or 1.4 and 0.006 mol mol h , respectively
Pd
(
Table 2). In either case, main products are CHONE and CHANOL. These results
indicate that there is no correlation between the above-mentioned parameters of sol-
vents and TOFs under the conditions used in this work. In the phenol hydrogenation
used for the preparation of CHONE, diꢁerent solvents were reported, such as scCO
2
[
31], CH Cl [22] and CH CH OH [30]. As reported, polar solvents are known
2
2
3
2
to enhance the adsorption of non-polar reactants and non-polar solvents enhance
the adsorption of polar reactants [33]. Phenol, as a polar reactant, has an aꢂnity
Fig. 6 Phenol hydrogenation in
the diꢁerent solvents over Pd/
Al O -1 at 333 K; blue line =
100
80
2
3
cyclohexane, red line = ethanol
and green line = dichlorometh-
ane; square = phenol, cycle
=
=
cyclohexanone and triangle
cyclohexanol. (Color ꢀgure
60
online)
4
2
0
0
0
0
50
100
150
200
250
300
350
Time / min
1
3

Selective phenol hydrogenation under mild condition over…
for polar solvents and, so, is favorable for concentration in polar solvents, such as
CH Cl and CH CH OH in this work. For example, the salvation of phenol with eth-
2
2
3
2
anol increases so as to consequently hinder phenol adsorption on the surface of cata-
lyst. The transfer of phenol from liquor to the surface of catalysts becomes diꢂcult.
Non-polar cyclohexane does not have good compatibility with phenol, compared
with the surface of catalysts with hydroxyl groups. It can be expected that phenol
diꢁuses in cyclohexane through the stationary layer onto the surface of Pd cata-
lysts more easily than in polar solvents. The diꢁerence of phenol diꢁusion through
the stationary layer in cyclohexane and ethanol can be attributed to solvent–phenol
interaction. Additionally, the solvent–catalyst interaction should be considered. As
known, hydroxyl groups resting on Al O surface are hydrophilic so that ethanol can
2
3
interact with catalysts, leading to the decrease in the active sites accessible for sub-
strates adsorption. A similar trend for the hydrogenation of CHONE to CHANOL
over Pd/Al O -1 in non-polar and polar solvents is observed. Considering the fact
2
3
that CHONE is non-polar, has good compatibility with cyclohexane and poor com-
patibility with ethanol, the activity for CHONE hydrogenation in ethanol should be
higher, that is, TOF
should be higher in ethanol. However, the interaction of
CHONE
ethanol with the surface of Pd catalysts exists to some extent. As a result, the reac-
tion almost cannot occur. The apparent activation energies for the hydrogenation of
phenol in the ethanol are estimated from the original rates within the temperature of
2
83–333 K to be 26 and 34.5 kJ/mol over Pd/Al O -1 and Pd/SiO -1, respectively,
2 3 2
which are similar to those obtained in cyclohexane. Similar results are observed over
Pd/Al O -0.5 and Pd/SiO -0.5 catalysts. These phenomena indicate that the diꢁer-
2
3
2
ence of hydrogenation eꢂciency in non-polar and polar solvents can be ascribed to
phenol diꢁusion, and not to reaction routes.
The reaction in DCM shows the lowest eꢂciency and the TOF is only
−1
−1
0
.002 mol mol h . Here, there are two possibilities: one is related to interaction
Pd
of phenol with DCM (salvation), and the other is strong adsorption of Cl species on
active sites (poisoning catalyst). In order to further investigate this low eꢂciency in
DCM, we add DCM in cyclohexane with diꢁerent amounts and modify the extent
of reaction with cyclohexane-rich solutions. The results show that the reaction is
greatly inhibited by adding only small amount of DCM and a great decrease of TOF
can be observed (Fig. 7). The deposition of Cl species on the surface of Pd/Al O
2
3
catalyst is detected by EDS analysis, implying that low eꢂciency in CH Cl results
2
2
from poisoning of Pd/Al O catalyst by Cl species adsorption. In other experiments,
2
3
we found that the increase in temperature and P can promote the removal of Cl
H2
species from the surface of Pd catalysts, and thus the activity for the hydrogenation
of phenol to a large extent (for example, at 6 bar for 2 h, the conversion increased to
2
0%).
Catalyst reusability
Catalyst recyclability in heterogeneous reactions is very important from the cost
point of view. In recycling study, the hydrogenation of phenol over Pd/Al O -1 or
2
3
Pd/SiO -1 catalysts was performed by repeatedly using the same catalyst at 1 bar.
2
1
3

X. Li et al.
Fig. 7 TOFs of phenol hydro-
genation as a function of DCM
content in cyclohexane over Pd/
Al O -1 at 333 K
2
5
2
3
20
1
1
5
0
5
0
0
20
40
60
80
100
DCM in CHX / %
After ꢀve successive runs, the concentration curves via time over both used Pd cata-
lysts are similar to those of the corresponding fresh Pd catalyst. A low Pd leaching
amount (less than 0.5% of Pd content of catalyst) was detected by ICP analysis of
the reaction liquor after the removal of catalyst. The result shows that Pd/Al O -1 or
2
3
Pd/SiO -1 catalyst are suitable for practical application.
2
Hydrogenation of tert‑butyl‑phenol
In order to investigate the activity of Pd catalysts for the hydrogenation of phenol via
the formation of σ–c, 2- or 4-tert-butyl-phenol (2-t-BPH or 4-t-BPH) are chosen as a
reactant for hydrogenation, as it adsorbs on the active site only via the formation of
σ–c because of a bulky tert-butyl group on the aromatic ring. As is generally known,
the adsorption energies of t-BPH via the formation of σ–c with C–H or C–OH on
the active sites is smaller than those via the formation of π–c with an aromatic ring,
implying that t-BPH can only be activated to a small extent. Therefore, the fact that
the conversion of 2-t-BPH or 4-t-BPH (Table 3) is much lower than that of phenol
may verify that phenol molecules adsorb on the active sites via the formation of
Table 3 The conversion of t-Butyl-phenols and selectivity for t-Butyl-CHONE and t-Butyl-CHANOL
t-Butyl-phenols
Pd/Al O -1
Pd/SiO -1
2
3
2
Conversion/%
Selectivity
C=O
Conversion/%
Selectivity
C=O
C–OH
C–OH
2
4
2
4
-t-Butyl-phenola 26.0
100
0
17.9
30.3
34.6
61.2
>99.9
98.1
96.7
<0.01
1.9
3.3
-t-Butyl-phenola 42.7
98.1
98.5
68.9
1.9
1.5
31.2
b
-t-Butyl-phenol
49.4
b
-t-Butyl-phenol
93.8
78.1
21.9
a
PH2: 1 bar; time: 6 h
PH2: 6 bar; time: 2 h; 750 rpm; mt-Butyl-phenols:mcatalysts =0.4
b
1
3

Selective phenol hydrogenation under mild condition over…
π–c with an aromatic ring. Moreover, the conversion of 4-t-Butyl-phenol is almost
2
times as 2-t-Butyl-phenol over either Pd/Al O -1 or Pd/SiO catalyst because of
2 3 2
steric eꢁects. The activity of Pd/Al O -1 for hydrogenation of t-BPH is higher than
2
3
that of Pd/SiO , indicating that Lewis acidity is favorable for the activation of t-
2
BPH. The increase in P can promote the conversion of t-BPH. The products are
H2
composed of t-butyl-CHONE and t-butyl-CHANOL. Compared with the hydrogen-
ation of phenol, the decrease in selectivity for ketone is observed at 1 bar and 6 bar
P , that is, t-butyl-CHONE can be hydrogenated when a signiꢀcant amount of t-
H2
BPH exists. This phenomenon indicates that the adsorption energies of t-BPH via
the formation of σ–c are smaller than that of phenol via the formation of π–c so that
t-butyl-CHONE can compete with t-BPH to some extent.
Conclusion
Pd/Al O and Pd/SiO catalysts were prepared by wet impregnation of H PdCl
4
2
3
2
2
aqueous solution, and used in the test of selective hydrogenation of phenol into
CHONE. 98% or more yield of CHONE can be obtained over Pd/Al O and Pd/
2
3
SiO at 333 K under atmospheric pressure in cyclohexane solvent. Lewis acidity of
2
Pd/Al O can promote the conversion of phenol, but does not aꢁect obviously the
2
3
selectivity for CHONE. The hydrogenation of CHONE occurs until the conversion
of phenol is high. Pd/SiO with smaller Pd nano-particles presents higher selectiv-
2
ity. The increase in the polarity of solvent decreases greatly the eꢂciency of phenol
hydrogenation, due to strong interaction of solvent with phenol. Using t-BPH as a
probe, it can be deduced that that phenol adsorbs on Pd catalysts via the formation
of π–c with an aromatic ring. Increased hydrogen pressure not only promotes sig-
niꢀcantly the rates of hydrogenation, but also increases the selectivity for CHONE,
especially over Pd/SiO -1 catalyst.
2
Acknowledgements We would like to acknowledge the National Key Research and Development
Program of China (no. 2016YFC0204300) and National Natural Science Foundation of China (nos.
2
1477036 and 21777043).
References
1
2
3
.
.
.
K. Weissermel, H.J. Arpe, Industrial Organic Chemistry (VHC, Weinheim, 1993), p. 251
M. Howe-Grant (ed.), Kirk-Othmer Encyclopedia of Chemical Technology (Wiley, New York, 1991)
L. Liu, Y. Li, H. Wei, M. Dong, J. Wang, A.M. Slawin, J. Li, J. Dong, R.E. Morris, Angew. Chem.
Int. Ed. 48, 2206 (2009)
4
5
.
.
S.G. Shore, E. Ding, C. Park, M.A. Keane, Catal. Commun. 3, 77 (2002)
L.M. Sikhwivhilu, N.J. Coville, D. Naresh, K.V.R. Chary, V. Vishwanathan, Appl. Catal. A Gen.
3
24, 52 (2007)
6
7
.
.
S. Scirè, S. Minicò, C. Crisafulli, Appl. Catal. A Gen. 235, 21 (2002)
A.M. Barrios, C.A. Teles, P.M. de Souza, R.C. Rabelo-Neto, G. Jacobs, B.H. Davis, L.E.P. Borges,
F.B. Noronha, Catal. Today 302, 115 (2018)
8
9
.
.
P.M. de Souza, R.C. Rabelo-Neto, L.E.P. Borges, G. Jacobs, B.H. Davis, U.M. Graham, D.E.
Resasco, F.B. Noronha, ACS. Catal. 5, 7385 (2015)
S. Narayanan, K. Krishna, Appl. Catal. A Gen. 198, 13 (2000)
1
3

X. Li et al.
1
1
1
1
1
0. Y.Z. Chen, C.W. Liaw, I. Lee, Appl. Catal. A Gen. 177, 1 (1999)
1. N. Mahata, V. Vishwanathan, Catal. Today 49, 65 (1999)
2. P. Claus, H. Berndt, C. Mohr, J. Radnik, E.J. Shin, M.A. Keane, J. Catal. 192, 88 (2000)
3. N. Mahata, V. Vishwanathan, J. Catal. 196, 262 (2000)
4. K.A. Resende, C.A. Teles, G. Jacobs, B.H. Davis, D.C. Cronauer, A.J. Kropf, C.L. Marshall, C.E.
Hori, F.B. Noronha, Appl. Catal. B Environ. 232, 213 (2018)
1
5. N. Mahata, K.V. Raghavan, V. Vishwanathan, C. Park, M.A. Keane, Phys. Chem. Chem. Phys. 3,
2
712 (2001)
1
1
6. K.V.R. Chary, D. Naresh, V. Vishwanathan, M. Sadakane, W. Ueda, Catal. Commun. 8, 471 (2007)
7. K.A. Resende, C.E. Hori, F.B. Noronha, H. Shi, O.Y. Gutierrez, D.M. Camaioni, J.A. Lercher, Appl.
Catal. A Gen. 548, 128 (2017)
1
1
2
8. S. Hu, G.X. Yang, H. Jiang, Y.F. Liu, R.Z. Chen, Appl. Surf. Sci. 435, 649 (2018)
9. Y.Y. Zhu, G.Q. Yu, J. Yang, M. Yuan, D. Xu, Z.P. Dong, J. Colloid Interface Sci. 533, 259 (2019)
0. S. Hu, X. Zhang, Z.Y. Qu, H. Jiang, Y.F. Liu, J. Huang, W.H. Xing, R.Z. Chen, J. Ind. Eng. Chem.
5
3, 333 (2017)
2
1. J.F. Van Peppen, W.B. Fisher, C.H. Chan, in Phenol Hydrogenation Process in Catalysis of Organic
Reactions, ed. by R.L. Augustine (Marcel Dekker, New York, 1980), p. 355
2. H.H. Liu, T. Jiang, B.X. Han, S.G. Liang, Y.X. Zhou, Science 326, 1250 (2009)
3. R.D.D. Putra, H.L. Trajano, S.D. Liu, H. Lee, K. Smith, C.S. Kim, Chem. Eng. J. 350, 181 (2018)
4. N. Itoh, W. Xu, Appl. Catal. A Gen. 107, 83 (1993)
2
2
2
2
2
2
2
2
3
3
3
3
5. H. Li, J. Liu, S. Xie, M. Qiao, W. Dai, Y.F. Lu, H.X. Li, Adv. Funct. Mater. 18, 3235 (2008)
6. C. Park, M.A. Keane, J. Colloid Interface Sci. 266, 183 (2003)
7. X. Yang, S.J. Liao, J.H. Zeng, Z.X. Liang, Appl. Surf. Sci. 257, 4472 (2011)
8. Y. Zheng, P. He, Y. Fang, X. Yang, H. Liang, RSC. Adv. 7, 31582 (2017)
9. J. Matos, A. Corma, Appl. Catal. A Gen. 404, 103 (2011)
0. X. Yang, L. Du, S.J. Liao, Y.X. Li, H.J. Song, Catal. Commun. 17, 29 (2012)
1. H. Yoshida, S. Narisawa, S. Fujita, M. Arai, J. Mol. Catal. A Chem. 379, 80 (2013)
2. Y.J. Lao, N.X. Zhu, X.X. Jiang, J. Zhao, Q.G. Dai, X.Y. Wang, Catal. Sci. Technol. 8, 4797 (2018)
3. J.A. Anderson, A. Athawale, F.E. Imrie, F.-M.M. Kenna, A.M. Cue, D. Molyneux, K. Power, M.
Shand, R.P.K. Wells, J. Catal. 270, 9 (2010)
1
3