Deactivation causes of supported palladium catalysts for the oxidative carbonylation of phenol

Article abstract of DOI:10.1016/j.molcata.2016.07.028

Palladium catalysts supported on manganese oxides octahedral molecular sieves (OMS-2), Pb doped OMS-2 (Pb-OMS-2) and perovskite-type La_0.5Pb_0.5MnO₃ were prepared for investigating the deactivation causes of catalysts during the oxidative carbonylation of phenol. The catalytic experiments on three different sets of catalysts demonstrated that they were inactive after continuous reaction for 22?h. XRD patterns and HRTEM images demonstrated that carbonaceous pollution and palladium leaching occurred during the reaction process. The solvent and oxygen have significant impact on the palladium leaching by ICP-AES analysis. In addition, hot filtration tests and SH-HZSM-5 adsorption studies indicated that the leached palladium species are not active species. The results of XPS patterns and H₂-TPR profiles showed that aggregation and reduction of Pd species was another deactivation cause. It was discovered that lattice oxygen species decreased and palladium aggregated, which hindered the re-oxidation of Pd(0) to Pd(II). In summary, carbonaceous pollution, Pd leaching, aggregation and consumption of oxygen species hindered the redox recycling of palladium, which resulted in the deactivation of supported palladium catalysts for the oxidative carbonylation of phenol.

Full text of DOI:10.1016/j.molcata.2016.07.028

Accepted Manuscript
Title: Deactivation causes of supported palladium catalysts for
the oxidative carbonylation of phenol
Author: Chaofan Yin Jie Zhou Qiming Chen Jinyu Han
Yuanxin Wu Xiaojun Yang
PII:
S1381-1169(16)30278-3
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2016.07.028
MOLCAA 9962
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
11-4-2016
9-7-2016
15-7-2016
Please cite this article as: Chaofan Yin, Jie Zhou, Qiming Chen, Jinyu Han, Yuanxin
Wu, Xiaojun Yang, Deactivation causes of supported palladium catalysts for the
oxidative carbonylation of phenol, Journal of Molecular Catalysis A: Chemical
http://dx.doi.org/10.1016/j.molcata.2016.07.028
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

1
2
Deactivation causes of supported palladium catalysts for
the oxidative carbonylation of phenol
3
4
5
Chaofan Yina, Jie Zhoua, Qiming Chen a, Jinyu Hanb,Yuanxin Wua,b , Xiaojun Yanga*
6
7
8
9
10
11
12
aKey Laboratory for Green Chemical Process of Ministry of Education, Wuhan Institute of Technology，Wuhan 430073,
P R China
bSchool of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, P R China
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29

30
31
32
HIGHLIGHTS
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
•
•
•
•
•
Leaching and reduction of Pd species are main causes of catalyst deactivation.
The palladium leaching is attributed to the presence of O2, not CO.
High temperature promotes Pd agglomeration, leading to the reduction of active sites.
The decreases of average manganese valence leads to the weakening of redox ability.
Lattice oxygen take part in the regeneration of reduced Pd (0) in the catalytic redox cycle.

60
61
62
63
64
65
66
67
68
69
70
71
72
73
Abstract
Palladium catalysts supported on manganese oxides octahedral molecular sieves (OMSꢀ2), Pb doped
OMSꢀ2 (PbꢀOMSꢀ2) and perovskiteꢀtype La0.5Pb0.5MnO3 were prepared for investigating the deactivation
causes of catalysts during the oxidative carbonylation of phenol. The catalytic experiments on three
different sets of catalysts demonstrated that they were inactive after continuous reaction for 22h. XRD
patterns and HRTEM images demonstrated that carbonaceous pollution and palladium leaching occurred
during the reaction process. The solvent and oxygen have significant impact on the palladium leaching by
ICPꢀAES analysis. In addition, hot filtration tests and SHꢀHZSMꢀ5 adsorption studies indicated that the
leached palladium species are not active species. The results of XPS patterns and H2ꢀTPR profiles showed
that aggregation and reduction of Pd species was another deactivation cause. It was discovered that lattice
oxygen species decreased and palladium aggregated, which hindered the reꢀoxidation of Pd(0) to Pd(II). In
summary, carbonaceous pollution, Pd leaching, aggregation and consumption of oxygen species hindered
the redox recycling of palladium, which resulted in the deactivation of supported palladium catalysts for
the oxidative carbonylation of phenol.
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
Keywords: Diphenyl carbonate; Oxidative carbonylation; Deactivation causes; Pd leaching; Oxygen
species

89
90
1. Introduction
Palladiumꢀcatalyzed oxidative carbonylation reaction has become a novel and powerful tool for diphenyl
carbonate (DPC) synthesis [1, 2]. The process of oxidative carbonylation has several advantages over the
traditional phosgene process as the latter comprises the use of toxic and corrosive reagents. The process of
oxidative carbonylation is environmentallyꢀfriendly and utilizes easily available raw materials with
molecular oxygen as the oxidant. Many efforts were made to develop a highꢀperformance catalyst for the
oxidative carbonylation reaction which could be easily separated and repeatedly used. It is generally
accepted that palladium is active species and its catalytic performance is better than the other metal
catalysts[3ꢀ4]. At the previous stage, homogenous palladium catalysts were referred and PdCl2, Pd(OAC)2,
or Pd(acac)2 were common homogeneous catalysts. Because the ligands are beneficial to cisꢀreaction
between phenol and CO, the Pdꢀcarbene complex with bis(heterocyclic carbene) ligands and Pdꢀ2,2’ꢀ
bipyridyl complex with 2,2’ꢀbipyridyl ligands were prepared and became another kind of high efficient
homogeneous catalysts. Homogeneous catalysts have not been commercialized because of one major
disadvantage compared with heterogeneous catalysts: the difficulty encountered when trying to separate
the catalyst from reaction product or solvent. In recent years, heterogeneous palladium catalysts turn into
mainstream according to the reported literature. A number of heterogeneous palladium catalysts anchored
on supports including activated carbon, polystyrene, silicon dioxide, layered double hydroxides, organicꢀ
inorganic hybrid materials and mixed metal oxides were investigated [5].
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
A number of homogeneous or heterogeneous palladium catalysts were reported to be effective, but
catalyst deactivation has not been addressed in detail until now [5ꢀ11]. Among the several catalyst
deactivation causes described in the literature, five main causes of deactivation are poisoning, fouling,
thermal degradation initiated by high temperature, mechanical damage, and corrosion/leaching by the
reaction mixture. In palladiumꢀcatalyzed oxidative carbonylation of phenol，the deactivation of palladium
catalysts is generally attributed to the reduction and aggregation of the active center Pd(II) to Pd(0). The
formation of metallic Pd nanoparticles in significant amount occurred with homogeneous palladium
systems [10, 12],which indicated that Pd（II）transformed into Pd(0) in the reaction and led to catalyst
deactivation. Song [13] observes the peak representing metallic Pd in XRD patterns of carbonꢀsupported
palladium catalyst and speculated that metallic Pd catalyst was readily sintered in an enriched reducing
CO atmosphere. He even considered that zeroꢀvalent palladium particles were an active species. Leaching
is another deactivation cause to be considered [14]. The leached compounds rather than the solid phase

119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
maybe the actual active species. Takagi [15] reports that soluble Pd existed in Pd/CꢀPbO catalytic system.
Pd leaching was also observed in the system containing acetate salts, but the activity of PdCl2ꢀ
OIH(organic–inorganic hybrid) catalysts remained virtually unchanged when the catalyst was recycled and
reused [16]. Unlike PdCl2ꢀOIH catalysts, on account of Pd leaching and destruction of the tubular
structure, the activity of the PdꢀO/CeO2ꢀnanotube catalyst decreased substantially when the catalyst was
recycled and reused [17].
Our research group has always placed emphasis on the enhanced activity of Pbꢀdoped supported
palladium catalysts in oxidative carbonylation of phenol [18]. Nevertheless, the importance of catalyst
stability is often underestimated. Therefore, a better understanding of deactivation causes is essential for
improving and optimizing the performance of palladium catalysts, and several measures were taken to
evaluate the deactivation behavior of catalyst. Herein, deactivation causes for supported palladium
catalysts were studied, and some fundamental conclusions were obtained.
2. Experimental
2.1. Catalyst preparation
Three sets of oxide supports were obtained as suggested in literature. Manganese oxide octahedral
molecular sieves(abbreviated as KꢀOMSꢀ2) supports were synthesized by the reflux method. Lead doped
OMSꢀ2 supports(labeled as PbꢀOMSꢀ2) were prepared by an impregnation method. The KꢀOMSꢀ2 was
impregnated with a suitable volume of Pb(NO3)2 solution and then dried at 70ºC under vacuum.
Perovskiteꢀtype La0.5Pb0.5MnO3 supports were obtained by a solꢀgel method, using metal nitrates and
citric acid as starting materials in stoichiometric amounts. Reagents of KMnO4 (99.0%), MnSO4·H2O
(99.0%), La(NO3)3·nH2O(La2O3≥44%), Pb(NO3)2 (99.0%), Mn(NO3)2 solution(60%), citric acid
(98.0%), nitric acid (AR, 60%) were used as starting materials purchased from Sinopharm Chemical
Reagent Co.,Ltd.
KꢀOMSꢀ2 was prepared by a reflux method according to the literature [19]. To a 500 ml of roundꢀ
bottomed flask with a condenser, potassium permanganate solution (0.4 M, 225 ml) was added to a
mixture of manganese sulfate hydrate solution (1.75 M, 67.5 ml) and concentrated nitric acid (6.8 ml). The
dark brown slurry was refluxed for 24 h, then filtered and washed with deionized water for several times.
The KꢀOMSꢀ2 was dried at 120ºC overnight before use.
To prepare PbꢀOMSꢀ2, 7.000 g KꢀOMSꢀ2 prepared by the reflux method was impregnated with 70.0
ml of Pb(NO3)2 aqueous solutions with different concentrations. And then the mixture was stirred for 24 h,

149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
followed by evaporation. PbꢀOMSꢀ2 were obtained by calcination in a muffle furnace at 450ºC for 6
h[18].
The Perovskiteꢀtype La0.5Pb0.5MnO3 supports were obtained by a solꢀgel method, using metal nitrates
and citric acid as starting materials with nominal atomic ratios. Then, the aqueous solutions of metal
nitrates with nominal atomic ratios La: Pb: Mn=0.5:0.5:1 were mixed together in deionized water. Citric
acid(CA) was proportionally added to the metal solution to have the same amounts of equivalents. Then,
the solution was concentrated by evaporation at approximately 50ºC with stirring for 1 h to convert them
to stable (La, Mn)/CA complexes. The solution, while being stirred, heated at approximately 75ºC to
remove excess water. During the evaporation, reddishꢀbrown gases corresponding to NXOY removed from
the solution and a viscous and bubbly mass was produced upon cooling. Then, the dry gel was obtained by
letting the sol into an oven and heated slowly to 120ºC. The gel pieces were ground to form a fine
powder. Finally, La0.5Pb0.5MnO3 nanoparticles were gained by thermal treatment of the precursor at 650ºC
for 9 h in air. [20].
0.42 g PdCl2 was dissolved in 50.0 ml aqueous solution when the pH value was adjusted to 1.0 with
concentrated hydrochloric acid. KꢀOMSꢀ2, PbꢀOMSꢀ2 and La0.5Pb0.5MnO3 supports were dipped into the
PdCl2 solution, respectively. 5.0% NaOH solution as a precipitant was dropped slowly till the pH value
was kept at 9.5. After stirred for 2 h, the solid was separated by filtration, then dried at 80ºC overnight and
calcined at 300ºC for 3 h. The powdery catalysts prepared were labeled as Pd/KꢀOMSꢀ2, Pd/PbꢀOMSꢀ
2and Pd/La0.5Pb0.5MnO3, respectively.
2.2. Characterization
The phase identification and crystalline structure analysis were determined by Xꢀray diffraction
(XRD) on a Panalytical X'Pert PRO XꢀRay diffractometer with highꢀintensity Cu Kα radiation (λ=1.54060
Å).
The chemical shift, valence of element and Pd content on surface were studied by Xꢀray
photoelectron spectra (XPS) in a PerkinꢀElmer PHI 1600 ESCA system with Mg Kα Xꢀray radiation
(1253.6 eV, 250 W).
The bulk chemical composition of the samples was determined by inductively coupled plasma atomic
emission spectroscopy (ICPꢀAES) on a Perkin Elmer ELAN DRCꢀe. Solid samples were dissolved in the
mixture of HCl and HNO3, and then diluted with ultrapure water.
H2ꢀTPR experiments were performed on a Micromeritics AutoChem II 2920. Each sample (50mg of

179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
fresh catalyst) was pretreated at 300ºC for 0.5 h and subsequently cooled to ambient temperature under an
Ar flow (30 mL/min). Then, a 10% H2ꢀ95% Ar flow (30 mL/min) was passed over the catalyst bed by
heating at a rate of 10ºC /min up to 900ºC. Finally, the catalyst was cooled under an Ar flow (30
mL/min). The consumption of H2 was monitored by a thermal conductivity detector (TCD).
High resolution transmission electron microscopy (HRTEM) images of the samples were obtained
using a FEI Tecnai G20 electron microscope operated at an accelerating voltage of 200 kV with an energy
dispersive Xꢀray analysis (EDXA) system.
2.3. Activity test
The oxidative carbonylation reaction was performed in a 250mL stainless steel autoclave equipped
with a magnetic stirrer. A typical reaction condition was as follows: phenol 47g (0.5mol),
(TBAB) 1g (3mmol) and Pd (in the catalyst) 0.9mmol were introduced into
the autoclave. Then the autoclave was sealed and heated up to 65ºC. Subsequently, the gas mixture of CO
and O2 (CO/O2 = 12/1 molar ratio) was charged. The conditions employed in this procedure fell within the
flammability range of CO in O2. After the reaction sustained at the pressure of 4.8MPa, the autoclave was
cooled and the products were taken out every 2 hours. The reaction products were quantified by a
capillary gas chromatography with a FID detector and a SEꢀ54 capillary column.
3. Results and discussion
3.1 Catalytic performance of supported palladium catalysts
Figure 1 describes the DPC yields of different supported palladium catalysts and also shows that the
DPC yields increase in the order of Pd/OMSꢀ2<Pd/La0.5Pb0.5MnO3<Pd/PbꢀOMSꢀ2. Further the DPC
yields for all the catalysts increased slightly when the oxidative carbonylation reaction extended up to 22h.
The majority of catalysts was inefficient and could be deemed of deactivated. Moreover, the Pd/OMSꢀ2
catalyst is found to be deactivated much faster than the other two supported catalysts. The DPC yields of
Pd/OMSꢀ2 catalyst almost remained constant after 10h reaction, while the DPC yields of Pd/PbꢀOMSꢀ2
catalyst continued to increase from 39.8% to 51.2% when reacted from 10h to 24h. To find out the causes
of fast deactivation of supported palladium catalysts, a comprehensive experimental study was planned
and utilizing various characterization techniques the causes were explored.
3.2 Structure stability of supported palladium catalysts
XRD patterns of fresh and used supported palladium catalysts are represented in Figure 2. Based on
the literature and our past research[18ꢀ20], the XRD patterns of Pd/OMSꢀ2 catalyst, Pd/PbꢀOMSꢀ2

209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
catalyst and Pd/La0.5Pb0.5MnO3 catalyst correspond well to the cryptomelaneꢀtype phase of K2ꢀ
xMn8O16(JCPDS 44ꢀ1386), the hollanditeꢀtype phase of Pb2ꢀxMn8O16 (JCPDS 42ꢀ1349)and the perovskiteꢀ
type phase of La0.5Pb0.5MnO3(JCPDS 22ꢀ1123) respectively. Figure 2(a) and 2(b) shows that no peak shift
is observed for the Pd/OMSꢀ2 catalyst, but its peak intensity at 2θ = 28.7 o, 37.5 o weakens. As illustrated
in Figure 2(c)ꢀ2(f), the crystal structures of the Pd/PbꢀOMSꢀ2 catalyst and the Pd/La0.5Pb0.5MnO3 catalyst
remain original structure before and after the reaction. Noticeably, no additional peaks related to active
palladium species appear in any of the catalysts, indicating that the palladium species is wellꢀdispersed on
the surface of supports.
The structure stability of the supported palladium catalysts was analyzed using TEM images. Figure
3(a) relates to the fresh Pd/OMSꢀ2 catalyst and demonstrates that highly dispersed palladium species with
particle (size of about 3nm) is loaded onto the OMSꢀ2 nanorod supports. Figure 3(b) relates to the used
Pd/OMSꢀ2 nanorod catalysts and demonstrates that catalysts are covered with membraneꢀlike compounds
which contain a large amount of carbon element as determined by EDXA analysis. This is consistent with
the weakening of its diffraction peaks in the XRD patterns caused by the embedding of carbon on the
surface. It is important that the active palladium species seems to have disappeared on the surface of
OMSꢀ2 nanorod supports. The palladium species also drops off of the surface of Pd/PbꢀOMSꢀ2 nanorod
catalysts similar to the Pd/OMSꢀ2 nanorod catalysts (Figure 3(c) and 3(d)). Further Figure 3(e) and 3(f)
shows that Pd/La0.5Pb0.5MnO3 catalysts look rather spongy due to a fast release of a large amount of gas in
the decomposition of the reacting precursors. Unfortunately, the palladium species could not be
distinguished in both fresh and used Pd/La0.5Pb0.5MnO3 catalysts. The above TEM images reveal that part
of the palladium species broke away from the catalyst carrier and this precisely indicates that leaching can
occur in oxidative carbonylation reaction. Thus, leaching is an important deactivation cause of supported
palladium catalysts. In order to explore whether palladium leaching results in the deactivation, we
conducted a verification test.
3.3 Palladium leaching
The palladium content of the fresh and used supported catalysts was measured by ICPꢀAES analysis
and the results are listed in Table 1. The data clearly indicate palladium leaching occurs during an
oxidative carbonylation reaction. Pd loss ratio is 6.2%, 9.1% and 35.0% for Pd/OMSꢀ2, Pd/PbꢀOMSꢀ2 and
Pd/La0.5Pb0.5MnO3 catalysts, respectively. Because there is a serious palladium leaching, Pd/PbꢀOMSꢀ2
catalyst was selected as only target and their parameters that influence the palladium leaching were then

239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
explored and are summarized in Table 2. It is clear from the data that the solvent and oxygen atmosphere
have a significant impact on the palladium leaching (Table 2, entries 7~9) [21]. The palladium content in
solution reached 3.51% for DCM and 8.90% for DMF (Table 2, entries 8~9). At O2 atmosphere (Table 2,
entries 7), 8.73% of the total palladium amount removed from the catalyst and were found in solution
compared to only 0.13% under CO and O2 mixed atmosphere (Table 2, entries 1). While the amount of Pd
leaching at CO+O2 atmosphere is much the same as that at air atmosphere (Table 2, entry 6). Thus, the
palladium leaching is attributed to the presence of O2, not CO. Reaction temperature has little bearing on
palladium leaching, and the palladium contents in solution dangle just below 1% at different temperatures
(Table 2, entries 1~5). We verified that the leached palladium was not a catalytically active species by the
hot filtration test. Hot filtration tests were conducted after running the normal 4h reaction and then the
solid catalysts were filtered under hot conditions [22]. The composition of the filtrate was monitored in the
same reaction environment to determine the DPC yield in the absence of the supported palladium catalyst.
As shown in Table 1, all catalysts in this study showed no activity after the hot filtration test. The reaction
never proceeded in the filtrate without the supported palladium catalyst and a slight decline of DPC yields
suggests that the DPC is being hydrolyzed instead of generation. Further, SHꢀHZSMꢀ5 was added into the
reactor and the corresponding DPC yield was 11.6%, compared to 12.1% without SHꢀHZSMꢀ5 under the
same condition [23, 24]. The results indicate that the leached palladium species is not active.
3.4 Reduction and aggregation of active palladium species
Both homogeneous and heterogeneous palladium catalyst are reported to be active in an oxidative
carbonylation of phenol [12]. Palladium species that leached into the solution became inactive which
clearly indicates that palladium species may be transforming into reduced state. XPS analysis is required
to obtain the information on the reduction degree of palladium species. The XPS spectra in Figure 4 show
a peak at 335.5 eV characteristic of Pd(0) and the other peak at 337.2 eV corresponding to Pd(II). The
binding energies of the entire used supported palladium catalysts shift toward lower binding energies as
compared to the fresh catalysts, indicating the reduction of active Pd(II) species to Pd(0). As shown in
Figure 4, the initial valency of palladium could have impact on the DPC yield. The initial valency of Pd
species is two in Pd/PbꢀOMSꢀ2 catalysts (Figure 4(c)), which shows the best catalytic activity. In contrast,
since a large amount of Pd(0) exists in Pd/OMSꢀ2 catalysts (Figure 4(a)), the DPC yield is the lowest and

269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
the deactivation is the fastest among all the catalysts. A preꢀreduction experiment was used to further
ensure that the reduction of palladium can deactivate the catalyst. The fresh Pd/PbꢀOMSꢀ2 catalyst was
then reduced with formaldehyde (Figure 1s). Results show that the catalytic activity sharply reduced and
the DPC yield was just 0.31%.
Take Pd/OMSꢀ2 catalyst as a model to investigate the aggregation of palladium species. Figure 5(a),
5(b) and 5(c) manifest that no XRD peaks of Pd(0) or Pd(II) appear in the fresh Pd/OMSꢀ2 catalyst when
calcined at 300ºC，indicating that calcination time can not affect the aggregation of palladium species.
After treating at 400ºC, the catalyst showed an XRD peak at 33. 9°(Figure 5(d)), which corresponds
well to the 101 plane of Pd(II) (JCPDS 43ꢀ1024) and indicates that high calcination temperature prompted
the reunion of Pd(II) on the surface. As illustrated in Figure 5(e), the aggregation of Pd(II) also occurred in
the used catalyst. Besides the peak of Pd(II) species , the spectra shows an additional peak at 40.1°,
which corresponds with the character of metal Pd(111)(JCPDS 46ꢀ1403). The aggregation of Pd(0) and
Pd(II) hindered the reꢀoxidation of palladium, which is another deactivation cause of the catalyst.
3.5 The change of oxygen species
Our research findings have demonstrated that oxygen species played an important role in the redox
cycle of palladium species [18], so the change of oxygen species for the fresh and used catalysts were
described by H2ꢀTPR profiles (Figure 6) and XPS spectra (Figure 7).
In the H2ꢀTPR profiles of Pd/OMSꢀ2 and Pd/PbꢀOMSꢀ2 catalysts three reduction peaks could be
deconvoluted according to the literature[25]. The α reduction peak can be possibly attributed to the
reduction of surface oxygen species. The β and γ reduction peaks stand for the reduction processes of
Mn4+→Mn3+→Mn2+. Figure 6 (a) and (c) shows that the reduction peaks of Pd/PbꢀOMSꢀ2 catalysts shift
to the high temperature regions relative to that of Pd/OMSꢀ2. It may be due to the higher PbꢀO bond
energy than MnO2, the present of PbꢀO bonds in the MnꢀOꢀPb makes the mobility of the oxygen atoms in
the octahedral crystal slow. The addition of the lead weakens the bulk redox ability of the OMSꢀ2. For the
La0.5Pb0.5MnO3 catalysts, each H2ꢀTPR profiles exhibit two reduction peaks. The first peak is located at
150~250℃and the other is positioned at 250~380℃. The reduction of Mn4+ to Mn3+ occurs in the first
process, and the reduction of Mn3+ to Mn2+ occurs in the second process. Figure 6 shows that the
reduction peaks of the entire used catalyst shift to higher temperature regions in comparison to that of the
fresh catalyst, implying that the Mn4+ reduces and low valence Mn3+ and Mn2+ increase for the used
catalysts. The average manganese valence decreases after reaction, which leads to the weakening of redox

299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
ability [18]. A significant shift is observed for the Pd/PbꢀOMSꢀ2 catalyst during reaction process, so the
corresponding redox ability reflects fully in catalytic cycle of active palladium species.
As shown in Figure 7, the oxygen species derived from fitted O1s spectra was also divided into two
groups: surface oxygen (Os, 531.7V) and lattice oxygen (OL, 528.7eV). From the peak area of three sets of
catalysts, it can be inferred that the lattice oxygen species reduces when catalysts are used. This is
consistent with the change of manganese valence. When the lattice oxygen species reduced, the low
valence manganese increased for charge equilibrium. It can be inferred that the oxygen species, especially
lattice oxygen, take part in the regeneration of reduced Pd (0) in the catalytic redox cycle. This is because
surface oxygen is underꢀcoordinated so that it is less reduced than that in lattice. Thus, the catalytic
performance of fresh Pd/PbꢀOMSꢀ2 catalyst which more lattice oxygen reduced is superior over the other
two catalysts.
3.6 Possible deactivation cause of supported palladium catalyst
From the results of these tests, possible deactivation causes of supported palladium catalysts can be
proposed:
Palladium leaching and reduction is the main cause of catalyst deactivation. Active palladium species
binds to outerꢀmost oxygen atom to form PdꢀO bonds through strong interaction. Presence of higher
number of Pd(II) species leads to higher interaction between Pd and O species and this makes the leaching
of palladium difficult in case of Pd/PbꢀOMSꢀ2 catalysts. However, lower palladium leaching for Pd/OMSꢀ
2 catalyst in spite of less Pd(II) species founded in Figure.3 (b), may be due to coating of carbonaceous
pollution on the catalyst. Further, the leached palladium is not catalytically active species. Meanwhile,
Pd(II) reduced to Pd(0) with the concomitant formation of DPC during the reaction. In order to regenerate
the catalytic cycle, the formed Pd(0) species needs to be reꢀoxidized to Pd(II) species. Oxygen species in
catalysts is effective to regenerate reduced Pd(0) in catalytic redox cycle.
Apart from causes mentioned above, the carbonaceous pollution can also lead to catalyst
deactivation. OMSꢀ2 has excellent hydrophobicity and strong affinity for organic compounds [26].
Surface bonded OH groups on OMSꢀ2 increases the adsorption of phenolic compounds. In oxidative
carbonylation reaction, organic reactants are absorbed on the surface of the Pd/OMSꢀ2 catalyst and
impeded the contact between reactants and active species.
4. Conclusions
Supported palladium catalysts were found to be deactivated in less than 22h during oxidative

329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
carbonylation of phenol and the deactivation causes have been studied. The structure stability of Pd/OMSꢀ
2, Pd/PbꢀOMSꢀ2 and Pd/La0.5Pb0.5MnO3 catalysts were found to be good in reaction environment. The
main deactivation causes are leaching and reduction of the palladium species. During the course of
reaction, the palladium was partially leached into the solution and experimental analysis has confirmed
that the leached palladium was not catalytically active species. Due to reduction in the amount of lattice
oxygen, the redox ability of supports to regenerate Pd(II) was reduced and more Pd(0) formed resulting in
the loss of catalytic activity.
Acknowledgement
We are grateful for the financial support from Key Laboratory of Green Chemical Process Ministry
of Education (Grant No.GCP201301) and Hubei Provincial Natural Science Foundation of China (Grant
No.2013CFB322). The authors also thank the Hubei Provincial Department of Education (Grant
No.Q20131503), the Opening Project of Hubei Key Laboratory of Lipid Chemistry and Nutrition
(201505) and Wuhan Institute of Technology (Grant No.10112061) for their supports.

359
Reference
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
[1] S.T. Gadge, B.M. Bhanage, RSC Adv. 4 (2014) 10367.
[2] X. Wu, H. Neumann, M. Beller, ChemSusChem. 6 (2013) 229.
[3] A.J. Chalk, US Patent. (1978): 4096169.
[4] A.J. Chalk, US Patent. (1980): 4187242.
[5] K.J.L. Linsen, J. Libens, P.A. Jacobs, Chem. Commun. 22 (2002) 2728.
[6] W. Xue, J. Zhang, Y. Wang, X. Zhao, Q. Zhao, Catal. Commun. 6 (2005) 431.
[7] L. Zhang, Y. He, X. Yang, H. Yuan, Z. Du, Y. Wu, Che. Eng. J. 278 (2015) 129.
[8] W. Lu, Z. Du, H. Hua, Q. Tian, Y. Wu, Chinese J Chem Eng. 21 (2013) 8.
[9] H. Ishii, K. Takeuchi, M. Asai, M. Ueda, Catal Commun. 2 (2001) 145.
[10] A. Vavasori, L. Toniolo, J Mol Catal AꢀChem. 139 (1999) 109.
[11] M. Goyal, R. Nagahata, J.i. Sugiyama, M. Asai, M. Ueda, K. Takeuchi, Catal Lett. 54 (1998) 29.
[12] L. Ronchin, A. Vavasori, E. Amadio, G. Cavinato, L. Toniolo, J Mol Catal AꢀChem. 298 (2009) 23.
[13] H.Y. Song, E.D. Park, J.S. Lee, J Mol Catal AꢀChem. 154 (2000) 243.
[14] Z. Li, Y. Wang, X. Ding, X. Zhao, J. Nat Gas Chem. 18 (2009) 104.
[15] M. Takagi, H. Miyagi, T. Yoneyama, Y. Ohgomori, J Mol Catal AꢀChem. 129 (1998) L1.
[16] G.Z. Fan, T. Li, G.X. Li, Appl Organomet Chem. 20 (2006) 656.
[17] Y. Yuan, Z. Wang, H. An, W. Xue, Y. Wang, Chinese J Catal. 36 (2015) 1142.
[18] X. Yang, J. Han, Z. Du, H. Yuan, F. Jin, Y. Wu, Catal Commun. 11 (2010) 643.
[19] Y. Li, Z. Fan, J. Shi, Z. Liu, J. Zhou, W. Shangguan, Catal Today. 256 (2015) 178.
[20] H. Tavakkoli, D. Beiknejad, T. Tabari , Water Treat. 52 (2014) 7377.
[21] RG. Heidenreich, EGE. Krauter, J. Pietsch, K. Kohler, J Mol Catal AꢀChem. 182 (2002) 499.
[22] Y. Ji, S. Jain, R. Davis, J Phys Chem B. 109 (2005) 17232.
[23] J. Richardson, C. Jones, J Catal. 251 (2007) 80.
[24] J. Webb, S. Macquarrie, K. Mceleney, C. Crudden, J Catal. 252 (2007) 97.
[25] L. Sun, Q. Cao, B. Hu, J. Li, J. Hao, G. Jing, X. Tang, Appl Catal A- Gen. 393 (2011) 323.
[26] B. Hu, C. Chen, S. Frueh, L. Jin, R. Joesten, S. Suib, J Phys Chem C. 114 (2010) 9835.
401

402 Figure Caption
403
404 Figure 1.
405
406
407
408
409
Figure 1 Catalytic performance of supported palladium catalysts(a) Pd/OMSꢀ2; (b) Pd/La0.5Pb0.5MnO3; (c)
Pd/PbꢀOMSꢀ2
410 Figure 2.
411
412
413
Figure 2 XRD patterns of fresh and used supported palladium catalysts (a) fresh Pd/OMSꢀ2; (b) used
Pd/OMSꢀ2; (c) fresh Pd/PbꢀOMSꢀ2; (d) used Pd/PbꢀOMSꢀ2; (e) fresh Pd/La0.5Pb0.5MnO3; (f) used
Pd/La0.5Pb0.5MnO3
414
415

416 Figure 3.
417
418
419
Figure 3. HRTEM images of fresh and used supported palladium catalysts (a) fresh Pd/OMSꢀ2; (b)
used Pd/OMSꢀ2; (c) fresh Pd/PbꢀOMSꢀ2; (d) used Pd/PbꢀOMSꢀ2; (e) fresh Pd/La0.5Pb0.5MnO3; (f) used
Pd/La0.5Pb0.5MnO3
420
421
422
423
424
425