The direct reaction between CO2 and phenol catalyzed by bifunctional catalyst ZrO2

Article abstract of DOI:10.1016/j.molcata.2007.07.041

The systematic investigation was carried out to elucidate the direct reaction between CO₂ and phenol over multiphase catalyst. Reactions carried out over ZrO₂, TiO₂, and KF/NaY + Mn²⁺ gave the high selectivity t

Full text of DOI:10.1016/j.molcata.2007.07.041

Available online at www.sciencedirect.com
Journal of Molecular Catalysis A: Chemical 277 (2007) 180–184
The direct reaction between CO2 and phenol catalyzed
by bifunctional catalyst ZrO2
Zhenhuan Li∗, Kunmei Su
College of Materials and Chemical Engineering, and Tianjin Key Lab of Fiber Modification & Functional Fiber,
Tianjin Polytechnic University, Tainjin 300160, China
Received 4 June 2007; received in revised form 22 July 2007; accepted 23 July 2007
Available online 28 July 2007
Abstract
The systematic investigation was carried out to elucidate the direct reaction between CO2 and phenol over multiphase catalyst. Reactions carried
out over ZrO2, TiO2, and KF/NaY + Mn2+ gave the high selectivity to 4-[4-hydroxybenzyl]phenol and 2-[2-hydroxybenzyl]phenol (HBP), and the
maximum HBP selectivity of 68% was obtained with ZrO2 as catalyst. XRD patterns showed that metastable tetragonal ZrO2 was the active phase
for HBP synthesis, and the bidentate carbonate on ZrO2 surface was responsible for high HBP selectivity. If Hbeta, HZSM-5, and Al2O3 were
used as catalyst, the main product was diphenyl ether (DPE). But when reactions were catalyzed by KY and KF/NaY, xanthone became the main
product; however, xanthone disappeared when KF/NaY coexisted with Mn(OCOCH3)2 as catalyst. The optimized reaction conditions for HBP
synthesis over tetragonal ZrO2 are 350 ◦C, 12 h, and 4.0 MPa CO2 pressure.
© 2007 Published by Elsevier B.V.
Keywords: Phenol; CO2; ZrO2; Bifunctional catalyst
1. Introduction
ate from methanol and CO2 [13,14]. Furthermore, the reaction
between phenol and CO2 might also provide a new method for
the treatment of organic pollutant phenol [15].
Carbon dioxide is an attractive C1 building block in organic
synthesis as it is highly functional, abundant, inexpensive, non-
toxic, and nonflammable. As petroleum reserves are depleted,
the development of efficient catalytic processes employing CO2
as a feedstock has become increasingly important as evidenced
by the intense research in this area in recent years [1]. For
example, it has been shown that using CO2 as reactant one
can generate formic acid [2], dimethyl formamide [3], formic
acid [4,5], methanol [6], dialkyl carbonate [7], urea, carbamate
[8,9], copolymer, cyclic carbonate [10], and salicylic acid [11].
Among the above mentioned researches, the synthesis of sali-
cylic acid from phenoxide (Na or K) and carbon dioxide through
Kolbe–Schmitt method has been practiced commercially for
more than one century [12]; however, no studies on the direct
reaction between phenol and CO2 under supercritical condi-
tions were carried out until now. The reaction between phenol
and CO2 might provide a new application of phenol and CO2
in organic synthesis just like the synthesis of dimethyl carbon-
In the catalytic reactions of carbon dioxide, ZrO2 is widely
used as active catalyst for hydrogenation of carbon dioxide
and carbon oxide. Previous studies [16–18] indicated that CO2
reacted with amorphous ZrO2 to give HCO3− which largely dis-
appeared as temperature increased to 400 ◦C. When tetragonal
ZrO2 reacted with CO2 at 25 ◦C, CO32− and HCO3− ions were
formed, and bidentate carbonate became the main species after
reaction temperature increased to 400 ◦C. If monoclinic ZrO2
reacted with CO2 at 400 ◦C, a large amount of unidentate car-
bonate existed on the zirconia surface in addition to bidentate
carbonate. The formation of bidentate and unidentate carbon-
ate was due to coordinative unsaturated Zr4+–O2− pairs which
was detected by the temperature-programmed desorption (TPD)
experiment [18,19]. Modes of the phenol adsorption on ZrO2
surface are non-planar and co-planar to the surface [20] and are
governed by acid–base properties of ZrO2 [21,22].
Phenol and CO2 could compete for adsorption on the ZrO2
surface or reaction with ZrO2, since both phenol and CO2 are
acids. If phenol adsorbs onto ZrO2, non-planar or co-planar
phenoxide will be formed which might further react with
CO2 to give salicylate, phenyl salicylate, or other derivatives
∗
Corresponding author. Tel.: +86 22 24528359; fax: +86 22 24528504.
E-mail address: Zhenhuanli@sohu.com (Z. Li).
1381-1169/$ – see front matter © 2007 Published by Elsevier B.V.
doi:10.1016/j.molcata.2007.07.041

Z. Li, K. Su / Journal of Molecular Catalysis A: Chemical 277 (2007) 180–184
181
through the Kolbe–Schmitt reaction. If CO2 adsorbs onto ZrO2,
bidentate carbonate and unidentate carbonate will be formed
at higher temperature, which might stop the Kolbe–Schmitt
reaction and result in new products.
Here, studies were carried out on the reaction between CO2
and phenol under supercritical conditions. Among the catalyst
screened, ZrO2 is the best catalyst for this reaction. At the same
time, effects of ZrO2 phase, reaction temperature, reaction dura-
tion, and CO2 pressure were investigated.
Quantitative analysis was carried out on a gas chromatograph
(ShimadzuGC-14BwithaFIDdetector, DB-1capillarycolumn)
with cetane as interior standard.
Powder X-ray diffraction (XRD) pattern characterization of
the catalyst samples was measured on a Bruker AXS (Germany)
diffractometer using Cu K␣ radiation. The data were recorded
from 20◦ to 70◦ (2θ).
N2 adsorption–desorption isotherms of the mesoporous ZrO2
samples were measured at 77 K on a micromeritics Tris-
tar 3000 sorptometer (Micromeritics Instrument Corporation,
USA). Prior to the measurement, all samples were outgassed at
473 K and 1.33 × 10−4 Pa over night. The specific surface areas
of the mesoporous samples were calculated by the BET method.
The pore size distribution of the samples was determined from
the adsorption branch of the isotherms using the BJH method,
and the pore size was obtained from the peak position of the
distribution curves.
2. Experimental
2.1. Catalyst preparation
The preparation of zirconia by the hydrolysis of zirconium
chloride with excess of NH4OH has been described previously
[23]. A 10% solution of zirconium chloride was slowly added
to a well-stirred precipitating solution of NH4OH (5 M, ca. 50%
excess base) at room temperature. After the addition of all the
zirconium chloride, the base concentration was 0.55 M. The pre-
cipitate was placed in a round-bottomed pyrex flask and refluxed
at 100 ◦C in the supernatant liquid for 48 h. The precipitate was
filtered and dried at 80 ◦C for 6 h. Pure ZrO2 was obtained and
activated at certain temperature (400 ◦C) for 3 h before use.
TiO2 was synthesized according to an earlier paper [24].
Titanium chloride (TiCl4) was dissolved in deionized water.
Then, titanium hydroxide was precipitated from TiCl4 solution
by adding ammonium hydroxide until pH 8.0. The precipitate
was filtered and washed with distilled water until complete elim-
ination of chlorine ions in the liquid phase. The precipitate was
filtered and dried at 80 ◦C for 6 h. Pure TiO2 was obtained and
activated at 550 ◦C for 3 h before use.
Commercial Hbeta (Si/Al = 30) and HZSM-5 (Si/Al = 50)
were purchased from Nankai University. ␥-Al2O3 was obtained
from Shan Xi Commodity Chemistry Institute and NaY was
purchased from QiLu Corporation. KY was obtained by
exchange of NaY with a KCl solution (1 mol L−1) for three
times. KF/NaY (10 wt.%) was prepared by impregnation.
Hbeta, HZSM-5, ␥-Al2O3, KY, and KF/NaY were activated at
550 ◦C in air for 3 h before use.
3. Results and discussion
3.1. Effect of catalyst on the product distribution
The reaction results are listed in Table 1. The product dis-
tribution was greatly affected by the use of the catalyst (see
Scheme 1).
When Hbeta, HZSM-5, and ␥-Al2O3 were used as catalyst,
the main products were DPE and HBP. In addition, a trace of
ortho- or para-phenyl phenol, bis(4-hydroxyphenyl)methanone,
bis(2-hydroxyphenyl)methanone, salicylic acid, para-hydroxyl
benzoic acid, phenyl salicylate, and phenyl 4-hydroxy benzoate
were detected. If KY and KF/NaY were used as catalyst, xan-
thone was found as the main product. For example, xanthone
selectivity was 64 and 46% with KY and KF/NaY as catalyst,
respectively. However, xanthone disappeared and HBP became
the main product when KF/NaY coexisted with Mn (OOCCH3)2
as catalyst. When ZrO2, TiO2, and KF/NaY + Mn2+ were used
as catalyst, the high selectivity to HBP was obtained, and ZrO2
was the best catalysts for HBP formation of 67%.
Table 1
The activity of various catalysts for the reaction between carbon dioxide and
phenol
2.2. Synthesis procedure
Catalysts
Product distribution (mol%)
Yield (%) (HBP)
A definite quantity of catalyst and phenol were placed in
an autoclave of 250 ml (inner volume). The autoclave was then
sealed and flushed with CO2 to remove the air. After that, the
CO2 was pressurized in the autoclave to a given pressure and the
autoclave was heated to a given temperature. After reaction, the
autoclave was cooled down to room temperature. The contents
was diluted with THF and discharged to separate the catalyst by
the simple filtration.
DPE HBP Xanthone Others
H␤
69
65
40
11
9
16
28
23
68
36
9
–
–
–
10
15
64
37
–
15
8
5.2
5.9
3.9
4.7
2.9
0.8
2.4
6.3
HZSM-5
␥-Al2O3
ZrO2
TiO2
KY
37
12
41
12
15
25
14
19
8
KF/NaY
KF/NaY + Mn2+
29
60
Reaction condition: phenol, 0.2 mol; catalyst, 3 g (activated at 550 ◦C for 3 h);
PCO2 , 5.0 MPa; temperature, 400 ◦C; time, 6 h. DPE: diphenyl ether; HBP: 2-
(2-hydroxybenzyl)phenol and 4-(4-hydroxybenzyl)phenol. Others include little
of ortho- or para-phenyl phenol (PPH), bis(4-hydroxyphenyl)methanone, bis(2-
hydroxyphenyl)methanone, salicylic acid, para-hydroxyl benzoic acid (PHBA),
phenyl salicylate (PS), phenyl 4-hydroxy benzoate (PHB), and xanthene.
2.3. Analysis
The structure of the products was confirmed by GC–MS
(HP5972) (capillary column: 30 m SE-30, 0.25 mm i.d., and
0.25 ␮m film thickness) and compared with authentical samples.

182
Z. Li, K. Su / Journal of Molecular Catalysis A: Chemical 277 (2007) 180–184
Scheme 1. The reaction between CO2 and phenol.
Table 2
3.2. Effect of ZrO2 morphology
The texture properties of ZrO2
ZrO2 was activated at 200, 300, 400, 500, and 600 ◦C for 3 h.
XRD patterns of ZrO2 are shown in Fig. 1. When ZrO2 was
calcined at 200 and 300 ◦C, the morphology of ZrO2 was amor-
phous. Both monoclinic and tetragonal phases were observed
in the ZrO2 calcined at 500 and 600 ◦C, and monoclinic phase
became the main component. But only tetragonal phase was
found in the ZrO2 calcined at 400 ◦C.
ZrO2
BET surface
Total pore volume
Pore diameter
(nm)
(m2 g−1
)
(cm3 g−1
)
200
300
400
500
600
368
284
173
99
0.47
0.42
0.36
0.33
0.33
6.3
6.4
7.5
8.9
12.2
73
The textural and physical properties of the samples are sum-
marized in Table 2. The elevating of calcination temperature led
to a continuous decrease of the specific surface area and the
cumulative pore volume, while pore diameter increased from
6.3 to 12.2 nm.
Reactions catalyzed by ZrO2 were carried out, and results
are listed in Fig. 2. When amorphous ZrO2 calcined at 200 and
300 ◦C was used as catalyst, HBP selectivity was 29 and 31%.
The by-products including DPE, phenyl phenol, phenyl sali-
cylate, xanthone, and xanthene were found. When tetragonal
ZrO2 calcined at 400 ◦C was used as catalyst, HBP selectivity
was 45%. While ZrO2 calcined at 500 and 600 ◦C were used
as catalyst, HBP selectivity declined to 24 and 20%, but DPE
selectivity increased to 5 and 19%. In addition, the others includ-
ing xanthone and phenyl salicylate increased to 58 and 35%.
Those results indicated that the most active phase for HBP for-
mation was metastable tetragonal, while the BET surface and
pore diameter were not the determined factor in HBP formation.
Previous study [15] indicated that CO2 reacted with amor-
phous ZrO2 at 25 ◦C to give HCO3− which largely disappeared
Fig. 2. Effect of ZrO2 phase on the reaction between phenol and CO2 (reaction
condition: ZrO2, 2 g; phenol, 0.1 mol; PCO2 , 4.0 MPa; reaction temperature,
350 ◦C; reaction time, 3 h).
Fig. 1. XRD patterns of ZrO2 calcined at different temperature. Crystal struc-
ture: metastable tetragonal (ꢀ); monoclinic (ꢀ).

Z. Li, K. Su / Journal of Molecular Catalysis A: Chemical 277 (2007) 180–184
183
as temperature increased to 400 ◦C. The tetragonal ZrO2 reacte−d
with CO2 at 25 ◦C to produce two principal CO32− and HCO3
ions. After temperature increased to 400 ◦C, bidentate carbon-
ate became the main species. When monoclinic ZrO2 reacted
with CO2 at 400 ◦C, a large amount of unidentate carbonate
existed on the zirconia surface in addition to bidentate carbon-
ate. According to above described results, the conclusion could
be drawn that the formation of HBP might be relevant with
bidentate carbonate.
Zr4+O2− Lewis acid/base pairs [25,26] might serve as active
centers for the synthesis of HPB from phenol and carbon dioxide
over zirconia. The formation of bidentate carbonate required an
acid–base pair center associated with an anionic vacancy which
would generate electron-rich zirconium cations [17]. As for
amorphous ZrO2, there were less anionic vacancy and acid–base
pair centers on the surface of ZrO2. The scarcity of bidentate
carbonate on surface of amorphous zirconia might result in less
HBP formation. Monoclinic ZrO2 had the strong Lewis acidity
and Lewis basicity, which were investigated by the temperature-
programmed desorption (TPD) experiment [18,19]. Lewis acid
site Zr4+ is responsible for DPE formation from phenol, while
O2− promotes the Kolbe–Schmitt reaction between phenol and
CO2 into phenyl salicylate and xanthone.
Fig. 4. EffectoftimeonthereactionbetweenphenolandCO2 undersupercritical
condition (reaction condition: phenol, 4.7 g; metastable tetragonal ZrO2, 0.5 g;
temperature, 350 ◦C; PCO2 , 5.0 MPa).
3.4. Effect of duration
Effect of reaction duration on the reaction between CO2 and
phenol at 350 ◦C was investigated and the results are listed in
Fig. 4. As the reaction time extended, both the selectivity to HBP
and the conversion of phenol increased. The optimized reaction
time for HBP synthesis was 12 h, and HBP selectivity amounted
to 87%. When reaction time exceeded to 12 h, the selectivity to
HBP declined due to ZrO2 lost activity from coke.
3.3. Effect of temperature
Reactions were carried out at different temperature and
results are shown in Fig. 3. The reaction was hardly observed
at 300 ◦C because of the inert property of CO2. When reaction
was carried out at 350 ◦C, the highest HBP selectivity of 90%
was achieved and the conversion of phenol amounted to 1.4%.
After the reaction temperature increased to 400 ◦C, much more
coke, xanthene, xanthone, bis(2-hydroxyphenyl)methanone,
and bis(4-hydroxyphenyl)methanone produced. In addition, the
methylation of phenol detected at 400 ◦C was due to the reaction
among CO2, phenol and hydrogen. The hydrogen came from the
thermolysis of phenol [27].
3.5. Effect of CO2 pressure
The effects of CO2 pressure on reaction are listed in Fig. 5.
When reactions were carried out at low CO2 pressure, HBP was
difficult to be synthesized while DPE was the main production.
In addition, by-products including ortho-, para-phenyl phenol
and benzofuranl were detected. After CO2 pressure increased
Fig. 3. Effect of temperature on the reaction between phenol and CO2 over
Fig. 5. Effect of CO2 pressure on the reaction between phenol and CO2 (reaction
condition: phenol, 4.7 g;metastabletetragonal ZrO2, 0.5 g;reactiontemperature,
350 ◦C; reaction time, 4 h).
ZrO2 (reaction condition: phenol, 4.7 g; metastable tetragonal ZrO2, 0.5 g; PCO
,
2
5.0 MPa; reaction time, 4 h).

184
Z. Li, K. Su / Journal of Molecular Catalysis A: Chemical 277 (2007) 180–184
to 4.0 MPa, HBP selectivity achieved maximum of 72%. But
whenCO2 amountedto5.0 MPa, theselectivitytoHBPdeclined,
which was contrary to our primary expectation. High HBP selec-
tivity was obtained at 4 MPa due to the following reasons: first,
high CO2 pressure was in favor of the formation of bidentate
carbonate species; secondly, high density of CO2 facilitated the
removal of compounds from the catalyst surface to keep cata-
lyst from being deactivated. The reason for the decline of HBP
selectivity at 5 MPa might be that ZrO2 surface was occupied
by CO2 molecule which kept phenol from reaction.
References
[1] E.J. Beckman, J. Supercrit. Fluids 00 (2003) 1.
[2] W. Leitner, Angew. Chem. Int. Ed. Engl. 34 (1995) 2207.
[3] P.E. Jessop, Y. Hsiao, T. Ikariya, R. Noyori, J. Am. Chem. Soc. 118 (1996)
344.
[4] M. Shi, K.M. Nicholas, J. Am. Chem. Soc. 119 (1997) 5057.
[5] P. Braunstein, D. Matt, D. Nobel, Chem. Rev. 88 (1988) 747.
[6] P.G. Jessop, T. Ikariya, R. Noyori, Chem. Rev. 95 (1995) 259.
[7] W. McGhee, D. Riley, J. Org. Chem. 60 (1995) 6205.
[8] D.B. Dell Amico, F. Calderazzo, L. Labella, F. Marchetti, G. Pampaloni,
Chem. Rev. 103 (2003) 3857.
[9] M. Aresta, E. Quaranta, Tetrahedron 48 (1992) 1515.
[10] D.J. Darensbourg, M.W. Holtcamp, G.E. Struck, J. Am. Chem. Soc. 121
(1999) 107.
4. Conclusion
[11] Y. Kosugi, Y. Imaoka, F. Gotoh, M.A. Rahim, Y. Matsui, K. Sakanishi, Org.
Biomol. Chem. 1 (2003) 817.
[12] A.S. Lindsey, H. Jeskey, Chem. Rev. 57 (1957) 583.
[13] K. Tomishige, T. Sakaihori, Y. Ikeda, K. Fujimoto, Catal. Lett. 58 (1999)
225.
[14] C. Jie, Y. Guo, C. Wang, C. Hu, Y. Wu, E. Wang, Appl. Catal. A 256 (2003)
203.
[15] J. Zhao, Z. Liu, D. Sun, J. Catal. 227 (2004) 297.
[16] W. Hertl, Langmuir 5 (1989) 96.
[17] B.B. Baeza, I.R. Ramos, A.G. Ruiz, Langmuir 14 (1998) 3556.
[18] K. Pokrovski, K.T. Jung, A.T. Bell, Langmuir 17 (2001) 4297.
[19] D. He, Y. Ding, H. Luo, C. Li, J. Mol. Catal. A 208 (2004) 267.
[20] S. Velu, M.P. Kapoor, S. Inagaki, K. Suzuki, Appl. Catal. A 245 (2003)
317.
[21] G. Neri, A.M. Visco, A. Donato, C. Milone, M. Malentacchi, G. Guibitosa,
Appl. Catal. A 110 (1994) 49.
[22] N. Mahata, K.V. Raghavan, V. Vishwanathan, C. Park, M.A. Keane, Phys.
Chem. Chem. Phys. 3 (2001) 2712.
In the reaction between phenol and CO2 under supercritical
conditions, high HBP selectivity was achieved with ZrO2, TiO2
and KF/Ac + Mn2+ as catalysts, which might be ascribed to their
bifunctional sites. If Hbeta, HZSM-5, and Al2O3 were used as
catalyst, reactions gave a large amount of DPE. But when alkali
KY and KF/NaY were used as catalyst, xanthone becomes the
major compound because alkali site promotes reaction to adopt
the Kolbe–Schmitt reaction.
The high HBP selectivity was achieved over ZrO2, and
metastable tetragonal ZrO2 was suggested to be active. The high
HBP selectivity achieved over tetragonal ZrO2 may be relevant
with bidentate carbonate on the surface of ZrO2. The optimized
reaction conditions for HBP synthesis over tetragonal ZrO2 are
350 ◦C, 12 h, and 4.0 MPa for reaction temperature, duration,
and CO2 pressure, respectively.
[23] G.K. Chuah, S. Jaenicke, B.K. Pong, J. Catal. 175 (1998) 80.
[24] H. Zou, Y.S. Lin, Appl. Catal. A 265 (2004) 35.
[25] J. Fung, I. Wang, J. Catal. 164 (1996) 166.
Acknowledgment
[26] K. Tanabe, W.F. Holderich, Appl. Catal. A 181 (1999) 399.
[27] W.B. Kim, Y.G. Kim, J.S. Lee, Appl. Catal. A 194/195 (2000) 403.
TheauthorsaregratefulforthefinancialsupportoftheTianjin
Polytechnic University.