Ortho-selective alkylation of phenol with 1-propanol catalyzed by CeO2-MgO

Article abstract of DOI:10.1006/jcat.1999.2416

Vapor-phase alkylation of phenol with 1-propanol was investigated over CeO₂-MgO catalysts prepared utilizing a molten mixture of the corresponding nitrates and citric acid. The CeO₂-MgO had a stable catalytic activity at 475°C, and had an excellent selectivity to 2-propylphenol, higher than 82% based on phenol. Although a portion of the 2-propylphenol produced was decomposed into o-cresol and 2-ethylphenol, a sum of selectivities to the monoalkylated phenols exceeded 97%. During the alkylation, propanal and 3-pentanone was observed. In the results of 1-propanol conversion without phenol, it was found that 1-propanol was dehydrogenated to propanal and that the propanal produced was dimerized to 3-hydroxy-2-methylpentanal via aldol addition, followed by the deformylation into 3-pentanone. Namely, in the reaction of phenol and 1-propanol, the propylation of phenol and the dehydrogenation of 1-propanol occurred concurrently over the CeO₂-MgO. The pure CeO₂, having both the redox property with Ce⁴⁺-Ce³⁺ and weak basic sites, catalyzed both the propylation of phenol and the 1-propanol transformation into 3-pentanone, while the pure MgO with strong basicity was less active for the 1-propanol transformation and had a low reaction rate in the alkylation. The reaction mechanism of the ortho-propylation over the CeO₂-MgO catalyst is speculated as follows. The ortho position of phenol adsorbed perpendicularly on weak basic sites of the catalyst is selectively alkylated by 1-propanol, which is possibly activated in the form of 1-hydroxypropyl radical on CeO₂ species rather than as a form of n-propyl cation. The redox property of CeO₂ is probably attributed to the homolitic activation of 1-propanol to produce 1-hydroxypropyl radical.

Full text of DOI:10.1006/jcat.1999.2416

Journal of Catalysis 184, 180–188 (1999)
Article ID jcat.1999.2416, available online at http://www.idealibrary.com on
Ortho-Selective Alkylation of Phenol with 1-Propanol
Catalyzed by CeO2–MgO
Satoshi Sato,1 Ryoji Takahashi, Toshiaki Sodesawa, Kotaro Matsumoto, and Yoichiro Kamimura
Department of Applied Chemistry, Faculty of Engineering, Chiba University, Yayoi, Inage, Chiba 263-8522, Japan
Received October 14, 1998; revised January 4, 1999; accepted January 5, 1999
HY (6, 7, 10), and other mixed metal oxides such as MgO-
based (1, 2, 4), Fe2O3-based (11–15), and V2O5-based ox-
ides (15–18) and Al-containing hydrotalcites (19–21) have
been investigated. They are divided into three groups with
respect to the reaction temperatures at which the cata-
lyst acts efficiently. First, acidic catalysts such as Al2O3,
H-ZSM5, and HY have methylation activities at temper-
atures lower than 300 C, whereas they are less selective be-
cause of their catalysis for both oxygen methylation and the
ringmethylation. Second, variousmixed oxides(1, 2, 11–21)
show catalytic activity at a temperature range between 300
and 450 C, and they are selective to the ortho-methylation
of phenol. Third, simple oxides such as MgO, CeO2, and
Mn3O4 show catalytic activity at temperatures higher than
450 C, and most of them are highly selective to the ortho-
position. Because the desorption process of acidic products
as well as the activation of methanol is important in the
catalytic reaction, acid–base properties of catalysts are sug-
gested to be a major factor of determining the optimum
reaction temperature.
In the previous report (4), CeO2–MgO catalyst systems,
which were prepared using a molten mixture of cerium (III)
nitrate, magnesium nitrate, and citric acid, were found to
have attractive catalytic performance without decay of ac-
tivity and had excellent selectivities to the sum of o-cresol
and 2,6-xylenol, higher than 98% at a temperature range
between 450 and 550 C. The catalytic activity correlated
well to the number of weak basic sites of CeO2 dispersed
in the MgO matrix: the turnover frequencies based on the
weak basic sites are constant for the CeO2–MgO system re-
gardless of preparation methods. The reaction mechanism
of the ortho-methylation over the CeO2–MgO catalyst has
been proposed: the ortho-position of phenol adsorbed per-
pendicularly on the weak basic site of CeO2 species is selec-
tively alkylated by methanol which is activated in the form
of a formyl or hydroxymethyl group rather than as a methyl
cation.
Vapor-phase alkylation of phenol with 1-propanol was investi-
gated over CeO2–MgOcatalystsprepared utilizinga molten mixture
of the corresponding nitrates and citric acid. The CeO2–MgO had
a stable catalytic activity at 475 C, and had an excellent selectiv-
ity to 2-propylphenol, higher than 82% based on phenol. Although
a portion of the 2-propylphenol produced was decomposed into
o-cresol and 2-ethylphenol, a sum of selectivities to the monoalky-
lated phenols exceeded 97%. During the alkylation, propanal and
3-pentanone was observed. In the results of 1-propanol conversion
without phenol, it was found that 1-propanol was dehydrogenated
to propanal and that the propanal produced was dimerized to 3-
hydroxy-2-methylpentanal via aldol addition, followed by the de-
formylation into 3-pentanone. Namely, in the reaction of phenol
and 1-propanol, the propylation of phenol and the dehydrogena-
tion of 1-propanol occurred concurrently over the CeO2–MgO. The
pure CeO2, having both the redox property with Ce4+–Ce3+ and
weak basic sites, catalyzed both the propylation of phenol and the
1-propanol transformation into 3-pentanone, while the pure MgO
with strong basicity was less active for the 1-propanol transforma-
tion and had a low reaction rate in the alkylation. The reaction
mechanism of the ortho-propylation over the CeO2–MgO catalyst
is speculated as follows. The ortho position of phenol adsorbed
perpendicularly on weak basic sites of the catalyst is selectively
alkylated by 1-propanol, which is possibly activated in the form of
1-hydroxypropyl radical on CeO2 species rather than as a form of
n-propyl cation. The redox property of CeO2 is probably attribu-
ted to the homolitic activation of 1-propanol to produce 1-hydro-
c
xypropyl radical.
1999 Academic Press
Key Words: alkylation; phenol; 1-propanol; dehydrogenation;
aldol addition; 3-pentanone; CeO2–MgO; base; citric acid.
INTRODUCTION
Alkylated phenols are important chemicals as interme-
diates in the agrochemical and polymer industries. In the
vapor-phase alkylation of phenol with methanol, various
oxides such as MgO (1–3), CeO2 (3, 4), Mn3O4 (5), and
Al2O3 (6–8), acidic zeolites such as H-ZSM5 (6, 7, 9, 10) and
Other alcohols such as ethanol, propanol, and butanol
were also used as alkylating reagents for the production
of alkylphenols (22–25). In the alkylation of phenol with
propanol, many alkylated products were observed over
1 To whom correspondence should be addressed. E-mail: satoshi@
planet.tc.chiba-u.ac.jp.
180
0021-9517/99 $30.00
c
Copyright
1999 by Academic Press
All rights of reproduction in any form reserved.

ORTHO-SELECTIVE PROPYLATION OF PHENOL OVER CeO2–MgO
181
Al2O3 (23, 24) and Al-containing hydrotalcites (25). In this was preheated in a glass tube reactor at 500 C for 1 h.
paper, we describe the characteristics of CeO2–MgO cata- 1-Propanol (17.6 mmol h 1) was fed into the reactor with
lysts for the alkylation of phenol with 1-propanol. Dur- helium carrier (50 mmol h 1). Liquid products were ana-
ing the alkylation, we found selective transformation of lyzed by the same GC unit as the propylation mentioned
1-propanol into 3-pentanone. Thus, we additionally report above, and a gaseous product such as propylene was an-
on the transformation of 1-propanol without phenol. As alyzed by FID-GC with a packed column of PEG-HT
we have previously demonstrated the phenol adsorption (1 m) at a constant temperature of 40 C. The conversion of
on weak basic sites (4), here we discuss the activation of 1-propanol is defined as a fraction of reacted 1-propanol.
alcohol and the nature of active species.
Characterization of Catalyst
EXPERIMENTAL
The temperature-programmed desorption (TPD) of ad-
sorbed CO2 was measured by neutralization titration using
an electric conductivity cell immersed in an aqueous solu-
tion, as has been described previously (4). The total TPD
profile was able to be deconvoluted into several Gausian
peaks. The number of total basic sites was obtained by the
amount of total desorbed CO2, and the number of weak
basic sites was calculated from peaks whose tops were at
temperatures lower than 150 C in the deconvoluted TPD
profile. The total surface area of the sample was calculated
by BET method using nitrogen isotherm at 196 C.
Catalyst Preparation
All reagents were supplied by Wako Chemical Ltd.
(Japan). CeO2–MgO samples with different CeO2 contents
were prepared by using cerium (III) nitrate hexahydrate,
magnesium nitrate hexahydrate, and citric acid monohy-
drate. The preparation procedure is essentially the same as
that for an amorphous citrate process (26), and the detail
was described in the previous paper (4). After the citrate
precursor had been heated in air at 170 C for 2 h, it was cal-
cined in air at 550 C for 2 h to give a CeO2–MgO sample.
Pure CeO2 and MgO were prepared from the molten salt of
the individual nitrate and citric acid. A reference catalyst,
DC-2282, a commercial alumina with a specific surface area
of 195 m2 g 1, was supplied by Dia Catalyst and Chemicals
Ltd.
RESULTS
Figure 1 shows changes in the phenol conversion of the
CeO2(11.2 mol% )–MgO with process time at 475 C, to-
gether with the results of pure CeO2 and MgO, for the
alkylation of phenol with 1-propanol. The catalytic activ-
ities decreased with process time in the initial period for
2 h and reached a stable conversion level. No decay in the
catalytic activities of each catalyst was observed after the
reaction started for 2 h. For the CeO2(11.2 mol% )–MgO
(curve b), the conversion level was higher than those of the
pure MgO (curve a) and the pure CeO2 (curve c).
Catalytic Reaction
The alkylation of phenol with 1-propanol was carried out
in a usual fixed bed flow reactor under atmospheric pres-
sure of helium at a temperature range between 425 and
500 C. Prior to the reaction, a catalyst sample (0.15 g) was
preheated in a glass tube reactor at 500 C for 1 h. A mix-
ture of phenol and 1-propanol with a molar ratio of 1 : 4
was fed into the reactor at a total flow rate of 17 mmol
h
1, together with helium flow of 50 mmol h 1. An effluent
collected in an ice trap was periodically extracted and an-
alyzed by FID-GC with a packed column of Silicon OV-17
(2 m) at temperatures controlled from 50 to 220 C at a
heating rate of 5 K min 1. The reaction products were iden-
tified by GC-MS, HP5973, equipped by Hewlett Packard.
The conversion of phenol is defined as a fraction of reacted
phenol, and the selectivity to a product is as a fraction of
the respective product to the reacted phenol. The catalytic
activity was evaluated in terms of an average conversion
of phenol for every hour. The reaction rate was calculated
at 475 C from a slope of a W/F-conversion curve, where
W and F were the catalyst weight (g) and the flow rate of
phenol (3.4 mmol h 1), respectively.
The transformation of 1-propanol was also performed in
the fixed bed flow reactor at temperatures ranging from 400
FIG. 1. Comparison of catalytic activities among CeO2–MgO at a re-
to 525 C. Prior to the reaction, a catalyst sample (0.15 g) action temperature of 475 C. a, MgO; b, CeO2(11.2 mol% )–MgO; c, CeO2.

182
SATO ET AL.
TABLE 1
maximized at the CeO2 content of 11.2 mol% (the sec-
ond row of Table 2). For the CeO2–MgO catalysts except
for the pure MgO, the selectivity to 2-n-propylphenol was
higher than 82% , and it decreased slightly with increasing
CeO2 content regardlessofthe phenolconversion level. The
ortho-selectivity to the total of 2-n-propylphenol, o-cresol,
and 2-ethylphenol exceeded 97% regardless of CeO2 con-
tent while small amounts of 2,6-di-n-propylphenol and 2,6-
xylenol were produced. For the pure MgO, however, the
conversion of phenol was very low, and the selectivity to 2-
propylphenol was low because of the production of 4-n-
propylphenol. Because other products such as propene,
propanal, and 3-pentanone were observed during the alky-
lation, both the conversion of 1-propanol and the prod-
uct distribution based on 1-propanol were also calculated
(Table 2). The 1-propanol conversion showed a maximum
at a CeO2 content of 72.1 mol% , in contrast to the phe-
nol conversion showing a maximum at a CeO2 content of
11.2 mol% . For the consumption of 1-propanol, the se-
lectivity to the alkylation decreased with increasing CeO2
content.
Effect of Reaction Temperaturea
CeO2(11.2 mol% )–MgO
Aluminab
Catalyst:
Reaction temperature ( C): 425 450 475 500 500c 300 300c
Conversion of phenol (% )
0.2 2.9 22.8 49.8
4.2 27.9 55.3
Selectivity (mol% )
o-Cresol
2.2 8.7 12.5 13.3 28.8 18.5
0
2-Ethylphenol
1.2 0.9 1.8 2.5 66.8
96.6 90.4 84.8 71.6
0
42.0
1.3
0
2-n-Propylphenol
2-Isopropylphenol
n-Propyl phenyl ether
2,6-Di-n-propylphenol
4-n-Propylphenol
Others
0
0
0
0
0
0
0
0
0
0
0
0.1 0.6
0.3 0.1
0.5 2.8
4.4 7.4 82.0
0
0
0
0.1
19.5
5.6
0
0
0
0
0
7.6
1.5d 1.0 6.9e 16.7f
a Average conversion between 2 and 5 h, catalyst 0.15 g, phenol
flow rate 3.4 mmol h 1, 1-propanol/phenol = 4, carrier helium flow rate
50 mmol h 1, selectivity was based on phenol.
b Commercial alumina.
c
2-Propanol was used as a alkylating reagent instead of 1-propanol,
2-propanol/phenol = 4.
d Contained 2-isopropyl-6-n-propylphenol (0.4% ), 2,4,6-tri-n-pro-
pylphenol (0.3% ), n-propyl 2-n-propylphenyl ether (0.5% ), 2,6-xylenol
(0.3% ).
Because 1-propanol was converted into propanal and
3-pentanone during the alkylation, the transformation of
1-propanol without phenol was performed at temperatures
rangingfrom 400to 525 C. Table 3summarizesthe resultsof
MgO, CeO2, and CeO2(40.2 mol% )–MgO catalysts. Several
products, such as propanal, 3-hydroxy-2-methylpentanal, 3-
pentanone, propene, and di-n-propyl ether, were observed.
e Contained 2-isopropyl-6-n-propylphenol (4.5% ), 2,4,6-tri-n-propyl-
phenol (2.3% ), n-propyl 2-n-propylphenyl ether (0.1% ).
f Contained 4-isopropylphenol (3.8% ), 2,6-diisopropylphenol
(10.4% ), 2,4-diisopropylphenol (2.6% ).
Table 1 summarizes the effect of reaction temperature
for the propylation of phenol over CeO2(11.2 mol% )–MgO
catalyst at a temperature range between 425 and 500 C.
The data are average activities between 2 and 5 h. For the
CeO2–MgO catalyst, the average conversion of phenol in-
creased with reaction temperature, and a main product was
2-n-propylphenol. The selectivity to 2-n-propylphenol de-
creased with increasing reaction temperature, and those to
the decomposition products of 2-n-propylphenol such as
o-cresol and 2-ethylphenol increased. At temperatures
higher than 475 C, 2,6-di-n-propylphenol was produced,
while a para-alkylated product like 4-n-propylphenol was
produced at 500 C. In addition, we also examined the
propylation of phenol over a commercial alumina at 300 C.
Although the alumina catalyzed the reaction at the low
temperature, many products were formed. Furthermore,
we examined the alkylation of phenol with 2-propanol, in-
stead of 1-propanol, over the alumina and the CeO2–MgO.
Over the alumina, the conversion ofphenolwith 2-propanol
exceeded that with 1-propanol, and the main product
was 2-isopropylphenol. However, a very small amount of
2-isopropylphenol was produced over the CeO2–MgO with
the main products of o-cresol and 2-ethylphenol, in addition
to a low conversion of phenol.
TABLE 2
Catalytic Properties of CeO2–MgO Catalysts with Different CeO2
Content at 475 Ca
CeO2 content (mol% ):
0
6.7 11.2 40.2 72.1 100
Conversion of phenol (% )
2.9 12.7 22.8 12.7 11.9 12.8
Selectivity based on phenol (mol% )
o-Cresol
9.1 10.8 12.5 12.7 12.1 13.0
2-Ethylphenol
0
2.1 1.8 2.0 1.7 2.1
2-n-Propylphenol
2-Isopropylphenol
n-Propyl phenyl ether
2,6-Dipropylphenol
2,6-Xylenol
46.5 85.0 84.8 83.2 84.4 82.6
4.3 0.2 0.1 0.3 0.1 0.2
4.3 0.5 0.3 0.9 0.9 1.3
2.3 0.4 0.5
12.4
0
0
0
0
0
0
0.9 0.8 0.8
4-n-Propylphenol
21.1 1.0
0
0
0
Conversion of 1-propanol (% )
1.3 9.0 17.5 17.8 26.6 24.6
Selectivity based on 1-propanol (mol% )
Total alkylation
Propanal
3-Pentanone
Propene
Othersb
56.0 35.4 31.4 8.0 11.2 13.0
30.6 16.3 11.4 0.8 30.1 25.1
0
11.0 13.7 15.6 7.6 5.6 5.3
2.4 2.9 2.8 4.0 2.8 4.1
31.7 38.9 49.7 50.2 52.5
a Phenol flow rate 3.4 mmol h 1, 1-propanol/phenol = 4, carrier helium
flow rate 50 mmol h 1, catalyst 0.15 g, average conversion between 2 and
5 h.
Table 2 summarizes the catalytic results of several CeO2–
MgO catalysts with various CeO2 content at 475 C. In the
b Contained di-n-propylether, 3-hydroxy-2-methylpentanal, and propyl
CeO2–MgO catalysts, the average conversion of phenol was propionate, etc.

ORTHO-SELECTIVE PROPYLATION OF PHENOL OVER CeO2–MgO
183
TABLE 3
1-Propanol Transformation over CeO2–MgOa
Selectivityb (mol% )
Reaction
temperature
( C)
Conversion of
1-propanol
(% )
Catalyst
MgO
MgO
MgO
MgO
MgO
MgO
(1) (2) (3) (4) (5)
400
425
450
475
500
525
0.9
1.5
2.6
8.0
23.1
50.9
99.0
13.4 85.3
0
0
0
1.0
1.3
0
0
36.2 56.7 4.6 1.6 0.9
47.8 36.0 5.6 1.6 0.3
65.2 16.9 4.4 1.5 0.6
83.4 7.1 2.4 2.1 0.4
CeO2–MgOc
CeO2–MgOc
CeO2–MgOc
CeO2–MgOc
CeO2–MgOc
CeO2–MgOc
400
425
450
475
500
525
1.6
4.3
12.5
37.3
70.0
87.9
9.8 10.4 74.1 5.6
6.5 5.5 84.7 3.3
5.9 1.5 88.6 2.2 1.8
2.7 0.6 76.3 1.0 1.4
0
0
0.6
0.3
0
0
71.0 0.7 1.3
56.1 0.7 9.4
FIG. 2. Arrhenius plots for the data in Tables 1 and 3. a, 1-Propanol
consumption for the alkylation, calculated from the phenol conversion in
Table 1 of CeO2–MgO; b, from the 1-propanol conversion in Table 3 of
MgO; c, CeO2–MgO; d, CeO2.
CeO2
CeO2
CeO2
CeO2
CeO2
CeO2
400
425
450
475
500
525
2.1
4.5
20.8
30.4
44.4
80.1
10.9 18.2 66.9 4.0
6.5 6.0 83.6 3.9
5.4 2.0 89.1 2.0
3.4 0.4 82.9 2.0 1.3
2.1 0.1 75.1 2.1 0.8
0
0
0
it was higher than that over MgO by about one order of
magnitude (Fig. 2b). An apparent activation energy can
be calculated from a portion of straight line for the plots
(solid line). The apparent activation energy for the phe-
3.7
0
64.8 2.1 0.2
a Average conversion between 1 and 2 h at the prescribed temperature.
b (1) Propanal, (2) 3-hydroxy-2-methylpentanal, (3) 3-pentanone,
(4) propene, (5) di-n-propyl ether.
1
nol propylation over CeO2–MgO was about 410 kJ mol
c The CeO2 content is 40.2 mol% .
(Fig. 2a), while that for 1-propanol conversion was 180 kJ
mol 1 (Fig. 2c). The activation energies for 1-propnaol con-
1
version over MgO and CeO2 were 190 and 160 kJ mol
The conversion of 1-propanol increased with increasing re-
action temperature. For the pure MgO, the selectivity to
propanal increased with increasing reaction temperature.
Although the selectivity to 3-hydroxy-2-methylpentanal
decreased with increasing reaction temperature, its yield,
multiplied the selectivity by the 1-propnanol conversion,
increased from 1 to 4% . The selectivity to 3-pentanone
was very low regardless of the reaction temperature. For
CeO2–MgO and the pure CeO2, however, the selectivity
to 3-pentanone was maximized at 450 C, while the selec-
tivities to propanal and 3-hydroxy-2-methylpentanal de-
creased with increasing reaction temperature. In addition,
CO and CO2 were observed in the transformation of 1-
propanol. Regardless of the catalysts used and the reac-
tion temperature, the selectivity to propene was constantly
low, together with low selectivity to di-n-propyl ether. At
525 C, however, di-n-propyl ether selectivity increased for
the CeO2–MgO catalyst.
Figure 2 shows Arrhenius plots for the conversion data in
Tables 1 and 3. 1-Propanol consumption based on unit sur-
face area was logarithmically plotted against the inverse of
reaction temperature. For the CeO2–MgO catalyst (Figs. 2a
and 2c), the 1-propanol consumption for the propylation
was decreased about two and three orders of magnitude
over that for 1-propanol transformation. The 1-propanol
consumptions for 1-propanol transformation over CeO2
and CeO2–MgO were almost same (Figs. 2c and 2d), and
(Figs. 2b and 2d), respectively.
Table 4 shows the catalytic results of alkylation of 2-n-
propylphenol with 1-propanol, instead of the propylation of
TABLE 4
Propylation of 2-n-Propylphenola
b
Catalyst:
Reaction temperature ( C):
CeO2
500
Al2O3
300
Conversion of 2-n-propylphenol (% )
31.3
27.9
Selectivity to (mol% )
Phenol
o-Cresol
2,6-Xylenol
2-Ethylphenol
4.8
15.2
4.3
16.8
15.9
4.1
1.4
1.5
0
0
0.5
0
0.1
0
7.7
0.6
70.6
5.1
8.8
2.0
0.2
2-Isopropylphenol
4-n-Propylphenol
n-Propyl 2-n-propylphenyl ether
2,6-Dipropylphenol
2-Isopropyl-6-n-propylphenol
2,4,6-Tripropylphenol
2,3,4,6-Tetrapropylphenol
2,3,4,5,6-Pentapropylphenol
0
0
0
a
Average conversion between 1 and 2 h, catalyst 0.15 g, 2-n-propyl-
phenol flow rate 3.1 mmol h 1, 1-propanol/2-n-propylphenol = 4, carrier
1
helium flow rate 50 mmol h
b Commercial alumina.
.

184
SATO ET AL.
TABLE 5
Physical Properties of Various CeO2–MgO Catalysts
Number of total
Number of weak
Specific
CeO2 content
basic sitesa
basic sitesb
surface area
1
1
1
Catalyst
MgOc
(mol% (wt.% ))
(
mol g
)
(
mol g
)
(m2 g
)
0
1075
1151
1202
1100
471
177
119
205
154
75
96
109
118
99
48
24
CeO2–MgO
CeO2–MgOc
CeO2–MgOc
CeO2–MgOc
CeO2–MgOc
6.7 (23.5)
11.2 (35.0)
22.3 (55.1)
40.2 (74.2)
72.1 (91.2)
100
172
115
46
115
c
CeO2
44
a Total amount of CO2 desorbed from the sample in the TPD profiles.
b Amount of CO2 desorption peak at lower than 150 C.
c The data were referenced in Ref. 4.
phenol, over typical samples of CeO2 and Al2O3. Although sites are effective for the methylation (4). Thus, we calcu-
a dialkylated product, 2,6-di-n-propylphenol, was preferen- lated two sets of turnover frequency (TOF) by dividing the
tial to Al2O3 at 300 C, CeO2 was inactive for the alkylation: reaction rate for the propylation at 475 C (the third column
the decomposition of 2-n-propylphenol into o-cresol and in Table 6) by the number of total basic sites (the third col-
2-ethylphenol and the isomerization to 2-isopropylphenol umn in Table 5) as well as by that of the weak basic sites
proceeded rather than the alkylation.
(the fourth column in Table 5). The TOF values are also
In the previous work (4), we elucidated the acid–base summarized in Table 6. The reaction rate based on catalyst
properties of CeO2–MgO systems using TPD technique. weight was maximized at the CeO2 content of 11.2 mol%
The deconvolution results in the TPD experiment of ad- in a similar manner as the phenol conversion shown in the
sorbed CO2 indicate that the basic sites of the pure MgO second row of Table 2. The reaction rate per unit surface
and the CeO2–MgO consist of at least four kinds of basic area, however, had a maximum at the CeO2 content of
sites with different strengths and that CeO2 has only weak 72.1 mol% . Figure 3 exhibits the variations in the two se-
basic sites of uniform strength (Fig. 7 in Ref. 4). Table 5 ries of TOF with CeO2 content. The TOF values based on
summarizes the number of weak basic sites calculated from both the total basic sites (TOFt) and the weak basic sites
a peak whose top was at temperatures lower than 150 C in (TOFw) of the pure MgO were much smaller than those
the deconvoluted TPD profile, together with the number of of the pure CeO2. The TOFt monotonously increased with
total basic sites and the specific surface area.
increasing CeO2 content (curve a). In contrast, the TOFw
We have demonstrated that the strongbasicsitesare irrel- value greatly increased with addition of CeO2 into MgO
evant to the catalytic methylation of phenol with methanol (curve b), and it showed a maximum at CeO2 content of
at temperatures lower than 500 C and that only weak basic 72.1 mol% .
TABLE 6
Reaction Rates of Various CeO2–MgO Catalysts
Reaction ratea
CeO2 content
(mol% )
TOFtb
TOFwc
1
2
1
1
Catalyst
MgO
CeO2–MgO
CeO2–MgO
CeO2–MgO
CeO2–MgO
CeO2–MgO
CeO2
mmol h 1 g
mmol h 1 m
(h
)
(h
)
0
6.7
0.3
3.4
5.7
4.1
2.5
2.3
2.7
0.003
0.031
0.048
0.041
0.052
0.096
0.061
0.3
1.7
29
3.0
4.7
3.7
5.3
13
23
11.2
22.3
40.2
72.1
100
28
27
33
50
23
a Reaction rate of the propylation at 475 C, 1-propanol/phenol = 4; the rate was calculated from a slope
of a W/F-conversion curve, where W and F were the catalyst weight (g) and the flow rate of phenol
(3.4 mmol h 1), respectively, W = 0.075, 0.10, and 0.15 g.
b Turnover frequency per a basic site.
c Turnover frequency per a weak basic site.

ORTHO-SELECTIVE PROPYLATION OF PHENOL OVER CeO2–MgO
185
2-methylpentanal is considered to be deformylized into
3-pentanone. Therefore, three reactions, dehydrogenation,
aldol addition, and deformylation, successively proceed
over the CeO2-containing catalysts to produce propanal,
3-hydroxy-2-methylpentanal, and 3-pentanone, respec-
tively. The reaction pathway can be proposed as Scheme 1.
The scheme can be also supported by the following: a
phenyl alkyl ketone such as 1-phenyl-2-propanone was
produced by the decomposition of 3-hydroxy-3-phenyl-2-
methylpropanal via aldol addition of benzaldehyde with
propanal over thoria catalyst at 450 C (27).
Although 1-propanol is dehydrogenated into propanal
both on the strong basic MgO and on the weak basic CeO2
(Table 3), 3-pentanone is formed on CeO2 not on MgO.
In the decomposition of methanol over the CeO2–MgO,
methanol was decomposed into CO, CO2, and methane
without forming dimethyl ether, whereas MgO had no ac-
tivity in the decomposition (4). In addition, CeO2 itself is re-
ported to have the ability of the hydrogenation of CO (28).
The deformylation of 3-hydroxy-2-methylpentanal into 3-
pentanone is possibly concerned with the redox property of
CeO2. Because aldol condensation is usually catalyzed by
strong base (29, 30), strong basic MgO should catalyze the
aldol addition of propanal. However, weak basic CeO2 was
more efficient for the formation of 3-pentanone (Table 3).
In the dimerization process of aldol addition, redox prop-
erty of CeO2 is considered to be attributed to the activation
of propanol in addition to the basic property. This will be
discussed in connection with the nature of active species.
FIG. 3. Variation in the TOF value with CeO2 content. a, TOF per a
basic site (TOFt); b, TOF per a weak basic site (TOFw).
DISCUSSION
3-Pentanone Formation
We have reported on the methylation of phenol with
methanol over the CeO2–MgO systems (4). The active
CeO2 species have been proved to consist of the intersti-
tial solid solution of fluorite-type MgxCe1 x/2O2 dispersed
in the MgO matrix. In addition to the methylation of
phenol, the CeO2–MgO systems have a catalytic activ-
ity for the methanol decomposition to CO and methane.
As shown in Table 2, propanal and 3-pentanone were
produced during the propylation of phenol. The conver-
sion of 1-propanol increased with increasing CeO2 con-
tent, and the selectivity to the alkylation decreased with
increasing CeO2 content. From the facts, it is speculated
that the CeO2 component contributes to the activation
of 1-propanol. In the transformation of 1-propanol with-
out phenol (Table 3), for the pure MgO, the yields of
both propanal and 3-hydroxy-2-methylpentanal increased
with increasing 1-propanol conversion. Thus, 1-propanol is
dehydrogenated into propanal at the first step of the reac-
tion, and 3-hydroxy-2-methylpentanal is produced via al-
dol addition of propanal on the basic sites of MgO. How-
ever, 3-pentanone was not produced over the MgO. On the
contrary, over the CeO2-containing catalysts, the yield of
3-pentanone increased with increasing the 1-propanol con-
version, while the selectivities to propanal and 3-hydroxy-2-
methylpentanal decreased. Thus, the produced 3-hydroxy-
Phenol Propylation Related with Basic Strength
In our previous report on the basicity of the CeO2–
MgO catalysts (4), the basic sites of the pure MgO and
the CeO2–MgO consist of at least four kinds of basic sites
with different strengths, whereas the pure CeO2 has only
weak basic sites of uniform strength. The other TPD re-
sults also indicated that acidic substrates such as phenol
and o-cresol adsorbed strongly on the surface of the pure
MgO, while they readily desorbed from the basic sites of
CeO2–MgO and CeO2 at temperatures higher than 450 C.
For the methylation of phenol with methanol, when we cal-
culated the TOFt value by dividing the reaction rate by the
number of total basic sites, the TOFt value monotonously
increased with increasing CeO2 content. This has indicated
that total basic sites including strong basic sites are irrel-
evant to the methylation of phenol at temperatures lower
than 500 C, because acidic phenol and products cannot be
SCHEME 1. Reaction pathway of 3-pentanone formation.

186
SATO ET AL.
desorbed from the strong basic sites. In the present propy- an alkylating reagent (Table 1). This indicates that n-propyl
lation, the TOFt monotonously increased with increasing cation produced on the acidicalumina can readilytransform
CeO2 content (Fig. 3a) in a similar manner as the methyla- to isopropyl cations, because secondary cation is more sta-
tion. Namely, the total basic sites are also indicated to be ble. As has been suggested by Klemm et al. (23) and Velu
irrelevant to the propylation of phenol.
et al. (25), propyl cations are said to be alkylating species
In the CeO2–MgO system, the variation in the number of of phenol on the acidic catalyst.
weak basic sites with CeO2 content resembles the variation
On the other hand, no acidic site is concluded to be on
in the reaction rate for the methylation of phenol at 450 C the CeO2–MgO catalysts because neither dimethyl ether
(4), and the TOFw values for the methylation, calculated by formation in the methylation nor ammonia adsorption in
dividing the reaction rate by the number of weak basic sites, the TPD experiment is observed (4). Although propene
are constant at CeO2 content higher than 3.8 mol% (Fig. 8b and di-n-propyl ether are produced over the CeO2–MgO
in Ref. 4). This means that the methylation proceeds on the catalyst during both the propylation and the transformation
weak basic sites which act as adsorption sites of phenol. In of 1-propanol, the dehydrogenation into propanal is domi-
contrast to the methylation, the TOFw value for the present nant (Tables 2 and 3). The facts do not mean that n-propyl
propylation seems to be maximized at CeO2 content of cation can be produced on the nonacidic CeO2–MgO
72.1 mol% . Except for the CeO2 content of 72.1 mol% and at temperatures lower than 500 C. In addition, neither
the pure MgO, the TOFw values can be regarded as roughly 2-n-propylphenol nor 2-isopropylphenol is produced over
constant (Fig. 3b). Thismeansthat the propylation proceeds the CeO2–MgO catalyst in the alkylation of phenol with
on the weak basic sites which provide adsorption sites of 2-propanol (the sixth column of Table 1). The fact also
phenol. The reason why the TOFw is maximized at CeO2 suggests that isopropyl cation cannot be produced on the
content of 72.1 mol% will be discussed in the last section.
One of the catalytic features of the CeO2–MgO catalyst significant difference from the behavior of acidic alumina.
is the high selectivity of ortho-position of phenol. It is re- Both the conversion of phenol (Table 2) and the reaction
CeO2–MgO at temperatures lower than 500 C. This is a
ported that the ortho-position of the phenol perpendicu- rate (Table 6) are maximized at CeO2 content of 11.2 mol% .
larly adsorbed on a basic site faces the surface of MgO However, only a third of the consumed 1-propanol is used
on which the ortho-selective methylation proceeds (31). for the alkylation at the CeO2 content. On the other hand,
The CeO2–MgO catalyst also provides effective adsorption the 1-propanol conversion in the propylation is highest at
sites for phenol: the ortho-position of the phenol perpen- CeO2 content of 72.1 mol% (Table 2), while a fraction of
dicularly adsorbed on the weak basic site faces the surface the 1-propanol consumed for the propylation is at most
of the CeO2–MgO catalyst. Although dimethylated phe- 11% . Thus, the propylation of phenol and the formation
nol such as 2,6-xylenol was produced over the CeO2–MgO of 3-pentanone can be said to be catalyzed by different
system in the methylation of phenol (3, 4), the total selectiv- functions.
ity to monoalkylated phenols was higher than 97% in the
In the transformation of 1-propanol without phenol
propylation (Table 2). In addition, 2-n-propylphenol was (Table 3), propanal is initially produced by the dehydro-
not alkylated with 1-propanol over the pure CeO2, while the genation of 1-propanol. In addition to the 1-propanol con-
decomposition and the isomerization proceeded (Table 4). sumption over MgO being obviously lower than those
2-n-Propylphenol is probably isomerized to 2-isopropyl- on the CeO2 containing catalysts (Figs. 2b, 2c, and 2d),
phenol followed by decomposition to o-cresol and 2-ethyl- the apparent activation energies for 1-propanol conver-
phenol, since o-cresol and 2-ethylphenol are produced sion were slightly decreased with increasing CeO2 content.
through the decomposition of 2-isopropylphenol in the Over CeO2, aldol addition and the following deformyla-
alkylation of phenol with 2-propanol (the sixth column of tion are assumed to be faster than the dehydrogenation of
Table 1). The n-propyl group of 2-n-propylphenol perpen- 1-propanol. In, contrast, the deformylation was not cata-
dicularly adsorbed on the catalyst is so large that the other lyzed by MgO even at higher temperatures, although an
ortho-position of phenol is probably blocked by itself from aldol product, 3-hydroxy-2-methylpentanal, was formed at
another 1-propanol molecule. Thus, 2,6-di-n-propylphenol higher temperatures (Table 3). It can be said that the calcu-
is not formed on the CeO2–MgO catalyst surface.
lated activation energiesof160–190kJ mol 1 for 1-propanol
conversion are for the dehydrogenation of 1-propanol to
propanal. Although the activation energy for the propyla-
tion is as high as 410 kJ mol 1 (Fig. 2a), we cannot discuss
Nature of Active Species
Acidic alumina is reported to be active for the alkyla- about it any further at the present time.
tion of phenol with propanol (23, 24), while more than
The weak basic CeO2 is more effective for the dehy-
15 alkylphenols are produced at high phenol conversions drogenation than the strong basic MgO (Figs. 2b and 2d).
of nearly 100% . Over a commercial alumina, 2-isopropyl- Because the density of weak basic sites of the pure MgO is
phenol is produced at 300 C even if 1-propanol is used as similar to those measured in the CeO2–MgO samples (4),

ORTHO-SELECTIVE PROPYLATION OF PHENOL OVER CeO2–MgO
187
basic sites of the catalysts are not assumed to be related to 3-pentanone to release CO through a catalysis of CeO2 with
the dehydrogenation of alcohol. It has been reported that Ce4+–Ce3+ redox character because of no activity of MgO.
only CeO2 is effective for the methylation of phenol among
We assume that the alkylation mechanism over the CeO2-
various rare earth oxides (3). Although Pr and Tb also have containing catalysts is different from that of acidic catalysts
the oxidation state of 4+, PrO2 and TbO2 are unstable to in a viewpoint of the activation of alcohol. It can be said
produce partially reduced oxides as Pr6O11 and Tb4O7, re- that bifunctional catalysis, such as adsorption of phenol on
spectively (32). They are inactive for the methylation of weak basic sites and homolytic activation of 1-propanol by
phenol (3). Only Ce in the rare earth metals provide stable the redox property, efficiently work in the propylation over
oxide as CeO2 with the highest oxidation state of 4+ (32). the CeO2–MgO systems. In the methylation of phenol, we
The dehydrogenation is probably attributed to the hydro- have speculated that methanol is activated on the CeO2
genation of CO (28). Thus, CeO2 with the redox property species as a form of formyl or hydroxymethyl group rather
is said to be effective for the activation of 1-propanol in the than methyl cation on acidic sites (4). From the present dis-
present CeO2–MgO system.
cussion, a hydroxymethyl radical rather than formyl group
The reaction mechanism can be speculated as Scheme 2. can be also proposed as an active methylating species in the
A hydrogen radical is probably eliminated from 1-propanol methylation of phenol.
to produce radical species such as 1-hydroxypropyl radical
rather than propoxy radical. The hydrogen radical can re-
Different Catalytic Activities in the System
duce Ce4+ into Ce3+ to convert proton. In other words,
Ce4+ cation oxdatively eliminates proton from 1-propanol.
The TOFw value for the propylation is exceptionally
The 1-hydroxypropyl radical possibly attacks the electron- high at a CeO2 content of 72.1 mol% , as mentioned above,
rich ortho-positions of phenol perpendicularly adsorbed on and the TOFw value of the pure CeO2 is somewhat less
the weak basic sites in the CeO2–MgO system because the than that for the other CeO2–MgO catalysts (Table 6 and
alpha-position of 1-hydroxypropyl radical is slightly posi- Fig. 3b). The structure of the CeO2–MgO has been clarified
tive. A hydroxy radical is readily eliminated from the alky- by XRD measurement: the CeO2–MgO catalyst is a mixture
lated intermediate to produce 2-n-propylphenol and water, of MgO and an interstitial solid solution of MgxCe1 x/2O2,
together with the oxidation of Ce3+ to Ce4+
.
and the CeO2(72.1 mol% )–MgO consists of only the solid
The reaction mechanism of 3-pentanone formation also solution of Mg0.4Ce0.8O2 without MgO phase (4). For the
can be speculated (Scheme 2). When phenol is not adsorbed CeO2(72.1 mol% )–MgO, the dehydrogenation is more
on the neighboring sites of the activated 1-hydroxypropyl preferential than the alkylation (Table 2). This indicates
radical, it is dehydrogenated into propanal to release that the solid solution of Mg0.4Ce0.8O2 has both high de-
hydrogen. If the produced propanal is activated by ba- hydrogenating power of 1-propanol and high alkylation
sic sites to produce enolate anion, the enolate anion at- potential of the 1-hydroxypropyl radical. We can provide
tacks another propanal or 1-hydroxypropyl radical to pro- another explanation: the basic strength of optimum adsorp-
duce 3-hydroxy-2-methylpentanal as an aldol product. The tion sites must be dependent on the reaction temperature.
enolate anion possibly attacks the alpha-position of an- Although the TOFw values for the methylation at 450 C are
other 1-hydroxypropyl radical, since MgO without redox constant at CeO2 content higher than 3.8 mol% , basic sites
property is not efficient for aldol addition. 3-Hydroxy- with medium strengths of the solidsolution of Mg0.4Ce0.8O2,
2-methylpentanal can be probably decomposed into stronger than the weak basic sites of the pure CeO2, may act
SCHEME 2. Probable reaction mechanism of phenol propylation and 3-pentanone formation over the CeO2–MgO system.

188
SATO ET AL.
as adsorption sites of phenol at temperatures higher than ortho-position of phenol adsorbed perpendicularly on weak
475 C. basic sites on the catalyst is selectively alkylated by 1-pro-
Variations in the catalytic behavior with process time panol, which is possibly activated in the form of 1-hydro-
over the CeO2–MgO system are different between the xypropyl radical rather than propyl cation, as illustrated in
methylation (Fig. 1 in Ref. 4) and the present propylation Scheme 2.
(Fig. 1). The catalytic activity for the methylation shows a
gradual increase in the initial reaction period for 2 h, in
contrast to the propylation in which the catalytic activities
ACKNOWLEDGMENTS
decrease with process time in the initial period for 2 h. They
reach a stable conversion level after the reactions have been
running for 2 h: the activation of methanol probably has
an induction period. On the other hand, because a part of
the basic sites are used for aldol addition in the propyla-
tion, adsorption sites for phenol are reduced. Actually, an
induction period of 2 h was observed in the 1-propanol con-
version over the pure CeO2 (data not shown), although the
conversion was retarded over the pure MgO in the initial
period. The effective sites for the propanal formation and
the methanol activation are suggested to be attributed to an
optimum redox state of CeO2, which is possibly produced
by reduction with 1-propanol and methanol in the initial re-
action period. It is our next interest to clarify the optimum
redox state of CeO2 catalysts.
We thank Mr. Masahiko Iijima, Tonen Co. (Japan), for measuring
GC-MS.
REFERENCES
1. Fukuda, Y., Nishizaki, T., and Tanabe, K., Nihon Kagaku Kaishi, 1754
(1972).
2. Tanabe, K., Hattori, H., Sumiyoshi, T., Tamaru, K., and Kondo, T.,
J. Catal. 53, 1 (1978).
3. Sato, S., Koizumi, K., and Nozaki, F., Appl. Catal. A 133, L7 (1995).
4. Sato, S., Koizumi, K., and Nozaki, F., J. Catal. 178, 264 (1998).
5. Bezouhanova, C., and Al-Zihari, M. A., Appl. Catal. 83, 45 (1992).
6. Pierantozzi, R., and Nordquist, A. F., Appl. Catal. 21, 263 (1986).
7. Belitrame, P., Belitrame, P. L., Carniti, P., Castelli, A., and Forni, L.,
Appl. Catal. 29, 327 (1987).
8. Santacesaria, E., Grasso, D., Gelosa, D., and Carra, S., Appl. Catal.
64, 83 (1990).
9. Santacesaria, E., Diserio, M., Ciambelli, P., Gelosa, D., and Carra, S.,
Appl. Catal. 64, 101 (1990).
10. Balsama, S., Beltrame, P., Beltrame, P. L., Carniti, P., Forni, L., and
Zuretti, G., Appl. Catal. 13, 161 (1984).
11. Kotanigawa, T., Yamamoto, M., Shimokawa, K., and Yoshida, Y., Bull.
Chem. Soc. Jpn. 44, 1961 (1971).
CONCLUSION
Vapor-phase alkylation of phenol with 1-propanol was
investigated over CeO2–MgO catalysts prepared utilizing
a molten mixture of the corresponding nitrates and cit-
ric acid. The CeO2–MgO had a stable catalytic activity at
12. Kotanigawa, T., and Shimokawa, K., Bull. Chem. Soc. Jpn. 47, 950
(1974).
475 C and had an excellent selectivity to 2-n-propylphenol 13. Kotanigawa, T., and Shimokawa, K., Bull. Chem. Soc. Jpn. 47, 1535
(1974).
higher than 82% based on phenol. Although a portion of
2-propylphenol was decomposed, a sum of selectivities to
monoalkylated phenols such as 2-propylphenol, o-cresol,
14. Grabowska, H., Kaczmarczyk, W., and Wrzyszcz, J., Appl. Catal. 47,
351 (1989).
15. Misono, M., and Nojiri, N., Appl. Catal. 64, 1 (1990).
and 2-ethylphenol exceeded 97% . The highest catalytic ac-
tivities for the propylation, high reaction rate, and high se-
lectivity to 2-n-propylphenol were obtained at CeO2 con-
tent of 11.2 mol% .
During the propylation, 1-propanol was converted into
propanal and 3-pentanone. The dehydrogenation was max-
16. Narayanan, S., Venkatrao, V., and Durgakumari, V., J. Mol. Catal. 52,
L29 (1989).
17. Venkatrao, V., Durgakumari, V., and Narayanan, S., Appl. Catal. 49,
165 (1989).
18. Venkatrao, V., Chary, K. V. R., Durgakumari, V., and Narayanan, S.,
Appl. Catal. 61, 89 (1990).
19. Velu, S., and Swamy, C. S., Appl. Catal. A 119, 241 (1994).
imized at CeO2 content of 72.1 mol% . In the results of the 20. Velu, S., and Swamy, C. S., Appl. Catal. A 145, 141 (1996).
21. Velu, S., and Swamy, C. S., Appl. Catal. A 145, 225 (1996).
22. Karuppannasamy, S., Narayanan, K., and Pillai, C. N., J. Catal. 66, 281
transformation of 1-propanol, it was found that 1-propanol
was dehydrogenated into propanal, and that the produced
propanal was converted to 3-hydroxy-2-methylpentanal via
aldol addition, followed by the production of 3-pentanone
(1980).
23. Klemm, L. H., and Taylor, D. R., J. Org. Chem. 45, 4320 (1980).
24. Klemm, L. H., and Taylor, D. R., J. Org. Chem. 45, 4326 (1980).
through the deformylation of 3-hydroxy-2-methylpentanal. 25. Velu, S., and Swamy, C. S., Catal. Lett. 40, 265 (1996).
26. Marcilly, C., Courty, P., and Delmon, B., J. Am. Ceram. Soc. 53, 56
The pure CeO2, which had only weak basic sites to-
gether with the redox property, selectively catalyzed the
1-propanol transformation to 3-pentanone, while the pure
MgO with strong basicity showed a very low reaction rate in
(1970).
27. Ananthan, S., Venkatsubramanian, N., and Pillai, C. N., J. Catal. 89,
489 (1984).
28. Li, C., Domen, K., Maruya, K., and Onishi, T., J. Catal. 141, 540 (1993).
the alkylation and inefficient formation of 3-pentanone in 29. Dabagh, G., and Davis, B., J. Mol. Catal. 48, 117 (1988).
30. Zhang, G., Hattori, H., and Tanabe, K., Appl. Catal. 36, 189 (1988).
31. Tanabe, K., and Nishizaki, T., “Proceedings, 6th International
Congress Catalysis, London, 1976” (G. C. Bord, P. B. Wells, and F. C.
Tompkins, Eds.). The Chemical Society, London, 1977.
32. Moeller, T., in “Comprehensive Inorganic Chemistry” (A. F. Trotman-
the transformation.
The redox property of Ce4+–Ce3+ is probably attributed
with the activation of 1-propanol to produce 1-hydro-
xypropyl radical. We speculate the reaction mechanism of
the ortho-propylation over the CeO2–MgO catalyst: the
Dickenson, Ed.), Vol. 4, p. 99. Pergamon, Oxford, 1973.