Liquid-phase oxidation of 2,6-di-tert-butylphenol with Cu-impregnated MCM-41 catalysts in the presence of alkali metals

Article abstract of DOI:10.1006/jcat.1999.2667

Mesoporous silicate MCM-41, with a uniform pore diameter of ca. 35 A, was used as a support for impregnated Cu catalysts for liquid-phase oxidation of 2,6-di-tert-butylphenol (BOH) in the presence of a base such as KOH. The oxidation products were 4,4′-dihydroxy-3,3′,5,5′-tetra-tert-butyl-biphenyl (H₂DPQ) and 3,3′,5,5′-tetra-tert-butyl-4,4′-diphenoquinone (DPQ). An alkali such as a potassium salt, which promotes the phenol oxidation activity of copper ion-impregnated MCM-41 (Cu/MCM-41) catalyst, was found to be more effective as an additive to the reaction solution than impregnation on the Cu/MCM-41 catalyst. The added alkali was found to play a role in generating the corresponding phenolate anion by dissolving a BOH molecule. H₂DPQ is formed through tautomerization of an intermediate dimer obtained by the C-C coupling of the corresponding phenoxy radicals. DPQ is formed via the consecutive oxidation of H₂DPQ and/or via the oxidative dehydrogenation of the intermediate dimer. The BOH molecules in the mesopores of Cu/MCM-41, rather than the BOH in bulk solution in the presence of CuCl₂, were found to favor H₂DPQ production. Cu/MCM-41 with added potassium [(K-Cu)/MCM-41] was much more active for phenol oxidation than Cu-impregnated NaZSM-5 with added potassium [(K-Cu)/NaZSM-5] or the corresponding NaY [(K-Cu)/NaY], each catalyst having only uniform micropores. This result indicates that the oxidation of sterically bulky BOH occurs mainly at the active sites in the mesopores and is difficult to carry out in the micropores of (K-Cu)/NaZSM-5 and (K-Cu)/NaY zeolites because of the steric bulkiness of the oxidation products, H₂DPQ and DPQ. The liquid-phase adsorption amounts of BOH on NaZSM-5 and NaY were found to be comparable to that on MCM-41. The shape selectivity by the oxidation products in the micropores of (K-Cu)/NaZSM-5 and (K-Cu)/NaY zeolites was thus suggested to inhibit the BOH oxidation activities of both catalysts, based on the results of the oxidation reaction and liquid-phase adsorption of BOH.

Full text of DOI:10.1006/jcat.1999.2667

Journal of Catalysis 188, 417–425 (1999)
Article ID jcat.1999.2667, available online at http://www.idealibrary.com on
Liquid-Phase Oxidation of 2,6-Di-tert-butylphenol with Cu-Impregnated
MCM-41 Catalysts in the Presence of Alkali Metals
�
�
�
Hirofumi Fujiyama, Ichitaro Kohara, Keisuke Iwai,† Satoru Nishiyama,
�
,1
Shigeru Tsuruya, and Mitsuo Masai‡
�
Department of Chemical Science and Engineering, Faculty of Engineering, and †Division of Molecular Science, Graduate School of Science
and Technology, Kobe University, Nada, Kobe, 657-8501, Japan; and ‡Department of Applied Physics and Chemistry,
The Fukui University of Technology, Gakuen, 3-6-1, Fukui, 910 Japan
Received July 30, 1999; accepted August 18, 1999
INTRODUCTION
Mesoporous silicate MCM-41, with a uniform pore diameter of
ca. 35 A˚ , was used as a support for impregnated Cu catalysts for
Since Mobil’s report (1, 2) on the syntheses of meso-
liquid-phase oxidation of 2,6-di-tert-butylphenol (BOH) in the pres- porous silicates M41S, which is classified as MCM-41
0
0
ence of a base such as KOH. The oxidation products were 4,4 - (hexagonal prism structure), MCM-48 (cubic structure),
0
0
0
dihydroxy-3,3 ,5,5 -tetra-tert-butyl-biphenyl (H2DPQ) and 3,3 ,5,5 -
and MCM-51 (sheet structure), many studies on the syn-
thesis and characterization of the mesopore silicates and
their counterparts with transition metals inserted into the
tetrahedral framework of silica have been reported (3–8).
In particular, MCM-41, which possesses one-dimensional
uniform mesopores, the size of which can be controlled
by selecting the template molecules, has been noted as a
provider of a mesopore reaction field (9–14). The applica-
0
tetra-tert-butyl-4,4 -diphenoquinone (DPQ). An alkali such as a
potassium salt, which promotes the phenol oxidation activity of cop-
per ion-impregnated MCM-41 (Cu/MCM-41) catalyst, was found
to be more effective as an additive to the reaction solution than
impregnation on the Cu/MCM-41 catalyst. The added alkali was
found to play a role in generating the corresponding phenolate an-
ion by dissolving a BOH molecule. H2DPQ is formed through tau-
tomerization of an intermediate dimer obtained by the C–C cou-
pling of the corresponding phenoxy radicals. DPQ is formed via tion of these mesopore materials to catalyst supports with a
the consecutive oxidation of H2DPQ and/or via the oxidative dehy- mesosize reaction field will be expected in liquid-phase fine
drogenation of the intermediate dimer. The BOH molecules in the chemical reactions including sterically bulky molecules.
mesopores of Cu/MCM-41, rather than the BOH in bulk solution
in the presence of CuCl2, were found to favor H2DPQ production.
Cu/MCM-41 with added potassium [(K–Cu)/MCM-41] was much
more active for phenol oxidation than Cu-impregnated NaZSM-
We have been studying liquid-phase oxidation of phenol
and/or catechol derivatives catalyzed by homogeneous cop-
per complexes and heterogeneous copper-supported poly-
mers (15). Through the study of the design and develop-
ment of the copper catalyst for the oxidation of phenol
derivatives, basic copper catalytic systems such as CuCl2–
KOH have been found (16, 17) to be effective for liquid-
phase oxidation of phenol derivatives including phenols
5
with added potassium [(K–Cu)/NaZSM-5] or the corresponding
NaY [(K–Cu)/NaY], each catalyst having only uniform micropores.
This result indicates that the oxidation of sterically bulky BOH oc-
curs mainly at the active sites in the mesopores and is difficult to
carry out in the micropores of (K–Cu)/NaZSM-5 and (K–Cu)/NaY
zeolites because of the steric bulkiness of the oxidation products, with sterically bulky substituent groups such as 2,6-di-tert-
H2DPQ and DPQ. The liquid-phase adsorption amounts of BOH on butylphenol using oxygen molecules as an oxidant under
NaZSM-5 and NaY were found to be comparable to that on MCM- mild reaction conditions. The liquid-phase oxidation of
4
1. Theshapeselectivitybytheoxidation productsin themicropores
3
,5-di-tert-butylphenol catalyzed by Ti- or V-incorporated
of (K–Cu)/NaZSM-5 and (K–Cu)/NaY zeolites was thus suggested
to inhibit the BOH oxidation activities of both catalysts, based on
the results of the oxidation reaction and liquid-phase adsorption of
mesoporous silicates has been reported (4, 13) using H2O2
or tert-butylhydroperoxide as an oxidant under various re-
action conditions. We have already reported (14) the liquid-
phase oxidation of benzene catalyzed by Cu-supported
mesoporous silicates and their aluminosilicate counterparts
using oxygen molecules and ascorbic acid as an oxidant and
a reductant, respectively. Phenol has been produced by the
direct oxygenation of benzene though the yield of produced
phenol was rather low.
BOH.
�c 1999 Academic Press
Key Words: 2,6-di-tert-butylphenol; oxidation; Cu/MCM-41;
Cu/NaZSM-5; Cu/NaY; diphenoquinone; dihydroxylbiphenyl; al-
kali metals; shape selectivity.
1
In this study, we report the liquid-phase oxidation of
the titled phenol catalyzed by Cu-supported mesoporous
To whom correspondence should be addressed. Fax: +81–78–803–
6
171. E-mail: tsuruya@cx.kobe-u.ac.jp.
417
0021-9517/99 $30.00
Copyright �c 1999 by Academic Press
All rights of reproduction in any form reserved.

4
18
FUJIYAMA ET AL.
silicates in the presence of alkali using oxygen molecules as Liquid-Phase Adsorption of 2,6-Di-tert-butylphenol
an oxidant. The usefulness of the mesopore field in the reac-
tion including the sterically bulky molecule is emphasized
on the basis of the comparison of the oxidation activities
of the Cu catalysts having mesopores and micropores. The
high selectivity for dihydroxydiphenyl (H2DPQ) derivative
in the mesopore reaction field is shown in the comparison
with a bulk solution as a reaction field. The role of added
alkali on phenol oxidation is also discussed in terms of the
reaction scheme of the phenol oxidation.
A 0.2 g portion of support (MCM-41, NaZSM-5, NaY)
was added to 20 cm of chloroform solution containing
mmol of dissolved 2,6-di-tert-butylphenol (0.206 g). Af-
ter magnetically stirring the whole system at 313 K for 24 h,
the adsorbed amount of phenol was determined by measur-
ing the amount of phenol remaining in the solution using
HPLC (Hitachi L-4200).
3
1
Liquid-Phase Oxidation of 2,6-Di-tert-butylphenol
Oxidation of 2,6-di-tert-butylphenol (0.103 g, 0.5 mmol)
3
EXPERIMENTAL
was performed in 12 cm of chloroform solvent including
0
.2 g catalyst at 313 K under air atmosphere for 24 h us-
Synthesis of MCM-41
ing a magnetic stirrer. When an alkali salt was added to the
3
reaction solution, a mixed solvent consisting of 10 cm chlo-
MCM-41 was synthesized by a sol–gel method at room
temperature in the atmosphere: tetraethyl orthosilicate
3
roform and 2 cm methanol containing the homogeneously
dissolved alkali salt was used, instead of pure chloroform
solvent. To minimize the effect of the reactant and the prod-
ucts adsorbed on the catalyst, the reaction system, after the
reaction, was centrifuged and immediately transferred to
the HPLC analyses. The products were analyzed by HPLC
[Nacarai Tesque, extra pure, 0.1 mol (20.83 g)] was added
dropwise for 10 min with stirring to an aqueous ammonia
solution (2.56% , 234 cm ) containing dissolved hexadode-
3
cyltrimethylammonium bromide [Nacarai Tesque, guaran-
teed reagent, 0.012 mol (4.37 g)] as a template. The en-
tire mixture was kept at room temperature for 24 h. The
white precipitate produced was filtered off and washed
under vacuum with warm (333–343 K) ion-exchanged
water until the filtrate became neutral. The precipitate
was dried at 393 K overnight, treated at 773 K for 1 h
in a nitrogen flow (250 cm /min), and calcined at 773 K
in an air flow (250 cm /min) to remove the template
moieties.
(
Hitachi L-4200) with a GC-C18 packed column using ace-
3
tonitrile (degassed) as a carrier (1.0 cm /min). The amount
of leached Cu or K during the oxidation was determined
by measuring the amount of Cu or K in the catalyst, which
was separated from the reaction solution by centrifugation
followed by immediate filtration, before and after the oxi-
dation using an atomic absorption spectrometer (Shimazu
AA-630-01).
3
3
0
0
0
Synthesis of 3,3 ,5,5 -Tetra-tert-butyl-4,4 -diphenoquinone
Preparation of Cu-Impregnated and Cu–K-
Coimpregnated Catalysts
0
0
0
3
,3 ,5,5 -Tetra-tert-butyl-4,4 -diphenoquinone was syn-
thesized using a homogeneous CuCl2–pyridine–KOH cata-
lytic system according to the literature (16). The re-
sultant reddish solid was recrystallized from methanol
containing a small amount of chloroform [IR: strong C == O
MCM-41(synthesized in our laboratory), NaZSM-5[syn-
thesized in our laboratory (18), Si/Al = 53], and NaY (Toso,
Si/Al = 2.8) were used as supports. Cu-impregnated cata-
lysts were prepared by impregnation with the prescribed
amount of copper acetate dissolved in ethanol. Cu–K-
coimpregnated catalysts were prepared using the ethanol
solution with both dissolved copper acetate and potassium
acetate. Cu-impregnated and Cu–K-coimpregnated cata-
lysts were dried at 393 K overnight and calcined at 773 K
for 5 h in an air flow. The standard impregnated amount of
Cu was 1 wt% .
� 1
stretching peak at 1702 cm (16); UV-VR: �max = 430 nm
(
2)].
0
0
0
Synthesis of 4,4 -Dihydroxy-3,3 ,5,5 -tetra-
tert-butylbiphenyl
0
0
0
4,4 -Dihydroxy-3,3 ,5,5 -tetra-tert-butylbiphenyl was syn-
0
0
thesized by catalytically reducing 3,3 ,5,5 -tetra-tert-butyl-
,4 -diphenoquinone. A 0.1-g sample (0.25 mmol) of the
0
4
corresponding diphenoquinone, 0.1 g of 3 wt% reduced
Ru-impregnated MCM-41 catalyst, and 5 cm of chloro-
Preparation of Reduced Ru-Impregnated MCM-41
3
MCM-41 was poured into an ethanol solution containing form solvent were added to a high-pressure stainless-steel
3
RuCl3 � 3H2O. After ethanolwasevaporated to drynesswith reactor (Taiatsu Glass Co., 30 cm ). After several hydro-
stirring, the resulting Ru-impregnated MCM-41 was dried gen purges in the reactor, the reaction was conducted at
at 393 K overnight, calcined at 673 K for 5 h, and reduced at 338 K for 24 h under 1.0 MPa hydrogen atmosphere. The
7
73K for 5h in a hydrogen flow. The amount ofimpregnated resultant off-white solid product was analyzed by HPLC
Ru was 3 wt% .
and obtained in almost 100% yield after centrifugation of

OXIDATION OF PHENOLS WITH Cu/MCM-41 CATALYSTS
419
the reaction mixture and evaporation of chloroform solvent
IR: no C == O stretching peak, a strong OH stretching peak
[
�
1
at 3647 cm (19); UV-VR: �max = 270 nm (2); LC-MS: 409
(
parent dihydroxybiphenyl-2 H)].
RESULTS
The MCM-41 synthesized here had the same XRD pat-
tern as that reported in the literature (1, 2). The physical
properties of the obtained MCM-41 were as follows: XRD
d100 spacing, 37.2 A˚ ; pore diameter, 34.6 A˚ ; wall thickness,
2
8
.4 A˚ ; BET surface area, 1023 m /g. The mechanical and
thermal stabilities of MCM-41 are listed in Table 1. MCM-
1 showed a relatively strong resistance to the calcination
4
treatment at higher temperature but was not highly resis-
tant to damage to its mechanical strength.
Through the present study, no oxidation products, such as
monomer counterparts(benzoquinone and/or hydrobenzo-
0
FIG. 1. Yield and conversion versus reaction time in BOH oxidation
quinone derivatives), other than two dimer products, 4,4 -
0
0
catalyzed by (K–Cu)/MCM-41. BOH, 5 mmol; catalyst, (K–Cu)/MCM-41,
dihydroxy-3,3 ,5,5 -tetra-tert-butylbiphenyl (H2DPQ) and
3
�
2
0
3
.2 g (Cu: 3.12 � 10 mmol); K/Cu atomic ratio, 1; reaction temperature,
0
0
0
,3 ,5,5 -tetra-tert-butyl-4,4 -diphenoquinone (DPQ), were
13 K; solvent, 12 ml of chloroform. ᮀ, Conversion; ᭹, H2DOQ yield;
detected using the Cu catalysts supported on MCM-41 or
ZSM-5 zeolite in the presence of alkali metal under the
present reaction conditions.
Leaching of Cu species supported on MCM-41, NaY,
and NaZSM-5 into the reaction solution was not detected
during BOH oxidation under the reaction conditions at-
tempted in this study.
᭡
, DPQ yield.
The relationship between reaction time and yield of both
the productsand the conversion ofBOH usinga basicCuCl2
catalytic sytem (CuCl2 + KOH) without a catalyst support
is illustrated in Fig. 2. The yield of H2DPQ passed through a
maximum value at the reaction time of ca. 1 h and decreased
with further reaction time. Differing from the supported
Oxidation of 2,6-Di-tert-butylphenol
Figure 1 illustrates the time dependence of the yields
of the oxidation products in the oxidation of 2,6-di-tert-
butylphenol (BOH) catalyzed by the K–Cu-coimpregnated
MCM-41 [(K–Cu)/MCM-41] system. The yield of H2DPQ
increased with the increase in reaction time from the initial
stage, but the H2DPQ yield tended to almost level off after
the reaction time of around 10 h. On the other hand, DPQ
was hardly detected in the initial stage of the oxidation;
thus an induction period was observed in the production of
DPQ.
TABLE 1
Mechanical and Thermal Properties of MCM-41
Applied
Calcination
temperature
(K)
Relative intensity
of (100) diffraction
peak
BET surface
area
pressure
2
2
(
kg/cm )
(m /g)
None
773
773
773
100
70
50
1023
638
621
4
6
00
00
FIG. 2. Yield and conversion versus reaction time in BOH oxida-
tion catalyzed by unsupported CuCl2. BOH, 5 mmol; catalyst, CuCl2,
None
None
None
773
1073
1273
100
96
76
1023
1023
889
�
2
� 2
3
.12 � 10 mmol; KOH, 3.12 � 10 mmol; KOH (K)/Cu mole ratio,
; reaction temperature, 313 K; solvent, 10 ml of chloroform + 2 ml of
1
methanol. ᮀ, Conversion; ᭹, H2DPQ yield; ᭡, DPQ yield.

4
20
FUJIYAMA ET AL.
(
K–Cu)/MCM-41 catalyst, DPQ was found to be produced
TABLE 3
almost selectively in a reaction time of more than 5 h using
unsupported CuCl2–KOH catalyst.
Liquid-Phase Adsorption of BOH in the Supports
a
with Mesopores and Micropores
To investigate the cause of the difference in the time
dependence of the oxidation products between the sup-
ported (K–Cu)/MCM-41and the unsupported CuCl2–KOH
catalysts, oxidation of H2DPQ, in place of the starting
BOH, was carried out using both MCM-41 supported and
unsupported Cu catalysts. The yields of DPQ catalyzed
by both (K–Cu)/MCM-41 and CuCl2–KOH catalysts were
Amount of adsorption
(mmol/g support)
Support
Si/Al atomic ratio
—b
53
2.8
MCM-41
NaZSM-5
NaY
0.93
0.88
0.36
a
3
BOH, 1 mmol; support, 0.2 g; solvent, 20 cm of chloroform; adsorp-
tion temperature, 313 K; adsorption time, 24 h.
5
8% and ca. 100% for a reaction time of 24 h, respec-
tively. Thus the higher oxidation rate of H2DPQ to DPQ
catalyzed by the unsupported CuCl2–KOH, in comparison
with the supported (K–Cu)/MCM-41 catalyst, will explain
the lower yield and selectivity of H2DPQ (Fig. 2) over the
prolonged reaction time. Also, the formation of 2,6-di-tert-
butylphenolate, which will preferentially lead to DPQ as
described later, is considered to be easier with the unsup-
ported CuCl2–KOH catalytic system, rather than with the
b
Al is not present.
stricted, in contrast to diffusion into the (K–Cu)/MCM-41
catalyst with mesopores.
Liquid-Phase Adsorption of 2,6-Di-tert-
butylphenol (BOH)
(
K–Cu)/MCM-41 catalyst, because of the presence of free
To investigate the capacity of the support to accommo-
date the BOH molecule, the liquid-phase adsorptions of
BOH on MCM-41, NaZSM-5, and NaY were measured
at 313 K (Table 3). The amount of BOH adsorption on
the MCM-41 support was found to be larger than those on
the NaZSM-5 and NaY supports, but the difference in ad-
sorption amounts of BOH was not so high. Thus the BOH
molecules may be able to penetrate considerably into the
micropores of the NaZSM-5 and NaY zeolites and to be
accommodated in the micropores. The percentage of ex-
ternal surface area occupied in the total surface area of a
support with high BET surface area is thought to be low.
So the contributions of the external surface to liquid-phase
adsorption using the three supports (MCM-41, NaY, and
NaZSM-5) will be rather low and of similar extents, even if
they could not be ignored. It is of interest to note that the
amount of BOH adsorption on NaZSM-5 was higher than
that on NaY, despite the smaller pore size of the former ze-
olite. This seems to be due in part to the higher hydropho-
bicity of the micropores in the NaZSM-5 zeolite because of
its higher Si/Al atomic ratios than NaY.
KOH in the reaction solution.
Comparison of the Catalytic Activities of Cu-Supported
Catalysts with Mesopores and Micropores
To investigate the influence of the difference between
the meso- and microreaction fields on the oxidation of
BOH, which has two sterically bulky substituents and a ki-
netic diameter (the minimum diameter of the molecule) of
ca. 7 A˚ , we attempted to use NaZSM-5 and NaY ze-
olites as supports for the catalyst in place of MCM-
4
1 (Table 2). The oxidation activities of both catalysts
with a micropore support [(K–Cu)/NaZSM-5 and (K–
Cu)/NaY] were found to be considerably lower than
that of the (K–Cu)/MCM-41 catalyst. Because the ki-
netic parameter of both oxidation products, H2DPQ and
DPQ, is around 11 A˚ , it is difficult for both H2DPQ
and DPQ to be formed in the micropores of both zeo-
lites by BOH oxidation. Diffusion of the starting mate-
rial, BOH, into the micropores of the (K–Cu)/NaZSM-5
and (K–Cu)/NaY catalysts may also be considered to be re-
Effect of the Added Alkali Metals on BOH Oxidation
TABLE 2
The effect of the method of adding CH3COOK on oxi-
dation activity is shown in Table 4. At first, it must be noted
that Cu/MCM-41 without added K was found to have
catalytic activity for BOH oxidation though the oxidation
activity was considerably low. K/MCM-41 in which only
alkali metal, K, was impregnated had no catalytic activity
for BOH oxidation. (K–Cu)/MCM-41 had higher oxidation
activity than did Cu/MCM-41 without the added K. The
catalytic system of Cu/MCM-41 + CH3COOK, in which
CH3COOK is dissolved in the reaction solution rather than
being impregnated into MCM-41, had higher oxidation ac-
tivity than (K–Cu)/MCM-41. The higher concentration of
Effect of Supports with Mesopores and Micropores
a
on BOH Oxidation
Pore
size of
support
( A˚ )
BET surface
area of
BOH
conversion
(% )
H2DPQ DPQ
yield
(% )
support
yield
(% )
2
Catalyst
(m /g)
(
(
(
K–Cu)/MCM-41
K–Cu)/NaZSM-5
K–Cu)/NaY
35
5.5
7.4
1020
360
750
17.8
0.4
0.2
11.0
0.4
0.2
6.8
0.0
0.0
a
� 2
Catalyst, 0.2 g (Cu, 3.12 � 10 mmol); K/Cu atomic ratio, 1; reaction
3
time, 24 h; reaction temperature, 313 K; solvent, 12 cm of chloroform.

OXIDATION OF PHENOLS WITH Cu/MCM-41 CATALYSTS
421
TABLE 4
TABLE 5
Effect of the Addition Method of CH3COOK on the Catalytic
Effect of KOH Added in the Reaction Solution on the Catalytic
a
a
Activity of BOH Oxidation
Activity of BOH Oxidation
BOH
conversion
(% )
H2DPQ DPQ
BOH
conversion
(% )
H2DPQ
yield
(% )
DPQ
yield
(% )
K/Cu
atomic ratio
yield
(% )
yield
(% )
K/Cu
atomic ratio
Run
Catalyst
Run
Catalyst
1
2
3
Cu/MCM-41
—
1
1
5.7
18.0
34.0
5.1
11.2
15.0
0.6
6.8
19.0
1
2
Cu/MCM-41
—
1
5.7
11.0
5.1
8.9
0.6
2.1
(K–Cu)/MCM-41
Cu/MCM-41 +
Cu/MCM-41 +
b
KOH
b
CH3COOK
3
Cu/MCM-41 +
16
0.7
0.6
0.1
b
4
5
6
Cu/MCM-41 +
3
3
50.0
45.0
21.0
11.0
13.0
0.4
39.0
32.0
20.6
KOH
b
CH3COOK
4
5
6
CuCl2
—
1
16
4.8
13.0
10.0
0.2
1.0
1.5
4.6
12.0
8.5
(K–Cu)/MCM-41 +
CuCl2 + KOH
b
CH3COOK
CuCl2 + KOH
Cu/MCM-41 +
—
�
+ c
a
� 2
BO K
Cu, 3.12 � 10 mmol;reaction time, 24h;reaction temperature, 313K;
3
solvent, chloroform 12 cm .
a
� 2
b
3
3
Catalyst, 0.2 g (Cu, 3.12 � 10 mmol); reaction time, 24 h; reaction
Solvent, chloroform 10 cm + methanol 2 cm .
3
temperature, 313 K; solvent, chloroform 12 cm .
b
3
3
Solvent, chloroform 10 cm + methanol 2 cm .
Isolated potassium di-tert-butylphenolate.
c
oxidation. The Cu/MCM-41 catalyst without an added al-
kali and the unsupported CuCl2 catalyst in the absence of
an added alkali showed high selectivities for H2DPQ and
CH3COOK (higher K/Cu atomic ratio) caused higher BOH DPQ, respectively. H2DPQ was thus found to be almost
conversion. Oxidation of the corresponding isolated potas- selectively synthesized, though the yield was not high, by
sium phenolate of BOH was carried out using Cu/MCM-41 supporting Cu on MCM-41 in the absence of alkalis, rather
catalyst without added K (Table 4, run 6). Potassium phe- than using unsupported Cu ions. Figure 4 illustrates the in-
nolate was found to be oxidized almost selectively to DPQ fluence of the K/Cu atomic ratio of the (K–Cu)/MCM-41
with the Cu/MCM-41 catalyst without added K. One of the catalyst on the yields of both H2DPQ and DPQ. Both yields
main roles played by the added alkali metal is to directly had maximum values at a K/Cu ratio of around 1. Further
cause BOH to form the corresponding phenolate anion, addition of K caused decreases in both DPQ and H2DPQ
which will be more easily oxidized than the neutral BOH yields. The color of the (K–Cu)/MCM-41 catalyst was visu-
molecule (16), in addition to the interaction of the added ally observed to vary from blue to off-white with an in-
alkali metal with the Cu species supported on MCM-41, crease in K/Cu atomic ratio. High K/Cu ratios are thus
the interaction of which is considered to make oxidative considered to cause coverage of Cu species with the al-
dehydrogenation activity high (20–25). The DPQ selec- kali and/or reduction of Cu(II) to Cu(I) through electron
�
+
tivity (98% ) of the Cu/MCM-41 + BO K system (run 6) transfer from the alkali to Cu(II), variations of which in the
was considerably higher than that (38% ) of the (K–Cu)/
MCM-41 + BOH system (run 2), despite the similar
conversions (18 and 21% , respectively) of both catalytic
systems.
The effect of adding KOH to the reaction solution
on BOH conversion and oxidation products is shown in
Table 5, together with the data for unsupported CuCl2
catalytic systems. Although the addition of KOH to the
reaction solution caused an increase in BOH conversion
(
Table 5, run 2), the degree of the increase in BOH con-
version was not as high as that with CH3COOK addition
Table 4, run 2). Furthermore, too much KOH inversely
(
caused a significant decrease in the conversion (Table 5,
run 3). The XRD pattern of the used Cu/MCM-41 catalyst
(
Table 5, run 3) showed that the intensities of the diffrac-
tion peaks based on a hexagonal prism abruptly decreased.
Fig. 3b). Thus, the destruction of the mesopore structure
caused by strong alkali in the reaction solution is considered
FIG. 3. XRD patterns of (a) Fresh Cu/MCM-41 catalyst (Cu: 3.12 �
(
� 2
� 2
10
mmol) and (b) Cu/MCM-41 catalyst (Cu, 3.12 � 10 mmol) used in
BOH oxidation in the presence of methanolic KOH solution (KOH/Cu =
to cause the disappearance of the catalytic activity for BOH 16) followed by calcination at 773 K for 5 h.

4
22
FUJIYAMA ET AL.
We are now attempting the liquid-phase oxidation of
BOH using Cu ion-exchanged Na � MCM-41 with added
K (K/Cu–Na � MCM-41) as catalyst, in place of the Cu-im
pregnated MCM-41 with added K ((K–Cu)/MCM-41)
counterparts studied here; the results will be reported else-
where in the near future.
BOH Oxidation Catalyzed by Used (K–Cu)/MCM-41
BOH oxidation was attempted using used (K–Cu)/
MCM-41 catalyst, in place of the fresh one, to investigate
the durability of the catalyst under the present reaction
conditions. Two used (K–Cu)/MCM-41 catalysts were uti-
lized for the second BOH oxidation run (all the reaction
conditions are same as those of the first run, except the
catalyst). The first used catalyst [used (K–Cu)/MCM-41
(
A)] was washed only with CHCl3 after the filtration, and
the second one [used (K–Cu)/MCM-41 (B)] was further
calcined at 773 K for 5 h in air flow after washing with
CHCl3. Conversion of BOH using both the used catalysts
FIG. 4. Influence of K/Cu mole ratio on yield and conversion using
K–Cu)/MCM-41 catalyst. BOH, 5 mmol; catalyst, (K–Cu)/MCM-41, 0.2 g
(
(
�
2
Cu: 3.12 � 10 mmol); K/Cu mole ratio, 1; reaction temperature, 313 K; were considerably lower than that using the fresh catalyst
solvent, 12 ml of chloroform. ᮀ, Conversion; ᭹, H2DOQ yield; ᭡, DPQ
yield.
(
Table 6). In particular, the decrease in the yield of DPQ
was conspicuous in both used catalysts. There were hardly
any differences in oxidation activity and selectivity between
the used catalyst treated by only CHCl3 washing and that
treated with CHCl3 washing followed by calcination, as
evidenced in Table 6. To investigate the cause of the de-
crease in catalytic activity for BOH oxidation, the amounts
of both Cu and K on used (K–Cu)/MCM-41 (A) and (B)
were compared with those on the fresh catalyst. No change
in the amounts of Cu supported on both the fresh and the
used catalysts was detected as described previously, but the
amounts of K on the used (K–Cu)/MCM-41 (A) and (B)
catalysts were found to decrease to about 45% that on the
fresh (K–Cu)/MCM-41 catalyst. The decrease in oxidation
activity is thought to be brought about by leaching of the
K during the first BOH oxidation run. It is of interest to
note that the catalytic activity and the product selectivity of
the used catalystswith lower-impregnated K (Table 6, runs2
and 3) were almost comparable to those of the Cu/MCM-41
catalyst without added K (Table 5, run 1). The deactivating
catalyst will cause the catalytic activity for BOH oxidation
to decrease.
The influence of K (as CH3COOK) in the reaction solu-
tion to the Cu ratio on both H2DPQ and DPQ yields using
the Cu/MCM-41 catalyst is illustrated in Fig. 5. The yield of
DPQ passed through a maximum value at a K/Cu atomic
ratio of around 3, though the H2DPQ yield was almost con-
stant with the increase in K/Cu atomic ratio.
TABLE 6
BOH Oxidation Catalyzed by Used (K–Cu)/MCM-41a
BOH
conversion
(% )
H2DPQ
yield
(% )
DPQ
yield
(% )
Run
Catalyst
1
2
3
Fresh (K–Cu)/MCM-41
Used (K–Cu)/MCM-41 (A)
Used (K–Cu)/MCM-41 (B)
18.3
3.7
4.1
11.4
3.5
3.8
6.9
0.2
0.3
b
c
FIG. 5. Influence of K/Cu mole ratio on yield and conversion using
a
� 2
unsupported CuCl2 catalyst with methanolic CH3COOK. BOH, 5 mmol;
Cu, 3.12 � 10 mmol;reaction time, 24h;reaction temperature, 313K;
�
2
3
catalyst, Cu/MCM-41 (Cu: 3.12 � 10 mmol), 0.2 g; reaction temperature,
13 K; reaction time, 24 h; solvent, 10 ml of chloroform + 2 ml of methanol.
, Conversion; ᭹, H2DOQ yield; ᭡, DPQ yield.
solvent, chloroform 12 cm .
b
3
ᮀ
Washed with CHCl3 before use as a used catalyst.
Washed with CHCl3 and calcined at 773 K for 5 h in air flow.
c

OXIDATION OF PHENOLS WITH Cu/MCM-41 CATALYSTS
423
behavior of (K–Cu)/MCM-41 catalyst observed in the DMF solvent has been reported (27) to give DPQ almost
conversion-versus-reaction time curve after a reaction time selectively (more than 99% ) in the BOH oxidation.
of about 5 h (Fig. 1) is in part explained by the K leaching
during the oxidation.
The scheme of formation of the dimeric oxidation prod-
ucts, H2DPQ and DPQ, by the oxidative coupling reac-
tion of 2,6-disubstituted phenol has been known (28) to
be a consequent one in which DPQ is formed via H2DPQ.
Also, DPQ formation via the oxidative dehydrogenation of
DISCUSSION
The oxidation of BOH using CuCl catalyst in acetonitrile an intermediate obtained from the C–C coupling reaction
in the presence of oxygen has been reported (26) to pro- of the corresponding phenoxy radicals has been reported
duce the corresponding p-benzoquinone at high catalyst to (27, 29) in the oxidation of 2,6-disubstituted phenol cata-
BOH ratios. However, the Cu/MCM-41, Cu/NaZSM-5, and lyzed by Co(salen). Based on actual knowledge of the ox-
K-added counterparts used in this study gave only dimeric idation of phenol derivatives obtained so far, a plausible
oxidation products, H2DPQ and DPQ, and not a corre- reaction path from BOH to both H2DPQ and DPQ prod-
sponding monomeric benzoquinone. The selectivities for ucts is depicted in Scheme 1.
both benzoquinone and DPQ derivatives of BOH oxida-
tion catalyzed by Co(salen) derivatives have been reported BOH is easily deprotonated, giving the corresponding
27) to depend on the nature of axially coordinated ligand phenolate anion (A). The corresponding phenoxy radical
In a basic medium or the presence of alkali metal,
(
as well as on the solvent used. In particular, Mn(salen) in (B, C) is formed via a one-electron transfer from the
SCHEME 1

4
24
FUJIYAMA ET AL.
phenolate anion (A) to Cu(II) ion. (The following oxida- fluence of the mesoporous field itself, it is worthwhile to
tion of BOH is also possible: via at first a one-electron loss note that Cu/MCM-41 (Table 5, run 1) and unsupported
to form the corresponding cation radical followed by de- CuCl2 (Table 5, run 4) catalysts without added alkalis had
protonation to form the phenoxy radical.) This oxidation catalytic activity for BOH oxidation and the selectivities for
process proceeds more easily by dissolving an alkali such H2DPQ and DPQ of both the catalysts were just opposite.
as CH3COOK in the reaction medium, instead of being sup- Thus Cu/MCM-41 favored H2DPQ formation; on the other
ported onto the MCM-41, which explains the higher activity hand, the oxidation product DPQ was preferentially ob-
of the Cu/MCM-41 + CH3COOK system than the (K–Cu)/ tained using unsupported CuCl2 catalyst. The results seem
MCM-41 catalyst (Table 4, runs 2 and 3). The corresponding to clearly indicate that the mesopores offers a reaction field
dimer (D) is formed through C–C coupling of the phenoxy substantially preferable for H2DPQ formation in compari-
radicals (C). The dimer (D) can easily rearrange to the more son with the bulk solution, though we explain the difference
stable tautomer, product H2DPQ. The product DPQ is ob- in selectivity. It must be noted from a synthetic chemical
tained by two consecutive oxidation steps, each one involv- point of view that the catalytic Cu species in the mesopores
ing both the withdrawal of an electron and deprotonation, of MCM-41 give higher selectivity for H2DPQ formation,
to form species (E) first and then the product DPQ after the in contrast to the reaction field in the bulk solution, in which
second oxidation step. The product DPQ can result through field DPQ was preferentially formed.
an alternative step (step g) via no H2DPQ intermediate:
The catalytic activity of the (K–Cu)/MCM-41 catalyst in
the oxidative dehydrogenation of dimer (D) directly gives BOH oxidation was found to be considerably higher, com-
DPQ without H2DPQ intermediate (29). The possibility of pared with (K–Cu)/NaZSM-5 and (K–Cu)/NaY catalysts
formation of DPQ by oxidative dehydrogenation of dimer (Table 2). The difference in the catalytic activity of the BOH
(
D) cannot be ignored in the presence of Cu catalysts and oxidation using the catalytic supports with different-sized
alkaline and in an O2 atmosphere. The appearance of the pores indicates that the reaction field of the oxidation of
induction period in the plots between the DPQ yield and BOH is mainly the interior of the mesopores or microp-
reaction time using the (K–Cu)/MCM-41 catalyst (Fig. 1) ores, rather than the outer surface. The difference in cata-
is explained by considering DPQ to be formed through a lytic activity caused by using Cu catalysts to which K is
consecutive step involving H2DPQ as an intermediate. On added supported by both mesopore and micropore sup-
the contrary, the unsupported CuCl2 catalyst in the pres- ports is considered to be due to a kind of shape selectiv-
ence of dissolved KOH brought no induction period for ity. On the other hand, the liquid-phase adsorption amount
the formation of DPQ as illustrated in Fig. 2, the results of of BOH on the MCM-41 support was not very different
which suggest that the production of DPQ can also occur via from those on both the NaZSM-5 and the NaY supports
step g, in addition to the route passing through the H2DPQ (Tables 3), though the former support had a higher capac-
intermediate (steps d → e → f). Steps e and f are faster in ity for liquid adsorption of BOH than the latter two zeolite
basic medium or in the presence of alkalis. DPQ formation supports. The results of both BOH oxidation (Table 2) and
via the oxidative dehydrogenation of (D) (step g) is also liquid-phase adsorption of BOH (Table 3) suggest that the
enhanced in the presence of alkalis, based on the results formation of sterically bulky H2DPQ and DPQ (a kinetic
(
20–25) that the addition of alkali metals to supported Cu parameter of 11 A˚ ) and/or their transfers from the microp-
catalysts has considerably promoted the oxidative dehydro- ores to the reaction solution will be strongly restricted in the
genation of alcohols. From the Cu catalysts and the reaction micropores of both zeolites, rather than the transfer of the
conditions used in this study, it is reasonable to suppose that reactant, BOH, into the micropores being restricted. Thus
product DPQ is formed by two consecutive oxidation steps the shape selectivity, particularly the product restriction, is
of H2DPQ derived from (D) (route d → e → f) and also considered to be the main cause of the difference in oxida-
by the oxidative dehydrogenation of (D) (route g), though tion activity between the MCM-41 and zeolite (NaZSM-5,
the contribution of each route for DPQ formation depends NaY) supports.
on the catalysts and the reaction conditions, and cannot be
estimated quantitatively at the present stage.
In this study, we have not yet obtained information on the
state of Cu impregnated on the MCM-41 support using var-
The selectivity of H2DPQ catalyzed by the Cu/MCM- ious spectrometric methods. As an extension of the present
4
1 + KOH system (Table 5, run 2) is considerably higher study, Cu ion-exchanged MCM-41 (Cu–Na � MCM-41) was
than that catalyzed by the CuCl2 + KOH system (Table 5, prepared using Na � MCM-41 (Si/Al atomic ratio, 59.5)
run 5) in the comparison of the two BOH conversions as a support, and K-impregnated Cu–Na � MCM-41 (K/
with almost similar values. The results mean that the route Cu–Na � MCM-41) has been used as a catalyst for the BOH
e → f and/or route g are retarded using a MCM-41 sup- oxidation (Table 7). BOH conversion and yields of both
port. One of the reasons may be that the mesopore field H2DPQ and DPQ were comparable for the supported
of the MCM-41 with hydrophobicity causes a lower con- Cu catalysts prepared by both impregnation [(K–Cu)/
centration of KOH than in the bulk reaction medium, in MCM-41] and ion-exchange (K/Cu–Na � MCM-41) meth-
which KOH promotes steps e, f, and also g. To see the in- ods (K was supported by impregnation for the catalyst).

OXIDATION OF PHENOLS WITH Cu/MCM-41 CATALYSTS
ACKNOWLEDGMENTS
425
TABLE 7
BOH Oxidation Using Cu-Supported MCM-41 Catalysts Prepared
We expressour thanksto one ofthe reviewersfor their usefulsuggestion
in Scheme 1. Thanks are also extended to Dr. I. Kawafune of Osaka Mu-
nicipal Technical Research Institute for his measurement of the pore size
distribution of MCM-41 synthesized in this study. We express our thanks to
Mr. Kenji Nomura of Kobe University for his technical assistance during
this work.
a
by Both Impregnation and Ion-Exchange Methods
BOH
conversion
(% )
H2DPQ
yield
(% )
DPQ
yield
(% )
Run
Catalyst
1
2
(K–Cu)/MCM-41b
K/Cu–Na � MCM-41
18.3
14.3
11.4
9.0
6.9
5.3
REFERENCES
c
1
2
. Kresge, C. T., Leonowicz, M. E., Roth, W. J., Vartuli, J. C., and Beck,
J. S., Nature 359, 710 (1992).
. Beck, J. S., Vartuli, J. C., Roth, W. J., Leonowicz, M. E., Kresge, C. T.,
Schmit, K. D., Chu, C. T.-W., Olson, D. H., Sheppard, E. W., McCullen,
S. B., Higgins, J. B., and Schlenker, J. L., J. Am. Chem. Soc. 114, 10834
a
� 2
Cu, 3.12 � 10 mmol;reaction time, 24h;reaction temperature, 313K;
3
solvent, chloroform 12 cm , K/Cu atomic ratio, 1.
b
Cu was impregnated followed by impregnation of K.
Cu was ion-exchanged followed by impregnation of K.
c
(
1992).
3. Chen, C.-Y., Li, H.-X., and Davis, M. E., Microporous Mater. 2, 17
The results imply that the state of Cu impregnated on
the MCM-41 support with relatively low amount of Cu of
(1993).
4
5
. Tanev, P. T., Chibwe, M., and Pinnavaia, T. J., Nature 368, 321 (1994).
. Huo, Q., Margolese, D. I., Ciesia, U., Feng, P., Gier, T. E., Sieger, P.,
Leon, R., Petroff, P. M., Schuth, F., and Stucky, G. D., Nature 368, 317
1
wt% may be similar to that of ion-exchanged Cu, the
Cu ions of which are well dispersed on the mesoporous
matrix.
(
1994).
. Blasco, T., Corma, A., Navarro, M. T., and Pariente, J. P., J. Catal. 153,
5 (1995).
6
7
2
. Yuan, Z. Y., Liu, S. Q., Chen, T. H., Wang, J. Z., and Li, H. X., J. Chem.
Soc. Chem. Commun., 973 (1995).
. Zhao, D., and Goldfarb, D., J. Chem. Soc. Chem. Commun., 875 (1995).
. Corma, A., Martinez, A., Martinez-Soria, V., and Monton, J. B.,
J. Catal. 153, 25 (1995).
CONCLUSIONS
8
9
Cu-impregnated MCM-41 with added alkali metal [(K–
Cu)/MCM-41] was found to catalyze the liquid-phase ox- 10. Inui, T., Kim, J. B., and Seno, M., Catal. Lett. 29, 271 (1994).
0
1
1
1
1. Kozhevnikov, I. V., Sinnema, A., Jansen, R. J. J., Pamin, K., and
Bekkum, H. V., Catal. Lett. 30, 241 (1995).
2. Liu, A. M., Shido, T., and Ichikawa, M., J. Chem. Soc. Chem. Commun.,
idation of 2,6-di-tert-butylphenol (BOH) to form 4,4 -
0
0
dihydroxy-3,3 ,5,5 -tetra-tert-butylbiphenyl (H2DPQ) and
,3 ,5,5 -tetra-tert-butyl-4,4 -diphenoquinone (DPQ). Both
Cu-supported MCM-41 (Cu/MCM-41) and unsupported
0
0
0
3
507 (1995).
3. Reddy, J. S., Liu, P., and Sayari, A., Appl. Catal. A 148, 7 (1996).
CuCl2 catalysts without added alkalis were also found to 14. Okamura, J., Nishiyama, S., Tsuruya, S., and Masai, M., J. Mol. Catal.
A 135, 133 (1998).
5. Tsuruya, S., Trends Inorg. Chem. 3, 71 (1993) and the references cited
therein.
have catalytic activity for BOH oxidation. The former cata-
1
1
lyst favored H2DPQ formation and the latter catalyst gave
preferentially the DPQ product. In the presence of alkalis,
the selectivities of H2DPQ and DPQ greatly depended on
6. Tadokoro, H., Nishiyama, S., Tsuruya, S., and Masai, M., J. Catal. 138,
24 (1992).
the presence of the added alkali metal and/or the method 17. Miyagi, A., Nishiyama, S., Tsuruya, S., and Masai, S., J. Mol. Catal. 55,
79 (1989).
3
of alkali metal addition. A scheme for BOH oxidation was
proposed. At first the corresponding phenoxy radical was
generated. H2DPQ was thought to be formed due to a tau-
tomeric rearrangement of a unstable dimer produced by
the C–C coupling of the phenoxy radical generated. DPQ
was proposed to be formed via the consecutive oxidation
steps of H2DPQ and/or via the oxidative dehydrogenation
of the unstable dimer without passing through the H2DPQ
intermediate. Cu-impregnated NaZSM-5 with added al-
kali metal [(K–Cu)/NaZSM-5] and NaY [(K–Cu)/NaY] ze-
olites, which differed from the (K–Cu)/MCM-41 catalyst,
were found to have hardly any catalytic activity for liquid-
phase oxidation of BOH. The amounts of liquid-phase ad-
sorption of BOH on NaZSM-5 and NaY supports, both of
which have micropores, were found to be comparable to
that on the MCM-41 support which has mesopores. These
results indicate that the difference in BOH oxidation activ-
ities between Cu-supported catalysts with mesopores and
those with micropores is governed by the shape selectivity
in product restriction.
1
1
8. Araguer, R. J., and Landolf, G., U.S. Patent 3,702,886 (1972).
9. Hozumi, K., and Kitamura, K., “Kikibunseki Tsuron” [Japanese],
p. 126. Hirokawa shotenn, Tokyo, 1993.
0. Hayashibara, H., Nanbu, T., Nishiyama, S., Tsuruya, S., and Masai, M.,
in “Proceedings of 9th International Zeolites Conference, Montreal”
2
(
R. von Ballmoos, J. B. Higgins, and M. M. J. Treaty, Eds.), p. 575.
Butterworth–Heinemann, London, 1993.
1. Hayashibara, H., Nishiyama, S., Tsuruya, S., and Masai, M., J. Catal.
2
2
2
2
2
153, 254 (1995).
2. Genta, M., Nishiyama, S., Tsuruya, S., and Masai, M., J. Chem. Soc.
Faraday Trans. 92, 1267 (1996).
3. Arai, M., Nishiyama, S., Tsuruya, S., and Masai, M., J. Chem. Soc.
Faraday Trans. 92, 2631 (1996).
4. Sueto, S., Nishiyama, S., Tsuruya, S., and Masai, M., J. Chem. Soc.
Faraday Trans. 93, 659 (1997).
5. Jian Xu, Ekblad, M., Nishiyama, S., Tsuruya, S., and Masai, M.,
J. Chem. Soc. Faraday Trans. 94, 473 (1998).
6. Capdevielle, P., and Maumy, M., Tetrahedron Lett. 24, 5611 (1983).
7. Kothari, V. M., and Tazuma, J. J., J. Catal. 41, 180 (1976).
8. Musso, H., in “Oxidative Couping of Phenols” (W. I. Taylor and
A. R. Battersby, Eds.), p. 1. Marcel Dekker, New York, 1967.
9. Simandi, L. I., in “Catalytic Activation of Dioxygen by Metal Com-
plexes,” p. 196. Kluwer Academic, Dordrecht, 1992.
2
2
2
2