Oxidation of 2,3,6-trimethylphenol over Ti- and V-containing mesoporous mesophase catalysts: Structure-activity/selectivity correlation

Article abstract of DOI:10.1006/jcat.2001.3264

The oxidation of 2,3,6-trimethylphenol (TMP) has been carried out over well-organized Ti- and V-containing mesoporous mesophase silicate catalysts (Ti- and V-MMM) using hydrogen peroxide and tert-butyl hydroperoxide (TBHP) as oxidants. Vanadium leaching was observed in the V-MMM/H₂O₂ and V-MMM/TBHP systems, whereas no titanium leaching occurred for Ti-MMM with both oxidants. TMP oxidation runs on a surface of Ti-MMM catalyst, producing 2,3,5-trimethyl-1,4-benzoquinone (TMBQ) with selectivity as high as 86% at 100% TMP conversion. Titanium content in Ti-MMM determines the state of the active catalytic site and influences the structure regularity, thus affecting the catalytic behavior. The catalysts with titanium loading in the 1.5-2 wt% range show the highest activity and selectivity. Some loss of catalytic properties observed after recycling may occur due to collapsing of the catalyst structure caused by water. The lower the water concentration in the reaction mixture, the more stable the catalyst.

Full text of DOI:10.1006/jcat.2001.3264

Journal of Catalysis 202, 110–117 (2001)
doi:10.1006/jcat.2001.3264, available online at http://www.idealibrary.com on
Oxidation of 2,3,6-Trimethylphenol over Ti- and V-Containing Mesoporous
Mesophase Catalysts: Structure–Activity/Selectivity Correlation
N. N. Trukhan, V. N. Romannikov, E. A. Paukshtis, A. N. Shmakov, and O. A. Kholdeeva1
Boreskov Institute of Catalysis, Acad. Lavrentiev Av., 5, Novosibirsk 630090, Russia
Received January 24, 2001; accepted April 27, 2001
geneous catalysts that leads to the known problems with
catalyst separation and, therefore, contamination of the
goal product with transition metal compounds. An efficient
method for TMBQ production based on the employment of
a solid, heterogeneous catalyst has not yet been reported.
Recently, an attempt was made to use Fe(III) phthalocya-
nine dimeric catalyst immobilized onto amorphous SiO2 in
TMP oxidation with TBHP; however, the selectivity with
this catalyst was good (72–80% ) only in the first catalytic
cycle (13).
The oxidation of 2,3,6-trimethylphenol (TMP) has been car-
ried out over well-organized Ti- and V-containing mesoporous
mesophase silicate catalysts (Ti- and V-MMM) using hydrogen per-
oxide and tert-butyl hydroperoxide (TBHP) as oxidants. Vanadium
leaching was observed in the V-MMM/H2O2 and V-MMM/TBHP
systems, whereas no titanium leaching occurred for Ti-MMM with
both oxidants. TMP oxidation runs on a surface of Ti-MMM cata-
lyst, producing 2,3,5-trimethyl-1,4-benzoquinone (TMBQ) with se-
lectivity as high as 86% at 100% TMP conversion. Titanium content
in Ti-MMM determinesthe state of the active catalytic site and influ-
ences the structure regularity, thus affecting the catalytic behavior.
The catalysts with titanium loading in the 1.5–2 wt% range show
the highest activity and selectivity. Some loss of catalytic properties
observed after recycling may occur due to collapsing of the catalyst
structure caused by water. The lower the water concentration in the
c
To date titanium-silicalites TS-1 and TS-2 are the most
efficient catalysts for H2O2-based oxidations of organic
˚
compounds with kinetic diameter <6 A (14, 15 and refe-
rences therein); however the steric restriction in the micro-
porous zeolite (pore size ca. 0.55 nm) prevents its
application in the field of fine chemistry. Mesoporous sil-
icate materials, containing transition metal ions (M) in a
heterogeneous matrix (M-MCM-41, M-MCM-48, M-HMS,
M-Si-mixed oxides, etc.) have attracted much attention as
catalysts for selective oxidation of bulky organic substrates
(16–34). Among other reactions, oxidation of 2,6-di-tert-
butyl phenol to a mixture of corresponding monoquinone
and diphenoquinone over M-HMS and M-MCM-41 (M =
V, Ti) has been reported (19, 23, 26, 27). Recently, oxida-
tions of 2,3,5-trimethylphenol with H2O2 over (Cr)MCM-
41 (35) and TiAPO-5 (36) have been reported to give
TMBQ with yield’s of 64 and 85% , respectively. However,
substantial chromium leaching was observed in the for-
mer system, while the heterogeneity of the latter system
has not been demonstrated and seems to be problematic
(29).
Recently, we reported the alkene and thioether oxidation
with H2O2 over well-organized mesoporous mesophase
silicate materials (Ti- and V-MMM) (37, 38). In the present
work, we considered the possibility of TMP oxidation to
TMBQ by cheap and environmentally friendly oxidants,
H2O2 and TBHP, in the presence of Ti- and V-MMM
catalysts. Special attention was devoted to the problem of
transition metal leaching during the oxidation process. The
question on structure/activity and structure/selectivity re-
lationships has been addressed. Note that when we started
reaction mixture, the more stable the catalyst.
2001 Academic Press
Key Words: titanium; vanadium; mesoporous mesophase cata-
lysts; oxidation; 2,3,6-trimethylphenol; 2,3,5-trimethyl-1,4-benzo-
quinone; H2O2; TBHP.
INTRODUCTION
The selective oxidation of 2,3,6-trimethylphenol (TMP)
is of great interest as a method for the preparation of 2,3,5-
trimethyl-1,4-benzoquinone (TMBQ), a key intermediate
in the synthesis of vitamin E (1, 2). Until recently, stoi-
chiometric oxidation of TMP with MnO2, HNO3, or other
hazardous reagents (3, 4) was the main route for TMBQ
industrial production. In the past decade, the development
of environmentally friendly catalytic methods has become
a challenging goal. A few catalytic systems have been
developed using“clean” oxidants, such asmolecular oxygen
and hydrogen peroxide: (1) copper halides and other
copper salts/O2, (5–7); (2) cobalt complexes with Schiff
bases/O2 (8, 9); (3) ruthenium salts/H2O2 (10); and
(4) heteropoly acids/O2 or H2O2 (11, 12). However, all the
above-mentioned systems are based on the use of homo-
1 To whom correspondence should be addressed. Fax:+7(3832) 343056.
E-mail: khold@catalysis.nsk.su.
110
0021-9517/01 $35.00
c
Copyright
2001 by Academic Press
All rights of reproduction in any form reserved.

OXIDATION OF 2,3,6-TRIMETHYLPHENOL OVER Ti- AND V-MMM
111
our investigation, no works concerning TMP oxidation over fied by GC. After the reaction, catalysts were filtered off,
Ti-containing catalysts had been reported (39). While we washed with methanol, dried in air at room temperature,
were working on this paper, Sorokin and Tuel published and then reused. Pure TMBQ as well as by-products were
the paper devoted to TMP oxidation over anchored ph- separated from the reaction mixture bymeansofadsorption
talocyanine catalysts (40), where they briefly advertised column chromatography on silica gel using gradual elution
the results on TMP oxidation with H2O2 in the pres- (hexane/ethyl acetate). 1H NMR and UV–Vis spectra of the
ence of Ti-containing silica-based molecular sieves. How- obtained TMBQ coincided completely with the reference
ever, since no experimental details or catalyst charac- spectra.
terization were reported, the results manifested in (40)
Instrumentation
cannot be compared with our results and thus collectively
discussed.
Gas chromatographic analyses were performed using a
gas chromatograph “Tsvet-500” equipped with a flame ioni-
zation detector and a quartz capillary column (25 0,3)
filled with Carbowax 20 M. GC-MS analyses of organic
products were conducted using an LKB-4091 instrument.
1H NMR spectra were recorded on an MSL-400 Bruker
spectrometer. DRS-UV measurements were performed on
a Shimadsu UV–VIS 2501PC spectrophotometer.
EXPERIMENTAL
Materials
2,3,6-TMP (Fluka) was recrystallized from hexane.
Hydrogen peroxide (30% ) was concentrated in vacuum up
to 84% and was determined iodometrically prior to use.
TBHP (6.95 M solution in decane, Aldrich) was commer-
cial sample and was titrated prior to experiments. All the
other reactants were obtained commercially and used with-
out further purification.
RESULTS AND DISCUSSION
Catalyst Characterization
The synthesis and characterization of the most of Ti-
MMM (samples 1–6) as well as the V-MMM (sample 7)
Catalyst Synthesis and Characterization
Ti-MMM (samples 1, 2, 3, 4, 5, and 6, containing 4.32, used in this study were reported elsewhere (37, 38). The
2.50, 1.90, 1.53, 1.00, and 0.58 wt% of Ti, respectively) sample codes and basic physical characteristics are listed
and V-MMM (sample 7, 0.88 wt% of V) were prepared in Table 1. Textural parameters were calculated as de-
by the procedure described in (37, 38) under weakly al- scribed in (41, 42 ). In previous work (38), we have shown
kaline conditions by hydrothermal synthesis at 140 C for that XRD reflections observed for Ti- and V-MMM are
40 h (titaniumsilicates) and for 140 h (vanadiumsilicate) in of highly symmetric shape with the full-width-at-half-
the presence of C16H33N(CH3)3Br. The surfactant was re- maximum (FWHM) values of about 0.08–0.09 2 . This
moved from as-synthesized forms by calcination in air flow allowed us to describe the mesostructure of these sam-
at 550–600 C. The calcined forms of the catalysts were char- ples as highly ordered hexagonal arrangement of meso-
acterized by X-ray diffraction (XRD) using synchrotron pores. This was supported also by a narrow interval of rel-
radiation, nitrogen adsorption at low temperature, and ative pressure 1(P/Po) = 0.06–0.08, along with relatively
elemental analysis. The state of the transition metal ions small values of the external surface area (37). For samples
before and after treatment with H2O2 was characterized by 1–3, which have higher Tiloading, the structure perfection is
DRS-UV at ambient conditions. To estimate hydrostability, less. However, comparing XRD patterns of all the Ti-MMM
samples were treated with water at room temperature for samples with XRD patterns reported for other mesoporous
2 h. All the samples were calcined at 560 C before running silicate materials, like Ti-MCM-41 and Ti-HMS (16, 18–20,
UV spectra and before XRD measurements.
24, 27, 28, 33), one can point out considerably more regu-
lar structure of Ti-MMM. The DR-UV spectra for samples
Catalytic Experiments
2–6 (
at 215–228 nm) indicated site-isolated titanium
max
species according to (14, 16, 18, 27). For sample 1, the UV
absorption maximum was shifted to longer wavelengths,
and the band was broader, thus indicating the appearance
of partially olygomerized titanium-oxygen species (14, 16,
27, 33).
Catalytic experiments were performed under vigorous st-
irringin thermostated glassvesselsat 30–80 C. Typically, the
reactions were initiated by adding 0.33–2.64 mmol of H2O2
(as 30–84% aqueous solution) or 1.16 mmol of TBHP (7 M
solution in decane) to a mixture, containing 0.3–0.15 mmol
of TMP, 25–107 mg of a catalyst (0.013 mmol of M), an
internal standard (biphenyl), and 3 ml of a solvent. The ox-
idation products were identified by GC-MS, UV-Vis and
Catalytic Results
Experimentaldata on the catalyticoxidation ofTMP with
1H NMR. TMP conversion and TMBQ yield were quanti- H2O2 and TBHP over the V-MMM catalyst are summarized

112
TRUKHAN ET AL.
TABLE 1
Structural and Textural Parameters of Ti- and V-MMM Catalysts
Structural parameters
Textural parameters
Sample
Si/M atomic
max in DR–UV
spectra (nm)
no. (M) ratio (wt% M) aoa (nm) FWHMb (deg. 2 ) AMec (m2/g) Aextd (m2/g) VMee (cm3/g) dMef (nm) hWg (nm)
1 (Ti)
H-1h
19 (4.32)
19 (4.32)
29 (2.50)
39 (1.89)
49 (1.53)
70 (1.00)
124 (0.58)
124 (0.58)
70 (0.88)
4.47
—
0.117
0.62
911
841
1209
1260
1059
—
1068
927
922
86
81
32
29
53
—
48
38
34
0.738
0.615
0.852
0.903
0.908
0.900
0.932
0.704
0.877
3.50
—
0.97
—
240
—
228
226
218
215
216
—
255, 370
2 (Ti)
3 (Ti)
4 (Ti)
5 (Ti)
6 (Ti)
H-6h
4.23
4.23
4.64
4.40
4.68
—
0.183
0.110
0.078
0.090
0.086
0.46
3.42
3.45
3.79
3.59
3.84
—
0.81
0.78
0.85
0.81
0.84
—
7 (V)
5.18
0.090
4.20
0.98
a Unit cell parameter.
b Full-width-at-half-maximum of the (100) reflection.
c Specific mesopore surface area.
d Specific external surface area.
e Specific mesopore volume.
f Mesopore diameter.
g Wall thickness calculated from the equation ao = dMe + hW
.
h After treatment with water.
in Table 2. The stoichiometry of TMP oxidation with H2O2 complete TMP conversion. The selectivity to TMBQ was
to produce TMBQ is 1 : 2.
moderate (46% ) and decreased with increasing peroxide
concentration. The use of TBHP instead of H2O2 gave only
31% of TMBQ, the reaction time being increased (Table 2).
The problem of transition metal ion leaching from meso-
porous catalysts has been addressed in many papers (23, 26,
28, 29, 33). We examined the V-MMM catalyst using exper-
iments with fast catalyst removal by filtration during the
oxidation process and found the same activity both in
the presence of V-MMM and in the filtrate. This was evi-
dence that vanadium leaching takes place, and the reaction
actually proceeds in the solution. With TBHP used as the
oxidant, the reaction also proceeded in the filtrate but
the oxidation rate was lower compared to the reaction in
the presence of V-MMM (Fig. 1a). Thus, the leaching of
vanadium active species occurs with both oxidants under
the conditions studied and the problem of vanadium leach-
ing from MMM as well as other mesoporous matrixes has
yet to be solved.
Only 90% of TMP were converted when twofold excess
of H2O2 was employed. This indicated some unproductive
decomposition of the oxidant in the presence of V-MMM.
Threefold or higher excess of H2O2 was needed to achieve
TABLE 2
TMP Oxidation with 30% Aqueous H2O2
over V-MMM Catalysta
In sharp contrast to V-MMM-based systems, no further
activity was exhibited in the filtrate after fast removal of
Ti-MMM during TMP oxidation in both Ti-MMM/H2O2
and Ti-MMM/TBHP catalytic systems, even at 80 C. The
catalyst filtration was carried out at the reaction tempera-
ture in order to prevent readsorption of metal ions from
the solution on cooling. Figure 1b clearly shows that the
TMP oxidation process is truly heterogeneous. Addition-
ally, no titanium leaching was found when comparing the
elemental analysis data of the catalyst before and after
several catalytic cycles. Taking into account that titanium
leaching was observed for Ti-MCM-41/H2O2 by other re-
search groups (28, 33), the present work unambiguously
Molar ratio
Yieldb of
TMP
[TMP]/[H2O2]
TMBQ (% )
conversion (% )
1/2
1/3
1/5
47
46
38
27
31
90
99
100
100
98
1/7
1/3.5c
a Reaction conditions: TMP 0.1 M, sample 7, 35 mg,
MeCN 3 ml, 50 C, 1 h.
b GC yield based on TMP consumed.
c TBHP was used instead of H2O2; reaction time 2.7 h.

OXIDATION OF 2,3,6-TRIMETHYLPHENOL OVER Ti- AND V-MMM
113
FIG. 2. lnRo vs 1/T plot for the TMP oxidation with H2O2 over Ti-
MMM (sample 4). Reaction conditions: TMP 0.1 M, Ti-MMM 42 mg,
[TMP]/[H2O2] = 1/7, MeCN 3 ml.
FIG. 1. TMP oxidation with TBHP over V-MMM (a) (sample 7) and
with H2O2 over Ti-MMM (b) (sample 4). Reaction conditions: TMP 0.1 M,
TBHP 0.35 M (or H2O2 0.7 M), V-MMM 35 mg (or Ti-MMM 42 mg),
MeCN 3 ml, 80 C.
demonstrates the advantage of the well-organized Ti-
The selectivity to TMBQ considerably increased with in-
MMM materials. The data on the catalytic TMP oxidation creasing temperature: 55 and 70% at 65 and 80 C, respec-
with H2O2 and TBHP over Ti-MMM (sample 4) are given in tively (runs 5 and 6). Effective activation energy Ea esti-
Table 3. Since the selectivity of TMBQ formation was con- mated from the lnRo – 1/T plot (Fig. 2) was 53.6 kJ/mol.
siderably higher with H2O2 than with TBHP (77 and 32% , This shows that the reaction rate is not limited by the diffu-
respectively), our attempts to optimize reaction conditions sion of reactants to the active catalytic sites, because Ea for
were directed to the reaction with H2O2. Like V-MMM, the diffusion-limited process should be expected to be consid-
Ti-MMM catalyst induced some unproductive peroxide de- erably lower.
composition, and thus a more than stoichiometric amount
of H2O2 was needed to attain complete TMP conversion tivity of TMP oxidation. At 65 C, the yield of TMBQ was
(runs 1–3).
found to be 55, 63, and 71% for MeCN, AcOH, and MeOH,
Generally, the nature of the solvent affected the selec-
TABLE 3
TMP Oxidation with 30% Aqueous H2O2 over Ti-MMM Catalysta
[TMP]/[H2O2]
(molar ratio)
Yieldb of
TMBQ (% )
TMP
conversion (% )
Run
[TMP] (M)
Solvent
T ( C)
Time (h)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.05
0.05
0.1
1/3
1/5
1/7
1/7
1/7
1/7
1/7
1/3
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
AcOH
MeOH
AcOH
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
50
50
50
30
65
80
65
65
80
80
80
80
80
80
80
8
8
8
6
6
6
6
6
39
47
51
48
55
70
63
71
70
71
77
32
82
86
82
86
95
97
33
98
100
98
98
99
90
100
97
100
100
100
1/7
1/3
6
0.4
0.4
3.3
0.4
0.5
0.8
1/3.5
1/3.5c
1/3.5
1/3.5d
1/3.5d
a Sample 4, 42 mg.
b GC yield based on TMP consumed.
c TBHP was used instead of H2O2.
d TMP solution was added by portions to the reaction mixture.

114
TRUKHAN ET AL.
respectively (runs 5, 7, and 8), whereas at 80 C the selec-
tivity attained 70% for both MeCN and AcOH (runs 6
and 9). While TMBQ was fairly stable to overoxidation
at 50 C and thus the TMP/H2O2 ratio did not affect the
selectivity (runs 2 and 3), at 80 C some further oxida-
tion of TMBQ took place when sevenfold molar excess
of H2O2 was employed. As a result, the selectivity value
passed through a maximum (82 and 70% at 93 and 99%
conversion, respectively). Therefore, high excess of H2O2
should be avoided when the temperature is raised. The op-
timal TMP/H2O2 molar ratio to get complete TMP con-
version at high selectivity was found to be 1/3.5 (run 11).
At a lower ratio the phenol conversion was not complete
(run 10).
3
FIG. 3. UV–Vis spectra (l = 1 cm): (a) TMBQ (2.06 10 M) in
It has been well-documented that the first step in
alkylphenol oxidation by one-electron oxidants is the for-
mation of phenoxyl radical ArO (43):
MeCN; (b) reaction mixture after 3 h. Reaction conditions: TMP 0.1 M,
Ti-MMM 42 mg (sample 1), [TMP]/[H2O2] = 1/6, MeCN 3 ml, 50 C. Be-
fore running spectra the catalyst was separated and the reaction mixture
was 25-fold diluted with MeCN.
ArOH + Ox → ArO + Red + H+.
The formation of C–C and C–O coupling products
allowed us to suggest that TMP oxidation in the Ti-
MMM/H2O2 catalytic system proceeds via the formation
of phenoxyl radical. Taking this hypothesis in mind, it was
reasonable to expect that lowering TMP initial concen-
tration would result in suppression of the polymerization
process, thusincreasingTMBQ yield. Indeed, the selectivity
to TMBQ raised from 77 to 82% when TMP concentra-
tion was twice reduced (runs 11 and 13). In turn, stepwise
addition of TMP to the reaction mixture also enhanced
the selectivity (runs 14 and 15), while slow dosing of H2O2
gave an opposite effect. Meanwhile, we cannot exclude that
an alternative mechanism, involving TMP electrophylic
hydroxylation with titanium hydroperoxocomplex, takes
place to yield TMBQ. It should be mentioned, however,
that the mechanism of interaction of titanium hydroper-
oxocomplex with organic substrates, including phenol, still
remains a matter of discussion (14). Probably, whether ho-
molyticor heterolyticroute takesplace, dependson the sub-
strate nature. A heterolytic mechanism seems to be more or
less proved for alkene epoxidation (14, 15, 29), while a ho-
molytic mechanism most likely operates in alkane (14, 29)
and thioether (46, 47) oxidations. Moreover, the reaction
mechanism seems to be dependent on the catalyst nature.
Thus when using amorphous titanosilicate, the oxidation of
the side chain in toluene was found to predominate thus
indicating a radical mechanism, while TS-1 and TS-2 pro-
duced only cresols indicating electrophylic hydroxylation
(21). To elucidate the actual mechanism of TMP oxidation
in the Ti-MMM/H2O2 system, a thorough mechanistic study
is needed.
The following routes of ArO transformations lead to
different oxidation products:
(1) ArO + Ox + ...... + quinone
(2) C–C coupling: 2ArO + biphenol
+Ox
+ diphenoquinone
(3) C–O coupling: nArO + polyphenylene oxide
The tail-to-tail dimer, 2,20,3,30,5,50-hexamethyl-4,40-
biphenol (BP), and the head-to-tail polymer, polypheny-
lene oxide, have been established to be typical by-products
formed in TMP oxidations to TMBQ (11 and refer-
ences cited therein). Note that the corresponding diphe-
noquinone, which could in principal result from further
oxidation of BP, had never been found in TMP oxida-
tion processes. At the same time, the formation of 3,30,5,
50-tetramethyldiphenoquinone from 2,6-dimethylphenol is
well-precedented (19, 23, 26, 27, 41–45). Sterical reasons
are most likely responsible for this phenomenon. More-
over, we have found earlier that BP can be an intermediate
in TMP oxidation to TMBQ (11). The only product of TMP
oxidation in the Ti-MMM/H2O2 system detected byGC-MS
was TMBQ. Meanwhile, the comparison of UV–Vis spec-
tra of a reference TMBQ and the reaction mixture clearly
showed the presence of a yellow product (or products) dif-
ferent from TMBQ (Fig. 3). The by-product fraction was
1
separated on silica gel and BP was identified by H NMR
(11). However, the strong absorption in the visible region
(Fig. 3) could not be due to BP, which is colorless. The yel-
low by-product could not result from TMBQ overoxidation
because the quinone was found to be stable at the reac-
tion conditions (50 C). We suppose that this is, probably
a C–O-coupling product, polyphenylene oxide, or another
nonvolatile polymeric compound.
Optimization of the reaction conditions allowed us to
produce TMBQ with the yield of86% at 100% TMP conver-
sion using Ti-MMM/H2O2 system. Note that this selectivity
value is similar to that observed for the best homogeneous

OXIDATION OF 2,3,6-TRIMETHYLPHENOL OVER Ti- AND V-MMM
115
structure regularity loss and partial titanium oligomeriza-
tion occur with increasing titanium content, it seems to
be problematic to estimate separately the effect of each
of these two factors on the Ti-MMM catalytic properties.
We believe that a comparison of Ti-MMM samples with
Ti-HMS may be profitable. Thus we have found that the
Ti-HMS catalyst with 2.60 wt% Ti, which has a titanium
dispersion very close to that of sample 1 according to the
DR-UV data, but does not have a regular structure (37),
showed considerably poorer activity and selectivity com-
pared sample 1. Thus, we may conclude that both the tita-
nium state and structure regularity (titanium accessibility)
determine catalyticactivityand selectivityofTi-MMM cata-
lysts in the TMP oxidation.
Less understandable was the decrease in the catalytic ac-
tivity and selectivity observed for samples with low Ti con-
tent (66and 39% selectivityfor sampleswith 1.00and 0.58%
Ti, respectively), given the fact that the structure of these
Ti-MMM samples has been established to be perfect and
site-isolated titanium species have been found to predom-
inate (Table 1). Therefore, structure perfection and high
dispersion of titanium are necessary but not sufficient fac-
torswhich determine the catalyticperformance ofTi-MMM
catalysts. We suppose that the reason for the reduced ac-
tivity of Ti-MMM samples with low titanium content might
be their lower hydrolytic stability compared to the samples
with high Ti content. The problem of the hydrolytic un-
stability, resulting in collapsing of silicate walls of MMM
and MCM-41, has been addressed in few papers (28, 30,
51–53). XRD patterns (Fig. 5) and N2 adsorption data for
hydrated Ti-MMM samples (Table 1) show that after treat-
ment with water the structure of Ti-MMM samples is con-
siderably damaged. However, this process seems to be less
pronounced for the sample with higher titanium content if
one compares intensity values of the [100] reflections. Re-
cently, it has been found that introduction of about 5 wt%
FIG. 4. Oxidation of TMP with H2O2 over Ti-MMM samples with
different Ti loading. Reaction conditions: TMP 0.1 M, Ti 1.3 10 2 mmol,
[TMP]/[H2O2] = 1/3.5, MeCN 3 ml, 80 C.
catalytic systems known to date for TMP oxidation
(5, 9–11).
Structure/Activity and Selectivity Relationships
In order to establish the factors which determine the cata-
lyst activity and selectivity in the reaction studied, we have
tested a series of Ti-MMM samples with different Ti content
(Fig. 4). In all experiments, specified amounts of catalyst
were used in order to introduce 0.013 mmol of Ti to each re-
action. The most active samples were found to be the most
selective samples. Samples 3 and 4 with titanium loading
in the 1.5–2 wt% range exhibited superior catalytic prop-
erties (Fig. 4). A similar trend was previously observed
for cyclohexene oxidation with H2O2 over Ti-MMM (37, 38)
and grafted Ti-MCM-41 (48). It has been reported that
the increase of titanium content in Ti-MMM as well as
Ti-MCM-41 results in the loss of structure regularity
(33, 37, 38, 49) and the formation of less reactive Ti–O–Ti
bonds (16, 33, 48). Both structure perfection of Ti-MMM
catalyst and high dispersion of titanium in the silicate ma-
trix were proposed to be crucial factors determining cata-
lytic behavior (37, 38). The structure regularity seems to be
important from the point of view of titanium accessibility.
The Ti-MMM wall-thickness values calculated as described
in (41, 42) are given in Table 1. Takinginto account that silica
tetrahedron size was estimated to be about 0.4 nm (50),
most of the introduced titanium should be expected to be
accessible to reactants for samples 2–6. With increasing ti-
tanium loading, the wall-thickness also increases according
to (49) and our results (sample 1) and thus a part of titanium
species is expected to be located inside the silica walls. In-
deed, the reaction selectivity to TMBQ was found to be
76 and 39% for the Ti-MMM samples, containing 1.89 and
4.32 wt% of Ti, respectively, while the reaction time was
FIG. 5. The XRD patterns for hydrated and recalcined forms of
increased in the same order (Fig. 4). However, since both sample 1 (H-1), sample 6 (H-6), and pure siliceous MMM.

116
TRUKHAN ET AL.
Al largely enhanced hydrostability of MMM compared to
pure siliceous MMM (51, 52). We suppose that a similar
relationship may exist between titanium content and hy-
drolytic stability of Ti-MMM. However, a combined XRD
and kinetic study is needed to confirm the difference in the
rates of structure damage for Ti-MMM samples with differ-
ent Ti loading under the conditions used for TMP oxidation.
Catalyst Recycling
Ti-MMM catalysts can be easily separated by simple fil-
tration and then reused. We finally investigated the possi-
bility of catalyst recycling using the best Ti-MMM samples
(3 and 4) and found that stability of the catalytic behavior
depends on H2O2 and, therefore, H2O concentration em-
ployed. When using 30% H2O2 , at [H2O2] = 0.175 M, both
the reaction rate and TMBQ yield remained unchanged
during three catalytic cycles, whereas at [H2O2] = 0.35 M,
subsequent activity and selectivity decay was observed
(Table 4). Again, contrary to the Ti-MCM-41/H2O2 system
(28, 33), no titanium leaching occurred from the Ti-MMM
catalysts. The total titanium content in the sample remained
unchanged according to the elemental analysis data. We
suppose that the catalyst deactivation results most likely
from the Ti-MMM structure collapsing during the oxida-
tion process caused by water (vide supra). The structure
collapse, in turn, may result in the blockage of some of
the active titanium sites inside the silica walls as well as
the formation of titanium–oxygen clusters. Indeed, DR-UV
spectra of the Ti-MMM catalyst (sample 3) recorded after
treatment with 30% aqueous H2O2 showed the long-wave
shift of the absorption maximum to 270 nm (Fig. 6). Note-
FIG. 6. UV–Vis diffuse reflectance spectra for Ti-MMM (sample 3):
(1) before treatment; (2) after first treatment with 30% H2O2; (3) after
second treatment with 30% H2O2; (4) after first treatment with 74% H2O2;
Ti-MMM 33 mg, H2O2 0.35 M, MeCN 3 ml, 80 C, 1 h.
worthy, after treatment with 74% H2O2 the spectrum re-
mained practically unchanged (Fig. 6). Since all the samples
were calcined before running UV spectra and the condi-
tions used to record the spectra were identical for the initial
sample and the treated samples, the long-wave shift cannot
be attributed to simple hydration of titanium species and
can be ascribed to the appearance of partially oligomerized
titanium species (14, 16, 27, 33). To verify the hypothesis,
concerning negative effect of water on the catalyst stabil-
ity, we performed the oxidation using 84% H2O2, instead
of 30% H2O2 (0.35 M H2O2 in the reaction mixture in both
cases), and found that the stability of the Ti-MMM cata-
lysts enhanced and they can be reused repeatedly for at
least three catalytic cycles without suffering a loss in activity
and selectivity (Table 4). However, it should be mentioned
that complete TMP conversion was not achieved when us-
ing 1/3.5 TMP/H2O2 molar ration and the reaction rate was
lower for concentrated H2O2 than for dilute one. The latter
fact is consistent with our results obtained earlier for H2O2-
based oxidation in the presence of Ti-containing polyox-
ometalate (54). Thus, water seems to play two contradictory
roles: on one hand, it enhances catalytic activity in the first
catalytic cycle but, on the other hand, it destroys the cata-
lyst structure and thus worsens the catalytic properties in
the following cycles. Therefore, a compromise should be
achieved between the catalyst activity and its hydrolytic
stability.
TABLE 4
The Effect of H2O2 Concentration on TMP Oxidation
over Ti-MMMa
TMP
% H2O2 [H2O2] (M) Cycle Time (h) conversion (% ) TMBQ (% )
Yieldb of
30
30
30
84
0.175
0.35
0.35
0.35
I c
II
III
I c
II
III
I
II
0.4
0.4
0.4
0.4
1.2
1.8
0.6
2.5
5.7
1.7
1.7
1.7
100
100
100
100
99
99
100
97
82
79
79
77
60
47
76
37
32
64
68
66
III
I
II
95
77d
80
CONCLUSIONS
The well-organized Ti-MMM are effective catalysts for
TMP oxidation to TMBQ with aqueous H2O2. High selec-
tivity (up to 86% at 100% TMP conversion) can be attended
at optimal reaction conditions. This is a rare example of
really heterogeneous process, producing TMBQ. The cata-
lysts with titanium content in the 1.5–2 wt% range possess
III
79
a Reaction conditions: [TMP]/[H2O2] = 1/3.5, sample 3, 33 mg, MeCN
3 ml, 80 C.
b GC yield based on TMP consumed.
c Sample 4, 42 mg.
d Complete TMP conversion was not achieved.

OXIDATION OF 2,3,6-TRIMETHYLPHENOL OVER Ti- AND V-MMM
117
the best catalytic properties. No titanium leaching takes 24. Corma, A., Iglesias, M., and Sanchez, F., Catal. Lett. 39, 153 (1996).
25. Zhang, W., and Pinnavaia, T. J., Catal. Lett. 38, 261 (1996).
26. Reddy, J. S., Liu, P., and Sayari, A., Appl. Catal. A: General 148, 7
place during the oxidation process; however, some loss of
the catalytic properties may occur, which is most likely due
to collapsing of the catalyst structure caused by water. The
(1996).
27. Zhang, W., Fro¨ba, M., Wang, J., Tanev, P. T., Wong, J., and Pinnavaia,
catalyst stability increases with reducing concentration of
water in the reaction mixture.
T. J., J. Am. Chem. Soc. 118, 9164 (1996).
28. Rhee, C. H., and Lee, J. S., Catal. Lett. 40, 261 (1996).
29. Sheldon, R., Stud. Surf. Sci. Catal. 110, 151 (1997).
30. Koyano, K. A., and Tatsumi, T., Micropor. Mater. 10, 259 (1997).
31. Kochkar, H., and Figueras, F., J. Catal. 171, 420 (1997).
32. Van der Waal, J. C., Rigutto, M. S., and van Bekkum, H., Appl. Catal.
A: General 167, 331 (1998).
ACKNOWLEDGMENTS
We are thankful to A. Yu. Dereviankin for N adsorption measure-
2
33. Chen, L. Y., Chuah, G. K., and Jaenicke, S., Catal. Lett. 50, 107 (1998).
34. Di Wei, Wei-Te Chueh, and Haller, G. L., Catal. Today 51, 501
(1999).
35. Mahalingam, R. J., Badamali, S. K., and Selvam, P., Chem. Lett. 1141
(1999).
ments, V. A. Utkin for GC-MS analysis, and R. I. Maksimovskaya for
1
H NMR measurements. We also thank A. Tuel for supplying us with the
Ti-HMS sample. The financial support of Russian Foundation for Basic
Research (Grant N 01-03-32852) is highly appreciated.
36. Mahalingam, R. J., and Selvam, P., Chem. Lett. 455 (1999).
37. Kholdeeva, O. A., Derevyankin, A. Yu., Shmakov, A. N., Trukhan,
N. N., Paukshtis, E. A., Tuel, A., and Romannikov, V. N., J. Mol. Catal.
A: Chemical 158, 417 (2000).
38. Trukhan, N. N., Derevyankin, A. Yu., Shmakov, A. N., Paukshtis,
E. A., Kholdeeva, O. A., and Romannikov, V. N., Micropor. Meso-
por. Mater. 44–45, 603 (2001).
REFERENCES
1. Schudel, P., Mayer, H., and Isler, O., in “The Vitamins” (W. H. Sebrell
and R. S. Harris, Eds.), Vol. 5, p. 165. Academic Press, New York, 1972.
2. Yamada, S., Takeshita, T., and Tanaka, J., J. Syn. Org. Chem. Jpn. 40,
268 (1982).
39. Kholdeeva, O. A., Romannikov, V. N., Trukhan, N. N., and Parmon,
V. N., RU Patent Application 2000105325. Priority 03.03.2000.
40. Sorokin, A., and Tuel, A., Catal. Today 57, 45 (2000).
41. Fenelonov, V. N., Romannikov, V. N., and Derevyankin, A. Yu., Mi-
cropor. Mesopor. Mater. 28, 57 (1999).
42. Romannikov, V. N., Fenelonov, V. B., Nosov, A. V., Derevyankin,
A. Yu., Tsybulya, S. V., and Kryukova, G. N., Russ. Chem. Bull. 48,
1821 (1999).
43. Taylor, W. I., “Oxidative Coupling of Phenols.” Dekker, New York,
1967.
44. Gorodetskaya, T. A., Kozhevnikov, I. V., and Matveev, K. I., RU Ap-
plication 1155580 (1985).
3. Bushmilyev, V. A., Kondratyeva, T. A., Lipkin, M. A., and Kondratyev,
A. V., Khim-Farm Zh. 5, 65 (1991). [In Russian]
4. Constantini, M., Krumenaker, L., and Igersheim, F., French Patent
2552425 (1983).
5. Chao-Yang, H., and Lyons, J. E., European Patent 93540 (1983).
6. Tu¨rk, D., European Patent 0 127 888 B1 (1987).
7. Chiba, K., and Fukuoka, N., Japanese Patent 202297 A2 (2000).
8. Tomaja, D. L., Vogt, L. H., and Wirth, J. G., J. Org. Chem. 35, 2029
(1970).
9. Laas, H., Tavs, P., and Hannebaum, H., Dt. Patent 3302498 (1984).
10. Ito, S., Aihara, K., and Matsumoto, M., Tetrahedron Lett. 24, 5249
(1983).
11. Kholdeeva, O. A., Golovin, A. V., Maksimovskaya, R. I., and
Kozhevnikov, I. V., J. Mol. Catal. 75, 235 (1992).
12. Shimizu, M., Hayakawa, T., and Takehira, K., Tetrahedron Lett. 30,
4711 (1989).
13. Sorokin, A., and Tuel, A., New. J. Chem. 23, 473 (1999).
14. Notari, B., Adv. Catal. 41, 253 (1996).
15. Clerici, M. G., and Ingallina, P., J. Catal. 140, 71 (1993).
16. Corma, A., Chem. Rev. 97, 2373 (1997).
17. Schneider, M., and Baiker, A., Catal. Rev. Sci. Eng. 37(4), 515 (1995).
18. Corma, A., Navarro, M. T., Perez-Pariente, J., and Sanchez, F., Stud.
Surf. Sci. Catal. 84, 69 (1994).
19. Tanev, P. T., Chibwe, M., and Pinnavaia, T., Nature 368, 321 (1994).
20. Blasco, T., Corma, A., Navarro, M. T., and Perez-Pariente, J., J. Catal.
156, 65 (1995).
45. McNelis, E., J. Org. Chem. 31, 1255 (1966).
46. Kholdeeva, O. A., Kovaleva, L. A., Maksimovskaya, R. I., and
Maksimov, G. M., J. Mol. Catal. A: Chemical 158, 223 (2000).
47. Kholdeeva, O., Maksimov, G. M., Maksimovskaya, and Kovaleva,
L. A., Kinet. Catal. 42, 217 (2001).
48. Marchese, L., Gianotti, E., Dellarocca, V., Maschmeyer, T., Rey, F.,
Coluccia, S., and Thomas, M., Phys. Chem. Chem. Phys. 1, 585 (1999).
49. Alba, M. D., Becerro, A. I., and Klinowski, J., J. Chem. Soc., Faraday
Trans. 92(5), 849 (1996).
50. Iler, R. K., “The Chemistry of Silica,” Vol. 1. Wiley, New York, 1979.
51. Solovyov, L. A., Kirik, S. D., Shmakov, A. N., and Romannikov, V. N.,
Micropor. Mesopor. Mater. 44–45, 17 (2001).
52. Kodenev, E. G., Shmakov, A. N., Derevyankin, A. Yu., Lapina, O. B.,
and Romannikov, V. N., J. Mol. Catal. A: Chemical 158, 349 (2000).
53. Zhao, X. S., Lu, G. Q., and Hu, X., Micropor. Mesopor. Mater. 41, 37
(2000).
21. Keshavaraja, A., Ramaswamy, V., Soni, H. S., Ramaswamy, A. V., and
Ratnasamy, P., J. Catal. 157, 501 (1995).
54. Kholdeeva, O. A., Maksimov, G. M., Maksimovskaya, R. I., Kovaleva,
L. A., Fedotov, M. A., Grigoriev, V. A., and Hill, C. L., Inorg. Chem.
39, 3828 (2000).
22. Contier, S., and Tuel, A., J. Catal. 157, 124 (1995).
23. Reddy, J. S., and Sayari, A., J. Chem. Soc., Chem. Commun. 2231
(1995).