Surface chemistry and reactivity of well-defined multilayered supported M1Ox/M2Ox/SiO2 catalysts

Article abstract of DOI:10.1016/j.jcat.2008.06.002

A series of group 5-7 transition metal oxides (TMOs) were supported on SiO₂ and surface-modified SiO₂ containing surface AlO_x, ZrO_x, and TiO_x species. The surface reactivity of these silica supported oxides was chemically probed with CH₃OH-temperature-programmed surface reaction (TPSR) spectroscopy. The selectivity of the model supported MO_x catalytic active sites on SiO₂ generally reflect the same product distribution as their corresponding bulk MO_x counterparts toward dimethyl ether (DME), formaldehyde (HCHO) and CO₂ from surface acidic, redox, and basic sites, respectively. The reactivity of the surface MO_x sites generally was suppressed by anchoring of the surface MO_x species onto the SiO₂ support. The general surface chemistry trend followed the known inorganic chemistry of the corresponding bulk MO_x TMOs. For the multilayered supported M₁O_x/M₂O_x/SiO₂, with M₁ representing the group 5-7 TMOs and M₂ representing Al, Zr or Ti, the selectivity of the catalytic active sites was generally comparable to that for the model-supported M₁O_x/SiO₂ catalysts. The reactivity of the surface VO_x, MoO_x, and ReO_x redox sites increased by one to four orders of magnitude with the introduction of the surface modifiers; however, the reactivity of the surface WO_x acidic site was mildly suppressed by the presence of the surface modifiers. The reactivity of the basic CrO_x site was only mildly perturbed by the surface modifiers. The reactivity trend of the catalytic TMOs sites was related to the electronegativity properties of the anchoring substrate cations (Si > Al > Zr ～ Ti).

Full text of DOI:10.1016/j.jcat.2008.06.002

Journal of Catalysis 258 (2008) 103–110
Contents lists available at ScienceDirect
Journal of Catalysis
www.elsevier.com/locate/jcat
Surface chemistry and reactivity of well-defined multilayered supported
M1Ox/M2Ox/SiO2 catalysts
∗
Edward L. Lee, Israel E. Wachs
Operando Molecular Spectroscopy and Catalysis Laboratory, Chemical Engineering Department, 111 Research Drive, Iacocca Hall, Lehigh University, Bethlehem, PA 18015, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of group 5–7 transition metal oxides (TMOs) were supported on SiO2 and surface-modified SiO2
containing surface AlOx, ZrOx, and TiOx species. The surface reactivity of these silica supported oxides
was chemically probed with CH3OH-temperature-programmed surface reaction (TPSR) spectroscopy. The
selectivity of the model supported MOx catalytic active sites on SiO2 generally reflect the same product
distribution as their corresponding bulk MOx counterparts toward dimethyl ether (DME), formaldehyde
(HCHO) and CO2 from surface acidic, redox, and basic sites, respectively. The reactivity of the surface
MOx sites generally was suppressed by anchoring of the surface MOx species onto the SiO2 support.
The general surface chemistry trend followed the known inorganic chemistry of the corresponding bulk
MOx TMOs. For the multilayered supported M1Ox/M2Ox/SiO2, with M1 representing the group 5–7 TMOs
and M2 representing Al, Zr or Ti, the selectivity of the catalytic active sites was generally comparable
to that for the model-supported M1Ox/SiO2 catalysts. The reactivity of the surface VOx, MoOx, and ReOx
redox sites increased by one to four orders of magnitude with the introduction of the surface modifiers;
however, the reactivity of the surface WOx acidic site was mildly suppressed by the presence of the
surface modifiers. The reactivity of the basic CrOx site was only mildly perturbed by the surface modifiers.
The reactivity trend of the catalytic TMOs sites was related to the electronegativity properties of the
anchoring substrate cations (Si > Al > Zr ∼ Ti).
Received 19 March 2008
Revised 30 May 2008
Accepted 3 June 2008
Available online 17 July 2008
Keywords:
Catalysts
Supported
SiO2
Oxides
V2O5
CrO3
MoO3
WO3
Re2O7
Reactivity
Kinetics
© 2008 Elsevier Inc. All rights reserved.
Temperature programmed surface reaction
TPSR
CH3OH
HCHO
CH3OCH3
CO
CO2
1. Introduction
metal oxide catalysts. The literature reports various industrial
applications for these supported catalysts, including ammoxida-
tion of 3-picoline (V2O5/Al2O3–SiO2) [4], selective catalytic re-
duction (SCR) of nitrogen oxide with ammonia and NO reduc-
tion with CO (V2O5/TiO2/SiO2) [3,5–11], ethylene polymerization
(CrO3/TiO2–SiO2) [12,13], denitrogenation of nitrogen containing
heteroaromatic compounds (Mo/Al2O3–SiO2) [14,15], HDS of thio-
phene (Mo/ZrO2–SiO2 and MoO3/TiO2–SiO2) [16–18], and alkene
metathesis and epoxidation with H2O2 (Re2O7/Al2O3–SiO2) [19,20].
These promoted mixed-oxide support materials have been found
to have more favorable catalytic properties than the more conven-
tional supported MOx/SiO2 catalysts.
The molecular and electronic structures of the supported group
5–7 transition metal oxides on SiO2 and surface modified SiO2
(where surface modification is achieved with surface AlOx, ZrOx,
and TiOx species) have been successfully determined [21–23].
The group 5 metal oxides maintain the monoxo surface structure
(M=O) for both the native SiO2 and the surface modified SiO2
supports. The surface monoxo (M=O)/dioxo (M(=O)2) ratio of the
Various promoters or additives are generally added to the
supported MOx/SiO2 catalyst systems to enhance their catalytic
performance (e.g., enhanced activity, improved selectivity, ther-
mal stability). Some of the typical additives are oxides of AlOx,
ZrOx, and TiOx. The interaction between Al2O3–SiO2 generates
new Brønsted acidic sites at bridging Al–OH–Si bond [1]. The
ZrO2–SiO2 interaction also results in enhanced surface acid-
ity, as well as excellent chemical resistance to alkaline corro-
sion and low thermal expansion [2], and the interaction be-
tween TiO2–SiO2 also yields high thermal stability, excellent me-
chanical strength, and generation of new catalytic active acidic
sites [3]. These surface-modified SiO2 mixed oxides are further
used as oxide supports for the supported group 5–7 transition
Corresponding author. Fax: +1 (610) 758 6555.
*
E-mail address: iew0@lehigh.edu (I.E. Wachs).
0021-9517/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2008.06.002

104
E.L. Lee, I.E. Wachs / Journal of Catalysis 258 (2008) 103–110
group 6 metal oxides (CrOx, MoOx, and WOx) is controlled by the
surface modifiers, however. The group 7 metal oxide of surface
ReO4 maintains the trioxo structure (Re(=O)3) in the presence and
absence of the surface modifiers. The promoters also have a sig-
nificant effect on the corresponding electronic structure and result
in a narrower distribution of isolated surface metal oxide species,
especially for the group 6 metal oxides. These structural changes
in the surface species also demonstrate that the surface group 5–7
transition metal oxides preferentially anchor to the surface modi-
fiers (AlOx, ZrOx, and TiOx) over the exposed SiO2 support sites.
The objective of this investigation was to determine the influ-
ence of the surface modifiers on the molecular/electronic structure-
activity/selectivity relationships for the multilayered SiO2-sup-
ported group 5–7 metal oxide catalysts. Surface AlOx, ZrOx, and
TiOx species were used as the surface modifiers in this study.
The surface chemistry and reactivity was chemically probed with
CH3OH-temperature-programmed surface reaction (TPSR) spec-
troscopy. The CH3OH-TPSR experiments provide information about
the nature of the catalytic active sites (redox, acidic, or basic)
and their specific surface reactivity (krds) toward methanol, in
which krds represents the first-order kinetic constant of the rate-
determining-step (RDS) [24–27]. The catalytic activities of the
model supported MOx/SiO2 catalyst systems were assessed and
used as benchmarks for the multilayered supported metal oxide
catalysts.
V2O5/SiO2, in keeping with earlier studies), henceforth designated
MOx/SiO2.
The multilayered catalysts were synthesized similarly to the
model catalysts; however, the incipient wetness impregnation of
the group 5–7 transition metal oxide overlayer was added to an
existing calcined alumina-, titania-, and zirconia-surface modified
SiO2 support. The calcination procedures followed that for the
model catalysts for the corresponding systems. The CrO3, MoO3,
WO3, and Re2O7 on the surface modified SiO2 supports were syn-
thesized at 3 wt% (with V2O5 synthesized at 5 wt%, in keeping
with earlier studies) unless specified otherwise and are designated
as M1Ox/M2Ox/SiO2 (e.g., CrO3/Al2O3/SiO2).
2.2. CH3OH-TPSR spectroscopy
The surface reactivity of the model supported MOx/SiO2 and
multilayered supported M1Ox/M2Ox/SiO2 catalysts was determined
by CH3OH-TPSR spectroscopy. CH3OH is a “smart” chemical probe
molecule that distinguishes among surface acidic, redox, and ba-
sic sites [24]. The dissociative chemisorption of methanol forms
∗
surface methoxy (CH3O ) intermediate species, the most abundant
reaction intermediate (MARI) [31,32]. The surface methoxy inter-
mediate undergoes different reaction pathways that depend the
nature of the catalytic active site: formaldehyde (HCHO) from re-
dox sites, dimethyl ether (CH3OCH3, DME) from acidic sites, and
CO/CO2 from basic sites [24,33]. No additional reaction products,
such as methyl formate (MF) and dimethoxy methane (DMM),
were found in the present study. Formation of H2O always accom-
panies the CH3OH surface chemistry, but this was not evaluated in
the study because it does not provide any additional insight. In ad-
dition, the formation of CO was not explicitly investigated because
of its similarity to the CO2 combustion product.
2. Experimental
2.1. Catalyst synthesis
The SiO2-supported catalysts consist of highly dispersed metal
oxides (Al2O3, TiO2, ZrO2, V2O5, CrO3, MoO3, WO3, and Re2O7) that
were successfully prepared by incipient wetness impregnation, as
described in detail elsewhere [21,28–30]. The SiO2 support mate-
rial, amorphous SiO2 (Cabot, Cab–O–Sil fumed silica EH-5, S.A. =
332 m2/g), was found to be more easily handled by an initial wa-
The CH3OH-TPSR spectra were obtained with a commercial
TPSR system (Altamira Instruments, AMI-200) equipped with an
on-line quadrupole mass spectrometer (Ametek Dycor Dymaxion
with Dycor System 2000 software). The catalyst samples, typically
200–300 mg of loose powder, were placed in a quartz bubble U-
tube and held in place by glass wool. The U-tube was placed in a
◦
ter pretreatment and calcination at 500 C for 4 h with no change
in the material properties. The SiO2 support was impregnated with
aqueous and nonaqueous (toluene) solutions of the corresponding
precursors: aluminum sec-butoxide (Al[O(CH3)CH2H5]3, Alfa Aesar,
95%), titanium isopropoxide (Ti[OCH(CH3)2]4, Alfa-Aesar, 99.999%),
zirconium tert-butoxide (Zr[OC(CH3)3]4, Alfa Aesar, 97%), vana-
dium triisopropoxide (VO[CHO(CH3)2]3, Alfa Aesar, 97%), chromium
(III) nitrate (Cr(NO3)3·9H2O, Alfa Aesar, 98.5%), ammonium hep-
tamolybdate ((NH4)6Mo7O24·4H2O, Aldrich, 99.98%), ammonium
metatungstate ((NH4)6H2W12O40·xH2O, Pfaltz and Bauer, 99.5%),
or perrhenic acid (HReO4, Alfa Aesar, 75–80%). The SiO2 was ini-
◦
clamshell furnace capable of linear heating rates from 1-30 C/min
◦
up to 1200 C, with the temperatures measured by thermocouples
both at the top of the sample bed and at the center of the fur-
nace. The gas flow, accurately metered by mass flow controllers
(Brooks, model 5850E series), and the temperature setpoints were
fully computer-automated (controlled by a LabVIEW-based appli-
cation software). The exhaust line from the U-tube reactor to the
◦
mass spectrometer (MS) was maintained at ∼100 C to prevent
condensation of the methanol and reaction products. The typi-
cal protocol for obtaining the CH3OH-TPSR spectra was as fol-
◦
tially dried for 2 h at 115 C for the nonaqueous preparations
◦
lows. The samples were first pretreated at 450 C under flow-
before synthesis inside a glove box (Vacuum Atmospheres, Omni-
Lab VAC 101965) under a nitrogen environment. After impregna-
tion, the samples were allowed to dry overnight under the nitro-
gen atmosphere. Calcination of the samples entailed ramping at
ing air (Airgas, Zero grade) for 1 h at 30 sccm, then cooled to
◦
110 C to prevent moisture condensation. The flowing gas was
switched to helium (Airgas, UHP) as the samples were allowed
◦
◦
◦
to cool further to 100 C for 30 min to remove any physically
1 C/min to 110 C and holding for 5 h under flowing N2 (Air-
adsorbed oxygen. Then a CH3OH/He (Airgas, Certified, 2000 ppm
CH3OH/He) gas mixture was introduced and chemisorbed at 100 C
gas, ultra-high purity) in a programmable furnace (Thermolyne
◦
◦
model 48000), followed by another 1 C/min ramp under flow-
◦
◦
ing air (Airgas, Zero grade) to 500 C (with 450 C used for the
V2O5/SiO2, in keeping with earlier studies) and holding for 6 h.
The procedure for the aqueous preparations was the same as for
the nonaqueous preparations, except that the drying and initial
calcination steps were performed in ambient air and under flow-
ing air (Airgas, zero grade), respectively. The surface modified SiO2
supports, Al2O3/SiO2, TiO2/SiO2, and ZrO2/SiO2, were synthesized
with 5 wt% metal oxide loadings on SiO2. The model catalysts,
V2O5/SiO2, CrO3/SiO2, MoO3/SiO2, WO3/SiO2, and Re2O7/SiO2, were
synthesized with 3 wt% metal oxides on SiO2 (with 5 wt% used for
(30 sccm) for 30–45 min, depending on the catalyst weight [24,
34]. A methanol breakthrough curve was obtained with the MS for
each run to ensure saturation of the catalyst surface. The catalyst
samples were purged with helium for 1 h to remove physically ad-
sorbed methanol, and then the sample temperature was ramped
◦
◦
at 10 C/min from room temperature to 500 C under either he-
lium or blended 1% O2/He (Airgas, Certified, 9.735% O2/He) gas.
The 1% O2/He carrier flow was used for the easily reducible sup-
ported metal oxide systems because reduction produces a second,
high-temperature TPSR peak from the reduced sites, and the 1%

E.L. Lee, I.E. Wachs / Journal of Catalysis 258 (2008) 103–110
105
Fig. 1. CH3OH-TPSR relative selectivity of supported MOx/SiO2 (left) at maximum attainable surface metal oxide coverage, without the presence of crystalline MOx NPs, and
their corresponding bulk MOx (right). The volatile nature of bulk Re2O7 prevented the study of this unsupported MOx.
O2/He mixture prevents surface reduction and provides MS spectra
for the fully oxidized surface metal oxides.
3. Results
−
The m/e values used to detect the various TPSR products were
3.1. Model supported MOx/SiO2 catalysts
−
−
m/e = 30 (primary) and 29 (secondary) for HCHO, m/e = 45
(primary) and 46 (secondary) for CH3OCH3 (DME), m/e− = 31 for
The CH3OH-TPSR spectrum of the pure SiO2 reflects the rela-
tively low reactivity of the native SiO2 support and the absence
of any significant surface redox (HCHO formation), acidic (DME
formation), and basic (CO2 formation) sites (see Supporting Infor-
mation, Fig. S1). The model SiO2-supported metal oxide catalyst
samples used in the present study (AlOx/SiO2, TiOx/SiO2, ZrOx/SiO2,
VOx/SiO2, NbOx/SiO2, TaOx/SiO2, CrOx/SiO2, MoOx/SiO2, WOx/SiO2,
and ReOx/SiO2) were previously shown to be 100% dispersed as
surface oxides on the SiO2 support by Raman, IR, UV–vis, XANES,
and solid-state 51V and 27Al NMR spectroscopy [21–23,38–52]. The
CH3OH-TPSR spectra of the model SiO2 catalyst systems reveal
that adding the surface metal oxides introduced surface redox,
acidic, and basic catalytic active sites (see Supporting Information,
Figs. S2–S11). The peak temperature, Tp, and krds of the model sup-
ported MOx/SiO2 catalytic systems for the formation of DME (see
Table 1), HCHO, and CO2 (see Table 2), along with the relative peak
area ratios of HCHO:DME:CO2 (see Table 3) for each system, are
summarized below. The selectivity and krds information are plot-
ted in Figs. 1 and 2, to facilitate visual inspection of the data. The
corresponding Eact values are tabulated elsewhere (see Supporting
Information, Table S1). In general, the CH3OH-TPSR spectral inten-
sity of the reaction products for each supported MOx/SiO2 system
increased with surface MOx coverage and decreased slightly in Tp
with surface coverage. The increase in MS signal intensity (area
under the TPSR curve) was proportional to the number of catalytic
active sites, and the slight drop in Tp values was related to the
decreased concentration of the less reactive surface Si–OCH3 inter-
mediates with increasing surface MOx coverage. A decrease in the
−
−
CH3OH, m/e = 28 for CO, and m/e = 44 for CO2. In addition,
◦
reaction-limited CH3OH also was formed at Tp ∼ 170–200 C from
∗
∗
the recombination of the surface CH3O and H species. The low
Tp value indicates the ease at which the recombination occurred,
whereas previous transient kinetic isotopic studies with CH3OH
and CD3OD revealed the efficient formation of CH3OD and CD3OH
species [35]. The contribution of the HCHO cracking fraction of
CH3OH was carefully subtracted out of the apparent HCHO MS sig-
nal when it overlapped with the true HCHO signal. The relative
selectivity toward each desorption product was determined from
the areas under the TPSR curves using a Lorentzian fit.
The surface kinetic parameters (Eact and krds) for the surface
methoxy reactions to HCHO, CH3OCH3, and CO/CO2 were obtained
directly from the CH3OH-TPSR spectra. The RDS for the unimolec-
∗
ular surface CH3O dehydrogenation to HCHO involves breaking
the surface methoxy C–H bond [33]. The RDS for CH3OCH3 for-
mation involves unimolecular surface methoxy C–O bond scission
[26,36]. Formation of CO/CO2 proceeds through conversion of the
∗
∗
surface CH3O to surface formate (HCOO ), and the RDS involves
breaking either the surface formate C–H or C–O bond unimolecular
reactions [25]. The unimolecular aspect of the RDS for the differ-
ent reaction pathways allows application of the first-order Redhead
equation [37] to determine the Eact for the surface reactions,
ꢀ
ꢁ
ꢀ
ꢁ
Eact
RTp2
ν
β
−Eact
=
exp
,
(1)
RTp
in which Tp is the CH3OH-TPSR peak temperature of the reaction
Tp corresponds to a decrease in Eact and an increase in krds
.
−1
product, R is the gas constant (1.987 cal/mol K), ν = 1013
s
for
Figs. 1 and 2 also compare the selectivity and activity from
the CH3OH-TPSR studies of the model supported MOx/SiO2 cat-
alysts with those of their corresponding unsupported, bulk MOx
catalysts. The selectivity patterns of the surface MOx catalytic ac-
tive sites on SiO2 were comparable to those of their corresponding
bulk MOx oxides, with one exception: The supported TiO2/SiO2 cat-
alyst system exhibited significant redox character in addition to
DME formation, whereas the bulk TiO2 system yielded DME ex-
clusively. With few exceptions, the bulk MOx catalysts tended to
be more active than their corresponding SiO2-supported surface
MOx species. The difference reflects the changes associated with
◦
first-order kinetics [34], and β is the heating rate (10 C/min). The
RDS of the surface methoxy intermediate conversion to DME and
HCHO, krds, is a function of Eact (and hence Tp) and is determined
by
ꢀ
ꢁ
−Eact
krds = v exp
,
(2)
RT
in which T is a reference temperature used for the comparison of
krds values. In keeping with previous investigations, this study also
◦
used T = 230 C as the reference temperature.

106
E.L. Lee, I.E. Wachs / Journal of Catalysis 258 (2008) 103–110
◦
Fig. 2. krds of HCHO, DME and CO2 of MOx/SiO2 at 230 C for low and high (maximum dispersion) metal oxide coverage. The krds values of bulk oxides are shown for
reference with notation as stars [24–27].
anchoring the isolated surface MOx species on the SiO2 surface.
In general, the SiO2-supported MOx species had comparable selec-
tivity to and slightly lower surface reactivity than their bulk MOx
counterparts.
Table 1
CH3OH-TPSR surface kinetics (krds) of model supported MOx/SiO2 catalysts (M = Al,
Ti, Zr, V, Nb, Ta, Cr, Mo, W, and Re) for surface methoxy dehydration to DME
Wt Tp
krds (DME) Tp
krds (DME) Tp
krds (DME) Tp
krds (DME)
−1
◦
−1
◦
−1
◦
−1
◦
(%) ( C) (s
)
( C) (s
)
( C) (s
)
( C) (s
)
The krds (DME) trends for the model supported MOx/SiO2 metal
oxides at low and high maximum dispersion as a function of the
metal oxide oxidation state or group number also are shown in
Fig. 2. The supported Al2O3/SiO2, CrO3/SiO2, and WO3/SiO2 sys-
tems had the fastest surface kinetics for DME formation and in-
cluded the most active surface acid sites. The supported TiO2/SiO2,
ZrO2/SiO2, and Re2O7/SiO2 catalysts were two to four orders of
magnitude slower in krds for DME formation; therefore, the krds
values indicate the following periodic group trend of the metal
oxide cation oxidation state: (+3) > (+6) > (+5) > (+4) ∼ (+7).
Each group in this trend was approximately one order of magni-
tude greater than the next, starting with the fastest krds (DME) for
supported Al2O3/SiO2. The krds values also were generally higher
at higher metal oxide loading, and the periodic group trend held
true for all loading coverages. The krds (DME) values for the bulk
oxides of Al2O3, TiO2, ZrO2, Nb2O5, Ta2O5, Cr2O3, and WO3, indi-
cated by the star symbol in Fig. 2 were slightly greater (within one
order of magnitude) than their corresponding SiO2-supported ox-
ides [24]. Only the bulk V2O5 and MoO3 bulk metal oxides did not
exhibit activity for DME formation during CH3OH-TPSR.
The krds trends for the formation of HCHO of the model
MOx/SiO2 systems also are shown in Fig. 2 for the same surface
metal oxide coverages. The supported CrO3/SiO2 system had the
most active surface redox catalytic sites, and its krds value was
approximately three orders of magnitude greater than those for
the supported V2O5/SiO2, Re2O7/SiO2, and MoO3/SiO2 systems. The
bulk oxides of ZrO2, V2O5, Nb2O5, Ta2O5, Cr2O3, and MoO3 and
their krds (HCHO) values, indicated by the star symbol in Fig. 2,
generally were several orders of magnitude greater than their cor-
responding SiO2-supported oxide catalysts. Only the bulk Al2O3,
TiO2, and WO3 systems did not give rise to HCHO formation dur-
ing CH3OH-TPSR. The least active redox catalytic sites were found
for the supported TiO2/SiO2, ZrO2/SiO2, and Nb2O5/SiO2 catalysts,
which were another order of magnitude lower in krds (HCHO).
Only the supported V2O5/SiO2, CrO3/SiO2, and Re2O7/SiO2 cata-
lysts gave rise to CO2 production (see Figs. 1 and 2). Although CO2
formation is reflective of basic catalytic active sites, CO2 formation
also can arise from decomposition of surface Si–OCH3 interme-
Al2O3/SiO2
1
3
5
270 6.82E−4
248 3.21E−3
241 5.26E−3
234 8.60E−3
10
TiO2/SiO2
V2O5/SiO2
CrO3/SiO2
1
3
5
8
10
12
377 3.56E−7
340 4.88E−6
330 9.89E−6
278 3.88E−4
278 3.88E−4
372 5.07E−7
364 8.93E−7
310 4.06E−5
360 1.19E−6
ZrO2/SiO2
285 2.37E−4
Nb2O5/SiO2
MoO3/SiO2
1
5
8
10
15
380 2.88E−7
378 3.31E−7
357 1.47E−6
360 1.19E−6
357 1.47E−6
328 1.14E−5
326 1.31E−5
324 1.51E−5
320 2.00E−5
–
–
–
–
Ta2O5/SiO2
WO3/SiO2
Re2O7/SiO2
1
3
5
6
8
330 9.89E−6
330 9.89E−6
308 4.68E−5
295 1.17E−4
300 8.23E−5
–
–
380 2.88E−7
330 9.89E−6
366 7.75E−7
324 1.51E−5
324 1.51E−5
10
diates and readsorption/oxidation of the HCHO reaction product.
Note that these supported catalysts also have active redox sites for
HCHO production. We do not discuss CO2 formation further here,
because of the uncertain nature of the origin of CO2 production.
3.2. Multilayered supported M1Ox/M2Ox/SiO2 catalysts
The surface chemistry and kinetics of the group 5–7 surface
metal oxides (VOx
titania-, and zirconia-surface modified SiO2 supports were also
chemically probed with CH3OH-TPSR. The resulting catalytic activ-
ity and selectivity patterns are shown in Figs. 3 and 4, respectively,
and the krds values are listed in Table 4 (also see Supporting Infor-
, CrOx, MoOx, WOx, and ReOx) on the alumina-,

E.L. Lee, I.E. Wachs / Journal of Catalysis 258 (2008) 103–110
107
Fig. 3. krds of supported 3%M1Ox/5%M2Ox/SiO2 (M1 = V, Cr, Mo, W, and Re; M2 = Al, Zr, and Ti, with exception of V2O5 at 5%) for surface methoxy decomposition to DME,
HCHO, and CO2.
Table 2
CH3OH-TPSR surface kinetics (krds) of model supported MOx/SiO2 catalysts (M = Al, Ti, Zr, V, Nb, Ta, Cr, Mo, W, and Re) for surface methoxy dehydrogenation to HCHO (left
columns) and formation of CO2 (right column)
Wt
(%)
Tp
( C)
krds (HCHO)
−1
Tp
( C)
krds (HCHO)
−1
Tp
( C)
krds (HCHO)
−1
Tp
( C)
krds (HCHO)
−1
Tp
( C)
krds (CO2)
−1
◦
◦
◦
◦
◦
(s
)
(s
)
(s
)
(s
)
(s
)
a
Al2O3/SiO2
1
3
5
–
–
–
–
–
–
–
–
10
a
TiO2/SiO2
V2O5/SiO2
345
CrO3/SiO2
V2O5/SiO2
370
1
3
5
400
6.98E−8
3.42E−6
9.89E−6
200
183
9.38E−2
3.82E−1
5.84E−7
2.40E−6
418
386
1.95E−8
1.88E−7
330
350
8
10
12
305
292
5.78E−5
1.45E−4
346
344
3.19E−6
3.68E−6
386
1.88E−7
a
a
a
ZrO2/SiO2
Nb2O5/SiO2
460
MoO3/SiO2
CrO3/SiO2
1
3
5
–
–
9.87E−10
333
350
8.00E−6
2.40E−6
274
280
5.15E−4
3.37E−4
367
356
370
370
7.22E−7
1.57E−6
5.84E−7
5.84E−7
455
445
425
1.41E−9
2.86E−9
1.18E−8
8
10
15
350
2.40E−6
a
a
Ta2O5/SiO2
WO3/SiO2
Re2O7/SiO2
Re2O7/SiO2
1
3
5
8
10
–
–
–
–
–
–
–
–
–
–
330
355
338
9.89E−6
1.69E−6
5.62E−6
386
380
370
1.88E−7
2.88E−7
5.84E−7
338
340
5.62E−6
4.88E−6
a
◦
Denotes catalytic system with no formation of CO2 product below 500 C.
mation Figs. S12–S16 for CH3OH-TPSR spectra and Tp values and
Table S2 for Eact values). With the exception of the surface AlOx
modifier, which dominates the acidic character of the catalysts, the
selectivity of the multilayered catalysts are mainly comparable to
the model supported MOx/SiO2 catalysts without the surface mod-
ifiers. The presence of the surface modifiers generally enhanced
the activity of the surface VOx, MoOx, and ReOx redox sites and
depressed the activity of surface WOx acid sites. The DME forma-
tion for multilayered catalysts with surface AlOx modifiers was not
taken into consideration because of the significant contribution of
the surface AlOx sites to DME production.
The HCHO formation krds values for the multilayered supported
catalysts, as shown in Fig. 3, exhibited significantly enhanced sur-
face reactivity compared with the model reference supported sys-
tems (represented by the black circles) for the redox supported
V2O5/SiO2, MoO3/SiO2, and Re2O7/SiO2 catalysts. The surface mod-
ifiers increased krds (HCHO) by one to three orders of magnitude,
with the exception of the supported chromia system, which was
not significantly affected and decreased slightly in activity. These
results agree well with findings from previous steady-state CH3OH
oxidation studies in which the turnover frequency (TOF) [53] in-
creased by more than an order of magnitude with addition of
the surface modifiers for HCHO formation [1–3,54]. CH3OH oxi-

108
E.L. Lee, I.E. Wachs / Journal of Catalysis 258 (2008) 103–110
dation steady-state reaction studies for supported CrO3/TiOx/SiO2
confirmed that the TOF and selectivity toward HCHO did not differ
appreciably from those for the model CrO3/SiO2 system [55]. The
enhanced redox activity did not originate from the less active sur-
face titania or zirconia sites, because these formed HCHO at higher
addition of the surface modifiers. The presence of the very active
surface AlOx sites appeared to dominate the overall acidity; conse-
quently, the AlOx-containing multilayered catalyst systems cannot
be considered in any analysis. In contrast, the supported ZrOx and
TiOx sites exhibited surface acidic sites that were relatively un-
reactive and produced DME only at elevated temperatures, above
those seen for the group 5–7 supported metal oxides. The surface
modifiers dramatically enhanced the DME formation kinetics of the
◦
temperatures (Tp > 350 C). The surface modifiers did not appre-
ciably affect the redox character of the supported tungsta systems,
because they generally did not yield HCHO, with the exception of
the supported WO3/TiOx/SiO2 catalyst, which created weak redox
◦
◦
both surface VOx (Tp from ∼330–340 C to ∼235–250 C with krds
◦
surface sites at a high Tp ∼ 436 C.
Table 3
The DME formation krds for the multilayered supported metal
HCHO:DME:CO2 ratios for model supported MOx/SiO2 catalysts
oxide catalysts, as shown in Fig. 3, also were strongly influenced by
Wt (%)
HCHO:DME:CO2 ratio (AHCHO:ADME:ACO
)
2
Al2O3/SiO2
1
3
5
0:1:0
0:1:0
0:1:0
0:1:0
10
TiO2/SiO2
4:1:0
V2O5/SiO2
8:1:27
CrO3/SiO2
1
3
3:1:45
2:1:60
5
8
3:1:0
2:1:0
20:1:35
10
12
25:1:37
24:1:33
2:1:0
ZrO2/SiO2
0:1:0
Nb2O5/SiO2
0.20:1:0
MoO3/SiO2
1
3
1:0:0
1:0:0
5
8
10
15
0.20:1:0
0.20:1:0
0.20:1:0
0.17:1:0
0.13:1:0
0.08:1:0
0.08:1:0
1:0:0
Ta2O5/SiO2
0:1:0
WO3/SiO2
Re2O7/SiO2
1
3
5
0:1:0
0:1:0
0:1:0
0:1:0
1:0:9
13:1:98
12:1:90
0:1:0
Fig. 4. CH3OH-TPSR relative selectivity of supported 3%M1Ox/5%M2Ox/SiO2 (except
V2O5 with 5%) for multilayered catalysts. M1 represent the group 5–7 transition
metal oxides and M2 represents surface AlOx, TiOx and ZrOx. The column labeled
none indicates that no surface M2Ox species have been added.
6
8
10
0.07:1:0
0.07:1:0
Table 4
CH3OH-TPSR surface kinetics (krds) of supported 3%M1Ox/5%M2Ox/SiO2 (M1 = V, Cr, Mo, W, and Re; M2 = Al, Zr, and Ti, with exception of V2O5 at 5%) for surface methoxy
decomposition to DME, HCHO, and CO2. The peak area ratio of HCHO:DME for each system is also listed
M1Ox/M2Ox/SiO2 multilayered catalyst systems
Tp
( C)
krds
(s
Tp
( C)
krds
(s
Tp
( C)
krds
(s
Tp
( C)
krds
(s
Tp
( C)
krds
◦
−1
◦
−1
◦
−1
◦
−1
◦
−1
(s )
)
)
)
)
M2Ox
DME
M1Ox = V2O5
CrO3
MoO3
WO3
Re2O7
None
AlOx
ZrOx
TiOx
340
240
250
246
4.88E−6
5.64E−3
2.79E−3
3.70E−3
278
270
280
277
3.88E−4
6.82E−4
3.37E−4
4.17E−4
–
–
308
236
315
347
4.68E−5
7.47E−3
2.85E−5
2.97E−6
380
240
242
312
2.88E−7
5.64E−3
4.90E−3
3.53E−5
246
300
320
3.70E−3
8.23E−5
2.00E−5
HCHO
None
AlOx
ZrOx
TiOx
330
240
235
240
9.89E−6
5.64E−3
8.02E−3
5.64E−3
183
205
200
202
3.82E−1
6.60E−2
9.38E−2
8.15E−2
350
264
278
230
2.40E−6
1.04E−3
3.88E−4
1.14E−2
–
–
–
436
–
–
–
355
314
260
278
1.69E−6
3.06E−5
1.38E−3
3.88E−4
5.43E−9
CO2
None
AlOx
ZrOx
TiOx
350
280
262
262
2.40E−6
3.37E−4
1.20E−3
1.20E−3
280
270
287
280
3.37E−4
6.82E−4
2.06E−4
3.37E−4
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
380
334
286
308
2.88E−7
7.45E−6
2.21E−4
4.68E−5
HCHO:DME:CO2 ratio (AHCHO:ADME:ACO
)
2
None
AlOx
ZrOx
TiOx
20:1:58
2:1:18
18:1:150
7:1:54
2:1:42
2:1:60
2:1:70
2:1:90
1:0:0
0.5:1:0
2:1:0
4:1:0
0:1:0
0:1:0
0:1:0
2:1:0
13:1:82
0.33:1:1
1:1:10
8:1:52

E.L. Lee, I.E. Wachs / Journal of Catalysis 258 (2008) 103–110
109
−6
−3
increasing from ∼10
to ∼10
s
−1) and surface ReOx (Tp de-
bulk M1Ox transition metal oxides. Although some contribution
may come from the different molecular structures of the silica-
supported M1Ox species compared with their unsupported M1Ox
counterparts, the major effect on the surface M1Ox species ap-
pears to be their coordination to the silica support, which has a
high cation electronegativity [57].
◦
−7
creases from ∼380 to ∼240 C and krds increasing from ∼10
to
−3
∼10
s
−1) species. For the supported MoOx systems, the surface
modifiers created surface acidic character that were absent on the
model system, possibly originating from the surface ZrOx and TiOx
sites. The surface modifiers did not appear to appreciably affect
the acidic behavior of the supported chromia and tungsta systems.
CH3OH oxidation steady-state reaction studies for CrO3/TiOx/SiO2
confirmed that the TOF and selectivity toward DME were not ap-
preciably different from those for the model supported CrO3/SiO2
catalyst [55].
4.2. Multilayered supported M1Ox/M2Ox/SiO2 catalysts
The addition of the surface modifiers in the multilayered sup-
ported metal oxide catalysts generally did not significantly al-
ter the CH3OH-TPSR product selectivity (see Fig. 4). The acidic
supported WO3/TiO2/SiO2 catalyst exhibited some redox charac-
ter with the addition of surface TiOx, and the redox supported
MoO3/ZrO2/SiO2 and MoO3/TiO2/SiO2 catalyst systems exhibited
some acidic character with the addition of surface ZrOx and TiOx.
Consequently, the surface M1Ox catalytic active sites in the multi-
layered supported M1Ox/M2Ox/SiO2 catalysts were found to mostly
retain their intrinsic chemical properties, because the product se-
lectivity demonstrated no appreciable change from the introduc-
tion of the surface modifiers.
The introduction of surface AlOx, TiOx, and ZrOx modifiers
onto the SiO2-support dramatically enhanced the krds redox values
(by as much as a factor of ∼104) of the supported M1Ox cat-
alytic active sites. All of the multilayered supported VOx/M2Ox/SiO2
and ReOx/M2Ox/SiO2 catalyst systems contained the same sur-
face monoxo VO4 and trioxo ReO4 structures, respectively. Thus,
the enhanced redox reactivity cannot be related to local molec-
ular structural changes, but instead must be a consequence of
the ligand effect of the different substrate cations (Si, Al, Ti, and
Zr). We propose that the enhanced redox activity is related to
the lower electronegativity of substrate cations, Si > Al > Ti ∼ Zr,
which inversely affects the electron density on the bridging V–O-
support bond. Several recent theoretical DFT/ab initio calculations
have concluded that for supported vanadia catalytic systems, the
bridging V–O-support bond is the most energetically favorable
methanol chemisorption site on the surface VOx structure [51,58–
60]. The inverse surface reactivity trend with cation electronegativ-
ity also was seen for the acidic sites of the multilayered supported
WO3/M2Ox/SiO2 catalyst systems. For the acidic surface WOx cat-
alytic active sites, the fewer the number of electronegative cations
introduced, the greater the resulting electron density on the bridg-
ing W–O-support bonds, which appeared to mildly suppress their
acidic activity rather than produce catalytic enhancement. Similarly
for the basic multilayered supported CrO3/M2Ox/SiO2 catalysts, in-
troduction of the less electronegative surface Al, Zr, and Ti cations
had only a minor effect on the overall reactivity. Thus, the sur-
face AlOx, ZrOx, and TiOx modifiers appear to have the most pro-
nounced catalytic enhancing effect on the surface redox catalytic
active sites on SiO2.
4. Discussion
4.1. Model supported M1Ox/SiO2 catalysts
The selectivity pattern of the model silica-supported surface
M1Ox catalytic active sites during CH3OH-TPSR was similar to that
of the corresponding unsupported, bulk M1Ox transition metal ox-
ides (see Fig. 1). This similarity indicates that the intrinsic redox,
acidic, and basic characteristics of each transition metal oxide were
retained when anchored onto the SiO2 surface. One exception to
this trend is the supported TiOx species, which exhibited signifi-
cant redox character compared with bulk TiO2, dominated by its
acidic character. Additional characterization studies are needed to
fully understand the origin of the enhanced redox characteristics
of the surface TiOx species on SiO2.
The selectivity trend is related to the known inorganic chem-
istry properties of the bulk metal oxides [56]. First, the first-row
transition metal oxides (TMOs) were more easily reducible than
the second- and third-row TMOs [56], which qualitatively suggests
that the first-row transition metal oxides will be more active for
oxidation reactions. As demonstrated earlier, VOx was more active
and more reducible than NbOx and TaOx, and CrOx was more ac-
tive and reducible than both MoOx and WOx. Second, the redox
properties were enhanced with the transition metal oxide oxida-
tion state (+4 < +5 < +6 < +7) [56]. As shown in Fig. 2, the
redox krds values increased as Ti+4 < V+5 < Cr
for the first-row
+6
+4
+5
+6
TMOs, as Zr ∼ Nb < Mo
for the second-row TMOs, and as
+5
+6
+7
Ta < W < Re for the third-row TMOs. These redox reactiv-
ity values for the supported MOx/SiO2 catalysts correlate well with
the periodic trends from inorganic chemistry properties.
Combining the molecular structural information [21] with the
current CH3OH-TPSR chemical probe study findings allows us to
examine possible structure–activity relationships for the model
group 4–7 supported MOx/SiO2 catalysts. It does not appear
that the local molecular structure of the surface MOx species
on SiO2 was the dominant factor in determining the surface re-
dox, acidic, and basic characteristics. For example, although the
group 5 supported metal oxide catalysts (V2O5/SiO2, Nb2O5/SiO2,
Ta2O5/SiO2) exhibited the same surface molecular structure, an
isolated monoxo O=M(–O–Si)3 structure, the surface VOx was
dominated by redox characteristics, whereas the surface NbOx and
TaOx were primarily acidic sites. Similarly, although the group 6
supported metal oxide catalysts (CrO3/SiO2, MoO3/SiO2, and WO3/
SiO2) had the same surface molecular structures of predominantly
isolated dioxo (O=)2M(–O–Si)2 species, the three systems exhib-
ited different reactivity characteristics. Therefore, the local struc-
ture of the dehydrated surface MOx species on SiO2 does not
appear to determine the surface reactivity of the catalytic active
sites.
5. Conclusion
In this work, the surface reactivity of model supported M1Ox/
SiO2 and multilayered supported M1Ox/M2Ox/SiO2 catalysts (where
M1 represents the group 5–7 transition metal oxides and M2
represents Al, Zr, or Ti) were chemically probed by CH3OH-TPSR
spectroscopy. For the model supported M1Ox/SiO2 catalysts, the
krds activity trend was seen to follow the periodic trends for in-
organic chemistry properties; the first-row transition metal ox-
ides were more easily reducible than the second- and third-row
metal oxides, and redox activity increased with oxidation state
(+4 < +5 < +6 < +7). The inverse trend was found to hold for
the acidic activity of the model supported MOx/SiO2 catalysts. For
the multilayered supported M1Ox/M2Ox/SiO2 catalytic systems, the
surface M2Ox modifiers had no significant effect on the selectivity
However, the surface reactivity of the model silica-supported
M1Ox catalytic active sites, as chemically probed by CH3OH-TPSR,
was affected by anchoring of the M1Ox transition metal oxides to
the SiO2 support. The activity of the surface M1Ox catalytic active
sites on SiO2 was generally lower than that of their unsupported,

110
E.L. Lee, I.E. Wachs / Journal of Catalysis 258 (2008) 103–110
of the surface M1Ox catalytic active sites, but the surface modi-
fiers did have a significant effect on the catalytic activity of the
SiO2-supported metal oxide catalysts. The krds activity of the re-
dox surface vanadia, molybdena, and rhenia catalytic active sites
was enhanced by ∼101–104. The presence of the surface modi-
fiers only mildly perturbed the reactivity of the basic surface CrOx
sites, but suppressed the reactivity of the acidic surface WOx cat-
alytic actives sites. In general, the selectivity was controlled by the
intrinsic properties of each surface M1Ox site, and the surface re-
activity was significantly influenced by the anchoring surface M2Ox
ligands. Thus, the catalytic activity was controlled by the M2Ox
ligand according to the electronic requirements of the different re-
action pathways and the electronegativity (Si > Al > Ti ∼ Zr) of the
anchoring sites or ligands.
[17] S. Damyanova, M.A. Centeno, L. Petrov, P. Grange, Spectrochim. Acta
(2001) 2495.
[18] B.M. Reddy, B. Chowdhury, P.G. Smirniotis, Appl. Catal. A 211 (2001) 19.
[19] J.A. Moulijn, J.C. Mol, J. Mol. Catal. 46 (1988) 1.
[20] D. Mandelli, M.C.A. van Vliet, U. Arnold, R.A. Sheldon, U. Schuchardt, J. Mol.
Catal. A 168 (2001) 165.
[21] E.L. Lee, I.E. Wachs, J. Phys. Chem. C 111 (2007) 14410.
[22] E.L. Lee, I.E. Wachs, J. Phys. Chem. C 112 (2008) 6487.
[23] E.L. Lee, I.E. Wachs, J. Phys. Chem. C, submitted for publication.
[24] M. Badlani, I.E. Wachs, Catal. Lett. 75 (2001) 137.
[25] X. Wang, I.E. Wachs, Catal. Today 96 (2004) 211.
[26] T. Kim, A. Burrows, C.J. Kiely, I.E. Wachs, J. Catal. 246 (2007) 370.
[27] I.E. Wachs, J.-M. Jehng, W. Ueda, J. Phys. Chem. B 109 (2005) 2275.
[28] X. Gao, S.R. Bare, J.L.G. Fierro, M.A. Banares, I.E. Wachs, J. Phys. Chem. B 102
(1998) 5653.
A 57
[29] X. Gao, S.R. Bare, B.M. Weckhuysen, I.E. Wachs, J. Phys. Chem. B 102 (1998)
10842.
[30] J.M. Jehng, I.E. Wachs, J. Phys. Chem. 95 (1991) 7373.
[31] L.J. Burcham, L.E. Briand, I.E. Wachs, Langmuir 17 (2001) 6164.
[32] L.J. Burcham, G. Deo, X. Gao, I.E. Wachs, Top. Catal. 11/12 (2000) 85.
[33] J.M. Tatibouet, Appl. Catal. A 148 (1997) 213.
[34] L.E. Briand, W.E. Farneth, I.E. Wachs, J. Catal. 208 (2002) 301.
[35] J.M. Tatibouet, H. Lauron-Pernot, J. Mol. Catal. A 171 (2001) 205.
[36] I.E. Wachs, T. Kim, E.I. Ross, Catal. Today 116 (2006) 162.
[37] P.A. Redhead, Vacuum 12 (1962) 213.
Acknowledgments
Funding for this study was provided by the U.S. Department of
Energy, Basic Energy Sciences (grant DE-FG02-93ER14350).
Supplementary material
[38] J.-M. Jehng, W.-C. Tung, C.-H. Huang, I.E. Wachs, Microporous Mesoporous
Mater. 99 (2007) 299.
The online version of this article contains additional supple-
mentary material: Tables S1–S2 and Figs. S1–S16.
Please visit DOI: 10.1016/j.jcat.2008.06.002.
[39] T. Tanaka, H. Nojima, T. Yamamoto, S. Takenaka, T. Funabiki, S. Yoshida, Phys.
Chem. Chem. Phys. 1 (1999) 5235.
[40] M. Cornac, A. Janin, J.C. Lavalley, Polyhedron 5 (1986) 183.
[41] C.C. Williams, J.G. Ekerdt, J.-M. Jehng, F.D. Hardcastle, A.M. Turek, I.E. Wachs, J.
Phys. Chem. 95 (1991) 8781.
References
[42] H. Hu, S.R. Bare, I.E. Wachs, J. Phys. Chem. 99 (1995) 10897.
[43] M.A. Banares, H. Hu, I.E. Wachs, J. Catal. 150 (1994) 407.
[44] W. Zhang, A. Desikan, S.T. Oyama, J. Phys. Chem. 99 (1995) 14468.
[45] N. Ohler, A.T. Bell, J. Phys. Chem. B 109 (2005) 23419.
[46] N. Das, H. Eckert, H. Hu, I.E. Wachs, J.F. Walzer, F.J. Feher, J. Phys. Chem. 97
(1993) 8240.
[47] H. Eckert, I.E. Wachs, J. Phys. Chem. 93 (1989) 6796.
[48] M. Baltes, A. Kytokivi, B.M. Weckhuysen, R.A. Schoonheydt, P. van der Voort,
E.F. Vansant, J. Phys. Chem. B 105 (2001) 6211.
[1] X. Gao, I.E. Wachs, J. Catal. 192 (2000) 18.
[2] X. Gao, J.L.G. Fierro, I.E. Wachs, Langmuir 15 (1999) 3169.
[3] X. Gao, S.R. Bare, J.L.G. Fierro, I.E. Wachs, J. Phys. Chem. B 103 (1999) 618.
[4] B.N. Reddy, M. Subrahmanyam, Langmuir 8 (1992) 2072.
[5] R. Mariscal, M. Galan-Fereres, J.A. Anderson, L.J. Alemany, J.M. Palacios, J.L.G.
Fierro, in: G. Centi, C. Cristiani, P. Forzatti, S. Perathoner (Eds.), Environmental
Catalysis, SCI, Rome, 1995, p. 223.
[49] M.A. Vuurman, I.E. Wachs, D.J. Stufkens, A. Oskam, J. Mol. Catal. 80 (1993) 209.
[50] M.A. Vuurman, D.J. Stufkens, A. Oskam, J. Mol. Catal. 76 (1992) 263.
[51] R.Z. Khaliullin, A.T. Bell, J. Phys. Chem. B 106 (2002) 7832.
[52] B.M. Weckhuysen, I.E. Wachs, R.A. Schoonheydt, Chem. Rev. 96 (1996) 3327.
[53] M. Boudart, Chem. Rev. 95 (1995) 661.
[54] J.-M. Jehng, I.E. Wachs, Catal. Lett. 13 (1992) 9.
[55] J.-M. Jehng, I.E. Wachs, B.M. Weckhuysen, R.A. Schoonheydt, J. Chem. Soc. Fara-
day Trans. 91 (1995) 953.
[56] N.N. Greenwood, A. Earnshaw, Chemistry of the Elements, Pergamon Press,
Elmsford, NY, 1989.
[57] R.T. Sanderson, J. Chem. Educ. 65 (1988) 112.
[58] J.L.Bronkema, A.T. Bell, J. Phys. Chem. C 111 (2007) 420.
[59] J. Dobler, M. Pritzsche, J. Sauer, J. Am. Chem. Soc. 127 (2005) 10861.
[60] N. Magg, B. Immaraporn, J.B. Giorgi, T. Schroeder, M. Baumer, J. Dobler, Z.
Wu, E. Kondratenko, M. Cherian, M. Baerns, P.C. Stair, J. Sauer, H.J. Freund, J.
Catal. 226 (2004) 88.
[6] M.D. Amiridis, J.P. Solar, Ind. Eng. Chem. Res. 35 (1996) 978.
[7] R.A. Rajadhyaksha, G. Hausinger, H. Zeilinger, A. Ramstetter, H. Schmeiz, H.
Knözinger, Appl. Catal. 51 (1989) 67.
[8] R.A. Rajadhyaksha, H. Knözinger, Appl. Catal. 51 (1989) 81.
[9] P. Wauthoz, T. Machej, P. Grange, Appl. Catal. 69 (1991) 149.
[10] E.T.C. Vogt, A. Boot, A.J. van Dillen, J.W. Geus, F.J.J.G. Janssen, F.M.G. van den
Kerkhof, J. Catal. 114 (1988) 313.
[11] M. Galan-Fereres, R. Mariscal, L.J. Alemany, J.L.G. Fierro, J. Chem. Soc. Faraday
Trans. 90 (1994) 3711.
[12] M.P. McDaniel, M.B. Welsh, M.J. Dreiling, J. Catal. 82 (1983) 98.
[13] S.J. Conway, J.W. Falconer, C.H. Rochester, J. Chem. Soc. Faraday Trans. 85 (1989)
71.
[14] S. Rajagopal, T.L. Grimm, D.J. Collins, R. Miranda, J. Catal. 137 (1992) 453.
[15] M. Henker, K.P. Wendlandt, J. Valyon, P. Bornmann, Appl. Catal. 69 (1991) 205.
[16] S. Damyanova, A. Docteva, C. Vladov, L. Petrov, P. Grange, B. Delmon, Belg.
Chem. Commun. 30 (1999) 306.