Solid-state NMR study of the kinetics and mechanism of dimethyl ether carbonylation on cesium salt of 12-tungstophosphoric acid modified with Ag, Pt, and Rh

Article abstract of DOI:10.1016/j.jcat.2013.08.024

By using ¹³C magic angle spinning (MAS) NMR, the mechanism of dimethyl ether (DME) carbonylation with carbon monoxide has been studied on solid metal-containing acidic cesium salt of 12-tungstophosphoric acid, M/Cs₂HPW₁₂O₄₀ (M/HPA, M = Ag, Pt, and Rh). The kinetics of the reaction has been monitored with ¹H MAS NMR in situ. Activation of DME occurs on acidic OH groups of M/HPA and gives rise to the surface methoxy groups at 293-473 K. Carbon monoxide forms the surface metal-carbonyl complexes on M/HPA at 293-473 K. The insertion of CO from the carbonyl into the CH₃O bond of surface methoxide results in the acetate group attached to the Keggin anion, from which the target product, methyl acetate, is produced under the interaction with dimethyl ether. The reaction rate decreases in the following order: Rh/Cs₂HPW ₁₂O₄₀ a‰ Pt/Cs₂HPW ₁₂O₄₀ > Ag/Cs₂HPW₁₂O ₄₀, which correlates well with concentration of the reaction intermediates, methoxy groups, and metal-carbonyls. The higher concentration of these intermediates detected on Rh/Cs₂HPW₁₂O₄₀ is in charge of the higher carbonylation rate on this catalyst with respect to M/HPA (M = Ag, Pt) catalysts.

Full text of DOI:10.1016/j.jcat.2013.08.024

Journal of Catalysis 308 (2013) 250–257
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Solid-state NMR study of the kinetics and mechanism of dimethyl ether
carbonylation on cesium salt of 12-tungstophosphoric acid modified
with Ag, Pt, and Rh
Mikhail V. Luzgin a,b, Maxim S. Kazantsev a, Galina G. Volkova a, Alexander G. Stepanov a,b,
⇑
a Boreskov Institute of Catalysis, Siberian Branch of Russian Academy of Sciences, Prospekt Akademika Lavrentieva 5, Novosibirsk 630090, Russia
b Department of Natural Sciences, Novosibirsk State University, Pirogova Street 2, Novosibirsk 630090, Russia
a r t i c l e i n f o
a b s t r a c t
By using 13C magic angle spinning (MAS) NMR, the mechanism of dimethyl ether (DME) carbonylation
with carbon monoxide has been studied on solid metal-containing acidic cesium salt of 12-tungstophos-
phoric acid, M/Cs2HPW12O40 (M/HPA, M = Ag, Pt, and Rh). The kinetics of the reaction has been monitored
with 1H MAS NMR in situ. Activation of DME occurs on acidic OH groups of M/HPA and gives rise to the
surface methoxy groups at 293–473 K. Carbon monoxide forms the surface metal–carbonyl complexes on
M/HPA at 293–473 K. The insertion of CO from the carbonyl into the CH3AO bond of surface methoxide
results in the acetate group attached to the Keggin anion, from which the target product, methyl acetate,
is produced under the interaction with dimethyl ether. The reaction rate decreases in the following order:
Rh/Cs2HPW12O40 ꢀ Pt/Cs2HPW12O40 > Ag/Cs2HPW12O40, which correlates well with concentration of the
reaction intermediates, methoxy groups, and metal–carbonyls. The higher concentration of these inter-
mediates detected on Rh/Cs2HPW12O40 is in charge of the higher carbonylation rate on this catalyst with
respect to M/HPA (M = Ag, Pt) catalysts.
Article history:
Received 26 February 2013
Revised 27 June 2013
Accepted 18 August 2013
Available online 23 September 2013
Keywords:
Dimethyl ether
Carbon monoxide
Carbonylation
Mechanism
Kinetics
12-Tungstophosphoric acid
Metal-carbonyls
Solid-state NMR
Ó 2013 Elsevier Inc. All rights reserved.
1. Introduction
target products, acetic acid, or methyl acetate, have been moni-
tored [12,15].
Up to date, the carbonylation of methanol mediated by homo-
geneous iodine Rh and Ir-complexes in the presence of iodide pro-
moter [1–5] represents the main method of the acetic acid
production at the industrial level. Taking into account the corrosive
properties of iodide compounds, the development of heteroge-
neous systems that catalyze this reaction in the absence of iodide
additives is of great importance. Halide-free carbonylation of
methanol and dimethyl ether (DME) has been recently demon-
strated to be realized on the acid forms of zeolites (H-MOR, H-
BEA, H-ZSM-5) [6–8], solid 12-tungstophosphoric acid (hetero-
polyacid, HPA), and its cesium salts [9–12]. These results have
encouraged studies of the reaction mechanism using kinetic, IR,
desorption experiments [7,8], NMR spectroscopy [12–15], as well
as theoretical methods [16]. Among all intermediates, suggested
for the carbonylation of methanol and DME, only surface methoxy
groups have been reliably identified for zeolite-catalyzed reaction
[13]. As to the carbonylation on heteropolyacid catalysts is con-
cerned, both the methoxy group and the surface acetate, as well
as the processes of their interconversion and transformation into
We have demonstrated recently that the surface methoxy
groups and Rh-carbonyls represent the main intermediates formed
from dimethyl ether and carbon monoxide during carbonylation
on Rh/Cs2HPW12O40 [15]. The insertion of carbon monoxide at
the CH3AO bond of methoxide occurs at considerably higher tem-
peratures in comparison with that required for the methoxy group
formation and further conversion of the acetate fragment gener-
ated at this insertion. So, we suppose that the reaction of carbon
monoxide insertion represents a rate-determining stage of the car-
bonylation. The main role of Rh in this catalytic system consists in
trapping of carbon monoxide from gaseous phase and its further
transfer to the center of dimethyl ether activation [12,15].
In the present paper, we have tried and established the effect of
the metal modifier (Ag, Pt, and Rh) loaded on Cs2HPW12O40 on the
rate of the carbonylation reaction. The 13C solid-state (MAS, magic
angle spinning) NMR and in situ 1H MAS NMR measurements of
the reaction kinetics could be the powerful tools to achieve the re-
search objectives. NMR allows establishment the mechanisms of
bifunctional action of such catalysts. The knowledge on the mech-
anism assists in understanding the principles of the process pro-
ceeding and could serve as the base for goal-seeking search of
the new carbonylation processes as well as the improvement of
existing ones.
⇑
Corresponding author. Fax: +7 383 330 8056.
E-mail address: stepanov@catalysis.ru (A.G. Stepanov).
0021-9517/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2013.08.024

M.V. Luzgin et al. / Journal of Catalysis 308 (2013) 250–257
251
low as 13C CP/MAS NMR and 13C MAS NMR. The following condi-
tions were used for recording the spectra with CP: the proton
2. Experimental
high power decoupling field strength was 11.5 G (4.9 ls length of
2.1. Materials
90° 1H pulse), contact time was 4 ms at the Hartmann–Hahn
matching condition of 50 kHz, and the delay time between scans
was 3 s. The single pulse excitation 13C MAS NMR spectra were re-
Cesium salts of 12-tungstophosphoric acid promoted with met-
als (M/Cs2HPW12O40, where M = Ag, Pt, and Rh) were prepared by
drop-wise addition of stoichiometric amounts of 0.1 M aqueous
solution of cesium nitrate to a mixture of 0.1 M solutions of
H3PW12O40 (HPA) and AgNO3 or H2PtCl6 or RhCl3. The resulting
suspension was kept under vigorous stirring for 24 h and then
was evaporated at 353–373 K to the solid. The synthesized samples
of M/Cs2HPW12O40 contained 1 wt% Ag or Rh and 1.9 wt%, which
corresponds to the same atomic concentration of metals, ca.
corded with 90° flip angle 13C pulses of the 4.9
ls duration and 12 s
recycle delay, which satisfies 10 ꢃ T1 condition. High power proton
decoupling in these experiments was used only during the acquisi-
tion time. A few thousands scans were collected for each 13C CP/
MAS NMR and 13C MAS NMR spectrum. The spinning rate was 5–
8 kHz. 1H, and 13C chemical shifts were referenced with respect
to TMS as an external standard with accuracy 0.5 ppm. The preci-
sion in the determination of the relative line position was 0.1–
0.15 ppm. The sample temperature was controlled by the Bruker
BVT-2000 variable-temperature unit. The calibration of the tem-
perature inside the rotor was performed with an accuracy of 2 K
by using lead nitrate as a 207Pb MAS NMR chemical shift thermom-
eter [17].
100 l
mol gꢁ1. The chemical analysis confirmed the desired compo-
sition specified by the preparation process. X-ray diffraction pat-
tern of the sample revealed the typical body center cubic
structure of Cs2HPW12O40. The BET surface area of these samples
was ca. 100 m2 gꢁ1. The concentration of acidic protons in anhy-
drous samples, measured from 1H MAS NMR spectra using TMS
as internal standard, was ca. 340 l
mol gꢁ1, which is in a good
accordance with chemical composition.
Dimethyl ether-13C2 (99 atom% 13C), dimethyl ether (of P99.0%
purity), carbon monoxide (of P99.0% purity), and carbon monox-
ide-13C (99 atom% 13C) were purchased from Aldrich Chemical
Company Inc. and were used without further purification.
3. Results and discussion
3.1. Activation of dimethyl ether on M/Cs2HPW12O40 (M=Ag, Pt, and
Rh)
2.2. Sample preparation
In our recent studies, we have shown that the activation of di-
methyl ether (DME) occurs on Brønsted acid centers regardless of
the presence or the absence of rhodium species on the surface of
Cs2HPW12O40 [15]. So, we suppose that the conversion of DME
on the Ag- and Pt-containing heteropolyacids proceeds in a similar
way with the formation of the surface methoxy species and a side
product – trimethyloxonium cation (TMOC). Fig. 1 shows the 13C
CP/MAS NMR spectra recorded for the products formed from the
dimethyl ether-13C2 (DME-13C2) on the surface of M/Cs2HPW12O40
(M = Ag, Pt, and Rh).
The samples of Rh and Pt/Cs2HPW12O40 (ꢂ80 mg) were preli-
minary activated at 473 K in a stream of H2 for 2 h to reduce par-
tially metal cations species, while Ag/Cs2HPW12O40 was activated
in air atmosphere at 473 K. Further, the samples of M/Cs2HPW12-
O40 (M = Ag, Pt, and Rh) were calcined at 523 K for 16 h under vac-
uum with the residual pressure less than 10ꢁ2 Pa. Further, the
adsorption of DME (ca. 60–110
l
mol gꢁ1) and carbon monoxide
(160
l
mol gꢁ1) was performed on this sample under vacuum at
the temperature of liquid nitrogen. The glass tube of 4 mm o.d.
and 10 mm length (microreactor) with the catalyst sample and
adsorbates was then sealed off from the vacuum system at the
temperature of liquid nitrogen. This axially high symmetric sealed
glass tube could be tightly inserted into the 4 mm zirconia rotor for
subsequent in situ NMR analysis of the reaction intermediates and
products. The reaction was carried out in this sealed glass tube by
sample heating directly in NMR probe, if in situ monitoring of the
reaction kinetics was performed. In most cases, the reaction was
carried out by sample heating outside NMR probe, and further
NMR analysis of the reaction intermediates and products formed
in the sealed glass microreactor was performed at room
temperature.
The signal at 63 ppm with a shoulder at 67 ppm is predomi-
nantly observed at room temperature (Fig. 1a). These resonances
have earlier attributed to DME adsorbed on different Brønsted acid
sites of HPA, e.g., on terminal and bridged oxygen atoms of the
Keggin anion [15]. However, the signal at 67 ppm is rather close
to that from the protonated dimethyl ether in superacid solution
(68 ppm [18]). So, we attribute this resonance to DME protonated
by acidic OH group of HPA, whereas that at 63 ppm belongs to the
adsorbed ether. The resonance at 80 ppm is the characteristic of
trimethyloxonium cation (TMOC) both in solution [18] and on
the surface of the solid acid catalysts [12,15,19]. Heating the sam-
ple results in appearance of the resonances at 59 and 76 ppm
(Fig. 1b). The first signal belongs to the ordinary methoxy group
bound to the HPA surface [12,20]. The second one is usually ob-
served only for cesium salts of HPA and is attributed to another
type of the surface methoxy group formed on terminal oxygen
atom of the Keggin unit and containing cesium cation bound to
the oxygen atom (see Scheme 1) [15]. Thus, the conversion of
DME on M/Cs2HPW12O40 occurs according to Scheme 1 producing
methoxy groups, protonated ether, and TMOC. Methanol, which
should be formed during the DME-to-methoxide conversion
(Scheme 1), is transformed further to the methoxy group and
water. It should be noted that the concentration of the methoxy
group is rather low for both Ag- and Pt/Cs2HPW12O40 in compari-
son with that for Rh/Cs2HPW12O40 catalyst even at the highest
temperature of the reaction (see Fig. 1, cf. the intensity of signal
at 59 ppm for Ag, Pt, and Rh/Cs2HPW12O40). This may have an
essential effect on the carbonylation reaction (vide infra).
2.3. NMR experiments
1H and 13C MAS NMR spectra were recorded on the Bruker
AVANCE-400 spectrometer (Larmor frequencies of 400.000 and
100.613 MHz, respectively) within the temperature range of 293–
493 K for 1H NMR and at 293 K for 13C NMR. For recording 1H
MAS NMR spectra, Hahn-echo pulse sequence (90-
tion) was used, where was equal to the period of rotor spinning,
200 s. This echo pulse sequence was used to suppress background
broad signal from the NMR probe. The following conditions were
used for recording the spectra: 4.9
s length of 90° 1H pulse, the
s-180-s-acquisi-
s
l
l
delay time between scans was 3 s. Twenty to forty scans were col-
lected for each 1H MAS NMR spectrum.
13C MAS NMR spectra with the high power proton decoupling
were recorded with or without cross-polarization (CP) denoted be-

252
M.V. Luzgin et al. / Journal of Catalysis 308 (2013) 250–257
Pt/Cs2HPW12O40
Rh/Cs2HPW12O40
63
Ag/Cs2HPW12O40
63
63
a
67
CH3
80
63
CH3
O
CH3OCH3
67
80
H
CH3
O
80
76
293 K
H3C
CH3
80
*
*
59
59
b
CH3
76
76
O
O
59
59
59
59
W
W
W
423 K
c
*
*
*
*
*
*
76
Cs
O
CH3
63
76
63
150
100
50
0
O
W
O
ppm
O
O
80
473 K
76
W
O
*
*
*
5
150
0
100
0
0
150
100
50
ppm
ppm
Fig. 1. 13C CP/MAS NMR spectra of the products formed from DME-13C2 on M/Cs2HPW12O40 (M = Rh (left), M = Ag (middle), and M = Pt (right)) at: (a) 293 K; (b) 423 K; (c)
473 K. Asterisks (ꢄ) denote spinning side-bands.
13CO
a
183
Ag/Cs2HPW12O40
178
183
170 150
190
230
210
*
183
Pt/Cs2HPW12O40
188
b
Rh/Cs2HPW12O40
Rh-CO
175
c
*
CO2
*
*
*
300
250
200
150
100
50
0
ppm
Fig. 2. 13C MAS NMR spectra of the products formed from 13CO at 293–473 K on: (a)
Ag/Cs2HPW12O40, (b) Pt/Cs2HPW12O40; (c) Rh/Cs2HPW12O40. The spinning rate was
8.0 kHz. Asterisks (ꢄ) denote spinning side-bands.
Scheme 1. The common steps of the dimethyl ether activation on M/Cs2HPW12O40
(M = Ag, Pt, and Rh) at 293–473 K [15,40].
cies, which signal is observed at 188 ppm [23] (Fig. 2b). Heating the
samples at 373–473 K does not cause any change in the spectra of
Ag- and Pt-carbonyls. It should be noted in advance that under
similar conditions (loaded metal concentration and amount of CO
adsorbed), only partial, ca. 60%, conversion of CO into metal–car-
bonyl is reached in the case of Ag and Pt/Cs2HPW12O40 at room
temperature. At the same time, almost quantitative conversion of
CO into RhACO species was observed for Rh/Cs2HPW12O40 [15]
(Fig. 2c). Both AgACO and PtACO surface complexes exist only un-
der the atmosphere of carbon monoxide, and the evacuation of the
samples leads to complete disappearance of their signals. So, we
were not able to follow the direct interaction of metal–carbonyl
with dimethyl ether as in the case of the rhodium catalyst [15].
We have analyzed further the interaction between DME and CO
3.2. Activation of carbon monoxide on M/Cs2HPW12O40 (M=Ag, Pt, and
Rh): the formation of carbonyl complexes
We have previously established that carbon monoxide forms
rhodium-carbonyls on the surface of Rh/Cs2HPW12O40. This car-
bonyl serves as the intermediate to transfer CO from gaseous phase
to the surface methoxide during the DME carbonylation on this
catalyst [15]. So, we have further studied the process of the CO
activation on Cs2HPW12O40 containing Ag or Pt.
Fig. 2 shows 13C MAS NMR spectra of carbon monoxide-13C ad-
sorbed on M/Cs2HPW12O40 (M = Ag, Pt, and Rh). For Ag/Cs2HPW12-
O40, a new signal at 178 ppm, different from that for gaseous 13CO
at 183 ppm, is observed already at room temperature (Fig. 2a).
Similar signals were found for Ag-carbonyl complexes in solution
[21,22]. Hence, carbon monoxide forms Ag-carbonyl complexes
on the surface of Ag/Cs2HPW12O40
The adsorption of carbon monoxide on Pt/Cs2HPW12O40 at room
temperature also leads to the formation of surface Pt-carbonyl spe-
on Ag/Cs2HPW12O40 and Pt/Cs2HPW12O40
.
It should be noted that the state of various metals in the metal–
carbonyls under study is different. For Rh- and Pt-HPA activated in
the hydrogen stream, zero-valence of metal would be expected
[24]. However, chemical shift of RhACO, 175 ppm, is typical for
Rh(I) carbonyl species [25–31]. This implies that Rh(0) is
.

M.V. Luzgin et al. / Journal of Catalysis 308 (2013) 250–257
253
transformed into Rh(I) in the formed Rh-carbonyl species [32–35].
In contrast to Rh, chemical shift of PtACO species, 188 ppm, indi-
cates that the metal exists in the form of highly-dispersed Pt(0)
particles [23]. For Ag/HPA, the observed 13C NMR signal is typical
for Ag(I)ACO species [21,22].
In principle, one cannot exclude the possibility of the interme-
diate formation of metal–methyl species, COAMACH3, especially
in the case of rhodium catalyst, via oxidative addition of DME to
the metal–carbonyls. This reaction is analogous to oxidative addi-
tion of methyl iodide promoter during methanol carbonylation by
rhodium complexes in solution [26,39]. Further processes could oc-
cur on metal center and include the migratory insertion reaction,
affording metal–acetyl complex. Rapid migration of acetyl group
from metal center to Brønsted acid site of Rh/Cs2HPW12O40 could
give rise to acetate group attached to the Keggin unit (Keggin ace-
tate). Indeed, such processes are realized onRh/Cs2HPW12O40 during
the DME carbonylation, but only in the presence of methyl iodide
promoter [40]. Methyl iodide forms cis- and trans-Rh-methyl com-
plexes even at room temperature, which are stable in the absence of
carbon monoxide and characterized by the resonances at 5 and
14 ppm. In the presence of CO, they readily converts into rhodium-
acetyls at its typical position in 13C NMR spectrum at 49 (CH3-group)
and 215 ppm (CO-group) [40]. In the presence of Brønsted acid site,
Rh-acetyl interacts with OH group to produce Keggin acetate. HI
formed at this stage interacts with DME to regenerate reactive
CH3I and, thus, involves the ether into the reaction. However, the sit-
uation is quite different in the absence of promoter. For DME itself,
the formation of both Rh-methyl species, as well as Rh-acetyls, is
never observed on metal-containing HPA used in our study (see
Figs. 1,3, Refs. [15,40]). So, this pathway of DME carbonylation on
M/Cs2HPW12O40, if realizes, gives minor contribution to the process.
Methyl acetate forms from the Keggin acetate under the reac-
tion with DME molecule (Scheme 2, stage 2). Contrary to Ag and
3.3. Interaction of DME with carbon monoxide on Ag and Pt/
Cs2HPW12O40
Fig. 3 shows the results of the reaction between CO and DME
on Ag/Cs2HPW12O40 and Pt/Cs2HPW12O40. The reaction does not
proceed up to 423 K, and only small conversion takes place on
Pt catalyst (Fig. 3a and b). The reaction starts at 473 K, and
methyl acetate is the main product, which is identified by typical
signals at 22 (CH3), 54 (CH3AO), and 179 ppm (>C@O) [36,37].
Small intensity resonances at 193 and 22 ppm indicate on the for-
mation of the surface acetate attached to Keggin anion (Keggin
acetate) [38]. The formation of the Keggin acetate is the result
of the CO insertion into the OACH3 bond of surface methoxy
group. On Rh/Cs2HPW12O40, we observe the direct transformation
of rhodium–carbonyl into the >C@O moiety of the Keggin acetate,
and no free CO exists in gaseous phase. On Ag- and Pt-containing
catalysts, metal–carbonyl complexes are in equilibrium with gas-
eous carbon monoxide (Fig. 1). So, the Keggin acetate can be
formed either from the metal–carbonyl or immediately from the
gaseous CO on Ag/Cs2HPW12O40 and Pt/Cs2HPW12O40 (Scheme 2,
stages 1a and 1b). The contribution of stage 1a and 1b will be dis-
cussed later (see Section 3.4).
Ag/Cs2HPW12O40
Pt/Cs2HPW12O40
63
63
a
80
80
293 K
b
76
59
22
76
59
179
423 K
*
*
*
O
c
22
21 179
CH3
54
C
54
CH3
O
193
179
54
473 K 30 min
179
*
22
d
193
C
O
CH3
193
54
50
193
O
W
W
473 K 2 h
*
*
0
*
0
*
300
200 150
50
-50
250
300
200 150
-50
100
250
100
ppm
ppm
Fig. 3. 13C CP/MAS NMR spectra of the products formed from DME-13C2 and 13CO on Ag/Cs2HPW12O40 and Pt/Cs2HPW12O40 at: (a) 293 K; (b) 423 K for 60 min; (c) 473 K for
30 min; (d) 473 K for 2 h. Asterisks (ꢄ) denote spinning side-bands.

254
M.V. Luzgin et al. / Journal of Catalysis 308 (2013) 250–257
Scheme 2. The common steps of the dimethyl ether carbonylation on M/Cs2HPW12O40 (M = Ag or Pt) at 293–473 K [15,40].
Pt/Cs2HPW12O40 catalysts, the Keggin acetate is the main product
of the DME carbonylation on Rh/Cs2HPW12O40 under similar reac-
tion conditions. This is the result of the lower concentration of the
unreacted ether, which could interact with Keggin acetate (see
Fig. 1). On Rh/Cs2HPW12O40, methyl acetate is formed only under
the adsorption of additional portion of DME [15].
The carbonylation of DME on Ag and Pt/Cs2HPW12O40 proceeds
at higher temperature than on Rh/Cs2HPW12O40 catalyst (vide infra,
Section 3.4). We have further monitored the kinetics of the DME
carbonylation on these three catalysts using 1H MAS NMR to obtain
a detailed information on the reaction rate.
gen-bonded with oxygen atoms of adjacent Keggin anions. The
intensity of the signal of hydrogen-bonded OH groups increases
and shifts to 9.8 ppm during the reaction. This is due to water
formed at the DME conversion to the methoxy groups. Water mol-
ecules are bounded to the acid OH groups of HPA to form H2nþ1Onþ
cation (n = 1, 2) [47], which results to the additional low-field shift
of the signal at 8.8 ppm.
Almost quantitative conversion of DME into the carbonylation
products occurs (Fig. 4). For obtaining information on the reaction
kinetics, we have followed with time the intensity of the signal at
2.5 ppm from the methyl groups. Based on the known initial con-
centration of DME, the total concentration of the Keggin acetate
and acetic acid was calculated.
3.4. In situ 1H MAS NMR monitoring the kinetics of the DME
carbonylation on M/ Cs2HPW12O40 (M=Ag, Pt, and Rh)
Fig. 5 shows experimental and simulated kinetics of the DME
carbonylation. The experimental points were fitted using a simple
function of exponential growth:
Fig. 4 shows the 1H MAS NMR spectra of DME reacting with car-
bon monoxide on Rh/Cs2HPW12O40 in dependence of the reaction
time. The decreasing signal at 4.0 ppm is due to the CH3AO moiety
of the mixture of adsorbed DME, methoxy group, protonated ether,
and TMOC [18,41–43]. The increasing resonance at ca. 2.5 ppm
arises evidently from the CH3 groups of both the Keggin acetate
and the acetic acid [43]. The signals at 8.8 and 5.0 ppm belong to
the acid OH groups of HPA. The first signal is typical for solid het-
eropolyacids [44–46] and is attributed to strongly acidic or supe-
racidic OH groups [47], which are hydrogen-bonded with
neighboring Keggin anions [48]. According to its position in the
spectrum, the upfield shifted signal at 5.0 ppm should be assigned
to less acidic ‘‘free’’ (like in zeolites) OH groups [49], i.e. without
hydrogen bonds. Similar chemical shift of 5.2 ppm was reported
for the acid OH groups of H3PW12O40 dispersed on various supports
[45,46]. The signal was attributed to the OH groups of the isolated
Keggin anions or small clusters of them, in which no hydrogen-
bonding between the Keggin anions exists [45,46]. Cs2HPW12O40
CAcetateþAcOH ¼ C0ð1 ꢁ eꢁkt
Þ
ð1Þ
where CAcetate+AcOH is current, C0 is final total concentration of the
surface acetate and acetic acid, and k is the effective rate constant.
Simulation of experimental kinetics offers the effective rate con-
stants
k for the DME carbonylation and its initial rates
W0DME ¼ k ꢅ C0 at different reaction temperatures (Table 1). Since
the reaction is complex and includes the formation of intermediate
species, the temperature dependence of initial reaction rate, WD0ME
,
was used for the determination of the activation energy. The Arrhe-
nius plot for W0DME (Fig. 6) gives the activation energy 88 kJ molꢁ1
.
Fig. 7 shows the 1H MAS NMR spectra of DME reacting with car-
bon monoxide on Ag/Cs2HPW12O40 in dependence of the reaction
time. The signal at 4.0 ppm, which belongs to the CH3AO groups
of the mixture of adsorbed DME, methoxy group, protonated ether,
and TMOC [18,41–43], is decreasing in the course of the reaction.
The signal of DME in the gas phase observed at 3.2 ppm in the
beginning of reaction disappears in the sequel. The increasing res-
onances at ca. 2.5 and 4.2 ppm arise from the CH3 groups in methyl
acetate [50]. The signals at 5.4 ppm belong to the acid OH groups of
Ag/Cs2HPW12O40. We have used the intensity of the signal at
2.5 ppm, and the known initial concentration of DME for obtaining
the kinetic curve shown in Fig. 8.
possesses appreciable porosity and surface area, ca. 100 m2 gꢁ1
,
in contrast to 5–10 m2 gꢁ1 typical for ‘‘bulk’’ H3PW12O40. So, ce-
sium salt can indeed contain more protons, which are not hydro-
2.5
4.0
5.0
As in the case of Rh/Cs2HPW12O40 catalyst, we have used the
temperature dependence of the initial reaction rate to determine
the activation energy (Table 1). The Arrhenius plot for WD0ME
(Fig. 6) affords the activation energy 85 kJ molꢁ1, which is similar
to that for Rh-catalyst.
9.8
8.8
Similar experiments have been carried out for Pt/Cs2HPW12O40
(spectra and kinetics curves not shown). The results are presented
in Table 1 and Fig. 6.
The activation energies for the DME carbonylation proved to be
similar for three catalysts under study. However, Rh/Cs2HPW12O40
is significantly more active than Pt/Cs2HPW12O40 and Ag/
180
t, min
0
12
10
8
6
4
2
ppm
Cs2HPW12O40
.
Fig. 4. Stack plot of the 1H MAS NMR spectra at 449 K of DME adsorbed on Rh/
Cs2HPW12O40 and reacting with carbon monoxide. The first spectrum (bottom) was
recorded 5 min after the temperature was raised to 449 K and the last spectrum
(top) after 180 min of the reaction duration. The time between subsequent spectra
recording was 4 min.
To demonstrate the promoting influence of metal on the carbon-
ylation reaction, as well as to accent the role of protons, the reaction
rates have been also determined for Cs2HPW12O40 and Rh/Cs3PW12-
O40 samples. The latter contains the same concentration of the

M.V. Luzgin et al. / Journal of Catalysis 308 (2013) 250–257
255
180
180
160
140
120
100
80
444 K
434 K
160
140
120
100
80
60
60
40
40
20
20
0
0
0
50
100 150 200 250 300
0
50
100 150 200 250 300
t, min
t, min
180
160
140
120
100
80
180
160
140
120
100
80
462 K
454 K
60
60
40
40
20
20
0
0
0
20 40 60 80 100 120 140
0
10 20 30 40 50 60 70 80 90
t, min
t, min
Fig. 5. Experimental and simulated (solid curves) kinetics of DME carbonylation on Rh/Cs2HPW12O40
.
Table 1
Initial rates of the DME carbonylation (WD0ME
, l
mol gꢁ1 minꢁ1) on M/Cs2HPW12O40 catalysts (M = Ag, Pt, and Rh).
T, K
Ag/Cs2HPW12O40
Pt/Cs2HPW12O40
Rh/Cs2HPW12O40
Cs2HPW12O40
Rh/Cs3PW12O40
434
444
449
454
462
473
482
2.1 0.1
3.6 0.2
4.2 0.3
6.9 0.4
8.8 0.4
0.51 0.03
0.90 0.06
1.25 0.05
2.1 0.2
1.0 0.06
1.38 0.07
3.2 0.2
3.7 0.2
61.0
60.14
2.5
O
4.0
5.4
4.2
2.5
CH3
C
Rh
=
88 8 kJ mol-1
4.2
Ea
10
O
CH3
80
t, min
0
1
11 10
9
8
7
6
5
4
3
2
ppm
Pt
=
89 15 kJ mol-1
Ea
DME
0
1
Fig. 7. Stack plot of the 1H MAS NMR spectra at 482 K of DME adsorbed on Ag/
Cs2HPW12O40 and reacting with carbon monoxide. The first spectrum (bottom) was
recorded 2 min after the temperature was raised to 482 K and the last spectrum
(top) after 80 min of the reaction duration. The time between subsequent spectra
recording was 1.9 min.
W
Ag
Ea
=
85 7 kJ mol-1
0.0023
0.0021
0.0022
T-1, K-1
of centers on the catalysts surface: Brønsted acid sites and metal
centers. The formation of methoxy group from DME occurs on
Brønsted acid sites, while the metal centers provide the delivery of
carbon monoxide to the center of DME activation. Thus, stage 1a
(Scheme 2) does provide the main pathway of DME carbonylation
on metal-containing Cs2HPW12O40. The insertion of CO from MACO
moiety into the CH3AO bond of methoxy group appears to be indi-
rect and demands for partial cleavage of MACO bond due to proper-
ties of metal–carbonyls, in which the electronic density on carbon
atom is lower in comparison with that on carbon atom of free CO.
However, the role of the formation of metal–carbonyl species
Fig. 6. Arrhenius plots for the initial rate, W0DME, of DME carbonylation on: (s) – Rh/
Cs2HPW12O40; (h) – Pt/Cs2HPW12O40; ( ) – Ag/Cs2HPW12O40
r
.
metal, but only traces of OH groups. Since the reaction on this sam-
ples proceeds very slowly, we can present the results obtained at
highest temperature studied (see Table 1). For un-promoted Cs2-
HPW12O40, the reaction rate is notably lower in comparison with
the rates for all M/Cs2HPW12O40. The lowest active catalyst proved
to be Rh/Cs3HPW12O40. All these data indicate that the efficiency
of metal-containing HPA is determined by the presence of two types

256
M.V. Luzgin et al. / Journal of Catalysis 308 (2013) 250–257
90
80
70
60
50
40
30
20
10
0
90
80
454 K
462 K
70
60
50
40
30
20
10
0
0
20
40
60
80 100 120
0
20 40 60 80 100 120 140 160
t, min
t, min
90
80
70
60
50
40
30
20
10
0
90
80
70
60
50
40
30
20
10
0
482 K
473 K
0
20
40
60
0
20
40
60
80
100
t, min
t, min
Fig. 8. Experimental and simulated (solid curves) kinetics of DME carbonylation on Ag/Cs2HPW12O40
.
consists in increasing lifetime (i.e., increasing the effective concen-
tration) of carbon monoxide in close proximity to surface methox-
ide, as it was suggested earlier [9]. This favors carbonylation.
The difference in the activity between M/Cs2HPW12O40 can be
rationalized in terms of relative concentrations of the intermediates
formed at the surface of these catalysts. The rate-determining step of
the carbonylation reaction on Rh/Cs2HPW12O40 is the interaction of
Rh-carbonyl with surface methoxide [15]. The situation is similar for
Pt- and Ag-containing catalysts. Indeed, the reaction intermediates,
the methoxy group and metal–carbonyl, are formed already at 293–
423 K, while appreciable interaction between these species and the
formation of the carbonylation products occurs at P453 K. The rate
of this interaction, as well as the rate of the entire process, is propor-
tional to the concentrations of the intermediate methoxy group and
metal–carbonyl: WDME ꢂ [CH3AOAW] ꢃ [MACO]. Rhodium cata-
lyst provides higher concentration of both methoxy intermediate
and carbonyl complex. This can account for the higher reaction rate
Rh-carbonyl provides its higher concentration in the reaction
with the rhodium catalyst. The possible reason for such order in
stability and concentration of metal–carbonyls may be the differ-
ence in energy of binding of CO with metal-particles. In this case,
the higher binding energy, which could be expected for Rh-cata-
lyst, should increase the activation energy for this catalyst. The
second parameter, which could have an influence on the activa-
tion energy, is enthalpy of methoxy group formation, as we have
already demonstrated for the case of DME carbonylation on
H3PW12O40 [51]. Lower enthalpy of methoxy group formation
for Rh-catalyst can compensate, in principle, the higher binding
energy, and we observe nearly the same activation barrier for
all samples under study. However, further studies are needed to
clarify this issue.
Thus, one should distinguish two principal factors determining
the activity of metal-containing heteropolyacids. They reflect the
bifunctional character of these catalysts. First, OH groups of strong
acidity of HPA structure are needed to provide the DME activation
and the formation of the intermediate methoxy groups. The second
factor is determined by the choice of metal modifier. The modifier
should be capable to perform the capture of carbon monoxide from
the gas phase to transfer CO to the center of the DME activation in
the form of carbonyl.
observed for Rh/Cs2HPW12O40
.
As far as methoxy intermediate is concerned, the possible factor
determining its concentration can be the nature and the acidity of
OH groups of HPA unit. According to 1H MAS NMR spectra, all the
catalysts under study contain a similar concentration of OH groups
(ca. 340 l
mol gꢁ1). However, only Rh/Cs2HPW12O40 has a signifi-
cant concentration of strongly acidic OH groups with the signal
at 8.8 ppm, whereas Ag- and Pt/Cs2HPW12O40 contain presumably
less acidic OH groups with the resonances at 5.0–5.4 ppm (cf. bot-
tom spectra of Figs. 4 and 7). Un-promoted HPA, Cs2HPW12O40, re-
vealed similar distribution of the OH groups as Rh/HPA (spectrum
not shown) that resulted in similar behavior of both catalysts in the
DME activation [12–15]. So, we suppose that the activation of DME
on more strong acidic OH groups of Rh/HPA may be the reason for
the higher concentration of the methoxy group on this catalyst.
The second point to note is that the concentration of surface
carbonyl complexes for Ag and Pt/Cs2HPW12O40 is also lower than
for Rh-catalyst (see Fig. 2). This fact also contributes to the higher
reaction rate achieved on the Rh/Cs2HPW12O40. The reason for the
difference in concentrations can be reasonably explained in terms
of stability of carbonyl-intermediates. The higher stability of
4. Conclusions
By using 13C and 1H solid-state NMR, the carbonylation of di-
methyl ether has been studied on Ag-, Pt-, and Rh-containing acidic
cesium salt of 12-tungstophosphoric heteropolyacid.
The mechanism of bifunctional action of these catalysts has
been established. Activation of dimethyl ether occurs on Brønsted
acid site of the Keggin unit with the formation of the intermediate
methoxy groups. Carbon monoxide is transferred to the center of
the DME activation through the formation of metal–carbonyls.
The insertion of CO from the metal–carbonyl to the CH3AO bond
of the methoxy group results in acetate fragment attached to the
Keggin anion. The later, under the interaction with the DME, leads
to the target reaction product, methyl acetate.

M.V. Luzgin et al. / Journal of Catalysis 308 (2013) 250–257
[22] Y. Souma, H. Kawasaki, Catal. Today 36 (1997) 91.
257
The kinetic studies revealed the similar values of the activation
energy for all three catalysts, 85–89 kJ molꢁ1, the reaction rate
decreasing in the following order: Rh/Cs2HPW12O40 ꢀ Pt/Cs2
HPW12O40 > Ag/Cs2HPW12O40. Such a sequence is the result of dif-
ferent concentrations of the reaction intermediates, the methoxy
groups, and metal–carbonyls, that are the highest in the case of
[23] J.S. Bradley, J.M. Millar, E.W. Hill, S. Behal, J. Catal. 129 (1991) 530.
[24] A.E. Newkirk, D.W. McKee, J. Catal. 11 (1968) 370.
[25] A. Haynes, B.E. Mann, D.J. Gulliver, G.E. Morris, P.M. Maitlis, J. Am. Chem. Soc.
113 (1991) 8567.
[26] A. Haynes, B.E. Mann, G.E. Morris, P.M. Maitlis, J. Am. Chem. Soc. 115 (1993)
4093.
[27] Q. Xu, H. Nakatani, Y. Souma, J. Org. Chem. 65 (2000) 1540.
[28] T.M. Duncan, J.T. Yates, R.W. Vaughan, J. Chem. Phys. 71 (1979) 3129.
[29] T.M. Duncan, J.T. Yates, R.W. Vaughan, J. Chem. Phys. 73 (1980) 975.
[30] T.M. Duncan, T.W. Root, J. Phys. Chem. 92 (1988) 4426.
[31] I.D. Gay, J. Phys. Chem. 94 (1990) 1207.
Rh/Cs2HPW12O40
.
References
[32] C.A. Rice, S.D. Worleya, J. Chem. Phys. 74 (1981) 6487.
[33] R.R. Cavanagh, J.T. Yates, J. Chem. Phys. 74 (1981) 4150.
[34] H.F.J. van ’t Blik, J.B.A.D. van Zon, T. Huizinga, J.C. Vis, D.C. Koningsberger, R.
Prim, J. Am. Chem. Soc 107 (1985) 3139.
[35] P. Basu, D. Panayotov, J. Yates, J. T., J. Phys. Chem. 91 (1987) 3133.
[36] E. Breitmaier, W. Voelter, 13C NMR Spectroscopy, Methods and Applications in
Organic Chemistry, VCH, Weinheim, 1978.
[37] D.E. Dorman, B. Bauer, J.D. Roberts, J. Org. Chem. 40 (1975) 3729.
[38] M.V. Luzgin, V.A. Rogov, S.S. Arzumanov, A.V. Toktarev, A.G. Stepanov, V.N.
Parmon, Angew. Chem. Int. Ed. 47 (2008) 4559.
[1] F.E. Paulik, J.F. Roth, Chem. Commun. (1968) 1578.
[2] J.F. Roth, J.H. Craddock, A. Hershman, F.E. Roth, Chemtech 1 (1971) 600.
[3] R.G. Schultz, P.D. Montgomery, J. Catal. 13 (1969) 105.
[4] R.G. Schultz, US Patent 3 717 670, Monsanto Company, 1973.
[5] M.J. Howard, M.D. Jones, M.S. Roberts, S.A. Taylor, Catal. Today 18 (1993) 325.
[6] B. Ellis, M.J. Howard, R.W. Joyner, K.N. Reddy, M.B. Padley, W.J. Smith, Stud.
Surf. Sci. Catal. 101 (1996) 771.
[7] P. Cheung, A. Bhan, G.J. Sunley, E. Iglesia, Angew. Chem. Int. Edit. 45 (2006)
1617.
[8] P. Cheung, A. Bhan, G.J. Sunley, D.J. Law, E. Iglesia, J. Catal. 245 (2007) 110.
[9] R.W. Wegman, J. Chem. Soc. Chem. Commun. (1994) 947.
[10] G.G. Volkova, L.M. Plyasova, A.N. Salanov, G.N. Kustova, T.M. Yurieva, V.A.
Likholobov, Catal. Lett. 80 (2002) 175.
[11] G.G. Volkova, L.M. Plyasova, L.N. Shkuratova, A.A. Budneva, E.A. Paukshtis, M.N.
Timofeeva, V.A. Likholobov, Stud. Surf. Sci. Catal. 147 (2004) 403.
[12] M.V. Luzgin, M.S. Kazantsev, W. Wang, A.G. Stepanov, J. Phys. Chem. C 113
(2009) 19639.
[39] A. Haynes, Top. Organomet. Chem. (2006) 179.
[40] M.S. Kazantsev, M.V. Luzgin, G.G. Volkova, A.G. Stepanov, J. Catal. 291 (2012) 9.
[41] G.A. Olah, D.H. O’Brien, J. Am. Chem. Soc 89 (1967) 1725.
[42] M.J. Lacey, C.G. Macdonald, A. Pross, J.S. Shannon, S. Sternhell, Aust. J. Chem. 23
(1970) 1421.
[43] H.E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 62 (1997) 7512.
[44] Y. Kanda, K.Y. Lee, S. Nakata, S. Asaoka, M. Misono, Chem. Lett. (1988) 139.
[45] V.M. Mastikhin, S.M. Kulikov, A.V. Nosov, I.V. Kozhevnikov, I.L. Mudrakovskii,
M.N. Timofeeva, J. Mol. Catal. 60 (1990) 65.
[13] Y. Jiang, M. Hunger, W. Wang, J. Am. Chem. Soc. 128 (2006) 11679.
[14] A. Bhan, A.D. Allian, G.J. Sunley, D.J. Law, E. Iglesia, J. Am. Chem. Soc. 129 (2007)
4919.
[46] V.M. Mastikhin, V.V. Terskikh, M.N. Timofeeva, O.P. Krivoruchko, J. Mol. Catal.
A 95 (1995) 135.
[15] M.V. Luzgin, M.S. Kazantsev, G.G. Volkova, W. Wang, A.G. Stepanov, J. Catal.
277 (2011) 72.
[47] U. Filek, A. Bressel, B. Sulikowski, M. Hunger, J. Phys. Chem. C 112 (2008)
19470.
[16] M. Boronat, C. Martinez-Sanchez, D. Law, A. Corma, J. Am. Chem. Soc. 130
(2008) 16316.
[48] M. Misono, K. Sakata, Y. Yoneda, W.Y. Lee, Acid-Redox Bifunctional Properties
of Heteropoly Compounds of Molybdenum and Tungsten Correlated with
Catalytic Activity for Oxidation of Methacrolein, Elsevier, Kodansha (Tokyo),
1980. 1047.
[49] M. Hunger, Catal. Rev.-Sci. Eng. 39 (1997) 345.
[50] J. McMurry, Fundamentals of Organic Chemistry, Cengage Learning, Stamford,
2011.
[17] D.B. Ferguson, J.F. Haw, Anal. Chem. 67 (1995) 3342.
[18] G.A. Olah, H. Doggweiler, J.D. Felberg, S. Fronlich, J. Org. Chem. 50 (1985) 4847.
[19] E.J. Munson, J.F. Haw, J. Am. Chem. Soc. 113 (1991) 6303.
[20] H.L. Zhang, A.M. Zheng, H.G. Yu, S.H. Li, X. Lu, F. Deng, J. Phys. Chem. C 112
(2008) 15765.
[21] P.K. Hurlburt, J.J. Rack, J.S. Luck, S.F. Dec, J.D. Webb, O.P. Anderson, S.H. Strauss,
J. Am. Chem. Soc. 116 (1994) 10003.
[51] M.S. Kazantsev, M.V. Luzgin, A.G. Stepanov, J. Phys. Chem. C 117 (2013) 11168.