Selective carbonylation of dimethyl ether to methyl acetate catalyzed by acidic zeolites

Article abstract of DOI:10.1002/anie.200503898

(Chemical Equation Presented) A simpler way: Acidic zeolites catalyze dimethyl ether carbonylation to methyl acetate at low temperatures with high selectivity and catalyst stability. This approach provides a halide-free heterogeneous catalytic route to acetic acid and methyl acetate.

Full text of DOI:10.1002/anie.200503898

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Angewandte
Chemie
Heterogeneous Catalysis
DOI: 10.1002/anie.200503898
Selective Carbonylation of Dimethyl Ether to
Methyl Acetate Catalyzed by Acidic Zeolites**
Patricia Cheung, Aditya Bhan, Glenn J. Sunley, and
Enrique Iglesia*
Methanol carbonylation is currently used to produce acetic
acid with Rh and Ir organometallic complexes and iodide co-
catalysts.[1–3] Zeolites[4–9] and Keggin polyoxometallate clus-
ters[10–13] also catalyze the carbonylation of alcohols and ethers
(Koch reaction) to form carboxylic acids and esters. Surface
alkyl groups, formed by alcohol dissociation, alkene proto-
nation, or alkane deprotonation at acidic sites react with CO
to form acylium ions, which then form carboxylic acids and
esters, as in the reactions of isobutane on sulfated zirconia[14]
and tert-butyl alcohol[5,7,8] on acidic ZSM5, MOR, BEA, and
Y-zeolites to form pivalic acid. Carbonylation of methanol
and dimethyl ether (DME) on acidic zeolites and polyoxo-
metallate clusters occurs concurrently with side reactions
(Scheme 1) and significant catalyst deactivation.[4,6,9–13]
Scheme 1. Network of carbonylation, hydration, dehydration, and
methanol-to-hydrocarbon chemistries.
We report herein stable and highly selective (> 99%)
halide-free catalysts based on zeolites for DME carbonylation
to methyl acetate at low temperatures (423–463 K). The
reaction rates are much higher than for methanol carbon-
ylation because water may adsorb competitively at CO
binding sites and/or cause parallel methanol dehydration
[*] P. Cheung, A. Bhan, Professor E. Iglesia
Department of Chemical Engineering
University of California at Berkeley
Berkeley, CA 94720 (USA)
Fax: (+1)510-642-4778
E-mail: iglesia@cchem.berkeley.edu
G. J. Sunley
BP Chemicals Limited
Hull Research and Technology Centre
Saltend, Hull HU128DS (UK)
[**] This work was supported by BP as part of the Methane Conversion
Cooperative Research Programat the University of California at
Berkeley. We acknowledge helpful technical discussions with Drs.
Theo Fleisch, David Law, Sander Gaemers, and Ben Gracey of BP.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or fromthe author.
Angew. Chem. Int. Ed. 2006, 45, 1617 –1620
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1617

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Communications
reactions. DME carbonylation involves the formation of
methyl groups, which reform after reaction with CO to
generate acetyl species and reactions of DME with acetyl
species to give methyl acetate; these propagation steps avoid
the formation of methyl groups by the initial direct reaction of
DME with H+, which causes an induction period during the
early stages of reaction (see the Supporting Information).
DME carbonylation proceeds at approximately 423–
463 K with 99% selectivity to methyl acetate; higher temper-
atures favor homologation and oligomerization reactions that
form hydrocarbons. Figure 1 shows that the rate of methyl
Figure 2. Effects of DME (squares: 0.5 MPa CO, 438 K; triangles:
0.5 kPa H2O, 0.5 MPa CO, 438 K) and CO (diamonds: 2–16 kPa
DME, 438 K) concentration on the rate of methyl acetate formation
on H-MOR (Si/Al=10:1).
The addition of water (0.5–1.1 kPa) to DME–CO reac-
tants led to methanol formation and a lower rate of methyl
acetate synthesis (Figure 2 and Table 1), without detectable
formation of acetic acid. The rate of carbonylation gradually
returned to initial values after the removal of water, which is
consistent with reversible kinetic inhibition. During methanol
carbonylation, parallel dehydration reactions form stoichio-
metric DME and H2O, with the latter inhibiting the carbon-
ylation steps. Inhibition by H2O does not reflect its compet-
Figure 1. The rate of methyl acetate formation (per total Al content) on
H-MOR (Si/Al=10:1; diamonds), H-FER (Si/Al=33.5:1; triangles),
H-MOR (Si/Al=45:1; circles), and H-ZSM5 (Si/Al=12.5:1; squares).
Conditions: 0.93 MPa CO, 20 kPa DME, 50 kPa Ar.
acetate synthesis (per total
Al content) is highest on
H-MOR (Si/Al = 10:1;
Table 1: Steady-state carbonylation rates with methanol, DME, and DME–H2O reactants on H-MOR (Si/Al=
10:1).
Reactants
T [K]
PMeOH or PDME [kPa]
PCO [kPa]
PH O [kPa]
Rate of methyl acetate
2
lower on H-FER (Si/Al =
33.5), H-MOR (Si/Al =
45:1), and H-ZSM5 (Si/
Al = 12.5:1); and unde-
tectable on H-USY (Si/
Al = 3:1), H-BEA (Si/
Al = 12.5:1), and amor-
phous SiO2–Al2O3 (Si/
synthesis [mol (gatomAl)ꢀ1 hꢀ1
]
DME–CO
DME–CO–H2O
DME–CO
DME–CO–H2O
CH3OH–CO
423
423
438
438
423
26
26
0.8–8
8–66.8
3.7
120
120
500
500
86
0
1.1
0
0.5
0
0.034
0.0025
0.50
0.12
not detected (<0.001)
Al = 3:1) at 0.93 MPa CO, 20 kPa DME, and 420–513 K.
The carbonylation rates did not show a clear correlation with
the number of framework or extraframework Al atoms
measured by 27Al MAS NMR spectroscopic analysis, thus
suggesting that neither type of acidic site is sufficient by itself
or uniform in reactivity for carbonylation turnover rates.
Kinetic and transient studies were conducted on H-MOR
(Si/Al = 10:1) under conditions that led to > 99% selectivity
for methyl acetate without detectable deactivation. Figure 2
shows that the rate of methyl acetate synthesis does not
depend on DME pressure (0.8–8.0 kPa), which indicates that
the active sites are saturated with DME-derived intermedi-
ates, and is proportional to CO pressure (0–0.93 MPa), thus
showing that kinetically relevant steps involve reactions of
gas-phase or adsorbed CO with DME-derived intermediates.
The requirement for Brønsted acid sites is consistent with the
stoichiometric effects of 2,6-dimethyl pyridine, a titrant
selective to H+ sites, on the rate of methyl acetate synthesis.
itive adsorption with DME on Brønsted acid sites or the
displacement of DME-derived intermediates by H2O because
the rates remain zero order with respect to DME pressure,
even when H2O is present (Figure 2). H2O competes with CO
for adsorption at the Lewis acid sites; this competition is
possibly required to bind CO coreactants[15–18] or inhibit CO
reactions with adsorbed methyl intermediates. IR spectra
showed that H2O (pre-adsorbed at 298 K) prevents the
binding of CO on Lewis acid sites on H-MOR at 123 K, but
does not influence CO interactions with H+.[19] The involve-
ment of Lewis acid sites in DME carbonylation reactions
remains equivocal at this time.
The potential role of surface methyl intermediates and the
kinetic relevance of the CO addition steps were examined by
DME pulsing studies (addition of DME/Al = 0.73:1–1.00:1)
at 438 K on samples treated in flowing dry air at 773 K. Only
DME and water were detected in the effluent during
subsequent He treatment. DME adsorption ratios per Al
1618
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 1617 –1620

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Angewandte
Chemie
atom were approximately 0.5:1 (ꢁ 0.05) on H-MOR (Si/Al =
6.5:1, 10:1) and H-ZSM5 (Si/Al = 12.5:1; Table 2), which is
consistent with DME reactions with H+ to form two methyl
species [Eq. (1)]. The formation of the methyl species is
species. The initial rate derived from Figure 3b is 0.42 methyl
acetate (gatom Al)ꢀ1 hꢀ1, which is approximately 50% of the
steady-state rate (0.82 methyl acetate (gatom Al)ꢀ1 hꢀ1). The
reasonable agreement between the steady-state and transient
rates indicates that the reaction of CO with the methyl-
saturated sites is the sole kinetically relevant step in DME
carbonylation reactions; the remaining differences may
reflect a nonuniform reactivity of methyl groups or a different
concentration of activated CO near such methyl groups,
which would cause only some methyl species to form methyl
acetate, while steady-state rates on the saturated surfaces
reflect carbonylation rates of the most reactive methyl groups
instead.
CH3OCH3 þ 2 ½SiOðHÞAlŠ Ð 2 ½SiOðCH3ÞAlŠ þ H2O
ð1Þ
consistent with IR and NMR spectral evidence of DME
interactions with H-zeolites.[20–24]
Table 2: Ratios of dosed and adsorbed DME per Al atomon zeolites at
438 K.
A plausible chain-transfer sequence involves initiation
through reaction of DME with H+ to form hydrogen-bonded
DME [Eq. (2)] and methyl groups [Eqs. (3) and (4)] in steps
Zeolite
Si/Al
10:1
10:1
10:1
6.5:1
12.5:1
DMEdosed/Al
DMEads/Al
H-MOR
1.00:1
0.80:1
0.73:1
0.88:1
0.83:1
0.45:1
0.51:1
0.45:1
0.54:1
0.51:1
CH3OCH3 þ ½SiOðHÞAlŠ Ð CH3OCH3 ꢀ ½SiOðHÞAlŠ
CH3OCH3 ꢀ ½SiOðHÞAlŠ Ð ½SiOðCH3ÞAlŠ þ CH3OH
CH3OH þ ½SiOðHÞAlŠ Ð ½SiOðCH3ÞAlŠ þ H2O
ð2Þ
ð3Þ
ð4Þ
H-ZSM5
Interactions of CO with these methyl groups were probed
by replacing DME–CO reactants with pure CO for a specific
time interval and then reintroducing DME–CO mixtures. CO
reacts with methyl-saturated surfaces to form acetyl com-
plexes that desorb only after DME is reintroduced, at which
time the rate of methyl acetate formation increases sharply
and then returns gradually to the steady-state values (Fig-
ure 3a). Exposing methyl-saturated samples to He led to an
immediate return to steady-state rates after DME–CO
mixtures were reintroduced. Precursors to methyl acetate
form slowly during the contact of methyl groups with CO, but
cannot desorb without methylation by DME to form methyl
acetate and a new methyl intermediate.
The number of CH3–CO complexes, measured from the
excess methyl acetate formed as the DME–CO reactions
approach steady-state, increased with CO exposure time
(Figure 3b). The local slope in Figure 3b gives the rate of
CH3–CO formation; the dashed line shows predictions for
rates proportional to the residual number of unreacted methyl
that lead to an induction period during DME–CO reactions
(see the Supporting Information). CO and water adsorb
competitively at sites responsible for CO binding [Eqs. (5)
and (6)], and the propagation steps [Eqs. (6)–(8)] dominate in
*
*
H2Oþ Ð H2O
ð5Þ
ð6Þ
*
*
COþ Ð CO
the absence of water. The catalyst surface becomes saturated
with stable methyl intermediates that react with CO in the
kinetically relevant step to form adsorbed acetyl intermedi-
ates [Eq. (7)]. Methyl acetate and subsequent methyl species
form by reactions of acetyl species with DME [Eq. (8)].
*
*
½SiOðCH3ÞAlŠ þ CO ! ½SiOðCOCH3ÞAlŠþ
ð7Þ
½SiOðCOCH3ÞAlŠ þ CH3OCH3 ! ½SiOðCH3ÞAlŠ þ CH3COOCH3
ð8Þ
In summary, acid zeo-
lites catalyze DME car-
bonylation to methyl ace-
tate at low temperatures
(423–463 K) with high
selectivity and catalyst sta-
bility. DME reacts with
Brønsted protons to form
methyl-saturated surfaces,
which react with CO to
form
These
acetyl
species
moieties.
form
methyl acetate by reaction
with DME, which also
restores
the
methyl
Figure 3. a) The rate of methyl acetate synthesis upon reintroduction of DME–CO reactants on H-MOR
(Si/Al=10:1) after exposure to 0.95 MPa CO for varying intervals (0.93 MPa CO, 20 kPa DME, 50 kPa Ar,
438 K). b) Rate of excess methyl acetate formation (0.93 MPa CO, 20 kPa DME, 50 kPa Ar, 438 K) on H-MOR
(Si/Al=10:1) integrated over time per Al atom as a function of exposure time in 0.95 MPa CO.
groups initially formed by
direct reaction of DME
with H+ during the induc-
Angew. Chem. Int. Ed. 2006, 45, 1617 –1620
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1619

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Communications
tion period. DME carbonylation preserves anhydrous con-
ditions, as conversion occurs and avoids inhibition of carbon-
ylation pathways by H2O, a process that leads to much lower
carbonylation rates when methanol is used as the reactant.
Kinetic and spectroscopic studies are on-going to address the
nature of the CO binding sites responsible for these reactions
and the apparent requirement for both Brønsted and CO
binding centers in DME carbonylation catalysis. The current
productivities are below the expected commercial targets but
do increase linearly with CO pressure. Taken together with
the remarkable selectivity for methyl acetate and the
unprecedented low reaction temperatures, these initial
reports show significant promise for the practical use of this
catalytic chemistry as our knowledge of the nature of CO
binding sites increases.
[1] F. E. Paulik, J. F. Roth, J. Chem. Soc. Chem. Commun. 1968,
1578.
[2] J. F. Roth, J. H. Craddock, A. Hershman, F. E. Paulik, Chem.
Technol. 1971, 600.
[3] G. J. Sunley, D. J. Watson, Catal. Today 2000, 58, 293.
[4] B. Ellis, M. J. Howard, R. W. Joyner, K. N. Reddy, M. B. Padley,
W. J. Smith, Stud. Surf. Sci. Catal. 1996, 101, 771.
[5] T. Li, N. Tsumori, Y. Souma, Q. Xu, Chem. Commun. 2003, 2070.
[6] W. J. Smith (BP Chemicals Limited), US 5420345, 1995.
[7] A. G. Stepanov, M. V. Luzgin, V. N. Romannikov, V. N. Side-
lnikov, K. I. Zamaraev, J. Catal. 1996, 164, 411.
[8] Q. Xu, S. Inoue, N. Tsumori, H. Mori, M. Kameda, M. Tanaka,
M. Fujiwara, Y. Souma, J. Mol. Catal. A 2001, 170, 147.
[9] K. Fujimoto, T. Shikada, K. Omata, H. Tominaga, Chem. Lett.
1984, 2047.
[10] A. Sardesai, S. Lee, T. Tartamella, Energy Sources 2002, 24, 301.
[11] G. G. Volkova, L. M. Plyasova, A. N. Salanov, G. N. Kustova,
T. M. Yurieva, V. A. Likholobov, Catal. Lett. 2002, 80, 175.
[12] G. G. Volkova, L. M. Plyasova, L. N. Shkuratova, A. A. Bud-
neva, E. A. Paukshtis, M. N. Timofeeva, V. A. Likholobov, Stud.
Surf. Sci. Catal. 2004, 147, 403.
[13] R. W. Wegman, J. Chem. Soc. Chem. Commun. 1994, 947.
[14] A. G. Stepanov, M. V. Luzgin, A. V. Krasnoslobodtsev, V. P.
Shmachkova, N. S. Kotsarenko, Angew. Chem. 2000, 112, 3804;
Angew. Chem. Int. Ed. 2000, 39, 3658.
Experimental Section
Amorphous silica alumina (Si/Al = 3:1) was obtained from Aldrich.
NH4-MOR (Si/Al = 10:1; Zeolyst), H-MOR (Si/Al = 45:1; Zeolyst),
NH4-FER (Si/Al = 33.5:1; Zeolyst), and NH4-BEA (Si/Al = 12.5:1;
Zeolyst), NH4-ZSM5 (Si/Al = 12.5:1; AlSi-Penta Zeolithe), and NH4-
USY (Si/Al = 3:1; Engelhard) were treated in flowing dry air at 773 K
for 3 h to form acid zeolites. Na-MOR (Si/Al = 6.5:1; Zeolyst) was
converted into its NH4 form by exchanging Na-MOR (10 g) with 1m
NH4NO3 (4” 0.2 L) at 353 K for 12 h with washing and filtering in
deionized water (0.2 L) after each exchange. After the last exchange,
the samples were dried overnight in ambient air at 393 K before
treatment in flowing dry air for 3 h at 773 K.
[15] M. Maache, A. Janin, J. C. Lavalley, E. Benazzi, Zeolites 1995,
15, 507.
[16] L. Benco, T. Bucko, J. Hafner, H. Toulhoat, J. Phys. Chem. B
2004, 108, 13656.
[17] J. A. van Bokhoven, A. M. J. van der Eerden, D. C. Konings-
berger, J. Am. Chem. Soc. 2003, 125, 7435.
Carbonylation rates and selectivities were measured using a
packed-bed stainless steel reactor (9.5 mm OD) equipped with a
multipoint thermocouple held within a 1.6-mm outer thermowell
aligned along the tube center. Catalysts (0.2–0.6 g, 125–250 mm) were
treated in flowing dry air ( ꢂ 1.67 cm3 sꢀ1 gꢀ1) for 2 h at 773 K and
cooled to reaction temperature (420–513 K) in flowing He (UHP,
Praxair) before introducing 2% DME/93% CO/5% Ar (99.5%
DME, UHP CO/Ar; Praxair), 16.7% DME/CO (99.5% DME, UHP
CO; Praxair), and/or 95% CO/Ar (UHP, Praxair). Methanol was
added through a saturator using He as the carrier gas, and water was
added by a syringe pump (Cole-Parmer, Model 100 series) or by
reaction of 1.25% H2/Ar (UHP, Praxair) with CuO at 673 K. Heated
lines (423–473 K) transferred the reactor effluent to a mass spec-
trometer (MKS Spectra Minilab) and a gas chromatograph (Agilent
[18] A. Zecchina, C. O. Arean, Chem. Soc. Rev. 1996, 25, 187.
[19] P. Cheung, A. Bhan, E. Iglesia, unpublished results.
[20] W. Wang, M. Seiler, M. Hunger, J. Phys. Chem. B 2001, 105,
12553.
[21] W. Wang, A. Buchholz, M. Seiler, M. Hunger, J. Am. Chem. Soc.
2003, 125, 15260.
[22] I. I. Ivanova, A. Corma, J. Phys. Chem. B 1997, 101, 547.
[23] S. M. Campbell, X. Z. Jiang, R. F. Howe, Microporous Meso-
porous Mater. 1999, 29, 91.
[24] V. A. Veefkind, J. A. Lercher, Appl. Catal. A 1999, 181, 245.
6890) equipped with
a methyl silicone column (HP-1, 50 m ”
0.32 mm ” 1.05 mm) connected to a flame-ionization detector and a
Porapak Q column (80–100 mesh, 12 ft. ” 1/8 in.) connected to a
thermal conductivity detector.
Transient experiments were carried in the same equipment. He
( ꢂ 3.34cm 3 sꢀ1 gꢀ1, UHP; Praxair) was used to introduce DME
(99.5%, Praxair) pulses (1.07 cm3) at 120-s intervals. Samples were
then flushed with He ( ꢂ 3.34cm 3 sꢀ1 gꢀ1) for 1.5–2 h to remove
physisorbed DME and any water formed. CO, Ar, and He streams
were purified of oil and water (Matheson, 451) and of metal carbonyls
(Matheson, 454) prior to addition to mixtures containing DME. These
streams were further dried using CaH2 (Aldrich, 99%) held at
ambient temperature.
Received: November 4, 2005
Published online: January 30, 2006
Keywords: carbonylation · carboxylic acids · ethers ·
.
heterogeneous catalysis · zeolites
1620 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 1617 –1620