Vapor-phase ethanol carbonylation with heteropolyacid-supported Rh

Article abstract of DOI:10.1016/j.jcat.2015.02.004

Ethanol carbonylation is a potential route to valuable C3 products. Here, Rh supported on porous, Cs-exchanged heteropolyacid Cs₃PW₁₂O_40, is demonstrated as an effective catalyst for vapor-phase ethanol carbonylation, with higher selectivity and conversion to propionates than existing catalysts. Residual acidity or a Mo polyatom was strongly detrimental to yields. Propionate selectivity was maximized at 96% at 170 °C and with added H₂O. The catalyst displayed stable selectivity over 30 h on stream and up to 77% conversion. Ethyl iodide is a required co-catalyst but at levels as low as 2% relative to ethanol. XPS and in situ XANES indicate partial Rh reduction, consistent with the formation of low-valent reactive intermediates and slow deactivation through formation of Rh nanoparticles. With further optimization and understanding, these Rh/heteropolyacid catalysts may lead to stable and selective catalysts for the production of propionates through ethanol carbonylation.

Full text of DOI:10.1016/j.jcat.2015.02.004

Journal of Catalysis 325 (2015) 1–8
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Vapor-phase ethanol carbonylation with heteropolyacid-supported Rh
a
a,1
b
b
a,
⇑
Sara Yacob , Sunyoung Park , Beata A. Kilos , David G. Barton , Justin M. Notestein
a
Northwestern University, Department of Chemical and Biological Engineering, Evanston, IL 60208, USA
The Dow Chemical Company, Core R&D, Midland, MI 48674, USA
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Ethanol carbonylation is a potential route to valuable C3 products. Here, Rh supported on porous,
Cs-exchanged heteropolyacid Cs PW12 40, is demonstrated as an effective catalyst for vapor-phase etha-
nol carbonylation, with higher selectivity and conversion to propionates than existing catalysts. Residual
acidity or a Mo polyatom was strongly detrimental to yields. Propionate selectivity was maximized at
Received 26 November 2014
Revised 29 January 2015
Accepted 10 February 2015
3
O
2
96% at 170 °C and with added H O. The catalyst displayed stable selectivity over 30 h on stream and
up to 77% conversion. Ethyl iodide is a required co-catalyst but at levels as low as 2% relative to ethanol.
XPS and in situ XANES indicate partial Rh reduction, consistent with the formation of low-valent reactive
intermediates and slow deactivation through formation of Rh nanoparticles. With further optimization
and understanding, these Rh/heteropolyacid catalysts may lead to stable and selective catalysts for the
production of propionates through ethanol carbonylation.
Keywords:
Carbonylation
Heteropolyacid
Ethanol
Iodide
Rhodium
Ó 2015 Elsevier Inc. All rights reserved.
1
. Introduction
Alcohol carbonylation is an important commercialized process,
carbon [23,24] or ligand-modified oxides [25] for carbonylation of
methanol in the vapor phase. Several other insoluble resins have
been developed for methanol carbonylation, both in the vapor
and liquid phases. Critically, Christensen and Scurrell found
decreasing selectivity for propionates with increasing ethanol con-
version [15,16]. This was not due to sequential reactions, but
rather due to the strong competition from ethanol dehydration
to ethylene and diethyl ether. Although the references cited above
demonstrate that ethylene and ethers can participate in carbonyla-
tion chemistry, the water that forms as a co-product appears to be
detrimental to carbonylation reactivity in those systems. Given
ethanol’s strong propensity to dehydrate, this must be addressed
for any acid catalyst [26,27]. Therefore, there is a need for develop-
ment and testing of new catalyst formulations that suppress etha-
nol dehydration and promote carbonylation in the vapor phase.
An important other class of carbonylation catalysts are those
based on heteropolyacids. Heteropolyacids have been widely used
as homogeneous and heterogeneous acid and redox catalysts
[28–30]. Limited reports have shown that these catalysts are cap-
able of methanol [31,32] or dimethyl ether [33] carbonylation to
acetic acid and methyl acetate, respectively. Both reports consid-
ered it essential to have a Brønsted acidic support. There are many
structural classes of heteropolyacids such as Keggin [34], Wells–
Dawson [35], Finke–Droege [36], and Pope–Jeannin–Preyssler
particularly for the production of acetic acid from methanol, which
is one of the largest industrial processes to use homogeneous
catalysts [1–4]. A common system for alcohol carbonylation uses
a group VIII metal complex and an iodide co-catalyst in the con-
densed phase. Alkyl iodides are formed in situ, which oxidatively
add to the transition metal complex, followed by CO insertion into
the M–C bond, and reductive elimination [5–7]. Much less studied
are processes for the ethanol carbonylation to propionates, alcohol
carbonylation in the vapor phase, or alcohol carbonylation using
supported, heterogeneous catalysts. Vapor-phase heterogeneous
catalysts include the mechanistically distinct carbonylation of
methanol and dimethyl ether with mordenite-type zeolites
[
[
8–11], and variants on the Reppe carbonylation of ethylene
10–14]. Only a very limited number of studies have investigated
solid catalysts for ethanol carbonylation to propionic acid or ethyl
propionate, and all are for group VIII metals promoted with alkyl
iodides [15–18]. Both Nefedov and Scurrell separately showed that
Rh supported on X-type zeolites are effective heterogeneous
catalysts for vapor-phase alcohol carbonylation with added iodide
[
15–22]. Rh has also been supported on various materials including
[
37]. Among the various heteropolyacids, Keggin heteropolyacids
⇑
Corresponding author. Fax: +1 847 491 3728.
E-mail address: j-notestein@northwestern.edu (J.M. Notestein).
Current address: Korean Research Institute of Chemical Technology, Yueseong-gu,
have been broadly investigated because they are the most stable
and readily synthesized. Keggin heteropolyacids are comprised of
heteropolyanions of the formula [XM12
1
nÀ
O
40
]
, where X is the
Deajeon 305-600, South Korea.
http://dx.doi.org/10.1016/j.jcat.2015.02.004
0
021-9517/Ó 2015 Elsevier Inc. All rights reserved.

2
S. Yacob et al. / Journal of Catalysis 325 (2015) 1–8
heteroatom (P, Si, etc.) and M is the polyatom (W, Mo, etc.)
28–30,34]. Changing the heteroatom and polyatom controls the
acidic and redox properties of these heteropolyacids, and these
properties have been investigated in various catalytic reactions
A silica-supported 1 wt% Rh/Cs
3
PW12
2
O40/SiO catalyst was syn-
[
thesized as follows. 1 g H
(Selecto Scientific, ꢀ500 m /g, ꢀ6 nm pore diameters, 40–63
3
PW12
O40 was loaded onto 2 g of silica gel
2
lm
particle size) by incipient wetness impregnation from water, dried
[
29,30]. The use of H
3
PW12
O
40 in alcohol and ether carbonylation
at 150 °C for 1–3 h, and calcined at 300 °C for 2 h. 0.19 g of CsNO
was dissolved in 30 mL of distilled water. 1.9 g of H 40/SiO
was then dispersed in the solution with constant stirring, and the
mixture was stirred for 12 h. The solid product was filtered,
washed with distilled water, dried at 150 °C for 1–3 h, and calcined
3
has been reported [38]. Heteropolyacids are highly soluble in polar
solvents including water and alcohols, while heteropolyacid salts
with large cations are insoluble and have high surface area by
forming a tertiary structure [28,29]. Examples of cations common-
ly used include K , Rb , Cs , and NH
heteropolyacid salt can be tuned by the amount of the large cation
exchanged on the heteropolyanion, leading to wide use as solid
3
PW12O
2
+
+
+
+
4
, and the acidity of a
at 350 °C for 2 h to obtain Cs
was then supported onto Cs
impregnation as above at a loading of 1 wt% relative to the Cs
40. The impregnated solid was dried at 150 °C for 1–3 h
3
PW12
O40/SiO
2
3 2 x
support. RhCl (H O)
3
PW12O
40/SiO
2
by incipient wetness
3
-
+
acid catalysts [28]. Cs -modified heteropolyacids have been
PW12O
demonstrated in dimethyl ether carbonylation [39].
and calcined at 350 °C for 2 h.
This work describes the novel application of Rh supported on
heteropolyacids as a promising heterogeneous catalyst for vapor-
phase ethanol carbonylation, including identification of suitable
operating conditions and catalyst compositions, spectroscopic ana-
lysis of the active catalyst, and mechanistic insight into the work-
ing state of the catalyst. The mechanism appears to be broadly
analogous to that of condensed-phase methanol carbonylation by
similar, soluble catalysts. As compared to existing catalysts known
for vapor-phase ethanol carbonylation, the catalysts described here
have higher selectivity to the desired carbonylation products,
including at high conversion.
2.2. Catalyst characterization
Cs and W contents in the catalysts were determined by induc-
tively coupled plasma-mass spectrometry (ICP-MS). Rh contents
on the catalysts were measured by inductively coupled plasma-
atomic emission spectrometry (ICP-AES) analysis. Materials were
found to dissolve well in a solution of 95 wt% water, 3 wt% nitric
acid, 1 wt% hydrofluoric acid, and 1 wt% hydrochloric acid. For
calibration, samples were compared to known concentrations of
Rh, Cs, and W in the same stock solutions that samples were dis-
solved in. All elemental ICP standard solutions were acquired from
Sigma–Aldrich at an original concentration of 1000 ppm.
2
. Experimental
Surface areas of the catalysts were measured using N
physisorption on Micromeritics 2010 ASAP. Prior to
2
a
N
2
2.1. Catalyst preparation
physisorption, all samples were degassed under vacuum overnight
at 140 °C. X-ray diffraction (XRD) patterns of the catalysts were
Heteropolyacid (HPA) salts were prepared through an
acquired using Cu Ka radiation operated at 40 kV and 20 mA using
ion-exchange method. Keggin HPAs, phosphotungstic acid
O)x, and phosphomolybdic acid H PMo12 O)
were purchased and used as received from Sigma–Aldrich. Cs or
a Rigaku Geigerflex X-ray powder diffractometer. X-ray photoelec-
tric spectroscopy (XPS) was performed on a Thermo Scientific
ESCALAB 250 Xi. Peak locations were identified using Advantage
Software v5.5.3. The spectrometer binding energy was calibrated
through energy shifts to the reference C 1s (284.9 eV). Thermo-
gravimetric analysis (TGA) was done on a Q500 from TA Instru-
ments. An 80 mg sample was subjected to a temperature ramp
from ambient to 350 °C at a rate of 5 °C/min under oxygen flow.
Rh K-edge X-ray absorption near edge structure (XANES) spec-
tral measurements were performed at the Advanced Photon
Source, Argonne National Laboratory. The DuPont–Northwestern–
Dow Collaborative Access Team (DND-CAT) bending magnet D
beamline at Sector 5 was used. XANES spectra were recorded in
transmission mode by employing a Si(111) double crystal
monochromator. Transmission intensities were measured with
Canberra ionization chambers. For the Rh K-edge, all XANES spec-
tra were scanned in the range of 23,100–23,450 eV. The photon
energy for each scan was calibrated using Rh foil, setting the first
H
3
PW12
O
40(H
2
3
O40(H
2
x
+
+
+
NH
with stoichiometries from x = 1.5–3.0 in Cs
Cs (NH 40 to cover the range where surface area and
acidity vary significantly [28]. NH
stoichiometries from x = 0–1.5 in Cs (NH
purified to 18 M resistivity using a Barnstead Nanopure Infinity
system and passed through a 0.2 m filter before use. A known
amount of cesium nitrate (CsNO , Sigma–Aldrich) and/or ammoni-
um nitrate (NH NO , Sigma–Aldrich) was dissolved in 20 mL of dis-
tilled water to form 0.13–0.15 M solutions and slowly added to
0 mL of 0.05 M aqueous solution of phosphotungstic acid or phos-
4
was used as exchange cations. Cs was incorporated
x
H3ÀxPW12O40 or
x
4
)3ÀxPW12O
+
4
was incorporated with
4
)3ÀxPW12O40. Water was
x
X
l
3
4
3
2
phomolybdic acid while stirring. The resulting composite solution
was heated at 60 °C overnight to obtain a solid product. The solid
product was dried for 1–3 h in a 150 °C oven and then calcined
at 350 °C in static air for 2 h ramping at 5 °C/min.
0
Rh was then loaded on HPA salts by incipient wetness impreg-
inflection point at the known edge energy of Rh , 23,220 eV. 2–5
nation using RhCl
3
(H
2
O)
x
obtained from Sigma–Aldrich and used as
scans of each reference were averaged to optimize signal-to-noise
ratio. XANES of the as-synthesized catalyst was compared to
received. Rh content was controlled in the range of 0.5–5 wt%. The
impregnated solid was dried for 1–3 h in a 150 °C oven and then
calcined at 350 °C in static air for 2 h ramping at 5 °C/min. A com-
parison Rh catalyst was synthesized on Na13X zeolite by ion
metallic Rh, Rh
2 3 3 2 x 3
O , RhCl (H O) , and RhI .
Temperature-programmed desorption (TPD) of NH
3
was carried
out in an Altamira Instruments AMI-200 to measure catalyst acid-
ity. 0.1 g of each catalyst charged into a quartz reactor and pre-
treated at 200 °C for 1 h with a stream of He (25 mL/min). A
3 2 x
exchange between RhCl (H O) and Na13X molecular sieve (Sig-
ma–Aldrich). The Rh precursor was dissolved in water, and a
known amount of Na13X molecular sieve was added such as to
make 1 wt% Rh. The resulting solution was heated to 80 °C and stir-
red overnight. Filtration yielded solid particles, which were
washed with ꢀ200 mL of purified water and dried in a 150 °C oven
overnight. After drying, the catalyst was calcined at 400 °C in static
air for 2 h ramping at 5 °C/min. These materials showed no indica-
tion of framework disruption by 27Al MAS SS NMR (Appendix A,
Fig. S1).
mixed stream of NH
3
and He (25 mL/min) was then introduced
into the reactor at 50 °C for 30 min. Physisorbed NH
3
was removed
at 100 °C for 1 h under a flow of He (25 mL/min). After cooling the
catalyst, furnace temperature was increased from 50 °C to 600 °C
at a heating rate of 20 °C/min under a flow of He (25 mL/min). Des-
orbed NH
(TCD). For calibration, known amounts of NH
the empty reactor under a flow of He (25 mL/min).
3
was detected using a thermal conductivity detector
3
were injected into

S. Yacob et al. / Journal of Catalysis 325 (2015) 1–8
3
2
.3. Ethanol carbonylation flow reactor
3. Results and discussion
3.1. Catalytic performance of Rh/(Cs,NH ) H PM12O40 (M = W, Mo)
Gas-phase reactions were carried out in quartz downflow
4
x 3Àx
microreactors in an Altamira Instruments BenchCAT 4000R-HP.
Mixtures of ethanol and ethyl iodide co-catalyst were delivered
via liquid syringe pump (KD Scientific KDS-100) at total liquid flow
A series of Rh-containing catalysts were prepared and com-
pared for activity in vapor-phase EtOH carbonylation (Table 1). In
this section, the ratio of CO:EtOH:EtI in the feed was fixed at
164:10:1, the WHSV was held constant at 24 h , and the tem-
perature was 170 °C. All reactions are at nominal 1 atmosphere
rates of 2–5 lL/min. The ethanol-to-ethyl iodide ratio was con-
À1
trolled in the range of 5–100:1. The liquid feed was vaporized in
line with CO reactant. CO-to-ethanol molar ratios varied from 2
to 18:1, and CO flow rates were varied between 5 mL/min and
total pressure.
+
+
+
+
3
0 mL/min. The catalyst bed consisted of 0.1–1.5 g of catalyst held
4
Cs /H - and Cs /NH -exchanged HPAs were used as supports for
1 wt% Rh catalysts. The use of either Cs or NH during synthesis
resulted in surface areas >160 m /g, as compared to H PW12O40
with a surface area <10 m /g. This shows that the porous, tertiary
+
+
in place by quartz wool, and the reactions were conducted at reac-
tor temperatures of 150–210 °C and nominally at atmospheric
pressure. No pretreatment was performed on the catalyst bed
unless otherwise stated. Weight hourly space velocities (WHSVs)
were obtained from total mass flow rate divided by mass of
catalyst.
4
2
3
2
structure was developed correctly [28]. N2 physisorption curves
+
+
are shown in Appendix A Fig. S2. After calcination, the Cs /NH
4
+
+
form is chemically indistinguishable from the Cs /H form. Regard-
less of whether H or NH was used during synthesis, 1 wt% Rh/Cs -
4 x
+
+
Reactant and product concentrations were measured every
2
0 min during time on stream (TOS) using an Agilent 7890A GC-
FID with helium carrier through HP InnoWax column
50 m  0.2 mm  0.4 m). The GC program included a multi-step
ramp from 60 to 200 °C. Ethylene, diethyl ether (DEE), ethyl iodide
EtI), ethanol (EtOH), ethyl propionate (EP), and propionic acid (PA)
H3 PW12O40 (x = 0, 1.5, 2.0, 2.5, 2.8, 3.0) gave EtOH conversions
that decreased monotonically with increasing Cs exchange at a
fixed WHSV. However, for all but the fully exchanged sample, only
dehydration products were synthesized with appreciable selec-
Àx
+
a
(
l
(
tivity. The marked increase in EP selectivity for Rh/Cs PW12O40
3
eluted in that order and were the only products quantified, and
they were used to report an ethyl species mass balance. CO and
(from 2.3% for Cs2.5, to 5.9% for Cs2.8, to 81.6% for Cs ) was due to
both a significant decrease in rate of dehydration and also a sig-
3
CO
this mass balance was 100 ± 10% and formally defined as: (nEtOH
EtI + nethylene + 2nDEE + 2nEP + nPA)/(nEtI,0 + nEtOH,0). Not able to be
2
are not included in this mass balance. Unless otherwise noted,
nificant rise in the rate of EP production. The marked drop in dehy-
dration products parallels the complete loss of acid sites for Rh/
+
n
Cs PW12O40. The x = 2.5 and 2.8 materials still contain 38 and
3
+
quantified were amounts of CO, CO
2
, H
2
, or H
2
O due to the use of
25 lmol titratable H sites per gram of catalyst, respectively,
an FID detector. Conversions are reported relative to EtOH and
EtI in an empty reactor and are generally ±5% over multiple trials
and for a given catalyst over several hours at steady operation.
Selectivity is reported relative to the total moles of products
detected, as opposed to ethanol converted, and is formally given
whereas x = 3.0 contains no significant acidity. NH -TPD curves
for x = 2.5, 2.8, and 3 are shown in Appendix A Fig. S3. In the
3
absence of Rh, all of these HPA salts gave only dehydration prod-
ucts, indicating that the Rh species is required for CO adsorption
and activation [31]. For increasing Rh between 0.5 and 5 wt%, EtOH
conversion slightly increased throughout the range, with EP selec-
tivity changing little above 1 wt%. For this reason, the 1 wt% load-
ing was the focus of all other catalytic investigations.
x x
as n /(nethylene + nDEE + nEP + nPA), where n refers to the molar con-
centrations of a given species. Unless otherwise stated, catalyst
performance is given as nominally steady state values averaged
over all data from 40 min to 4–5 h TOS at a given condition. In early
studies, it was seen that catalyst changes were slow after 4 h.
An important role of the polyatom was seen for 1 wt%
Rh/Cs PM12O40 with M = W or Mo. Under standard conditions,
3
Table 1
4 x
) H3ÀxPM12O
40 (M = W, Mo).a
EtOH carbonylation with Rh/(Cs,NH
2
Catalyst
S.A. (m /g)
Acidity (lmol-NH
3
/g)
% Conversion
% Selectivity
Ethylene
Mass balance (%)
EtOH
EtI
DEE
EP
PA
1
1
1
1
1
1
1
wt% Rh/(NH
wt% Rh/Cs1.5(NH
wt% Rh/Cs (NH
wt% Rh/Cs2.5(NH
4
)
3
PW12
1.5PW12
PW12
40
0.5PW12
O
40
201
189
170
165
159
159
162
87.7
90.0
88.2
66.3
52.6
38.0
8.7
4.8
6.9
4.6
7.0
2.7
<1
83.7
84.1
86.3
43.5
39.5
24.8
11.4
10.6
9.5
9.3
52.3
57.9
69.3
7.0
4.2
4.5
3.3
3.0
2.3
5.9
81.6
1.5
2.0
1.1
1.2
0.4
0
112.5
111.4
108.3
89.8
93.9
92.1
4
)
O
40
2
4
)
1
O
4
)
O
40
wt% Rh/Cs2.5
wt% Rh/Cs2.8
H
H
0.5PW12
0.2PW12
O
O
40
40
37.7
25.4
Trace
wt% Rh/Cs
3
PW12
O
40
6.1
0
95.3
Cs2.5
Cs2.8
H
H
0.5PW12
0.2PW12
40
O
O
40
40
64.0
52.0
<1
3.0
<1
<1
43.6
13.4
17.9
56.4
86.6
82.1
0
0
0
0
0
0
93.7
99.3
102.4
Cs
3
PW12
O
0
1
3
5
.5 wt% Rh/Cs
3
PW12
O
40
6.8
8.7
11.0
11.3
8.4
6.1
11.1
7.4
22.8
11.4
9.8
4.8
7.0
10.2
10.1
72.5
81.6
80.0
78.4
0
0
0
0
95.9
95.3
93.5
92.8
wt% Rh/Cs
wt% Rh/Cs
wt% Rh/Cs
3
PW12
3
PW12
3
PW12
O
O
O
40
40
40
11.4
1
1
wt% Rh/Cs
wt% Rh/Cs
3
PMo12
PMo12
O
O
40
1.9
2.3
<1
2.8
26.4
28.7
7.4
12.1
20.5
29.4
45.7
29.9
99.0
98.0
b
3
40
1
1
wt% Rh/Cs
3
PW12
O
40/SiO
2
23.4
12.3
<1
14.2
59.9
81.5
28.4
2.9
1.3
0
93.5
94.2
wt% Rh/Na13X
890
850
9.9
11.7
a
À1
Reaction conditions: CO:EtOH:EtI = 164:10:1, WHSV = 24 h , 170 °C, average of 5 h TOS.
b
À1
WHSV = 1.6 h , average of 7 h TOS, otherwise same conditions.

4
S. Yacob et al. / Journal of Catalysis 325 (2015) 1–8
À1
EtOH conversion was very low over Rh/Cs
onate selectivities were somewhat lower than for Rh/Cs
3
PMo12
O
40 and propi-
40
PW12O .
conditions were 1 atm nominal total pressure, 170 °C, 24 h , 5 h
time on stream, and a feed ratio of CO:EtOH:EtI = 164:10:1. A sin-
gle parameter was varied at a time. First, the temperature was con-
trolled between 150 and 210 °C increasing in 20 °C increments, and
the results are shown in Fig. 1. EtOH conversion increased with
increasing reaction temperature up to 190 °C. Ethylene selectivity
increased monotonically, while DEE selectivity decreased. Car-
bonylation selectivity (EP + PA) passed through a clear maximum
at 170 °C. PA was only detectable at 210 °C, presumably a conse-
quence of the high water partial pressure in the reactor outlet aris-
ing from the high conversion to ethylene and attendant release of
water.
3
At lower space velocities, neither conversions nor PA selectivities
improved, and the catalyst became dark blue after 7 h on stream,
indicating reduction of the phosphomolybdic acid. It has been
reported that acidic and redox properties of HPAs can be controlled
by changing polyatom [29,30]. Since both catalysts had all acidity
titrated away, the electronic/redox properties of the W-containing
HPA anion itself are specific for increasing stability, activity, and
selectivity in EtOH carbonylation.
+
One material was prepared consisting of H
3
PW12
O
40, Cs , and
Rh3+ sequentially deposited on silica gel. EtOH conversion over
Rh/Cs 40/SiO was higher than that over unsupported Rh/
Cs 40, while EP selectivity was much lower than that over
unsupported Rh/Cs 40. Even though the composition of this
3
PW12
PW12
O
2
Second, the dependence of CO partial pressure was examined at
1 atm and below, and at standard conditions, the results are shown
in Table 2. N replaced CO to maintain constant WHSV. EtOH con-
2
3
O
3
PW12
O
catalyst is in the same ratios as for the unsupported HPA salts, use
of the silica support may lead to incomplete exchange of Cs for
acid sites and poor contact between Rh and the HPA salt, which
would both be detrimental to propionate selectivity.
Finally, Table 1 also includes a benchmark 1 wt% Rh/Na13X cat-
alyst, known for EtOH carbonylation [15–18]. The Rh/Na13X gave
EtOH conversions and propionate selectivity comparable to pub-
lished work (e.g. 12% selectivity) and markedly inferior to the
results disclosed here [18]. Selectivity patterns of this reference
catalyst appear to be weakly impacted by calcination conditions
version is essentially independent of CO partial pressure over this
range, consistent with the rate limiting step over Rh-iodides being
the oxidative addition of the ethyl species [5,7,15,16]. EP selectivity
begins to decrease for partial pressures below 0.6 atm, indicating a
nearly CO-saturated catalyst under typical conditions. Similar
dependences have been seen for molecular catalysts in solution,
consistent with the elimination of most of the parallel, acid-cat-
alyzed reactivity of the support [1,2,4].
Table 3 shows the effect of EtOH:EtI molar feed ratio for values
between 5:1 and 100:1, as well as for pure EtI and EtOH liquid
feeds, at otherwise standard conditions and constant total liquid
flow rates. With only EtOH or EtI, the catalyst was essentially inert,
with neither carbonylation nor dehydration products observed.
Without the iodide co-catalyst, the ethyl species is assumed to
be unable to enter the Rh catalytic cycle. The lack of products in
the EtOH-free case indicates that the acyl iodide of classical Rh
mechanisms does not appear to be able to exit from the catalytic
cycle, or it indicates that the catalyst rapidly deactivates under
these high halogen feed conditions. The moderately acidic
+
(Appendix A, Table S1).
3.2. Effect of temperature and feed composition
3
A 1 wt% Rh/Cs PW12O40 catalyst was used to investigate the
dependence of temperature and other process conditions. Standard
1
00%
9
8
7
6
5
4
3
2
1
0%
0%
0%
0%
0%
0%
0%
0%
0%
Rh/Cs2.5H0.5PW12O40 catalyst is similarly unreactive with EtI and
results only in dehydration products for EtOH-only feeds. With
the mixtures of both EtOH and EtI, neither EtOH conversion nor
EP selectivity changed substantially until the EtOH:EtI ratio
Table 3
3
EtOH carbonylation with 1 wt% Rh/Cs PW12O
40 and various EtOH:EtI feed ratios.a
EtOH:EtI
molar
ratio
EtOH
conv.
(%)
EtI
conv.
(%)
Ethylene
sel. (%)
DEE
sel.
(%)
EP
sel.
(%)
PA
sel.
(%)
Mass
balance
(%)
EtIb
EtI
–
–
8.7
5.8
<1
4.6
100
0
13.0
0
0
7.8
0
0
79.2
0
0
0
102.5
104.6
96.7
5:1
0
%
10:1
20:1
8.7
8.1
8.2
5.6
1.4
6.1
7.0
3.4
2.5
–
11.4
9.6
10.4
9.6
0
40.8
7.0
7.6
8.3
10.2
0
81.6
82.8
81.4
80.2
0
0
0
0
0
0
0
95.3
95.3
93.7
95.6
98.6
93.9
1
50
170
190
210
5
1
0:1
00:1
Temperature (°C)
EtOH
Fig. 1. EtOH carbonylation with 1 wt% Rh/Cs
3
PW12
O
40 at various temperatures and
EtOHb
67.3
–
59.2
0
differential
conversion.
Reaction
conditions:
CO:EtOH:EtI = 164:10:1,
À1
a
À1
WHSV = 24 h , average of 5 h TOS. Ethyl species mass balance 95 ± 5%. ꢀ EtOH
conversion, j ethylene selectivity, N DEE selectivity, and d EP + PA selectivity.
Reaction conditions: WHSV = 24 h , 170 °C, average of 5 h TOS.
With 1 wt% Rh/Cs2.5H0.5PW12O
40.
b
Table 2
3
EtOH carbonylation with 1 wt% Rh/Cs PW12O
40 at various CO partial pressures.a
CO partial pressure
CO:EtOH molar ratio
EtOH conv. (%)
EtI conv. (%)
Ethylene sel. (%)
DEE sel. (%)
EP sel. (%)
PA sel. (%)
Mass balance (%)
0
0
0
0
0
.92b
.61
.46
.31
.15
16.4:1
10.9:1
8.2:1
5.5:1
2.7:1
8.7
9.4
9.2
7.6
9.9
6.1
5.4
5.2
6.5
3.0
11.4
13.9
23.4
34.6
70.7
7.0
3.4
3.9
4.7
4.8
81.6
82.7
72.3
60.7
24.5
0
0
0
0
0
95.3
95.0
95.3
95.6
92.6
a
À1
Reaction conditions: EtOH:EtI = 10:1, WHSV = 24 h , 170 °C, average of 5 h TOS.
b
At these conditions, CO conversion is approximately 0.5% estimated from detected carbonylation products.

S. Yacob et al. / Journal of Catalysis 325 (2015) 1–8
5
Table 4
EtOH carbonylation with 1 wt% Rh/Cs
3
PW12
O
40 and added H
2
O.a
EtI conv. (%)
CO:H
2
O:EtOH:EtI molar ratio
EtOH conv. (%)
Ethylene sel. (%)
DEE sel. (%)
EP sel. (%)
PA sel. (%)
Mass balance (%)
1
1
1
1
6
64:0:10:1
8.7
8.6
5.0
3.4
–
6.1
6.9
6.5
2.5
11.4
7.5
4.3
1.6
4.3
7.0
10.3
4.8
1.8
0.4
81.6
77.6
32.8
10.4
0.3
0
4.6
58.2
86.2
72.2
95.3
93.3
97.7
105.1
87.8
64:19:10:1
64:37:10:1
64:56:10:1
0:9:0:1b
25.0
a
À1
Reaction conditions: WHSV = 24 h , 170 °C, average of 5 h TOS.
b
Not shown in the table is 22.9% selectivity toward EtOH.
Table 5
EtOH carbonylation with 1 wt% Rh/Cs
3
PW12
O
40 under various WHSV.a
EtI
À1
WHSV (h
)
Conversion %
EtOH
Selectivity %
Mass balance (%)
Ethylene
DEE
EP
PA
2
1
3
1
1
4
6
.2
.6
.1
8.7
10.7
33.9
59.9
77.2
6.1
5.4
5.5
4.4
1.2
11.4
14.9
17.9
12.6
14.4
7.0
5.7
6.8
8.1
4.8
81.6
79.4
75.3
79.3
80.8
0
0
0
0
0
95.3
95.4
88.0
79.9
72.6
a
Reaction conditions: CO:EtOH:EtI = 164:10:1, 170 °C, average of 5 h TOS.
exceeded 50:1. This catalyst is able to produce EP selectively at
lower levels of iodide than previously reported catalysts for EtOH
carbonylation [15].
100%
9
0%
Finally, 1 wt% Rh/Cs
of H O concentration on EtOH carbonylation at standard condi-
tions as shown in Table 4. The CO:EtOH:EtI feed ratio was fixed
at 164:10:1. H O was added as a second liquid feed at in the feed
ratio range of 0 – 56:10 for H O:EtOH. Selectivity to dehydration
products decreased with increasing H O feed, consistent with the-
se reactions being reversible. Ethylene and DEE were almost sup-
pressed completely at the highest H O feed rates, and without
3
PW12
O40 was used to determine the effect
80%
2
7
0%
60%
2
2
5
0%
2
40%
2
3
2
1
0%
0%
0%
the competitive dehydration, the selectivity to EP and PA increased
to ꢀ97%. Within the propionates, PA was increasingly favored with
increasing H
increasing H
precision at these WHSV. Additionally, during a test when no EtOH
was included (CO:H O:EtOH:EtI = 60:9:0:1), virtually all propi-
onate selectivities went to PA, as shown in the last entry of Table 4.
2
O feed rates. Conversion appears to decrease with
2
O feed, although the changes may be within available
0%
0
5
10
15
20
25
30
2
Time (hr)
Fig. 2. Catalytic performance with time on stream in EtOH carbonylation over
3
.3. Effect of weight hourly space velocity (WHSV) and time on stream
À1
1
1
wt% Rh/Cs
3
PW12
O
40 for 30 h. WHSV = 1.6 h , 1 atm nominal total pressure,
70 °C, feed ratio of CO:EtOH:EtI = 164:10:1. ꢀ EtOH conversion, j ethylene
selectivity, N DEE selectivity, and d EP + PA selectivity, À mass balance.
The results show high propionate selectivity at low conversions
typical of high space velocities. Prior work has shown decreasing
propionate selectivity with increasing conversion [15,16], and the
same conclusion would be drawn using only the data in Table 1.
An extended reaction of 30 h TOS was carried out at a WHSV of
1.6 h and standard conditions as shown in Fig. 2. EtOH conver-
À1
À1
Instead, in Table 5, WHSV was controlled between 1.1 and 24 h
with 1 wt% Rh/Cs
3
PW12
O
40 at 170 °C and 1 atm for 5 h TOS with
sion decreased from ꢀ65% to ꢀ50% over the first 6 h, but decayed
much slower over the subsequent 24 h. EP selectivity remained
nearly constant at ꢀ80% for the entire run. The mass balance
improved progressively throughout the run, eventually stabilizing
out at ꢀ85%.
a CO:EtOH:EtI feed ratio of 164:10:1. As expected, EtOH conversion
increased significantly with decreasing WHSV, but EP selectivity
was remarkably constant at all space velocities tested. It should
be noted that even for the highest conversions in Table 5, reactive
ethyl (EtOH + EtI) concentrations in the effluent are comparable to
A sample of catalyst recovered after a similar experiment at
those of H
2
O. Thus, substantial excess H
2
O, such as in Table 4, is
2
10 h on stream lost 3.7 wt% during subsequent TGA in O up to
required to see any significant EP hydrolysis to PA for this system.
These results are consistent with dehydration and carbonylation
being parallel pathways. We must note, however, that at the low-
est space velocities, ethyl species mass balances are poor, indicat-
ing the formation of undetectable products. Although individual
runs show poor mass balance at low WHSV, conversions and selec-
tivities were reproducible to within 5% and 3% error, respectively,
over multiple catalyst preparations and catalytic runs, showing
that the reaction is not intrinsically unstable. Explicit batch-to-
batch variability is shown in Appendix A Table S2.
350 °C, the typical initial calcination temperature. This mass loss
corresponds to approximately 1.6 mmol carbon per 50 mg catalyst
or approximately 3% of the EtOH feed over the 10 h run. Thus, only
a fraction of the incomplete mass balance can be accounted for by
deposition of easily removed carbon. Presumably, the remaining
unaccounted species are heavier species trapped elsewhere in
the reactor system or species undetected by the current GC
method. Given the large amounts of HI generated in the reactor
at low WHSV, rearrangement to heavier, undetected products such
as ethylene oligomers is not unexpected, and a gradual baseline

6
S. Yacob et al. / Journal of Catalysis 325 (2015) 1–8
rise is noticed in the GC traces of these experiments. All thermal
treatments intended to regenerate the catalyst were ultimately
detrimental to performance, but none altered the crystal structure
of the HPA (Appendix A Table S3 and Fig. S4), suggesting that for-
mation of Rh(0) nanoparticles and subsequent aggregation is one
possible origin of deactivation, which will be discussed further
below [40,41]. Increasing conversion with decreasing WHSV is
Table 6
Rh 3d5/2 binding energies for Rh/Cs
3
PW12
O
40 before and after various treatments.
Rh 3d5/2 binding energy (eV)
Catalyst and treatment
Rh/Cs
Rh/Cs PW12O40 used
Rh/Cs
Rh/Cs
3
PW12
O
40 fresh
308.3
307.0
306.5
307.9
3
3
PW12
O
O
40 fresh, reduced at 400 °C
3
PW12
40 regenerated
expected to generate H
the Rh catalyst [42], which could exacerbate this process.
2
from residual water gas shift activity of
Rh
Rh foil [45]
Supported Rh nanoparticles [46]
2
O
3
[43,44]
308.5
307.2
306.8
0
3.4. Spectroscopic analysis
Samples of the fresh and used catalysts were also examined
using XPS to support the changes in the average Rh oxidation state
seen from in situ XANES. Table 6 shows Rh 3d5/2 binding energies
In situ XANES was used to examine the oxidation state of the
catalyst during reaction conditions. A least-squares linear combi-
nation fit was conducted using known reference standards: metal-
(
±0.5 eV due to low Rh loadings), and XPS scans are collected in
lic Rh, Rh
Cs
and RhCl
2
O
O
3
, RhCl
40 catalyst was well fit as a combination of Rh
(79%) character, both trivalent Rh species. This fit
3
, and RhI
3
(Figs. S5 and S6a). The fresh 5 wt% Rh/
Appendix A Fig. S8. The fresh catalyst has a Rh 3d5/2 binding energy
of 308.3 eV, within uncertainty of the literature assignment of
Rh(III) oxide at 308.5 eV [43,44]. After the catalyst has been used
at standard conditions and re-exposed to air during sample trans-
fer, the binding energy decreased to 307.0 eV. Analysis of a fresh
3
PW12
2 3
O
(21%)
3
implies the freshly prepared catalyst contains Rh(III) species in a
mixture of Cl and O coordination. Good agreement could not be
reached when excluding RhCl
ear combination fit. As anticipated, the fresh catalyst contains no
metallic Rh or RhI character.
3
standard from the least-squares lin-
2
catalyst deliberately reduced in H at 400 °C for 2 h gave a similar
binding energy of 306.5 eV. Both measurements are within error of
typical binding energies of the bulk metal, 307.2 eV [45] and of
reported values for supported Rh nanoparticles, 306.8 eV [46]. An
attempt was made to regenerate the used catalyst by calcination
at 350 °C in static air for 2 h. The regenerated catalyst was only
partially re-oxidized with a binding energy of 307.9 eV. Overall,
the XPS results are consistent with partial reduction of the catalyst
during reaction witnessed via in situ XANES. Specifically, XPS sug-
gests that the catalyst contained significant Rh(0) content and that
these Rh(0) nanoparticles could not be re-oxidized under typical
calcination conditions. This further supports the hypothesis that
reduction to Rh(0) nanoparticles may be the root cause for deacti-
vation in these materials.
3
The catalyst was examined with XANES during standard reac-
tion conditions using an in situ cell, Fig. 3. Spectra were fit every
hour as a combination of the standards. In this manner, the
Rh
immediately after the in situ reaction began. Likewise, principal
component analyses revealed RhI character shortly after the in situ
2 3 3
O and RhCl species content decreased significantly, beginning
3
reaction began, indicating the formation of metal–iodide bonds
when feeding EtI (Appendix A Fig. S7). Also, evident via the princi-
pal component analysis is the decrease in the average oxidation
state during the first two hours. After two hours on stream, the cat-
alyst did not show any additional significant changes. The used cat-
alyst after 10 h on stream was fit to a mixture of Rh, RhI
RhCl
, corresponding to an average oxidation state of ꢀ1.8 (Appen-
dix A Fig. S6b). Fits of the used catalyst excluding metallic Rh or
RhI standards were both found inadequate. This decrease in aver-
3
, and
3
3.5. Isotopic labeling study
3
age oxidation state is consistent both with slow activation of the
catalyst to form Rh(I) reactive intermediates analogous to
solution-phase catalysts and the possible formation of Rh(0)
nanoparticles as a deactivation route.
EtOH carbonylation was also carried out with isotopically
labeled reagents to help understand the reaction network. 99% of
1
3
13
the feed EtI was labeled at the primary carbon, EtI = CH
3 2
CH I.
Standard reaction conditions were used consisting of 1 atm nom-
1
3
À1
inal pressure, CO:EtOH: EtI of 164:10:1, WHSV of 24 h , and
temperature of 170 °C using the reference 1 wt% Rh/Cs
catalyst. Products are analyzed by an Agilent 5975C GC–MS after
h TOS. Under these conditions, the exiting organic liquids are
3
PW12O
40
2
0
0
1
2
3
4
5
6
7
8
9
hr
.2 hr
hr
approximately 8.5% EtI, 83% EtOH, 1% ethylene, 0.5% DEE, and 7%
EP. Mass spectra fragmentation patterns are collected in Appendix
A Fig. S9.
12
The mass spectrum for EtI peak showed a mixture of C (38%)
hr
and 13C (62%) EtI; the appearance of unlabeled 12C EtI indicates
1
2
13
hr
that EtI is formed from the C EtOH. Further, C is incorporated
into the propionate fragment of EP at nearly the same level of iso-
hr
1
2
13
tope incorporation ( C 35% and C 65%), consistent with a
mechanism where the EtI adds to the Rh center, eventually leading
hr
hr
13
12
13
to formation of an acyl group. C is found in DEE ( C 66%, C 31%,
1
3
13
hr
C double incorporation 2%), indicating the reaction of EtI and
EtOH. No significant label is seen in EtOH or in the ethoxy fragment
of EP. Preferred condensation reactions with EtI are therefore con-
sistent with the much higher concentration of EtOH (83%) than
propionate groups (7%) or water (1–2% by stoichiometry).
Considering the results presented here, we propose the
3
mechanism for EtOH carbonylation over Rh/Cs PW12O40 catalyst
to be analogous to established mechanisms for methanol carbony-
lation shown in Fig. 4 [5–7]. In this analogous mechanism, a Rh–
ethyl complex is formed by addition of EtI to a Rh(I) anion. The
hr
hr
1
0 hr
23100
23150
23200
23250
23300
23350
23400
23450
Photon Energy (eV)
À1
Fig. 3. In situ XANES spectra over 5 wt% Rh/Cs
atm nominal total pressure, 170 °C, feed ratio of CO:EtOH:EtI = 164:10:1.
3
PW12O
40 for 10 h. WHSV ꢀ 24 h
,
1

S. Yacob et al. / Journal of Catalysis 325 (2015) 1–8
7
I
the fresh catalyst and immediate gains in I coordination after the
reaction begins. XPS and XANES also show that changes in coordi-
nation during reaction are coupled with an overall drop in average
oxidation state consistent with a presumed Rh(I) active state.
Dependencies on reactants and isotopic incorporation indicate that
the mechanism is Rh-centered and includes the oxidative addition
of an alkyl iodide co-catalyst to the metal. Overall, the evidence pre-
sented above suggests that the HPA anion is primarily a stable sup-
port with no detrimental redox chemistry little direct role as an acid
catalyst. Thus, other soft cations and large, weakly coordinating
counter-anions may therefore be alternate methods for supporting
an active catalyst.
+
HI
I
CO
CO
I
CO
CO
Cs+
Cs+
Rh
Rh
H
-
HI
X
X
+
C H I
2 5
-
C H COI
-C H
2
5
2
4
I
I
Regarding process conditions, these catalysts give remarkably
stable selectivity to propionates (80–85%) when all acidity is
removed. Selectivity is unchanged at up to ꢀ75% EtOH conversion
holding all other standard conditions constant apart from WHSV.
Selectivity to propionates increases further to ꢀ95% with added
water, and PA is the overwhelming product, at the expense of some
conversion. EtI feed can be reduced at least fivefold relative to
standard feed compositions of 10:1 EtOH:EtI with no detrimental
reactivity effects. At low space velocities, ethyl species mass bal-
ances begin to suffer, indicating generation of undetectable car-
bon-containing species. Similarly, these catalysts deactivate
slowly and have worse mass balances at low WHSV. The exact
cause of deactivation and loss of mass balance is not clear at this
time, but neither appears to be related to extensive coking. There
is some spectroscopic evidence from changes to the Rh average
oxidation state that deactivation may be caused by reduction of
the Rh to metallic nanoparticles. Additional experiments are
underway to use X-ray absorption spectroscopy, FTIR, and other
tools to understand catalyst speciation under process conditions.
Further optimizing catalyst precursors, calcination temperatures
and other synthesis parameters may improve activity by placing
more of the Rh in an active form. Likewise, process optimization
of EtI and water feed rates, WHSV, and other operating parameters
may further stabilize this catalytic system for the selective conver-
sion of ethanol to valuable propionate oxygenates.
I
CO
CO
C H
I
CO
CO
Cs+
Cs+
Rh
Rh
+CO
X
X
C H
2 5
O
2
5
Fig. 4. Proposed mechanism for EtOH carbonylation. In addition, acid-catalyzed,
reversible reactions interconnect acyl iodide to PA and EP via reaction with HI, H O,
or EtOH. A similar network exists for the interconversion of EtI, EtOH, DEE, and
ethylene. The X ligand may be HPA, iodide, or some other species, with the exact
nature of the coordination not known at this time.
2
resulting Rh(III) species can either undergo migratory insertion of
CO into the Rh–ethyl bond, as is typical, or b-hydride elimination
to produce ethylene, and eventually HI and the starting Rh(I) com-
plex. The latter steps would not be found in the methanol mechan-
ism but are possible here. The Rh–acyl complex then eliminates the
acyl iodide to complete the catalytic cycle. In steps not likely cat-
alyzed by the Rh center, the acyl iodide is condensed with EtOH
(
or water in certain cases) and HI reacts with EtOH to regenerate
EtI. In side reactions that are likely close to equilibrium, EtOH
and/or EtI can form DEE or ethylene, and PA can condense with
EtOH or EtI. In addition, large amounts of gas-phase HI, as found
at the lower space velocities, would lead to a number of acid-cat-
alyzed side reactions, such as the formation of oligomers and rear-
rangement side products, not necessarily all detected, and
contribute to the poor ethyl species mass balance at low WHSV.
The exact role of the HPA salt remains less clear. Its most impor-
tant features are that it is a redox-stable, porous support, free of
acidity. Most direct roles of the HPA acid sites can be ruled out,
as even small amounts of acidity result almost exclusively in etha-
nol dehydration. Given that alkali iodide salts such as CsI stabilize
and promote Rh carbonylation catalysts [47], the HPA salt may
simply be a particularly effective source of Cs. Direct coordination
of the polyatom to an active Rh(I) complex also cannot be ruled
out. In either case, the possible structures, and thus mechanisms,
are quite analogous to those of homogenous methanol
carbonylation.
Acknowledgments
This work was supported by The Dow Chemical Company. Por-
tions of this work were performed at the DuPont–Northwestern–
Dow Collaborative Access Team (DND-CAT) located at Sector 5 of
the Advanced Photon Source (APS). DND-CAT is supported by E.I.
DuPont de Nemours & Co., The Dow Chemical Company and North-
western University. Use of the APS, an Office of Science User Facility
operated for the U.S. Department of Energy (DOE) Office of Science
by Argonne National Laboratory, was supported by the U.S. DOE
under Contract No. DE-AC02-06CH11357. The authors highly
appreciate Dr. Qing Ma at DND-CAT of APS for his X-ray absorption
spectroscopy expertise. Metal analysis (ICP-MS) was performed at
the Northwestern University Quantitative Bioelemental Imaging
Center generously supported by NASA Ames Research Center
NNA06CB93G. Metal analysis (ICP-AES) and Solid State NMR were
performed at the Northwestern University Integrated Molecular
Structure Education and Research Center with funding provided
by NSF DMR-0521267. The CleanCat Core facility acknowledges
funding from the Department of Energy (DE-FG02-03ER15457)
used for the purchase of the Altamira AMI-200. This work made
use of the J.B. Cohen X-Ray Diffraction Facility supported by the
MRSEC program of the National Science Foundation (DMR-
1121262) at the Materials Research Center of Northwestern Univer-
sity. This XPS work was performed in the Keck-II facility of NUANCE
Center at Northwestern University. The NUANCE Center is support-
ed by NSEC (NSF EEC-0647560), MRSEC (NSF DMR-1121262), the
Keck Foundation, the State of Illinois, and Northwestern University.
4
. Conclusion
This study has demonstrated that the requirements for a high
yielding, vapor-phase EtOH carbonylation catalyst are negligible
acidity coming from >95% exchange of the protons in the HPA sup-
port with alkali, a W polyatom, and Rh. Other known catalysts for
ethanol carbonylation in the gas phase, such as Rh/Na13X, are
markedly slower and less selective, and those same supported Rh
catalysts for methanol carbonylation report rates only approximate-
ly threefold that of the ethanol carbonylation rates reported here
15], making ethanol carbonylation a competitive process. Sequen-
tially impregnating all catalyst components on an inert SiO sup-
port was strongly detrimental to selectivity, indicating that
intimate mixing of Rh, Cs, and the HPA anion is required. Spectro-
scopic studies using in situ XANES reveal Cl and O coordination in
[
2

8
S. Yacob et al. / Journal of Catalysis 325 (2015) 1–8
[
[
24] R.G. Schultz, P.D. Montgomery, J. Catal. 13 (1969) 105.
25] J. Evans, S.L. Scruton, To BP Chemicals Limited, The University of Southampton,
US Patent 5 185 462, 1993.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2015.02.004.
[26] J.F. DeWilde, H. Chiang, D.A. Hickman, C.R. Ho, A. Bhan, ACS Catal. 3 (2013)
798.
[
[
27] H. Chiang, A. Bhan, J. Catal. 271 (2010) 251.
28] T. Okuhara, N. Mizuno, M. Misono, in: D.D. Eley, W.O. Haag, B. Gates (Eds.),
Advances in Catalysis, vol. 41, Elsevier Academic Press Inc., San Diego, 1996, p.
References
113.
[
[
[
[
1] J.F. Roth, J.H. Craddock, A. Hershman, F.E. Paulik, Chem. Technol. 1 (1971) 600.
2] F.E. Paulik, J.F. Roth, Chem. Commun. (1968) 1578.
3] D.J. Forster, J. Chem. Soc. 98 (1976) 846.
[29] I.V. Kozhevnikov, Chem. Rev. 98 (1998) 171.
[30] B.S. Jaynes, C.L. Hill, J. Am. Chem. Soc. 117 (1995) 4704.
[31] L.D. Dingwall, A.F. Lee, J.M. Lynam, K. Wilson, L. Olivi, J.M.S. Deeley, S. Gaemers,
G.J. Sunley, ACS Catal. 2 (2012) 1368.
4] P.M. Maitlis, A. Haynes, G.J. Sunley, M.J. Howard, J. Chem. Soc. Dalton Trans.
(
1996) 2187.
[32] R.W. Wegman, A.G. Abatjoglou, A.M. Harrison, J. Chem. Soc. Chem. Commun.
24 (1987) 1981.
[33] M.V. Luzgin, M.S. Kazantsev, G.G. Volkova, W. Wang, A.G. Stephanov, J. Catal.
(2010) 72.
[34] J.F. Keggin, Nature 131 (1933) 908.
[35] B. Dawson, Acta Crystallogr. 6 (1953) 113.
[36] R.G. Finke, M.W. Droege, Inorg. Chem. 22 (1983) 1006.
[37] M.H. Alizadeh, S.P. Harmalker, Y. Jeannin, J. Martinfrere, M.T. Pope, J. Am.
Chem. Soc. 107 (1985) 2662.
[
[
[
[
[
5] T.W. Dekleva, D. Forster, J. Am. Chem. Soc. 107 (1985) 3565.
6] T.W. Dekleva, D. Forster, J. Mol. Catal. 33 (1985) 269.
7] D. Forster, T.W. Dekleva, J. Chem. Educ. 63 (1986) 204.
8] P. Cheung, A. Bhan, G.J. Sunley, E. Iglesia, Angew. Chem. Int. Ed. 45 (2006) 1617.
9] E. Iglesia, J.G. Sunley, D.J. Law, A. Bhan, To the Regents of the University of
California, BP Chemicals Ltd., US Patent 7 507 855, 2009.
[
[
[
[
[
[
[
[
[
10] G. Kiss, Chem. Rev. 101 (2001) 3435.
11] C.W. Bird, Chem. Rev. 62 (1962) 283.
12] W.F. Gresham, R.E. Brooks, To Du Pont, US Patent 2 448 368, 1948.
13] W. Reppe, H. Kröper, German Patent 863 194, 1953.
14] W. Reppe, H. Kröper, Justus Liebigs Ann. Chem. 582 (1953) 38.
15] B. Christensen, M.S. Scurrell, J. Chem. Soc. Faraday Trans. I 73 (1977) 2036.
16] B. Christensen, M.S. Scurrell, J. Chem. Soc. Faraday Trans. I 74 (1978) 2313.
17] M.S. Scurrell, T. Hauberg, Appl. Catal. 2 (1982) 225.
18] B.K. Nefedov, R.V. Dzhaparidze, O.G. Mamaev, Bull. Acad. Sci. USSR Div. Chem.
Sci. 27 (1978) 1448.
[38] R.W. Wegman, US Patent 5 218 140, To Union Carbide Chemicals & Plastics
Technology Corporation, 1988.
[39] G.G. Volkova, L.M. Plyasova, A.N. Salanov, G.N. Kustova, T.M. Yurieva, V.A.
Likholobov, Catal. Lett. 80 (2002) 175.
[40] K. Paredis, L.K. Ono, F. Behafarid, Z.F. Zhang, J.C. Yang, A.I. Frenkel, B.R. Cuenya,
J. Am. Chem. Soc. 133 (2011) 13455.
[41] D.H. Mei, R. Rousseau, S.M. Kathmann, V.A. Glezakou, M.H. Engelhard, W.L.
Jiang, C.M. Wang, M.A. Gerber, J.F. White, D.J. Stevens, J. Catal. 271 (2010) 325.
[42] E. Trabuco, P.C. Ford, J. Mol. Catal. A – Chem. 148 (1999) 1.
[43] Y. Okamoto, N. Ishida, T. Imanaka, S. Teranishi, J. Catal. 58 (1979) 82.
[44] V.I. Nefedov, M.N. Firsov, I.S. Shaplygin, J. Electron Spectrosc. 26 (1982) 65.
[45] M. Cardona, L. Ley, Photoemission in Solids I. General Principles, vol. 26,
Springer, Berlin Heidelberg, 1978.
[
[
[
19] S.L.T. Andersson, M.S. Scurrell, J. Catal. 71 (1981) 233.
20] S.L.T. Andersson, M.S. Scurrell, J. Catal. 59 (1979) 340.
21] B.K. Nefedov, R.V. Dzhaparidze, Y.T. Eidus, Bull. Acad. Sci. USSR Div. Chem. Sci.
2
6 (1977) 1310.
[
[
22] B.K. Nefedov, R.V. Dzhaparidze, O.G. Mamaev, N.S. Sergeeva, Bull. Acad. Sci.
USSR Div. Chem. Sci. 28 (1979) 348.
23] R.G. Schultz, US Patent 3 689 553, 1972.
[46] A. Berko, I. Ulrych, K.C. Prince, J. Phys. Chem. B 102 (1998) 3379.
[47] B.L. Smith, G.P. Torrence, M.A. Murphy, A. Aguilo, J. Mol. Catal. 39 (1987) 115.