Carbon dioxide fixation into chemicals (methyl formate) at high yields by surface coupling over a Pd/Cu/ZnO nanocatalyst

Article abstract of DOI:10.1021/ja0706302

Chemical fixation of CO₂ has recently attracted attention as a possible way to manufacture useful chemicals. Here, we demonstrate a one-step catalytic fixation of CO₂ to methyl formate (MF) in liquid phase under mild industrial applicable conditions by using a new concept of coupling surface formate species formed from CO₂ on solid catalyst with excess methanol in hydrogen to form the MF, and the quantity of the product is found to greatly exceed the surface coverage of the catalyst by >45-fold. As a result, we report, for the first time, that the optimum Pd/Cu/ZnO/alumina nanocatalyst prepared from simple co-precipitation is capable of activating gaseous CO₂ to condensable MF at high yields (>20%) with excellent selectivity (>96%). The present study also offers directions to rationally design and optimize the required catalyst, reactor, and experimental conditions for this new reaction. Copyright

Full text of DOI:10.1021/ja0706302

Published on Web 04/28/2007
Carbon Dioxide Fixation into Chemicals (Methyl Formate) at High Yields by
Surface Coupling over a Pd/Cu/ZnO Nanocatalyst
K. M. Kerry Yu, Connie M. Y. Yeung,† and Shik Chi Tsang*
Surface and Catalysis Research Centre, Department of Chemistry, UniVersity of Reading, Reading RG6 6AD, U.K.
Received January 28, 2007; E-mail: s.c.e.tsang@reading.ac.uk
In recent years, the carbon dioxide (CO2) issue has become the
focus of attention because of the position of CO2 as the primary
greenhouse gas and the implication of its emissions on the problem
of climate change. Burning fossil fuels releases the CO2 stored
millions of years ago. Deforestation releases the carbon stored in
trees, resulting in more CO2 in the atmosphere. The resulting climate
change is an immediate threat to our security and prosperity, as
the global food supplies will be compromised. Therefore, in pursuit
of a stable climate, there is an urgent need to construct a low carbon
global economy. For example, the U.K. government targets a 60%
cut in emissions by 20501 as a necessary step for the development
of a sustainable economy. Thus, an instantaneous response for
research studies in the chemistry of CO2 including its activation,
utilization, and fixation is urgently needed. Apart from the physical
storage of CO2 into depleted oil/gas/coal reservoirs, chemical
fixation of CO2 has attracted additional attention as a possible way
to manufacture useful chemicals in some specific locations. As a
result, the activation of CO2 at high yields appears to be the essential
step to achieve the above objective. However, there has been limited
work in literature reporting a catalytic approach to activate and to
fix CO2 under industrial applicable conditions, despite the fact that
this process has been recognized as a significant innovation to
chemical industry.2 Most previous attempts were unable to attract
industrial attention as they suffer from either the use of expensive
but non-robust homogeneous catalysts or the use of peculiar reaction
conditions (requires excess ligand(s) and solvents at extreme
conditions).3 Here, we demonstrate a one-step catalytic fixation of
CO2 to methyl formate (MF) in liquid phase under mild industrial
applicable conditions by using a new concept of coupling surface
formate species formed from CO2 on solid catalyst with excess
methanol in hydrogen to form the MF, and the quantity of the
product is found to greatly exceed the surface coverage of the
catalyst by >45-fold. As a result, we report, for the first time, that
the optimum Pd/Cu/ZnO/alumina nanocatalyst prepared from simple
co-precipitation is capable of activating gaseous CO2 to condensable
MF at high yields (>20%) with excellent selectivity (>96%).
As an initial effort, co-precipitated Cu/ZnO/alumina (Cat) was
used as the catalyst for the proof of concept of surface coupling
with blended molecule, which is an industrial catalyst for water
gas shift reaction, WGS (and also reversed reaction, RWGS), where
formate species is believed to be the intermediate species.4 The
use of hydrogen to activate CO2 to useful chemicals at low
temperature could be commercially viable with regard to future
hydrogen availability from hydrocarbon reformations with carbon
capture/storage and from catalytic water splitting via solar energy.
Thus, screening of different solvent molecules in order to identify
a candidate that can couple with surface formate to form chemicals
under hydrogen and carbon dioxide in presence of Cat was carried
out. As a result, preliminary studies with 10 bar CO2, 50 bar H2,
N2 (as diluent), 10.0 mL of solvent (hexane, triethylamine, methanol,
dichloromethane, DMF, pyridine, no solvent, etc.), and 0.2 g of
Cat in a 100 mL high-pressure Parr autoclave reactor at the total
pressure of 160 bar at 150 °C for 25 h were carried out. It was
1
interesting to find from GC/MS and H NMR that methanol was
the only solvent that gave a clean, exclusive condensable product,
which was identified to be methyl formate (MF). The noncatalytic
methanol conversion to MF and the direct decomposition of
methanol to MF in the absence of CO2 (replaced by N2) were
discounted since only trace MF was observed in both cases.5 The
yield of MF under CO2/H2 in methanol was estimated to be over
45 times that of the complete monolayer of Cat, implying that the
methanol molecule is capable of coupling the surface formate,
dragging the reversed WGS forward (see Supporting Information).
It is thus believed that esterification (by thermal means or acid sites
from the catalyst) may have taken place quickly between the surface
formate and the adsorbed methanol, giving MF which can desorb
to the solvent, see Scheme 1. Therefore, a new but clean catalytic
reaction route of fixing CO2 into condensable MF is revealed.
The conditions for optimum MF production were then assessed
at 150 °C (Figure 1) which showed a maximum CO2 conversion at
23.2% through variation of the H2 to CO2 molar ratio (R) from 0.5
to 12.5 with the MF as a primary product (selectivity >95%). Trace
of carbon monoxide (CO) was detected as a byproduct possibly
from the RWGS. However, the highest turnover frequency, TOF
(evaluated at the first 10 min of the reaction), was achieved at R )
5 with a volcano relationship observed, which was attributed to
the competition of adsorption sites between H2 and CO2 on the
catalyst. Under excess H2 (90 bar, 255.9 mmol), an optimum
conversion of CO2 to MF in function of CO2 pressure was also
assessed (Table 1). It was noted that the % CO2 conversion to MF
could be raised from 10.9 to 19.1% without any noticeable
selectivity attenuation simply by reducing the CO2 pressure under
the conditions listed in Table 1.
With reference to the tentative mechanism in Scheme 1, the
formation of MF requires chemisorption of at least H2, CO2, and
methanol on the surface. Doping of the catalyst with more than
one metal promoter for co-adsorption of different surface species
usually improves the overall reaction rate as this concept has been
clearly demonstrated in catalytic hydrogenations.6 Thus, an attempt
to increase the TOF by doping the Cat with 1 w/w % of different
metal promoters was carried out, and the results are summarized
in Table 2. It is noted that doping the material with precious metals
such as Ru, Ni, Au, and Pd metal showed a dramatic effect on
TOF: Pd proved to be the best doper that enhanced the TOF by
23.6%, giving activity of 38.15 h-1. This order may relate to the
ability of hydrogen activation and transfer of the promoter on the
catalyst surface since Pd is also well-known to show the best
hydrogen spill-over activity among all precious metals.7 Au doper
appeared to decrease the TOF, which could be related to its high
affinity for Cu, blocking surface sites.
† Present address: Johnson Matthey Catalyst, Belasis Avenue, Billingham,
Cleveland TS23 1LB, U.K.
9
6360
J. AM. CHEM. SOC. 2007, 129, 6360-6361
10.1021/ja0706302 CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Figure 1. A plot of the total quantity of MF produced (also % CO2 conv.
in white boxes) and initial TOF against H2 to CO2 molar ratio, R: 0.20 g
of Cat, 10 mL of CH3OH, 1 bar CH4 as internal standard, 10 bar CO2
(28.4 mmol) with variation of H2 using N2 as diluent at 160 bar and 150
°C for 25 h.
Figure 3. A plot of MF produced against time: (4) 35 bar CO2, 85 bar
H2, 0.830 mL (13.46 mmol) of MF with 113.0 mmol total carbon; (O) 40
bar CO2, 90 bar H2 with 113.7 mmol total carbon (0.20 g of Cat; 1 bar
CH4 as internal standard; N2 as diluent in 100 mL of reactor at 160 bar at
150 °C for 25 h).
Table 2. Effect of TOF by Doping the Cat with 1% Precious
Metalsa
1
catalyst
TOF (h-
)
Cu/ZnO/alumina
30.85
33.62
37.27
13.42
38.15
1% Ru-Cu/ZnO/alumina
1% Ni-Cu/ZnO/alumina
1% Au-Cu/ZnO/alumina
1% Pd-Cu/ZnO/alumina
Figure 2. HRTEM image of Pd-doped Cat showing the intimate interface
between Pd and Cu (PdO <101> and CuO <111> observed8 with crystal
sizes of 2.66 and 2.55 nm, respectively, after the sample was exposed to
air).
a Conditions: 10 bar CO2, 50 bar H2 at 150 °C.
To conclude, it is demonstrated that, with introduction of
methanol to the Pd/Cu/ZnO nanocatalyst under H2/CO2, carbon
dioxide can be activated and fixed as methyl formate with
impressive TOF, yield, and selectivity. Thus, a new route for CO2
activation by heterogeneous means at ambient conditions is
described. This preliminary study also offers directions to rationally
design and optimize the required catalyst, reactor, and experimental
conditions for this reaction.
Scheme 1. Proposed Capture of Surface Formate over Cat by
Methanol to Methyl Formate under H and CO2
Table 1. Attempts to Optimize % Conversion of CO2 to MF by
Reducing the CO2 Pressure in the Reaction Mixturea
Acknowledgment. We are grateful to the EPSRC, U.K. (EP/
D023238/1), for support under the consortium of Formic Acid
Economy. Useful discussions from Davidson (Bath), Rayner
(Leeds), Davis (Oxford) and Hall (Strathclyde), and S. Ellis (JM)
of the low carbon technology team at Johnson Matthey Technology
Centre are acknowledged.
CO2/bar
(mmol)
MF yield
(mmol)
CO
TOF
selectivityb
(%)
conv.c
(%)
1
(mmol)
(h-
)
1
2
3
4
40 (113.7)
20 (56.9)
10 (28.4)
5 (14.2)
12.4
9.2
5.0
2.7
0.27
0.20
0.16
n.d.
26.1
20.7
14.6
n.d.
97.9
97.8
96.8
n.d.
10.9
16.2
17.6
19.1
a Same conditions as quoted in Figure 1; captions were applied except
Supporting Information Available: Synthesis of Cat and doping,
setup, and experimental procedure are included. This material is
available free of charge via the Internet at http://pubs.acs.org.
with variation of CO2 pressure at 90 bar H2 (255.9 mmol). b Selectivity to
MF. c % Conversion of CO2 to MF.
Detailed HRTEM imaging of the Pd-doped Cat (Figure 2) was
conducted, which showed the intimate interface of Cu and Pd in
our co-precipitated nanocatalyst.
References
(1) Department of Trade and Industry. Energy White Paper: Our Energy
FuturesCreating a Low Carbon Economy; The Stationery Office, London,
2003.
(2) Behr, A.; Ebbinghaus, P.; Naendrup, F. Chem.-Ing.-Tech. 2003,
75, 877.
(3) (a) Gassner, F.; Leitner, W. Chem. Commun. 1993, 1465. (b) Graf, E.;
Leitner, W. Chem. Commun. 1992, 623.
(4) (a) Lazo, N. D.; Murray, D. K.; Kieke, M. L.; Haw, J. F. J. Am. Chem.
Soc. 1992, 114, 8552. (b) Collins, S. E.; Baltana´s, M. A.; Bonivardi, A.
L. J. Catal. 2004, 226, 410. (c) Shido, T.; Iwasawa, Y. J. Catal. 1992,
136, 493.
(5) Choi, Y.; Stenger, H. G. Appl. Catal. B 2002, 38, 259.
(6) (a) Raja, R.; Sankar, G.; Hermans, S.; Shephard, D. S.; Bromley, S.;
Thomas, J. M.; Johnson, B. F. G. Chem. Commun. 1999, 1571. (b)
Alexeev, O. S.; Gates, B. C. Ind. Eng. Chem. Res. 2003, 42, 1571. (c)
Goodman, D. W.; Houston, J. E. Science 1987, 236, 403. (d) Ichikawa,
M. AdV. Catal. 1992, 38, 283.
(7) Tsang, S. C.; Bulpitt, C. D. A.; Mitchell, P. C. H.; Ramirez-Cuesta, A. J.
J. Phys. Chem. B 2001, 105, 5737.
(8) (a) JCPDS-International Centre for Diffraction Data 1996, PDF Nos. 45-
0937 and 01-1117 for CuO; PDF Nos. 43-1024 and 41-1107 for PdO. (b)
Liu, Y.; Chu, Y.; Li, M. Y.; Li, L. L.; Dong, L. H. J. Mater. Chem. 2006,
16, 192. (c) Thevenin, P. O.; Alcalde, A.; Pettersson, L. J. Jarås S. G.;
Fierro, J. L. G. J. Catal. 2003, 215, 78.
Of all catalysts tested, it is interesting to note that the quantity
of MF produced was found to level off after a prolonged period of
time (see Figure 3). The kinetic poisoning effect by byproducts
such as CO and water was discounted since adding excess CO and
water prior to reaction showed no significant negative effects on
the TOF and the quantity of MF produced. Thus, it is likely the
MF production was somehow restricted under a new equilibrium.
As shown in Figure 3, the curve with the pre-addition of MF (13.46
mmol) reached the same steady (equilibrium) position as the one
without blending the MF, achieving 12.29 mmol MF in both cases.
Thus, the results indicate that the CO2 activation and fixation to
MF by these catalytic means with the coupling of methanol can
achieve equilibrium for a prolonged period of time. Thus, the future
challenge to increase MF productivity from CO2 activation is to
avoid the establishment of equilibrium in liquid by either changing
the batch process to a continuous process or/and removing products
from recycled mixture.
JA0706302
9
J. AM. CHEM. SOC. VOL. 129, NO. 20, 2007 6361