From organometallic zinc and copper complexes to highly active colloidal catalysts for the conversion of CO2 to methanol

Article abstract of DOI:10.1021/cs502038y

A series of zinc oxide and copper(0) colloidal nanocatalysts, produced by a one-pot synthesis, are shown to catalyze the hydrogenation of carbon dioxide to methanol. The catalysts are produced by the reaction between diethyl zinc and bis(carboxylato/phosphinato)copper(II) precursors. The reaction leads to the formation of a precatalyst solution, characterized using various spectroscopic (NMR, UV-vis spectroscopy) and X-ray diffraction/absorption (powder XRD, EXAFS, XANES) techniques. The combined characterization methods indicate that the precatalyst solution contains copper(0) nanoparticles and a mixture of diethyl zinc and an ethyl zinc stearate cluster compound [Et₄Zn₅(stearate)₆]. The catalysts are applied, at 523 K with a 50 bar total pressure of a 3:1 mixture of H₂/CO₂, in the solution phase, quasi-homogeneous, hydrogenation of carbon dioxide, and they show high activities (>55 mmol/g_ZnOCu/h of methanol). The postreaction catalyst solution is characterized using a range of spectroscopies, X-ray diffraction techniques, and transmission electron microscopy (TEM). These analyses show the formation of a mixture of zinc oxide nanoparticles, of size 2-7 nm and small copper nanoparticles. The catalyst composition can be easily adjusted, and the influence of the relative loadings of ZnO/Cu, the precursor complexes and the total catalyst concentration on the catalytic activity are all investigated. The optimum system, comprising a 55:45 loading of ZnO/Cu, shows equivalent activity to a commercial, activated methanol synthesis catalyst. These findings indicate that using diethyl zinc to reduce copper precursors in situ leads to catalysts with excellent activities for the production of methanol from carbon dioxide.

Full text of DOI:10.1021/cs502038y

Research Article
pubs.acs.org/acscatalysis
From Organometallic Zinc and Copper Complexes to Highly Active
Colloidal Catalysts for the Conversion of CO to Methanol
2
†
†
†
†
†
́
Neil J. Brown, Andres García-Trenco, Jonathan Weiner, Edward R. White, Matthew Allinson,
†
‡,§
†
‡,§
⊥
,†
Yuxin Chen, Peter P. Wells, Emma K. Gibson, Klaus Hellgardt, Milo S. P. Shaffer,*
,
and Charlotte K. Williams*
†
Department of Chemistry, Imperial College London, South Kensington, London SW7 2AZ, United Kingdom
The UK Catalysis Hub, Research Complex at Harwell, Harwell, Oxon OX11 0FA, United Kingdom
‡
§
Kathleen Lonsdale Building, Department of Chemistry, University College London, Gordon Street, London WC1H 0AJ, United
Kingdom
⊥
Department of Chemical Engineering, Imperial College London, South Kensington, London SW7 2AZ, United Kingdom
S Supporting Information
*
ABSTRACT: A series of zinc oxide and copper(0) colloidal
nanocatalysts, produced by a one-pot synthesis, are shown to
catalyze the hydrogenation of carbon dioxide to methanol. The
catalysts are produced by the reaction between diethyl zinc and
bis(carboxylato/phosphinato)copper(II) precursors. The reaction
leads to the formation of a precatalyst solution, characterized using
various spectroscopic (NMR, UV−vis spectroscopy) and X-ray
diffraction/absorption (powder XRD, EXAFS, XANES) techni-
ques. The combined characterization methods indicate that the
precatalyst solution contains copper(0) nanoparticles and a mixture
of diethyl zinc and an ethyl zinc stearate cluster compound
[
Et Zn (stearate) ]. The catalysts are applied, at 523 K with a 50 bar total pressure of a 3:1 mixture of H /CO , in the solution
4
5
6
2
2
phase, quasi-homogeneous, hydrogenation of carbon dioxide, and they show high activities (>55 mmol/gZnOCu/h of methanol).
The postreaction catalyst solution is characterized using a range of spectroscopies, X-ray diffraction techniques, and transmission
electron microscopy (TEM). These analyses show the formation of a mixture of zinc oxide nanoparticles, of size 2−7 nm and
small copper nanoparticles. The catalyst composition can be easily adjusted, and the influence of the relative loadings of ZnO/
Cu, the precursor complexes and the total catalyst concentration on the catalytic activity are all investigated. The optimum
system, comprising a 55:45 loading of ZnO/Cu, shows equivalent activity to a commercial, activated methanol synthesis catalyst.
These findings indicate that using diethyl zinc to reduce copper precursors in situ leads to catalysts with excellent activities for the
production of methanol from carbon dioxide.
KEYWORDS: hydrogenation of CO , CO reduction, methanol synthesis, colloidal catalysts, Cu-ZnO catalysts, nanoparticles,
2
2
nanocatalysts, catalysts from organometallic
INTRODUCTION
is also a good precedent for these same catalysts to be used for
3b,6
■
the hydrogenation of carbon dioxide.
They are generally
Using waste CO is a high priority: among the most promising
2
prepared by metal salt coprecipitation, following by calcination,
options are the synthesis of polymers and the reduction to
alcohols which may serve as fuels. By applying renewable or
off-peak power to make hydrogen, for example, via electrolysis
of water, it is envisaged that carbon dioxide hydrogenation to
methanol could reduce our dependence on fossil fuels.
Methanol is an attractive energy vector and a promising
renewable transport fuel, particularly because its use is
7
1
aging, nanostructuring, and reduction cycles. Controlling the
conditions for these steps is essential to maximize catalyst
activity. Detailed investigations indicate that maximizing the
ZnO/Cu interface is crucial, through controlling the Zn/Cu
loading, introducing a suitable support, and ensuring intimate
2
5e,7
mixing of the phases, on the nanometre scale.
However,
3
current synthesis methods require the preformation of mineral
phases, most commonly zincian malachite, which fundamen-
tally limit the loading of Zn/Cu (in the case of zincian
compatible with existing fuels infrastructures. In China,
methanol is already being applied, in blends with gasoline, as
4
a fuel.
Methanol is currently produced on a huge scale (>40 million
tonnes, globally in 2009), mostly from synthesis gas (CO/H2),
generated by coal gasification or methane reforming, converted
Received: December 18, 2014
Revised: February 17, 2015
5
using ternary heterogeneous Cu/ZnO/Al O catalysts. There
2
3
©
XXXX American Chemical Society
2895
DOI: 10.1021/cs502038y
ACS Catal. 2015, 5, 2895−2902

ACS Catalysis
Research Article
Scheme 1. Representation of the Formation of the Ligand-Stabilized Cu and ZnO Nanoparticles from Diethyl Zinc and
Bis(stearate)copper(II) Precursor Using the One-Pot Synthesis Method
malachite, to 3:7) and predetermine the products of aging and,
Subsequently, Fischer and co-workers reduced Cu/Zn stearate
mixtures with hydrogen gas to produce copper nanoparticles,
stabilized by zinc stearate.
5b
hence, the extent and quality of the interface. Furthermore,
such coprecipitation routes require forcing conditions to
decompose the mineral, thereby limiting the nanomorphology
of the catalyst. Recent elegant structural studies of these ternary
11n
Our group has focused on CO rather than syn-gas as the
2
carbon source. An alternative zinc oxide synthesis involving the
hydrolysis of diethyl zinc, in the presence of substoichiometric
quantities of zinc carboxylates/phosphinates produces 3−5 nm
ZnO nanoparticles which are soluble in various organic
5c−e
catalysts have revealed that some surface sites are dynamic.
For these reasons, there is an impetus to develop new catalyst
8
syntheses. For example, Tsang and co-workers have mixed (in
14
the solid state) isolated nanoparticles of ZnO and Cu to
produce high-activity CO2 hydrogenation heterogeneous
catalysts and have correlated activity to exposure of particular
solvents/polymers. A high-activity carbon dioxide hydro-
genation catalyst system was prepared by combining solutions
of ZnO nanoparticles, surface coordinated by di(octyl)-
phosphinate groups, with copper nanoparticles, surface-
9
crystal facets.
15
Methanol can also be efficiently produced by slurry phase
coordinated with stearate groups. However, the optimum
composition and nature of the precatalyst solution remains
under-investigated. Furthermore, uncovering the relative
importance of particle sizes, oxidation states of the metals,
and interfaces between particles are central to improving
catalytic activity. One drawback of this system is the
requirement to pre-prepare separate solutions of the nano-
particles. A “one-pot” method to prepare the catalyst would be
desirable, as would be the replacement of hydrazine as the
reductant in the preparation of copper nanoparticles. This
paper presents a “one-pot” approach to catalyst synthesis,
applying diethyl zinc as both the reductant to produce copper
nanoparticles and the precursor to zinc oxide (by hydrolysis of
Zn−C bonds). This method provides highly active carbon
dioxide reduction catalysts, and through characterization of the
pre- and postcatalyst solutions, the effects of organometallic
precursor compound, relative loading, and metal oxidation state
can be explored.
methods, whereby the active catalyst is dissolved or suspended
10
in a high-boiling solvent (such as squalane). Of particular
interest is the use of inorganic/organometallic precursor
compounds to produce well-defined, quasi-homogeneous
̈
nanoparticles for catalysis. The groups of Fischer and Schuth
have pioneered the application of colloidal solutions of
nanoparticles (Cu and/or ZnO) as high-activity catalysts for
11
such solution phase syn-gas to methanol processes. Simon et
al. reported, as early as 1988 that the reaction between
Cu(acac) and Et Zn in the presence of 1,3-butadiene in an
2
2
organic solvent formed an efficient syn-gas catalyst system with
1
2
a high selectivity for ethanol and methanol formation. Corker
and Evans used EXAFS to investigate the precatalyst solution.
They proposed it comprised a mixture of colloidal copper and
an oxozinc cluster compound (of undefined structure and
stoichiometry); however, detailed characterization of the cluster
and changes to the system during and after catalysis were not
1
3
carried out.
RESULTS AND DISCUSSION
In the past 10 years, Fischer and co-workers have pioneered
various routes to catalytic nanoparticles, including hydro-
thermal decomposition, hydrogenolysis, photochemical decom-
position, chemical vapor deposition (CVD), and atomic layer
■
Precatalyst Synthesis and Characterization. An in situ,
one-pot catalyst preparation method that involved mixing
together equimolar quantities of copper(II) bis(stearate) and
diethyl zinc in toluene at 298 K (Scheme 1) was discovered.
This solution showed several color changes consistent with the
reduction of the copper(II) precursor and formation of a red/
brown solution and characteristic of Cu(0) nanoparticles. It is
proposed that ligand exchange occurs between the copper and
zinc complexes, leading to transient formation of copper−alkyl
species which are known to undergo rapid decomposition at
11a−q
deposition (ALD), using organo-Zn and Cu precursors.
In
2
005, Schuth and co-workers applied trioctyl aluminum
̈
reagents as the reductants and ligands for copper nanoparticles;
these systems showed high activities for liquid phase methanol
catalysis. Fischer and Muhler have also applied colloidal
nanoparticles of ZnO/Cu prepared by hydrothermal decom-
1
1r
11b,g,k,n,p
position of well-defined inorganic compounds,
some of
1
6
which show excellent activities for the liquid phase synthesis of
room temperature to produce copper(0) species. Ligand
exchange would also be expected to lead to the formation of
some zinc carboxylate species; such species have recently been
shown to react with any excess dialkyl zinc to form heteroleptic
11i
methanol from syn-gas. In 2006, they reported a high-activity
system prepared by the addition of diethyl zinc to a hot (200
°
C) solution of a copper(II) alkoxide precursor. This system
14b,d
formed small, unsupported nanoparticles of both Cu and ZnO
and showed up to 85% of the activity, in a squalane suspension,
of the ternary reference. Other alkyl zinc alkoxides were also
shown to undergo thermolysis in the presence of amine
alkyl zinc carboxylate complexes.
The characterization data
(vide infra) support this proposed ligand exchange and indicate
the reaction proceeds to the formation of a solution containing
copper(0) nanoparticles and well-defined zinc organometallic
complexes.
11i
11k
surfactants to yield quasi-homogeneous ZnO nanoparticles.
2
896
DOI: 10.1021/cs502038y
ACS Catal. 2015, 5, 2895−2902

ACS Catalysis
Research Article
The UV−vis spectrum of the red solution confirmed the
The EXAFS data and calculated fitting parameters are
evidence of the formation of copper nanoparticles and are
illustrated in Figure S3 and Table S1. These data also indicate
that the precatalyst solution contains small copper nano-
particles, again with no evidence for the formation of either of
the oxides. The first-shell Cu−Cu coordination number can be
used to estimate the particle size, according to the method
formation of copper nanoparticles, with a copper surface
17
plasmon absorbance observed at 580 nm (Figure S1). The
spectrum also clearly indicated the complete consumption of
the Cu(II) precursor complex, as shown by the disappearance
of its characteristic absorbance at 700 nm.
The speciation of the zinc was revealed by 1H NMR
spectroscopy (Figure S2), which showed the complete
consumption of the paramagnetic species (e.g., Cu(II)) and
the formation of well resolved signals consistent with the
presence of diethyl zinc and an ethyl zinc stearate cluster
compound, Zn Et (stearate) , the structure of which has been
19
developed by Beale et al. The estimated mean particle size,
from the EXAFS data, is 1.2 nm, consistent with the plasmon
2
0
absorption observed in the UV−vis spectrum.
The zinc XANES spectrum (Figure 2) of the precatalyst
solution shows features in the absorption edge that are
5
4
6
1
4b,d
reported previously.
The pentanuclear zinc cluster
compound is known to form by reaction between diethyl
zinc and bis(stearate)zinc and to retain its structure in
1
4b,d
noncoordinating solvent solutions.
Analysis of the peak
1
integrals in the H NMR spectrum revealed a 3:2 molar ratio of
Et Zn/[Et Zn (stearate) ] (Figure S2). We have also pre-
2
4
5
6
viously established that the hydrolysis of mixtures of diethyl
zinc and the zinc stearate cluster, at analogous molar loadings,
leads to the efficient formation of 3−5 nm zinc oxide
14c
nanoparticles, capped with stearate ligands.
The precatalyst solution was also analyzed by XAFS
measurements, conducted in toluene solutions, under an inert
atmosphere. The copper speciation was studied (Figure 1) by
Figure 2. Normalized XANES spectra of diethyl zinc (black), the
precatalyst solution (red), Zn(stearate) (green), the post catalysis
2
solution (light blue), ZnO nanoparticles with stearate capping groups
(dark blue), and Zn foil (yellow).
consistent with the presence of both bis(stearate) zinc and
diethyl zinc. A linear combination fit of the XANES data was
calculated using Zn(stearate) and diethyl zinc as reference
2
compounds (Figure S4). A reasonable fit was obtained, having
an R factor of 0.0025, based on a composition of ∼84% diethyl
zinc and ∼16% Zn(stearate) . This estimation shows close
2
agreement with the relative ratio of ethyl/stearate groups
determined by analysis of the integrals for Et Zn and
Figure 1. Normalized XANES spectra of Cu(stearate) (black), CuO
2
2
1
(
red), Cu O (green), the precatalyst solution (light blue), the post
Et Zn (stearate) in the H NMR spectrum (∼86% of signals
4
5
6
2
catalysis solution (blue), and Cu foil (yellow).
are due to zinc bound ethyl groups; ∼14% of signals, to zinc
bound stearate moieties). It is apparent that the anaerobic
techniques were sufficiently robust because the organozinc
carboxylate cluster compound (Et Zn (stearate) ) was success-
comparing the Cu K-edge X-ray absorption near edge structure
(XANES) spectra of the catalyst colloids with appropriate
4
5
6
reference compounds, including Cu(stearate) , CuO, Cu O,
fully detected, a compound that is known to hydrolyze readily
2
2
14b,d
and Cu metal. The precatalyst solution spectrum shows features
similar to the Cu foil reference, implying the presence of
metallic copper. Moreover, the XANES spectra lack any
features typically associated with oxidic forms of Cu, primarily
the peak at 8983.5 eV, indicative of linear Cu O (1s → 4p ),
in the presence moisture.
Furthermore, the zinc carbon
bond lengths determined by the fits to the EXAFS data are in
good agreement with those obtained for a sample of diethyl
zinc (Table S2, Figure S5), providing good support for the
presence of zinc−ethyl moieties in the precatalyst solution. The
large Debye−Waller factor, compared with diethyl zinc, is also
indicative of other species having Zn−C or Zn−O bonds, as
would be expected from the ethyl zinc stearate cluster
compound.
2
x,y
and the peak at 8986.5 eV, which is evidence for CuO (1s →
1
8a
4
pz). The main edge peaks are less well resolved than those
of the copper foil, likely due to the nanoparticle size.
Theoretical XANES simulations and experimental measure-
ments of copper clusters have shown related effects, attributed
to the absence of higher coordination shells and to a reduced
X-ray diffraction measurements were conducted on the
precatalyst solution (after removal of all volatiles including
diethyl zinc) (Figure S6). The samples are only weakly
1
8
particle size.
2
897
DOI: 10.1021/cs502038y
ACS Catal. 2015, 5, 2895−2902

ACS Catalysis
Research Article
b
Table 1. Effect of Varying the Weight Ratio ZnO/Cu on the Catalytic Activity for Carbon Dioxide Hydrogenation to Methanol
molar ratio precursors:
ZnEt /Cu(stearate)
expected weight ratio
ZnO/Cu
methanol peak activity,
methanol activity after 12 h,
mmol gZnOCu−
1
h
−1
mmol gZnOCu
−1 −1
h
entry
2
2
1
2
3
4
5
1:3
1:1
2:1
7:1
30:70
55:45
70:30
90:10
39
60
43
18
60
39
58
37
13
46
a
ternary benchmark (expected weight ratio of ZnO/
Cu is 1:2)
a
Alfa Aesar ternary methanol synthesis catalyst (product code: 45776), comprising (by weight) 24.7% ZnO, 63.5% CuO, 11.4% Al O , and MgO;
2
3
see ESI for catalyst preactivation details and activity calculations. Error for each measurement is ±4.5% as determined by runs in triplicate. Figures S7
b
and S8 illustrate the stability of the catalysts showing activity vs time. Reaction conditions: 523 K, 50 bar (3:1, H /CO ), in a solution of squalane/
2
2
−1
toluene (90:10) at a fixed total volume of 100 mL, a total gas flow of 166 mL min , over 16 h. The total catalyst concentration (ZnO + Cu) was
−3
constant at 0.20 g dm .
diffracting, but do confirm the presence of crystalline Cu metal.
introduction of the reaction gases (CO /H ) resulted in a
2 2
Some oxidation peaks (for CuO and Cu O) are observed, likely
catalyst system showing equivalent activity. Regardless of the
source of the moisture, the hydrolysis reaction is expected to
generate the ZnO nanoparticles, which in combination with the
copper nanoparticles comprise the active catalyst (vide infra).
The method was applied to prepare a series of precatalyst
solutions, using different molar ratios of diethyl zinc/copper-
2
as a result of air exposure during sample preparation and
measurement. The average copper particle size, determined for
the Cu (111) peak using the Scherrer equation, is 4.8 nm,
somewhat larger than the value obtained by EXAFS. However,
EXAFS is more sensitive to detection of smaller clusters and
particles, and XRD is volume-weighted to larger particles;
therefore, a difference in the obtained particle size between the
two measurements is to be expected in a polydisperse system.
The XRD measurements also showed additional peaks at low
angle, which correspond to signals expected for zinc-bound
stearate groups, consistent with the presence of the ethyl zinc
stearate cluster.
(
stearate) precursors, while keeping the overall loading of
2
metals constant. The catalytic activities were benchmarked
against the total mass of Cu and ZnO present, deduced from
the molar quantities of the precursors and assuming complete
conversions. The activities obtained using the precatalyst
solutions were high and also very reproducible. For example,
when a 1:1 molar mixture of diethyl zinc and copper stearate
The combined spectroscopic data, therefore, indicate that the
red precatalyst solution contains a mixture of 1−5 nm copper
nanoparticles, presumably surface-capped with stearate ligands,
and a mixture of diethyl zinc and ethyl zinc stearate clusters.
Britten and co-workers have previously studied the reaction
(
55:45 weight ratio of ZnO/Cu) was applied, a methanol
−1 −1
−1 −1
activity of 60 mmol gZnOCu
h
(133 mmol gCu h ) was
obtained (Table 1, Figure S7). This activity is similar to that
obtained using a heterogeneous ternary syn-gas catalyst
21
(
preactivated according to the standard protocols) as a
16
between diethyl zinc and Cu(II) precursors. They established
that diethyl zinc rapidly and irreversibly reduced the Cu(II)
species to Cu, which formed as a mirror. This reaction was
attributed to rapid ligand exchange and the instability of the
resulting copper alkyl complexes, which underwent decom-
position even at very low temperatures, purportedly by one-
electron/radical processes. In the present case, the mechanisms
are proposed to be similar, with the stearate ligands capping the
nanoparticles and preventing copper mirror formation.
benchmark material; however, after 10 h, the best catalyst
system showed activity superior to the ternary heterogeneous
catalyst control. These findings underscore the stability of the
colloidal catalyst because the peak activity does not diminish
significantly over the run (12 h), unlike the reference system.
It is clear that activity is strongly dependent on the relative
quantities of zinc oxide and copper, at a fixed overall metal
concentration (Figure S8). Optimum activities were observed
for a 1:1 molar ratio of ZnEt /Cu(stearate) precursors, which
corresponds to an expected weight ratio of 55:45 ZnO/Cu,
assuming complete conversions to the relevant oxide/metal
nanoparticles (as indicated by the spectroscopic data). The
activity rapidly decreased when either an excess of diethyl zinc
2
2
Carbon Dioxide Hydrogenation Catalysis. Catalyst
Composition. The precatalyst solution was injected into a
degassed solution of squalane in a continuously stirred tank
reactor (CSTR), under an atmosphere of N at 298 K. A 3:1
2
H /CO mixture, at 50 bar, was added, and the reactor was
2
2
(
therefore, an increase in the ZnO loading) or an excess of
heated to 523 K. The reaction was monitored using an in-line
GC to detect the formation of methanol. During the first 120
min, the quantity of methanol produced steadily increased (a
phenomenon partly resulting from reactor parameters, which
introduce a delay in methanol detection by the GC instru-
ment), reaching a steady state after ∼2 h and remaining at this
level for the duration of each experiment (12 h). Initially,
ethane was also detected, but production ceased by 5 h (Figure
S7). It is proposed that the ethane evolution results from the
reaction of zinc ethyl groups in the precatalyst with water,
which either is formed by carbon dioxide hydrogenation or is
present in sufficient quantities in the gases/solvents. Indeed, a
control experiment in which the precatalyst solution was
exposed only to a continuous nitrogen flow in the reactor also
resulted in ethane evolution over 5 h, after which time the
copper stearate was applied. On one hand, sufficient diethyl
zinc is required to reduce the copper completely; on the other
hand, sufficient stearate (initially coordinated to the copper
precursor) is required for nanoparticle stabilization. The
colloidal nanoparticle catalysts show good stability over 12 h
of the run, with activities remaining at or only slightly lower
than the peak value for the duration of the run in the best case.
In contrast, the ternary heterogeneous suspension shows a
significant loss in activity after 12 h on-stream. Furthermore, in
cases where an excess of diethyl zinc is added, significant
deactivation is also observed. This is tentatively attributed to a
reduction in the stability of the colloidal particles due to the
reduced amount of stearate capping ligand available to stabilize
the particles.
2
898
DOI: 10.1021/cs502038y
ACS Catal. 2015, 5, 2895−2902

ACS Catalysis
Research Article
Catalyst Concentration. The influence of overall nano-
particle concentration in the reactor was investigated at the
optimum ratio of ZnO/Cu, 55:45 (1:1, Et Zn/Cu(stearate) )
nature of the precursor. In all cases, the catalysts operated in
the same manner, with the evolution of ethane and formation
of methanol over the first 5 h, followed by a steady state
methanol production that does not significantly decrease over
the following 12 h (Figure S10). Substituting diethyl zinc with
diphenyl zinc, using bis(stearate) copper(II) as the copper
source reduced the catalytic activity by ∼25%. This reduction is
tentatively attributed to differences in the reactivity of diphenyl
zinc, both toward the Cu(II) precursor and toward water
during the formation of zinc oxide in the reactor.
2
2
(Figure 3 and Table S3). As the nanoparticle concentration
The influence of the copper precursor on catalytic activity,
using diethyl zinc as the zinc oxide source, was more subtle.
Using bis(laurate) or bis(octanoate) copper(II) precursors
resulted in slightly more active catalysts than using bis(stearate)
copper(II). It is likely that this effect relates to the balance
between optimum solubility of the nanoparticles and steric
hindrance and protection of the surface active sites. Thus, the
optimum C-chain length for these carboxylate capped nano-
particles appears to be 8 (octanoate).
We have previously reported a catalyst system prepared by
15
mixing solutions of preformed nanoparticles of ZnO/Cu. In
that system, it was observed that increasing the reductive
stability of the zinc oxide nanoparticle capping ligand by
substituting carboxylate for di(octyl)phosphinate resulted in
increased catalytic activity. In this study, changing the copper
precursor to di(octyl)phosphinate has a much lesser influence
Figure 3. Influence of the overall catalyst concentration (ZnO + Cu in
−1
−1
gZnOCu/L) on the methanol synthesis activity (mmol gZnOCu h ).
The error for each measurement is ±4.5% as determined by runs in
triplicate. The complete data set is presented in Table S3, and plots for
each concentration of activity vs time are presented in Figure S9.
15
on activity, resulting in only a small increase. The Et Zn/
2
Cu(di(octyl)phosphinate) system was also tested at the 0.2
2
increases from 0 to 0.2 gZnOCu/L, the activity also steadily
increases, presumably as the probability of an interaction
between particles increases, allowing the formation of a
catalytically active interface between ZnO and Cu. However,
at nanoparticle concentrations above 0.2 gZnOCu/L, the activity
decreases. At the relatively higher concentrations, aggregation
of the nanoparticles appears to increase, presumably reducing
the availability of active sites (which must be on the catalyst
gZnOCu/L catalyst concentration (optimum for stearate),
−1
−1
resulting in a peak activity of 60 mmol gZnOCu h , which is
close to the value for the Cu(stearate) precursor (Figure S11).
2
Characterization. The optimum precatalyst solution was
formed by the in situ reduction of copper stearate/phosphinate
complexes with diethyl zinc. TEM analysis showed small
nanoparticles (5−10 nm in diameter) in agglomerates
containing from two to hundreds of nanoparticles (Figure 4).
Some agglomeration is likely to occur during the catalyst
precipitation, washing, and deposition required for TEM
sample preparation, but the primary nanoparticles do appear
to be intrinsically fused to form small clusters (Figure 4).
Measurements of the lattice spacings in the high-resolution
22
surface) and thereby reducing activity.
Influence of the Organometallic Precursor Com-
plexes. Various copper(II) and di(organo)zinc precursor
complexes were investigated to establish the generality of the
protocol to synthesize the catalyst (Table 2, Figure S10). All
the complexes resulted in the formation of effective catalysts for
methanol synthesis, with the activity depending weakly on the
0
TEM images reveal that the nanoparticles consist of ZnO, Cu ,
and Cu O (Figure 4). Some nanoparticles showed a spacing of
2
only ∼2.47 Å, which corresponds to either ZnO(101) or
Cu O(111). Such nanoparticles cannot be identified unambig-
2
Table 2. Influence of the Organometallic Precursor
uously. A lattice is not resolved in the remaining unidentified
nanoparticles. Phases with larger lattice spacings are easier to
resolve and, hence, easier to identify. ZnO(100) corresponds to
a spacing 35% larger than Cu(111), which may be why the
concentration of ZnO appears higher.
a
Complexes on Methanol Catalytic Activity
zinc
precursor
methanol peak activity, mmol
1
−1
copper precursor
gZnOCu−
h
Cu(stearate)2
ZnEt2
ZnPh2
ZnEt2
ZnEt2
ZnEt2
ZnEt2
42
32
43
44
39
50
Cu(stearate)2
The high-resolution imaging was performed using an
Cu(laurate)2
environmental TEM holder that prevented any exposure of
0
Cu(octanoate)2
Cu(2-ethyl hexanoate)2
the sample to air. Although the direct oxidation of Cu to Cu O
2
by water or CO is unlikely, dissociative adsorption of CO
2
2
0
Cu(dioctyl
phosphinate)2
onto Cu , interacting with ZnO or other promoters, has been
23
frequently reported to form Cu O. Nevertheless, it is possible
2
that trace O during the washing sample preparation could also
a
2
Reaction conditions: 523 K, 50 bar (3:1, H /CO ), in squalane/
2
2
contribute to Cu oxidation. HAADF STEM−EDX further
confirms that Cu-, Zn-, and O-containing nanoparticles are in
close proximity (Figure S12). Intimate interaction between the
copper and zinc oxide surfaces is thought to be important for
toluene (90:10) at a fixed total volume of 100 mL, a total gas flow of
1
5
the overall catalyst concentration (ZnO + Cu) was fixed at 0.4 gZnOCu
L. Error for each measurement is 4.5% as determined by runs in
−1
66 mL min , over 16 h. The weight ratio of ZnO/Cu was fixed at
5:45, assuming complete conversion of the precursor complexes, and
/
5c,7,24
highly active methanol synthesis catalysts
and should be
triplicate. Figure S10 shows the activity vs time data for all catalysts.
encouraged by the formation of zinc oxide in situ with the
2
899
DOI: 10.1021/cs502038y
ACS Catal. 2015, 5, 2895−2902

ACS Catalysis
Research Article
Figure 4. TEM images of the catalyst postreaction. (left) Low-resolution TEM image of discrete nanostructures. (right) High-resolution image of a
nanoparticle cluster containing ZnO, Cu, and Cu O; colors indicate the phase where unambiguously identified by lattice analysis. Catalysis
2
−3
conditions: 1:1 ZnEt /Cu(stearate) at fixed concentration (0.2 g dm ), 523 K, 50 bar (3:1, H /CO ), in squalane at a fixed total volume of 100
2
2
2
2
−
1
mL, a total gas flow of 166 mLmin , over 16 h.
copper nanoparticles. The presence of a contact interface
between Cu-containing and ZnO nanoparticles is clearly
observed in Figure 4.
The Zn K-edge XANES spectrum of the postcatalysis
solution (Figure 2) also indicates that ZnO is present. The
main edge peak observed at 9669 eV and a feature at 9663 eV
are entirely consistent with the 1s → 4p and the 1s → 4pz
x,y
25
transitions of ZnO. The Cu K-edge XANES spectrum of the
same postcatalysis mixture indicates that the major copper-
containing species continue to be Cu(0) nanoparticles (Figure
1). Indeed, the spectrum is essentially the same as that for the
precatalysis mixture. On the other hand, the EXAFS data for
the postcatalysis copper species demonstrate a small
component of oxidic Cu species (as evidenced by the feature
at ∼1.8 Å in the Fourier transform data) as well as the Cu(0)
species. The distance (1.84 Å) is consistent with a Cu(I)−O
bond length, suggesting there may be minor contamination by
Figure 5. XRD of postcatalysis material for 55:45 ZnO/Cu (Table 1,
entry 2). The spectrum is referenced to Cu (PDF no. 01-0851326
Cu O in the postcatalysis sample (Figure S3, Table S1),
2
ICDD), Cu O (PDF no. 034-1354 ICDD), CuO (PDF no. 045-0937
consistent with the TEM observations. The EXAFS also shows
features at R > 3 Å, which are indicative of metallic Cu. These
features are more pronounced compared with the precatalysis
mixture, likely due to an increase in nanoparticle size after
catalysis; however, because of the mixed speciation of the
copper postcatalysis, it is not possible to accurately quantify the
copper nanoparticle size.
Powder X-ray diffraction measurements, conducted on the
samples after centrifugation, also confirmed the presence of
ZnO and copper nanoparticles in the postreaction mixtures.
Figures S13−S15 illustrate the XRD patterns obtained for
different loadings of ZnO/Cu; an illustrative example at 55:45
ZnO/Cu is shown in Figure 5. In every case, there are clearly
peaks that can be indexed against the patterns expected for
ZnO and copper nanoparticles. In most spectra, some copper
2
ICDD), and ZnO (PDF no. 36-1451, ICDD).
the Cu particles appears rather more variable with low loadings
of either ZnO/Cu (relating to lower molar quantities of either
Et Zn or Cu(stearate) ), resulting in the formation of
2
2
significantly larger copper nanoparticles (∼12−15 nm). When
the relative loading of ZnO/Cu is more closely balanced (e.g.,
5
5:45), then the sizes of the nanoparticles are also close, both
being ∼7 nm. This value is slightly higher than in the
precatalyst solution, in line with the TEM images. It is
tentatively proposed that the matching between the ZnO and
Cu nanoparticle sizes may in part rationalize the improved
activity shown at these loadings, possibly by maximizing the
active interface between the two components.
oxidic species are also present (both Cu O and CuO), although
2
CONCLUSIONS
Highly active, quasi-homogeneous CO hydrogenation catalysts
were produced by reaction between discrete inorganic
complexes, namely, bis(carboxylate) copper(II) and diethyl
zinc, to form precatalyst solutions. These solutions were
applied (at 523 K, 50 bar of a 1:3 mixture of CO /H ) to
the oxidation is proposed to occur predominantly during
sample preparation, because the quantities present differ
nonuniformly, and in one case, these species are absent
■
2
(30:70, ZnO/Cu). The Scherrer equation can be used to
estimate the relative particle sizes, and Table S4 lists these
values for the various different loadings. In all cases, ∼7 nm
ZnO nanoparticles are estimated; on the other hand, the size of
2
2
produce methanol in a continuously stirred slurry reactor
2
900
DOI: 10.1021/cs502038y
ACS Catal. 2015, 5, 2895−2902

ACS Catalysis
squalane) with in-line GC detection equipment. The pre- and
Research Article
(
beamtime (SP8071-8). The RCaH are also acknowledged for
postcatalysis solutions were characterized using UV−vis,
use of facilities and support of their staff.
EXAFS, and XANES spectroscopy; microscopy (TEM); and,
1
where possible, by H NMR spectroscopy. The combined
REFERENCES
■
techniques indicate that the precatalysis mixture contains small
(1) (a) MacDowell, N.; Florin, N.; Buchard, A.; Hallett, J.; Galindo,
A.; Jackson, G.; Adjiman, C. S.; Williams, C. K.; Shah, N.; Fennell, P.
Energy Environ. Sci. 2010, 3, 1645−1669. (b) Kember, M. R.; Buchard,
A.; Williams, C. K. Chem. Commun. 2011, 47, 141−163. (c) Aresta, M.
Carbon Dioxide as a Chemical Feedstock. Wiley-VCH: Weinheim, 2010.
d) Darensbourg, D. J. Inorg. Chem. 2010, 49, 10765−10780.
e) Peters, M.; Kohler, B.; Kuckshinrichs, W.; Leitner, W.;
ller, T. E. ChemSusChem 2011, 4, 1216−1240.
f) Wesselbaum, S.; vom Stein, T.; Klankermayer, J.; Leitner, W.
(1−4 nm) copper nanoparticles and organometallic zinc
complexes (diethyl zinc and ethyl zinc stearate cluster). On
the other hand, after catalysis, the organozinc reagents were
hydrolyzed to produce zinc oxide nanoparticles and the copper
nanoparticles of sizes <10 nm. These spectroscopic findings are
in line with ethane evolution (consistent with ethyl−zinc bond
hydrolysis) during the initial phase of catalysis. The catalyst
synthesis method was easily tuned, and the influence of the
relative loadings of ZnO/Cu, the nature of the precursor
compounds, and the effect of overall catalyst concentration
were reported.
(
(
̈
Markewitz, P.; Mu
(
̈
Angew. Chem., Int. Ed. 2012, 51, 7499−7502. (g) Balaraman, E.;
Gunanathan, C.; Zhang, J.; Shimon, L. J. W.; Milstein, D. Nat. Chem.
2011, 3, 609−614. (h) Menard, G.; Stephan, D. W. J. Am. Chem. Soc.
2010, 132, 1796−1797.
The best systems, which were formed using equimolar ratios
of diethyl zinc/bis(stearate) copper in dry toluene gave
activities for methanol production equivalent to those observed
with a commercially available heterogeneous catalyst bench-
mark (mostly comprising ZnO, CuO, and Al O mixtures).
(2) Kondratenko, E. V.; Mul, G.; Baltrusaitis, J.; Larrazabal, G. O.;
Perez-Ramirez, J. Energy Environ. Sci. 2013, 6, 3112−3135.
(
3) (a) Olah, G. A. Angew. Chem., Int. Ed. 2013, 52, 104−107.
(
b) Pontzen, F.; Liebner, W.; Gronemann, V.; Rothaemel, M.; Ahlers,
B. Catal. Today 2011, 171, 242−250. (c) Olah, G. A.; Goeppert, A.;
2
3
Prakash, G. K. S. J. Org. Chem. 2009, 74, 487−498.
(
(
1
Thus, it is clear that formation of the copper nanoparticles, by
reduction with diethyl zinc and subsequent in situ hydrolysis of
the organozinc complexes results in a highly active and selective
catalyst mixture. It is tentatively proposed that this increased
activity may result from a preferential formation of ZnO/Cu
interfaces using the in situ methodology (as compared with
mixing together discrete solutions of nanoparticles). The facility
with which variables can be changed as well as the ease of
catalyst preparation make this route a rather attractive method
for further optimization and investigation. In contrast to other
4) Yang, C.-J.; Jackson, R. B. Energy Policy 2012, 41, 878−884.
̈
5) (a) Strunk, J.; Kahler, K.; Xia, X.; Muhler, M. Surf. Sci. 2009, 603,
776−1783. (b) Baltes, C.; Vukojevic, S.; Schuth, F. J. Catal. 2008,
258, 334−344. (c) Behrens, M.; Studt, F.; Kasatkin, I.; Kuhl, S.;
Havecker, M.; Abild-Pedersen, F.; Zander, S.; Girgsdies, F.; Kurr, P.;
Kniep, B. L.; Tovar, M.; Fischer, R. W.; Norskov, J. K.; Schlogl, R.
Science 2012, 336, 893−897. (d) Hansen, P. L.; Wagner, J. B.; Helveg,
S.; Rostrup-Nielsen, J. R.; Clausen, B. S.; Topsoe, H. Science 2002, 295,
2
053−2055. (e) Fichtl, M. B.; Schumann, J.; Kasatkin, I.; Jacobsen, N.;
Behrens, M.; Schlogl, R.; Muhler, M.; Hinrichsen, O. Angew. Chem.,
Int. Ed. 2014, 53, 7043−7047.
6) (a) Saito, M.; Murata, K. Catal. Surv. Asia 2004, 8, 285−294.
b) Saito, M.; Takeuchi, M.; Fujitani, T.; Toyir, J.; Luo, S. C.; Wu, J.
G.; Mabuse, H.; Ushikoshi, K.; Mori, K.; Watanabe, T. Appl.
Organomet. Chem. 2000, 14, 763−772. (c) Wang, W.; Wang, S. P.;
Ma, X. B.; Gong, J. L. Chem. Soc. Rev. 2011, 40, 3703−3727. (d) Wu, J.
G.; Saito, M.; Takeuchi, M.; Watanabe, T. Appl. Catal., A 2001, 218,
̈
(
high temperature) methods to prepare Cu/ZnO particles, this
(
synthesis operates at room temperature in organic solvents and
yields highly reproducible, small nanoparticles surface-capped
with organic ligands. Given the synthetic simplicity and
potential to form nonequilibrium products, this method may
well be applicable to a range of different catalytic systems,
including those used for the production of methanol from
synthesis gas.
(
2
3
35−240. (e) Li, C.; Yuan, X.; Fujimoto, K. Appl. Catal., A 2014, 469,
06−311. (f) Natesakhawat, S.; Lekse, J. W.; Baltrus, J. P.; Ohodnicki,
P. R.; Howard, B. H.; Deng, X. Y.; Matranga, C. ACS Catal. 2012, 2,
1
667−1676. (g) Saito, M.; Fujitani, T.; Takeuchi, M.; Watanabe, T.
ASSOCIATED CONTENT
Supporting Information
The following file is available free of charge on the ACS
Publications website at DOI: 10.1021/cs502038y.
■
Appl. Catal., A 1996, 138, 311−318. (h) Fisher, I. A.; Bell, A. T. J.
Catal. 1997, 172, 222−237. (i) Liu, X. M.; Lu, G. Q.; Yan, Z. F.;
Beltramini, J. Ind. Eng. Chem. Res. 2003, 42, 6518−6530. (j) Fujita, S.;
Usui, M.; Ito, H.; Takezawa, N. J. Catal. 1995, 157, 403−413.
*
S
(k) Koeppel, R. A.; Baiker, A.; Wokaun, A. Appl. Catal., A 1992, 84,
Experimental details, 15 figures, 4 tables of results;
additional references (PDF)
7
7−102. (l) Dutta, G.; Sokol, A. A.; Catlow, C. R. A.; Keal, T. W.;
Sherwood, P. ChemPhysChem 2012, 13, 3453−3456. (m) Grabow, L.
C.; Mavrikakis, M. ACS Catal. 2011, 1, 365−384.
AUTHOR INFORMATION
Corresponding Authors
E-mail: m.shaffer@imperial.ac.uk.
E-mail: c.k.williams@imperial.ac.uk.
(7) Behrens, M.; Schloegl, R. Z. Anorg. Allg. Chem. 2013, 639, 2683−
■
2
695.
(
8) Weiss, R.; Guo, Y. Z.; Vukojevic, S.; Khodeir, L.; Boese, R.;
Schuth, F.; Muhler, M.; Epple, M. Eur. J. Inorg. Chem. 2006, 1796−
802.
9) (a) Liao, F. L.; Huang, Y. Q.; Ge, J. W.; Zheng, W. R.; Tedsree,
K.; Collier, P.; Hong, X. L.; Tsang, S. C. Angew. Chem., Int. Ed. 2011,
0, 2162−2165. (b) Liao, F. L.; Zeng, Z. Y.; Eley, C.; Lu, Q.; Hong, X.
L.; Tsang, S. C. E. Angew. Chem., Int. Ed. 2012, 51, 5832−5836.
c) Zhou, X. W.; Qu, J.; Xu, F.; Hu, J. P.; Foord, J. S.; Zeng, Z. Y.;
*
*
1
(
Notes
The authors declare no competing financial interest.
5
ACKNOWLEDGMENTS
■
(
At Imperial College London, the EPSRC are acknowledged for
funding (EP/H046380, EP/K035274/1), as well as the Energy
Futures Lab, the Alan Howard Studentship (Y.W.), and the
EPSRC Centre for Doctoral Training in Plastic Electronics
Hong, X. L.; Tsang, S. C. E. Chem. Commun. 2013, 49, 1747−1749.
(
(
10) Lee, S. G.; Sardesai, A. Top. Catal. 2005, 32, 197−207.
11) (a) Hambrock, J.; Rabe, S.; Merz, K.; Birkner, A.; Wohlfart, A.;
Fischer, R. A.; Driess, M. J. Mater. Chem. 2003, 13, 1731−1736.
(
M.A.). Dr Peter Wells and Emma Gibson acknowledge
(b) Hambrock, J.; Schroter, M. K.; Birkner, A.; Woll, C.; Fischer, R. A.
EPSRC funding for the XAFS measurements (EP/I019693/1,
EP/K014714/1) and Diamond Light Source for provision of
Chem. Mater. 2003, 15, 4217−4222. (c) Lu, L. H.; Wohlfart, A.; Parala,
H.; Birkner, A.; Fischer, R. A. Chem. Commun. 2003, 40−41.
2
901
DOI: 10.1021/cs502038y
ACS Catal. 2015, 5, 2895−2902

ACS Catalysis
Research Article
(
d) Tkachenko, O. P.; Klementiev, K. V.; Loffler, E.; Ritzkopf, I.;
Mol. Catal. A: Chem. 1996, 113, 455−468. (c) Chinchen, G. C.;
Spencer, M. S.; Waugh, K. C.; Whan, D. A. J. Chem. Soc., Faraday
Trans. 1987, 83, 2193−2212. (d) Chinchen, G. C.; Waugh, K. C.;
Whan, D. A. Appl. Catal. 1986, 25, 101−107.
Schuth, F.; Bandyopadhyay, M.; Grabowski, S.; Gies, H.; Hagen, V.;
Muhler, M.; Lu, L. H.; Fischer, R. A.; Grunert, W. Phys. Chem. Chem.
Phys. 2003, 5, 4325−4334. (e) Kurtz, M.; Bauer, N.; Buscher, C.;
Wilmer, H.; Hinrichsen, O.; Becker, R.; Rabe, S.; Merz, K.; Driess, M.;
Fischer, R. A.; Muhler, M. Catal. Lett. 2004, 92, 49−52. (f) Lu, L. H.;
Fischer, R. A. Chem. Lett. 2004, 33, 1318−1319. (g) Cokoja, M.;
Parala, H.; Schroter, M. K.; Birkner, A.; van den Berg, M. W. E.;
Klementiev, K. V.; Grunert, W.; Fischer, R. A. J. Mater. Chem. 2006,
(24) (a) Behrens, M. J. Catal. 2009, 267, 24−29. (b) Behrens, M.;
Zander, S.; Kurr, P.; Jacobsen, N.; Senker, J.; Koch, G.; Ressler, T.;
̈
Fischer, R. W.; Schlogl, R. J. Am. Chem. Soc. 2013, 135, 6061−6068.
(25) Grandjean, D.; Castricum, H. L.; Van Den Heuvel, J. C.;
Weckhuysen, B. M. J. Phys. Chem. B 2006, 110, 16892−16901.
1
6, 2420−2428. (h) Schroeder, F.; Hermes, S.; Parala, H.; Hikov, T.;
Muhler, M.; Fischer, R. A. J. Mater. Chem. 2006, 16, 3565−3574.
i) Schroter, M. K.; Khodeir, L.; van den Berg, M. W. E.; Hikov, T.;
(
Cokoja, M.; Miao, S. J.; Grunert, W.; Muhler, M.; Fischer, R. A. Chem.
Commun. 2006, 2498−2500. (j) Lu, L.; Hu, S.; Lee, H.-I.; Woell, C.;
Fischer, R. A. J. Nanoparticle Res. 2007, 9, 491−496. (k) Hikov, T.;
Rittermeier, A.; Luedemann, M.-B.; Herrmann, C.; Muhler, M.;
Fischer, R. A. J. Mater. Chem. 2008, 18, 3325−3331. (l) Mueller, M.;
Hermes, S.; Kaehler, K.; van den Berg, M. W. E.; Muhler, M.; Fischer,
R. A. Chem. Mater. 2008, 20, 4576−4587. (m) Lee, B. H.; Hwang, J.
K.; Nam, J. W.; Lee, S. U.; Kim, J. T.; Koo, S.-M.; Baunemann, A.;
Fischer, R. A.; Sung, M. M. Angew. Chem., Int. Ed. 2009, 48, 4536−
4
539. (n) Rittermeier, A.; Miao, S.; Schroeter, M. K.; Zhang, X.; van
den Berg, M. W. E.; Kundu, S.; Wang, Y.; Schimpf, S.; Loeffler, E.;
Fischer, R. A.; Muhler, M. Phys. Chem. Chem. Phys. 2009, 11, 8358−
8
366. (o) Schimpf, S.; Rittermeier, A.; Zhang, X.; Li, Z.-A.; Spasova,
M.; van den Berg, M. W. E.; Farle, M.; Wang, Y.; Fischer, R. A.;
Muhler, M. ChemCatChem 2010, 2, 214−222. (p) Sliem, M. A.; Hikov,
T.; Li, Z.-A.; Spasova, M.; Farle, M.; Schmidt, D. A.; Havenith-Newen,
M.; Fischer, R. A. Phys. Chem. Chem. Phys. 2010, 12, 9858−9866.
(
q) Sliem, M. A.; Turner, S.; Heeskens, D.; Kalidindi, S. B.; Van
Tendeloo, G.; Muhler, M.; Fischer, R. A. Phys. Chem. Chem. Phys.
012, 14, 8170−8178. (r) Vukojevic, S.; Trapp, O.; Grunwaldt, J. D.;
Kiener, C.; Schuth, F. Angew. Chem., Int. Ed. 2005, 44, 7978−7981.
12) Simon, M.; Mortreux, A.; Petit, F. J. Chem. Soc., Chem. Commun.
988, 1445−1446.
13) Corker, J. M.; Evans, J. J. Chem. Soc., Chem. Commun. 1994,
027−1029.
14) (a) Gonzalez-Campo, A.; Orchard, K. L.; Sato, N.; Shaffer, M. S.
2
(
1
(
1
(
P.; Williams, C. K. Chem. Commun. 2009, 4034−4036. (b) Orchard, K.
L.; Harris, J. E.; White, A. J. P.; Shaffer, M. S. P.; Williams, C. K.
Organometallics 2011, 30, 2223−2229. (c) Orchard, K. L.; Shaffer, M.
S. P.; Williams, C. K. Chem. Mater. 2012, 24, 2443−2448. (d) Orchard,
K. L.; White, A. J. P.; Shaffer, M. S. P.; Williams, C. K. Organometallics
2
009, 28, 5828−5832.
15) Brown, N. J.; Weiner, J.; Hellgardt, K.; Shaffer, M. S. P.;
Williams, C. K. Chem. Commun. 2013, 49, 11074−11076.
16) (a) Vidjayacoumar, B.; Emslie, D. J. H.; Blackwell, J. M.;
Clendenning, S. B.; Britten, J. F. Chem. Mater. 2010, 22, 4854−4866.
b) Vidjayacoumar, B.; Emslie, D. J. H.; Clendenning, S. B.; Blackwell,
J. M.; Britten, J. F.; Rheingold, A. Chem. Mater. 2010, 22, 4844−4853.
17) (a) Curtis, A. C.; Duff, D. G.; Edwards, P. P.; Jefferson, D. A.;
Johnson, B. F. G.; Kirkland, A. I.; Wallace, A. S. J. Phys. Chem. 1988,
2, 2270−2275. (b) Papavassiliou, C.; Kokkinakis, T. J. Phys. F: Met.
Phys. 1974, 4, L67.
18) (a) Rothe, J.; Hormes, J.; Bonnemann, H.; Brijoux, W.; Siepen,
(
(
(
(
9
(
K. J. Am. Chem. Soc. 1998, 120, 6019−6023. (b) Montano, P. A.;
Shenoy, G. K.; Alp, E. E.; Schulze, W.; Urban, J. Phys. Rev. Lett. 1986,
5
(
1
(
1
(
6, 2076−2079.
19) Beale, A. M.; Weckhuysen, B. M. Phys. Chem. Chem. Phys. 2010,
2, 5562−5574.
20) Creighton, J. A.; Eadon, D. G. J. Chem. Soc., Faraday Trans.
991, 87, 3881−3891.
21) Sawant, A.; Ko, M. K.; Parameswaran, V.; Lee, S.; Kulik, C. J.
Fuel Sci. Technol. 1987, 5, 77−88.
22) (a) Selvam, N. C. S.; Narayanan, S.; Kennedy, L. J.; Vijaya, J. J. J.
(
Environ. Sci. 2013, 25, 2157−2167. (b) Huang, M.; Xu, C.; Wu, Z.;
Huang, Y.; Lin, J.; Wu, J. Dyes Pigm. 2008, 77, 327−334.
(
23) (a) Fujitani, T.; Saito, M. Catal. Lett. 1994, 25, 271−276.
(
b) Yurieva, T. M.; Plyasova, L. M.; Makarova, O. V.; Krieger, T. A. J.
2
902
DOI: 10.1021/cs502038y
ACS Catal. 2015, 5, 2895−2902