Palladium nanoparticles supported on carbon nanotubes from solventless preparations: Versatile catalysts for ligand-free Suzuki cross coupling reactions

Article abstract of DOI:10.1039/c3ta12512b

Palladium nanoparticles supported on single- or multi-walled carbon nanotubes (Pd/SWCNT and Pd/MWCNT) were prepared by a rapid, solventless method that does not require reducing agents or electric current. The method involves a straightforward process using dry mixing of a precursor Pd salt (e.g., palladium acetate) with carbon nanotubes at ambient temperature by ball-milling (mechanochemical route) or with subsequent annealing at 300°C (thermal route) in an inert atmosphere. The Pd/MWCNT sample with Pd nanoparticle size of 1-3 nm and uniform dispersion prepared by mechanochemical ball-milling at room temperature [designated as (Pd/MWCNT)_M] displayed remarkable catalytic activity towards Suzuki cross coupling reactions with a high turn over number (TON) of 7250 and turn over frequency (TOF) of 217-500 h^-1. These nanoparticles were characterized by a variety of techniques including transmission electron microscopy (TEM), X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS). Additionally, the (Pd/MWCNT)_M sample was successfully employed in Suzuki cross coupling reactions with a wide variety of functionalized substrates.

Full text of DOI:10.1039/c3ta12512b

View Article Online
View Journal
JMouarntaelroiaf lsChemistryA
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: A. R. Siamaki, Y. Lin, K. Woodberry, J.
W. Connell and F. Gupton, J. Mater. Chem. A, 2013, DOI: 10.1039/C3TA12512B.
This is an Accepted Manuscript, which has been through the RSC Publishing peer
Journalof
review process and has been accepted for publication.
Materials ChemistryA
Accepted Manuscripts are published online shortly after acceptance, which is prior
Materials for energy and sustainability
www.rsc.org/MaterialsA
Volume 1 | Number 1 | January 2013 | Pages 0000–0000
to technical editing, formatting and proof reading. This free service from RSC
Publishing allows authors to make their results available to the community, in
citable form, before publication of the edited article. This Accepted Manuscript will
be replaced by the edited and formatted Advance Article as soon as this is available.
To cite this manuscript please use its permanent Digital Object Identifier (DOI®),
which is identical for all formats of publication.
More information about Accepted Manuscripts can be found in the
Information for Authors.
Please note that technical editing may introduce minor changes to the text and/or
graphics contained in the manuscript submitted by the author(s) which may alter
content, and that the standard Terms & Conditions and the ethical guidelines
that apply to the journal are still applicable. In no event shall the RSC be held
responsible for any errors or omissions in these Accepted Manuscript manuscripts or
any consequences arising from the use of any information contained in them.
ISSN 2050-7488
0959-9428(2010)20:1;1-A
www.rsc.org/materialsA
Registered Charity Number 207890

Page 1 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
Palladium Nanoparticles Supported on Carbon Nanotubes from
Solventless Preparations: Versatile Catalysts for Ligand-Free
Suzuki Cross Coupling Reactions
Ali R. Siamaki,1 Yi Lin,2 Kendra Woodberry,1 John W. Connell,3 and B. Frank
Gupton1
1 Department of Life Science and Chemical Engineering, Virginia Commonwealth
University, Richmond, VA, 23284, USA
2 National Institute of Aerospace, 100 Exploration Way, Hampton, VA 23666
3Advanced Materials and Processing Branch, NASA Langley Research Center, Hampton,
VA, 23681ꢀ2199, USA
Palladium nanoparticles supported on singleꢀ or multiꢀwalled carbon nanotubes
(Pd/SWCNT and Pd/MWCNT) were prepared by a rapid, solventless method that
does not require reducing agents or electric current. The method involves a
straightforward process using dry mixing of a precursor Pd salt (e.g., palladium
acetate) with carbon nanotubes at ambient temperature by ballꢀmilling
(mechanochemical route) or with subsequent annealing at 300 oC (thermal route)
in an inert atmosphere. The Pd/MWCNT sample with Pd nanoparticle size of 1ꢀ3
nm and uniform dispersion prepared by mechanochemical ballꢀmilling at room
temperature [designated as (Pd/MWCNT)M] displayed remarkable catalytic
activity towards Suzuki cross coupling reactions with a high turn over number
(TON) of 7250 and turn over frequency (TOF) of 217,500 hꢀ1.
These
nanoparticles were characterized by a variety of techniques including
transmission electron microscopy (TEM), Xꢀray diffraction (XRD) and Xꢀray
photoelectron spectroscopy (XPS). Additionally, the (Pd/MWCNT)M sample was
successfully employed in Suzuki cross coupling reactions with wide variety of
functionalized substrates.
1

Journal of Materials Chemistry A
Page 2 of 31
View Article Online
DOI: 10.1039/C3TA12512B
Introduction
Carbon nanotubes including singleꢀwalled (SWCNT) and multiꢀwalled (MWCNT)
nanotubes have attracted considerable interests due to their remarkable thermal
conductivity,1ꢀ3 mechanical,4ꢀ6 and electrical properties.6,7 In addition, their excellent
chemical and mechanical stability and large surface area make them an ideal and
potentially useful choice for applications in optics,7ꢀ8 electronics,9 catalysis,10ꢀ11 polymer
composites,12 and many others.13 In this context, the deposition of the metal
nanoparticles on carbon nanotubes (also referred to as “decoration”) has provided further
promising utilities for these materials in a wide range of applications from hydrogen
storage, solar cells, sensor devices, energy storage, as well as catalysis.14,15 As such, one
of the potential areas in which carbon nanotubes may be utilized as an effective solid
support is in cross coupling reactions.16ꢀ20
Metal catalyzed carbonꢀcarbon bond forming reactions such as Suzuki, Heck, and
Sonogashira and many others have rapidly become one of the most effective tools in
organic synthesis for the assembly of highly functionalized molecules.21ꢀ25 These
reactions are typically performed under homogeneous reaction conditions, which require
ligands to stabilize the catalyst and broaden the range of their reactivity. 21ꢀ25 The use of
homogeneous catalysis in cross coupling reactions, however, has some major limitations
in large scale pharmaceutical and industrial synthesis mainly due to the lack of
recyclability, difficulty in separation of the catalyst from the reaction mixture, and
therefore the potential concern over residual contamination of the metals in the reaction
products.26,27 One promising solution to address this issue is to employ solidꢀsupported
metal nanocatalysts capable of not only efficiently catalyzing the reactions, but also
2

Page 3 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
providing a unique opportunity to leverage the reactivity and stability of these materials
along with the ease of separation and reusability in subsequent reactions. Thus,
numerous efforts have been undertaken in recent years to develop novel metal
nanocatalysts for cross coupling reactions in which the metal is fixed on a variety of solid
supports such as zeolites,28,29 polymers,30,31 mesoporous silica,32,33 inorganic oxides,34,35
and activated carbons.36,37 Among these support systems, graphene38ꢀ45 and carbon
nanotubes16ꢀ20 have attracted significant attention due to the high specific surface area,
thermal stability, and remarkable structural and electronic features capable of stabilizing
the metal nanoparticles.
The preparation of metal nanoparticles supported on graphene or carbon nanotubes
49,50
has typically required the use of a reducing agent46ꢀ48 or electrochemical reduction
for the conversion of metal ions to zero valence metal. Alternatively, we have recently
reported a series of solventless methods to prepare bulk quantities of metal nanoparticles
supported on carbon nanotubes starting from solid mixtures of carbon nanotubes and
precursor metal salts (typically acetates or acetylacetonates).51ꢀ53 The mixture was
subjected to either conventional heating in a nitrogen oven (the “thermal” route)51 or
rapid heating using a microwave reactor (the “microwave” route).53 Without the presence
of any other additional reagent, the metal salts thermally decomposed into metals (mostly
noble metals including Ag, Au, Pt, Pd) or metal oxides (most other transition metals). An
interesting observation when using a mechanical mixer for the saltꢀnanotube blending
was that the precursor salt would undergo partial decomposition just by the mechanical
energy provided during the aggressive ballꢀmilling process and form ultrafine (typically <
10 nm) nanoparticles without the need of additional heating (the “mechanochemical”
3

Journal of Materials Chemistry A
Page 4 of 31
View Article Online
DOI: 10.1039/C3TA12512B
route).52 We present herein the preparation of Pd nanoparticles supported on singleꢀ
walled and multiꢀwalled carbon nanotubes using the thermal [(Pd/MWCNT)T,
(Pd/SWCNT)T] and mechanochemical routes [(Pd/MWCNT)M (Pd/SWCNT)M], as well
as the excellent catalytic activity of the latter in Suzuki cross coupling reactions.
Importantly, these methods are facile, lowꢀcost, and environmental friendly, allowing for
a large scale preparation of Pd nanoparticles supported on carbon nanotubes which can be
further employed in industrial and pharmaceutical applications.
4

Page 5 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
Experimental
Materials and methods
MWCNTs of a broad diameter distribution (~20ꢀ150 nm) were purchased from the
University of Kentucky. SWCNTs were obtained from Carbon Solutions, Inc. Palladium
(II) acetate (99.9+%) were obtained from Aldrich Chemical Co. Aryl bromides and
iodides, potassium carbonate, aryl substituted boronic acid were also purchased from
Aldrich and used asꢀreceived. A mixture of ethanol/deionized water was used for the
Suzuki cross coupling reactions.
Transmission electron microscopy (TEM) images were acquired using a JEOL JEMꢀ
1230 electron microscope operated at 120 kV equipped with a Gatan UltraScan 4000SP
4K x 4K CCD camera. Samples for TEM were prepared by placing a droplet of a colloid
suspension in toluene onto a Formvar carbonꢀ or holey carbonꢀcoated copper grid (Ted
Pella or Structure Probes, Inc.) and allowed to evaporate in air at room temperature. The
Xꢀray diffraction (XRD) patterns were measured using PANalytical Xꢀray diffraction
system.
The Xꢀray photoelectron spectroscopy (XPS) analysis was performed on a Thermo
Fisher Scientific ESCALAB 250 using a monochromatic Al KR Xꢀray. The Pd content
in each of the Pd nanoparticles supported on carbon nanotubes were determined using an
Inductively Coupled Plasma equipped with Mass Spectrometry (ICPꢀMS, Varian 820ꢀ
MS). Gas chromatography ꢀ mass spectroscopy (GCꢀMS) analyses were performed on an
Agilent 6890 gas chromatograph equipped with an Agilent 5973 mass selective detector.
A CEM Discover microwave instrument was used for cross coupling reactions. The
reactions were performed at operator selectable power output of 250 W. 1H and 13C NMR
5

Journal of Materials Chemistry A
Page 6 of 31
View Article Online
DOI: 10.1039/C3TA12512B
were acquired on a Mercury 300 MHz spectrometer. High resolution mass spectrometry
analyses were obtained for the products of the Suzuki reactions in Table 2 from the
Virginia Commonwealth University mass spectrometry facilities.
Syntheses of Pd-carbon nanotube catalysts
Both Pd/SWCNT and Pd/MWCNT catalysts were prepared using the same
procedures except for variations in the PdꢀtoꢀC ratio. Examples are given below for
catalysts prepared using the mechanochemical route and the thermal route, respectively.
Mechanochemical Route. In a typical experiment to prepare Pd/MWCNT with a Pdꢀ
toꢀC weight ratio of 1:10 (or ~9 wt% Pd), MWCNT powder (500 mg) and palladium
acetate (105 mg) were added to a 20ꢀmL volume ceramic vial (SPEX CertiPrep). After
adding two ceramic balls (d = 1.3 cm), the vial was subjected to mechanical shaking in a
ballꢀmill mixer (SPEX CertiPrep 8000D HighꢀEnergy Shaker Mill) for 30 min to yield
the final product (Pd/MWCNT)M. When in operation, the ballꢀmill mixer typically
provides mechanical motions of ~1060 lateral cycles per minute with 5.7 cm backꢀandꢀ
forth and 2.5 cm sideꢀtoꢀside movements.
Thermal Route. In a typical experiment, the product from the mechanochemical
reaction, (Pd/MWCNT)M, was transferred to a glass vial or an aluminum pan and heated
in a nitrogen oven (Blue M Electric Aꢀ5245ꢀQ Inert Gas Oven, nitrogen flow rate at ~80
cm3/min) to 300 °C over 1 h and held isothermally for 3 h. The product was then
obtained as (Pd/MWCNT)T.
6

Page 7 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
Procedure for determining catalyst concentration effects (Fig. 4)
Bromobenzene (50 mg, 0.32 mmol, 1 eq.) was dissolved in a mixture of 4 mL
H2O:EtOH (1:1) and placed in a 10 mL microwave tube. To this was added phenyl
boronic acid (47 mg, 0.382 mmol, 1.2 eq.), and potassium carbonate (133 mg, 0.96
mmol, 3 eq.). Palladium catalysts (X mmol, Y mol% as indicated in Fig. 4) were then
added, and the tube was sealed and stirred at room temperature (r.t.). An aliquot of the
reaction mixtures was taken after 5, 15, 30, 60, 120, 180, 240, 300 minutes, diluted with
10 mL of CH3CN and injected into the GC/MS. The percent conversions of the products
were then calculated based on the consumption of bromobenzene starting material as
determined from GCꢀMS spectroscopy.
Procedure for recycling the (Pd/MWCNT)M and (Pd/SWCNT)M nanocatalysts using
bromo and iodobenzenes (Table 1)
Bromobenzene (50 mg, 0.32 mmol, 1 eq.) was dissolved in a mixture of 4 mL
H2O:EtOH (1:1) and placed in a 10 mL microwave tube. To this was added phenyl
boronic acid (47 mg, 0.382 mmol, 1.2 eq.), potassium carbonate (133 mg, 0.96 mmol, 3
eq.). (Pd/MWCNT)M (2 mg, 1.6 ꢁmol, 0.5 mol%) or (Pd/SWCNT)M (2.3 mg, 1.6 ꢁmol,
0.5 mol%) were then added, the tube was sealed, and heated at 80 ºC for 10 min under
microwave irradiation (250 W, 2.45 MHz). The progress of the reaction was monitored
by GCꢀMS analysis. After the completion of the reaction, the mixture was diluted with
10 mL of EtOH and shaken. The entire mixture was centrifuged and the solvent above
the Pd nanoparticle catalyst was completely decanted. The EtOH washing followed by
centrifugation was repeated two additional times to ensure the removal of all the products
from the catalyst surface. The (Pd/MWCNT)M catalyst was then directly transferred to
7

Journal of Materials Chemistry A
Page 8 of 31
View Article Online
DOI: 10.1039/C3TA12512B
another microwave tube along with fresh reagents for the next run. This procedure was
repeated for every run and the percent conversion of product was determined by the
means of GCꢀMS spectroscopy. The same experimental details were applied for
recycling the (Pd/MWCNT)M and (Pd/SWCNT)M nanoparticles using iodobenzene with
the amount of reagents as follows: Iodobenzene (50 mg, 0.245 mmol), phenylboronic
acid (36 mg, 0.294 mmol, 1.2 eq.), potassium carbonate (102 mg, 0.735 mmol, 3 eq.), and
(Pd/MWCNT)M (1.5 mg, 1.2 ꢂmol, 0.5 mol%) or (Pd/SWCNT)M (1.7 mg, 1.6 ꢁmol, 0.5
mol%) in 4 mL (H2O:EtOH) (1:1), heated at 80 °C under microwave irradiation for 10
min.
General procedure for diversity scope in Suzuki reactions (Table 2)
Aryl bromide (0.32 mmol, 1 eq.) was dissolved in a mixture of 4 mL H2O:EtOH 1:1
and placed in a 10 mL microwave tube. To this was added the arylboronic acid (0.384
mmol, 1.2 eq.) and potassium carbonate (0.96 mmol, 3 eq.). A (Pd/MWCNT)M sample
(2 mg, 1.6 ꢁmol, 0.5 mol%) was then added, and the tube was sealed and heated under
microwave irradiation (250 W, 2.45 MHz) at a specific temperature and time indicated in
Table 2. After completion, the reaction mixture was diluted with H2O (30 mL) and
extracted with dichloromethane (CH2Cl2 3 × 50 mL). The organic layers were combined,
dried over MgSO4, and filtered. The solvent in the filtrate was then removed in vacuo to
give a solid which was further purified by flash chromatography using hexane:ethyl
acetate as eluent.
8

Page 9 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
Results and Discussion
Pd/MWCNT samples were prepared using the solventless procedures that were
previously established in our lab with the most detailed investigations conducted mainly
on Ag/MWCNT systems.51,52 In those systems, silver acetate was the precursor salt for
metallic Ag nanoparticles. As previously discussed in those published reports,51ꢀ53 the
Pd/MWCNT system behaved very similarly. Mechanistically, the formation of Pd
nanoparticles on MWCNT surface was due to the direct decomposition of the precursor
salt, palladium acetate, into palladium metal along with volatile byꢀproducts such as CO,
CO2, acetic acid, and water. The salt decomposition was triggered by either the
mechanical energy (mechanochemical route) or subsequent heat treatment (thermal
route). The decomposition reaction can be viewed as a partial disproportionation process,
thus no reducing agent was needed. The nanotubes used were pristine and therefore
contained minimal surface functional groups. However, there were intrinsic surface
defect sites on the nanotube surface, which provided strong Pd nanoparticle anchoring
especially in the thermal route.51 Other factors that might have affected the nanoparticle
anchoring include the nanotube diameter and impurity content.
The average size, distribution and the composition of Pd nanoparticles decorated on
the MWCNT surfaces from the two routes investigated were distinctively different. As
shown in the TEM images in Fig. 1, the nanotubeꢀattached Pd nanoparticles from the
mechanochemical route (Pd/MWCNT)M were much smaller (and denser in decoration)
than those from thermal route (Pd/MWCNT)T (3.4 ± 1.3 nm vs. 9.4 ± 3.6 nm) and
displayed a more uniform distribution of Pd nanoparticles on the surface of MWCNT. In
contrast, significant increase in sizes of the Pd nanoparticles was apparent over the large
9

Journal of Materials Chemistry A
Page 10 of 31
View Article Online
DOI: 10.1039/C3TA12512B
areas of several microns on the TEM image of Pd nanoparticles prepared by the thermal
route (Pd/MWCNT)T (Fig. 1b). The size increase of Pd nanoparticles in this sample was
apparently the result of the additional heating at 300 °C. Such heating in the thermal
route induced the complete decomposition of palladium acetate (see below) and the
simultaneous Ostwald ripening of the existing subꢀ5 nm nanoparticles from
mechanochemical mixing. Compared to the kinetically formed smaller Pd nanoparticles
during mixing, the formation of larger nanoparticles from the subsequent heating should
be more thermodynamically driven. Similarly, the TEM images of two Pd/SWCNT
samples prepared by mechanochemical and thermal methods exhibited average particle
sizes of 6.6 ± 2.1 nm and 10.7 ± 3.6 nm for (Pd/SWCNT)M and (Pd/SWCNT)T,
respectively. Significant size increase of the Pd nanoparticles was also observed on the
broad areas of the TEM image of the thermally obtained sample (Pd/SWCNT)T in
comparison to the mechanochemically obtained sample (Pd/SWCNT)M (Fig. 1cꢀd). It
should be noted that it was challenging to identify individual or even bundles of
SWCNTs after ballꢀmilling because the nanotubes were partially amorphortized by the
mechanical forces. In addition, the impurities (amorphous carbon, catalyst nanoparticles,
etc) from the pristine SWCNT sample might have also masked the nanotubes.
Nevertheless, the Pd content in all Pdꢀcarbon nanotube samples were well controlled by
the amount of Pd salt used. This was further confirmed by inductively coupled plasma
equipped with mass spectroscopy (ICPꢀMS). For Pd/nanotube samples with a targeted
Pd content of ~9 wt%, ICPꢀMS analysis indicated Pd content values of 8.5 wt%,7.3 wt%,
7.3 wt%, 7.0 wt% for (Pd/MWCNT)M, (Pd/MWCNT)T, (Pd/SWCNT)M, and
(Pd/SWCNT)T, respectively.
10

Page 11 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
Fig. 1 TEM images of Pd/MWCNT
and Pd/SWCNT nanoparticles a) (Pd/MWCNT)M b)
(Pd/MWCNT)T c) (Pd/SWCNT)M d) (Pd/SWCNT)T
a)
b)
c)
11

Journal of Materials Chemistry A
Page 12 of 31
View Article Online
DOI: 10.1039/C3TA12512B
d)
Fig. 2a displays the XRD patterns of Pd/MWCNT nanoparticles prepared by
mechanochemical and thermal routes. The peaks at 2θ ~ 26° were assigned to the
signature graphitic layered structure, the (002) peak, of MWCNTs in (Pd/MWCNT)M and
(Pd/MWCNT)T, respectively (Fig. 2a, b). The typical XRD patterns of Pd(0) (ICPDS
#05ꢀ0681) can be observed in the (Pd/MWCNT)T spectrum in Fig. 2b. For example, the
peak at 2θ value of 40.1° corresponds to the (111) dꢀspacing of 2.2 Å of metallic Pd.
Other peaks were of lower intensity, consistent with the standard pattern. The broadening
of the peaks was due to the nanoscale nature of the crystalline phase. For the
(Pd/MWCNT)M sample, most peaks from Pd(0) were discernible but of much lower
intensity (Fig. 2a). This was likely due to the lower saltꢀtoꢀmetal conversion and the
smaller particle size (broadening effect) in this sample. XRD results on the Pd/SWCNTs
were similar (Fig. S1, Supporting Information), despite significant impurity peaks from
the pristine nanotubes.
12

Page 13 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
Fig. 2 XRD spectra of a) (Pd/MWCNT)M b) (Pd/MWCNT)T
In our previous work,51 it was established that the mechanical energy during ballꢀ
milling might not be sufficient to fully decompose the metal salt used. Unlike the case of
silver acetate, the XRD patterns of residual palladium acetate were not readily
identifiable in the products from room temperature mechanochemical reactions (Fig. 2 a).
Therefore, the XPS data was further employed to determine the composition of the Pd
species in both samples (Fig. 3). The XPS spectrum of (Pd/MWCNT)M indicates the
presence of a mixture of Pd(0) and Pd(II) (52% to 48%) as evident from the measured
binding energy of the Pd 3d5/2 and 3d3/2 electrons at 335.8 and 341 eV for Pd(0) and
338.1 and 343.2 eV for Pd(II), respectively (Fig. 3a). Consistently, the XPS data of
(Pd/MWCNT)T shows a dominating Pd(0) (~90%) due to two significant binding
energies of the Pd 3d5/2 and 3d3/2 electrons at 336 and 341.4 eV, indicating the further
decomposition of palladium acetate into Pd metal under the thermal conditions used. As
13

Journal of Materials Chemistry A
Page 14 of 31
View Article Online
DOI: 10.1039/C3TA12512B
previously discussed, the nearly complete decomposition of the Pd salt under the thermal
conditions was likely a major contribution to the size increase of the Pd nanoparticles.
Fig. 3 XPS Spectra of a) (Pd/MWCNT)M b) (Pd/MWCNT)T c) (Pd/SWCNT)M d) (Pd/SWCNT)T
Similarly, the XPS spectra of (Pd/SWCNT)M shows the measured binding energies
of the Pd 3d5/2 and 3d3/2 electrons at 338.2 and 343.6 eV that correspond to the presence
of Pd(II) as the major palladium species (74%) in these nanoparticles. Further
decomposition of Pd salt under thermal conditions of 300 oC led to the formation of Pd(0)
14

Page 15 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
as the dominant peaks (80%) in the XPS data of (Pd/SWCNT)T as evident from the
measured binding energies at 335.4 and 340.7 eV (Fig. 3d).
Catalytic activity of Pd/MWCNT in Suzuki cross coupling reactions
The catalytic activity of Pd/MWCNT prepared by mechanochemical and thermal
routes were investigated using the Suzuki cross coupling reaction of bromobenzene and
phenylboronic acid in a mixture of H2O:EtOH (1:1) at room temperature using various
catalyst concentrations. Fig. 4 illustrates the % conversion of the reaction in Eq. 1 for
different loadings of nanocatalysts. At 0.5 mol% loading, the (Pd/MWCNT)M
nanoparticles showed complete conversion (100%) of aryl bromide to the biphenyl
product within 5 min, at ambient temperature. Under the same concentration, the thermal
route product (Pd/MWCNT)T with the same catalyst loading demonstrated lower
reactivity and afforded only 78% conversion after 6 h at room temperature. Notably,
commercially available Pd supported on activated carbon (10% Pd/C), the most often
used catalyst in heterogeneous Pdꢀcatalyzed coupling reactions, exhibits only 75%
conversion after 5 hours at room temperature. When 0.5 mol% of (Pd/SWCNT)M was
employed in Suzuki cross coupling reaction, a 95 % conversion was obtained after 30
min at room temperature. Under the same reaction conditions, the (Pd/SWCNT)T sample
exhibited a very weak catalytic activity and provided only 70% biphenyl product within 6
h at room temperature. Thus, the (Pd/MWCNT)M catalyst system appears to be the most
reactive towards the Suzuki cross coupling reaction of bromobenzene and phenylboronic
acid at room temperature.
15

Journal of Materials Chemistry A
Page 16 of 31
View Article Online
DOI: 10.1039/C3TA12512B
Lowering the catalyst loading of (Pd/MWCNT)M to 0.05 mol%, afforded a 98%
conversion in 30 minutes. At an even lower loading of 0.008 mol%, (Pd/MWCNT)M
provided 100% conversion after 3 h at room temperature. Interestingly, at this low
concentration, (Pd/MWCNT)M is capable of converting 58% of the bromobenzene to the
biphenyl product at 80 °C under microwave heating for 2 min, and 100% formation of
product was observed after 5 min. These results demonstrate the remarkable catalytic
activity of (Pd/MWCNT)M with a turnover number (TON) of 7250 and turnover
frequency (TOF) of 217,500 hꢀ1 in microwave assisted Suzuki cross coupling reaction at
Fig. 4 Effect of catalyst concentrations on catalysis in Suzuki cross coupling reaction
Br
B(OH)2
Pd Cat.
Eq. 1
K2CO3 3 eq.
H2O:EtOH (1:1)
r.t.
100
80
60
40
20
0
(Pd/MWCNT)M: 0.5 mol%
(Pd/MWCNT)M: 0.05 mol%
(Pd/MWCNT)M: 0.008 mol%
(Pd/MWCNT)T: 0.5 mol%
(Pd/SWCNT)M: 0.5 mol%
(Pd/SWCNT)T: 0.5 mol%
10% Pd/C: 0.5 mol%
0
50
100
150
200
250
30
16

Page 17 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
80oC. The excellent catalytic activity of (Pd/MWCNT)M can be attributed in part to the
small particles size of Pd nanoparticles (1ꢀ3 nm), and uniform distribution on MWCNT
with no observed agglomeration during the mild preparation method of ballꢀmilling at
ambient temperature for 30 min.53 However, the MWCNT support system may also play
a significant role in accelerating reaction rates by activating the Pd nanoparticles on the
electron rich surface of the carbon nanotubes. It should also be noted that microwave
irradiation can have a significant impact on the efficiency of the Suzuki reactions due to
increasing the reaction rates by providing a direct, rapid and consistent energy for the
cross coupling reactions. The remarkably high TON and TOF values observed with the
(Pd/MWCNT)M catalyst may also be ascribed to the ability of the Pd particles to more
efficiently absorb microwave irradiation and thus function effectively even at very low
catalyst concentrations (e.g. 0.008 mol%) and a short reaction times.
To further investigate the mechanism of this catalytic reaction, the Suzuki cross
coupling of bromobenzene and phenyl boronic acid was performed in the presence of 0.5
mol% of the (Pd/MWCNT)M and potassium carbonate at 80oC for 10 min using a mixture
of H2O:EtOH (1:1) as solvents (Fig. 5). The reaction mixture was then hot filtered over
celite and the filtrate solution was analyzed by ICPꢀMS in which the amount of Pd
content was determined to be 120 ppb. Moreover, after the removal of the supported Pd
nanoparticles from the reaction mixture by filtration over celite, the resulting filtrate
solution was subjected to the microwave irradiation in the presence of fresh amount of
bromobenzene, phenyl boronic acid, and potassium carbonate. No further catalytic
activity was observed in this solution after heating the reaction mixture under microwave
at 80oC for 10 min. While the small amount of leached Pd (120 ppb) may not be
17

Journal of Materials Chemistry A
Page 18 of 31
View Article Online
DOI: 10.1039/C3TA12512B
consistent with the complete heterogeneity of the catalytic mechanism, the lack of
reactivity after the removal of the supported catalyst indicates the vivid interaction of the
carbon nanotube support system with Pd nanoparticles in inducing the catalytic activity.
Thus, these results combined with our recycling experiments (Table 1) provide consistent
evidence for the proposed interfacial releaseꢀredeposition mechanism, by which a small
quantity of Pd nanoparticles leach off of the support surface, catalyzes the reaction, and
reꢀdeposit onto the surface of the carbon nanotube. We are currently investigating the
surface interactions between the solvent system, the Pd particles and the support surface
in order to obtain a more fundamental understanding of the extraordinary catalytic
activity as well as the remarkable recyclability of this catalyst system.
Fig. 5 Investigation of the catalytic reaction
Effect of the ball-milling time on the catalytic performance
The duration of ballꢀmilling time for the preparation of (Pd/MWCNT)M and its affect
on Suzuki cross coupling catalytic activity was also investigated. Samples of
18

Page 19 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
(Pd/MWCNT)M were prepared at different ballꢀmill time of 10 min, 30 min and 60 min
and their catalytic activity were tested for Suzuki cross coupling reaction of
bromobenzene with phenylboronic acid at room temperature (reaction in Eq. 1).
Interestingly, using 0.5 mol% of each of these nanocatalysts in Suzuki reactions provided
the biphenyl product in a quantitative yield (100%) after 5 minutes at room temperature,
thus indicating no significant difference in catalytic activity between the three samples.
The TEM images of all of these (Pd/MWCNT)M samples demonstrated an average
particles size of 3ꢀ4 nm with uniform distributions of the Pd nanoparticles on the surface
of MWCNT (Fig. S2, Supporting Information). Alternatively, to evaluate the effect of
the ballꢀmilling technique in providing the necessary mechanical energy for facile
decomposition of the Pd salt in the presence of carbon nanotubes, samples of the catalysts
were prepared by mixing the Pd salt using a mortar and pestle at both room temperature
o
and thermal condition of 300 C. Both samples demonstrated poor catalytic activity
(10% conversion after 60 min at room temperature) in Suzuki coupling reaction described
in Eq. 1. Unlike the (Pd/MWCNT)M samples prepared by ballꢀmilling, the manually
mixed Pd/MWCNT sample did not contain any significantly active Pd nanoparticles,
which were required for the catalysis. Therefore, the ballꢀmilling operations proved not
only to be a facile and scalable methodology for catalyst preparation, but also to provide
a consistent and necessary mechanical force to produce catalytically active materials.
Recyclability of (Pd/MWCNT)M and (Pd/SWCNT)M for Suzuki cross coupling
reaction
The recyclability of (Pd/MWCNT)M was also examined using the microwave assisted
Suzuki cross coupling reaction of bromobenzene with phenylboronic acid at 80 °C for 10
19

Journal of Materials Chemistry A
Page 20 of 31
View Article Online
DOI: 10.1039/C3TA12512B
min (Eq. 1). In this case, full conversions were obtained in runs 1ꢀ4. The activity
slightly dropped in the run 5 yielding 85% conversion and further dropped in runs 6, 7 to
60% and 42 % conversions, respectively. We examined the (Pd/MWCNT)M samples by
TEM after the 6th run (Fig. 6) and found a significant amount of agglomeration of the Pd
nanoparticles. Utilizing iodobenzene as the arylhalide, the (Pd/MWCNT)M (0.5 mol%)
c
Table 1 Recycling experiments with (Pd/MWCNT)M and (Pd/SWCNT)M
Run Conversion (%)a Conversion (%)b Conversion (%)a Conversion (%)b
(Pd/MWCNT)M (Pd/MWCNT)M (Pd/SWCNT)M (Pd/SWCNT)M
1
100
100
100
100
85
60
42
ꢀ
100
100
100
100
100
100
100
100
100
70
100
100
98
78
60
45
ꢀ
100
100
100
100
100
100
100
100
62
2
3
4
5
6
7
8
ꢀ
9
ꢀ
ꢀ
10
11
ꢀ
ꢀ
48
ꢀ
52
ꢀ
ꢀ
a) bromobenzene (50 mg, 0.32 mmol boronic acid, phenylboronic acid (47 mg, 0.382 mmol, 1.2 eq.),
potassium carbonate (133 mg, 0.96 mmol, 3 eq.), and (Pd/MWCNT)M (2 mg, 1.6 ꢁmol, 0.5 mol%) or
(Pd/SWCNT)M (2.3 mg, 1.6 ꢂmol, 0.5 mol%) in 4 mL (H2O:EtOH) (1:1) was heated at 80 °C (MWI) for 10
min. b) Iodobenzene (50 mg, 0.245 mmol), phenylboronic acid (36 mg, 0.294 mmol, 1.2 eq.), potassium
carbonate (102 mg, 0.735 mmol, 3 eq.), and (Pd/MWCNT)M (1.5 mg, 1.2 ꢁmol, 0.5 mol%) or (Pd/SWCNT)M
(1.7 mg, 1.6 ꢂmol, 0.5 mol%) in 4 mL (H2O:EtOH) (1:1) was heated at 80 °C (MWI) for 10 min. c)
Conversions were determined by GCꢀMS.
successfully afforded quantitative conversion in Suzuki cross coupling reaction with
o
phenylboronic acid at 80 C microwave irradiation in 10 min for 9 consecutive recycles
20

Page 21 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
without any loss of reactivity. The activity of the catalyst dropped in runs 10 and 11 to
70 and 52% conversions, respectively. The (Pd/SWCNT)M demonstrated slightly lower
recyclability in Suzuki cross coupling reactions with iodoꢀ and bromobenzene. As
indicated in Table 1 with bromobenzene, the (Pd/SWCNT)M can be reꢀused three times
affording quantitative yields of the product in each run. The catalytic activity
subsequently diminished to 78% in run 4. Using iodobenzene, the catalyst was recycled
7 times yielding a 100% conversion in each run. The catalytic activity diminished to 62
and 48 % in runs 9 and 10, respectively.
TEM images of (Pd/MWCNT)M after the 6th run and (Pd/SWCNT)M after the 5th run
revealed significant agglomeration of Pd nanoparticles on the surface of carbon
nanotubes indicating the possible mechanism for deactivation of the catalyst due to the
decrease in the surface area and saturation of the coordination sites (Fig. 6). It should also
be noted that the diameter of the MWCNTs (~20ꢀ150 nm) is significantly greater than the
SWCNTs (~1.5 nm) which may provide the opportunity for stronger surface interactions
and a more effective anchor for the Pd particles, thus providing less opportunity for
agglomeration. The higher catalytic activity and improved recyclability of
(Pd/MWCNT)M compared to (Pd/SWCNT)M may also be attributed to the higher
concentration of Pd nanoparticles in this sample, greater degree of dispersion on the
surface of the MWCNT, and smaller particle size (1ꢀ3 nm) as is evident by TEM images
(Fig. 1a, c).
Notably, the Pd content measured by ICPꢀMS in the reaction solution of
(Pd/MWCNT)M and (Pd/SWCNT)M after the 6th and 5th recycling experiments were
determined to be 150 and 350 ppb, respectively. The smaller amount of Pd leaching in
21

Journal of Materials Chemistry A
Page 22 of 31
View Article Online
DOI: 10.1039/C3TA12512B
(Pd/MWCNT)M once again suggests a stronger surface interaction between Pd
nanoparticles with MWCNT in this sample, which increases the catalystꢀsupport
interaction, and thus contributes to the enhancement of the catalytic activities of these
nanoparticles.
Fig. 6 Significant agglomeration of a) (Pd/MWCNT)M after the 6th run b) (Pd/SWCNT)M after the
5th run
As shown from the recycling experiments, the (Pd/MWCNT)M exhibited excellent
catalytic activity and recyclability towards Suzuki cross coupling reactions. Since this
catalyst contained a significant amount of Pd(II) species as determined by XPS, the in
situ reduction of Pd(II) to Pd(0) is most likely facilitated during the cross coupling
reaction by the protic solvents under basic conditions to increase the concentration of
Pd(0) for further recycling reactions.38,40,41 Alternatively, since the carbon nanotubes are
effective microwave energy absorbers, the microwaveꢀinduced in situ
reduction/decomposition of Pd(II) to Pd(0) on the carbon nanotube surface could also
occur during the catalytic reaction,53 thus increasing the concentration of the Pd(0) in
these nanocatalysts as an active species in the Suzuki coupling reaction. Fig. 7 displays
22

Page 23 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
the XPS spectra of (Pd/MWCNT)M and (Pd/SWCNT)M prepared before and after the first
Suzuki reactions during the recycling experiments.
Fig. 7 XPS spectra of (Pd/MWCNT)M a) before the Suzuki reaction b) after the Suzuki reaction,
and (Pd/SWCNT)M c) before the Suzuki reaction d) after the Suzuki reaction
Pd(II)
Pd(0)
b)
a)
350
345
340
335
330
Binding Energy(eV)
d)
Pd(II)
Pd(0)
c)
350
345
340
335
330
Binding Energy(eV)
These results indicate that the majority of Pd(II) concentration was reduced in situ to
Pd(0) during the first reaction for both catalytic systems which is consistent with our
previous observations.40 The presence of Pd(0) in both Pd nanoparticles is confirmed by
the major binding energies of 3d5/2 and Pd 3d3/2 electrons at 335.8 and 341.1eV
correspond to Pd(0) species formed during the first Suzuki recycling reaction.
23

Journal of Materials Chemistry A
Page 24 of 31
View Article Online
DOI: 10.1039/C3TA12512B
Diversity of substrates in Suzuki cross coupling reactions using (Pd/MWCNT)M
The versatility of Suzuki cross coupling reactions was further investigated using a
broader range of functionalized aryl bromides and phenyl boronic acids. (Pd/MWCNT)M
nanoparticles were employed exclusively in these reactions due to the greater catalytic
activity established in our previous studies (Table 2). Thus, the reactions of diversely
substituted aryl bromide and phenyl boronic acids were performed in the presence of
potassium carbonate (3 eq.) and 0.5 mol% of (Pd/MWCNT)M catalyst using a mixture of
H2O:EtOH (1:1), which is an environmentally benign solvent system. The reactions were
o
either conducted at room temperature or heated at 80 C for 10 min under microwave
irradiation. As shown in Table 2, a series of aryl bromides containing various functional
groups on the phenyl rings including cyano (1a, b), aldehyde (1c), ketone (1e), ester (1h),
and nitro (1d) can effectively undergo Suzuki cross coupling reactions providing high
yields of the corresponding biphenyl products. In addition, phenyl boronic acids bearing
functional groups such as dimethylamino (1c, h), thiomethyl (1d), electron donors
methoxy and ethoxy groups (1e, f), as well as 4ꢀaminocarbonyl (1g) all worked well in
this coupling reaction scheme and afforded high yields of the corresponding Suzuki
products.
24

Page 25 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
Table 2 Diversity of Suzuki cross coupling reactions
R2
Br
R2
B(OH)2
(Pd/MWCNT)M 0.5 mol%
K2CO3 3 eq.
R1
R1
80 oC uw / 10 min
(1)
H2O:EtOH (1:1)
Boronic acid
B(OH)2
cpd
ab
Aryl-halide
1(%)c
Br
N(Me)2
NC
Me2N
95%
NC
bb
O
Br
O
O
B(OH)2
O
NC
H
92%
90%
NC
H
cb
N(Me)2
Br
B(OH)2
Me2N
O
O
SMe
d
e
Br
B(OH)2
B(OH)2
O2N
MeS
MeO
O2N
87%
Br
OMe
O
90%
92%
O
O
f
Br
Br
B(OH)2
O
O
NC
NC
NC
O
g
B(OH)2
H2N
NH2
NC
87%
25

Journal of Materials Chemistry A
Page 26 of 31
View Article Online
DOI: 10.1039/C3TA12512B
N(Me)2
h
Br
B(OH)2
O
Me2N
O
O
O
85%
Br
i
O
O
B(OH)2
NC
a) Aryl halide (0.51 mmol), Boronic acid (0.61 mmol), potassium carbonate (0.96 mmol), and Pd/G (0.3
mol%) in 4 mL (H2O:EtOH) (1:1) was heated at 80 °C (MWI) for 10 min. b) Reactions were completed at r.t.
after 30 min. c) Isolated yields.
Conclusions
In summary, a mild, rapid and solventless mechanochemical ballꢀmilling method was
developed to prepare Pd nanoparticles supported on singleꢀ and multiꢀwalled carbon
nanotubes (Pd/SWCNT and Pd/MWCNT) at ambient temperature as well as heating at
300 oC without the necessity for any reducing agent. The nanoparticles prepared by ballꢀ
milling at ambient temperature [(Pd/MWCNT)M, (Pd/SWCNT)M] demonstrated greater
catalytic activity towards the Suzuki cross coupling reactions than those prepared under
thermal conditions [(Pd/MWCNT)T, (Pd/SWCNT)T], mainly due to the smaller particle
size, uniform dispersion of Pd nanoparticles on the surface of singleꢀ and multiwalled
carbon nanotubes, and lack of agglomeration of these nanoparticles during the
mechanochemical preparation at room temperature. While both (Pd/MWCNT)M and
(Pd/SWCNT)M exhibited excellent catalytic activity in Suzuki cross coupling reaction
under ligand free ambient conditions in an environmentally benign solvent system,
(Pd/MWCNT)M showed a slightly better reactivity and recyclability with the TON of
26

Page 27 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
7250 and TOF of 217,500 hꢀ1 The superior catalytic activity and increased recyclability
.
of (Pd/MWCNT)M may be attributed to the larger diameter of the MWCNTs (~20ꢀ150
nm) which provides stronger surface interactions and a more effective anchor for the Pd
particles, thus facilitating the deposition of the greater number of Pd nanoparticles on the
surface of MWCNTs. In addition, the higher Pd content of 8.5 wt% in this sample,
smaller particles size (1ꢀ3 nm), and the more uniform distribution of the Pd nanoparticles
on the surface are also considered to be contributing factors to the overall catalyst
performance of the (Pd/MWCNT)M. This catalyst was effective over a broad range of
Suzuki cross coupling reactions with a wide variety of functionalized substrates. Studies
towards applying the same mechanochemical ballꢀmilling methods to prepare Pd
nanopaticles deposited on other solid supports such as graphene and investigating their
catalytic activities is currently underway in our laboratory.
Acknowledgements
We thank Virginia Commonwealth University and NASA Langley Research Center for
the support of this work.
27

Journal of Materials Chemistry A
Page 28 of 31
View Article Online
DOI: 10.1039/C3TA12512B
References
1 Z. Han, A. Fina, Prog. Polym. Sci., 2011, 36, 914.
2 K. Sun, M. A. Stroscio, M. Dutta, J. Appl. Phy., 2009, 105, 1
3 J. Che, T. Cagin, W. A. Goddard III, Nanotechnology, 2000, 11, 65.
4 A. L. MartinezꢀHernandez, C. VelascoꢀSantos, V. M. Castano, Nanoscience, 2010, 6,
12.
5 H. Miyagawa, M. Misra, A. K. Mohanty, J. Nanosci. & Nanotech., 2005, 5, 1593.
6 V. Popoy, Mat. Sci. & Eng. R. Reports, 2004, R43, 61.
7. C. Kramberger, T. Pichler, Adv. in Carbon nanomaterials, 2012, 131.
8 K. Liu, J. Deslippe, F. Xiao, R. B. Capaz, X. Hong, S. Aloni, A. Zettl, W. Wang, X.
Bai, S. G. Louie, E. Wang, F. Wang, Nat. Nano, 2012, 7, 325.
9 C. Biswas, Y. H. Lee, Adv. Func. Mat., 2011, 21, 3806.
10 P. Serp, M. Corrias, P. Kalck, Appl. Cat. A: General, 2003, 253, 337.
11 P. Serp, E. Castillejos, ChemCatChem., 2010, 2, 41.
12 M. Moniruzzaman, K. I. Winey, Macromolecules, 2006, 39, 5194.
13 R. H. Baughman, A. A. Zakhidov, W. A. de Heer, Science, 2002, 297, 787.
14 G. G. Wildgoose, C. E. Banks, R. G. Compton, Small, 2006, 2, 182.
15. V. Georgakilas, D. Gournis, V. Tzitzios, L. Pasquato, D. M. Guldi, M. Prato, J. Mat.
Chem., 2007, 17, 2679.
16 H. B. Pan, C. H. Yen, B. Yoon, M. Sato, C. M. Wai, Synth. Commun., 2006, 36,
3473.
17 J. A. Sullivan, K. A. Flanagan, H. Hain, Catal. Today, 2009, 145, 108.
18 D. H. Turkenburg, A. A. Antipov, M. B. Thathager, G. Rothenberg, G. B.
Sukhorukov, E. Eiser, Phys. Chem. Chem. Phys., 2005, 7, 2237.
19 S. W. Kim, M. Kim, W. Y. Lee, T. J. Hyeon, J. Am. Chem. Soc., 2002, 124, 7642
20 P. P. Zhang, X. X. Zhang, H. X. Sun, R. H. Liu, B. Wang, Y. H. Lin, Tetrahedron
Lett., 2009, 50, 4455.
28

Page 29 of 31
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C3TA12512B
21 S. L. Buchwald, Acc. Chem. Res., 2008, 41, 1439.
22 L. Yin, J. Liebscher, Chem Rev., 2007, 107, 133.
23 R. F. Heck, Acc. Chem. Res., 1979, 12, 146.
24 N. Miyaura, A. Suzuki, Chem. Rev., 1995, 95, 2457.
25 I. P. Beletskaya, A.V. Cheprakov, Chem. Rev., 2000, 100, 3009.
26 C. J. Welch, J. AlbanezeꢀWalker, W. R. Leonard, M. Biba, J. DaSilva, D. Henderson,
B. B. Laing, D. J. Mathre, S. Spencer, X. Bu, T. Wang, Org. Process Res. Dev., 2005, 9,
198.
27 C. Garrett, K. Prasad, Adv. Synth. Catal., 2004, 346, 889.
28 L. Djakovitch, K. Koehler, J. Am Chem. Soc., 2001, 123, 5990.
29 L. Djakovitch, K. Koehler, J. Mol. Catal. A: Chem., 1999, 142, 275.
30 J. Li, A. W. H. Mau, C. R. Strauss, Chem. Commun., 1997, 1275.
31 H. Erdoğan, Ö. Metin, S. Özkar, Phys. Chem. Chem. Phys., 2009, 11, 10519.
32 L. Djakoyitch, V. Dufaud, R. Zaidia, Adv. Synth. Catal., 2006, 348, 715.
33 S. El. Hankari, A. El Kadib, A. Finiels, A. Bouhaouss, J. J. E. Moreau, C. M.
Crudden, D. Brunel, P. Hesemann, Chem. Eur. J., 2011, 17, 8984.
34 Z. Shuai, J. Melanie, J. G. C. Veinot, Chem. Commun., 2010, 46, 2411.
35 E. Formo, M. S. Yavuz, E. P. Lee, L. Lane, Y. Xia, J. Mat. Chem., 2009, 19, 3878.
36 B. M. Bhanage, M. Arai, Catal. Rev., 2001, 43, 315.
37 K. Köhler, R.G. Heidenreich, S. S. Soomro, S. S. Pröckl, Adv. Synth. Catal., 2008,
350, 2930.
38 G. M. Scheuermann, L. Rumi, P. Steurer, W. Bannwarth, R. Mülhaupt, J. Am. Chem.
Soc., 2009, 131, 8262.
39 O. Metin, S. F. Ho, C. Alp, H. Can, M. N. Mankin, M. S. Gueltekin, M. Chi, S. Sun,
Nano Research, 2013, 6, 10.
40 A. R. Siamaki, A. E. R. S. Khder, V. Abdelsayed, M. S. ElꢀShall, B. F. Gupton, J.
Catalysis, 2011, 279, 1.
41 S. Moussa, A. R. Siamaki, B. F. Gupton, M. S. ElꢀShall, ACS Catal., 2012, 2, 145.
29