New environmentally friendly catalysts containing Pd-interstitial carbon made from Pd-glucose precursors for ultraselective hydrogenations in the liquid phase

Article abstract of DOI:10.1039/c1cc12681d

We report a novel preparation of a Pd nanocatalyst modified with subsurface C via blending a glucose precursor at the molecular level: the catalyst is demonstrated for the first time to be stereoselective in the hydrogenation of alkynes to cis-alkenes in the liquid phase.

Full text of DOI:10.1039/c1cc12681d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 7971–7973
www.rsc.org/chemcomm
COMMUNICATION
New environmentally friendly catalysts containing Pd–interstitial carbon
made from Pd–glucose precursors for ultraselective hydrogenations in the
liquid phasew
Chun Wong Aaron Chan,a Yaling Xie,a Nick Cailuo,a Kai Man Kerry Yu,a James Cookson,b
Peter Bishopb and Shik Chi Tsang*a
Received 6th May 2011, Accepted 2nd June 2011
DOI: 10.1039/c1cc12681d
We report a novel preparation of a Pd nanocatalyst modified
with subsurface C via blending a glucose precursor at the
molecular level: the catalyst is demonstrated for the first time
to be stereoselective in the hydrogenation of alkynes to
cis-alkenes in the liquid phase.
carbon atoms encapsulating Pd particles can also take residence
on the octahedral interstitial sites of the metal, hence blocking
the formation of b-hydride under hydrogen. We show that the
alkyne reductions are stereoselective, occurring via syn addition to
give mainly cis-alkenes by this new class of stable ‘Pd–C’
catalysts.
The aforementioned synthesis of the final Pd catalyst is
described in the experimental section (ESIw). Fig. 1 shows the
transmission electron micrographs (TEM) used to verify the
carbon encapsulation on Pd at pH 10. The Pd particles
are clearly seen inside the amorphous carbon derived from
glucose. Furthermore, they appear to be dense and form
clusters. The mean particle size was found to be 6.0 nm
(ꢀ1.9 nm, number of particles = 37) with the mode at
5 nm. Energy dispersive X-ray analysis (EDX) confirmed that
the dense clusters are rich in Pd.
Selective conversion of organic compounds is a key trans-
formation in industrial organic synthesis.1 The use of
stoichiometric chemical reagents such as LiAlH4 and NaBH4
generates a lot of inorganic wastes. This poses limitations from
both economical and environmental viewpoints regarding product
purification and waste management. The use of more efficient
protocols employing non-toxic and non-waste producing reagents,
for example, molecular hydrogen for chemical reduction is more
preferable. The employment of easily separable heterogeneous
catalytic systems must also be considered in the development of
novel environmentally aware protocols.2
Calculation of h111i peak broadening from powder X-ray
diffraction (PXRD) confirms with the size obtained from
TEM. Analysis of higher angle peaks for samples treated at
400 1C in N2 appears to shift to lower values compared to pure
Pd of comparable size as well as our as-synthesised sample
without heat treatment (Table 1). This is consistent with
previous reports by Zlemecki and Jones that the lattice parameter,
a0, of bulk Pd with carbon occupying interstitial octahedral
sites has increased to 0.3995 nm.5 This represents carbon modified
palladium with the formula Pd0.85C0.15 (independent of size) where
carbon saturation is reached at 15%. However, when treated with
Catalytic hydrogenation of alkynes to alkenes represents
one of the most important steps in chemical manufacture. It is
established that solid catalysts containing Pd in the presence
of hydrogen are suitable candidates, affording high alkene
selectivity in the gas phase.3 Recent work by Teschner et al.
demonstrated that the formation of subsurface C in Pd during
catalysis greatly promotes alkene selectivity.4 However,
interstitial subsurface carbon atoms were only formed under
transient conditions at elevated temperatures under an
atmosphere of reactive gas.5 We sought to develop a stable
Pd analogue, modified with subsurface carbon atoms for
liquid phase hydrogenations, where the conversion of alkynes
to alkenes is one of the key steps in fine chemicals catalysis.
Here we show that glucose can act as a reducing and capping
agent as well as the supporting matrix to stabilise small
Pd nanoparticles. Upon heat treatment, the glucose derived
a Wolfson Catalysis Centre, Department of Chemistry,
University of Oxford, Oxford, OX1 3QR, UK.
E-mail: edman.tsang@chem.ox.ac.uk
b Johnson Matthey Technology Centre, Blount’s Court Road,
Sonning Common, Reading, RG4 7NH, UK
w Electronic supplementary information (ESI) available: Synthesis of
glucose encapsulated Pd, effect of pH, FTIR, TGA, PXRD, ICP,
catalyst testing and reaction modelling. See DOI: 10.1039/c1cc12681d
Fig. 1 TEM of (a) glucose derived carbon embedding 5–6 nm Pd
nanoparticles after treatment at 400 1C, scale bar 100 nm (left),
(b) HRTEM of the same sample (right), scale bar 5 nm.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 7971–7973 7971

View Article Online
Table 1 Lattice parameter of Pd nanoparticles
Pd glucose catalyst
Lattice a0
Approx. C atom %
As-synthesised
Treatment at 400 1C N2
Treatment at 400 1C 5% H2/N2
0.3923(4)
0.3994(3)
0.3892(3)
4.7
15.1
0.3
H2 at 400 1C the interstitial carbon atoms are more unstable than
other surface carbons, which are then expelled from the metal
lattice, also consistent with previous observations.6
Carbon diffusion into a metal lattice has been previously
carried out by passing hydrocarbons (acetylene, ethylene) or
CO over supported or unsupported metal particles at elevated
temperatures in the gas phase. This process involves the
deposition of a carbonaceous overlayer, followed by an activated
diffusion of carbon atoms through the metal lattice.7 A major
drawback of this strategy is that the Pd–C phase only exists
under the transient reaction conditions4 and it is challenging to
control the degree of carbon saturation (rate of carbon diffusion is
size dependent) while maintaining the particle size without
sintering at high temperature. In addition, coke or graphitic
carbon can be formed on the metal surface which is difficult to
remove thus reducing the rate of carbon diffusion and usefulness
of the material for catalysis.8 The fundamental difference of
our synthesis is the blending of precursors at the molecular
level before Pd–C is gradually reduced and assembled, thus
facilitating the saturation of interstitial carbons. Fig. 2 shows
the temperature programmed reduction (TPR) of a sample
after treatment with and without interstitial carbon atoms.
The evolution of H2 (negative peak) at 70 1C represents the
decomposition of b-hydride species from the Pd lattice where
there are no interstitial carbon atoms. The carbon atoms
clearly inhibit the formation of b-hydride species by competing
for the same octahedral interstices.5
Fig. 3 H2 uptakes of a glucose carbonised Pd catalyst modified with
interstitial carbon (black) and commercial 5% Pd/C (red).
uptakes of the two catalysts for 3-hexyn-1-ol. The glucose
derived catalyst was markedly different from Pd/C, which
maintained an impressive selectivity to alkene (91.9% to
3-hexen-1-ol). Only one equivalent of H2 was taken up even
after prolonged exposure to excess H2 with complete alkyne
conversion while two equivalents of H2 were used over Pd/C
which led to 100% alkane formation (Table 2). The new
catalyst also showed high stability with no apparent deactivation
or loss in selectivity upon repeated testing.
This result suggests that the adsorption strength of alkene
on the metal surface over this new type of catalysts is relatively
attenuated, thus allowing for high alkene selectivity in the
liquid phase. Hydrogenation of the more polar 2-butyn-1,4-
diol in a polar solvent can reduce the corresponding polar
alkene diol adsorption from the surface as shown in the case of
Pd/C. Nevertheless, this was still evident in enhancing the
alkene diol selectivity from 72.5 to 97.2% (Table 2) using the
glucose derived catalyst without a significant drop in alkene
selectivity as the conversion increased along with the reaction
time (ESIw). Stereoselective syn addition to unsaturated multiple
bonds is well known to occur on the metal surface. However,
excessive adsorption of alkenes or intermediates in the liquid
phase can allow time for their isomerisation on surface to give
trans-alkenes. It is noted from Table 2 that the reduction in
alkene adsorption from alkyne hydrogenation over the glucose
derived catalyst diminishes bond rearrangements, thus giving
high selectivity to cis-alkenes.
Stereoselective hydrogenations of 3-hexyn-1-ol and 2-butyn-
1,4-diol were tested at Johnson Matthey Laboratory on the
glucose carbonised catalyst and compared to a commercial
Pd/C to explore their activity in low temperature, liquid phase
conditions. The former catalyst was pretreated with a mixture
of H2O2/NH4OH as a mild oxidation step, as it was found
necessary to partially remove the amorphous carbon on the
surface to increase the rate of reaction. Fig. 3 shows typical H2
Investigation of which carbon atom (surface, sub-surface
interstitial or both) causes the attenuation of alkene adsorption
was conducted. Stereoselective hydrogenation of 4-octyne was
tested on the glucose carbonised catalysts with (CSurface) and
without H2 pretreatment (CSurface + CSubsurface) and compared
to commercial JM 5% Pd/C (Type 39). Kinetic reaction
profiles were fitted based on the Langmuir–Hinshelwood
kinetic model. The model assumes equilibrium establishment
of hydrocarbons and weak H2 adsorption.9 Kinetic fittings to
the experimental data were excellent with good reproducibility,
reinforcing the validity of the model (Fig. 4 and ESIw). As seen
from Fig. 4, there is strong adsorption of 4-octyne over alkene
species (supported by the K constants derived from the model),
resulting in almost exclusively 4-octene production until
the 4-octyne is completely consumed in both cases when
Pd contains CSurface or CSurface + CSubsurface. This is due to
lower adsorption energy of the weaker adsorbate, in this case
Fig. 2 TPR shows decomposition of the b-Pd–H phase on the
catalyst with the interstitial carbon selectively removed (blue) and
the blockage of b-hydride by the catalyst with the interstitial carbon
(red).
c
7972 Chem. Commun., 2011, 47, 7971–7973
This journal is The Royal Society of Chemistry 2011

View Article Online
Table 2 3-Hexyn-1-ol and 2 butyn-1,4-diol hydrogenations
Substrate
Catalyst
Conv./%
Sel/% alkene
cis/trans
3-Hexyn-1-ola
5% Pd/C (Johnson Matthey)
PdGluC400[O]H2O2/NH4OHb
5% Pd/C (Johnson Matthey)
100
100
19.1
35.4
0
0
3-Hexyn-1-ola
91.9
72.5
97.2
10.8
5.83
18.45
2-Butyn-1,4-diolc
2-Butyn-1,4-diolc
d
PdGluC800[O]HNO3
a
100 mg catalyst, 50 mL, 0.5 M of 3-hexyn-1-ol in IPA, 3 bar H2, 30 1C with stirring at 450 rpm, 20 h, 1,4-dioxane as standard, cis and trans-
b
c
hexen-1-ol as alkene and 1-hexanol as alkane product. Catalyst pretreated with an oxidant to increase activity. 0.3 mg 5% Pd/C or
2 mg PdGluC800[O]HNO3, 5 mL, 0.46 M of 2-butyn-1,4-diol in IPA, 10 bar H2, 30 1C with stirring at 450 rpm, 1 h. cis and trans 2-butene-
1,4-diol as alkene and butane-1,4-diol as alkane product with traces of 1-butanol, butyraldehyde, E,Z-butene-1,4-diol, crotyl alcohol and
d
2-hydroxytetrahydrofuran. Catalyst pretreated with acid to increase activity.
Table 3 Kinetic analysis of two catalysts for rate ratios
PdGlu400[O]a PdGlu400[O]H2
JM Pd/C
a
Rate ratio
k cis-4-Octene/
k Full
7.53(ꢀ0.49)
2.66(ꢀ0.18)
3.82(ꢀ0.53)
0.79(ꢀ0.11)
0.43(ꢀ0.14)
P
hydrogenations
k cis-4-Octene/
k Isomerisations
0.25(ꢀ0.08)
P
a
Catalyst was pretreated with a mixture of H2O2/NH4OH to increase
the rate of reaction.
Fig. 4 Hydrogenation of 4-octyne in toluene over the glucose
carbonised catalyst modified with interstitial carbon pretreated with
H2O2/NH4OH (left) and Pd/C (right). Conditions: 40 mg (left) and
10 mg (right) catalyst, 50 mL, 0.188 M 4-octyne in toluene, 1 bar H2,
50 1C, n-heptane as standard (blue: 4-octyne; green: cis-4-octene, red:
octane, brown: trans-4-octene, dark: 3-octene). Experimental data
points with best fitted lines according to the model (ESIw) are given.
carbon and hydrogen atoms in Pd for alkyne hydrogenation
in the gas phase, by a new high pressure XPS technique.4 As
reported, the diffusion of C and H atoms from model
reactions with the Pd lattice is dynamic, substrate dependent
and transient. In contrast, at low temperatures in the liquid
phase, the combination of kinetically stable trapped carbon
atoms from our preparation, the solvent cage effect, stronger
molecular adsorption, and molecule re-adsorption render the
‘carbon effects’ more significant than gas phase reactions.
Here, we demonstrate that interstitial carbon atoms in the
Pd subsurface created by our green synthesis route using
glucose are indeed effective to catalyse the hydrogenation of
alkynes to alkenes in the liquid phase. Nevertheless, this new
class of catalysts could open up rational ways to tune adsorption
with subsurface interstitials for ultraselective catalysis.
an alkene over an alkyne (see K values) such that the alkyne
molecule forces the intermediate alkene molecule to desorb
from the metal surface by taking up the adsorption site. The
catalyst surface will favour re-adsorption of 4-octene after
near complete alkyne consumption, only then will it catalyse
isomerisation and over-hydrogenation (ESIw). On comparing
the slopes of cis/trans isomerisation and over-hydrogenation
between the two samples, the influence of subsurface carbon
atoms was clearly evident in reducing these reaction rates,
despite the fact that they have no contact with substrate
molecules. Table 3 shows resolved rate ratios of cis-4-octene/
full hydrogenations and cis-4-octene/isomerisations as
compared to commercial Pd/C. It was found to be about
17.5 and 10.6 times less in full hydrogenations and isomerisations
for the sample with CSurface + CSubsurface cf. 5% Pd/C,
respectively. Selective removal of subsurface carbon atoms
by H2 giving only CSurface gave 8.9 and 3.2 times in reduction.
Thus, apart from surface carbon effects, this study shows
that the interstitial carbon atoms in the subsurface of Pd play a
significant role in reducing the rates of alkene intermediates
from over-hydrogenation and isomerisation. They could be
related to the inhibition of the b-Pd–H phase, which is known
to alter the hydrogen atom equilibria. In addition, there is also
electronic modification of Pd by the reduction of alkene
adsorption energy. This is achieved by hybridisation of the
Pd d-band with C s–p bands.10,11 It is noted that despite the
high ratios and atomic blending of carbon to Pd atoms,
the octahedral occupancy is still limited to 15%, implying that
the filling is driven by hybridisation.5
Notes and references
1 Encyclopedia of Catalysis, ed. I. T. Horvath, VCH, 2003, vol. 6,
p. 1.
2 S. C. Tsang, V. Caps, I. Paraskevas, D. Chadwick and
D. Thompsett, Angew. Chem., Int. Ed., 2004, 43, 5645.
3 A. Drelinkiewicz, W. Stanuch, A. Knapik, A. Ghanem,
R. Kosydar, A. Bukowska and W. Bukowski, J. Mol. Catal. A:
Chem., 2009, 300, 8.
4 D. Teschner, J. Borsodi, A. Wootsch, Z. Revay, M. Havecker,
A. Knop-Gericke, S. D. Jackson and R. Schlogl, Science, 2008,
320, 86.
5 S. B. Zlemecki and G. A. Jones, J. Catal., 1985, 95, 621.
6 N. Krishnankutty and M. A. Vannice, J. Catal., 1995, 155,
312.
7 S. B. Ziemecki, G. A. Jones, D. G. Swartzfager, R. L. Harlow and
J. Faber, J. Am. Chem. Soc., 1985, 107, 4547.
8 G. C. Bond, Appl. Catal., A, 1997, 149, 3.
9 J. Wood, L. Bodenes, J. Bennett, K. Deplanche and
L. E. Macaskie, Ind. Eng. Chem. Res., 2009, 49, 980.
10 C. D. Gelatt, A. R. Williams and V. L. Moruzzi, Phys. Rev. B,
1983, 27, 2005.
11 J. Haglund, A. F. Guillermet, G. Grimvall and M. Korling, Phys.
Rev. B: Condens. Matter, 1993, 48, 11685.
Our results are enforced by Teschner et al. who recently
made a direct observation on the influence of subsurface
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 7971–7973 7973