Selective formation of lactate by oxidation of 1,2-propanediol using gold palladium alloy supported nanocrystals

Article abstract of DOI:10.1039/b823285g

The use of bio-renewable resources, such as glycerol, a by-product from bio-diesel manufacture, can provide a viable way to make valuable products using greener technology. In particular, glycerol can be reduced to give 1,2-propanediol that can then be se

Full text of DOI:10.1039/b823285g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/greenchem | Green Chemistry
Selective formation of lactate by oxidation of 1,2-propanediol using gold
palladium alloy supported nanocrystals
Nikolaos Dimitratos, Jose Antonio Lopez-Sanchez, Sankar Meenakshisundaram, Jinto Manjaly
Anthonykutty, Gemma Brett, Albert F. Carley, Stuart H. Taylor, David W. Knight and Graham J. Hutchings*
Received 24th December 2008, Accepted 29th April 2009
First published as an Advance Article on the web 28th May 2009
DOI: 10.1039/b823285g
The use of bio-renewable resources, such as glycerol, a by-product from bio-diesel manufacture,
can provide a viable way to make valuable products using greener technology. In particular,
glycerol can be reduced to give 1,2-propanediol that can then be selectively oxidised to lactate,
which has immense potential as a monomer for the synthesis of biodegradable polymers. We show
that gold-palladium alloy catalysts can be very effective for the selective oxidation of
1,2-propanediol to lactate. Two supports, TiO2 and carbon, and two preparation methods, wet
impregnation and sol-immobilisation, are contrasted. The addition of palladium to gold
significantly enhances the activity and retains the high selectivity to lactate using O2 as oxidant
(we observe 96% lactate selectivity at 94% conversion). Use of hydrogen peroxide is also possible
but lower activities are observed as a result of the reaction conditions, but in this case no marked
enhancement is observed on addition of palladium to gold. Comparison of the activity for C3
alcohols shows that the reactivity decreases in the order: glycerol > 1,2-propanediol >
1,3-propanediol ~ 1-propanol > 2-propanol. The use of a sol-immobilisation preparation method
as compared to impregnation leads to alloy catalysts with the highest activity for lactate formation
from the oxidation of 1,2-propanediol; the origins of these activity trends are discussed.
rate of production is very low and only dilute solutions can
Introduction
be produced as the enzyme is inhibited by the product. There
Lactic acid has the potential to become a major chemical for the
is also a potential tension between carbohydrate utilisation as
production of biodegradable polymers. Until recently, lactic acid
foodstuffs or plastics, although non-foodstuff carbohydrates can
had relatively minor uses in a range of fine chemical applications.
be used in the fermentation process. Hence the ‘non-green’
For example, in cosmetics it can be used as a moisturizer and is
chemical routes can produce lactic acid at high rates, whereas the
used in the textile industry as a mordant and also finds appli-
green fermentation route is hampered by the low productivity.
cations in tanning leather as well as in the manufacture of inks.
There is therefore a need to identify a new green production
It has uses in the foods industry, such as in yoghurt and cheese
strategy for the production of lactate and this is the topic that
production. However, interest in lactic acid has recently been
this paper addresses as selective oxidation of 1,2-propanediol
enhanced by the realisation that polylacetate is a biodegradable
presents a viable alternative pathway from a substrate that can
plastic, which has spurred the potential to transform lactic acid
be derived from a bio-renewable feedstock.
from being solely a fine chemical to being a commodity material,
Recently, glycerol has been identified as a useful starting
with annual production of >100 kt/annum.1
material for chemical synthesis.3 Glycerol is a bio-renewable
Lactic acid can be produced by number of chemical routes, e.g.
by the reaction of acetaldehyde with HCN followed by hydrolysis
with sulfuric acid, or by the reaction of CO with formadehyde
in water at high pressures with HF as catalyst. As noted by
Prati and Rossi,2 these chemical processes are non-green. For
this reason lactic acid is mainly produced by a fermentation
process in which a range of carbohydrates are transformed, in
the presence of calcium hydroxide, into calcium lactate that is
readily recovered from the fermentation broth.1 Lactic acid is
then produced by reaction with an acid. The main problem with
the fermentation route is the high dilutions required and the
feedstock that is a by-product of the manufacture of bio-diesel,4
and since there is an increasing drive towards greener fuels,
the supply of glycerol as a sustainable raw material is growing
steadily. Glycerol is an ideal substrate for the production of 1,2-
propanediol by dehydration/hydrogenation,5 and this can be
linked to biodiesel production as 1,2-propanediol is used as an
antifreeze. However, 1,2-propanediol is an ideal starting point
for the synthesis of lactate using selective oxidation. T.Tsujino
et al.6 reacted aqueous 1,2-propanediol at 90 ◦C at pH 8 using
Pd/C modified by Pb, Bi and/or Te as additives. They observed
that both the primary and the secondary hydroxyl groups were
reactive with their Pd/C catalysts and in addition to lactic acid,
hydroxyacetone and pyruvic acid were also produced. Pyruvate
was shown to be produced by sequential oxidation of lactate.
Pinxt et al.7 studied 1,2-propanediol oxidation using Pt/C mod-
ified by Pb, Bi and Sn and showed that pyruvate was produced
Cardiff Catalysis Institute, School of Chemistry, Cardiff University,
Main Building, Park Place, Cardiff, UK CF10 3AT.
E-mail: hutch@cardiff.ac.uk; Fax: +44 29 2087 4059; Tel: +44 29 2087
4059
This journal is
The Royal Society of Chemistry 2009
Green Chem., 2009, 11, 1209–1216 | 1209
©

View Article Online
in two parallel pathways from either lactate or hydroxyacetone.
Under the conditions used by the authors, it is clear that with
Pd or Pt catalysts only low selectivities of lactate are observed.
Supported gold catalysts have been shown to be very selective
for a range of oxidation reactions.8–11 In particular, they are effec-
tive for low temperature oxidation of CO,12 even in the presence
of H2, H2O and CO2 a reaction that is finding some applications
in fuel cells.13,14 Gold has also been found to be effective for the se-
lective oxidation of alkenes15,16 and alcohols,17 and for the selec-
tive hydrogenation of unsaturated carbonyl compounds and ni-
tro groups.18 Rossi and Prati were the first to show that supported
gold nanoparticles were effective for alcohol oxidation in the
presence of base.2,19,20 Gold was found to be very selective for the
oxidation of the primary alcohol group in diols and triols.2,33,34
We have shown that, although gold alone is a very effective cata-
lyst for the selective oxidation of an alcohol to an aldehyde under
solvent-free conditions,17 the alloying of gold with palladium
leads to a twenty five-fold enhancement in activity with simulta-
neous retention of selectivity.21 We have extended this work and
have demonstrated that the combination of gold and palladium
provides a pronounced synergistic effect in redox catalysis.22–28
Furthermore, using a sol immobilisation preparation method
affords even more reactive catalysts.29,30 In this paper we
investigate the synergistic effect of adding gold and palladium
for the oxidation of 1,2-propanediol and demonstrate that very
high selectivities to lactate can be obtained at high conversions.
brown sol. After 30 min of sol generation, the colloid was
immobilised by adding activated carbon (acidified to pH 1 by
sulfuric acid) under vigorous stirring conditions. The amount of
support material required was calculated so as to have a total
final metal loading of 1% wt. After 2 h the slurry was filtered,
the catalyst washed thoroughly with distilled water and dried
at 120 ◦C overnight. Further catalysts were prepared using a
1:1 molar ratio of Au:Pd denoted as Au-Pd/TiO2SIm and Au-
Pd/CSIm. Monometallic catalysts containing gold or palladium
were also prepared using a similar methodology, and these are
denoted Au/TiO2SI, Pd/TiO2SI, Au/CSI and Pd/CSI respectively.
Catalyst testing
Diol and alcohol oxidation
Autoclave reactor studies. Reactions were performed with a
range of substrates: glycerol, 1,2-propanediol, 1,3-propanediol,
1-propanol and 2-propanol. Catalytic reactions were carried
using a 50-mL Parr autoclave. The alcohol solution (0.6 M
and NaOH/alcohol ratio = 2, mol/mol, 20 ml) was added into
the reactor and the desired amount of catalyst (alcohol/metal
ratio = 500–2000, mol/mol) was suspended in the solution. The
autoclave wa◦s pressurised with oxygen (10 bar pressure) and
heated at 60 C. The reaction mixture was stirred at 1500 rpm
for 4 h. The reactor vessel was cooled to room temperature and
the reaction mixture was analysed by HPLC.
Glass reactor studies. Experiments were also conducted
with hydrogen peroxide as oxidant at atmospheric pressure.
In this case the alcohol solution (0.6 M and NaOH/alcohol
molar ratio = 2, 20 ml) was placed in a round bottom flask
(100 ml) and the desired amount of catalyst (alcohol/metal
molar ratio = 2000) was suspended in the solution. Hydrogen
peroxide was added (substrate/H2O2 molar ratio = 4) and the
mixture was reacted at 60 ◦C, in air atmospheric pressure for 3h.
Experimental
Catalyst preparation
The following notation is used for the catalyst samples we have
prepared: I denotes impregnation, SI denotes sol-immobilis-
ation, w denotes Au and Pd are present in a 1:1 ratio by weight,
and m denotes Au and Pd are combined in a 1:1 molar ratio.
Product analysis. Analysis was carried out using HPLC
with ultraviolet and refractive index detectors. Reactants and
products were separated using a Metacarb 67H column. The
eluent was an aqueous solution of H3PO4 (0.01M) and the flow
was 0.45 ml/min. Samples of the reaction mixture (0.5 ml)
were diluted (5 ml) using the eluent. Products were identified
by comparison with authentic samples. For the quantification
of the amounts of reactants consumed and products generated,
an external calibration method was used and the calibration
factor for each reactant/product was calculated. The substrate
conversion and the selectivity to the reaction products were
calculated on carbon basis. Furthermore the carbon mass
balance was closed for the reactions reported within 5%.
Impregnation method. Pd-only, Au-only and Au-Pd bimetal-
lic catalysts were prepared by impregnation of carbon (Darco
G60, Aldrich) and TiO2 (Degussa, P25), via an impregnation
method using aqueous solutions of PdCl2 (Johnson Matthey)
and/or HAuCl4.3H2O (Johnson Matthey). Materials contain-
ing a total of 0.5, 1 and 5 wt% total metal were prepared
and the Au and Pd were present in equal weights. For the
0.5%Au-0.5%Pd/carbon catalyst, the detailed preparation pro-
cedure employed is described below. An aqueous solution of
HAuCl4·3H2O [2 ml, 5 g dissolved in water (250 ml)] and an
aqueous solution of PdCl2 [0.83 ml, 1 g in water (25 ml)] were
simultaneously added to carbon (3.8 g). The paste formed was
◦
ground and dried at 110 C for 16 h and calcined in static air,
typically at 400 ◦C for 3 h.
Results and discussion
Sol-immobilisation method. Au, Pd and Au-Pd catalysts
supported on carbon or TiO2 were also prepared using a sol-
immobilisation method. An aqueous solution of PdCl2 (Johnson
Matthey) and HAuCl4·3H2O of the desired concentration was
prepared. Polyvinyl alcohol (PVA) (1 wt% solution, Aldrich,
weight average molecular weight MW = 9,000–10,000 g/mol,
80% hydrolysed) was added (PVA/Au (wt/wt) = 1.2); a
0.1 M freshly prepared solution of NaBH4 (>96%, Aldrich,
NaBH4/Au (mol/mol) = 5) was then added to form a dark-
For all the experiments we have selected catalysts that we have
been characterised extensively in our previous studies using
a combination of TEM, STEM-XEDS and XPS.21,27,30,31 We
aimed to contrast the reactivity of supported Au, Pd and Au-
Pd nanocrystals made by impregnation and sol immobilisation
for the selective oxidation of 1,2-propanediol. In addition we
have selected two supports for evaluation (activated carbon and
TiO2). For the materials prepared by impregnation, the metal
1210 | Green Chem., 2009, 11, 1209–1216
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 1 Effect of the metal composition and nature of support on the liquid phase oxidation of 1,2-propanediol with O2 using Au and Pd catalysts
synthesised by impregnationa
Selectivity (%)
Lactate
Catalyst
Conv. (%)
Acetate
Formate
TOF (h-1) b
1%Au/CI
1%Pd/CI
2.5%Au-2.5%Pd/CIw
1%Au/TiO2I
1%Pd/TiO2I
5%Au/TiO2I
5%Pd/TiO2I
2.5%Au-2.5%Pd/TiO2Iw
16
11
32
15
6
27
3
73.7
77.9
95.4
80.9
99.7
91.5
92.0
96.3
23.2
20.7
4.0
17.7
0.0
6.0
7.5
3.1
1.4
0.7
1.4
0.3
1.6
0.5
0.1
20
13
40
18
7
34
4
91
3.6
114
a Reaction conditions: water (20 ml), 0.6 M 1,2-propanediol, 1,2-propanediol/total metal molar ratio = 500, NaOH/1,2-propanediol molar ratio = 2,
T = 60 ◦C, pO2 = 10 bar, stirring rate 1500 rpm, reaction time = 4 h. b TOF (h-1) at 4 hours of reaction. TOF numbers were calculated on the basis
of total loading of metals.
particle sizes ranged between 2 and 14 nm, with the average size
being around 6 nm.30 Occasional larger particles in excess of
25 nm were also present in this material, although these were rel-
atively rare. By comparison, the metal particle size distribution in
catalysts prepared by sol-immobilisation was found to be much
narrower, i.e. between 4–7 nm.30,35 For the impregnated materials
the Au-Pd supported alloys have been characterised extensively
by STEM-XEDS21,27,30 and have been shown to have contrast-
ing morphologies. The TiO2-supported material, in common
with other oxide-supported Au-Pd catalysts prepared by wet
impregnation, have a core-shell structure with a palladium-rich
shell and a gold-rich core, whereas the carbon-supported Au-Pd
catalyst is a homogeneous alloy. We have recently found that the
supported Au-Pd catalysts prepared using sol-immobilisation
are also homogeneous alloys by using STEM-HAADF.35 XPS
analysis showed that the samples prepared by impregnation
are almost entirely composed of Pd2+ species, whereas those
prepared by sol-immobilisation contain both Pd2+ and Pd0 With
the latter predominating.35 We consider that these two sets of
well-characterised materials provide a useful basis for this study
of the selective oxidation of 1,2-propanediol.
The effect of loading amount of the metals was also in-
vestigated for the TiO2-supported catalysts (Table 1). Using
lower loadings of Au led to a lower conversion and selectivity
for lactate, and we have previously observed a similar effect
with glycerol oxidation.32 In contrast the Pd catalyst gave a
slightly higher conversion with an enhanced selectivity. These
observations suggest that parallel pathways are operating for
these products on these catalysts, and that specific sites may be
separately responsible for lactate and acetate formation.
When Au and Pd are combined using the impregnation
method, a significant enhancement in catalyst performance is
observed and, in particular, the catalyst activity is improved
whilst simultaneously increasing the selectivity for lactate, the
desired product (Table 1). The effect is most pronounced for
Au-Pd/TiO2Iw where the catalyst activity is increased by a factor
of ca. 3 when compared with the Au-only and by a factor of
ca. 30 with the Pd-only catalysts. With Au-Pd/TiO2Iw, after
a reaction time of 4 h, 91% conversion was achieved with a
selectivity of 96.3%.
The effect of the reaction time is shown for the 2.5%Au-
2.5%Pd/TiO2Iw and 2.5%Au-2.5%Pd/C2Iw materials in Fig. 1.
Comparison of Au, Pd and Au-Pd catalysts prepared by
impregnation
We initially investigated the catalyst activity of mono-metallic
TiO2-supported catalysts prepared by wet impregnation com-
prising 5 wt% Au or Pd (Table 1) using standard reaction
conditions in the presence of base. The main product was
lactate and some minor amounts of acetate and formate were
also observed. The Au/TiO2 catalyst prepared by impregnation
was considerably more active than the Pd/TiO2 material. The
selectivities were similar.
Au was markedly more reactive than Pd. For the carbon-
supported catalysts, and again the selectivities were very similar,
the main product being lactate. However, the carbon-supported
catalysts prepared by impregnation were much less selective than
the TiO2-supported catalysts and in this case high quantities of
acetate were observed. In addition, lower amounts of formate
were also formed. The analysis was carried out using HPLC and
the carbon mass balance was 100% and no CO2 was observed.
Fig.
1
Selective oxidation of 1,2-propanediol,
ꢀ
2.5wt%Au-
2.5wt%Pd/TiO2Iw; ꢁ 2.5wt%Au-2.5wt%Pd/CIw. Reaction conditions:
water (20 ml), 0.6 M 1,2-propanediol, 1,2-propanediol/total metal molar
ratio = 500, NaOH/glycerol molar ratio = 2, T = 60 ◦C, pO2 = 10 bar,
stirring rate 1500 rpm.
This journal is
The Royal Society of Chemistry 2009
Green Chem., 2009, 11, 1209–1216 | 1211
©

View Article Online
Table 2 Catalytic activity of Au, Pd and Au-Pd supported catalysts
It is clear that TiO2 is the more effective support for Au-Pd
alloys. This contrasts with the observations for the monometallic
catalysts, although these can only be directly compared at lower
loadings. The effect of time-on-line on selectivity is shown in
Figs. 2 and 3. For the carbon-supported bimetallic catalyst
(Fig. 2), the selectivity to lactate remains steady with reaction
time (Fig. 2). For the TiO2-supported catalyst (Fig. 3), the
selectivity to lactate remains constant within experimental error,
although it may be slightly lower at the very start of the reaction.
These are key observations and indicate that the acetate and
formate are formed in a parallel reaction pathway to lactate
with these bimetallic catalysts. No pyruvate was observed at
any reaction times with the bimetallic catalysts, indicating that
prepared by sol-immobilisation in liquid phase oxidation of 1,2-
a
propanediol with O2
Selectivity (%)
Catalyst
Conv. (%) Lactate Acetate Formate TOF (h-1) b
0.5%Au/TiO2SI 58
0.5%Pd/TiO2SI
1%AuPd/TiO2SIm 94
96.0
98.5
95.9
92.0
97.3
99.3
4.0
1.5
4.1
7.2
1.1
0.3
0
0
0
0.8
1.6
0.5
1156
135
1880
619
60
1066
7
1%Au/CSI
1%Pd/CSI
31
3
1%AuPd/CSIm
53
a Reaction conditions: water (20 ml), 0.6 M 1,2-propanediol, 1,2-
propanediol/total metal molar ratio = 2000, NaOH/1,2-propanediol
molar ratio = 2, T = 60 ◦C, pO2 = 10 bar, stirring rate 1500 rpm,
reaction time = 1 h. b TOF (h-1) at 1 hour of reaction. TOF numbers
were calculated on the basis of total loading of metals.
lactate is very stable over these catalysts, a crucial observation
for the design of a green synthetic process.
Comparison of Au, Pd and Au-Pd catalysts prepared by
sol-immobilisation
A series of TiO2- and carbon-supported catalysts were prepared
using the sol-immobilisation technique and their activity for
1,2-propanediol oxidation under standard reaction conditions
was contrasted with the catalyst prepared by wet impregnation
(Tables 1 and 2). In the case of sol-immobilisation, much lower
concentrations of the metals were used, since the method is
restricted by the concentrations of the sols that can be prepared.
Hence, catalysts containing 1 wt% of total metal have been
prepared and evaluated. The catalysts prepared by the sol-
immobilisation method are significantly more active for 1,2-
propanediol oxidation than the equivalent sample prepared
by wet impregnation. This effect is apparent for both the
monometallic and the bimetallic materials. In general, a sig-
nificant enhancement in activity by a factor of ca. 10–20 as
determined on a mol basis is observed for all the catalysts.
However, and most importantly, the selectivity to lactate is also
enhanced. For 1%AuPd/TiO2SIm after 1 h reaction a conversion
of 94% was observed with a lactate selectivity of 95.9% and with
1%AuPd/CSIm 53% conversion was observed with a selectivity
of 99.3%. In both case only minor amounts of by-products were
observed with the sol-immobilised catalysts.
Fig. 2 Products formed in the oxidation of 1,2-propanediol using
2.5wt%Au-2.5wt%Pd/CIw. Reaction conditions as figure 1. Key: ꢀ
Lactate selectivity (%), ꢁ Acetate selectivity (%), ꢀ Formate selectivity
(%).
Comparison of H2O2 and O2 as oxidants
In a subsequent set of experiments, we have set out to contrast
the catalyst performance using H2O2 as oxidant in place of
molecular oxygen. H2O2 is considered to be a green oxidant,
since the only by-product is water, and is often used as a more
reactive form of oxygen; the results are shown in Table 3. For
the monometallic catalysts prepared using impregnation, it is
apparent that much lower selectivities to lactate are observed
and higher amounts of acetate are apparent. However, this
is not observed with the bimetallic AuPd catalysts prepared
by impregnation, but in this case no synergistic enhancement
in activity is observed. These results tend to suggest that the
hydroperoxy species is not responsible for the selective formation
of lactate, since this species would be expected to be formed and
Fig. 3 Products formed in the oxidation of 1,2-propanediol using
2.5wt%Au-2.5wt%Pd/TiO2Iw. Reaction conditions as figure 1. Key: ꢀ
Lactate selectivity (%), ꢁ Acetate selectivity (%), ꢀ Formate selectivity
(%).
1212 | Green Chem., 2009, 11, 1209–1216
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 3 Liquid phase oxidation of 1,2-propanediol carried out at atmospheric pressure in the presence of hydrogen peroxide using Au and Pd
supported catalystsa
Selectivity (%)
Lactate
Catalyst
Conv. (%)
Acetate
Formate
TOF (h-1) b
5%Au/TiO2I
5%Pd/TiO2I
2.5%Au-2.5%Pd/CIw
2.5%Au-2.5%Pd/TiO2Iw
1%Au/TiO2SI
1%Pd/TiO2SI
1%AuPd/TiO2SIm
1%(Au-Pd)/CSIm
6
2
3
0.4
9
8
82.6
72.0
100.0
100.0
96.7
97.8
96.1
14.5
22.0
0.0
0.0
1.8
1.2
2.5
8.7
2.9
6.0
0
40
14
21
0
3
1.6
1.0
1.4
0.8
62c
54c
66c
151c
10
23
90.6
a Reaction conditions: water (20 ml), 0.6 M 1,2-◦propanediol, 1,2-propanediol/total metal molar ratio = 500, NaOH/1,2-propanediol molar ratio = 2,
H2O2/1,2-propanediol molar ratio = 4, T = 60 C, p(air) = 1 bar, reaction time = 3 h. b TOF (h-1) at 3 hours of reaction. TOF numbers were calculated
on the basis of total loading of metals. c 1,2-propanediol/total metal molar ratio = 2000.
the bimetallic catalysts prepared by impregnation are known
to be highly active and selective for H2O2 synthesis22–28,30 and
alcohol oxidation to aldehydes.21 In contrast to the catalysts
made by impregnation, the catalyst made by sol immobilisation
all show very high selectivity to lactate with H2O2 as oxidant,
but again no synergistic enhancement in activity is observed
for the bimetallic catalysts. Quantification of residual hydrogen
peroxide at the end of reaction showed that it had been totally
consumed, in agreement with previous reported data.36
catalyst system to oxidise primary alcohols at the expense of
secondary alcohols; hence, one would expect 1,3-propanediol to
show similar reactivity to glycerol, as both have two primary
alcohol groups. That it does not (Scheme 1) suggests that the
otherwise relatively unreactive secondary hydroxyl, which is
present in glycerol but not in 1,3-propanediol, promotes the
oxidation in some way. This could be due to some form of
anchiomeric assistance, for example by intramolecular hydrogen
bonding between the proton of the secondary hydroxyl group
and the presumed alkoxide, which is initially formed from
the primary alcohol prior to its oxidation. Alternatively, the
vicinal diol system would be much more capable of forming a
complex with the catalyst metal surface, which could facilitate
oxidation. Further, the electron withdrawal by the secondary
hydroxyl oxygen could weaken the C-H bond, which has to be
broken during oxidation of the primary hydroxyl. Such an effect
could be considerable, as such bond cleavage could be a rate
determining step.
Effect of the substrate on reactivity
To examine the effect of the substrate structure on reactivity, we
have examined the reactions of a range of C3 alcohols using a
catalyst prepared by sol immobilisation that has been found to
be the most active and selective (1%AuPd/TiO2SIm); the results
are shown in Table 4. Using O2 as the oxidant, for the primary
alcohols we observed very high selectivities to the mono-acid and
for 2-propanol, acetone was selectively, if more slowly formed;
the complete order of reactivity observed is shown in Scheme 1.
With H2O2 as oxidant, the order of reactivity was different: with
this oxidant, 1,3-propanediol was much less reactive.
Effect of the concentration of NaOH
The effect of the concentration of NaOH was investigated for
both O2 and H2O2 as oxidants (Table 5). Increasing the con-
centration increased the conversion and selectivity for lactate.
The observed order of reactivity with O2 as oxidant is
somewhat puzzling, given the significant preference for this
Table 4 Effect of the substrate structure and oxidant using 1% wt.Au-Pd/TiO2 catalyst prepared by sol-immobilisation
Substrate
Oxidant
Conv. (%)
Selectivity (%)
TOF (h-1)
a
Glycerol
O2
O2
O2
O2
100
94
62
62
35
13
10
3
61.3 (Glycerate)
95.9 (Lactate)
94.9 (3-hydroxy-propanoate)
100.0 (propanoate)
100.0 (acetone)
53.5 (Glycerate)
95.6 (Lactate)
100.0 (3-hydroxy-propanoate)
100.0 (propanoate)
100.0 (acetone)
1999c
1880c
1246c
1239c
691c
83d
a
1,2-propanediol
1,3-propanediol
1-propanol
2-propanol
Glycerol
1,2-propanediol
1,3-propanediol
1-propanol
a
a
a
O2
b
H2O2
H2O2
H2O2
H2O2
66d
b
17d
b
b
19
14
125d
93d
b
2-propanol
H2O2
a Reaction conditions: water (20 ml), 0.6 M Substrate, Substrate/total metal molar ratio = 2000, NaOH/Substrate molar ratio = 2, T = 60 ◦C,
pO2 = 10 bar, stirring rate 1500 rpm, reaction time = 1 h. b Reaction conditions: water (20 ml), 0.6 M Substrate, Substrate/total metal molar
ratio = 2000, NaOH/Substrate molar ratio= 2, H2O2/1,2-propanediol molar ratio = 4, T = 60 ◦C, p(air) = 1 bar, reaction time = 3 h. c TOF (h-1)
at 1 hour of reaction. TOF numbers were calculated on the basis of total loading of metals. d TOF (h-1) at 3 hour of reaction. TOF numbers were
calculated on the basis of total loading of metals.
This journal is
The Royal Society of Chemistry 2009
Green Chem., 2009, 11, 1209–1216 | 1213
©

View Article Online
Scheme 1
Table 5 Influence of NaOH/1,2 propanediol molar ratio in the liquid
phase oxidation of 1,2-propanediol with 1%Au/CSI
Table 6 Oxidation of intermediates (hydroxyacetone, pyruvaldehyde
and pyruvic acid) using 1%Au/CSI
Selectivity (%)
NaOH/M
molar ratio
Selectivity (%)a
Conv. (%) Lactate Acetate Formate TOF (h-1)
Substrate
Conv. (%)
Lactate
Acetate
Formate
Hydroxyacetoneb
Hydroxyacetonec
Pyruvaldehydeb
Pyruvaldehydec
Pyruvic acidb
100
100
100
100
0
84.0
69.9
100
100
0
15.4
26.6
0
0
0
0.6
3.5
0
0
0
0.5a
1a
15
21
31
23
6
7
8
81.0
88.0
92.0
95.7
80.8
73.5
89.2
84.3
18.4
10.7
7.2
3.3
0.8
1.6
0.6
3.0
0.6
0.5
0.8
295c
420c
619c
463c
37d
2a
4a
1.0
0.5b
1b
18.4
24.9
10.2
12.6
Pyruvic acidc
31
0
100
0
49d
2b
57d
a Analysis excluding carbon oxides, which may be formed in addition.
b Reaction conditions: water (20 ml), 0.3 M Substrate, Substrate/total
metal molar ratio = 2000, NaOH/Substrate molar ratio = 2, T = 60 ◦C,
pO2 = 10 bar, stirring rate 1500 rpm, reaction time = 0.5 h. c Reaction
conditions: water (20 ml), 0.3 M Substrate, Substrate/total metal mol
ratio = 2000, NaOH/Substrate mol ratio = 2, T = 60 ◦C, pO2 = 10 bar,
stirring rate 1500 rpm, reaction time = 1 h.
4b
19
124d
a Reaction conditions: water (20 ml), 0.3 M Substrate, Substrate/total
metal molar ratio = 2000, NaOH/Substrate molar ratio = 0.5–4,
T = 60 ◦C, pO2 = 10 bar, stirring rate 1500 rpm, reaction time = 1 h.
b Reaction conditions: water (20 ml), 0.3 M 1,2-propanediol, 1,2-
propanediol/total metal molar ratio = 2000, NaOH/1,2-propanediol
molar ratio = 0.5–4, H2O2/1,2-propanediol molar ratio = 4, T = 60 ◦C,
p(air) = 1 bar, reaction time = 3 h. c TOF (h-1) at 1 hour of reaction.
TOF numbers were calculated on the basis of total loading of metals.
d TOF (h-1) at 3 hours of reaction. TOF numbers were calculated on the
basis of total loading of metals.
are consistent with their regiochemistry and the propensity of
these catalysts to oxidise primary alcohols. In addition, for
1,2-propanediol, by-product generation (acetate and formate)
suggests that parallel pathways may be operating. In addition,
when H2O2 is used as the oxidant, no synergistic enhancement in
activity is observed, suggesting that a hydroperoxy species may
not be the selective oxidation species present on the surface of
the very active catalysts.
With H2O2, the effect was primarily on conversion whereas for
O2 the use of 2M NaOH was found to be optimal, although
acceptable rates were achieved with O2 as oxidant with lower
NaOH concentrations.
The most selective catalyst system discovered during the
present work would appear to operate in a very straightforward
fashion, by reason of the very high regioselectivity: oxidation
of 1,2-propanediol to lactaldehyde is presumably overall a slow
step, which is followed by a relatively fast second oxidation to
the observed lactate product (Scheme 2). Further oxidation to
pyruvate is not observed, but may occur (see Table 6), even
Comments on the mechanism of product formation
It is apparent that the oxidation of alcohols using the bimetallic
AuPd catalysts is very selective for the formation of mono-
acid products, even when additional oxidations are possible.
The reactivity patterns we have observed for the C3 alcohols
Scheme 2
1214 | Green Chem., 2009, 11, 1209–1216
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
with less selective systems, in line with previously reported
observations.
Acknowledgements
This work formed part of the Glycerol Challenge and Sasol and
the Technology Strategy Board are thanked for their financial
support. This project is co-funded by the Technology Strategy
Board’s Collaborative Research and Development programme,
following an open competition. The Technology Strategy Board
is an executive body established by the Government to drive
innovation. It promotes and invests in research, development
and the exploitation of science, technology and new ideas for
the benefit of business – increasing sustainable economic growth
in the UK and improving quality of life. For more information
visit www.innovateuk.org.
However, in less selective or less effective reactions, when the
major by-product was very largely acetate, other pathways must
be involved or any pyruvate formed is being rapidly transformed
into acetate (Table 6). Although not observed during the present
work, this may also feature the intermediacy of hydroxyace-
tone, formed either by direct oxidation of 1,2-propanediol or,
more likely, by base-induced tautomeric equilibrium with the
intermediate enediol derived from lactaldehyde. That the final
lactate product is stable with respect to oxidation to pyruvate
supports the contention that these catalyst systems are quite
unreactive with secondary alcohols and hence it seems unlikely
that oxidation of 1,2-propanediol to hydroxyacetone would be
an important contributor to this chemistry (see Scheme 2).
Subsequent oxidation of hydroxyacetone to pyruvaldehyde,
which was also not detected, would now be viable and could then
lead to more lactic acid by a Cannizzaro reaction,2 as well as to
both acetate and formate by oxidative cleavage. In the absence of
reliable and detailed knowledge of the true nature of the interme-
diates involved in these oxidations, we cannot comment further
on the reaction mechanism. However, we have carried out some
initial studies using hydroxyacetone, pyruvaldehyde and pyruvic
acid under the same reaction conditions in the presence of
NaOH at 60 ◦C (Table 6). Pyruvaldehyde reacted immediately on
addition of base to form lactate exclusively. This tends to explain
why the former was not detected but does not confirm it as a
central intermediate in the formation of lactate. This is wholly
consistent with earlier proposals.2,7 Hydroxyacetone also reacted
immediately on addition of NaOH to the reaction mixture.
However, although lactate was the major product, significant
quantities of acetate and formate were also formed. This indi-
cates that the route based on hydroxyacetone can also be a route
for the formation of acetate and formate that are observed as
products. In contrast, pyruvic acid was less reactive and reaction
in the presence of catalyst led to the formation of acetate. In this
case no formate was observed, but the carbon mass balance
was low and we suspect that carbon oxides are also formed.
Further work will be carried out to unravel these mechanistic
complexities.
References
1 R. Datta and M. Henry, J. Chem. Technol. Biotechnol., 2006, 81,
1119.
2 L. Prati and M. Rossi, J. Catal., 1998, 176, 552.
3 A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411.
4 G. W. Huber and A. Corma, Angew. Chem. Int. Ed., 2007, 46,
7184.
5 T. Miyazawa, Y. Kasunoki, K. Kunimori and K. Tomisihige, J. Catal.,
2006, 240, 213.
6 T. Tsujino, T. Ohigashi, S. Sugiyama, K. Kawashiro and H. Hayashi,
J Molec. Cat., 1992, 71, 25.
7 H. H. C. M. Pinxt, B. F. M. Kuster and G. B. Marin, Appl. Catal. A,
2000, 191, 45.
8 A. S. K. Hashmi and G. J. Hutchings, Angew. Chem. Int. Ed., 2006,
45, 7896.
9 G. C. Bond and D. T. Thompson, Catal. Rev.-Sci. Eng., 1999, 41,
319.
10 A. Corma and H. Garcia, Chem. Soc. Rev., 2008, 37, 2096.
11 C. Della, Pina, E. Falletta, L. Prati and M. Rossi, Chem. Soc. Rev.,
2008, 37, 2077.
12 M. Haruta, T. Kobayashi, H. Sano and N. Yamada, Chem. Lett.,
1987, 16, 405.
13 P. Landon, J. Ferguson, B. E. Solsona, T. Garcia, A. F. Carley, A. A.
Herzing, C. J. Kiely, S. E. Golunski and G. J. Hutchings, Chem.
Commun., 2005, 3385.
14 P. Landon, J. Ferguson, B. E. Solsona, T. Garcia, S. Al-Sayari,
A. F. Carley, A. Herzing, C. J. Kiely, M. Makkee, J. A. Moulijn,
A. Overweg, S. E. Golunski and G. J. Hutchings, J. Mat. Chem.,
2006, 16, 199.
15 A. K. Sinha, S. Seelan, S. Tsubota and M. Haruta, Angew. Chem.
Int. Ed., 2004, 43, 1546.
16 M. D. Hughes, Y.-J. Xu, P. Jenkins, P. McMorn, P. Landon, D. I.
Enache, A. F. Carley, G. A. Attard, G. J. Hutchings, F. King, E. H.
Stitt, P. Johnston, K. Griffin and C. J. Kiely, Nature, 2005, 437,
1132.
17 A. Abad, P. Conception, A. Corma and H. Garcia, Angew. Chem.
Conclusions
Int. Ed., 2005, 44, 4066.
18 A. Corma and P. Serna, Science, 2006, 313, 332.
19 F. Porta, L. Prati, M. Rossi and G. Scari, J. Catal., 2002, 211, 464.
20 F. Porta and L. Prati, J. Catal., 2004, 224, 397.
21 D. I. Enache, J. K. Edwards, P. Landon, B. Solsona-Espriu, A. F.
Carley, A. A. Herzing, M. Watanabe, C. J. Kiely, D. W. Knight and
G. J. Hutchings, Science, 2006, 311, 362.
22 P. Landon, P. J. Collier, A. J. Papworth, C. J. Kiely and G. J. Hutchings,
Chem. Commun., 2002, 2058.
23 P. Landon, P. J. Collier, A. F. Carley, D. Chadwick, A. J. Papworth, A.
Burrows, C. J. Kiely and G. J. Hutchings, Phys. Chem. Chem. Phys.,
2003, 5, 1917.
24 B. E. Solsona, J. K. Edwards, P. Landon, A. F. Carley, A. Herz-
ing, C. J. Kiely and G. J. Hutchings, Chem. Mater., 2006, 18,
2689.
25 J. K. Edwards, B. Solsona, P. Landon, A. F. Carley, A. Herzing, M.
Watanabe, C. J. Kiely and G. J. Hutchings, J. Mater. Chem., 2005,
15, 4595.
We have shown that AuPd catalysts prepared using a sol
immobilisation method are very active for the selective ox-
idation of 1,2-propanediol to lactate. The sol immobilised
materials are significantly more active than catalysts prepared
by impregnation and this is due to the smaller metal particle
sizes attained with the sol immobilisation method Alloying
gold with palladium has also been shown to improve activity,
and in some instances even to enhance selectivity to lactic
acid. Very high selectivities at high conversion are achieved.
This is possible because the formation of lactate and the by-
products, acetate and formate, occurs via parallel reaction
pathways. In view of this, we consider that this selective oxidation
pathway forms the basis of a new green process for lactate
synthesis, which could compete with the present fermentation
pathway.
26 J. K. Edwards, B. E. Solsona, P. Landon, A. F. Carley, A. Herzing,
C. J. Kiely and G. J. Hutchings, J. Catal., 2005, 236, 69.
This journal is
The Royal Society of Chemistry 2009
Green Chem., 2009, 11, 1209–1216 | 1215
©

View Article Online
27 J. K. Edwards, A. F. Carley, A. A. Herzing, C. J. Kiely and G. J.
Hutchings, Faraday Disc., 2008, 138, 225.
28 J. K. Edwards, A. Thomas, A. F. Carley, A. A. Herzing, C. J. Kiely
and G. J. Hutchings, Green Chem., 2008, 10, 388.
32 S. Carrettin, P. McMorn, P. Johnston, K. Griffin, C. J. Kiely and G. J.
Hutchings, Phys. Chem. Chem. Phys., 2003, 5, 1329.
33 S. Carrettin, P. McMorn, P. Johnston, K. Griffin and G. J. Hutchings,
Chem. Commun., 2002, 696.
29 N. Dimitratos, J. A. Lopez-Sanchez, D. Morgan, A. F. Carley, L.
Prati and G. J. Hutchings, Cat. Today, 2007, 122, 317.
30 J. A. Lopez-Sanchez, N. Dimitratos, P. Miedziak, E. Ntainjua, J. K.
Edwards, D. Morgan, A. F. Carley, R. Tiruvalam, C. J. Kiely and
G. J. Hutchings, Phys. Chem. Chem. Phys., 2008, 10, 1921.
31 A. A. Herzing, M. Watanabe, J. K. Edwards, M. Conte, Z.-R. Tang,
G. J. Hutchings and C. J. Kiely, Faraday Disc., 2008, 138, 337.
34 S. Demirel, P. Kern, M. Lucas and P. Claus, Catal. Today, 2007, 122,
292.
35 N. Dimitratos, J. A. Lopez-Sanchez, J. M. Anthonykutty, A. F. Car-
ley, R. C. Tiruvalam, A. A. Herzing, C. J. Kiely, D. W. Knight and G. J.
Hutchings, Phys. Chem. Chem. Phys., 2009, DOI: 10.1039/b904317a.
36 L. Prati, P. Spontoni and A. Gaiassi, Topics in Catalysis, 2009, 52,
288.
1216 | Green Chem., 2009, 11, 1209–1216
This journal is
The Royal Society of Chemistry 2009
©