Hetropolyacid-Catalyzed Oxidation of Glycerol into Lactic Acid under Mild Base-Free Conditions

Article abstract of DOI:10.1002/cssc.201501200

Lactic acid (LA) is a versatile platform molecule owing to the opportunity to transform this compound into useful chemicals and materials. Therefore, efficient production of LA based on inexpensive renewable feedstocks is of utmost importance for insuring its market availability. Herein, we report the efficient conversion of glycerol into LA catalyzed by heteropolyacids (HPAs) under mild base-free conditions. The catalytic performance of molecular HPAs appears to correlate with their redox potential and Bronsted acidity. Namely, H₃PMo₁₂O₄₀ (HPMo) exhibits the best selectivity towards LA (90 %) with 88 % conversion of glycerol. Loading of HPMo onto a carbon support (HPMo/C) further improves LA productivity resulting in 94 % selectivity at 98 % conversion under optimized reaction conditions. The reaction takes place through the formation of dihydroxyacetone/glyceraldehyde and pyruvaldehyde as intermediates. No leaching of HPMo was observed under the applied reaction conditions and HPMo/C could be recycled 5 times without significant loss of activity.

Full text of DOI:10.1002/cssc.201501200

DOI: 10.1002/cssc.201501200
Full Papers
Hetropolyacid-Catalyzed Oxidation of Glycerol into Lactic
Acid under Mild Base-Free Conditions
Meilin Tao,[a] Xiaohu Yi,[a, b] Irina Delidovich,[b] Regina Palkovits,*[b] Junyou Shi,*[c] and
Xiaohong Wang*[a]
Lactic acid (LA) is a versatile platform molecule owing to the
opportunity to transform this compound into useful chemicals
and materials. Therefore, efficient production of LA based on
inexpensive renewable feedstocks is of utmost importance for
insuring its market availability. Herein, we report the efficient
conversion of glycerol into LA catalyzed by heteropolyacids
(HPAs) under mild base-free conditions. The catalytic per-
formance of molecular HPAs appears to correlate with their
redox potential and Brønsted acidity. Namely, H3PMo12O40
(HPMo) exhibits the best selectivity towards LA (90%) with
88% conversion of glycerol. Loading of HPMo onto a carbon
support (HPMo/C) further improves LA productivity resulting in
94% selectivity at 98% conversion under optimized reaction
conditions. The reaction takes place through the formation of
dihydroxyacetone/glyceraldehyde and pyruvaldehyde as inter-
mediates. No leaching of HPMo was observed under the ap-
plied reaction conditions and HPMo/C could be recycled
5 times without significant loss of activity.
Introduction
The interest in platform molecules deriving from biomass re-
sources to produce fuels, chemicals, and materials is growing
because of the progressive depletion of fossil feedstocks and
the increase in greenhouse gas emission causing global warm-
ing.[1] Glycerol is an abundant and inexpensive compound as it
is a by-product of biodiesel production by trans-esterification
of vegetable oils.[1] The growing production of biodiesel is ac-
companied by increasing interest in the development of po-
tential valorization strategies of glycerol.[2] Glycerol contains
three hydroxyl groups that can undergo selective hydrogenoly-
sis, oxidation, esterification, and dehydration reactions.[2b,3]
Among these reactions, the oxidation of glycerol gives rise to
valuable oxygenated derivatives including lactic acid (LA), glyc-
eric acid (GlyA), dihydroxyacetone, or hydroxypyruvic acid. Sig-
nificant effort was recently devoted to the development and
optimization of catalysts for the selective oxidation of glycerol.
Among the listed products of selective oxidation, LA attracts
attention as a very promising platform chemical.[1,4] Nowadays
LA is utilized in production of poly(lactic acid) (PLA), as well as
in food industry, for leather and textile technologies, and in
pharmaceutics.[4a] Recent studies predicted considerable expan-
sion of the application field of LA and its esters. The latter bear
great potential as organic solvents and could be used for pro-
duction of highly pure lactic acid.[4] Currently LA is manufac-
tured biotechnologically with a capacity estimated to 300–
400 kt per annum.[5] The fermentative production of LA has the
advantage of very high enantiomeric purity of the product.
This is of utmost importance for the polymer industry, which
relies mainly on l-LA as substrate for the production of high-
quality crystalline PLA.[6] Nevertheless, the microbial production
of LA also possesses drawbacks, such as utilization of expen-
sive feedstocks and co-production of enormous amounts of
wastes. Although utilization of various carbon-containing sour-
ces for fermentation, including glycerol, is possible, most of
producers rely on pure and thus more expensive carbohydrate
feedstocks including sucrose, dextrose corn syrups, or starch
hydrolysates.[7] Additionally, fermentation proceeds most effi-
ciently in neutral medium, therefore calcium hydroxide or calci-
um carbonate is added to the broth during fermentation for
neutralization of the formed LA.[7] As a result, calcium lactate is
produced instead of LA. Traditionally, the calcium lactate solu-
tion is acidified with sulfuric acid to precipitate calcium sulfate
leaving LA in aqueous phase. The need for disposal of calcium
sulfate co-produced in equimolar amounts with LA presents
a major drawback of biotechnological processes.
[a] M. Tao,+ X. Yi,+ Prof. Dr. X. Wang
Key Lab of Polyoxometalate Science of Ministry of Education
Northeast Normal University
Changchun 130024 (P. R. China)
E-mail: wangxh665@nenu.edu.cn
[b] X. Yi,+ Dr. I. Delidovich, Prof. Dr. R. Palkovits
Chair of Heterogeneous Catalysis and Chemical Technology
RWTH Aachen University
Worringerweg 2, 52074 Aachen (Germany)
E-mail: Palkovits@itmc.rwth-aachen.de
[c] Prof. Dr. J. Shi
Wood Material Science and Engineering Key Laboratory of Jilin Province
Beihua University
Jilin 132013 (P. R. China)
Alternatively, glycerol can be transformed into LA chemoca-
talytically. Although different catalysts were proposed for this
transformation, it is remarkable that the same intermediates
were detected for various catalytic systems (Scheme 1).[8] First,
E-mail: bhsjy64@163.com
[+] These authors contributed equally to this work.
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201501200.
ChemSusChem 2015, 8, 4195 – 4201
4195
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Although glycerol can be converted in high yields to LA in
the presence of (metal)base-catalysts, application of soluble
bases entails separation and corrosion problems. Consequently,
transforming glycerol into LA under base-free conditions was
investigated. Lewis acids were recently found to be efficient
for the conversion of DHA into LA, analogous to basic cata-
lysts.[14] Tsuj et al. investigated a tandem of Au-Pd/TiO2 with
AlCl3 and reported 85.6% selectivity at 30% conversion.[15]
Later on, Cho et al. reported synthesis of a bifunctional Pt-Sn-
MFI catalyst possessing both metal (Pt) and Lewis acidic (Sn)
active sites. In the presence of Pt-Sn-MFI, 80.5% LA selectivity
at 89.8% conversion of glycerol was reported, but the reaction
lasted as long as 24 h to reach high conversion.[8j] Consequent-
ly, catalysts for a more efficient conversion of glycerol into LA
under base-free conditions are desirable.
Scheme 1. Transformation of glycerol into lactic acid (LA) through formation
of glyceraldehyde (GCA), dihydroxyacetone (DHA), and pyruvaldehyde (PAL).
glycerol undergoes oxidation to form dihydroxyacetone
(DHA) in equilibrium with glyceraldehyde (GCA). Dehydration
of DHA and GCA gives rise to pyruvaldehyde (PAL). The last
step is the benzylic acid rearrangement-like transformation of
PAL to LA.
Herein we report heteropolyacids (HPAs) as very active cata-
lysts for a selective conversion of glycerol to LA. HPAs proved
to be excellent catalysts for a number of processes owing to
very well-tuned Brønsted acidic and redox properties.[16] This
study addressed two main objectives: (i) the optimal composi-
tion of HPAs was identified to optimize the LA yield, and (ii) a
promising HPA was supported onto carbon to facilitate catalyst
recovery and recyclability.[17] Preparation of the supported HPA
was performed through an assembly based on ionic interac-
tions between a surfactant cation and the HPA anion.[18]
Glycerol was converted to LA when treated in aqueous solu-
tion of base under harsh conditions at 280–3008C.[8a,b,9] Strong
bases such as potassium and sodium hydroxide demonstrated
the highest catalytic activity among other alkali and alkali
earth hydroxides resulting in up to 90% yield of LA.[8a] For this
process, the first step, dehydrogenation of glycerol, appeared
to be rate limiting, whereas further conversion of DHA and
GCA to LA can be performed over basic catalysts under milder
conditions.[10] Thus, a combination of a dehydrogenation cata-
lyst with a base enables a decrease in the processing tempera-
ture to 90–2008C.[8c–i,11] Dehydrogenation catalysts such as sup-
ported noble metals,[8c,d,g,h,11b,c,12] molecular iridium complexe- Results and Discussion
s,[8i] and copper(I) oxide[8e] were reported to be catalytically
Screening molecular HPA catalysts
active in tandem with a base. The latter is believed to have
a double function. First, the base eliminates a proton from
glycerol facilitating dehydrogenation;[13] second, the base cata-
lyzes the transformation of DHA and GCA into LA.[10] Interest-
ingly, the LA synthesis was investigated under reducing (hydro-
gen),[8d,11] inert (nitrogen or helium),[8d,e] and oxidizing (oxy-
gen)[8c,f–j,12] atmosphere. In the presence of hydrogen, the for-
mation of LA competes with hydrogenation of the intermedi-
ates into 1,2-propanediol.[8d,11] This results in a low selectivity
towards LA of ~40%.[8d,11b] The performance under inert atmos-
phere is more selective and enables formation of LA in high
yields of 80 and 75% in the presence of base-promoted cop-
per(I) oxide[8e] and iridium on carbon (Ir/C),[8d] respectively. Fi-
nally, syntheses of LA in the presence of metal catalysts with
base under oxygen atmosphere were reported.[8c,f–j,12] A mech-
anistic study by Zope et al. demonstrated that the metal per-
forms dehydrogenation of the alcohol, whereas oxygen serves
as scavenger for the metal surface.[13] The great challenge of
the latter approach is that the aldehyde functionality of GCA
can be readily oxidized once this intermediate is formed
(Scheme 1). Therefore, over-oxidation of GCA into GlyA usually
accompanies the formation of LA.[8c,g,h,j] Nevertheless, tailoring
the catalyst and adjusting the reaction conditions enables high
yields of LA. The formation of LA with selectivities of 80–87%
was reported in the presence of bases in combination with Au-
Pt/TiO2,[8c] Au/CeO2,[8f] and Au-Pt/CeO2.[8h] 91% LA yield was ob-
tained with an iridium complex in the presence of a base.[8i]
Oxidation of glycerol into LA is a cascade of reactions that in-
cludes dehydrogenation/oxidation, dehydration, and isomeriza-
tion of a benzylic-acid-type rearrangement (Scheme 1). One-
pot implementation of these reactions was performed mostly
over a tandem of catalysts, that is, metal catalyst with
base[8c–i,11] or metal catalyst with Lewis acid.[15] At the same
time, being a multifunctional catalyst, HPAs can potentially cat-
alyze these transformations alone. This concept encouraged us
to screen molecular HPAs with different Brønsted acidity and
oxidation potential, namely the HPAs in proton form
H5PMo10V2O40, H3PMo12O40, and H3PW12O40 as well as cation-ex-
changed HPAs Ag3PMo12O40 and K3PMo12O40.
The results of the screening are summarized in Table 1, en-
tries 1–5 as well as in Table S1 in the Supporting Information,
showing that glycerol was converted to LA with rather high se-
lectivity. The stepwise transformation of glycerol to LA
(Scheme 1) suggests the following required functionalities of
the catalyst: (i) the first step demands the presence of a redox
catalyst to convert glycerol to DHA and GCA; (ii) the second
step is dehydration of DHA and GCA into PAL that proceeds ef-
ficiently over Brønsted acids; and (iii) the third step is rear-
rangement of PAL into LA that can efficiently be catalyzed by
Lewis acids. The results from Table 1 (entries 1–5) enable un-
derstanding the roles of different functionalities of HPAs in cat-
alysis. Selectivities for DHA and GCA were quantified by HPLC
and the formation of PAL was observed using GCMS. Addition-
ChemSusChem 2015, 8, 4195 – 4201
www.chemsuschem.org
4196
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
The last step of the transformation of glycerol into LA is
most probably catalyzed by molybdenum- or tungsten-con-
taining heteropolyanions. Indeed, complexation of LA with mo-
lybdenum and tungsten is well known.[20] Moreover, recent in-
vestigations uncovered an efficient rearrangement of PAL into
LA in the presence of metal salts.[14a,21] The Brønsted acidic
sites are known to be moderately active for the benzylic-acid-
type rearrangement.[14a]
Table 1. Oxidation of glycerol in the presence of HPAs as catalysts. Reac-
tion conditions: 5 mL of 10 wt% aqueous solution of glycerol, 50 mg cat-
alyst, 608C, 5 h, 5 bar O2, 800 rpm.
Entry Catalyst
Conv. LA yield TOF[a] Selectivity [%]
[%] [%] DHA GCA LA GlyA AcA
[hÀ1
]
1
2
3
4
5
6
7
8
9
H5PMo10V2O40 93 42
16
32
34
19
19
124
29
0.01
0
6
3
9
20
26
3
20
54
61
5
4
6
45 21
23
3
3
2
0
H3PMo12O40
Ag3PMo12O40
K3PMo12O40
H3PW12O40
HPMo/C[b]
90
89
68
48
98
70
9
79
72
44
29
92
43
0.3
0
88
81
2
2
2
1
The results of the screening of molecular HPAs were very
promising as these compounds exhibited excellent catalytic
performance for oxidation of glycerol into LA. The highest
yield of LA was obtained over H3PMo12O40 owing to optimal
redox and Brønsted acidic properties of this catalyst. Despite
very promising results, application of molecular catalysts is
problematic mainly owing to separation issues. Therefore, we
aimed at preparation of supported H3PMo12O40 (referred to as
HPMo in the following sections) to facilitate catalyst recycling.
Two solid supported catalysts, namely HPMo/C and HPMo/
SiO2, were prepared and tested under the screening conditions
(entries 6 and 7, Table 1). Remarkably, HPMo/C was very effi-
cient, even outperforming the molecular counterpart in terms
of activity and selectivity. The following studies were focused
on detailed investigations of HPMo/C.
11 65
14 60
3
14 61
42
39
93 0.4 0.6
[b]
HPMo/SiO2
2
0
0
3
0
0
Carbon
SiO2
3
0
6
[a] Turnover frequency calculated as TOF=(concentration of formed LA,
mollÀ1)/[(amount used HPA, molLÀ1)”(reaction time, h)]. [b] 30 wt%.
ally, the oxidation products such as GlyA and acetic acid (AcA)
were identified.
Redox properties of HPAs are determined by the nature of
heteropoly anion and acidity of the reaction medium.[16,19] Thus,
the order of redox potentials increases as follows: H3PW12O40 <
H3PMo12O40 <H5PMo10V2O40 with ~0.7 V for vanadium-contain-
ing HPAs at pH 1.[16,19] This means that H5PMo10V2O40 is the
strongest oxidant and H3PW12O40 exhibits the mildest oxidation
properties. As can be seen from Table 1, conversion of glycerol
correlates with the redox potential of the HPAs and increases in
a row: H3PW12O40 <H3PMo12O40 <H5PMo10V2O40. Meanwhile, the
highest selectivity for LA was observed for H3PMo12O40, pointing
at optimal redox properties of this catalyst. Thus, oxidation of
the GCA into GlyA and over-oxidation to AcA takes place in the
presence of H5PMo10V2O40 (entry 1, Table 1). At the same time,
the redox potential of H3PW12O40 is not sufficient for efficient
catalysis: conversion of glycerol over H3PW12O40 was much
lower compared to H3PMo12O40 and H5PMo10V2O40 (entry 5 vs.
entries 1 and 2, Table 1).
Conversion of glycerol in the presence of supported HPA
Preparation and characterization of HPMo/C
HPMo was loaded onto a carbon support by way of surfactant-
assisted synthesis.[18] This approach includes immobilization of
heteropolyanions by means of their electrostatic interactions
with a hydrophobic cation covalently bonded to the carbon
surface. Specifically, the protocol of the HPMo/C synthesis in-
cluded three main steps as shown in Scheme 2: (1) surface oxi-
dation of a carbon material by successive treatments with
H2SO4 +HNO3 and H2SO4 +H2O2 mixtures; (2) covalent binding
of ethanediamine to the surface carboxylic groups; and
(3) functionalization with 1-bromodecane and loading of HPMo
through self-assembly to form a lipid-like layer on the surface
of the carbon materials. Literature data suggest that surround-
ing the support with a lipid-like layer improves the resistance
of HPA against leaching as well as stability in case of pH and
temperature changes.[22] Loaded HPAs serve not only as cata-
lytically active sites, but also as hydrophilic moieties for tuning
Brønsted acidity is important for efficient dehydration of
DHA and GCA to obtain PAL. Indeed, K3PMo12O40 demonstrated
somewhat lower conversions than its protonated counterpart
H3PMo12O40 (entries 2 and 4, Table 1). Nevertheless, even the
heteropolysalts exhibited considerable conversion for the glyc-
erol oxidation. This is probably a result of dissociation of coor-
dinated water molecules that induces Brønsted acidity for neu-
tral HPAs.[19] Brønsted acidity of Keggin HPAs is known to
depend weakly on their composition.[16,19] Therefore, variation
in catalytic activities of the protonated HPAs for dehydration of
DHA and GCA is surprising. H3PW12O40 demonstrated lower ac-
tivity in dehydration of the DHA and GCA intermediates com-
pared to H5PMo10V2O40 and H3PMo12O40. After the reaction, the
summed selectivity for the intermediates DHA+GCA over
H3PW12O40 was 40%, whereas the selectivity to intermediates
did not exceed 11 % in the presence of H5PMo10V2O40 and
H3PMo12O40 (entries 1, 2 and 5, Table 1). Apparently, the forma-
tion of PAL depends not only on Brønsted acidity but also on
the structure of heteropolyanion. Further investigations are re-
quired for understanding this observation on a molecular level.
Scheme 2. Synthetic progresses of HPMo-modified carbon catalysts.
ChemSusChem 2015, 8, 4195 – 4201
www.chemsuschem.org
4197
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
the amphiphilic properties of the catalyst. Consequently, such
lipid-like bilayers can enhance the adsorption of oxygen and
glycerol molecules, which could cause an accelerated reaction.
The modified carbon materials were characterized by IR
spectroscopy (Figure S1). The IR spectrum of the oxidized
carbon exhibits vibration bands at 1089 and 1640 cmÀ1 corre-
sponding to the primary hydroxyl groups[23] and carboxylate
groups,[24] respectively (Figure S1a). The IR spectrum of the
NH2-modified carbon (Figure S1b) contains signals at 1564,
3016 and 1733 cmÀ1 for NÀH[25] and CÀH stretching vibrations
in the alkyl chain of ethanediamine and C=O vibrations of an
amide bond, respectively. Four characteristic peaks at 1137,
1036, 942, and 870 cmÀ1 were present in the IR spectrum of
HPMo/C (Figure S1c). These signals were attributed to the
characteristic stretching vibrations of the Keggin anion
3À
PMo12O40 at 1084, 977, 880, and 799 cmÀ1, respectively. This
result demonstrates that HPMo was successfully loaded on the
surface of carbon materials. Furthermore, vibrations at 1220–
1600 and 2800–3000 cmÀ1 confirm the presence of an alkyl
group belonging to the long carbon chain of 1-bromodecane
(Figure S1c).
The 31P magic angle spinning (MAS) NMR spectrum of
HPMo/C possesses two signals at À2.35 and À5.48 ppm origi-
3À
nating from the central PO4 unit of the PMo12O40 anion at
Figure 1. TEM micrographs of (a) carbon used as support and (b) HPMo/C;
(c) EDX of HPMo/C.
À3.8 ppm (Figure S2a). The split of this resonance can be ex-
3À
plained by interaction between PMo12O40 anion and grafted
cation.[26]
tion of oxygen on catalysts was performed to verify if the pres-
ence of lipid-like layer increases the oxygen uptake. Indeed,
the oxygen uptake was found to be significantly higher (2”
10À5 molgÀ1) for HPMo/C. This value exceeds those of HPMo
(2”10À7 molgÀ1), carbon support (6”10À7 molgÀ1), and HPMo/
SiO2 (5”10À7 molgÀ1). Probably, the structure of HPMo/C can
be described as a supported HPA ionic liquid. We prepared
and tested an unsupported HPA ionic liquid [DTAB]H2PMo12O40
(DTAB=dodecyltrimethylammonium bromide-phosphomolyb-
dic acid). For conversion of glycerol into LA, [DTAB]H2PMo12O40
demonstrated a TOF of 21 hÀ1(Table S1), which is comparable
with molecular HPAs (Table 1).
HPMo/C was further characterized by TEM and energy-disper-
sive X-ray (EDX) techniques. Figure 1a shows that the original
carbon material was tubular with uniform carbon walls. Particles
of irregular shape and overall sizes ranging from 50 to 150 nm
were observed on the surface of HPMo/C (Figure 1b). Moreover,
the wall of the functionalized carbon was surrounded by a layer
shown in the inset of Figure 1b, which might be a lipid-like bi-
layer containing the surfactant 1-bromodecane and HPMo. EDX
analysis (Figure 1c) revealed the presence of phosphorous, mo-
lybdenum and oxygen on HPMo/C. 30 wt% loading of HPMo
onto carbon was confirmed by inductively coupled plasma opti-
cal emission spectrometry (ICP-OES) analysis.
We investigated the reaction network for glycerol oxidation.
DHA and GCA were found as intermediates converted into LA
during the course of reaction (Figure 2). PAL was identified in
the reaction mixture by GCMS. AcA was found as side product
formed in minor amounts. Importantly, both DHA and GCA can
be converted into LA in the presence of HPMo/C.
Catalytic activity of HPMo/C for transformation of glycerol
into LA
HPMo/C-catalyzed glycerol oxidation was very efficient yielding
92% of LA. To the best of our knowledge, this is the highest
yield of LA achieved using a solid catalyst under very mild
base-free conditions with short reaction time. The high activity
of HPMo/C was attributed to the unique structure of the cata-
lyst: The lipid-like environment of HPAs on the surface of
carbon adsorbs glycerol and oxygen molecules to concentrate
the reaction reagent around the catalytic sites. Adsorption of
glycerol was performed on HPMo/C and the catalyst was char-
acterized by IR spectroscopy (Figure S3). Comparison of the IR
spectra of as-prepared HPMo/C and HPMo/C after adsorption
of glycerol reveals more intensive vibrational bands at 3287
and 994–1115 cmÀ1 for the latter. Most probably, these signals
correspond to ÀOH groups of glycerol. In addition, the adsorp-
The dependency of the catalyst productivity on the reaction
conditions was investigated. Increasing the reaction tempera-
ture from 60 to 908C results in gradual decrease of selectivity
for LA owing to over-oxidation of intermediates (Figure 3a).The
optimal weight ratio of glycerol to HPMo/C appeared to be
10:1 to complete the reaction within 5 h. When using more
concentrated solutions of substrate (Figure 3b) or lower
amounts of catalyst (Figure 3c), lower selectivity for LA was ob-
served after 5 h of reaction owing to incomplete conversion of
the intermediates. Use of high amounts of catalyst (0.06 or
0.07 g, Figure 3c) or performing the oxidation for 6 h (Fig-
ure 3d) did not result in a decrease of LA selectivity. This indi-
cates that LA is stable in the presence of HPMo/C under the
ChemSusChem 2015, 8, 4195 – 4201
www.chemsuschem.org
4198
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
lectivity for LA were 66% and 56% for 18 h, and 84% and
68% for 24 h, respectively. This result is very important for
practical applications. Another crucial aspect for industrial use
is purity of the substrate. Utilization of crude glycerol pro-
duced as by-products at bio-diesel plants is highly interesting
from a commercial point of view. Conversion of crude glycerol
is challenging owing to the impurities, which usually negative-
ly influence the catalytic upgrading of glycerol.[8i,27] In this
study, crude glycerol was obtained from biodiesel production
comprising 71 wt% of glycerol, 28 wt% of methanol, and
other minor organic admixtures. Surprisingly, HPMo/C dis-
played good activity in converting crude glycerol to LA with
almost the same conversion (98%) and selectivity (90%) under
mild conditions (10 wt% aqueous solution of crude glycerol,
50 mg 30 wt% HPMo/C, 608C, 5 bar O2, 5 h). This result sug-
gests HPA/C to be a methanol-tolerant catalyst capable of con-
verting crude glycerol.
Figure 2. Time course of glycerol and the products. Reaction conditions:
5 mL 10 wt% aqueous solution of glycerol, 50 mg 30 wt% HPMo/C, 5 bar
O2, 800 rpm.
The HPMo/C catalyst was successfully reused five times with-
out significant loss of catalytic activity and selectivity for LA
formation (Figure 4). After each catalytic run, the catalyst was
recovered by centrifugation, washed with ethanol and dried
before reuse. To investigate a potential leaching of HPMo
under the reaction conditions, the reaction mixture after the
catalytic experiment was studied by UV/Vis spectroscopy. As
the UV/Vis spectrum of the reaction mixture does not exhibit
characteristic peaks of HPMo in the range of 200–400 nm, we
suppose that no leaching of HPA takes place under our reac-
tion conditions (Figure S4). Therefore, minor loss of catalytic
activity of HPMo/C can be explained by loss of the catalyst
applied reaction conditions. An increase of HPA loading from
10 to 30 wt% improved glycerol conversion. Under the applied
reaction conditions, 30 wt% HPMo/C was efficient enough for
full conversion of the substrate with excellent selectivity for LA
(Figure 3e).
Interestingly, even neat glycerol could be successfully con-
verted to LA under the optimized reaction conditions [5 mL
neat glycerol, 50 mg 30 wt%HPMo/C, 5 bar (1 bar=0.1 MPa)
O2, 800 rpm, 608C]. The conversion of neat glycerol and the se-
Figure 3. Conversion of glycerol (black) and selectivity for LA (red) observed under the following reaction conditions: (a) 5 mL 10 wt% aqueous solution of
glycerol, 50 mg 30 wt% HPMo/C, 5 bar O2, 800 rpm, 5 h; (b) 5 mL aqueous solution of glycerol, 50 mg 30wt.% HPMo/C, 5 bar O2, 800 rpm, 608C, 5 h; (c) 5 mL
10 wt% aqueous solution of glycerol, 5 bar O2, 800 rpm, 608C, 5 h; (d) 5 mL 10 wt% aqueous solution of glycerol, 50 mg 30 wt% HPMo/C, 5 bar O2, 800 rpm,
608C; (e) 5 mL 10 wt% aqueous solution of glycerol, 50 mg HPMo/C, 5 bar O2, 800 rpm, 608C.
ChemSusChem 2015, 8, 4195 – 4201
www.chemsuschem.org
4199
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
the HPA. Very high potential of HPMo/C as catalyst for LA syn-
thesis can be highlighted by capability of this material to con-
vert crude glycerol (LA yield 88% under optimized conditions)
as well as neat glycerol (LA yield 58% under optimized condi-
tions for 24 h).
Experimental Section
Catalyst preparation
The following molecular HPAs were synthesized according to the
previously reported methods: H5PMo10V2O40,[28] H3PMo12O40,[29]
Ag3PMo12O40,[30] K3PMo12O40,[30] and H3PW12O40.[31]
Synthesis of HPMo/SiO2 and HPMo/C
The HPMo was supported on carbon or silica to produce HPMo/C
or HPMo/SiO2. HPMo/SiO2 was synthesized following the method
reported by Guo et al.[32] The Brunauer–Emmett–Teller (BET) surface
area and pore size of HPMo/SiO2 were measured as 645 m2 gÀ1 and
5.7 nm, respectively.
Figure 4. Reusability test catalyzed by HPMo/C in oxidation of glycerol. Reac-
tion conditions: 5 mL of 10 wt% aqueous solution of glycerol, 50 mg cata-
lyst, 608C, 5 h, 5 bar O2, 800 rpm.
The HPMo/C catalyst was prepared according to Scheme 2. Car-
bonized willow catkins were used as support. First, the surface of
carbon support was oxidized. Carbon (1 g) was dispersed under
stirring in concentrated sulfuric acid and concentrated nitric acid
with a volume ratio of 3:1 and sonicated for 1 h. Thereafter, the so-
lution was stirred for 24 h at room temperature. The reaction mix-
ture was diluted, filtered, and washed with deionized water. The
mixture was added then to 50 mL of a 4:1 (volume ratio) mixture
of concentrated sulfuric acid and hydrogen peroxide and stirred
for 30 min at 708C. The obtained oxidized carbon material was
washed with deionized water until the media of the effluent
became neutral. Finally, the product was dried at 808C.
during recycling. The IR spectrum of the used HPMo/C does
not significantly differ from that of fresh HPMo/C, indicating
that the structure of the heteropolyanion remains intact (Fig-
ure S1d). The characteristic vibration bands of heteropolyan-
ions with the Keggin structure at 1055, 951, 867, and 791 cmÀ1
were observed in IR spectrum of the used HPMo/C. The 31P
MAS NMR also confirms stability of the HPMo/C catalyst exhib-
iting two signals at À2.24 and À5.43 ppm (Figure S2b). Stabili-
ty of HPMo/C was also suggested by X-ray photoelectron spec-
troscopy (XPS) spectra, which do not differ for the fresh and
used catalysts (Figure S5).
Second, ethanediamine was grafted to the oxidized carbon materi-
al. COOH-functionalized carbon (0.2 g) was suspended in 20 mL of
2:1:1 (volume ratio) mixture of DMF, chloroform, and methanol
and sonicated for 1 h. Ethanediamine (2.24 mL) along with a 10”
molar excess 4-dimethylaminopyridine (DMAP) and 10”molar
excess 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochlo-
ride (EDCI) was added to the reaction mixture as the catalyst for
amide formation, sonicated for 1 h, and stirred for 48 h at 508C.
The reaction mixture was centrifuged for 20 min at 4000 rpm to
separate the carbons and remove unbound ethanediamine. The
modified carbon was washed with 1:1 (volume ratio) mixture of
chloroform and methanol until all the ethanediamine was re-
moved. Lipid-modified carbon was dried overnight under vacuum
at 308C.
Finally, HPMo was loaded on the produced carbon material. The
NH2-functionalized carbon (0.2 g) was mixed with 1-bromodecane
(5 mmol) in 10 mL toluene. HPMo (0.06 g) was added to the mix-
ture, sonicated for 1 h, and stirred for 24 h at 1108C. The catalyst
was recovered by filtration, washed with toluene (10 mL) five times
and dried under vacuum at 608C for 4 h.
Conclusions
This study demonstrates heteropolyacids (HPAs) as outstanding
catalysts for oxidation of glycerol into LA. Dihydroxyacetone
(DHA), glyceraldehyde (GCA), and pyruvaldehyde (PAL) were
identified as reaction intermediates. The first step of the trans-
formation is a heteropolyanion-catalyzed oxidation of glycerol
into DHA and GCA. Dehydration of these intermediates is the
second step of the process promoted by Brønsted acidic cen-
ters and leading to the formation of PAL. Finally, the rearrange-
ment of PAL gives rise to lactic acid (LA). Most probably, the re-
arrangement is also catalyzed by a heteropolyanion. Productiv-
ity of molecular HPAs is mainly a function of redox potential
and Brønsted acidity of the catalyst. The optimal oxidation
properties were demonstrated by H3PMo12O40 (HPMo) that pro-
duced LA by aerobic oxidation of glycerol in aqueous solution
in 79% yield at 608C. Loading of HPMo on carbon was per-
formed through a three-step procedure, resulting in inclusion
of the HPA in a lipid-like layer on the surface of the support.
The catalyst HPMo/C demonstrated excellent activity for glyc-
erol oxidation. 94% selectivity of LA was reached at 98% con-
version of glycerol under optimized reaction conditions:
10 wt% aqueous glycerol solution, 30 wt% HPMo/C, 10:1 w/w
glycerol/catalyst, 608C, 5 bar O2, 5 h. Most importantly, HPMo/
C was stable for 5 catalytic runs and exhibited no leaching of
Characterization of supported HPMo/C
HPMo/C was characterized by XPS, FTIR, 31P MAS NMR, and TEM.
XPS spectra were recorded on an Escalab-MK II photoelectronic
spectrometer with AlKa (1200 eV). IR spectra (4000–400 cmÀ1) were
acquired as potassium bromide discs on a Nicolet Magna 560 IR
spectrometer. The 31P MAS NMR spectra were obtained using
a Bruker AM500 spectrometer at 202.5 MHz using 85% H3PO4 as
ChemSusChem 2015, 8, 4195 – 4201
www.chemsuschem.org
4200
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Li, Y. Ren, H. Liu, Chem. Eur. J. 2010, 16, 7368–7371; d) F. Auneau, S.
Nol, G. Aubert, M. Besson, L. Djakovitch, C. Pinel, Catal. Commun.
2011, 16, 144–149; e) D. Roy, B. Subramaniam, R. V. Chaudhari, ACS
Catal. 2011, 1, 548–551; f) P. Lakshmanan, P. P. Upare, N.-T. Le, Y. K.
Hwang, D. W. Hwang, U. H. Lee, H. R. Kim, J.-S. Chang, Appl. Catal. A
2013, 468, 260–268; g) J. Xu, H. Zhang, Y. Zhao, B. Yu, S. Chen, Y. Li, L.
Hao, Z. Liu, Green Chem. 2013, 15, 1520–1525; h) R. K. P. Purushotha-
man, J. van Haveren, D. S. van Es, I. Meliµn-Cabrera, J. D. Meeldijk, H. J.
Heeres, Appl. Catal. B 2014, 147, 92–100; i) L. S. Sharninghausen, J.
Campos, M. G. Manas, R. H. Crabtree, Nat. Commun. 2014, 5, 5084;
j) H. J. Cho, C.-C. Chang, W. Fan, Green Chem. 2014, 16, 3428–3433.
[9] Z. Shen, F. Jin, Y. Zhang, B. Wu, A. Kishita, K. Tohji, H. Kishida, Ind. Eng.
Chem. Res. 2009, 48, 8920–8925.
[10] a) J. F. Harris, Carbohydr. Res. 1972, 23, 207–215; b) M. Gibbs, J. Am.
Chem. Soc. 1950, 72, 3964–3965; c) J. C. Sowden, E. K. Pohlen, J. Am.
Chem. Soc. 1958, 80, 242–244.
[11] a) E. P. Maris, R. J. Davis, J. Catal. 2007, 249, 328–337; b) E. P. Maris, W. C.
Ketchie, M. Murayama, R. J. Davis, J. Catal. 2007, 251, 281–294; c) J.
ten Dam, F. Kapteijn, K. Djanashvili, U. Hanefeld, Catal. Commun. 2011,
13, 1–5.
[12] Y. Li, S. Chen, J. Xu, H. Zhang, Y. Zhao, Y. Wang, Z. Liu, Clean 2014, 42,
1140–1144.
[13] B. N. Zope, D. D. Hibbitts, M. Neurock, R. J. Davis, Science 2010, 330, 74–
78.
[14] a) C. B. Rasrendra, B. A. Fachri, I. G. B. N. Makertihartha, S. Adisasmito,
H. J. Heeres, ChemSusChem 2011, 4, 768–777; b) M. S. Holm, S. Sarava-
namurugan, E. Taarning, Science 2010, 328, 602–605; c) P. P. Pescarmo-
na, K. P. F. Janssen, C. Delaet, C. Stroobants, K. Houthoofd, A. Philip-
paerts, C. De Jonghe, J. S. Paul, P. A. Jacobs, B. F. Sels, Green Chem.
2010, 12, 1083–1089.
external standard. TEM micrographs and EDX spectra were collect-
ed using a JEOL-2020F electron microscope with accelerating volt-
age of 200 kV. Powder samples of the catalysts were ultrasonicated
in ethanol and dispersed on copper grids. The loading of HPMo
onto HPMo/C and HPMo/SiO2 was determined by ICP-OES.
Catalytic tests
Glycerol oxidation was performed in a high-pressure batch auto-
clave of stainless steel with a polytetrafluoroethylene inlet (10 mL).
The autoclave was equipped with gas supply system and a magnet-
ic stirrer. A catalyst was suspended in 5 mL aqueous glycerol, and
the mixture was heated up to 608C. During the reaction, oxygen
pressure was maintained at 5 bar. When DHA or GCA were used as
substrates, the same reaction conditions as for glycerol transforma-
tion were used. Concentrations of glycerol, LA, DHA, GCA, GlyA,
and AcA were determined by HPLC. After the experiment, the reac-
tion mixture was diluted 10 times with distilled water and analyzed
by HPLC using a Shimadzu LC10A-VP chromatograph equipped
with a SPB-10 A Variable UV (210 nm) and a RID-10A R.I. detectors.
Prevail TM C18 column (4.6”250 mm) was used with a solution of
H2SO4 (0.1% w/w) in H2O/acetonitrile (1/2 v/v, 1.0 mLminÀ1) as the
eluent at 508C.
Acknowledgements
[15] A. Tsuji, K. T. V. Rao, S. Nishimura, A. Takagaki, K. Ebitani, ChemSusChem
2011, 4, 542–548.
[16] I. V. Kozhevnikov, Chem. Rev. 1998, 98, 171–198.
This work was supported by the National Natural Science Foun-
dation of China (No. 51078066), the major projects of Jilin Pro-
vincial Science and Technology Department (20086035,
20100416, and 20140204085GX). X.Y. thanks the China Scholar-
ship Council for financial support. I.D. thanks the Alexander von
Humboldt Foundation and the Bayer Foundation for the financial
support.
[17] L. Qian, Ind. Catal. 2012, 20, 60–62.
[18] a) J. Zhu, P.-c. Wang, M. Lu, J. Iran Chem. Soc. 2013, 10, 123–129; b) S.
Zhao, C. Sun, L. Wang, X. Xu, X. Wang, Dalton Trans. 2010, 39, 5087–
5090; c) C. Zhan, J. M. Cameron, J. Gao, J. W. Purcell, D.-L. Long, L.
Cronin, Angew. Chem. Int. Ed. 2014, 53, 10362–10366; Angew. Chem.
2014, 126, 10530–10534.
[19] I. V. Kozhevnikov, Catal. Rev. 1995, 37, 311 –352.
[20] a) M. M. Caldeira, M. L. Ramos, V. M. S. Gil, Can. J. Chem. 1987, 65, 827–
832; b) M. M. Caldeira, V. M. S. Gil, Polyhedron 1986, 5, 381–385.
[21] a) Z. Tang, W. Deng, Y. Wang, E. Zhu, X. Wan, Q. Zhang, Y. Wang, Chem-
SusChem 2014, 7, 1557–1567; b) Y. Wang, W. Deng, B. Wang, Q. Zhang,
X. Wan, Z. Tang, Y. Wang, C. Zhu, Z. Cao, G. Wang, H. Wan, Nat.
Commun. 2013, 4, 2141.
[22] a) Y. Dayani, N. Malmstadt, Langmuir 2012, 28, 8174–8182; b) S. E.
Baker, W. Cai, T. L. Lasseter, K. P. Weidkamp, R. J. Hamers, Nano Lett.
2002, 2, 1413–1417.
[23] O. Taran, E. Polyanskaya, O. Ogorodnikova, V. Kuznetsov, V. Parmon, M.
Besson, C. Descorme, Appl. Catal. A 2010, 387, 55–66.
[24] K. V. Singh, R. R. Pandey, X. Wang, R. Lake, C. S. Ozkan, K. Wang, M.
Ozkan, Carbon 2006, 44, 1730–1739.
Keywords: base-free · glycerol oxidation · heterogeneous
catalysis · heteropolyacids · lactic acid
[1] a) J. J. Bozell, G. R. Petersen, Green Chem. 2010, 12, 539–554; b) C. Tang,
J. Peng, X. Li, Z. Zhai, W. Bai, N. Jiang, H. Gao, Y. Liao, Green Chem.
2015, 17, 1159; c) J. Zhang, Y. Zhao, M. Pan, X. Feng, W. Ji, C. T. Au, ACS
Catal. 2011, 1, 32–41; d) Z. Zhai, X. Li, C. Tang, J. Peng, N. Jiang, W. Bai,
H. Gao, Y. Liao, Ind. Eng. Chem. Res. 2014, 53, 10318; e) J. Peng, X. Li, C.
Tang, W. Bai, Green Chem. 2014, 16, 108–111; f) C. Tang, Z. Zhai, X. Li, L.
Sun, W. Bai, J. Catal. 2015, 329, 206–217.
[2] a) S.-H. Chai, H.-P. Wang, Y. Liang, B.-Q. Xu, Appl. Catal. A 2009, 353,
213–222; b) J. M. Clomburg, R. Gonzalez, Trends Biotechnol. 2013, 31,
20–28.
[3] a) C.-H. Zhou, J. N. Beltramini, Y.-X. Fan, G. Q. Lu, Chem. Soc. Rev. 2008,
37, 527–549; b) B. Sels, E. D’Hondt, P. Jacobs in Catalysis for Renewables,
Wiley-VCH, 2007, pp. 223–255; c) J. Zhang, M. Sun, Y. Han, RSC Adv.
2014, 4, 35463–35466.
[4] a) M. Dusselier, P. Van Wouwe, A. Dewaele, E. Makshina, B. F. Sels, Energy
Environ. Sci. 2013, 6, 1415–1442; b) P. Mäki-Arvela, I. L. Simakova, T.
Salmi, D. Y. Murzin, Chem. Rev. 2014, 114, 1909–1971.
[25] M. G. Giorgini, M. R. Pelletti, G. Paliani, R. S. Cataliotti, J. Raman Spec-
trosc. 1983, 14, 16–21.
[26] R. Wei, M. Guo, J. Wang, Chin. J. Chem. Eng. 2009, 17, 58–63.
´
[27] a) E. Skrzynska, A. Wondołowska-Grabowska, M. Capron, F. Dumeignil,
Appl. Catal. A 2014, 482, 245–257; b) F. Yang, M. Hanna, R. Sun, Biotech-
nol. Biofuels 2012, 5, 13.
[28] G. A. Tsigdinos, C. J. Hallada, Inorg. Chem. 1968, 7, 437–441.
[29] H. Wu, J. Biol. Chem. 1920, 43, 189–220.
[30] S. Zhu, X. Gao, F. Dong, Y. Zhu, H. Zheng, Y. Li, J. Catal. 2013, 306, 155–
163.
[31] C. Rocchiccioli-Deltcheff, M. Fournier, R. Franck, R. Thouvenot, Inorg.
Chem. 1983, 22, 207–216.
[5] P. F. H. Harmsen, M. M. Hackmann, H. L. Bos, Biofuels Bioprod. Biorefin.
2014, 8, 306–324.
[6] G.-Q. Chen, M. K. Patel, Chem. Rev. 2012, 112, 2082–2099.
[7] P. Gruber, D. E. Henton, J. Starr in Biorefineries-Industrial Processes and
Products, Wiley-VCH, 2008, pp. 381–407.
[32] Y. Guo, K. Li, X. Yu, J. H. Clark, Appl. Catal. B 2008, 81, 182–191.
[8] a) H. Kishida, F. Jin, Z. Zhou, T. Moriya, H. Enomoto, Chem. Lett. 2005, 34,
1560–1561; b) C. A. Ramírez-López, J. R. Ochoa-Gómez, M. Fernµndez-
Santos, O. Gómez-JimØnez-Aberasturi, A. Alonso-Vicario, J. Torrecilla-
Soria, Ind. Eng. Chem. Res. 2010, 49, 6270–6278; c) Y. Shen, S. Zhang, H.
Received: September 4, 2015
Revised: October 21, 2015
Published online on November 27, 2015
ChemSusChem 2015, 8, 4195 – 4201
www.chemsuschem.org
4201
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim