Efficient synthesis of lactic acid by aerobic oxidation of glycerol on Au-Pt/TiO2 catalysts

Article abstract of DOI:10.1002/chem.201000740

(Chemical equation presented) Go for the burn! A one-pot approach to the efficient synthesis of lactic acid through the aerobic oxidation of glycerol in the presence of bases at an atmospheric pressure of O₂ is reported, which combines two cons

Full text of DOI:10.1002/chem.201000740

DOI: 10.1002/chem.201000740
Efficient Synthesis of Lactic Acid by Aerobic Oxidation of Glycerol on
Au–Pt/TiO2 Catalysts
Yihong Shen, Shenghong Zhang, Hongjia Li, Yuan Ren, and Haichao Liu*[a]
Biomass and its derivatives provide renewable alterna-
tives to fossil fuel resources for sustainable production of
chemicals and liquid fuels.[1–3] One such example is glycerol,
which is currently available in surplus as an inevitable by-
product from biodiesel production by transesterification of
vegetable oils, and can also be potentially obtained from
more sustainable microalgae or cellulose and its derivatives
in the near future.[1,3–6] Glycerol is polyfunctional and it is
reported to convert to a variety of valuable chemicals or in-
termediates by different catalytic reactions, such as selective
hydrogenolysis, oxidation, and dehydration.[1–3,5,6] These fea-
tures render glycerol viable as a versatile biobuilding
block,[1–3,5,6] and thus its new chemistry has been largely ex-
ploring to expand its outlets, which can also in turn optimize
the economy of biodiesel production. Herein, we report a
new oxidation reaction of glycerol to lactic acid.
Lactic acid (2-hydroxypropanoic acid) is an important
platform chemical and widely applied in the food, pharma-
ceutical, and chemical industries.[7–9] Its market demand
grows rapidly, in particular, because of its application in the
synthesis of polylactic acids useful as biodegradable plastics
and biocompatible medical materials.[7–9] Lactic acid can be
produced by chemical and fermentation routes.[7–9] The
chemical routes involve the reactions of petroleum-based
feedstocks, such as acetaldehyde with HCN in the presence
of H2SO4, which are clearly not green and sustainable.
Therefore, lactic acid is currently manufactured mainly by
fermentation of carbohydrates (e.g., starch-derived glucose).
However, the efficiency and productivity of the fermentative
method are low and need to be improved substantially. It is
thus imperative to develop more efficient methods, includ-
ing the use of new reactants for the large-scale production
of lactic acid.
Recently, Kishida, et al.[10] reported that a hydrothermal
reaction of glycerol with NaOH at 573 K forms lactic acid in
high yields. Davis et al.[11] also detected the formation of
lactic acid in glycerol hydrogenolysis to propylene glycol in
alkaline solutions at 473 K and 4.0 MPa H2. These previous
studies have shown the potential of glycerol as an alterna-
tive reactant for the synthesis of lactic acid, irrespective of
their low efficiencies or harsh reaction conditions, which
may present significant hurdles to their industrial practice.
Moreover, these studies proposed the formation of lactic
acid in the presence of bases via a glyceraldehyde intermedi-
ate formed from glycerol dehydrogenation,[10,11] which ap-
pears to be thermodynamically more favorable under oxida-
tion conditions. This proposition is in accordance with the
known isomerization of glyceraldehyde and its isomer dihy-
droxyacetone to lactic acid.[8,12] However, extensive studies
on glycerol oxidation have shown that glyceric acid is gener-
ally the dominant product in the presence of bases, without
the formation of lactic acid in any significant amount,[13–17]
apparently reflecting the susceptibility of glyceraldehyde or
dihydroxyacetone to oxidation. For instance, Hutchings,
et al.[13] reported glycerol oxidation to glyceric acid with
100% selectivity at 56% conversion on Au/C in the pres-
ence of NaOH (at 333 K and 0.3 MPa O2). Herein, we
report a one-pot approach to the efficient conversion of
glycerol into lactic acid at atmospheric pressure by a combi-
nation of glycerol oxidation with O2 on Au–Pt/TiO2 to glyc-
eraldehyde and dihydroxyacetone intermediates and their
instantaneous reactions with NaOH in water.
[a] Dr. Y. Shen, S. Zhang, H. Li, Y. Ren, Prof. Dr. H. Liu
Beijing National Laboratory for Molecular Sciences
State key Laboratory for Structural Chemistry
of Stable and Unstable Species
Table 1 shows the activities (normalized to total metal
atoms) and selectivities for the aerobic glycerol oxidation at
atmospheric pressure of O2 and 363 K in the presence of
NaOH at similar glycerol conversions (ca. 30%) on TiO2-
supported bimetallic Au–Pt with various Au/Pt atomic ratios
(3:1–1:3) and, for comparison, monometallic Au, Pt, and Pd
catalysts. Characterization of these catalysts by TEM
College of Chemistry and Molecular Engineering
Peking University, Beijing 100871 (China)
Fax : (+86)10-6275-4031
E-mail: hcliu@pku.edu.cn
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000740.
7368
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 7368 – 7371

COMMUNICATION
Table 1. Glycerol conversion activities and selectivities on different catalysts.[a]
Entry Catalyst Selectivity [%]
Activity[b]
lytic activity and lactic acid se-
lectivity. As shown in Table 1,
entry 1, Au/TiO2 catalyzed glyc-
A
]
Lactic
acid
Glyceric
acid
Tatronic
acid
Oxalic
acid
Glycolic
acid
Formic
acid
erol oxidation to lactic acid
with an activity of 374.4 hÀ1 and
1
2
3
4
5
6
Au/TiO2
Pt/TiO2
Pd/TiO2
374.4
405.2
9.1
73.8
84.8
10.3
85.6
84.3
85.3
81.5
21.0
9.5
3.5
0.7
6.2
0.5
0.6
0.7
1.4
0.3
0.2
0.0
0.2
0.2
0.2
0.2
0.5
3.2
14.4
1.7
1.7
3.2
0.8
1.6
5.2
1.5
1.7
1.6
2.0
73.8% selectivity; other prod-
[c]
63.9
10.6
11.5
11.5
11.7
ucts included glyceric acid
(21%), tatronic acid (3.5%),
and insignificant glycolic acid,
oxalic acid, and formic acid (in
total ꢀ1.6%). Pt/TiO2 was also
active for glycerol oxidation,
showing an activity of 405.2 hÀ1
and lactic acid selectivity of as
Au–Pt/TiO2 (1:1)[d]
Au–Pt/TiO2 (3:1)[d]
Au–Pt/TiO2 (1:3)[d]
Au/TiO2 +Pt/TiO2
(1:1)[e]
517.1
507.4
501.3
386.7
3.1
7
(389.8)[f]
(79.3)[f]
(15.3)[f]
(1.8)[f]
N
A
A
[a] 363 K, 1 atm O2, 2.5ꢂ10À3 mmol metal, 0.22 molLÀ1 glycerol in H2O, NaOH/glycerol=4:1 (mole ratio),
ꢀ30% glycerol conversion. [b] Glycerol conversion activities normalized per metal atom. [c] 9.4ꢂ10À3 mmol
metal was used because of its low activity. [d] Data in parentheses represent Au/Pt molar ratios. [e] Physical
mixture of Au/TiO2 and Pt/TiO2 with equal amounts. [f] Data in parentheses correspond to the theoretical
values calculated by the assumption that Au/TiO2 and Pt/TiO2 behave independently.
high
as
84.8%
(Table 1,
entry 2). In contrast, Pd/TiO2
was much less active (9.1 hÀ1)
showed that they possessed similar metal particle sizes (2.7–
3.8 nm, Figure 1 and Figures S1–S4 in the Supporting Infor-
mation). Au, Pt, and Pd were studied because of their wide
applications as active catalyst components for glycerol oxi-
dation.[13–17] The choice of TiO2 support was based on our
experiments showing that Au dispersed on TiO2, Al2O3,
CeO2, ZrO2 and Fe2O3 actively produced lactic acid
(Table S1 in the Supporting Information) in good selectivi-
ties (67.5–73.8%) at atmospheric pressure of O2 and TiO2
was superior to the other supports in term of both the cata-
with only 10.3% selectivity for the synthesis of lactic acid
under identical conditions (Table 1, entry 3), which is in
agreement with the low activity of Pd catalysts previously
reported in glycerol oxidation.[15] Notably, combination of
Au and Pt (Au/Pt=1:1) on TiO2 led to an enhanced activity
(517.1 hÀ1) with
a high lactic acid selectivity (85.6%;
Table 1, entry 4), which remained essentially constant over a
relatively broad range of Au/Pt ratios (1:3–3:1) (Table 1, en-
tries 4–6). Moreover, the Au–Pt/TiO2 catalysts retained such
high selectivity to lactic acid (ca. 86%) even at 100% glyc-
erol conversion (Figure S5 in the Supporting Information),
corresponding to approximately 86% yield of lactic acid. To
the best of our knowledge, this is the highest lactic acid
yield directly from glycerol conversion reported to date in
the literature under such mild conditions.
For comparison, a physical mixture of Au/TiO2 and Pt/
TiO2 (Au/Pt=1:1) was examined in glycerol oxidation. Its
activity and product selectivity (Table 1, entry 7) were essen-
tially equal to the average of the activity and of the selectiv-
ity for the two individual components tested separately.
Therefore, the improved performance of the bimetallic Au–
Pt catalysts, compared to the monometallic Au and Pt cata-
lysts, is not due to the coexistence of Au and Pt, but can be
tentatively ascribed to the interaction and synergism be-
tween the two metals, as evidenced from previously report-
ed IR spectra for CO adsorption.[18,19] The interaction leads
to electron transfer from Au to Pt on Au–Pt/TiO2, in agree-
ment with the finding by X-ray photoelectron spectrosco-
py.[20] Such an interaction between Au and Pt may originate
from their alloying on Au–Pt/TiO2, which can be seen from
the high-resolution TEM images, showing uniform lattice
spacings of 2.29 ꢁ (Figure 1a) between the values of 2.35
and 2.23 ꢁ (Figures S1a and S2a in the Supporting Informa-
tion), characteristic for Au (111) and Pt (111) planes,[21] as
reported previously.[19b,22]
Such interaction and synergism also led to excellent reus-
ability of the Au–Pt catalysts, which is an important feature
of solid catalysts in liquid-phase reactions. As shown in
Figure 2, no essential decline in lactic acid yields was ob-
served on Au–Pt/TiO2 (Au/Pt=1:1) after five successive
Figure 1. TEM images and histograms of Au–Pt particle size distribution
of Au–Pt/TiO2 (Au/Pt=1:1) before (a) and after (b) five reaction cycles
at 363 K. Inset in (a) represents HRTEM image of Au–Pt/TiO2 (Au/Pt=
1:1) before the reaction.
Chem. Eur. J. 2010, 16, 7368 – 7371
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7369

H. Liu et al.
NaOH was present. In the absence of NaOH, glycerol was
oxidized on Au–Pt/TiO2 (Au/Pt=1:1) predominantly to glyc-
eraldehyde, dihydroxyacetone, and glyceric acid with selec-
tivities of 61.3, 22.5, and 11.8%, respectively (at ꢀ10% con-
version), reflecting the preference of this catalyst for oxidiz-
ing the primary hydroxyl groups over the secondary ones in
glycerol. The activity was 411.6 hÀ1 (normalized by the total
Au and Pt atoms), which is comparable to that in the pres-
ence of NaOH and indicates that there is no involvement of
NaOH in the rate-determining step of glycerol oxidation.
With increasing the glycerol conversion, the glyceraldehyde
selectivity decreased concurrently with an increase in the se-
lectivity to glyceric acid, while the selectivity to dihydroxya-
cetone and also the sum of the selectivities to glyceralde-
hyde and glyceric acid remained essentially constant. Fur-
ther, we confirmed that no tautomerization occurred be-
tween glyceraldehyde and dihydroxyacetone in such neutral
aqueous solutions, although it is known to occur under basic
or acidic conditions, showing that these two products are all
primarily formed from glycerol. Taken together, these re-
sults demonstrate that glycerol oxidation requires the pres-
ence of both Au–Pt catalysts and an O2 atmosphere; NaOH
itself does not catalyze glycerol conversion under our reac-
tion conditions (especially at temperatures as low as 363 K),
consistent with the results reported by Kishida et al.[10] Glyc-
eraldehyde and dihydroxyacetone are formed primarily
from glycerol, while glyceric acid is a secondary product
from the oxidation of glyceraldehyde. Our separate reac-
tions of glyceraldehydes and dihydroxyacetone with NaOH
showed that these intermediates readily converted to lactic
acid and other products in the absence or presence of Au–
Pt/TiO2, at rates much faster than that of the glycerol oxida-
tion. Taken together, we propose that glycerol oxidation
proceeds, as shown in Scheme 1, by kinetically relevant oxi-
dative dehydrogenation of glycerol to glyceraldehyde and
dihydroxyacetone intermediates on Au–Pt nanoparticle sur-
faces, followed by base-catalyzed dehydration and benzilic
acid rearrangement of these intermediates to lactic acid, or
their further oxidation to glyceric acid by Au–Pt/TiO2. The
two secondary reaction steps are competitive, and accord-
ingly, dictate the final selectivities to lactic acid and glyceric
acid. Such reaction pathways offer directions to optimize
the reaction conditions and thus the catalytic activities and
Figure 2. Yields of lactic acid after five reaction cycles at 363 K on Au/
TiO2, Pt/TiO2, and Au–Pt/TiO2 (Au/Pt=1:1, 2.5ꢂ10À3 mmol metal,
0.22 mollÀ1 glycerol in H2O, NaOH/glycerol=4:1 (mole ratio), 1 atm O2)
cycles. This is consistent with the characterization results for
this catalyst. Analysis of the aqueous reaction solutions by
inductive coupled plasma emission spectroscopy (ICP) after
each cycle showed no detectable leaching of Au and Pt into
the reaction mixture. The mean diameters of the Au–Pt
nanoparticles and their size distributions were essentially
unaltered after the five cycles (Figure 1b). These results
demonstrate that the bimetallic Au–Pt/TiO2 catalysts are
stable and recyclable under the reaction conditions used in
this work, which are in sharp contrast to the monometallic
Au and Pt catalysts that showed dramatic decrease in their
activities (Figure 2). No leaching of Au or Pt was detected
by ICP for these two catalysts. Together with the TEM char-
acterization results (Figure S1 in the Supporting Informa-
tion), Au/TiO2 deactivation can be ascribed to the large ag-
gregation of the Au nanoparticles (3.8 vs. 7.6 nm). No signif-
icant growth of Pt nanoparticles for Pt/TiO2 after the recy-
cling was observed (Figure S2 in the Supporting Informa-
tion), and thus oxygen covering or oxidation of the Pt
surfaces may account for the deactivation of Pt/TiO2, as gen-
erally found for Pt catalysts in the literature.[15c,23] Taken to-
gether, it is clear that the interaction between Au and Pt can
efficiently prevent the agglomeration of Au particles and
the poisoning of Pt sites by O2, leading to the observed su-
perior stability of the Au–Pt/
TiO2 catalysts.
To understand the reaction
pathways and mechanism for
the glycerol reaction to lactic
acid, several experiments were
performed with no Au–Pt/TiO2
catalyst, NaOH, or O2 (other
conditions were the same as
those given in Table 1). It was
found that glycerol reaction did
not occur at all in the absence
of either Au–Pt/TiO2 (Au/Pt=
1:1) or O2 at 363 K, even when Scheme 1. Proposed glycerol reaction pathways on Au–Pt/TiO2.
7370
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 7368 – 7371

Efficient Synthesis of Lactic Acid
COMMUNICATION
lactic acid selectivities, as our preliminary results confirmed
that higher reaction temperatures and lower O2 concentra-
tions favor the formation of lactic acid (not shown here).
In conclusion, we have developed an efficient approach to
the synthesis of lactic acid from glycerol aerobic oxidation
in high yields on the bimetallic Au–Pt catalysts in alkaline
aqueous solutions, which proceeds through glyceraldehyde
and dihydroxyacetone intermediates formed from glycerol
oxidative dehydrogenation as rate-determining step. Further
optimization of the two competitive reactions of the dehy-
drogenated intermediates, for example, by design of the cat-
alysts and tuning of the reaction parameters will improve
the yield and productivity in this work and lead to potential
industrial practice of this approach.
[1] G. W. Huber, S. Iborra, A. Corma, Chem. Rev. 2006, 106, 4044–
4098.
[2] J. N. Chheda, G. W. Huber, J. A. Dumesic, Angew. Chem. 2007, 119,
7298–7318; Angew. Chem. Int. Ed. 2007, 46, 7164–7183.
[3] Y. Shen, S. Wang, C. Luo, H. Liu, Prog. Chem. 2007, 19, 431–436.
[4] C. Luo, S. Wang, H. Liu, Angew. Chem. 2007, 119, 7780–7783;
Angew. Chem. Int. Ed. 2007, 46, 7636–7639.
[5] a) P. Gallezot, Catal. Today 2007, 121, 76–91; b) M. Pagliaro, R. Cir-
iminna, H. Kimura, M. Rossi, C. D. Pina, Angew. Chem. 2007, 119,
4516–4522; Angew. Chem. Int. Ed. 2007, 46, 4434–4440.
[6] a) A. Behr, J. Eilting, K. Irawadi, J. Leschinski, F. Lindner, Green
Chem. 2008, 10, 13–30; b) C. Zhou, J. N. Beltramini, Y. Fan, G. Q.
Lu, Chem. Soc. Rev. 2008, 37, 527–549; c) Y. Zheng, X. Chen, Y.
Shen, Chem. Rev. 2008, 108, 5253–5277.
[7] Y. Fan, C. Zhou, X. Zhu, Catal. Rev. Sci. Eng. 2009, 51, 293–324.
[8] R. M. West, M. S. Holm, S. Saravanamurugan, J. Xiong, Z. Bevers-
dorf, E. Taarning, C. H. Christensen, J. Catal. 2010, 269, 122–130.
[9] a) J. Lunt, Polym. Degrad. Stab. 1998, 59, 145–152; b) K. Fukushi-
ma, Y. Kimura, Polym. Int. 2006, 55, 626–642.
Experimental Section
[10] H. Kishida, F. Jin, Z. Zhou, T. Moriya, H. Enomoto, Chem. Lett.
2005, 34, 1560–1561.
[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.
[12] a) J. C. Sowden, E. K. Pohlen, J. Am. Chem. Soc. 1958, 80, 242–244;
b) G. Machell, G. N. Richards, J. Chem. Soc. 1960, 1924–1931.
[13] a) S. Carrettin, P. Mcmorn, P. Johnston, K. Griffin, G. J. Hutchings,
Chem. Commun. 2002, 696–697; b) S. Carrettin, P. Mcmorn, P. John-
ston, K. Griffin, C. J. Kiely, G. J. Hutchings, Phys. Chem. Chem.
Phys. 2003, 5, 1329–1336.
[14] W. C. Ketchie, M. Murayama, R. J. Davis, J. Catal. 2007, 250, 264–
273.
[15] a) R. Garcia, M. Besson, P. Gallezot, Appl. Catal. A 1995, 127, 165–
176; b) P. Gallezot, Catal. Today 1997, 37, 405–418; c) M. Besson, P.
Gallezot, Catal. Today 2000, 57, 127–141.
[16] a) F. Porta, L. Prati, J. Catal. 2004, 224, 397–403; b) C. L. Bianchi, P.
Canton, N. Dimitratos, F. Porta, L. Prati, Catal. Today 2005, 102–
103, 203–212; c) N. Dimitratos, C. Messi, F. Porta, L. Prati, A. Villa,
J. Mol. Catal. A 2006, 256, 21–28.
[17] a) S. Demirel-Gꢃlen, M. Lucas, P. Claus, Catal. Today 2005, 102–103,
166–172; b) S. Demirel, K. Lehnert, M. Lucas, P. Claus, Appl. Catal.
B 2007, 70, 637–643.
[18] H. Lang, S. Maldonado, K. J. Stevenson, B. D. Chandler, J. Am.
Chem. Soc. 2004, 126, 12949–12956.
[19] a) C. Mihut, C. Descorme, D. Duprez, M. D. Amiridis, J. Catal. 2002,
212, 125–135; b) D. Mott, J. Luo, P. N. Njoki, Y. Lin, L. Wang, C.
Zhong, Catal. Today 2007, 122, 378–385.
TiO2-supported monometallic Au, Pt, and Pd and bimetallic Au–Pt cata-
lysts were prepared by a single-step borohydride reduction method.[24]
Upon treatment in air at 673 K for 4 h, TiO2 was dispersed in deionized
water at 298 K, into which aqueous solutions of HAuCl4, H2PtCl6, and
PdCl2, and of HAuCl4 and H2PtCl6 (Beijing Chemicals, AR grade) with
varying Au/Pt atomic ratios were added with vigorous stirring, followed
by adding an aqueous solution of excess NaBH4 (Beijing Chemicals, AR
grade) dropwise. The total metal loadings were kept around 1 wt%.
After agitation for 2–4 h, the mixtures were filtered and washed with
H2O until no ClÀ was detected in the filtrate by using AgNO3. All cata-
lysts were dried in an oven under vacuum at 298 K overnight, and then
treated in flowing air at 573 K for 2 h.
Oxidation reactions of glycerol (Beijing Chemicals, AR grade) were typi-
cally carried out in a three-necked flask (100 mL) at 363 K with atmos-
pheric pressure O2 flowing through the reaction solutions under vigorous
stirring. Liquid-phase products were analyzed by HPLC and ESIMS, and
gas-phase products were analyzed by GC. Glycerol reaction activities
were reported as molar glycerol conversion rates per mole of metal per
hour, and selectivities on a carbon basis; carbon mass balance was better
than (99Æ3)% in this work.
TEM and high-resolution TEM images for supported Au, Pt, Pd, and
Au–Pt catalysts were taken on a Philips Tecnai F30 FEG-TEM and a Hi-
tachi H-9000NAR microscope, respectively, operated at 300 kV. Samples
were prepared by uniform dispersion in ethanol and were then placed
onto carbon-coated copper grids. The average size of metal particles and
their size distributions were calculated by averaging of not less than 300
particles randomly distributed in TEM images.
[20] H. Tada, F. Suzuki, S. Ito, T. Akita, K. Tanaka, T. Kawahara, H. Ko-
bayoshi, J. Phys. Chem. B 2002, 106, 8714–8720.
[21] S. Y. Tsen, P. A. Crozier, J. Liu, Ultramicroscopy 2003, 98, 63–72.
[22] a) J. R. Croy, S. Mostafa„ L. Hickman, H. Heinrich, B. R. Cuenya,
Appl. Catal. A 2008, 350, 207–216; b) G. De, C. N. R. Rao, J. Mater.
Chem. 2005, 15, 891–894.
Acknowledgements
[23] T. Mallat, A. Baiker, Catal. Today 1994, 19, 247–283.
[24] K. Mallick, M. J. Witcomb, M. S. Scurrell, J. Mol. Catal. A 2004, 215,
103–106.
This work was supported by the National Natural Science Foundation of
China (grant no. 20825310, 20733009, 20673005 and 50821061) and Na-
tional Basic Research Project of China (grant no. 2006CB806100).
Keywords: biomass · gold · heterogeneous catalysis · lactic
acid · platinum
Received: February 25, 2010
Published online: May 18, 2010
Chem. Eur. J. 2010, 16, 7368 – 7371
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7371