Specific selectivity of Au-catalyzed oxidation of glycerol and other C 3-polyols in water without the presence of a base

Article abstract of DOI:10.1021/cs5005568

A big challenge in upgrading bio-oxygenate platform molecules is to develop catalysts for the selective oxidation of a nonterminal HO-bonded carbon atom in polyols. We report herein the first finding of a specific selectivity of oxide-supported nano-Au catalysts for dihydroxyacetone (DHA) production in glycerol oxidation in water without NaOH. Though the support nature (Al ₂O₃, TiO₂, ZrO₂, NiO, and CuO) significantly affects the Au activity, a highly active Au/CuO catalyst offering DHA yields up to 80% at 40-50 °C has been identified. Rich data are provided to clarify that DHA is the only primary product of glycerol oxidation. This propensity of nano-Au for oxidizing the HO-bonded secondary (central) carbon is further verified by comparing the oxidation of propanediols and propanols. Molecular insight into the reactions is given on the basis of the kinetic isotopic effect study of deuterium on the oxidation of 2-propanol, uncovering an unanticipated chemistry of Au catalysis.

Full text of DOI:10.1021/cs5005568

Letter
pubs.acs.org/acscatalysis
Specific Selectivity of Au-Catalyzed Oxidation of Glycerol and Other
C3‑Polyols in Water without the Presence of a Base
Shu-Sen Liu, Ke-Qiang Sun, and Bo-Qing Xu*
Innovative Catalysis Program, Key Lab of Organic Optoelectronics & Molecular Engineering, Department of Chemistry, Tsinghua
University, Beijing 100084, China
S
* Supporting Information
ABSTRACT: A big challenge in upgrading bio-oxygenate
platform molecules is to develop catalysts for the selective
oxidation of a nonterminal HO-bonded carbon atom in
polyols. We report herein the first finding of a specific
selectivity of oxide-supported nano-Au catalysts for dihydroxy-
acetone (DHA) production in glycerol oxidation in water
without NaOH. Though the support nature (Al2O3, TiO2,
ZrO2, NiO, and CuO) significantly affects the Au activity, a highly active Au/CuO catalyst offering DHA yields up to 80% at 40−
50 °C has been identified. Rich data are provided to clarify that DHA is the only primary product of glycerol oxidation. This
propensity of nano-Au for oxidizing the HO-bonded secondary (central) carbon is further verified by comparing the oxidation of
propanediols and propanols. Molecular insight into the reactions is given on the basis of the kinetic isotopic effect study of
deuterium on the oxidation of 2-propanol, uncovering an unanticipated chemistry of Au catalysis.
KEYWORDS: heterogeneous catalysis, supported gold catalyst, selective oxidation, glycerol, propane polyols, dihydroxyacetone,
reaction mechanism
ecent interest in the use of renewable bioresourced
coproduction of TTA, glycolic acid (GCA), oxalic acid (OA),
Roxygenate platform molecules for chemicals has been and/or lactic acid (LA) was never avoided.4−8,14−20 The
formation of GLA was assumed to be preceded by a step
forming GLD and/or DHA as the reactive intermedi-
ate(s).14,15,18 However, a clear detection of GLD during the
reaction has never been reported over any Au catalyst, although
under electrochemical conditions in alkaline media, GL
oxidation on gold seemed to proceed with GLD being an
intermediate.21 The formation of DHA was detected only when
challenging the catalysis community to develop heterogeneous
catalysts for selective conversion of polyols, among them,
glycerol (GL) would be the most popular molecule.1,2 GL is
also a coproduct of biodiesel manufacturing from either
vegetable or animal oils. Its highly functionalized molecular
structure and annual production rate (ca. 250 million tons) will
enable GL to be a viable feedstock for value-added chemicals
and materials.1−3 Among many possible catalytic reactions like
oxidation, hydrogenolysis, dehydration, etherification, trans-
esterification, and so forth,4−13 the oxidation of GL could
potentially produce a number of valuable products including
glyceric acid (GLA), tartronic acid (TTA), dihydroxyacetone
(DHA), or glyceraldehyde (GLD).14−16 This richness in
product candidates of catalytic GL oxidation would be
discouraging if highly selective catalysts could not be developed,
making it a greater challenge to develop efficient catalysts that
can offer a high selectivity toward a clean oxidation reaction to
produce a specific product.
The catalytic GL oxidation reaction has often been studied
using an aqueous solution of GL and with the presence of
NaOH or a base, employing supported Au nanoparticles (NPs)
or immobilized Au colloids as active catalysts.4−8,14−20 The
added NaOH (OH−, in essence) was believed to serve as an
indispensable activator or initiator, activating GL by abstraction
of a proton from one of the three hydroxyls.4−8,14−16,19,20
Regardless of the supporting material, size and morphology of
catalytic Au, the oxidation of GL usually produces fairly high
selectivity toward the formation of GLA (50−70%), though
24
carbon18,22,23 or CeO2 was the support material of Au NPs.
Considering that the measured selectivity for DHA was
dependent on GL conversion and the reactor type,18,22−24 it
would be possible that the presence of NaOH could modify the
overall reaction channel and make it difficult to detect the
primary product and to offer a high selectivity toward one
product.
We provide herein a clear observation on the formation of
DHA in the Au-catalyzed oxidation of aqueous GL under base
(NaOH)-free conditions. Even under weakly acidic conditions
(pH = 4), Au NPs on different supporting oxides (Al2O3, TiO2,
ZrO2, NiO, and CuO) are found to be very selective in
catalyzing GL oxidation for DHA formation under mild
conditions (30−100 °C). For the first time, DHA is identified
as the only primary product of GL oxidation, and Au NPs
supported on CuO are found to be the most active for this GL-
to-DHA chemistry. Oxidation of 1,2- and 1,3-propanediols are
Received: April 24, 2014
Revised: June 3, 2014
© XXXX American Chemical Society
2226
dx.doi.org/10.1021/cs5005568 | ACS Catal. 2014, 4, 2226−2230

ACS Catalysis
Letter
a
Table 1. Catalytic GL Oxidation in Water over Supported Au Catalysts without the Presence of NaOH
b
product sel. (C%)
f
catalyst
CuO
GL conv. (%)
DHA
GLA
GCA
OA
CO2
TOF (h−1
)
0
1Au/Al2O3
1Au/TiO2
1Au/ZrO2
1Au/NiO
1Au/CuO
2Au/CuO
4.9
96.1
76.4
68.5
83.7
83.3
82.5
74.9
82.3
81.7
3.5
16.9
13.7
5.2
0
0
0.4
1.1
6.6
4.7
3.0
5.6
14.5
1.5
1.6
1.2
1.3
1.6
1.7
0
2.9
0
3.3
4.1
10.5
19.4
20.6
34.6
20.0
19.5
3.7
2.6
2.4
1.4
2.7
2.1
6.0
19.9
39.2
40.6
1.3
11.2
12.3
21.8
11.3
11.9
1.6
c
2Au/CuO
0.7
d
2Au/CuO
2Au/CuO
2.2
e
2.7
a
b
Rxn conditions: 20 mL 0.1 M GL, GL/Au = 1000 mol/mol, pH = 6.7, 80 °C, PO2 = 10 bar, stirring speed = 900 rpm, time = 2 h. DHA =
c
d
dihydroxyacetone, GLA = glyceric acid, GCA = glycolic acid, OA = oxalic acid. This reaction was conducted with GL/Au = 500 mol/mol. pH =
6.0. pH = 4.0. TOF = GL consumption rate (mol/h)/(total number of exposed Au atoms (mol)). The dispersion (D) of Au was estimated
e
f
according to D = 1/d, where d refers to the average Au particle size in nanometer measured by TEM.
a
Table 2. Catalytic GL Oxidation in Water over Supported Au Catalysts with the Presence of NaOH
b
product sel. (C%)
c
catalyst
GL conv. (%)
GLA
GCA
LA
TTA
OA
AC
FA
CO2
1Au/Al2O3
1Au/TiO2
1Au/ZrO2
1Au/NiO
1Au/CuO
2Au/CuO
48.2
34.2
34.4
19.6
27.9
24.8
60.0
58.0
60.3
33.1
43.4
44.8
14.5
10.7
11.1
5.1
7.5
6.4
6.7
0
4.5
11.5
5.4
1.2
1.6
0.4
5.0
0.4
0.8
0
12.3
10.7
10.6
25.4
24.6
27.6
0
0
1.2
2.7
0
2.8
0.7
0
30.6
7.6
23.7
24.4
0
0.2
1.1
0
1.2
0
a
Rxn conditions: 20 mL 0.3 M GL, GL/Au = 8000 mol/mol, NaOH/GL = 2 mol/mol (pH = 13.8), 60 °C, PO2 = 10 bar, stirring speed = 900 rpm,
b
c
time = 30 min. LA = lactic acid, TTA = tartronic acid, AC = acetic acid, FA = formic acid; for others see the footnote of Table 1. Detected as
carbonate anions (CO32−) by analysis with ion chromatography of the product solution.
also studied, and the results demonstrate a specific selectivity
toward the oxidative dehydrogenation of the secondary alcohols
for acetone production. These findings point to a direct
activation of GL over the Au catalyst without the help of a base
activator, which sheds light on a new dimension in chemistry of
catalysis by gold for selective oxidation of nonterminal HO-
bonded secondary (central) carbon atoms in polyols and
perhaps other bio-oxygenates containing multiple functional
groups.
The GL oxidation reaction was conducted in a Parr autoclave
reactor (50 mL) using a catalyst suspension in 20 mL aqueous
GL (0.1 M) without addition of any base (NaOH) or at pH ≤
6.7. Unless otherwise specified, the reaction was conducted at
80 °C for 2 h with an initial molar GL/Au ratio of 1000, an O2
pressure (PO2) of 10 bar, and an acidity of pH = 6.7. The
products and unreacted GL were determined quantitatively by
HPLC and GC analyses. The catalysts used were 1Au/Al2O3,
1Au/TiO2, 1Au/ZrO2, 1Au/NiO, 1Au/CuO, and 2Au/CuO,
and the preceding numbers in these sample codes denote the
Au loadings in weight percent (Table S1). Au/Al2O3 and Au/
TiO2 were AuTEK products from the World Gold Council;
Au/ZrO2, Au/NiO, and Au/CuO catalysts were prepared by
deposition−precipitation with urea as the precipitant.25 The Au
particle sizes in these catalysts, as measured by transmission
electron microscopy (TEM, Figure S1), were in the range of
2−4 nm by diameters. Detailed information on the catalyst
preparation and characterizations, procedures of the catalytic
reaction, and product analysis is given in Supporting
Information.
Table 1 shows the catalytic data of GL oxidation over the
oxide-supported Au catalysts. Irrespective of the supporting
oxides, every catalyst produced a high selectivity (69−96%) for
the formation of DHA. However, the conversion of GL varied
widely over different catalysts; for instance, the GL conversion
over the two Au/CuO catalysts (ca. 20%) were 4−7 times
those of the other catalysts. In separate experiments, the
supporting oxides (Al2O3, TiO2, ZrO2, NiO, CuO) with no Au
were found totally inactive for the reaction; CuO is shown in
Table 1 as a representative of the supporting oxides. Obviously,
Au NPs functioned as the catalyst for GL oxidation. The
numbers in the last column of Table 1 give the catalytic GL
consumption rates normalized to the surface Au sites, TOFGL
(h−1), which reveal that Au NPs in Au/CuO were ca. 2−9 times
more active than those in the other catalysts (Au/Al2O3, Au/
TiO2, Au/NiO, and Au/ZrO2). Further, both the activity and
selectivity of 2Au/CuO were found to be essentially unaffected
by lowering with dilute nitric acid the initial pH of the reaction
solution to pH = 6.0 and 4.0, as shown in the last two rows of
Table 1.
The catalysts listed in Table 1 were also used to catalyze GL
oxidation in the presence of NaOH, under conditions well-
adopted in earlier studies (60 °C, 10 bar O2, GL/Au = 8000
mol/mol, NaOH/GL = 2 mol/mol).14−17 The results from
these experiments (Table 2) confirmed no formation of DHA
but a fairly high selectivity (58−60%) for GLA production of
the Au/Al2O3, Au/TiO2, and Au/ZrO2 catalysts; other products
detected were LA (ca. 6%) and TTA (5−10%), GCA (10−
15%), OA (ca. 1%), and formic acid (FA, 11−12%). These
results are similar to those documented frequently in earlier
2227
dx.doi.org/10.1021/cs5005568 | ACS Catal. 2014, 4, 2226−2230

ACS Catalysis
Letter
literature.4−8 Similar catalytic performance was also observed
over the two Au/CuO catalysts, though they produced
considerably lower selectivity for GLA (ca. 44%) and higher
selectivity for GCA (23.7% and 24.4%) and FA (24.6% and
27.6%) (Table 2). The Au/NiO catalyst showed a preference
for TTA formation, though its selectivity for GLA was the
lowest (33%). The reaction under the presence of NaOH also
produced small amounts of carbonate (CO32−) according to IC
analysis of the reacted solutions, though the formation of
carbonate was seldom considered earlier.14−16,19,20 A compar-
ison of the data in Tables 1 and 2 therefore demonstrates that
the high selectivity for DHA formation is distinctive of Au
catalysis for GL oxidation under the base (NaOH)-free
conditions, at least within the window of 4.0 ≤ pH ≤ 6.7.
The remarkably different activity data in Table 1 demonstrate
that the activation of GL on supported Au NPs under the base-
free conditions is sensitive to the nature of the supporting
materials, Au/CuO ≫ Au/NiO > Au/Al2O3 > Au/TiO2 ≈ Au/
ZrO2. Obviously, CuO was a highly demanded supporting
material for the GL-to-DHA chemistry. The same activity data
by TOFGL of Au NPs in 1Au/CuO and 2Au/CuO (Table 1)
are not a surprising because the Au NPs in these two samples
showed basically the same sizes, and these catalytic Au NPs
actually “occupied” only a few percent (<5%) of the support
(CuO) surface.
°C when the GL conversion was less than ca. 1%. This strongly
suggests that DHA was the primary product of GL oxidation
under the base-free conditions.
It was proposed previously22 that GL oxidation over Au
catalysts would occur according to an oxidative dehydrogen-
ation mechanism, hinting that the possible primary product(s)
could be either GLD or DHA, or both.14 We also attempted to
detect a possible formation of GLD, but the result was always
negative under the various reaction conditions in Tables 1 and
3. No formation of GLD was also confirmed by analyses with
HPLC-MS, which were detailed in Figures S2 and S3
(Supporting Information).
Increasing the reaction temperature from 30 to 100 °C with
invariant initial PO2 and fixed reaction duration (e.g., PO2 = 10
bar, 2 h) resulted in continued lowering of DHA selectivity
from 100% to 60.4%, though the GL conversion was improved
from ca. 1.0% to 23%. The data in the lower part of Table 3
demonstrate a fairly mild effect of PO2 on the product selectivity
of GL oxidation at 80 °C; a 4-fold increase in PO2 lowered the
DHA selectivity only by 5%. Interestingly, the significant
lowering in DHA selectivity at the higher temperatures was
compensated mainly by formation of gaseous CO2; GLA, GCA,
or/and OA, which were typical products in Au-catalyzed GL
oxidation with the presence of NaOH (Table 2),4−8 were
produced only in very small amounts (<ca. 3%). Consistently,
the other possible minor products of GL oxidation under the
basic conditions, TTA, LA, and acetic acid (AC),14 were not at
all produced under the present base (NaOH)-free conditions
(Table 3). These different product distribution data further
suggest that under the base-free conditions, the Au-catalyzed
oxidation of GL would take place with a different mechanism
from that operating with the presence of a base (NaOH).
With the presence of NaOH, DHA was previously
considered as an active intermediate of GL oxidation over the
Au catalyst, and the formation of GLA, LA, TTA, and GCA was
proposed to be produced from subsequent secondary DHA
oxidations.14,17 We then investigated the oxidation reaction of
DHA over 2Au/CuO under the base-free conditions by
replacing GL with DHA for the reaction, using the same
reactor and reaction procedure as for GL oxidation study
(Table 4). Without presence of the catalyst (2Au/CuO), DHA
itself was stable in the solution under the reaction conditions
(40−80 °C, 10 bar O2). Addition of 2Au/CuO mainly led to
nonselective or complete oxidation of DHA toward CO2
formation, though the conversion of DHA was not high; the
selective oxidation products (GLA, GCA, OA, and AC) were
formed only in small amounts. These results clearly
The 2Au/CuO sample was then used as the priority catalyst
to investigate effects of reaction conditions (temperature, PO2,
and duration of reaction) on the catalytic oxidation of GL,
DHA, propanediols, and propanols, because the higher Au
loading entitled us to use larger amounts of the alcohol
substrates for the reaction tests at constant molar GL/Au ratios.
Shown in Table 3 are the catalytic results of GL oxidation over
Table 3. Effects of Reaction Temperature and Oxygen
Pressure on Catalytic GL Oxidation over 2Au/CuO Catalyst
a
under Base-Free Conditions
c
product sel. (C%)
PO2
(bar)
rxn temp
(°C)
GL conv.
(%)
DHA GLA GCA OA
CO2
variation of reaction temperature
b
10
10
10
10
10
10
10
30
1.0
2.1
100
99.0
0
0
0
0
40
50
0
0
0.7
2.2
1.3
4.6
2.4
1.6
0.3
0.6
3.0
7.7
12.3
32.5
3.6
97.1
94.4
85.3
82.5
60.4
0
0
60
8.7
0.8
1.2
1.6
3.2
0.5
1.2
1.2
2.2
70
16.4
20.6
22.9
80
100
variation of oxygen pressure
Table 4. Catalytic Results of DHA Oxidation with the
5
10
15
20
80
80
80
80
17.1
20.6
23.2
25.3
82.8
82.5
79.2
78.4
0
0
3.3
2.4
2.3
1.8
13.9
12.3
15.7
17.4
Absence and Presence of Au/CuO Catalyst under Base-Free
1.6
2.0
1.6
1.2
0.8
0.8
a
Conditions
b
product sel. (C%)
a
Other conditions for the reaction: 20 mL 0.1 M GL, GL/Au = 1000
DHA
conv.
(%)
b
c
rxn temp
(°C)
mol/mol, stirring speed = 900 rpm, time = 2 h. Time = 1 h. See the
footnote of Table 1.
catalyst
GLA GCA OA
AC
CO2
40−80
40
0
2Au/CuO
2Au/CuO
2Au/CuO
0.2
4.4
13.1
13.9
14.0
18.5
9.9
6.0
3.7
0
0
0
76.2
73.9
62.9
the 2Au/CuO catalyst under diversified conditions. Very high
selectivity for DHA (>94%) was always obtained when the
reaction temperature was no higher than 60 °C and when the
GL conversion was lower than 10%. An important observation
of Table 3 is that the DHA selectivity increased with a decrease
in the GL conversion, and DHA became the only product at 30
60
6.0
4.7
80
10.3
a
Other conditions for the reaction: 20 mL 0.1 M DHA in water (pH =
6.0), DHA/Au = 1000 mol/mol, PO2 = 10 bar, stirring speed = 900
rpm, time = 2 h. See the footnote of Table 1.
b
2228
dx.doi.org/10.1021/cs5005568 | ACS Catal. 2014, 4, 2226−2230

ACS Catalysis
Letter
demonstrate that the oxidation of DHA over the Au/CuO
catalyst was nonselective, favoring a complete oxidation to form
CO2 instead of selective oxidations to form the organic acids
(GLA, GCA, OA), which would indicate that the GLA, GCA,
OA, and CO2 products in Tables 1 and 3 were indeed formed
by secondary oxidation of DHA.
To gain insights into the stability and reactivity of the
individual GLA, GCA, and OA over the Au/CuO catalyst under
the base-free condition, each of these individuals was also used
to replace GL for the reaction. The results (Table S2) disclose
that GLA was stable under the reaction conditions but OA was
the most reactive molecule, whose oxidation produced CO2 as
the only product. GCA could be easily oxidized to CO2,
probably with OA as the reaction intermediate. These results
entitle us to propose the reaction channels for GL oxidation
under the base-free conditions (Figure 1).
when the reaction time was extended for as long as 12 h.
Similarly, the reactions of 2-PO and 1-PO gave, respectively,
acetone and propionic acid (PA) as their only products. Note
that the PA yield was only half of the acetone yield, which
demonstrates a much lower reactivity of 1-PO than 2-PO.
Thus, all of our catalytic data point to the fact that under the
base-free conditions, supported Au catalysts (like Au/CuO)
have a special propensity to selectively oxidize the secondary
alcohols to form their corresponding ketones. The fact that no
aldehyde (and its drivatives) but only HA was produced in the
catalytic oxidation of 1,2-PDO (Table S3) would indicate that
under the base-free conditions, the terminal OH group could
remain essentially “unreactive”. We therefore have reason to
propose that the present observed oxidative dehydrogenation
reactions of GL and 1,2-PDO is a strong indication that the
catalytic activation of the secondary C−H bond dominated the
chemistry involved in their oxidation reactions because a
selected activation of the secondary OH without affecting the
terminal OH in both reactants would be difficult to imagine.
The importance of the catalytic activation of the secondary
C−H bond in the oxidation reactions of GL and 1,2-PDO was
further supported by studying the kinetic isotope effect (KIE)
on the oxidation of 2-PO over the 2Au/CuO catalyst, using
unlabeled (CH3CH(OH)CH3) and deuterium-labeled (2-d-2-
PO, CH3CD(OH)CH3) molecules for the reaction under the
base-free conditions. The conversion of both 2-PO and 2-d-2-
PO increased linearly with the reaction time unless the
conversion was higher than ca. 35−40%, which clearly
demonstrates a first order reaction kinetics in 2-PO (Figure
S4A). The first-order reaction rate constant measured with 2-
PO and 2-d-2-PO were kH = 0.176 h−1 and kD = 0.054 h−1,
respectively, which give a KIE of kH/kD = 3.5 (Figure S4B).
This very prominent KIE ascertains that the breakage of the
secondary C−H bond is the kinetically significant or rate-
determining step in the selective oxidation of 2-PO. These
results would thus confirm that the catalytic activation of the
secondary C−H bond is the key to the selective oxidation of
both GL and 1,2-PDO for the formation of acetones (DHA and
HA).
Figure 1. Glycerol oxidation over Au/CuO without the presence of
NaOH.
The high selectivity of Au/CuO for the oxidation of the
central carbon in GL would uncover a new propensity of
supported Au catalyst for a specific catalysis toward the
oxidation of secondary alcohols under the base-free conditions.
To confirm this possibility, other C3-alcohols including 1,2-
propanediol (1,2-PDO), 1,3-propanediol (1,3-PDO), 1-prop-
anol (1-PO), and 2-propanol (2-PO) were then used,
respectively, to replace GL for the reaction over 2Au/CuO.
As shown in Table S3, the reaction of 1,2-PDO produced 1-
hydroxylacetone (HA) as the only product. However, 1,3-PDO
with no secondary OH showed little reactivity, its reaction was
not detectable in 2 h but offered a conversion as low as 3.2%
The high selectivity for DHA formation of the Au/CuO
catalyst in GL oxidation under base-free conditions would be of
potential for application, because the market price of DHA was
ca. 250 times higher than that of GL.26 Specifically, DHA is an
important chemical in organic synthesis and a skin coloring
a
Table 5. Synthesis of DHA by Catalytic Oxidation of GL in Water under Base-Free Conditions
b
product sel. (C%)
rxn temp (°C)
Gl/Au (mol/mol)
PO2 (bar)
rxn time (h)
GL conv. (%)
DHA
GLA
GCA
OA
CO2
DHA yield (%)
40
40
40
50
50
50
50
50
50
60
60
80
100
100
50
10
30
30
5
10.0
8.0
10.0
8.0
5.0
3.0
4.0
3.0
2.5
4.5
3.5
2.0
56.4
81.3
90.7
100
100
90.6
97.8
100
100
100
93.3
100
91.4
86.5
82.5
63.9
70.6
82.5
81.8
79.8
68.4
60.1
66.7
54.6
0
0.9
1.9
1.3
1.1
1.4
3.3
0.9
3.9
0.2
1.2
1.4
0.9
0.4
0.2
0.5
0.2
0.2
1.1
1.0
2.6
6.3
0.1
0.2
0.1
7.3
11.4
15.7
34.8
27.8
13.1
16.2
13.7
25.1
38.6
31.7
43.6
51.5
70.3
74.8
63.9
70.6
74.7
80.0
79.9
68.4
60.1
62.2
54.6
0
0
100
100
50
0
10
20
20
20
20
10
20
10
0
0
50
0
20
0
10
0
100
100
100
0
0
0.8
a
b
Other conditions for the reaction: 20 mL 0.1 M GL, 2Au/CuO, stirring speed = 900 rpm. See the footnote of Table 1.
2229
dx.doi.org/10.1021/cs5005568 | ACS Catal. 2014, 4, 2226−2230

ACS Catalysis
Letter
agent in cosmetic industry.27 Under the conditions employed in
Table 3, prolongation of the reaction duration had only a
limited effect on improving the DHA yield, due to the
occurrence of secondary reactions of DHA (Table 4). Attempts
were then made to enhance the DHA yield by using more
catalyst (2Au/CuO) to reach a complete GL conversion in the
possible shortest reaction duration. On increasing the catalyst
loading up to GL/Au = 50 and 20 (mol/mol), we obtained
high DHA yields of 75−80% at nearly complete GL conversion
by adjusting the other reaction parameters (temperature, PO2,
and duration), as listed in Table 5.
ACKNOWLEDGMENTS
■
The authors would like to thank Prof. Ming-Yu Ding for his
kind help during the establishment of the HPLC and IC
methods for analyses of products from the oxidation reactions.
We also acknowledge the Natural Science Foundation of China
(NSFC, grants 21033004 and 21221062) and Tsinghua
University for financial support of this work.
REFERENCES
■
(1) Medlin, J. W. ACS Catal. 2011, 1, 1284−1297.
(2) Besson, M.; Gallezot, P.; Pinel, C. Chem. Rev. 2014, 114, 1827−
1870.
Only two research groups detected earlier a formation of
DHA in Au-catalyzed GL oxidation studies with the presence of
NaOH at 60 °C; the supporting materials for the catalytic Au
nanoparticles were carbons18,22,23 and CeO2.24 One group
worked with a batch (autoclave) reactor, and the obtained
DHA selectivity was in the range of 20−27% at GL conversion
levels of 30−50%.18,22,24 The other group conducted the
reaction in a continuous down-flow slurry bubble column
reactor and reported a DHA selectivity of 53% at 30% GL
conversion.23 Obviously, the DHA yields obtained in Table 5
under the base-free conditions are far higher than those
obtained over the earlier Au/C and Au/CeO2 catalysts (<16%).
In the literature, Bi- and Au-modified Pt, and bimetallic Pd−Ag
on various carbon supports were registered as the selective
catalysts for DHA synthesis from GL oxidation in water.28
Though the selectivity to DHA could be up to 80−85%, their
offered DHA yields were in the range of 20−52%,28 which are
also much lower than the high yields shown in Table 5.
Therefore, this study uncovers for the first time that DHA
was the only primary product of catalytic GL oxidation over
supported Au NPs under the base-free conditions in water. The
Au NPs also catalyzed further oxidations of DHA, which
produced CO2 as the main and final product with GCA and OA
being the intermediate products. On adjusting the catalyst
loading and other parameters for GL oxidation, however, DHA
yields up to 80% could be still obtained. Further studies on the
reactions of 1,2- and 1,3-PDO, unlabeled and deuterium-
labeled 2-PO, and 1-PO also confirmed the high activity and
specific selectivity of Au NPs toward the oxidation of secondary
alcohols under the base-free conditions, which strongly
demonstrate that a selective activation of the secondary C−H
bond is the key to the specific selectivity in the oxidation of
both GL and 1,2-PDO. These findings reveal an unprecedented
mode of Au catalyst for the selectivity control in polyol
activation and oxidation catalysis and point to a new dimension
in chemistry of Au-based oxidation catalysis. In particular, it
sheds a light on innovating Au catalysts for fine-tuning
oxidation catalysis toward special chemicals syntheses from
bioresourced polyols.
(3) Pagliaro, M.; Ciriminna, R.; Kimura, H.; Rossi, M.; Pina, C. D.
Angew. Chem., Int. Ed. 2007, 46, 4434−4440.
(4) Prati, L.; Spontoni, P.; Gaiassi, A. Top. Catal. 2009, 52, 288−296.
(5) Ketchie, W. C.; Fang, Y. L.; Wong, M. S.; Murayama, M.; Davis,
R. J. J. Catal. 2007, 250, 94−101.
(6) Dimitratos, N.; Villa, A.; Bianchi, C. L.; Prati, L.; Makkee, M.
Appl. Catal., A 2006, 311, 185−192.
(7) Veith, G. M.; Lupini, A. R.; Pennycook, S. J.; Villa, A.; Prati, L.;
Dudney, N. J. Catal. Today 2007, 122, 248−253.
(8) Villa, A.; Gaiassi, A.; Rossetti, I.; Bianchi, C. L.; Benthem, K.; van
Veith, G. M.; Prati, L. J. Catal. 2010, 275, 108−116.
(9) Maris, E. P.; Davis, R. J. J. Catal. 2007, 249, 328−337.
(10) Tao, L. Z.; Yan, B.; Liang, Y.; Xu, B. Q. Green Chem. 2013, 15,
696−705.
(11) Gu, Y. L.; Azzouzi, A.; Pouilloux, Y.; Jer
Green Chem. 2008, 10, 164−167.
(12) Kaufman, V. R.; Garti, N. J. Am. Oil Chem. Soc. 1982, 59, 471−
474.
́ ̂
ome, F.; Barrault, J.
(13) Zhou, C. H.; Beltramini, J. N.; Fan, Y. X.; Lu, G. Q. Chem. Soc.
Rev. 2008, 37, 527−549.
(14) Ketchie, W. C.; Murayama, M.; Davis, R. J. Top. Catal. 2007, 44,
307−317.
(15) Demirel-Gulen, S.; Lucas, M.; Waerna, J.; Salmi, T.; Murzin, D.;
̈
Claus, P. Top. Catal. 2007, 44, 299−305.
(16) Zope, B. N.; Hibbitts, D. D.; Neurock, M.; Davis, R. J. Science
2010, 330, 74−78.
(17) Carrettin, S.; McMorn, P.; Johnston, P.; Griffin, K.; Hutchings,
G. J. Chem. Commun. 2002, 696−697.
(18) Demirel-Gulen, S.; Lucas, M.; Claus, P. Catal. Today 2005,
̈
102−103, 166−172.
(19) Carrettin, S.; McMorn, P.; Johnston, P.; Griffin, K.; Kiely, C. J.;
Attard, G. A.; Hutchings, G. J. Top. Catal. 2004, 27, 131−136.
(20) Villa, A.; Veith, G. M.; Prati, L. Angew. Chem., Int. Ed. 2010, 49,
4499−4502.
(21) Kwon, Y.; Schouten, K. J. P.; Koper, M. T. M. ChemSusChem
2011, 3, 1176−1185.
(22) Demirel-Gulen, S.; Lehnert, K.; Lucas, M.; Claus, P. Appl. Catal.,
̈
B 2007, 70, 637−643.
(23) Pollington, S. D.; Enache, D. I.; Landon, P.;
Meenakshisundaram, S.; Dimitratos, N.; Wagland, A.; Hutchings, G.
J.; Stitt, E. H. Catal. Today 2009, 145, 169−175.
(24) Demirel-Gulen, S.; Kern, P.; Lucas, M.; Claus, P. Catal. Today
̈
2007, 122, 292−300.
ASSOCIATED CONTENT
(25) Hong, Y. C.; Sun, K. Q.; Han, K. H.; Liu, G.; Xu, B. Q. Catal.
Today 2010, 158, 415−422.
■
S
* Supporting Information
(26) Katryniok, B.; Kimura, H.; Skrzynska, E.; Girardon, J. S.;
Fongarland, P.; Capron, M.; Ducoulombier, R.; Mimura, N.; Paula, S.;
Dumeignil, F. Green Chem. 2011, 13, 1960−1979.
(27) Hekmat, D.; Bauer, R.; Fricke, J. Bioprocess Biosyst Eng. 2003, 26,
109−116.
Experimental details and supporting tables and figures. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
(28) Besson, M.; Gallezot, P.; Pinel, C. Chem. Rev. 2014, 114, 1827−
1870 and references therein..
■
Corresponding Author
*E-mail: bqxu@mail.tsinghua.edu.cn.
Notes
The authors declare no competing financial interest.
2230
dx.doi.org/10.1021/cs5005568 | ACS Catal. 2014, 4, 2226−2230