Promotion effect of Sn on Au/Sn-USY catalysts for one-pot conversion of glycerol to methyl lactate

Article abstract of DOI:10.1021/acscatal.7b02254

Herein, hierarchical bifunctional catalysts of Sn-USY-supported Au nanoparticles were designed for the one-pot conversion of glycerol (GLY) to methyl lactate (MLA). Over Au/Sn-USY catalyst, 79% MLA yield can be obtained with a high selectivity (～90%). The promotion effect of Sn was investigated, and the interaction between extraframework SnO_x and Au was confirmed by TEM, pyridine-FT-IR, CO-FT-IR, and XPS. The interaction promotes the dispersion of Au particles (smaller and much more uniform). This is important for the oxidation of GLY to 1,3-dihydroxyacetone (DHA) and glyceraldehyde (GA), which are the intermediate species for the conversion of GLY to MLA. Meanwhile, introduction of Sn modified the acid properties of the catalyst, which are critical for the selective conversion of DHA and GA to MLA.

Full text of DOI:10.1021/acscatal.7b02254

Subscriber access provided by UNIVERSITY OF THE SUNSHINE COAST
Article
Promotion Effect of Sn on Au/Sn-USY Catalysts for
One-Pot Conversion of Glycerol to Methyl Lactate
Tianliang Lu, Xiaomin Fu, Lipeng Zhou, Yunlai Su, Xiaomei
Yang, Li Han, Jianfeng Wang, and Chengying Song
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b02254 • Publication Date (Web): 12 Sep 2017
Downloaded from http://pubs.acs.org on September 12, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
Promotion Effect of Sn on Au/SnꢀUSY Catalysts for
OneꢀPot Conversion of Glycerol to Methyl Lactate
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Tianliang Lua,b, Xiaomin Fub, Lipeng Zhoub, Yunlai Sub, Xiaomei Yangb,*, Li Hana, Jianfeng
Wanga, Chengying Songa
a Research Center of Heterogeneous Catalysis and Engineering Sciences, School of Chemical
Engineering and Energy, Zhengzhou University, 100 Kexue Road, Zhengzhou 450001, China
b College of Chemistry and Molecular Engineering, Zhengzhou University, 100 Kexue Road,
Zhengzhou 450001, China
ABSTRACT: Herein, hierarchical bifunctional catalysts of SnꢀUSY supported Au nanoparticles
were designed for the oneꢀpot conversion of glycerol (GLY) to methyl lactate (MLA). Over
Au/SnꢀUSY catalyst, 79% MLA yield can be given with a high selectivity (~90%). Promotion
effect of Sn was investigated and the interaction between extra framework SnOx and Au was
confirmed by TEM, pyridineꢀFTꢀIR, COꢀFTꢀIR and XPS. The interaction promotes the
dispersion of Au particles (smaller and much more uniform). This is important for the oxidation
of GLY to 1,3ꢀdihydroxyacetone (DHA) and glyceraldehyde (GA) which are the intermediate
species for the conversion of GLY to MLA. Meanwhile, introduction of Sn modified the acid
property of the catalyst which is critical for the selective conversion of DHA and GA to MLA.
KEYWORDS: Au; Sn; Glycerol; Methyl lactate; Bifunctional catalysts; Zeolite
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 34
1
2
3
4
INTRODUCTION
5
6
7
8
9
Selective transformation of glycerol (GLY), which is the main byproduct of biodiesel industry, is
an attractive prospect.1,2 Because of highly functionalized molecular structure of GLY, various
high valueꢀadded chemicals can be obtained by kinds of processes including selective
hydrogenolysis,3 dehydration,4 oxidation5 and esterification.6 Lactic acid and alkyl lactates are
important chemicals widely used in food, pharmaceutical and chemical industries; and their
derivative polylactic acid is a promising biodegradable polymer.7,8 Production of lactic acid and
alkyl lactates from GLY attracts much attention of researchers nowadays.2,5,9ꢀ13 For both
environmental and economical reasons, heterogenously catalytic conversion of GLY to lactic
acid and alkyl lactates is desirable.14ꢀ27 Recently, a variety of supported heterogenous catalysts
including Au,14ꢀ17 Pt,18ꢀ21 Pd,22 and Cu23,24 metals were developed for the conversion of GLY to
lactic acid in water. However, most of the above mentioned methods need high reaction
temperature and/or addition of NaOH. Moreover, further esterification, distillation and
hydrolysis are needed to produce lactic acid with high quality.25 Methyl lactate (MLA) can be
obtained directly with methanol as a solvent. This process does not need a separate esterification
step compared to the aqueous medium. Furthermore, methanol is present in the crude GLY
obtained from biodiesel. Taking methanol as a solvent eliminates the separation process for
GLY/methanol. Oneꢀpot transfromation of GLY to MLA in methanol is attractive. Ye and
coworkers applied CaO and CuO as catalysts to catalyze the conversion of glycerol to calcium
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
lactate at 230 C firstly; then methanol and CO2 were charged to the reaction system, and
calcium lactate was converted into MLA with 57% yield through ester transfer reaction at 200
oC.26 Heeres and coworkers have reported the conversion of GLY to MLA over supported Au
catalysts in baseꢀfree methanol.27
2
ACS Paragon Plus Environment

Page 3 of 34
ACS Catalysis
1
2
Based on the previous reports,14ꢀ24,27 three continuous reactions are included in the conversion
of GLY to MLA (Scheme 1): (1) oxidation of GLY to 1,3ꢀdihydroxyacetone (DHA) and
glyceraldehyde (GA), (2) dehydration of DHA and GA to pyruvaldehyde (PA), (3) alcohol
addition of PA and isomerization to MLA. Thus, oxidation active sites (for reaction 1) and acid
sites, particularly Lewis (L) acid sites (for reactions 2 and 3) should coexist in the designed
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
catalysts for the conversion of GLY to MLA.
O
HO
OH
DHA
O
OH
O
-H2O
Methanol
O2
O
HO
OH
O
Isomerization
OH
MLA
GLY
PA
OH
O
HO
GA
Scheme 1. Transformation of GLY to MLA in methanol.
As typical L acids, Snꢀcontaining catalysts show excellent performance for selective
production of MLA from DHA due to the unique activation ability of Sn for the carbonyl
group.28ꢀ31 Furthermore, the L acid property of Snꢀcontaining catalysts, which is critical for
obtaining high MLA yield, can be controled by adjustment of Sn content. We have reported that
hierarchical SnꢀUSY is a highly active catalyst for the conversion of DHA to MLA. Even at
o
ambient temperature (25ꢀ40 C), DHA was completely converted to MLA with more than 95%
yield.28 So, we used hierarchical SnꢀUSY to provide the L acid sites. On the other hand, Au
based catalysts were studied extensively for the conversion of GLY, and showed good activity
and selectivity to produce DHA.32 Thus, in this work, the hierarchical bifunctional catalysts of
SnꢀUSY supported Au nanoparticles were designed for the oneꢀpot conversion of GLY to MLA
in methanol. By a modified colloidal deposition method, hierarchical Au/SnꢀUSY were
synthesized. The effects of Sn on Au particle size and its surface electronic property, which are
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 34
1
2
3
4
5
6
7
8
9
very important for the catalytic behavior, were studied in detail, and the influence of Au
nanoparticles on the acid properties of SnꢀUSY was also investigated. By TEM, pyridineꢀFTꢀIR,
COꢀFTꢀIR and XPS, the interaction between Au and Sn, and the mechanism for preparation of
Au/SnꢀUSY were established. Moreover, various preparation conditions were optimized to
obtain Au/SnꢀUSY with excellent catalytic performance for oneꢀpot conversion of GLY to MLA.
Finally, the reaction conditions for selective conversion of GLY to MLA were optimized.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL
Materials. Glycerol, methanol, HNO3, H2SO4, NaBH4, SnCl4·5H2O, HAuCl4·4H2O and
naphthalene were obtained from commercial sources (AR grade). HꢀUSY (Si/Al = 3.2) was
purchased from Nankai University Catalyst Co., Ltd. (China). Polyvinyl alcohol (PVA;
molecular weight = 9000 ꢀ 10000) was obtained from SigmaꢀAldrich. Methyl lactate (>98%) was
purchased from TCI Shanghai, China.
Preparation of Sn-DeAl-USY. SnꢀUSY zeolites were prepared using a method described in
our previous report.28 Typically, HꢀUSY (8.0 g) was added to a nitric acid solution (200 mL, 8
M). After stirring at 85 oC for 8 h, the solid was separated from the suspension by centrifugation
and washed to pH = 7 with deionized water. The obtained dealuminated solid was denoted as
o
DeAlꢀUSY. After drying at 100 C for 12 h in air, the solid along with SnCl4·5H2O was ground
thoroughly for 1 h in a mortar. Finally, the sample was calcined at 550 oC for 3 h after drying at
o
100 C for 12 h. The obtained sample was denoted as xSnꢀDeAlꢀUSY where x represents the
weight percent of Sn to SnꢀDeAlꢀUSY.
Preparation of Au/Sn-DeAl-USY. Au was loaded on SnꢀDeAlꢀUSY via a modified colloidal
deposition method as follows. First, desired amount of PVA was diluted with deionized water
(25 mL). Then, desired volume HAuCl4 (25 mmol Lꢀ1) was added to the diluted PVA aqueous
4
ACS Paragon Plus Environment

Page 5 of 34
ACS Catalysis
1
2
solution. After stirring for 20 min, calculated volume NaBH4 solution (0.1 mol Lꢀ1, mole ratio of
NaBH4 to HAuCl4 is 5) was added dropwise to the HAuCl4 solution, and pH of the solution was
adjusted to a certain value using H2SO4 (0.25 mol Lꢀ1). Then, SnꢀDeAlꢀUSY was added to the
solution. After stirring at room temperature for 6 h, the solid was obtained by rotary evaporation
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
o
o
of the suspension at 40 C. Finally, the solid was dried at 60 C for 12 h, and calcined at 350 C
for 3 h. The obtained sample was denoted as yAu/xSnꢀDeAlꢀUSY where y represents the weight
percent of Au to Au/SnꢀDeAlꢀUSY. For comparison, Au supported on HꢀUSY (1Au/HꢀUSY)
and DeAlꢀUSY (1Au/DeAlꢀUSY) were also prepared by similar procedure.
Characterization. Xꢀray diffraction (XRD) was performed using a Panalytical X’pert PRO
powder diffractometer with CuꢀKα radiation (λ = 0.15418 nm) at 40 kV. N2 adsorptionꢀ
desorption isotherms were measured with a Quantachrome Autosorb using N2 as adsorbate at ꢀ
o
196 C. Total surface area and pore size distributions (PSDs) were calculated according to
BrunauerꢀEmmetꢀTaller (BET) method and Barrett–Joyner–Halenda (BJH) method (desorption
branch of the isotherms), respectively. Transmission electron microscopy (TEM) images were
obtained on a JEM2100 microscope operated at 200 kV. The FTꢀIR spectra of different catalysts
with pyridine or deuterated acetonitrile (CD3CN) adsorption were performed on a Bruker Tensor
Ⅱ spectrometer, and the resolution of the spectrometer is 4 cmꢀ1. The samples (9ꢀ12 mg) were
o
evacuated in a pool under 350 C for 3 h to remove water. After cooling down to room
temperature, FTꢀIR spectra in the hydroxyl stretching vibration region of the sample were
collected, and air was used as background reference. Then, another IR spectrum of the sample
was collected and used as background reference for the following measurement. After that,
pyridine or CD3CN gas was introduced to the reactor pool at room temperature until adsorption
saturation (15 min). Subsequently, a vacuum pump was used to desorb probe molecule under
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 34
1
2
3
4
5
6
7
8
9
different temperature for some time. Finally, the FTꢀIR spectrum of the catalyst with pyridine or
CD3CN adsorption was collected using the obtained IR spectrum mentioned above as
background reference. FTꢀIR spectrum of CO adsorbed over 1Au/3SnꢀDeAlꢀUSY was collected
on a Bruker Tensor 27 IR spectrometer. The sample was pressed into a selfꢀsupporting disk (13
mm diameter, 26.7 mg) and placed in a homemade IR cell attached to a closed glassꢀcirculation
system. Prior to CO adsorption, the sample disk was pretreated by heating at 250 °C for 1 h in
vacuum (pressure < 10−3 Pa) and then cooled to 20 °C, followed by the recording of a
background at a resolution of 4 cm−1, the sample disk was exposed to CO. IR spectrum of the
chemisorbed CO was recorded after evacuation to eliminate physically adsorbed CO. Xꢀray
photoelectron spectroscopy (XPS) was used to study the oxidation state of the elements on the
catalyst surface. XPS experiments were carried out on a Kratos AXIS Supra photoelectron
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
spectrometer with an Al Kα radiation (hν = 1486.6 eV). The instrument was run at 15.0 kV with
a current of 10 mA. XPS data was analyzed on the software of XPSPEAK. The binding energy
of C1s was taken as a reference at 284.6 eV. The Sn content in reaction solution after reaction
was determined by induced coupled plasmaꢀatomic emission spectroscopy (ICPꢀAES).
Catalytic Tests. The catalytic conversion of GLY to MLA was carried out in a stainless steel
autoclave reactor with a magnetic stirrer and a temperature controller. Typically, GLY (1 mmol),
Au/SnꢀDeAlꢀUSY (0.2 g) and methanol (5 mL) were added to the reactor. After the autoclave
was sealed, the atmosphere over the solution was replaced four times with O2, and then the
pressure of O2 was charged to 0.5 MPa. Then, the reactor was heated to the reaction temperature
under magnetic stirring. After reaction, the reactor was cooled down to ambient temperature and
depressurized carefully.
6
ACS Paragon Plus Environment

Page 7 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
Product Analyses. Identification of the products in the reaction solution was based on Agilent
6890N GC/5973MS and compared with the authentic samples. Conversion of GLY was
determined by external standard method on a Shimadzu LCꢀ20AT HPLC with an Aminex HPXꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
87H column (300 × 7.8 mm, column temperature was 60 C) and a RIDꢀ10A refractive index
detector. Yield of MLA was analyzed on Agilent 7890A GC with HPꢀ5 capillary column and
flame ionization detector using naphthalene as the internal standard.
RESULTS AND DISCUSSION
Physicochemical Properties of Catalysts. XRD patterns of 3SnꢀDeAlꢀUSY and 1Au/3Snꢀ
DeAlꢀUSY shown in Figure S1 indicate that introduction of Sn and loading of Au did not destroy
the crystalline structure of USY zeolite. Type IV isotherms of N2 adsorption with H3 hysteresis
loop (Figure S2) indicates that all the samples are microꢀmeso porous materials. Textural
properties of the samples are summarized in Table 1. The external surface area and the mesopore
volume of DeAlꢀUSY increased dramatically compared to HꢀUSY due to dealumination.
However, the surface area and the pore volume of 3SnꢀDeAlꢀUSY decreased dramatically. This
might be caused by the reconstitution during incorporation of Sn into the zeolite through
grinding. Textural data of 1Au/HꢀUSY, 1Au/3SnꢀDeAlꢀUSY and 1Au/DeAlꢀUSY are similar to
that of HꢀUSY, 3SnꢀDeAlꢀUSY and DeAlꢀUSY, respectively, indicating that the load of Au did
not affect the textural properties of the samples dramatically.
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 34
1
2
3
4
Table 1. Textural Parameters of the Samples
Total pore volume
(mL gꢀ1)
Mesopore volumea
(mL gꢀ1)
5
6
7
Sample
SBET (m2 gꢀ1)
SExternala (m2 gꢀ1)
8
9
HꢀUSY
DeAlꢀUSY
427
505
421
418
486
422
0.31
0.40
0.34
0.30
0.39
0.33
77
121
82
0.12
0.20
0.17
0.12
0.20
0.16
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3SnꢀDeAlꢀUSY
1Au/HꢀUSY
1Au/DeAlꢀUSY
1Au/3SnꢀDeAlꢀUSY
a
62
115
96
SExternal = SBET – SMicropore; mesopore volume = total pore volume – micropore volume, where
the micropore surface area and volume were determined by the tꢀplot method at a relative
pressure of 0.05ꢀ0.70.
Morphology of Catalysts. Particle size is very important for the catalytic performance of gold.
In general, Au nanoꢀparticles (Au NPs) smaller than 10 nm are proper for many catalytic reaction.
To obtain small (< 10 nm) and uniform Au NPs, preparation conditions and parameters including
solution pH, Au/PVA weight ratio and Sn content were studied firstly.
8
ACS Paragon Plus Environment

Page 9 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. TEM images of 1Au/3SnꢀDeAlꢀUSY catalysts prepared with different solution pH
value (Au/PVA weight ratio = 4).
The solution acidity has a marked effect on the size of the prepared Au NPs.33,34 To illustrate
the acidity effects, 1Au/3SnꢀDeAlꢀUSY samples were synthesized at different pH value (pH = 1
to 7) with a fixed Au/PVA weight ratio. TEM images in Figure 1 show that the size of the Au
NPs decreased in the beginning as the pH increased, and followed by an increase of the Au NPs
size. When the pH values were 3 and 5, size of most of Au NPs was smaller than 10 nm. Lower
(pH = 1) or higher (pH = 7) pH value caused the formation of lager Au NPs (> 10 nm). In acidic
solution (pH < 7), hydroxyl groups on PVA chains were protonated. Thus, the polymer chains
are extended in acidic solution due to the electrostatic repulsions between the positive charges of
protonated PVA chains. This prevents Au particles from aggregating.33 However, much lower
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 34
1
2
3
4
5
6
pH value (pH = 1) might lead to partial decomposition of PVA and destruction of NaBH4, which
caused the growth of Au particles.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. TEM images of 1Au/3SnꢀDeAlꢀUSY prepared with different Au/PVA weight ratio
(solution pH value = 3).
TEM images and particle distributions of 1Au/3SnꢀDeAlꢀUSY catalysts prepared with
Au/PVA weight ratio of 4, 1 and 0.25 are presented in Figure 2. When the Au/PVA weight ratio
was adjusted to 4, average gold particle diameter of 7.8 nm was observed along with a broad
particle size distribution of 5ꢀ30 nm. At the same Au concentration, decreasing the Au/PVA
weight ratio from 4 to 1 produced smaller Au NPs. Average particle diameter of 6.2 nm and a
narrower particle size distribution (4ꢀ10 nm) were obtained. However, further decreasing the
Au/PVA weight ratio from 1 to 0.25 resulted in the formation of large Au NPs, and average
particle diameter of Au NPs increased from 6.2 nm to 8.5 nm with a broader particle size
distribution (4ꢀ22 nm). Thus, there is an optimum Au/PVA weight ratio of 1 for the formation of
small and uniform Au NPs in this work. When PVA is insufficient, Au NPs would be unstable
and growth of some particles might occur, which lead to the formation of a few large particles
and the broader Au NPs size distribution. Otherwise, when PVA is excess, partial aggregation
and coalescence of Au NPs would exist due to more difficult immobilization of the Au NPs onto
the support. Similar results were reported in literature.35,36
10
ACS Paragon Plus Environment

Page 11 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. TEM images of 1Au/xSnꢀDeAlꢀUSY catalysts prepared with Au/PVA weight ratio of
1 and solution pH of 3.
Sn dopped hierarchical USY was applied as the support in this study. Figure 3 shows the
morphology of 1Au/xSnꢀDeAlꢀUSY catalysts. Surprisingly, the average Au NPs diameter and
the particle size distribution were affected significantly by the Sn content in USY. When Sn was
absent, large Au NPs with average particle size of 18.7 nm was observed associated with a broad
particle size distribution in the range of 8 nm to 32 nm. Increasing the Sn content to 1 wt%, a
dramatic decrease of average Au NPs diameter from 18.7 nm to 9.4 nm was observed along with
a narrower particle size distribution from 4 nm to 16 nm. Further increase of the Sn content from
1 wt% to 2 wt%, 3 wt% and 4 wt% resulted in a decrease of average Au NPs size from 9.4 nm to
6.8 nm, 6.2 nm and 5.5 nm, respectively. These results demonstrate that the presence of Sn
promoted the formation of small and uniform Au NPs. Growth or aggregation of Au particles
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 34
1
2
3
4
5
6
7
8
9
mainly exist in the reduction process of Au(III) to Au(0) in the gel and immobilization of the Au
NPs onto the support (including the calcination process). Size of most Au NPs in the gel
synthesized at the fixed conditions (solution pH = 3, Au/PVA weight ratio = 1 and Sn content =
3 wt%) is in the range of 2 nm to 5 nm (Figure 4a). Taking the unsupported Au NPs sample
(Figure 4a) to calcine directly, huge lump of Au formed (Figure S3). Thus, immobilization of Au
NPs onto SnꢀDeAlꢀUSY depressed the growth or aggregation of Au caused by the calcination,
although size of Au NPs on Au/SnꢀDeAlꢀUSY was larger than that of unsupported Au NPs.
However, different Sn content in SnꢀDeAlꢀUSY brought on different degree of depression effect
as described above.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Introduction of Sn into DeAlꢀUSY was realized by the postꢀsynthesis method via
dealumination of HꢀUSY zeolite with nitric acid and solidꢀstate ionꢀexchange with SnCl4. Plenty
of mesopores formed in SnꢀDeAlꢀUSY due to the dealumination as described in our previous
work,28 diameter of mesopores in SnꢀDeAlꢀUSY is mainly in the range of 5 nm to 30 nm which
was further confirmed by the TEM images of SnꢀDeAlꢀUSY in Figure 4b. By this method, there
would be some Sn sites at the surrounding of the mesopores. By analyzing the sites of Au on
1Au/xSnꢀDeAlꢀUSY (for example, partial enlarged view of 1Au/4SnꢀDeAlꢀUSY in Figure 3), it
was found that Au NPs located at the surrounding of the mesopores. Thus, it is reasonable that
the interaction between Au and Sn made Au NPs tend to locate at the Sn sites around the
mesopores, and this prevented Au NPs from aggregation. Zhang and coworkers prepared Auꢀ
Sn/C catalyst, and illustrated that strong interaction between Au and Sn made the Au species
disperse well.37 In this work, the interaction between Au and Sn was further confirmed by COꢀ
FTꢀIR, pyridineꢀFTꢀIR and XPS in the following sections.
12
ACS Paragon Plus Environment

Page 13 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. TEM images of the unsupported Au NPs in the gel (a) and 3SnꢀDeAlꢀUSY (b).
Acid Property of Catalysts. Figure 5 shows the FTꢀIR spectra in the hydroxyl stretching
vibration region. Several characteristic bands at 3740 cmꢀ1, 3470 cmꢀ1, 3630 cmꢀ1 and 3561 cmꢀ1
were observed for these samples. Bands at 3740 cmꢀ1 and 3470 cmꢀ1 can be ascribed to the
external SiꢀOH groups and the internal SiꢀOH groups in the framework formed in the process of
dealumination, respectively.38ꢀ40 Bands at 3630 cmꢀ1 and 3561 cmꢀ1 belonged to bridging
hydroxyls SiꢀOHꢀAl in zeolite cages.38,39 Spectrum of 1Au/HꢀUSY (Figure 5b) is similar to that
of HꢀUSY (Figure 5a) indicating the introduction of Au has little impact on the hydroxyl of Hꢀ
USY without dealumination. For DeAlꢀUSY (Figure 5c), intensity of bands at 3740 cmꢀ1 and
3470 cmꢀ1 increased dramatically compared to the parent HꢀUSY zeolite (Figure 5a) implying
the formation of a large amount of SiꢀOH groups. After introduction of Sn, both the intensity of
band at 3470 cmꢀ1 and band at 3740 cmꢀ1 decreased (3SnꢀDeAlꢀUSY, Figure 5d). This
phenomenon indicates that the internal SiꢀOH groups react with SnCl4 to form framework Sn,
and the external SiꢀOH groups can also react with SnCl4 to form extra framework Sn.
Interestingly, after introduction of Au into DeAlꢀUSY (1Au/DeAlꢀUSY Figure 5e), band at 3470
cmꢀ1 also decreased indicating that Au species can react with the SiꢀOH groups too. FTꢀIR
spectrum in the hydroxyl stretching vibration region of 1Au/3SnꢀDeAlꢀUSY (Figure 5f) was
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 34
1
2
3
4
5
6
similar to that of 3SnꢀDeAlꢀUSY (Figure 5d). Thus, for 1Au/3SnꢀDeAlꢀUSY, SiꢀOH groups in
DeAlꢀUSY mainly react with Sn species.
7
8
9
3561
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3630
3470
3740
(a)
(b)
(c)
(d)
(e)
(f)
4000
3800
3600
3400
3200
3000
Wavenumber (cmꢀ1)
Figure 5. FTꢀIR spectra in the hydroxyl stretching vibration region for samples of (a) Hꢀ
USY, (b) 1Au/HꢀUSY, (c) DeAlꢀUSY, (d) 3SnꢀDeAlꢀUSY, (e) 1Au/DeAlꢀUSY and (f)
1Au/3SnꢀDeAlꢀUSY.
Figure S4 and Figure S5 show the pyridineꢀFTꢀIR spectra of 1Au/3SnꢀDeAlꢀUSY prepared
with different solution pH and different Au/PVA weight ratio, respectively. Three obvious IR
bands at 1453 cmꢀ1, 1543 cmꢀ1 and 1490 cmꢀ1 observed in all the 1Au/3SnꢀDeAlꢀUSY samples
were ascribed to pyridine adsorption on the Lewis (L) acid sites, Brönsted (B) acid sites and L/B
acid sites, respectively.28,38,41 According to the area of the corresponding peak, the amount of L
and B acid sites was calculated and listed in Table S1. In general, with a fixed Au/PVA weight
ratio of 4, high pH value (5 and 7) leads to more L acid sites, which are important for this
reaction, and fewer B acid sites compared to low pH value (1 and 3). Meanwhile, Au/PVA
weight ratio did not affect the amount of L acid sites of the catalysts significantly. Figure 6
14
ACS Paragon Plus Environment

Page 15 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
shows the effect of Sn content on the acid property of 1Au/xSnꢀDeAlꢀUSY. The amount of
acidic sites is listed in Table 2. There are few L and B acid sites in DeAlꢀUSY sample due to
dealumination. Additional IR bands at 1460 cmꢀ1 was observed obviously for 1Au/DeAlꢀUSY
and 1Au/1SnꢀDeAlꢀUSY. So, IR bands at 1460 cmꢀ1 can be ascribed to pyridine adsorption on
the L acid sites created by the introduction of Au. Introduction of Au into DeAlꢀUSY leads to the
increase in the amount of both L and B acid sites by 29.5 and 8.9 ꢁmol gꢀ1, respectively. When
the Sn content increased from 0 wt% to 2 wt%, the intensity of band at 1453 cmꢀ1 became
stronger, indicating the increase of L acid sites created by the formation of framework Sn.
Meanwhile, the intensity of the band at 1460 cmꢀ1 became lower, implying that L acid sites
created by the introduction of Au decreased. This might be caused by the competition of Sn.
Total amount of L acid sites (according to the total area of bands at 1453 and 1460 cmꢀ1)
increased dramatically from 43.2 to 83.5 ꢁmol gꢀ1 when the Sn content increased from 0 wt% to
2 wt%. Further increasing the Sn content to 4 wt%, the amount of L acid sites increased slightly
to 97.5 ꢁmol gꢀ1. The introduced Sn can react with the internal SiꢀOH and the external SiꢀOH.
When the Sn loading is low (< 2 wt%), introduced Sn mainly react with the internal SiꢀOH. With
the increase of the Sn loading (≥ 2 wt%), more and more introduced Sn would react with the
external SiꢀOH around the mesopores, and these extra framework Sn species (SnOx) are
favorable for the dispersion of Au NPs.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 34
1
2
3
4
5
6
1453
1460
1543
1490
7
8
9
(a)
(b)
(c)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(d)
(e)
(f)
1575
1550
1525
1500
1475
1450
1425
1400
Wavenumber ( )
cmꢀ1
Figure 6. FTꢀIR spectra of (a) DeAlꢀUSY, (b) 1Au/DeAlꢀUSY, (c) 1Au/1SnꢀDeAlꢀUSY, (d)
1Au/2SnꢀDeAlꢀUSY, (e) 1Au/3SnꢀDeAlꢀUSY and (f) 1Au/4SnꢀDeAlꢀUSY after pyridine
o
adsorption and evacuation at 150 C for 30 min.
Table 2. Amount of Lewis and Brönsted Acid Sites of Au/SnꢀUSY with Different Sn Content.a
Sample
Lewis acid sites
(ꢁmol gꢀ1)
Brönsted
acid sites
(ꢁmol gꢀ1)
L/B ratio
DeAlꢀUSY
13.7
43.2
53.3
83.5
85.2
97.5
4.4
13.3
9.2
3.1
3.3
5.8
3.4
3.4
3.6
1Au/DeAlꢀUSY
1Au/1SnꢀDeAlꢀUSY
1Au/2SnꢀDeAlꢀUSY
1Au/3SnꢀDeAlꢀUSY
1Au/4SnꢀDeAlꢀUSY
24.9
24.8
27.0
a Calculating by the equation given in literature.42 The integrated molar extinction coefficient of
pyridine adsorbed to Lewis acid sites is 1.42 cm µmol‒1.
16
ACS Paragon Plus Environment

Page 17 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
FTꢀIR spectra of adsorbed CD3CN over DeAlꢀUSY, 3SnꢀDeAlꢀUSY, 1Au/DeAlꢀUSY and
1Au/3SnꢀDeAlꢀUSY were also collected (Figure S6). According to the previous report, bands at
2310 cmꢀ1 and 2274 cmꢀ1 could be assigned to the interaction of the CN group with isolated
framework Sn sites and silanols in Snꢀsubstituted zeolites.28,42 As shown in Figure S6, no
obvious band can be observed at 2310 cmꢀ1 in DeAlꢀUSY (Figure S6a) and 1Au/DeAlꢀUSY
(Figure S6c) implying there is no framework Sn sites in these samples. L acid sites detected in
DeAlꢀUSY (13.7 ꢁmol gꢀ1) and 1Au/DeAlꢀUSY (43.2 ꢁmol gꢀ1) by pyridineꢀFTꢀIR were caused
by the residue Al species and the introduction of Au. For 3SnꢀDeAlꢀUSY and 1Au/3SnꢀDeAlꢀ
USY, a band at 2310 cmꢀ1 was observed indicating the formation of framework Sn sites. We
calculated the amount of framework Sn sites according to the FTꢀIR spectra of CD3CN
adsorption shown in Figure S6 by a reported method.42 The results are shown in Table S2.
Introduction of Au reduces the framework Sn sites in 3SnꢀDeAlꢀUSY. This might be due to that
Au NPs located at the surrounding of the mesopores hinder the interaction of the CN group with
some framework Sn sites nearby. Meanwhile, some introduced Au might interact with the
framework Sn leading to the shift of peak at 2310 cmꢀ1 to lower wave number region as shown in
Figure S6d.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chemical States. XPS was conducted to examine the chemical state of Au and Sn (Figure 7
and Figure S7). Figure 7 shows the Sn3d in 3SnꢀDeAlꢀUSY and 1Au/3SnꢀDeAlꢀUSY. In the
Sn3d region, signals positioned at 487.2, 487.8, 495.6 and 496.2 eV can be observed for both
3SnꢀDeAlꢀUSY and 1Au/3SnꢀDeAlꢀUSY samples. Binding energy values of 487.8 and 496.2 eV
were ascribed to 3d5/2 and 3d3/2 photoelectrons of tetrahedrally coordinated framework Sn
species, 487.2 and 495.6 eV should be ascribed to 3d5/2 and 3d3/2 photoelectrons of extra
framework Sn species which probably come mainly from the Sn sites at the surrounding of the
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 34
1
2
mesopores of USY mentioned above.43,44 Compared to 3SnꢀDeAlꢀUSY, two additional signals at
lower binding energy (486.6 and 495.0 eV) were observed for 1Au/3SnꢀDeAlꢀUSY indicating
the interaction of Au and Sn. When Au was introduced to 1Au/3SnꢀDeAlꢀUSY, Au NPs
preferentially located at the extra framework Sn sites around the mesopores of USY and
transferred some electrons into the extra framework SnOx. Similar results have been reported in
many previous literature.37,45 Au 4f in 1Au/DeAlꢀUSY and 1Au/3SnꢀDeAlꢀUSY are shown in
Figure S7. Because Au species are nanoparticles which including hundreds or thousands of Au
atoms, electron transfer between Au NPs and extra framework SnOx does not affect the electron
structure of Au NPs significantly. Thus, similar peak profile of Au4f (87.3/83.6 eV) were
obtained for 1Au/DeAlꢀUSY and 1Au/3SnꢀDeAlꢀUSY.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To confirm the existence of positively charged Au further, FTꢀIR spectrum of 1Au/3SnꢀDeAlꢀ
USY with CO adsorption was collected (Figure S8). According to the previous literature,46ꢀ48 the
band at 2131 cmꢀ1 can be assigned to adsorption of CO on positively charged Au. This result
proves the existence of positively charged Au in 1Au/3SnꢀDeAlꢀUSY.
1Au/3SnꢀDeAlꢀUSY
Sn 3d5/2
Sn 3d3/2
486.6
495.0
495.6
496.2
487.8
487.2
Sn 3d5/2
3SnꢀDeAlꢀUSY
Sn 3d3/2
500
495
490
485
480
Binding energy (eV)
Figure 7. XPS spectra of Sn3d in 3SnꢀDeAlꢀUSY and 1Au/3SnꢀDeAlꢀUSY.
18
ACS Paragon Plus Environment

Page 19 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
Interaction between Au and Sn. Based on the results of TEM, FTꢀIR and XPS discussed
above, the interaction between Au and Sn in the sample can be confirmed. After dealumination,
few L acid site can be observed in DeAlꢀUSY sample. The formation of L acidic Sn sites in Snꢀ
DeAlꢀUSY is due to the tetrahedrally coordinated framework Sn4+ centers. When the Sn content
(for example 3 wt% or 4 wt%) is high, many Sn species can react with the SiꢀOH around the
mesopores to form extra framwork Sn sites (SnOx). The electron on Au surface is transferable. In
the loading process, Au NPs preferentially located at the extra framework SnOx around the
mesopores of USY and transferred some electrons into the SnOx. This intereaction depressed the
further growth and agregation of Au NPs in the loading process. Based on the results mentioned
above, we proposed a scheme for the preparation of 1Au/3SnꢀDeAlꢀUSY catalyst (Scheme 2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Formation of 1Au/3SnꢀDeAlꢀUSY catalyst
Catalytic Results. Effects of the synthesis conditions of the catalysts, including pH of the Au
gel, weight ratio of Au/PVA, Sn content and Au loadings, on the conversion of GLY to MLA
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 34
1
2
3
4
5
6
7
8
9
were investigated (Table 3). The pH value of Au gel affects the catalytic performance of the
catalysts significantly (Table 3, entries 1 to 5). When the pH value of Au gel was 3, the prepared
catalyst showed the best catalytic performance; 72% GLY conversion and 60% yield of MLA
were obtained. Both lower and higher pH value led to the decrease of GLY conversion and MLA
yield. For example, when the pH = 1, 66% GLY conversion and 49% MLA yield were obtained.
GLY conversion and MLA yield were 51% and 34% at the pH of 7. As mentioned above (Figure
1), when the pH = 1 or 7, average particles size of Au became much larger than that at pH = 3,
and this caused the lower catalytic activity of Au NPs. Conversion of GLY to MLA includes
oxidation of GLY to 1,3ꢀdihydroxyacetone (DHA) and/or glyceraldehyde (GA), and conversion
of DHA and/or GA to MLA. Acid properties especially the L acid sites are important for the
selective conversion of DHA and/or GA to MLA. The pH value also affects the mole ratio of L
acid sites to B acid sites (L/B ratio) of the prepared catalyst (Table S1). L/B ratio might influence
the selectivity of MLA. When pH = 1, L/B ratio of the sample (1.8) is much lower than that (4.6)
synthesized at pH = 3. Although these two samples have comparable amount of L acid sites (71.4
and 67.1 ꢁmol gꢀ1), MLA selectivity for pH =1 (74%) is lower than that for pH = 3 (83%). The
effect of Au/PVA weight ratio was also studied (Table 3, entries 6 to 10). In a broad range of 2 to
0.5, high GLY conversion (~80%) and high MLA yield (~70%) can be obtained. Much more
PVA used in the catalyst synthesis would be unfavorable for the catalytic performance. GLY
conversion and MLA yield were only 50% and 41% respectively when the Au/PVA weight ratio
was 0.25. This is consistent with the TEM results (Figure 2) that size of Au NPs was larger when
the Au/PVA weight ratio was 0.25. Furthermore, as shown in Table S1, all the L/B ratios of the
samples with different Au/PVA weight ratio (4, 1 and 0.25) are larger than 3. The MLA
selectivity for these samples is higher than 80%. It is known that Sn content affects the acid
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20
ACS Paragon Plus Environment

Page 21 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
property of the catalysts. The L/B ratios of the samples with 1 wt%, 2 wt%, 3 wt% and 4 wt% Sn
content are all higher than 3. When these samples were applied as catalysts, 85% to 90% MLA
selectivity was obtained except for 1Au/1SnꢀDeAlꢀUSY. For 1Au/1SnꢀDeAlꢀUSY, the MLA
selectivity is 73%. The L/B ratio of 1Au/1SnꢀDeAlꢀUSY is 5.8, but the total amount of L acid
sites is 53.3 ꢁmol gꢀ1. Considering other L acid sites generated by the residue Al, the introduction
of Au and the extra framework Sn, the amount of L acidic Sn sites might be not enough. The
importance of L acidic Sn sites can also be proved by the catalytic result (19% MLA selectivity)
of 1Au/DeAlꢀUSY (L/B ratio = 3.3). Thus, both adequate amount of L acidic Sn sites and high
L/B ratio are necessary to obtain a high MLA selectivity. Considering the effect of Sn on the
oxidation performance of Au which determined the conversion of GLY, 3 wt% Sn content was
proper for this study. Au is the active component for the oxidation of GLY to DHA and/or GA,
so we also studied the effect of Au amount on GLY conversion. In the absence of Au, no MLA
was detected (Table 3, entry 15). It means that no oxidation of GLY to DHA and/or GA
occurred, although 10% GLY conversion was observed. For 0.5Au/3SnꢀDeAlꢀUSY, 1Au/3Snꢀ
DeAlꢀUSY and 2Au/3SnꢀDeAlꢀUSY, GLY conversions are 25%, 81% and 84%, and MLA
yields are 22%, 69% and 68%. It can be seen that when the amount of Au increased from 0.5
wt% to 1 wt%, GLY conversion and MLA yield increased dramatically. However, further
increasing the amount of Au to 2 wt%, similar GLY conversion and MLA yield were observed
as that for 1Au/3SnꢀDeAlꢀUSY. So, the proper Au content is 1 wt% in this work. Thus, the
optimized synthesis conditions of the catalysts including pH of the Au gel (pH = 3), weight ratio
of Au/PVA (1), Sn content (3 wt%) and Au loadings (1 wt%) were obtained.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 3. Modification of the Synthesis Conditions of the Catalysts.a
Entry
pH
Au/PVA weight Sn content Au loading Conversion of Yield of MLA
ratio (wt%) (wt%) GLY (%) (%)
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 34
1
2
3
4
5
6
7
8
9
1
2
1
2
3
5
7
3
3
3
3
3
3
3
3
3
ꢀ
4
4
3
3
3
3
3
3
3
3
3
3
0
1
2
4
3
3
3
1
1
66
62
72
61
51
76
81
81
81
50
48
62
54
63
10
25
84
49
48
60
52
34
58
68
69
67
41
9
3
4
1
4
4
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
4
1
6
8
1
7
2
1
8
1
1
9
0.5
0.25
1
1
10
11
12
13
14
15
16
17
1
1
1
1
45
46
57
0
1
1
1
1
ꢀ
0
3
3
1
0.5
2
22
68
1
a Reaction conditions: 0.2 g of catalyst, 1 mmol of GLY, 5 mL of methanol, 0.5 MPa O2, 140 oC,
10 h.
22
ACS Paragon Plus Environment

Page 23 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
100
80
60
40
20
0
GLY conversion
MLA yield
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. Conversion of GLY to MLA catalyzed by various catalysts. Reaction conditions: 1
o
a
mmol of GLY, 5 mL of methanol, 0.2 g of catalyst, 0.5 MPa O2, 140 C, 10 h. Physical
mixture of 1Au/DeAlꢀUSY (0.2 g) and 3SnꢀDeAlꢀUSY (0.2 g).
To get more information about the effect of Sn on the performance of 1Au/3SnꢀDeAlꢀUSY
catalyst, reactions over various catalysts including HꢀUSY, 3SnꢀDeAlꢀUSY, 1Au/HꢀUSY,
1Au/DeAlꢀUSY, 1Au/3SnꢀDeAlꢀUSY and physical mixture of 1Au/DeAlꢀUSY and 3SnꢀDeAlꢀ
USY were carried out (Figure 8). As anticipated, only 1% GLY was converted and no MLA was
observed without catalyst. No MLA formed with HꢀUSY or 3SnꢀDeAlꢀUSY as catalyst, although
15% or 10% conversion of GLY was obtained, respectively. In our previous report we proved
that SnꢀDeAlꢀUSY is an excellent catalyst for the conversion of DHA to MLA.28 Thus, it is
confirmed that Au plays a crucial role for the oxidation of GLY to DHA and/or GA. When
1Au/HꢀUSY and 1Au/DeAlꢀUSY were used as catalysts, 45% and 48% GLY conversion were
obtained. However, MLA yields were only 14% and 9% respectively. In contrast, 1Au/3Snꢀ
DeAlꢀUSY exhibited remarkable activity and selectivity for the conversion of GLY to MLA, and
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 34
1
2
3
4
5
6
7
8
9
81% GLY conversion with 69% MLA yield was obtained under the same reaction conditions as
that for 1Au/HꢀUSY and 1Au/DeAlꢀUSY. Physical mixture of 1Au/DeAlꢀUSY and 3SnꢀDeAlꢀ
USY was also applied as catalyst for this reaction, 38% GLY conversion and 13% MLA yield
were obtained. Thus, introduction of Sn promotes both the conversion of GLY and the selectivity
of MLA. First, GLY conversion is determined by the catalytic activity of Au. The presence of Sn
promoted the formation of small and uniform Au particles due to the interaction between Au and
extra framework Sn. It is known that Au particle size is important for its catalytic activity.
Secondly, introduction of Sn increased the L acid sites dramatically which are very beneficial for
the conversion of DHA to MLA. So, introduction of Sn is critical for the high GLY conversion
and MLA yield.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
100
GLY conversion
MLA yield
80
60
40
20
0
100
120
140
160
180
Reaction temperature (oC)
Figure 9. Catalytic performance of 1Au/3SnꢀDeAlꢀUSY at different reaction temperature.
Reaction conditions: 1 mmol of GLY, 5 mL of methanol, 0.2 g of catalyst, 0.5 MPa O2, 10 h.
Effects of the reaction conditions including reaction temperature, reaction time, substrate
concentration, catalyst amount and initial O2 pressure were also investigated. Figure 9 shows the
catalytic performance of 1Au/3SnꢀDeAlꢀUSY at different reaction temperature. With the
24
ACS Paragon Plus Environment

Page 25 of 34
ACS Catalysis
1
2
o
3
4
elevation of the reaction temperature from 100 to 140 C, GLY conversion increased sharply
5
6
from 12% to 81% and MLA yield increased from 12% to 69%. Further elevated the temperature
7
8
9
o
to 180 C, both GLY conversion and MLA yield increased gradually. When the reaction
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
o
temperature was 160 C or 180 C, the same MLA yield was obtained (79%). However, the
o
o
MLA selectivity obtained at 180 C was lower than that at 160 C. High reaction temperature
would lead to the formation of overꢀoxidation products, such as methyl glycerate (MGA). This
would be further discussed in the following section.
100
80
GLY conversion
MLA yield
MGA yield
60
40
20
0
0
2
4
6
8
10
Reaction time (h)
Figure 10. Time course of the reaction over 1Au/3SnꢀDeAlꢀUSY. Reaction conditions: 1
mmol of GLY, 5 mL of methanol, 0.2 g of catalyst, 0.5 MPa O2, 160 oC.
To understand the reaction mechanism over Au/SnꢀUSY catalyst, time courses of the reactants
and products in the presence of 1Au/3SnꢀDeAlꢀUSY are provided in Figure 10. 1Au/3SnꢀDeAlꢀ
USY shows high activity, 80% GLY conversion can be obtained in 1 h. Further increasing the
reaction time to 10 h, GLY conversion increased to 88%. Meanwhile, with the increase of time,
MLA yield also increased sharply in first 1.5 h, and then increased slowly in the following
reaction time. As a byꢀproduct, MGA was formed due to overꢀoxidation. 4% MGA yield was
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 34
1
2
3
4
5
6
7
8
9
obtained in first 1 h. Prolonged the reaction time to 10 h, 7% MGA yield achieved. When the
reaction time is 10 h, 88% GLY conversion, 79% MLA yield and 7% MGA yield were obtained,
and only 2% other unknown byꢀproduct formed. However, shorter reaction time caused lower
MLA selectivity, and MGA yield was minor. This is a consecutive reaction including oxidation
of GLY to intermediate species (DHA and/or GA) and further conversion to MLA. Because of
the high reaction temperature, no DHA and/or GA was detected in the products even in the first
0.5 h. Thus, some other unknown intermediate species formed from DHA and/or GA in this
process. Furthermore, the unknown intermediate species can also be converted to DHA and/or
GA reversibly when prolonged the reaction time. To confirm this assumption, DHA was applied
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
as reactant at 160 C under 0.5 MPa O2 without catalyst. After 0.5 h, 92% DHA conversion and
9% MGA yield were obtained. No other product was detected. This result implied that, under
160 oC, DHA was converted to the unknown intermediate species mentioned above which might
be the condensation or polymerization products. This would be investigated in detail in the future
work.
Effect of reactant concentration was studied (Figure S9). As anticipated, with the increase of
GLY concentration, both the GLY conversion and the MLA yield decreased. Effect of catalyst
amount was also investigated (Figure S10). Compared to 0.2 g of 1Au/3SnꢀDeAlꢀUSY, slight
decrease of GLY conversion and MLA yield was observed when 0.1 g of catalyst was applied.
Further decreasing the catalyst amount to 0.05 g, GLY conversion and MLA yield decreased
significantly. Initial O2 pressure has little impact on the catalytic performance of the catalyst
(Figure S11).
The yield and the selectivity of MLA are higher than those described in previous report.27 To
the best of our knowledge, this is the first example of such a high MLA selectivity (~90%) with a
26
ACS Paragon Plus Environment

Page 27 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
high MLA yield (> 75%) for the conversion of GLY to MLA directly. TOF value of 1Au/3Snꢀ
DeAlꢀUSY was tested (Table 4). To avoid the mass transfer limitations, the test was carried out
with a small amount of catalyst (0.05 g) and high stirring speed. After 0.5 h reaction time, the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ꢀ1
MLA yield was 12% which corresponds to a TOF of 94.6 molMLA molAu hꢀ1. For comparison,
TOFs of 1Au/HꢀUSY and 1Au/DeAlꢀUSY were also tested. However, no MLA was detected at
the same reaction conditions.
Table 4. Catalytic Activity of 1Au/HꢀUSY, 1Au/DeAlꢀUSY and 1Au/3SnꢀDeAlꢀUSYa
TOF
Entry
Sample
GLY conversion (%) MLA yield (%)
(molMLA molAuꢀ1 hꢀ1)
1
2
3
1Au/HꢀUSY
1Au/DeAlꢀUSY
7.5
4.6
0
0
0
0
1Au/3SnꢀDeAlꢀUSY
14.4
12
94.6
a Reaction conditions: 1 mmol of GLY, 5 mL of methanol, 0.05 g of catalyst, 0.5 MPa O2, 160
oC, 0.5 h.
Catalyst Stability. Stability of the prepared catalyst was investigated. Figure 11 shows the
XRD patterns of 1Au/3SnꢀDeAlꢀUSY before and after reaction. Crystalline structure of the
zeolite did not alter significantly. So, the crystalline structure of the catalysts is stable.
o
Furthermore, GLY conversion was 26% after 1 h under 160 C when the 1Au/3SnꢀDeAlꢀUSY
was used as catalyst. Then, the catalyst was removed from the reaction by hotꢀfiltration. After
reaction for another 1 h, the conversion of glycerol did not increase dramatically (27%). Thus,
this reaction is heterogeneous and there is no leaching of Au NPs in the reaction.
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 34
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(a) Before reaction
(b) After reaction
5
10
15
20
25
30
35
40
2Theta (o)
Figure 11. XRD patterns of 1Au/3SnꢀDeAlꢀUSY (a) before and (b) after the reaction.
We further examined the recyclability of 1Au/3SnꢀDeAlꢀUSY (Figure 12). 77% MLA yield
with 88% GLY conversion was obtained in the second run, which is comparable to the first run
(79% MLA yield and 88% GLY conversion). However, for the third run, 66% MLA yield and
81% GLY conversion were given. Figure 13 shows TEM images of 1Au/3SnꢀDeAlꢀUSY before
and after reaction. It can be seen that the average size of Au NPs increased slightly after reaction.
So, the decrease of GLY conversion might be caused by the aggregation of Au NPs and/or the
exfoliation of smaller Au NPs from the support. To explain the decrease of the MLA selectivity,
the Sn content in the reaction solution after reaction was tested by ICP, only 0.0037 wt% Sn
leaching was observed. Furthermore, the acid properties of the used catalyst after 2 runs were
measured (Figures S12 and S13 and Table S3). After reaction, the amount of L acid sites
decreased, especially the framework Sn sites. This might be caused by the strong chemisorption
of chemicals on the framework Sn sites, the active sites for the conversion of DHA to MLA. So,
the recyclability of the catalyst would be further improved in the future study.
28
ACS Paragon Plus Environment

Page 29 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
100
80
60
40
20
0
GLY conversion
MLA yield
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2
Run
3
Figure 12. Reuse test of 1Au/3SnꢀDeAlꢀUSY in the conversion of GLY to MLA. Reaction
conditions: 1 mmol of GLY, 5 mL of methanol, 0.2 g of catalyst, 0.5 MPa O2, 160 oC, 10 h.
Figure 13. TEM images of 1Au/3SnꢀDeAlꢀUSY (a) before and (b) after the reaction.
CONCLUSIONS
In summary, we developed hierarchical bifunctional catalysts of SnꢀUSY supported Au
nanoparticles for oneꢀpot conversion of glycerol to methyl lactate. High methyl lactate yield can
be obtained with a high selectivity (~90%). Interaction between Au and extra framework SnOx
around the mesopores of DeAlꢀUSY promotes the dispersion of Au particles (smaller and much
more uniform), which is important for the oxidation of glycerol to 1,3ꢀdihydroxyacetone and
29
ACS Paragon Plus Environment