Elucidation of the Roles of Re in Aqueous-Phase Reforming of Glycerol over Pt-Re/C Catalysts

Article abstract of DOI:10.1021/acscatal.5b01770

We report the investigation of surface properties of Pt/C and Pt-Re/C catalysts using in situ spectroscopic tools and a fundamental understanding of their catalytic performances in glycerol aqueous-phase reforming (APR). We found that adding Re to Pt/C improves its activity significantly, which is consistent with our previous observation in glycerol steam reforming. However, the difference in reaction selectivity is much more pronounced in APR. Compared to Pt/C, Pt-Re/C yielded significantly more liquid products, while the selectivity to H₂ and CO₂ decreased by more than 40%. In operando X-ray absorption spectroscopy and attenuated total reflectance infrared (ATR-IR) with in situ capability as well as Raman spectroscopy were employed to investigate the catalyst surface properties and the roles of Re. In operando X-ray absorption fine structure shows significant oxidation of Re, and ATR-IR of adsorbed pyridine revealed the formation of acid sites on PtRe after hot liquid water treatment. Additionally, ATR-IR using CO as a probe molecule demonstrated that desorption of CO from the Pt-Re/C surface is more facile than that from Pt/C in the aqueous phase. We propose that well-dispersed Re oxide species in the proximity of Pt work as the active sites, providing both metal and acid functionalities, while the terminal Re-O moiety has little contribution to the overall reactivity as evidenced by Raman spectroscopy.

Full text of DOI:10.1021/acscatal.5b01770

Subscriber access provided by EPFL | Scientific Information and Libraries
Article
Elucidation of the roles of Re in aqueous-phase
reforming of glycerol over Pt-Re/C catalysts
Zhehao Wei, Ayman M. Karim, Yan Li, and Yong Wang
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.5b01770 • Publication Date (Web): 27 Oct 2015
Downloaded from http://pubs.acs.org on November 2, 2015
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 22
ACS Catalysis
1
2
3
4
5
6
Elucidation of the roles of Re in aqueous-phase reforming of
glycerol over Pt-Re/C catalysts＃
†
‡
*
Zhehao Weia,b , Ayman Karimb , Yan Lia, Yong Wanga,b*
7
8
9
a The Gene and Linda Voiland School of Chemical Engineering and Bioengineering, Washington State University, Pullman,
WA 99164, USA
b Institute for Integrated Catalysis, Pacific Northwest National Laboratory, Richland, WA 99352, USA
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
ABSTRACT: We report the investigation of surface properties of Pt/C and PtꢀRe/C catalysts using in situ spectroscopic tools and
the fundamental understanding of their catalytic performances in glycerol aqueousꢀphase reforming (APR). We found that adding
Re to Pt/C improves its activity significantly, which is consistent with our previous observation in glycerol steam reforming. Howꢀ
ever, the difference in reaction selectivity is much more pronounced in APR. Compared to Pt/C, PtꢀRe/C yielded significantly more
liquid products, while the selectivity to H2 and CO2 decreased by more than 40%. Inꢀoperando Xꢀray absorption spectroscopy and
attenuated total reflectance infrared (ATRꢀIR) with in situ capability as well as Raman spectroscopy were employed to investigate
the catalyst surface properties and the roles of Re. Inꢀoperando Xꢀray absorption fine structure (XAFS) shows significant oxidation
of Re, and ATRꢀIR of adsorbed pyridine revealed the formation of acid sites on PtRe after hot liquid water treatment. Additionally,
ATRꢀIR using CO as probe molecule demonstrated that CO desorption from the PtꢀRe/C surface is more facile than that from Pt/C
in aqueous phase. We propose that wellꢀdispersed Re oxide species in close proximity with Pt work as the active site providing both
metal and acid functionalities while the terminal ReꢀO moiety has little contribution to the overall reactivity as evidenced by Raman
spectroscopy.
KEYWORDS: aqueousꢀphase reforming, bimetallic catalyst, platinumꢀrhenium, rhenium oxide, catalysis in condensed waꢀ
ter, surface acidity, selectivity, in situ and inꢀoperando spectroscopy
＃ Invited submission to the ACS Catalysis special issue on Catalysis at U.S. Department of Energy National Laboratories
____________________
Received on 08/12/2015
Revised on 10/12/2015 and 10/22/2015
Accepted on xx/xx/xxxx
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 22
1. Introduction
were prepared by incipient wetness impregnation. The imꢀ
1
2
3
4
5
6
7
8
pregnated samples were dried at 110 °C for 2 h and calcined at
260 °C for another 2 h. Bimetallic PtꢀRe/C was prepared by
impregnating Re precursor solution onto the calcined Pt/C
sample followed by the same drying and calcination steps.
2.2. Catalyst characterization
Glycerol has been reported to be a good model compound in
biomass catalytic conversion, with functionality that benefits
the mechanistic understandings. Cortright et al.[1] first conꢀ
firmed the feasibility of performing glycerol aqueousꢀphase
reforming (APR) with high hydrogen yield using a benchmark
Pt/Al2O3 catalyst. Wen et al. screened a few catalysts and
came up with the order of hydrogen production rate, in terms
of active metal, of Pt > Cu > Ni > Co[2]. Extensive research
has confirmed that adding a second metal (e.g., Mo[3, 4],
Ni[5, 6], or Re[7ꢀ12]) to Pt catalysts can enhance the catalytic
activity, and some new insights into the role of these second
metals have also been proposed. Dietrich et al.[4] observed 4
times higher turnover frequency (TOF) over PtMo/C and
higher selectivity to CꢀO bond cleavage than that over Pt/C,
and ascribed the role of Mo to altering the electronic properꢀ
ties of Pt, lowering the binding energy of CO and reducing the
activation energy of both dehydrogenation and C–O bond
cleavage. Inꢀoperando characterization indicated that the cataꢀ
lyst surface is mainly composed of Ptꢀrich bimetallic PtMo
nanoparticles and oxidized Mo[3]. We compared the perforꢀ
mances of Pt/C and PtꢀRe/C catalysts in glycerol APR and
found a higher TOF of about one order of magnitude over Ptꢀ
Re/C, along with higher selectivity to dehydration products
over PtꢀRe/C[8]. In situ characterizations using Xꢀray photoeꢀ
lectron spectroscopy (XPS) and ammonia temperature proꢀ
grammed desorption (NH3ꢀTPD) confirmed that Re becomes
oxidized, creating surface acidity upon hydrothermal treatment
(in this case steam treatment); the concentration of acid sites
was found to increase with Re addition, favoring C–O over C–
C cleavage while hydrogen selectivity was decreased[10].
However, the effect of liquid phase hydrothermal conditions
on the catalyst structure is still not well understood. Therefore,
it is worthwhile to separately investigate these two cases, i.e.,
vapor phase steam and aqueous phase. Our most recent invesꢀ
tigation[12] on steam treatment of PtꢀRe catalyst and its perꢀ
formances in steam reforming of glycerol indicated that upon
steam treatment, oxidized Re species could facilitate CO spillꢀ
over and desorption from neighboring Pt sites, making them
accessible to glycerol reactant, and leading to the overall enꢀ
hanced activity while selectivity to products remains the same.
In the present work, we extend our studies to elucidating the
roles of Re on PtꢀRe/C bimetallic catalyst in aqueous phase
reforming of glycerol. We particularly focus on understanding
the effects of water phase (i.e., condensed phase vs vapor
phase) on the structure and surface properties of PtꢀRe, and
correlating these properties with catalyst activity and selectiviꢀ
ty in glycerol APR reaction. In situ techniques such as Xꢀray
absorption fine structure (XAFS), attenuated total reflectance
infrared (ATRꢀIR), and Raman spectroscopy are used to charꢀ
acterize Pt/C and PtꢀRe/C catalysts to help understand their
properties under reaction conditions.
Attenuated total reflectance infrared (ATRꢀIR) characterizaꢀ
tion of the samples was performed using a homeꢀmade ATR
cell with the details previously reported[12]. The catalyst
sample was ground into fine powder and suspended in nanoꢀ
pure water under sonication to form an ink. This ink was then
coated onto the round top surface of the IRE and dried overꢀ
night to form a thin layer. The ATR cell can be heated to up to
300 °C by a cartridge heating element integrated with a Watꢀ
low temperature controller. A HPLC pump (LabAlliance Seꢀ
ries III) was used to deliver pressurized liquid into the cell. A
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
backpressure
regulator
(BPR,
KPB1L0A412P20000,
Swagelok) was used to control the system pressure downꢀ
stream.
The sample was reduced in 10% H2/Ar (10 mL/min) from
room temperature to 280 °C at 5 °C/min. After 1 h reduction,
the gas was switched to 10 mL/min He and purge for 10 min
before cooling down. The sample was held at room temperaꢀ
ture for at least 10 min before collecting a background specꢀ
trum. 10 mL/min 1% CO/He was then introduced and a series
of sample scans were taken every minute. The resolution of
each scan was set at 4 cmꢀ1 and 128 scans were averaged to
generate each spectrum. For the CO desorption measurement
as a function of temperature, a series of spectra under He at
200, 150, 100, 50 and 25 °C were collected before introducing
CO and used as the background spectra for the desorption
counterparts. For experiments requiring high pressure liquid
environment, liquid was delivered by the HPLC pump at a
certain flow rate (typically 0.1 mL/min if not otherwise speciꢀ
fied; catalyst layer was intact following high pressure water
flow) after the cell was pressurized with N2 and the backpresꢀ
sure set at 450 psi by the BPR. CO adsorption under liquid
environment was performed by coꢀfeeding CO with liquid
water (0.01 mmol CO/L H2O). CO desorption was carried out
by switching to pure water flow. The IR data collected in the
presence of liquid water was corrected by subtraction of a
scaled water background spectrum according to a procedure
described elsewhere[13].
Raman spectra of the catalyst samples were collected with a
Horiba LabRAM HR Raman microscope system reported beꢀ
fore[14]. The system is equipped with a 532 nm laser source
(Ventus LP 532) and a Synapse CCD (charge coupled device)
detector. Catalysts were loaded in an in situ reaction cell
(Linkam CCR1000) for Raman characterization. Sample reꢀ
duction was carried out under 10% H2/Ar (20 mL/min) at 280
°C for 1 h. For those requiring hot liquid water (HLW) expoꢀ
sure, the samples were first treated in a batch reactor in the
similar manner (hydrogen reduction followed by exposure to
water at APR reaction condition without feeding glycerol) as
described in section 2.3. After 2 h, the catalyst samples were
collected and immediately transferred to the reaction cell unꢀ
der N2 protection to minimize exposure to air. The samples
were then dried under He flow (20 mL/min) at 110 °C for 2 h.
The catalysts were probed by Xꢀray absorption spectroscopy
(XAS) in transmission mode using an inꢀhouse built inꢀsitu
cell rated to 1100 psi and 700 °C, as previously reported [12,
2.
Experimental
2.1. Catalyst preparation
The catalyst preparation procedure has been reported previꢀ
ously[12]. Briefly, active carbon support (TA 60, PICATAL,
60~100 mesh) was dried at 110 °C overnight prior to use.
Tetraammineplatinum(II)
nitrate
hexahydrate
((NH3)4Pt(NO3)2·6H2O, 99.999%, SigmaꢀAldrich) and perꢀ
rhenic acid (HReO4, 99.99%, in 65ꢀ70% aqueous solution,
SigmaꢀAldrich) were used as metal precursors. Pt/C and Re/C
2
ACS Paragon Plus Environment

Page 3 of 22
ACS Catalysis
15]. It consists of a flow reactor and the cell body made of catalyst, the reactor was flushed with Ar to remove air. Then it
1
2
3
4
5
6
7
8
stainless steel and heated by resistive heating cartridges. An
Xꢀray window (1 cm horizontal × 0.5 cm vertical) allows for
measurements in transmission mode. The catalyst bed (about
60 mg, 60~100 mesh) was packed with quartz wool on both
ends in a glassy carbon tube (o.d. = 6.35 mm, i.d. = 4.1 mm,
from HTW HochtemperaturꢀWerkstoffe GmbH, Germany).
and connected to the rest of the system via standard Swagelok
fittings using graphite ferrules (450 °C temperature rating).
The reaction temperature was controlled using a Watlow temꢀ
perature controller within ±1 °C, measured on the external
tube wall using a K type thermocouple. The XAS measureꢀ
ments were performed at beamline Xꢀ18A at the National
Synchrotron Light Source (NSLS) operated by the Synchroꢀ
tron Catalysis Consortium (SCC) at Brookhaven National
Laboratory (BNL). The measured Pt and Re absorption edge
steps for Pt and Re for the catalysts were about 0.65.
was pressurized with 10% H2/Ar and depressurized for three
times before a final pressurization to 20 psi. After the catalyst
was in situ reduced at 280 °C for 1 h and cooled down to room
temperature, the gas inside was vented and 20 mL 10% glycꢀ
erol aqueous solution was injected into the vessel by a syringe
without opening or exposing the reduced catalyst to air. After
pressurizing and depressurizing the reactor with N2 for three
times to remove residual H2/Ar, a final charge of N2 to 80 psig
was applied. The mixture was stirred at 300 rpm for 10 min
before heating; after reaching the reaction temperature, the
reaction was run under vigorous stirring at 600 rpm for 2 h
before quenching the reactor vessel in an ice bath. The inner
pressure at 225 °C reached above 450 psi, as typically required
for APR reaction. Under such reaction conditions, no signifiꢀ
cant mass transfer limitation is observed (i.e., no change in
activity at a higher stirring rate or smaller catalyst particle
size). Product analysis followed the same procedure reported
before[12]. Briefly, the gas products were collected in a Tedlar
gas bag (1 L, SigmaꢀAldrich) and analyzed offline with an
Agilent 490 micro GC equipped with four independent modꢀ
ules (a molsieve column, a CPꢀPoraPLOT U column, an aluꢀ
minum oxide column, and a CPꢀSil 5CB column, all with
TCDs; the molsieve used Ar as reference gas while all other
modules used He). After depressurizing and opening the vesꢀ
sel, liquid product was separated from solid catalyst by a 0.2
ꢂm syringe filter and analyzed using a Waters Breeze 2 HPLC
equipped with a BioꢀRad Aminex HPXꢀ87H ion exclusion
column (300 mm × 7.8 mm) and a refractive index detector.
The mobile phase was 5 mM H2SO4 aqueous solution at a flow
rate of 0.6 mL/min. The carbon balance is generally higher
than 90%.
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
Xꢀray absorption near edge structure (XANES) and Extended
Xꢀray absorption fine structure (EXAFS) data processing and
analysis were performed using Athena and Artemis programs
of the IFEFFIT data analysis package[16, 17]. Three scans
were collected under each condition and merged after alignꢀ
ment using Pt (or Re) foil spectrum collected simultaneously
for each scan. Scans were collected at the Pt L2, 3 and Re L3
edges. For the PtꢀRe catalyst, due to the overlap of the Re L2
(11959 eV) with the Pt L3 (11564 eV) which limits the
EXAFS signal to 9.6 Åꢀ1 in kꢀspace, we collected the EXAFS
signal at the Pt L2 edge which extends the kꢀrange to 12.1 Åꢀ1.
χ(k) (where k is the photoelectron wavenumber) was obtained
by subtracting smooth atomic background from the normalꢀ
ized absorption coefficient using the AUTOBK code. The
theoretical EXAFS signal was constructed using the FEFF6
code and fitted to the data in rꢀspace using the Artemis proꢀ
gram of the IFEFFIT package. The spectra were fitted in rꢀ
space by varying the coordination number (CN) of the single
scattering Pt–M (Re–M or ReꢀO) path, the bond length disorꢀ
der (Debye–Waller factor), σ2, the effective Pt–M (ReꢀM or
ReꢀO) scattering length, and the correction to the threshold
3.
Results and discussion
3.1. Catalytic reactivity in glycerol aqueousꢀphase reꢀ
forming
Batch reaction of glycerol APR was carried out over Pt/C and
PtꢀRe/C catalysts. The catalyst loading (50 mg of 3%Pt/C and
20 mg of 3%Ptꢀ3%Re/C) was varied to obtain similar glycerol
conversion (around 15%). The turnover frequency (TOF) was
calculated as either the consumption rate of glycerol or the
formation rate of each product divided by the Pt dispersion
measured in our previous report using the same catalysts[12].
We also note here that APR reaction is known to be catalyst
structure sensitive[20]. However, the Pt particle sizes in this
study are mostly 1~2 nm as we previously observed[10]. Acꢀ
cording to Kirilin et al.[20], within this narrow particle size
range the catalyst structure sensitivity for APR reaction is
negligible.
As we have previously reported[8, 10], the catalytic perforꢀ
mances of Pt/C and PtꢀRe/C in glycerol APR reaction are sigꢀ
nificantly different. Glycerol TOF is more than 5 times higher
over 3%Ptꢀ3%Re/C than over 3%Pt/C, as can be seen in Figꢀ
ure 1. Similar enhancement in catalytic activity with addition
of Re to Pt/C has been reported before[7, 8, 10ꢀ12]. Additionꢀ
ally, the product distributions are significantly different. As
shown in Table 1, the value XG/XL, representing the ratio beꢀ
tween conversion to gas products and liquid products, was 0.9
for 3%Pt/C and 0.4 for 3%Pt/ꢀ3%Re/C, indicating that producꢀ
tion of liquid products was more facile over PtꢀRe/C. From the
results shown in Figure 1, it can be observed that the addition
of Re causes a significant shift from the production of the gas
2
energy, ꢁE0. S0 (passive electron reduction factor) was obꢀ
tained by analyzing the spectrum for Pt and Re foils, and the
best fit value (0.93 for Pt L2, and 0.86 for both Pt L3 and Re
L3) was fixed during the fitting of the catalyst samples. The
values for S02 are consistent with our previous work[5, 15] and
values reported in the literature[18, 19]. Re consisted of Ptꢀ
Re/C sample contained a mixture of metallic Re and ReOx and
fitting the spectra to a single ReꢀO scattering path did not reꢀ
sult in good fits. We used a similar model to the one proposed
by Bare et al.[18] for Re/Al2O3 where they used two ReꢀO
scattering paths at about 1.73 Å and 2.08 Å. The model with
two ReꢀO scattering paths gave a much better statistical fit
(reduced χ2 was lower by a factor of 5ꢀ10) than the model with
a single ReꢀO scattering path. The kꢀrange used for Fourier
transform of the χ(k) was 2.5–12 Åꢀ1 for Pt and 2.5ꢀ13.9 Åꢀ1
for the Re edge and the rꢀrange for fitting was 1.8–3.3 Å for Pt
and 1.1–3.2 Å for Re unless otherwise specified.
2.3. Catalyst evaluation in batch reactor for glycerol
APR
The catalyst sample was ground and sieved to 325~400 mesh
before loading into the reactor for evaluation. A 50 mL micro
stirred reactor purchased from Parr Instrument was used for
this study. Typically, after the charge of a certain amount of
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 22
products H2 and CO2 to the production of the liquid products,
such as acetol and ethanol.
1
2
3
4
5
6
7
8
9
Table 1 Glycerol APR over Pt and PtꢀRe catalysts supported on active carbon (reaction temperature 225 °C, operated at
initial pressure at room temperature of 80 psi N2); catalysts were in situ reduced under 20 psi initial pressure at room temꢀ
perature of 10% H2/Ar at 280 °C for 1 h. 20 mL 10 wt.% glycerol aqueous solution was the reactant. Catalyst weight was 50
mg of 3%Pt/C and 20 mg of 3%Ptꢀ3%Re/C. The stirring rate was fixed at 600 rpm.
Gas product composition
Liquid product distribution (mol%)b
(mol% dry basis)a
XG
(%)
Catalyst
XL (%)
XG/XL
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
C2+
H2
CH4 CO2
LA
14.4 2.7 13.3 15.1
17.1 4.5 6.1 24.2
AA EG
acetol PG
45.3 1.2
35.3 0.8
MeOH EtOH PA
3%Pt/C
6.2
5.4
65.4 1.3
54.0 3.8
30.4 3.8
6.7
6.9
9.3
1.1 0.9
2.7 0.4
3%Ptꢀ3%Re/C
30.6 11.5
13.4
XG, XL: conversion to gas and liquid products, respectively; LA: lactic acid; AA: acetic acid; EG: ethylene glycol; PG: propylene
glycol; PA: 1ꢀpropanol.
C2+
a Gas product is mainly composed of H2, CH4, CO and CO2, C2H6 and C3H8.
includes C2H4, C2H6 and C3H8.
b
Experimental data shown in this table was obtained at total carbon loss within 10%, most of which was caused by unknowns in
liquid products.
Table 2 EXAFS fitting results for the Pt/C and PtꢀRe/C catalysts at the Pt L2 and Re L3 edges. The catalysts were reduced
at 300 °C in 100% H2 and the EXAFS spectra were collected at 225 °C under 100% H2 flow and atmospheric pressure or
during APR at 500 psig and 225 °C.a The numbers in parentheses indicate the statistical error in the most significant digit
obtained from the fit in Artemis (e.g., 6.2(8) ≡ 6.2±0.8).
Condition
(at 225 °C
Absorberꢀ
backscatterer
pair
Edge Sample
CN
R (Å)
σ2 (Å2)
ꢁE0 (eV) Reduced χ2
)
H2
PtꢀPt
8.5(9)
2.749(7) 0.012(1) 5.6(8)
25
49
86
43
29
Pt/C
Pt L2
APR
H2
PtꢀPt
10.4(1.3) 2.759(8) 0.012(1) 5.7(9)
PtꢀM
7.0(8)
9.1(1.1)
6.2(8)
1.1(3)
1.2(4)
8.9(9)
0.9(2)
2.6(5)
2.735(9) 0.009(1) 6.2(7)
2.750(7) 0.010(1) 6.3(9)
2.677(4) 0.013(1) 3.2(1.5)
PtꢀRe/C
APR
PtꢀM
ReꢀM
ReꢀO1
ReꢀO2
ReꢀM
ReꢀO1
ReꢀO2
H2
1.74(1)
2.00(1)
0.006(2) 1.0(1.5)
0.006(2) 1.0(1.5)
Re L3 PtꢀRe/C
2.716(6) 0.009(1) 5.3(1.0)
97
APR
1.75(2)
2.00(1)
0.007(2) 5.3(1.0)
0.007(2) 5.3(1.0)
a No significant difference in the catalyst performance or structure (XAFS) was found by increasing the reduction temperature from 280 to
300
°
C or by increasing the H2 partial pressure from 10% to 100% (not shown).
4
ACS Paragon Plus Environment

Page 5 of 22
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
Figure 1. TOF of glycerol (minꢀ1) and product selectivity for
Figure 2. CO adsorption at room temperature (time interval = 1.5
min) on Pt/C under (a) gas and (b) aqueous phase
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
APR of 20 mL 10 wt.% glycerol aqueous solution over
3%Pt/C and 3%Pt–3%Re/C at 225 °C. Loaded catalyst weight
was 50 mg of 3%Pt/C and 20 mg of 3%Ptꢀ3%Re/C to achieve
similar conversion level (<20%). H2 selectivity is defined as
the ratio of H2 produced to the maximum H2 that could be
produced from the converted glycerol. Selectivity to other
carbonꢀcontaining products (in both gas and liquid phases) are
normalized and calculated based on moles of carbon in the
product
The spectra of CO adsorption on Pt/C under dry and high
pressure water environments are compared in Figure 2 as an
example. In Figure 2(a), linearly adsorbed CO on Pt0 site was
initially observed at 2030 cmꢀ1, which gradually shifted to
2035 cmꢀ1 at saturation level due to dipoleꢀdipole coupling
caused by the increasing surface coverage[21]. In Figure 2(b),
in the presence of high pressure water, linearly adsorbed CO
on Pt0 site was detected at 2004 cmꢀ1 and gradually shifted to
2009 cmꢀ1, due to the same dipoleꢀdipole coupling[21]. The
slower profile development of CO adsorption spectrum in
Figure 2(b) is because of a much lower CO concentration. The
redshift of CO adsorption peak and about a factor of four specꢀ
trum intensity enhancement from gas to aqueous phase are
consistent with observations by Ebbesen et al. and can be atꢀ
tributed to the direct effect of water on the transition dipole
moment of the adsorbed CO[21]. Such redshift of CO adsorpꢀ
tion and spectrum intensity enhancement from gas to aqueous
phase were also observed at above room temperature of CO
adsorption (Figure 3).
3.2. Catalyst properties in aqueousꢀphase environꢀ
ment
3.2.1 CO adsorption
We recently reported on the in situ ATRꢀIR characterizations
of Pt/C and PtꢀRe/C catalyst surfaces under reducing and
steam environments, and found that CO desorption is signifiꢀ
cantly more facile on PtꢀRe/C after steam treatment[12]. In
this work, we further explored the utilization of this technique
and extended its capability to in situ high pressure liquid enviꢀ
ronment, simulating APR reaction condition.
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 22
Figure 5. Relative integrated peak area as a function to desorption
temperature of Pt/C and PtꢀRe/C in gas or aqueous phase
1
2
3
4
5
6
7
8
9
As shown in Figures 4, a CO adsorption peak shift of 25 cmꢀ1
was also observed on PtꢀRe/C between gas and aqueous phase
treatment, which is caused by the presence of condensed waꢀ
ter[21]. The intensity of spectra over PtꢀRe/C catalyst is about
half that of Pt/C for both gas and aqueous phases (Figures 3
and 4), which is consistent with our previous report that Pt
dispersion is decreased by half with the addition of Re to
Pt/C[12]. CO adsorption on Pt/C and PtꢀRe/C under gas or
aqueous phase displayed different vibrational features and the
adsorption peak maxima at 25 °C were centered at 2035, 2009,
2031, and 2006 cmꢀ1 for Pt/C gas, Pt/C aqueous, PtꢀRe/C gas,
and PtꢀRe/C aqueous respectively. These features can all be
ascribed to linear CO adsorption (COL) on Pt site[22], while
CO adsorbed on bridge or multiꢀfold sites was negligible. Upꢀ
on TPD, the peak maxima redꢀshifted and the peak areas deꢀ
creased. As shown in Figure 5, for Pt/C, the relative integrated
peak area of CO adsorption in aqueous phase dropped by less
than 10% at each temperature compared to its gas phase adꢀ
sorption profile. However, for PtꢀRe/C the amount of CO adꢀ
sorption in aqueous phase decreased by about half at 50 °C
and almost dropped to zero around 180ꢀ200 °C. The compariꢀ
son of CO TPD in the gas phase shows that CO is more
strongly adsorbed on PtRe than on Pt. On the other hand, unꢀ
der aqueous phase conditions the results are analogous to our
previous observation on steamꢀtreated samples where the CO
desorption from catalyst surface was more facile on PtRe
compared to Pt/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
Figure 3. CO adsorption at different temperatures on Pt/C in (a)
gas and (b) aqueous phase
Figure 4. CO adsorption at different temperatures on PtꢀRe/C in
(a) gas and (b) aqueous phase
Figure 6. XAFS spectra of Re L3 edge under APR and reducꢀ
tion conditions
We previously reported the investigation of the electronic and
structural properties of the catalysts by XANES and EXAFS
under different gas environments[12]. We showed that the
formation of PtꢀRe alloy results in an insignificant charge
transfer between Pt and Re. From the XANES spectra, we
showed that the alloy formation results in a possible narrowing
of the Pt dꢀDOS, likely as a result of the hybridization between
the Pt and Re dꢀbands. Moreover, since there is minimal
charge transfer between Pt and Re, a narrowing of the dꢀDOS
should be expected to result in the upward shift of the dꢀband
center relative to the Fermi level according to the “rectangular
band” model[23]. The conclusions from XAFS were conꢀ
sistent with the stronger CO adsorption as shown by CO ATRꢀ
6
ACS Paragon Plus Environment

Page 7 of 22
ACS Catalysis
IR and TPD[12]. To determine the effect of APR reaction cluster[12]. Because the existence of PtꢀOꢀRe linkage is not
1
conditions on the catalyst structural and electronic properties,
inꢀoperando XAFS of Pt/C and PtꢀRe/C catalysts was perꢀ
formed and compared to that under reducing environment. The
EXAFS fitting for both catalysts are listed in Table 2. It can be
seen that the APR conditions result in an increase in the PtꢀM
and ReꢀM coordination numbers which is consistent with an
increase in particle size during APR as we reported previously
by ex situ TEM[10]. Additionally, the ReꢀO coordination inꢀ
creased despite the increase in particle size which suggests
that the percentage of ReOx on the surface was significantly
higher during APR than after reduction. Figure 6 shows the
XANES spectra at the Re L3 edge under reducing conditions
and during APR. Clearly, it can be seen that the Re L3 edge
shifts to higher energy and the white line intensity increases,
which is consistent with the partial oxidation of Re and/or
adsorption of OH on Re[12] which is consistent with the highꢀ
er ReOx percentage concluded from EXAFS. The oxidation of
Re was investigated further by Raman spectroscopy as disꢀ
cussed below. The small differences between the Pt L2,3
XANES spectra of Pt and PtRe during APR (see SI) can be
mostly attributed to difference in the coverage of adsorbates.
No Pt oxidation was detected by XANES or EXAFS for either
catalyst, Pt/C or PtRe/C, consistent with previous reports on Pt
and PtMo[3].
supported by our XANES characterization at the Pt L3 edge
(Figure S1), assigning this feature to ReꢀOꢀRe linkage is more
likely. The low frequency band at 338 cmꢀ1 along with a
shoulder at 383 cmꢀ1 was assigned to the degeneracy of the Reꢀ
OꢀRe bending modes. Although the bands here were observed
at higher frequencies than previously reported, the band strucꢀ
ture is consistent[24, 25]; the shift is likely due to the close
association with Pt center. The Raman feature at 478 cmꢀ1 was
at the same frequency as that observed on both the reduced
and steamꢀtreated samples. This feature cannot be simply corꢀ
related with oxygen bonding between either Pt or Re as the
closest possible Re features is at ~478 cmꢀ1 for ReꢀOꢀRe symꢀ
metric stretching (very weak)[25] or the closest possible Pt
feature at ~514 cmꢀ1 for the A1g symmetric breathing mode of
αꢀPtO2[26]. In order to properly assign this feature, the steamꢀ
treated catalyst was subjected to HLW treatment with the aim
to leach out excess ReOx not in contact/alloyed with Pt[10]. As
shown in Figure 7, as exposure to HLW was extended, the
terminal ReꢀO feature decreased, whereas the feature 478 cmꢀ1
continued growing along with slight increase in the ReꢀOꢀRe
bending mode intensity. As a result, it appears that the termiꢀ
nal ReꢀO is most vulnerable to leaching treatment, which is
consistent with the fact that Oꢀterminated Re2O7 readily disꢀ
solves in water via hydrolysis. Our previous report showed
that although the Re content on the catalyst decreases by half
after a week of APR reaction, the timeꢀonꢀstream activity only
drops by 10% with little change in selectivity[10]. In contrast,
ReꢀOꢀRe increased slightly, this could be due to better disperꢀ
sion or more oxidation of Re. This feature is also present on
reduced and steamꢀtreated sample. However, the intensity is
relatively low, possibly due to the coverage by terminal ReꢀO.
The PtꢀRe/C catalyst surface structural changes after steam
and HLW exposures are depicted in Scheme 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
3.2.2 Raman study of PtꢀRe/C catalyst and Re leaching
Figure 7. Raman spectra of PtꢀRe/C catalyst after reduction,
steam or hot liquid water (HLW, for 0.5, 1, 1.5, and 2 h) treatment
In order to understand the impact of water treatment on the
catalyst surface structure, Raman study of PtꢀRe/C catalyst
was carried out. Figure 7 depicts the Raman spectra of Ptꢀ
Re/C catalyst after reduction, steam or HLW, respectively.
The reduced sample shows only one small peak at 478 cmꢀ1,
while after steam treatment a peak at 995 cmꢀ1 shows up as
previously reported[12]. Introduction of steam following reꢀ
duction had a negligible effect on the feature at 478 cmꢀ1 but
gave rise to the band at 995 cmꢀ1. Literature has reported[24]
that the latter could be assigned to terminal ReꢀO symmetric
stretching from tetrahedrally coordinated ReO4 unit[25] on the
catalyst surface. Upon HLW treatment, the 478 cmꢀ1 peak inꢀ
creases, whereas the 995 cmꢀ1 peak decreases as a function of
time. The sample with 2 h HLW treatment exhibits four Raꢀ
man bands (338, 383, 478, and 594 cmꢀ1) with the 478 cmꢀ1
band being most prominent. More specifically, the 478 cmꢀ1
feature has been assigned to the MꢀOꢀRe symmetric stretching
mode with lowꢀvalent Re species or low coordination ReOx
Scheme 1. Surface structural changes of PtꢀRe/C after
steam and HLW exposures
Following HLW treatment for up to 4 h, the samples were
tested for APR reaction to understand how the surface properꢀ
ties of catalyst impact the activity. Glycerol was injected into
the aqueous phase without exposing catalyst to air after the
HLW treatment. Figure S2 shows the comparison of glycerol
conversion over catalysts treated under HLW for 0.5, 1, 1.5, 2,
and 4 h (product selectivity was the same as observed over
samples not treated with HLW for both Pt/C and PtꢀRe/C).
Consistent with our prior report using a fixed bed reactor[10],
PtꢀRe/C activity appears to be less affected by this pretreatꢀ
ment, despite the leaching of Re occurring as evidenced by
Raman spectroscopy and the result in Figure S3 showing the
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 22
actual amount of Re that is leached out. Therefore, the leached of Lewis acidity. These data suggest that after HLW treatment
1
2
3
4
5
6
7
8
Re species from the catalyst surface must have a negligible
impact on catalyst activity. Additionally, because of the very
low amount of leached Re species in the aqueous phase, it is
very unlikely that the dissolved Re species are participating as
a homogenous catalyst.
the Pt nanoparticle surface becomes electron deficient (likely
due to adsorption of hydroxyl groups[10, 29]) and the exposed
Ptδ+ acts as Lewis acid (LPy) for the coordination with pyriꢀ
dine molecule. After adding Re, it was observed that a broad
peak (about 50 cmꢀ1 wide) appears at ~1535 cmꢀ1, indicating
CꢀN ring vibration of pyridium cation (PyH+) adsorbed on
Brønsted acid sites of catalyst surface (Pt/C does not show this
feature). An additional feature appeared at 1635 cmꢀ1, confirmꢀ
ing the presence of Brønsted acidity. At least three sets of exꢀ
periment were carried out to make sure the peaks observed
here are not due to spectrum noise. In addition to the peak at
1445 cmꢀ1, another feature of Lewis acidity appeared at 1438
cmꢀ1, forming a doublet. This new feature could arise from
pyridine molecule coordinately adsorbed on exposed Reδ+. The
difference in wavenumber (7 cmꢀ1) indicates that Lewis acidity
associated with Re is stronger than that with Pt. Moreover, by
deconvolution and comparison of the peak area for 1445 and
1438 cmꢀ1, the amount of Lewis acidity associated with Pt and
Re, respectively, appeared to be very similar. This is conꢀ
sistent with the chemisorption measurement showing that after
adding Re, the Pt dispersion decreased by about half.
3.2.3 Pyridine adsorption
In addition to surface acidity from the catalyst support, extra
surface acidity derived from metallic sites (oxide/hydroxide)
has been proposed to have a significant impact on the catalytic
behavior[8, 10, 27]. Due to the strong absorption of infrared
energy by carbon, conventional DRIFTS studies of the cataꢀ
lysts were only possible on Pt and PtRe on other nonꢀcarbon
based supports, e.g. SiO2 [10]. Thus, ATRꢀIR was here emꢀ
ployed to monitor the surface acidic properties of Pt/C and Ptꢀ
Re/C catalysts with in situ treatments. In Figure 8, the characꢀ
teristic features of different acid centers can be resolved and
identified with the peak assignment allowing for differentiaꢀ
tion among Brønsted sites (~1640 and 1540 cmꢀ1) and Lewis
sites (~1610 and 1440 cmꢀ1).
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
3.3
Role of Re species in APR
The primary function of Re species observed in this work is to
enhance the activity of Pt/C catalyst. This enhancement has
previously been related to the weakened interaction between
CO and Pt surface modified by Re. However, our in situ ATRꢀ
IRS study on reduced catalyst does not support this hypotheꢀ
sis. We previously reported that after steam exposure at steam
reforming temperature, CO desorption became more facile
from PtꢀRe/C than that from Pt/C[12]. Here we employed the
same in situ ATRꢀIRS technique under simulated APR condiꢀ
tion to investigate CO adsorption behavior in condensed water
environment. The results over Pt/C and PtꢀRe/C catalysts are
shown in Figure 5. Pt/C shows only linear CO desorption, and
about 20% of the peak area is retained at 200 °C. Facile CO
desorption occurs at low temperatures over PtꢀRe/C, and less
than 20% peak area remains at 100 °C. Considering the differꢀ
ence of the catalyst surface, it is possible that, due to its oxꢀ
ophilicity nature, ReOx may facilitate CO transfer from neighꢀ
boring Pt site via interacting with the O atom, leading to facile
CO desorption. This is consistent with our observation for
steamꢀtreated catalysts. The overall activity enhancement over
PtꢀRe/C catalyst in aqueousꢀphase reforming is also in good
agreement with our previous conclusion in glycerol steam
reforming. Since in condensed water the close proximity beꢀ
tween Pt and oxidized Re on PtꢀRe/C catalyst surface is not
affected and is comparable to that after steam treatment, CO
spillover from Pt site to the neighboring Re oxide species that
we previously reported[12] could be used to explain the obꢀ
served activity enhancement. Moreover, the leached Re oxide
species do not seem to contribute to this enhancement.
Figure 8. Pyridine adsorption on samples after HLW treatment
characterized by ATRꢀIRS
As can be seen in Figure 8, carbon support and Re/C exhibit
merely background signal. Although some organic groups,
including phenolic and carboxylic, could be present on active
carbon surface the concentration was too low to be detected
and their contribution to the overall surface acidity is negligiꢀ
ble. Zawadzki reported pyridine adsorption on carbon film and
concluded that without substantially oxidizing the carbon film
only physisorbed pyridine was present[28]. This is consistent
with what was observed on the carbon support even after
HLW treatment and also implies that the carbon support used
in this study is stable (resistant to oxidation) under HLW
treatment. The fact that there was no measureable acidity over
the Re/C sample could be due to limited Re dispersion, coorꢀ
dination saturation of Re, and little OH on terminal ReO speꢀ
cies[27]. Pt/C showed a broad feature centered at 1588 cmꢀ1
with a shoulder at 1611 cmꢀ1 which can be assigned to physiꢀ
cally and coordinately adsorbed pyridine on Lewis acid sites,
respectively. A less resolved weak peak with a center at 1482
cmꢀ1 indicated the presence of either Brønsted or Lewis acidity
as they can independently contribute to this feature. Because
there was no Brønsted feature observed around 1540 cmꢀ1, it
appears that only Lewis acidity exists. This was further conꢀ
firmed by the clear band at 1445 cmꢀ1 which is characteristic
The preferable reaction pathway of glycerol reforming to proꢀ
duce hydrogen should consist of primarily dehydrogenation
and decarbonylation. It is generally accepted that Pt surface
can catalyze dehydrogenation[30] and a small portion of deꢀ
hydration[31] of glycerol. Questions remain about how the
addition of a second metal affects the overall catalytic activiꢀ
ties of the catalyst and whether the reaction environment has
any effect. In contrast to the statement by Lin[30] that doping
8
ACS Paragon Plus Environment

Page 9 of 22
ACS Catalysis
Re acts as a basic promoter like KOH, under the hydrothermal could participate in the hydrogenolysis pathway along with Pt
1
2
3
4
5
6
7
8
condition Re is believed to present as oxide/hydroxide/hydrate
species providing (additional) acid site[8, 10, 27], or to coorꢀ
dinate with Pt presenting different types of acid site as eviꢀ
denced in this paper. As previously reported, Re in nonꢀzero
oxidation states (e.g., ReOx) could provide sites for CO oxidaꢀ
tion[32] and water activation[33]. In Figure 1, PtꢀRe/C catalyst
shows much lower H2 and CO2 selectivities compared to Pt/C
catalyst. This has been well correlated to the CꢀC and CꢀO
bond cleavage ability as we recently reported[10]. The preꢀ
ferred pathway (all the way including waterꢀgas shift) in glycꢀ
erol APR should produce H2 and CO2 according to equation
C3H8O3+3H2Oꢀ7H2+3CO2. However, the decrease in CO2
selectivity is more pronounced over PtꢀRe/C as shown in Figꢀ
ure 1. Besides, acetol selectivity over PtꢀRe/C is almost douꢀ
bled, whereas propylene glycol selectivity remains about the
same. Therefore, it can be suggested that the reaction pathway
is significantly shifted from dehydrogenation over Pt/C to
dehydration over PtꢀRe/C, while hydrogenation of the dehyꢀ
drated product is also less favored over PtꢀRe/C. The lower
ethylene glycol production and higher selectivity of ethanol
and C2+ alkanes also support the more favored dehydration
pathway.
It has been evidenced by both CO chemisorption and pyridine
adsorption measurement that compared with 3%Pt/C, Pt disꢀ
persion is decreased by about half on 3%Ptꢀ3%Re/C. It is posꢀ
sible that wellꢀmixed bimetallic surface enhances the catalytic
turnover and results in the observed higher TOF over PtꢀRe/C
catalyst. On such bimetallic surface, Pt acts as the adsorption
and main reaction site for dehydrogenation and CꢀC cleavage,
while Re provides the site for temporal accommodation of CO
molecule and its desorption or waterꢀgas shift. A possible reꢀ
action route is shown in Scheme 2. Scheme 2(b) shows the
APR reaction route of which the decomposition product CO
could be transferred from metallic Pt to the neighboring Re
site through the higher affinity of oxygen atom to Re. In aqueꢀ
ous phase reforming, the terminal ReꢀO structure may only
serve as spectator which also leaches out under hot liquid waꢀ
ter condition as evidenced by Raman (Figure 7), while the
active structure for APR reaction involves both Pt and Re govꢀ
erning the activity and selectivity despite the inactive Re speꢀ
cies leaching into the solution. The APR active structure conꢀ
sists of coordinately unsaturated Pt and Re atoms due to the
reaction environment, as evidenced the Raman and pyridine
ATRꢀIR (Figure 8). Acidity brought by these species faciliꢀ
tates the production of acetol as the major dehydration interꢀ
mediate[34]. Pyridine ATRꢀIR showed that coordinatively
unsaturated Pt and Re atoms on the catalyst surface could proꢀ
vide Lewis acidity, and their amounts are comparable based on
the result of 3%Ptꢀ3%Re/C catalyst shown in Figure 8. The
selectivity of acetol, dehydration product of glycerol catalyzed
by Lewis acid site[35], over PtꢀRe/C catalyst is about doubled
compared to that over Pt/C. Therefore, the amount of Lewis
acidity agrees well with the initial step of glycerol dehydration
via Lewis acid site. Although Brønsted acidity is also present
as characterized by pyridine adsorption by ATRꢀIRS in Figure
8, its participation in the initial dehydration step is negligible
because of the absence of propanal and acrolein (via dehydraꢀ
tion) or 1,3ꢀpropandiol (via hydrogenolysis), which are generꢀ
ally accepted as glycerol dehydration product via protonꢀ
catalyzed pathway[36, 37]. However, such Brønsted acidity
site[35]. Scheme 2 shows the reaction route of glycerol aqueꢀ
ousꢀphase reforming and the possible pathway on each site.
Glycerol decomposition is initiated via adsorption on Pt site.
Exposed coordinatively unsaturated Pt and Re atoms under
reaction condition can serve as Lewis acid sites, which are
responsible for the dehydration pathway, competing with the
CꢀC cleavage and thus suppress the formation of hydrogen and
CO2. The synergy between Pt sites and Brønsted acid sites
contribute to the hydrogenolysis reaction pathway.
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
Scheme 2. APR of glycerol over (a) Pt/C and (b) PtꢀRe/C
catalysts
4.
Conclusions
Bimetallic PtꢀRe/C catalyst showed enhanced catalytic activity
compared to monometallic Pt/C catalyst in glycerol APR.
Similar to glycerol steam reforming, the enhancement of activꢀ
ity was found to be due to more facile CO desorption from Ptꢀ
Re/C than that from Pt/C under simulated APR condition.
Contrary to glycerol steam reforming, liquid product is more
favored over PtꢀRe/C at the expense of H2 and CO2 selectivity
in glycerol APR compared to that on Pt/C. This could be due
to the extent of Re oxidation under APR reaction condition
since XANES shows that, compared to steam treatment, APR
reaction environment further oxidizes the surface Re species.
Raman characterization of PtꢀRe/C catalyst after HLW treatꢀ
ment, simulating APR environment, indicated the presence of
both wellꢀdispersed Re oxide species (characterized by the Reꢀ
OꢀRe linkage) associated with Pt sites and Re oxide species
with terminal ReꢀO. However, the Re oxide species with terꢀ
minal ReꢀO can be readily leached out under APR conditions
and have little influence on the product selectivity in glycerol
APR. The leaching of Re oxide species with terminal ReꢀO in
glycerol APR likely exposes the wellꢀdispersed Re oxide speꢀ
cies associated with Pt sites, which is selective towards liquid
product formation. Such structure with coordinately unsaturatꢀ
ed Pt and Re atoms could act as Lewis acid sites catalyzing the
initial dehydration of glycerol to hydroxyacetone (acetol),
leading to completely different product selectivity in glycerol
APR on PtꢀRe/C compared to that on Pt/C. Brønsted acidity is
also present on PtꢀRe/C, which could originate from the bridgꢀ
ing hydroxyl in the MꢀOHꢀM moiety. However, Brønsted
acidity does not seem to compete with Lewis acidity in the
initial dehydration step in glycerol APR due to the absence of
Brønsted acid catalyzed product formation.
AUTHOR INFORMATION
Corresponding Author
* Y. W.: Phone: +1 (509) 371ꢀ6273. Fax: +1 (509) 371ꢀ6242. Eꢀ
mail: yong.wang@pnnl.gov.
* A.M.K: phone: 540ꢀ231ꢀ9138, Eꢀmail: amkarim@vt.edu
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 22
[11] Ciftci, A., Ligthart, D.A.J.M., Sen, A.O., van Hoof, A.J.F., Friedrich,
H., Hensen, E.J.M., J. Catal., 2014, 311. 88ꢀ101.
[12] Wei, Z., Karim, A.M., Li, Y., King, D.L., Wang, Y., J. Catal., 2015,
322. 49ꢀ59.
Present Addresses
1
2
† Present address: Johnson Matthey Inc., 900 Forge Avenue, Suite
100, Audubon, PA 19403, USA
3
4
5
6
7
8
9
[13] Ebbesen, S.D., Mojet, B.L., Lefferts, L., Langmuir, 2005, 22. 1079ꢀ
1085.
[14] Li, Y., Wei, Z., Sun, J., Gao, F., Peden, C.H.F., Wang, Y., J. Phys.
Chem. C, 2013, 117. 5722ꢀ5729.
[15] Karim, A.M., Howard, C., Roberts, B., Kovarik, L., Zhang, L., King,
D.L., Wang, Y., ACS Catal., 2012, 2. 2387ꢀ2394.
[16] Newville, M., J. Synchrotron Radiat., 2001, 8. 96ꢀ100.
[17] Ravel, B., Newville, M., J. Synchrotron Radiat., 2005, 12. 537ꢀ541.
[18] Bare, S.R., Kelly, S.D., D.Vila, F., Boldingh, E., Karapetrova, E.,
Kas, J., Mickelson, G.E., Modica, F.S., Yang, N., Rehr, J.J., J. Phys.
Chem. C, 2011, 115. 5740ꢀ5755.
[19] Roldan Cuenya, B., Croy, J.R., Mostafa, S., Behafarid, F., Li, L.,
Zhang, Z., Yang, J.C., Wang, Q., Frenkel, A.I., J. Am. Chem. Soc., 2010,
132. 8747ꢀ8756.
[20] Kirilin, A.V., Hasse, B., Tokarev, A.V., Kustov, L.M., Baeva, G.N.,
Bragina, G.O., Stakheev, A.Y., Rautio, A.ꢀR., Salmi, T., Etzold, B.J.M.,
Mikkola, J.ꢀP., Murzin, D.Y., Catal. Sci. Technol., 2014, 4. 387ꢀ401.
[21] Ebbesen, S.D., Mojet, B.L., Lefferts, L., J. Catal., 2007, 246. 66ꢀ73.
[22] Mojet, B.L., Ebbesen, S.D., Lefferts, L., Chem. Soc. Rev., 2010, 39.
4643ꢀ4655.
[23] Schweitzer, N., Xin, H., Nikolla, E., Miller, J., Linic, S., Top. Catal.,
2010, 53. 348ꢀ356.
[24] Williams, K.P.J., Harrison, K., J. Chem. Soc., Faraday Trans., 1990,
86. 1603ꢀ1606.
[25] Hardcastle, F.D., Wachs, I.E., Horsley, J.A., Via, G.H., J. Mol.
Catal., 1988, 46. 15ꢀ36.
[26] McBride, J.R., Graham, G.W., Peters, C.R., Weber, W.H., J. Appl.
Phys., 1991, 69. 1596ꢀ1604.
[27] Chia, M., PaganꢀTorres, Y.J., Hibbitts, D., Tan, Q.H., Pham, H.N.,
Datye, A.K., Neurock, M., Davis, R.J., Dumesic, J.A., J. Am. Chem. Soc.,
2011, 133. 12675ꢀ12689.
[28] Zawadzki, J., Carbon, 1988, 26. 627ꢀ633.
[29] Yakovlev, A.L., Neyman, K.M., Zhidomirov, G.M., Rösch, N., J.
Phys. Chem., 1996, 100. 3482ꢀ3487.
[30] Lin, Y.ꢀC., Int. J. Hydrogen Energy, 2013, 38. 2678ꢀ2700.
[31] Wawrzetz, A., Peng, B., Hrabar, A., Jentys, A., Lemonidou, A.A.,
Lercher, J.A., J. Catal., 2010, 269. 411ꢀ420.
[32] Ishida, Y., Ebashi, T., Ito, S.ꢀi., Kubota, T., Kunimori, K., Tomishige,
K., Chem. Commun., 2009, 5308ꢀ5310.
[33] Azzam, K.G., Babich, I.V., Seshan, K., Mojet, B.L., Lefferts, L.,
ChemCatChem, 2013, 5. 557ꢀ564.
[34] Katryniok, B., Paul, S., BelliereꢀBaca, V., Rey, P., Dumeignil, F.,
Green Chem., 2010, 12. 2079ꢀ2098.
[35] Katryniok, B., Paul, S., Dumeignil, F., ACS Catal., 2013, 3. 1819ꢀ
1834.
[36] ten Dam, J., Hanefeld, U., ChemSusChem, 2011, 4. 1017ꢀ1034.
[37] Foo, G.S., Wei, D., Sholl, D.S., Sievers, C., ACS Catal., 2014, 4.
3180ꢀ3192.
‡ Present address: Department of Chemical Engineering, Virginia
Polytechnic Institute and State University, Blacksburg, Virginia
24061, USA
Author Contributions
The manuscript was written through contributions of all authors.
All authors have given approval to the final version of the manuꢀ
script.
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
ACKNOWLEDGMENT
The authors acknowledge financial support from US Department
of Energy (DOE), Office of Basic Energy Sciences, Division of
Chemical Sciences, Geosciences, and Biosciences. Use of the
National Synchrotron Light Source, Brookhaven National Laboraꢀ
tory, for the EXAFS experiments was supported by the US Deꢀ
partment of Energy, Office of Basic Energy Sciences (Grant# DEꢀ
FG02ꢀ05ER15688). Beam line X18A is supported, in part, by the
Synchrotron Catalysis Consortium. A.M.K. would like to thank
Yongchun Hong (Washington State University) for his help with
the XAS data collection.
Supporting Information Available
Effect of hot liquid water (HLW) treatment duration on glycerol
conversion in APR and the percentage of Re retained on the Ptꢀ
Re/C catalyst and leached into the liquid phase. EXAFS and
XANES spectra under APR and reducing environments. This
material is available free of charge via the Internet at
http://pubs.acs.org/.
REFERENCES
[1] Cortright, R.D., Davda, R.R., Dumesic, J.A., Nature, 2002, 418. 964ꢀ
967.
[2] Wen, G., Xu, Y., Ma, H., Xu, Z., Tian, Z., Int. J. Hydrogen Energy,
2008, 33. 6657ꢀ6666.
[3] Dietrich, P., LoboꢀLapidus, R., Wu, T., Sumer, A., Akatay, M.,
Fingland, B., Guo, N., Dumesic, J., Marshall, C., Stach, E., Jellinek, J.,
Delgass, W., Ribeiro, F., Miller, J., Top. Catal., 2012, 55. 53ꢀ69.
[4] Dietrich, P., Wu, T., Sumer, A., Dumesic, J., Jellinek, J., Delgass,
W.N., Ribeiro, F., Miller, J., Top. Catal., 2013, 56. 1814ꢀ1828.
[5] Tupy, S.A., Karim, A.M., Bagia, C., Deng, W., Huang, Y., Vlachos,
D.G., Chen, J.G., ACS Catal., 2012, 2. 2290ꢀ2296.
[6] Yu, W., Porosoff, M.D., Chen, J.G., Chem. Rev., 2012, 112. 5780ꢀ
5817.
[7] Kunkes, E.L., Simonetti, D.A., Dumesic, J.A., Pyrz, W.D., Murillo,
L.E., Chen, J.G.G., Buttrey, D.J., J. Catal., 2008, 260. 164ꢀ177.
[8] King, D.L., Zhang, L.A., Xia, G., Karim, A.M., Heldebrant, D.J.,
Wang, X.Q., Peterson, T., Wang, Y., Appl. Catal., B, 2010, 99. 206ꢀ213.
[9] Wei, Z., Sun, J., Li, Y., Datye, A.K., Wang, Y., Chem. Soc. Rev., 2012,
41. 7994ꢀ8008.
[10] Zhang, L., Karim, A.M., Engelhard, M.H., Wei, Z.H., King, D.L.,
Wang, Y., J. Catal., 2012, 287. 37ꢀ43.
10
ACS Paragon Plus Environment

Page 11 of 22
ACS Catalysis
1
2
3
4
5
Table of Contents
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
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 22
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. TOF of glycerol (min-1) and product selectivity for APR of 20 mL 10 wt.% glycerol aqueous solution
over 3%Pt/C and 3%Pt–3%Re/C at 225 °C. Loaded catalyst weight was 50 mg of 3%Pt/C and 20 mg of
3%Pt-3%Re/C to achieve similar conversion level (<20%). H2 selectivity is defined as the ratio of H2
produced to the maximum H2 that could be produced from the converted glycerol. Selectivity to other
carbon-containing products (in both gas and liquid phases) are normalized and calculated based on moles of
carbon in the product
204x153mm (150 x 150 DPI)
ACS Paragon Plus Environment

Page 13 of 22
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 2. CO adsorption at room temperature (time interval = 1.5 min) on Pt/C under (a) gas and (b)
aqueous phase
119x188mm (150 x 150 DPI)
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 22
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. CO adsorption at different temperatures on Pt/C in (a) gas and (b) aqueous phase
125x196mm (150 x 150 DPI)
ACS Paragon Plus Environment

Page 15 of 22
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. CO adsorption at different temperatures on Pt-Re/C in (a) gas and (b) aqueous phase
114x178mm (150 x 150 DPI)
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 22
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 5. Relative integrated peak area as a function to desorption temperature of Pt/C and Pt-Re/C in gas
or aqueous phase
207x160mm (150 x 150 DPI)
ACS Paragon Plus Environment

Page 17 of 22
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 6. XAFS spectra of Re L3 edge under APR and reduction conditions
142x100mm (150 x 150 DPI)
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 22
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
196x160mm (150 x 150 DPI)
ACS Paragon Plus Environment

Page 19 of 22
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
202x162mm (150 x 150 DPI)
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 22
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
144x73mm (150 x 150 DPI)
ACS Paragon Plus Environment

Page 21 of 22
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
164x61mm (109 x 110 DPI)
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 22
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
210x125mm (150 x 150 DPI)
ACS Paragon Plus Environment