Synthesis of 1,6-Hexanediol from Cellulose Derived Tetrahydrofuran-Dimethanol with Pt-WOx/TiO2 Catalysts

Article abstract of DOI:10.1021/acscatal.7b03593

Cellulose-derived tetrahydrofuran-dimethanol (THFDM) can be converted over Pt-WO_x/TiO₂ catalysts to 1,6-hexanediol (1,6-HDO) with up to 70% yield. This reaction involves ring-opening of THFDM to 1,2,6-hexanetriol (HTO) and then hydrogenolysis of HTO to 1,6-HDO. Hydrogen atoms spill over from Pt sites onto WO_x/TiO₂ to reduce the W=O functional group and create Br?nsted acid sites. Similar catalytic activity for THFDM conversion can be been obtained with a physical mixture of Pt/TiO₂ and WO_x/TiO₂ due to hydrogen spillover over spatially separate Pt and WO_x when a reducible support (TiO₂) is used.

Full text of DOI:10.1021/acscatal.7b03593

Subscriber access provided by READING UNIV
Article
Synthesis of 1,6-Hexanediol from cellulose derived
tetrahydrofuran dimethanol (THFDM) with Pt-WOx/TiO2 catalysts
jiayue he, Samuel P Burt, Madelyn R. Ball, Dongting Zhao,
Ive Hermans, James A. Dumesic, and George W. Huber
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b03593 • Publication Date (Web): 08 Jan 2018
Downloaded from http://pubs.acs.org on January 8, 2018
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 17
ACS Catalysis
1
2
3
4
5
6
Synthesis of 1,6-Hexanediol from cellulose derived tetrahydrofuran-
dimethanol (THFDM) with Pt-WO /TiO catalysts
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
x
2
†
#
†,§#
†
†
†,§
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Jiayue He, Samuel P. Burt, Madelyn Ball, Dongting Zhao, Ive Hermans,
†
*
†*
James A. Dumesic, and George W. Huber
†
Department of Chemical and Biological Engineering, University of WisconsinꢀMadison, Madison, Wisconsin 53706, Unitꢀ
ed States
§
Department of Chemistry, University of WisconsinꢀMadison, Madison, Wisconsin 53706, United States
Several academic and industrial researchers
have studied the production of 1,6ꢀHDO from celluloseꢀ
derived molecules. Faber reported on a threeꢀstep route
ABSTRACT:
Celluloseꢀderived
tetrahydrofuranꢀ
dimethanol (THFDM) can be converted over Ptꢀ
WO /TiO catalysts to 1,6ꢀhexanediol (1,6ꢀHDO) with
up to 70% yield. This reaction involves ringꢀopening of
THFDM to 1,2,6ꢀhexanetriol (HTO) and then hydrogenꢀ
olysis of HTO to 1,6ꢀHDO. Hydrogen spills over from
Pt sites onto WO /TiO to reduce the W=O functional
group and create Brønsted acid sites. Similar catalytic
activity for THFDM conversion can be been obtained
with a physical mixture of Pt/TiO and WO /TiO due to
x
2
8
for 1,6ꢀHDO production from biomass in 1981. This
route involves (I) biomass depolymerization to 5ꢀ
hydroxymethyl furfural (HMF) using an acid catalyst in
the aqueous phase, (II) HMF hydrogenation to tetrahyꢀ
drofuranꢀdimethanol (THFDM) with Raney®ꢀNickel,
and (III) THFDM hydrogenolysis to 1,6ꢀHDO over a
copper chromite catalyst. 1,6ꢀHDO yields were not reꢀ
ported in this patent, while low HMF yields (less than
8%) from cellulose have been reported because of the
x
2
9
2
x
2
hydrogen spillover over spatially separate Pt and WOx
when a reducible support (TiO ) is used.
2
10–12
formation of humins.
Similarly, Buntara et al. reꢀ
ported a method to convert HMF into 1,6ꢀHDO, which
begins with HMF hydrogenation to THFDM, followed
by the ringꢀopening of THFDM to 1,2,6ꢀhexanetriol
Keywords: heterogeneous catalysis, biomass upgrad-
ing, bimetallic catalyst, 1,6-hexanediol, catalyst charac-
terization, hydrogenolysis, C-O bond cleavage, H spill-
(
HTO), and HTO hydrogenolysis to 1,6ꢀHDO (selectiviꢀ
13
2
ty to 1,6ꢀHDO was 73% at 17% conversion). The auꢀ
thors showed a oneꢀpot reaction of THFDM to HDO
using both RhꢀReO /SiO and Nafion SACꢀ13 with a
,6ꢀHDO yield of up to 86%. Drawbacks of this process
include low catalyst activity (0.10 molꢁmol Pt ꢁ min
with THFDM conversion of 21% at 80 C) and rapid
catalyst deactivation due to the high solubility of the
catalyst support (SiO ) in water.
,6ꢀHDO production from HMF over a mixture of
Pd/SiO and IrꢀReO /SiO catalysts in a H O/THF mixꢀ
ture with a 57.8% 1,6ꢀHDO yield. Chen et al. and our
group have also demonstrated CꢀO cleavage of tetrahyꢀ
dropyranꢀ2ꢀmethanol (THPꢀ2M) to 1,6ꢀHDO with Rhꢀ
ReO /C in the aqueous phase.
demonstrated the same reaction from THP2M in a threeꢀ
step process without the use of precious metals.
over
x
2
1
ꢀ1 ꢀ1
1
. Introduction
Celluloseꢀderived from lignocellulosic biomass
is the most abundant renewable carbon resource in naꢀ
ture, making it a promising feedstock for the production
o
14,15
Xiao et al. studied
2
1
1
–5
of bioꢀbased chemicals. 1,6ꢀhexanediol (1,6ꢀHDO) is
2
x
2
2
a high value chemical with a market price of
16
6
$
4,400/ton, an annual market size of 138,000 tons, and
6
a global market value of approximately $600 MM. It is
primarily used as a monomer for polyesters and polyureꢀ
thanes. The most common industrial route for 1,6ꢀ
HDO production is catalytic hydrogenation of adipic
acid or its esters. This traditional route suffers from low
yields, homogeneous chemistry involving toxic chemiꢀ
cals, and the nonꢀrenewable use of petroleumꢀbased
feedstocks. It would be highly desirable to develop a
new route to 1,6ꢀHDO from renewable lignocellulosic
feedstocks.
17,18
Our group has also
7
x
19
7
DuPont patented a process for 1,6ꢀHDO synthesis by
hydrogenolysis of THFDM derived from levogluꢀ
cosenone (LGO) over a PtꢀWO /TiO catalyst with a
reported 83% yield. Little information was given about
the PtꢀWO /TiO2 catalyst used in the
x
2
20
x
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 17
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme
1.
Proposed
reaction
pathway
for
1,6ꢀHDO
synthesis
from
cellulose.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
DuPont patents, and no relationships between the strucꢀ
ture of the catalyst and performance were given. Renꢀ
novia patented a process for the production of a mixture
of HTO and 1,6ꢀHDO from THFDM in the presence of
uct mixture was then filtered using a 0.22 ꢂm PES (polꢀ
yethersulfone) syringe filter.
Products were quantified using a Shimadzu Gas
Chromatograph using a Flame Ionization Detector (FID)
with liquid injection. A Restek RTXꢀVMS capillary colꢀ
umn (Length: 30 m, ID: 0.25 mm, film thickness: 1.4
ꢂm) was used.
PtꢀWO /ZrO , however did not report overall yield or
x
2
21
catalytic activity. While 1,6ꢀHDO production from
biomass is currently receiving attention from both indusꢀ
try and academia there is a need to understand the basic
chemistry and develop a relationship between the strucꢀ
ture of the catalyst and performance. This knowledge
can be used to design improved catalytic materials.
THFDM hydrogenolysis was carried out in a
stainlessꢀsteel tubular flow reactor (30 cm long, 6.35
mm outer diameter), arranged in an upflow configuraꢀ
tion and heated using a tube furnace (Thermcraft Inc.
No. 114ꢀ12ꢀ1ZH). The catalyst (0.10 g) was added into
the reactor with quartz wool and quartz beads packed on
both sides without dilution. The catalyst was reduced
under a H flow (100 cm (STP)min−1) at 250 C for 2 h
before reaction. The reactor was cooled to 160 C and
pressurized to 500 psi H . An aqueous 1 wt% THFDM
solution was pumped into the reactor using an Agilent
The objective of this paper is to describe the
catalytic chemistry for production of 1,6ꢀHDO from
THFDM over a PtꢀWO /TiO catalyst. As we will show
in this paper, the PtꢀWO /TiO catalyst has higher activiꢀ
ty, stability and selectivity than other reported catalytic
materials for this reaction. THFDM can be produced
from cellulose by a route shown in Scheme 1. First, celꢀ
lulose is extracted from lignocellulosic biomass. 1,6ꢀ
HDO is then synthesized from cellulose in a three step
process: (I) cellulose dehydration to levoglucosenone
x
2
3
o
x
2
2
o
2
22
3
ꢀ1
Prostar HPLC pump with coꢀfed H (40 cm (STP)min ).
Two stainless steel tanks (150 cm (STP)each) were used
2
3
to accumulate the liquid products at the reactor outlet.
The accumulated liquid product was drained periodically
into a collector container, filtered through a 0.22 ꢂm
PES syringe filter and then analyzed by GC. The gaseꢀ
ous products continued to flow through a back pressure
regulator to maintain the reaction pressure. Both an exꢀ
cess flow shutꢀoff valve and a pressure relief valve were
also installed in the reactor system to ensure safe operaꢀ
tion. The feed solutions and the liquid products were
analyzed using a gas chromatograph (Shimadzu, GCꢀ
2010) equipped with a flame ionization detector (FID)
and a Restek RTXꢀVMS capillary column. The gas efꢀ
fluent from the continuous flow reactor was analyzed
using a gas chromatograph (Shimadzu, GCꢀ2014)
equipped with an FID and a Restek RTꢀQꢀBond capillary
column as well as a thermal conductivity detector and a
ShinCarbon ST micropacked column.
(
LGO) and 5ꢀhydroxymethyl furfural (HMF) in the
presence of dilute sulfuric acid in a tetrahydrofuran
2
3
(THF)/H O mixture, (II) LGO and HMF hydrogenaꢀ
2
tion and hydrogenolysis to THFDM catalyzed by
2
4
Ni/SiO and Pt/SiO ꢀAl O catalysts, and (III) THFDM
2
2
2
3
hydrogenolysis to 1,6ꢀHDO over a PtꢀWO /TiO cataꢀ
x
2
2
5
lyst. As we will show in this publication, the catalytic
chemistry involved in THFDM hydrogenolysis requires
interaction between Pt and WO sites. The key objective
of this paper is to elucidate the relationship between the
catalyst structure and performance for THFDM hydroꢀ
x
genolysis over PtꢀWO /TiO .
x
2
2
. Experimental
Catalytic Reactions. Reactions were carried
out in 50 mL HEL highꢀpressure batch reactors. Prior to
reaction kinetics studies, the catalysts were first reduced
Catalyst Characterization. Raman experiꢀ
ments were carried out with a highꢀperformance Renꢀ
ishaw InVia Raman Spectrometer equipped with a 514
nm (excitation) laser. All measurements used a 2400 l
3
ꢀ1
under flowing H (100 cm (STP)min ) (Airgas, UHP) at
2
o
o
ꢀ1
2
50 C for two hours (1 C min ). The feedstock (5 wt%
in H O) was then added to the reactor. The reactor was
2
ꢀ1
mm grating with an efficiency of approximately 30% at
purged three times with H prior to reaction, pressurized
to the desired H pressure, stirred at 700 rpm, and heated
to the reaction temperature. After the reaction, the reacꢀ
tor was cooled in ice water and depressurized. The prodꢀ
2
5
14 nm. In situ Raman studies used an Olympus
2
LMPlanFL N objective with 50x magnification and a
working distance of 10.6 mm. Scattered light was filꢀ
ACS Paragon Plus Environment

Page 3 of 17
ACS Catalysis
averaged to give one spectrum with a resolution of 4.0
tered into a UV enhanced (lumogen coated) deep depletꢀ
ꢀ
1
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ed array detector (Renishaw). The instrument was caliꢀ
brated to an internal Si standard giving a Raman signal
cm . A flowꢀthrough transmission cell was loaded with
approximately 50 mg of sample that was pressed into a
selfꢀsupporting wafer with a diameter of 13 mm. Samꢀ
ꢀ1
at 520.7 cm . In situ measurements were taken with a
fully open aperture and a 10 s exposure time, with 8 acꢀ
cumulations. Approximately 10 mg of sample was used
for each in situ trial. Experiments were performed in a
highꢀtemperature cell (Linkam CCR1000) designed for
ꢀ1
ples were dehydrated at 350 °C (1 °C min , hold for 1
ꢀ7
hour) under 10 mbar. Samples were then cooled to 25
°C, and the timeꢀzero spectrum was taken. Pyridine was
then flowed over the catalyst for 10 min by way of a
o
temperatures up to 1000 C. This setup uses a quartz
bubbler with a carrier gas of dry N at atmospheric presꢀ
2
window and waterꢀcooled Oꢀrings. The temperature was
controlled by a Linkam T95ꢀHT system. Gas flows durꢀ
ing in situ experiments were controlled by mass flow
sure. The cell was then flushed with dry N for 10 min to
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
remove physisorbed pyridine. Samples were then heated
o
ꢀ1
o
at 3 C min to 150 C and held for 1 hour under a flow
3
ꢀ1
controllers. A flow rate of 4 cm (STP) min hydrogen
Airgas, UHP) was used for all reductions, and 4
of N to further remove physisorbed pyridine. Spectra
2
o
(
were obtained at 30 C after cooling under flowing N .
2
3
ꢀ1
cm (STP) min of a 20% oxygen/helium (Airgas, UHP)
mixture was used for all calcinations. Heating ramps for
The Opus 7.0 software package was used to analyze the
data. Each spectrum was background subtracted. The
o
ꢀ
ꢀ1
in situ reductions and calcinations were kept at 1 C min
signals occurring at 1535 and 1447 cm were integrated
1
.
Once the desired temperature was reached, it was held
for Brønsted and Lewis acid sites, respectively, each
corresponding to the ν19b vibration for adsorbed pyriꢀ
dine. When needed, signals were deconvoluted using the
Fityk 0.9.8 software package. Extinction coefficients of
there for two hours. The sample was cooled to room
temperature after calcination or reduction. Spectra were
3
ꢀ1
then taken at room temperature under 4 cm (STP)min
6
ꢀ1
ꢀ1
6
ꢀ1
ꢀ1
helium flow. Data were processed with the Renishaw
WiRE software package.
1.3 × 10 cm mol and 1.5 × 10 cm mol were used
to compare quantities of Brønsted and Lewis acid sites,
26
respectively.
Metal loadings of catalysts used in this study
were determined by Inductively Coupled Plasmaꢀ
Atomic Emission Spectroscopy (ICPꢀAES) after digestꢀ
ing the solid samples in a mixture of HCl (Sigmaꢀ
Surface compositions of the carbon supports
were characterized by Xꢀray Photoelectron Spectroscopy
(XPS) using a Kꢀalpha XPS (Thermo Scientific) instruꢀ
ment with a microꢀfocused monochromatic Al Kα Xꢀray
source. Prior to analysis, the catalysts were reduced with
Aldrich #258148), HNO (SigmaꢀAldrich #438073), and
3
o
HF (SigmaꢀAldrich #339261) at 120 C overnight. NH ꢀ
3
o
o
ꢀ1
TPD measurements were carried out with a Micromeritꢀ
ics Autochem II Chemisorption Analyzer. Approximateꢀ
ly 150 mg of sample was used for these measurements.
flowing H at 250 C for two hours (1 C min ), and
2
stored in an oxygenꢀfree environment, where they could
be loaded in an airꢀtight sample holder for transport to
the spectrometer. When it was desired to analyze oxiꢀ
Prior to NH ꢀTPD analysis, samples were dehydrated at
3
o
o
−1
4
00 C for three hours (1 C min ) under a flow of dry
dized samples, catalysts were treated in 1% O /N at
2 2
air (Airgas). Samples that were to be preꢀreduced were
room temperature after reduction, and the same process
o
ꢀ7
then cooled to 50 C under a flow of He (Airgas, UHP),
was carried out. The samples were analyzed at 10 mbar
o
o
ꢀ1
and then heated to 250 C for two hours (1 C min )
pressure and room temperature. The spectra in the C1s,
O1s, Pt4f, W4d, and Ti2p regions were collected over
multiple scans; the number of scans was adjusted for
each element (C = 15, O = 20, Pt = 60, W = 60, Ti = 20)
to obtain an acceptable signal/noise ratio. The pass enerꢀ
gy was held at 50 eV, the dwell time at 50 ms, and the
energy step size at 0.2 eV for each region. Each region
was integrated using the Avantage (Thermo Scientific)
software package for determination of surface composiꢀ
tion.
under a flow of H (Airgas, UHP). Samples were then
2
cooled to 100 °C for NH adsorption. Adsorption was
3
performed for 30 min at 100 °C under a flow of 40
3
ꢀ1
cm (STP)min 10%NH –Ar (Airgas). NH was deꢀ
3
3
3
ꢀ1
sorbed under a 40 cm (STP)min flow of He while heatꢀ
ing at 10 C min .
o
ꢀ1
N physisorption measurements were performed
2
on a Micromeritics 3flex apparatus. The samples were
o
degassed prior to measurement for 4 h at 150 C under
vacuum. Adsorption isotherms were collected at 77 K
and analyzed using the BET method.
Scanning transmission electron microscopy (STEM)
imaging was performed using a FEI Titan STEM with
Cs aberration correction operated at 200 kV in highꢀ
angle
FTIR experiments were performed with a
Bruker Vertex 70 Fourier Transform Infrared Spectromeꢀ
ter equipped with an MCT detector. 3000 scans were
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 17
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. THFDM/HTO (5 wt% in H O, 20 mL) conversion over 10 wt% Ptꢀ10 wt% WO /TiO catalysts at 160˚C unꢀ
2
x
2
der 800 psi H , 700 rpm. (a) THFDM conversion in presence of 0.10 g catalysts, (b) HTO conversion into HDO in
2
presence of 0.10 g catalysts, (c) THFDM conversion with 0.50 g catalysts. (d) proposed reaction pathway for THFDM
conversion to 1,6ꢀHDO.
annular dark field (HAADF) mode. Catalyst samples
were suspended in ethanol then dropped on a holey carꢀ
bon Cu TEM grid. Samples were plasma cleaned before
being loaded into the microscope.
3. Results
Hydrogenolysis of THFDM with Pt-
WO /TiO catalysts. The conversion of THFDM over
x
2
10 wt% Ptꢀ10 wt% WO /TiO is shown in Figure 1 (a).
x
2
CO Chemisorption measurements were carried
out with a Micromeritics Autochem II Chemisorption
Analyzer. Approximately 100 mg of sample was used
for each measurement. Prior to CO chemisorption, samꢀ
HTO selectivity above 95% is obtained throughout the
reaction with a conversion up to 23%, while 1,6ꢀHDO
selectivity is around 5%. The HTO that is formed can be
selectively (90% + selectivity) converted into 1,6ꢀHDO,
as shown in Figure 1 (b). At THFDM conversions of
100% (Figure 1 (c)), 1,6ꢀHDO can be produced in
yields above 70%. 1ꢀhexanol and 1,2ꢀhexanediol (1,2ꢀ
HDO) are the major side products. A 1,6ꢀHDO yield of
>70% can be obtained with 10 wt.% THFDM feedstock
in 8 hrs (Figure S10).
3
ꢀ1
ples were reduced under 40 cm (STP)min 10%H /Ar
2
o
o
ꢀ1
(
Airgas) to 250 C (1 C min ) for two hours. Samples
o
3
ꢀ1
were then cooled to 35 C under 10 cm (STP)min He
Airgas, UHP) flow. 5%CO/He (Airgas) was then
(
3
pulsed onto the catalyst in 0.5 cm (STP) increments unꢀ
til consecutive signals were equal. Signals were then
integrated to determine total CO consumption.
Synthesis methods of catalysts and THFDM
with 13C NMR spectrum) can be found in Supporting
Information.
The rates of conversion of THFDM, HTO and
THPꢀ2M are shown in Table 3. The rate for THFDM
(
ACS Paragon Plus Environment

Page 5 of 17
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. THFDM hydrogenolysis over PtꢀWO /TiO catalysts (a) TOFs of 4 wt% PtꢀWO /TiO as a function of tungꢀ
x
2
x
2
sten loadings. (b) TOFs of 10 wt% PtꢀWO /TiO as a function of tungsten loadings.
x
2
ꢀ
1
ꢀ1
ꢀ1
ꢀ1
conversion is 0.29 molꢁmol Pt ꢁ min . The rate for HTO
wt% W loading to a maximum of 0.75 molꢁmol Pt ꢁ min
ꢀ1
ꢀ1
conversion is 0.15 molꢁmol
ꢁ min . As reported in
with 7 wt% W and then gradually decreases to 0.079
molꢁmol Pt ꢁ min with 20 wt% W, as shown in Figure 2
(b).
Pt
1
8,27
ꢀ1
ꢀ1
previous studies,
the hydrogenolysis of THFDM and
THPꢀ2M over a RhReꢀbased catalyst requires the presꢀ
ence of water to generate the Brønsted acidity of the
oxophilic promoter; however, as shown in Table 1, the
rate for THFDM conversion in THF (0.075 molꢁmol
min ) is 3ꢀ4 times lower than in H O (0.29 molꢁmol
Table 1 Reaction rate for conversion of THFDM, HTO
and tetrahydropyranꢀ2ꢀmethanol (THPꢀ2M) over Ptꢀ
ꢀ
1
ꢁ
ꢁ
Pt
ꢀ1
ꢀ1
Pt
WO /TiO .
x
2
2
ꢀ1
min ). THPꢀ2M undergoes hydrogenolysis in H O on 10
wt% Ptꢀ10 wt% WO /TiO at similar rates as HTO. This
2
Reactant Solvent
Rate
x
2
behavior indicates that there is a difference in the cataꢀ
lytic sites between PtꢀWO /TiO and RhReꢀbased cataꢀ
-1
-1
(
mol·mol Pt · min )
x
2
lysts.
THFDM H O
0.29
0.08
0.15
0.20
2
PtꢀWO /TiO catalysts with various tungsten
x
2
loadings and 10 wt% Pt were tested for THFDM converꢀ
sion in the aqueous phase, as shown in Figure S1.
THFDM conversion goes through a maximum at an inꢀ
termediate tungsten loading. At 8 h timeꢀonꢀstream the
THFDM conversion increases from 5.3% with 4 wt% W
loading to a maximum of 52.8% with 7 wt% W loading.
The THFDM conversion then decreases to 21.5% with
THFDM THF
HTO
H O
2
THPꢀ2M H O
2
1
0 wt% W loading and 4.5% with 20 wt% W loading.
The 10 wt% WO /TiO catalyst has no catalytic
x
2
Regardless of the tungsten loadings, the combined selecꢀ
tivity to 1,6ꢀHDO precursors and 1,6ꢀHDO is ~99%.
activity for THFDM conversion as shown in Table 2. A
0 wt% Pt/TiO catalyst is active for THFDM converꢀ
1
2
sion, however, the rate is only 15% of the rate over 10
wt% Pt – 10 wt% WO /TiO . THFDM can be converted
The turnover frequency (TOF) of THFDM hyꢀ
drogenolysis (on a CO chemisorption basis) as a funcꢀ
tion of tungsten loading on 4 wt% and 10 wt% Ptꢀ
x
2
to HTO with a selectivity of 95% with a physical mixꢀ
ture of 10 wt% Pt/TiO + 10 wt% WO /TiO . The rate of
2
x
2
WO /TiO2 catalysts passes through a maximum as
x
THFDM conversion of this physical mixture (0.25
shown in Figure 2(a) and (b) and Table S1. The TOF
ꢀ1
ꢀ1
molꢁmol Pt ꢁ min ) is near that of the bimetallic catalyst
for THFDM conversion over 4wt% PtꢀWO /TiO inꢀ
x
2
ꢀ
1
ꢀ1
ꢀ1
ꢀ1
1
0 wt% Pt – 10 wt% WO /TiO (0.29 molꢁmol ꢁ min )
x 2 Pt
creases from 0.258 molꢁmol Pt ꢁ min with 4 wt% W
loading to a maximum of 0.570 molꢁmol Pt ꢁ min with 7
wt% W loading and then gradually decreases to 0.089
molꢁmol Pt ꢁ min with 20 wt% W, as shown in Figure 2
(a). The TOF for THFDM conversion over 10 wt% Ptꢀ
WO /TiO increases from 0.10 molꢁmol ꢁ min with 4
ꢀ
ꢀ
1
ꢀ1
and 5ꢀ6 times higher than the Pt/TiO (0.044 molꢁmol
min ). Each catalyst system exhibits high selectivity
2
Pt
1
ꢀ1
ꢁ
ꢀ
1
ꢀ1
to HTO (95~99%).
ꢀ1
ꢀ1
x
2
Pt
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 17
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 2. Rate of THFDM conversion on Pt/Al O , Pt/TiO , Pt/C, WO /TiO , Amberlyst 70 and their
2
3
2
x
2
physical mixtures. Reaction conditions: THFDM (5 wt% in H O, 20 mL), 800 psi H , 160˚C, 700 rpm.
2
2
Catalyst
Pt cat. (g) W or acid TOF
-1
-1
cat. (g)
(mol·mol Pt · min )
5
wt% Pt/Al O
3
0.20
0.10
0
0
0.038
0.044
0.069
0.250
0.005
ꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
10 wt% Pt/TiO2
5 wt% Pt/ Al O + 10 wt% WO / TiO
2
0.20
0.10
0.20
0
0.10
0.10
0.10
0.10
0
2
3
x
10 wt% Pt/ TiO + 10 wt% WO / TiO
2
2
x
5 wt% Pt/C + 10 wt% WO /TiO
x
2
1
1
1
1
0 wt% WO /TiO
x 2
0 wt% Ptꢀ10 wt% WO /TiO
0.10
0.10
0.10
0.290
0.120
0.040
x
2
0 wt% Ptꢀ10 wt% WO / Al O
3
0
x
2
0 wt% Pt/ TiO + Amberlyst 70
0.10
2
Approximately 5% of the W leached from the
mixture of 10 wt% WO /TiO and 10 wt% Pt/TiO in an
PtꢀWO /TiO was studied in a continuous flow
x 2
reactor for 46 h as shown in Figure S9. The selectivity
to the sum of the desired products (1,6ꢀHDO, HTO, and
THPꢀ2M) ranged from 85 to 90%. The THFDM converꢀ
sion decreases from 45% at 11 h time on stream (TOS)
to 23% at 46 h TOS. The 1,6ꢀHDO selectivity increases
with time from 50 to 60%. The HTO selectivity goes
from 23.6% through a maximum of 33.7%, then deꢀ
creases to 27.6% with time. 1,2ꢀHDO and 1,5ꢀHDO are
two other side products produced with selectivities from
2.4 to 3.6%.
x
2
2
8
hr period as determined by ICP. The W leaching is
probably the reason for catalyst deactivation.
A physical mixture of Pt/TiO + Amberlyst 70
2
ꢀ
1
ꢀ1
has similar activity (0.042 molꢁmol ꢁ min ) as the Pt/
Pt
ꢀ1
ꢀ1
TiO catalyst (0.044 molꢁmol ꢁ min ). Physical mixꢀ
2
Pt
tures of Pt on nonꢀreducible supports (Al O or activated
2
3
carbon) and WO /TiO have much lower activity than
x
2
the physical mixture of Pt/TiO and WO /TiO . The
2
x
2
HTO selectivity drops from >95% to 77ꢀ84% with the
mixtures containing Pt/Al O . It is also noted that there
Pt-WO /TiO Characterization. For clarity in
x
2
2
3
the proceeding sections, we would like to define two
terms which will be frequently used. After the final calꢀ
cination of all catalysts, they were reduced under 100
is no conversion over the physical mixture of 5 wt%
Pt/C + 10 wt% WO /TiO after 30 mins. This behavior
x
2
indicates that Pt must be on a reducible support THFDM
hydrogenolysis with physical mixtures. However, the
reaction rate of bimetallic 10 wt% Pt – 10 wt%
3
ꢀ1
o
ꢀ1
cm (STP)min H to 250 C (1 ˚C min ) for 2 hours.
2
3
ꢀ1
Catalysts were then passivated under 100 cm (STP)min
1% O /Ar for 30 minutes at room temperature, so they
ꢀ1
Pt
ꢀ1
2
WO /Al O (0.12 molꢁmol ꢁ min ) is much higher
x
2
3
could be stored under ambient conditions. In some cases,
catalysts were then reduced once again at 250 ˚C for 2
hours to be characterized under reducing conditions.
Catalysts that have undergone this second reduction step
will from now on be referred to as “reduced”.
than that of the physical mixture of Pt/TiO and
2
ꢀ1
Pt
ꢀ1
WO /Al O (0.069 molꢁmol ꢁ min ) and about 40% of
x
2
3
the rate of the THFDM hydrolysis with 10 wt% Pt – 10
wt% WO /TiO .
x
2
ACS Paragon Plus Environment

Page 7 of 17
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Raman spectra of monometallic WO /TiO and PtꢀWO /TiO catalysts (a) WO /TiO after 250˚C calcination
x
2
x
2
x
2
3
3
under 8 cm (STP)/min flow of 20%O /N , (b) WO /TiO after 250˚C reduction under 8 cm (STP)/min flow of H , (c)
PtꢀWO /TiO after 250 ˚C calcination under 8 cm (STP)/min flow of 20%O /N , and (d) PtꢀWO /TiO after 250˚C reꢀ
duction under 8 cm (STP)/min flow of H . All spectra were taken at room temperature.
2
2
x
2
2
3
x
2
2
2
x
2
3
2
ꢀ1
Catalysts that have not undergone this second reduction
step will be referred to as “passivated”.
Fig. 3, the W=O signal at 1010 cm is visible in all cataꢀ
ꢀ1
lysts, and the WꢀOꢀW signal at 810 cm is visible in
some catalysts. No WO signals are visible in the reꢀ
x
Figure 3 shows the results of Raman characterꢀ
duced 4 wt% PtꢀWO /TiO catalysts (Fig 3d). Possible
x
2
ization on WO /TiO . The signals at 395, 511, and 633
x
2
causes of this disappearance in WO signals is the creaꢀ
x
ꢀ1
28–30
cm are representative of anatase.
The signal at 810
tion of WꢀOH sites and/or W metal, each of which are
Raman inactive. Comparison of Figs. 3 (a, b) and (c, d)
ꢀ1
ꢀ1
cm represents WꢀOꢀW, and the signal at 1010 cm
31–33
represents W=O.
the W=O signal increases with increasing W loading
linear trend shown in Figure S2). As the WO particles
The ratio of the WꢀOꢀW signal to
indicates that the W=O bond on WO is more readily
x
reduced in the presence of Pt, likely by way of hydrogen
spillover from a Pt particle onto the WO /TiO surface.
(
x
x
2
grow, the concentration of WꢀOꢀW groups increases and
the concentration of W=O groups stays relatively conꢀ
It was not possible to study bimetallic catalysts with
Raman spectroscopy using higher Pt loadings, because
there was too much light absorption to get an adequate
signal to noise ratio. As a result, this and most of the
proceeding characterization of bimetallic catalysts is
carried out with 4 wt% Pt.
33
stant.
Figs. 3 (a) and 3 (b) show spectra of each
WO /TiO catalyst after calcination and reduction, reꢀ
x
2
spectively. The spectra in Figs. 3 (a) and 3 (b) are esꢀ
sentially identical, indicating that the reduction proceꢀ
dure used here does not reduce WO /TiO without Pt on
Fourier Transform Infrared (FTIR) spectroscopy
of adsorbed pyridine was used to quantify the surface
x
2
acidity. Pyridine was adsorbed on each catalyst after
the catalyst. Figs. 3(c) and (d) show the Raman spectra
of the bimetallic 4 wt% PtꢀWO /TiO catalysts after calꢀ
o
being reduced at 250 C with flowing H and dehydrated
2
x
2
o
at 350 C under vacuum to quantify the Brønsted/Lewis
cination and reduction, respectively. The catalysts with
Pt have noticeably different signals upon reduction. In
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 17
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. (a) FTIR spectra of pyridine adsorbed on 4wt% PtꢀWO /TiO catalysts after 250˚C reduction for two hours
and 350˚C evacuation for 1 hour, quantification of (b) Brønsted, (c) Lewis, and (d) total acid sites at different W loadꢀ
ings. Open symbols represent passivated catalysts and closed symbols represent reduced catalysts.
x
2
Brønsted acid sites created by WO are important for the
x
reactivity of PtW catalysts for similar reactions, like hyꢀ
drogenolysis of glycerol, as the reactions studied here.
Fig. 4 (c) shows that Lewis acid concentrations follow
similar trends whether the catalyst is reduced or passivꢀ
acid ratios of each catalyst. The ν19b vibration was used
to quantify the relative amounts of each site. These
ꢀ1
bands appear at 1450 and 1540 cm for Lewis and
34–39
Brønsted acid sites, respectively.
Overall acid conꢀ
ated. It is possible that the deposition of WO occupies
x
centrations were then determined with Temperature Proꢀ
grammed Desorption of NH (NH ꢀTPD). These values
the Lewis acid sites of the TiO surface up to the WOx
2
3
3
monolayer. This would be shown—at low WO loadꢀ
x
are listed in Table 3.
ings—by a decrease in Lewis acid concentration with
Figure 4 shows the analysis of the FTIR spectra
of adsorbed pyridine on 4 wt% Pt catalysts. Fig. 4 (a)
shows the FTIR spectra of adsorbed pyridine in the 4
increasing WO loading, followed by a flattening out or
x
slight increase of the Lewis acid concentration at the
highest WO loadings, which is what we observe in our
x
wt% Pt catalysts after reduction under H and dehydraꢀ
2
samples. Literature suggests that a WO monolayer ocꢀ
x
tion under vacuum. As expected and shown in Fig. 4 (b),
curs in the range of WO loading where our Lewis acid
x
increasing
W loading corresponds to increasing
concentration reaches a minimum, given the TiO surꢀ
2
Brønsted acid concentration in reduced catalysts (filled
squares) and passivated catalysts (open squares), preꢀ
sumably due to the generation of WꢀOH sites. Based on
Raman and XPS (discussed later) data showing the reꢀ
face area determined by N physisorption (shown in the
2
31
Fig. S5). Finally, Fig. 4 (d) shows that the total acid
concentration remains fairly constant both in reduced
and passivated catalysts.
duction of WO in bimetallic catalysts, we believe Wꢀ
x
The FTIR spectra of adsorbed pyridine on 10
wt%Pt catalysts are shown in Figure 5. The Brønsted
acid concentrations increase with increasing W loading,
OH is formed during reduction, and these FTIR data
indicate that some WꢀOH remains after passivation. It
3
3,40,41
has been proposed in previous literature
that
ACS Paragon Plus Environment

Page 9 of 17
ACS Catalysis
similar at each W loading (Figure 7 & S7). The Pt parꢀ
with or without preꢀreduction, presumably due to the
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
creation of WꢀOH sites. Lewis and total acid concentraꢀ
tions also trendsimilar to those of the 4 wt% Pt catalysts
with and without passivation. We also note that Fig. 5b
shows that Brønsted acid concentrations are slightly
higher in 10 wt% Pt catalysts than 4 wt% Pt catalysts at
each W loading. This could be due to more generation of
ticle size of 10 wt% Pt catalysts measured with STEM is
1.57ꢀ1.74 nm with a standard deviation of 0.58ꢀ0.78 nm.
The Pt particle size of 4 wt% Pt catalysts measured with
STEM is 1.13ꢀ1.45 nm with a standard deviation of
0.31ꢀ0.72 nm. The Pt particle size calculated from the
CO chemisorption (assuming crystallites containing Pt
are spherical and the number of surface atoms is equal to
the area of the particle divided by the cross sectional
acidic TiO sites when there is more Pt on the catalyst to
x
dissociate H , which can spill over onto the surface.
2
42
area of an individual Pt site) is between 2.6 and 9.8 nm
for 10 wt% Pt (Table S1) and 3.1 and 6.6 for 4 wt% Pt
Table 3 shows that the total acid concentration
decreases by nearly 50% in all Ptꢀcontaining catalysts,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(Table 3). These results indicate that WOx and/or
including monometallic Pt/TiO , during reduction. The
2
WO /TiO migrates over a Pt nanoparticle during the
total acid concentration for WO /TiO does not decrease
x
2
x
2
catalyst pretreatment, and thereby blocks CO adsorption
during reduction at any W loading. It appears, then, that
the decrease in acid concentration during reduction is
17,18,43
sites.
There is no decrease in CO chemisorption
between monometallic 10 wt% Pt/TiO and 10 wt% Pt –
not related to WO species, whether on bimetallic or
2
x
4
wt% WO /TiO , indicating that the effect of WO
x 2 x
monometallic Wꢀcontaining catalysts. Figure 6 shows
the CO chemisorption over the 10 wt% Pt and 4 wt% Ptꢀ
WO /TiO catalysts. The quantity of adsorbed CO deꢀ
blocking hydrogenation sites is minimal at such low W
loadings.
x
2
creases as W is added to the catalyst. According to
STEM, the 4 wt% and 10 wt%Pt particle diameters are
Figure 5. (a) FTIR spectra of pyridine adsorbed on 10wt% PtꢀWO /TiO catalysts after 250˚C reduction for
two hours and 350˚C evacuation for 1 hour, quantification of (b) Brønsted, (c) Lewis, and (d) total acid sites at
different W loadings. Open symbols represent passivated catalysts and closed symbols represent reduced cataꢀ
lysts.
x
2
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 17
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. CO Chemisorption over (a) 4wt% PtꢀWO /TiO and (b) 10wt% PtꢀWO /TiO catalysts at different
W loadings.
x
2
x
2
Figure 7. STEM pictures of (a) 4 wt% Pt 4 wt% WO /TiO (b) 4 wt% Pt 7 wt% WO /TiO (c) 4 wt% Pt 10
x
2
x
2
wt% WO /TiO (d) 4 wt% Pt 20 wt% WO /TiO .
x
2
x
2
10
ACS Paragon Plus Environment

Page 11 of 17
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
Table 3. Oxidation states, Brønsted, Lewis, and total acid concentrations, CO chemisorption and THFDM conversion rate of 4wt% PtꢀWO /TiO .
x
2
Catalyst
Pt particle size
nm)
CO
Up- Oxidation States
Oxidation States Brønsted
(passivated cata- Concentration
Acid Lewis Acid Con- Total Acid Con- TOF
take (ꢀmol
centration
centration
(reduced catalysts) lysts)c
c
( mol·mol
-
1
(
g )
-1 d
-1 d
-1 e
-1
-1
(
ꢀmol g )
(ꢀmol g )
(ꢀmol g )
Pt min )
0
a
b
5+
6+
2+
0
5+
6+
2+
0
1
2
3
Reduced Passivated Re-
duced
Passivated
Reduced Passivat-
ed
STEM
CO
W :W
Pt :Pt
W :W
Pt :Pt
4
wt%WO /TiO2
NA
NA
3.1
NA
4.3
NA
5.6
NA
6.6
3.1
ꢀ
0
0.08
0.23
0.06
0.25
0.05
0.14
0.06
0.07
NA
NA
0.04
NA
0.07
NA
0.13
NA
0.12
0
0.08
0.04
0.02
0.09
0.02
0.09
0.01
0.05
NA
NA
45
47
208
210
402
122
315
109
317
115
330
466
389
253
336
253
315
241
290
292
333
370
350
257
595
255
577
221
621
338
609
576
577
ꢀ
x
4
5
4
wt%Pt-4wt%WO /TiO2
63
3
0.36
NA
63
193
119
262
112
303
223
279
110
188
273
136
187
112
155
156
175
319
235
0.258
ꢀ
x
1.35±0.31
NA
6
7
7
wt%WO /TiO2
119
128
129
135
136
158
51
x
8
9
4
wt%Pt-7wt%WO /TiO2
48
2
0.69
NA
0.570
ꢀ
x
1.45±0.40
NA
0
1
1
0wt%WO /TiO
x
2
2
3
4
wt%Pt-10wt%WO /TiO2
41
0
0.19
NA
0.488
ꢀ
x
1.24±0.35
NA
4
5
2
0wt%WO /TiO
x
2
6
7
8
9
0
4wt%Pt-20wt%WO /TiO2
x
30
177
54
0.40
0.30
0.30
0.089
0.044
0.250
1.13±0.72
10wt% Pt/TiO2
1.59
±0.38
1
10wt% Pt/TiO2 + 10wt% ꢀ
0.20
0.20
0.06
115
2
WO /TiO Physical Mix
x
2
(
1:1)
3
4
d
d
5
wt% Pt/Al O3 + 10wt%
25
0.13
NA
0.06
NA
74
53
339
202
413
255
0.069
2
5
WO /TiO Physical Mix
x
2
6
(
1:1)
ꢀ
ꢀ
7
8
9
0
1
2
3
4
5
6
7
8
1
1
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 17
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 8. Raman spectra of 1:1 mixture of (a) physical mixture of 10wt%Pt/TiO and WO /TiO and (b) WO /TiO ,
2
x
2
x
2
and (c) physical mixtures of 5 wt% Pt/C, 5 wt% Pt/Al O and 10wt%Pt/TiO and WO /TiO after reduction with 8
2
3
2
x
2
3
ꢀ1
cm (STP)min H 250˚C for two hours. Spectra taken at 30˚C after cooling under H . “Ambient” indicates these cataꢀ
2
2
lysts
have
not
been
reduced
after
passivation.
XꢀRay Photoelectron Spectroscopy (XPS) was
of the catalyst. These signals are constant during reducꢀ
performed on each catalyst (4 wt% Pt for bimetallics)
under reducing conditions to measure the W and Pt oxiꢀ
dation state information as shown in Table 3. Approxiꢀ
mately 90% of the Pt in bimetallic catalysts is reduced to
metallic sites under our reduction conditions, while
tion of the WO /TiO as shown in Fig. 8 (b). All mixꢀ
tures measured before reduction show WꢀOꢀW and W=O
x
2
signals. Fig. 8 (c) shows that reduction of WO is not
x
seen in the physical mixtures containing Pt/Al O or
2
3
Pt/C with WO /TiO . Each of these samples shows Wꢀ
x
2
6+
5+
some W is reduced to W . This will be discussed furꢀ
ther in section 4.
OꢀW and W=O signals before and after reduction.
The FTIR of adsorbed pyridine on the physical
Characterization of Physical Mixtures of
Monometallic Catalysts. Figure 8 shows the Raman
spectra of monometallic catalysts and their physical
mixtures. Fig. 8 (a) shows the Raman spectra from the
mixtures is summarized in Table 3. The mixture conꢀ
taining Pt/Al O contains roughly half the acidity of
2
3
10wt% WO /TiO . The acidity of the physical mixture
x
2
containing Pt/TiO before and after reduction is similar
2
physical mixture of Pt/TiO and WO /TiO before and
to the bimetallic PtꢀWO /TiO catalyst with similar Pt
2
x
2
x 2
after our reduction treatment. The W=O and WꢀOꢀW
signals disappear in this physical mixture upon reduction
and W loadings.
12
ACS Paragon Plus Environment

Page 13 of 17
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
ꢀ1
Figure 9. XPS spectra of TiO (a) and γꢀAl O (b) before and after 250 ˚C treatment under 8 cm (STP)min H for 2 h.
2
2
3
2
13
The XPS results for mixtures containing
catalysts reported in the literature. Buntara et al. reꢀ
Pt/TiO and Pt/Al O are shown in Table 3. An accurate
ported a RhꢀReO /SiO2 catalyzed ringꢀopening of
2
2
3
x
W oxidation state for the Pt/C mixed with WO /TiO2
could not be determined because the XPS spectra in this
study are calibrated to the adventitious carbon signal at
THFDM to 1,2,6ꢀhexanetriol (HTO) with a rate of 0.013
x
ꢀ1
ꢀ1
molꢁmol Rh ꢁ min , which is 57 times slower than the 10
ꢀ1
ꢀ1
wt% Ptꢀ7 wt% WO /TiO (0.75 molꢁmol ꢁ min )
x
2
Pt
2
84.8 eV. This signal is covered by the much larger carꢀ
(Figure 2(b)). Both catalysts have a HTO selectivity of
above 97%. Buntara et al. also showed HTO hydrogenꢀ
olysis to 1,6ꢀHDO at a selectivity of 73% at 17% conꢀ
version. As shown in Figure 1(b) the selectivity to 1,6ꢀ
HDO with 10 wt% Ptꢀ7 wt% WO /TiO is 90% at 21%
bon support signal in Pt/C. Table 3 shows that the physꢀ
ical mixture of Pt/TiO and WO /TiO has a similar
W /W ratio as PtꢀWO /TiO catalysts with similar Pt
and W loadings. The increase in W /W ratio of the
2
x
2
5+
6+
x
2
5
+
6+
x
2
physical mixture containing Pt/TiO during reduction is
conversion.
2
almost double that of the physical mixture containing
Pt/Al O . This will be discussed further in section 4.
Our results are consistent with the DuPont paꢀ
tents where they reported a process for 1,6ꢀHDO syntheꢀ
sis by hydrogenolysis of THFDM derived from LGO
2
3
We examined the Ti 2p and Al 2s regions of
TiO and γꢀAl O to study the reducibilities of the supꢀ
over a PtꢀWO /TiO catalyst with a reported 83% 1,6ꢀ
2
2
3
x 2
19
ports themselves. Figure 9 shows that the Ti 2p XPS
signal shifts by about 1 eV to lower binding energies
during reduction of TiO , while the reduction of γꢀAl O
HDO yield. Rennovia also has a patent for production
of a mixture of HTO and 1,6ꢀHDO from THFDM in the
20
presence of PtꢀW/ZrO catalysts; however, information
2
2
2
3
about the HTO and 1,6ꢀHDO yields and rates was not
provided in the Rennovia patent.
shows no appreciable shift in the Al 2s signal. This beꢀ
havior indicates that TiO is reduced under our reduction
2
conditions, while γꢀAl O is unaffected. This supports
2
3
The unique catalytic performance of Ptꢀ
our hypothesis that H atoms can be transferred over
TiO , but not Al O , surfaces during reduction of physiꢀ
WO /TiO catalysts is due to their complex surface
x
2
2
2
3
5+
chemistry which includes Pt metal surface sites, W and
cal mixtures, which is the main reason why the physical
mixture containing Pt/TiO can reduce WO and proꢀ
6+
W sites, and reduced TiO sites. Pt particle sizes calcuꢀ
2
2
x
lated by CO chemisorption are larger than the particle
sizes measured directly with STEM, indicating that WOx
duce reactivity, while the physical mixture containing
Pt/Al O cannot. This will be discussed further in secꢀ
2
3
and TiO sites overcoat Pt metal sites. This behavior is
x
tion 4. We also note that the signal positions for our
similar to RhꢀReO /C catalysts, where the CO adsorpꢀ
x
TiO and γꢀAl O samples without reduction treatment
2
2
3
tion quantity decreased linearly with increasing Re conꢀ
tent, whereas the metal particle sizes determined by
match literature values for TiO and Al O , respectiveꢀ
2
2
3
44,45
ly.
. Discussion
The PtꢀWO /TiO catalysts presented in this
17,18,44
TEM remained unchanged.
4
The addition of WO to Pt/TiO adds W=O
x
2
x
2
sites, shown by Raman spectra of oxidized catalysts
Figure 3). These Raman signals disappear upon reducꢀ
publication have superior activity, selectivity, and stabilꢀ
ity for conversion of THFDM to 1,6ꢀHDO than other
(
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 17
tion of bimetallic catalysts, but remain in the case of
monometallic WO /TiO catalysts, indicating that WO
x
follow clear trends with W loading in this range. Meanꢀ
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
while, CO uptake decreases consistently with W loading
in bimetallic catalysts in the range of 4ꢀ20 wt% W. Theꢀ
se results indicate migration of WO /TiO over Pt partiꢀ
x
2
is more readily reduced in the presence of Pt. This beꢀ
havior is likely due to hydrogen spillover from Pt onto
WO /TiO , which enhances the reduction of W=O sigꢀ
x
2
cles during high temperature calcination/reduction,
which lowers the quantity of hydrogenation sites. The
maximum reactivity at intermediate W loadings shows
the need for both Brønsted acid sites for CꢀO cleavage
and exposed Pt sites for hydrogenation. This intermediꢀ
ate loading allows for both necessary processes to occur
x
2
nals to a species that is not Raman active. XPS supports
5
+
6+
this conclusion, showing an increase in W /W ratios
during reduction treatment of bimetallic catalysts.
5+
6+
Meanwhile, W /W ratios remain below 0.1 before and
after reduction on WO /TiO . Furthermore, WO /TiO
x
2
x
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
has essentially no activity in THFDM conversion (Table
for reactivity. At higher W loadings, WO species will
x
5+
2
&3), indicating that W is a crucial piece of the active
cover up hydrogenation sites, thereby decreasing our
rates. At lower W loadings, the Brønsted acid concentraꢀ
tion will be low, and will thereby make CꢀO bond scisꢀ
sion slow and decrease our overall rates. Similarly, M.
site for this reaction. Further supporting this hypothesis,
5+
6+
the W /W ratio is maximized in 4 wt% Pt – 7 wt%
WO /TiO , which is the catalyst with the maximum
x
2
THFDM conversion among the 4 wt% Pt catalysts. The
Chia et al. reported that the RhꢀReO catalyzed hydroꢀ
x
ꢀ1
high THFDM conversion rate in H O (0.29 molꢁmol
ꢁ
genolysis of THPꢀ2M to 1,6ꢀHDO through a bifunctionꢀ
al catalyst which facilitates initial acidꢀcatalyzed carbeꢀ
nium ion chemistry followed by metalꢀcatalyzed hydroꢀ
2
Pt
ꢀ1
min ), which is 3ꢀ4 times higher than THF (0.075
molꢁmol Pt ꢁ min ) (Table 1), is further evidence for
ꢀ1
ꢀ1
17
hydrogen spillover on PtꢀWO /TiO contributing to
genation. A likely mechanism for conversion of
x
2
THFDM conversion. As reported by Benson et al., the
THFDM conversion to 1,6 HDO over Pt – WO /TiO2
x
reduction of WO by hydrogen spillover from Pt in
catalysts includes the following steps:
x
Pt/WO is accelerated by H O, which can facilitate the
x
2
+
ꢀ
46
i.
Hydrogen molecules adsorb on the Pt metal surface
and dissociate to hydrogen atoms.
migration of H atoms (H /e pair).
A physical mixture of Pt/TiO and WO /TiO
2
ii.
Hydrogen atoms at the interface of Pt and TiO2
2
x
shows comparable activity as PtꢀWO /TiO when the Pt
spillover onto TiO Hydrogen atoms travel in the
x
2
2.
ꢀ
+
4+
3+
and W loadings are similar. A physical mixture of
Pt/Al O or Pt/C and WO /TiO shows almost no reacꢀ
pair of e and H . The electrons reduce Ti to Ti
4+
3+
and travel through the support via Ti and Ti exꢀ
2
3
x
2
+
tivity. This comparison indicates that the Pt metal must
changes, while H diffuse across the surface of
4
7
be on a reducible support such as TiO in order to create
TiO2.
H atoms (H and e ) spillover/migrate to WO . A H
2
+
ꢀ
the active site if WO and Pt are not already in direct iii.
x
x
+
ꢀ
contact. This effect is shown in our Raman data of physꢀ
ical mixtures, where W=O signals completely disappear
during reduction of physical mixtures containing
Pt/TiO , but not in those containing Pt/Al O or Pt/C. iv.
atom (H and e ) reacts with a W=O species, reducꢀ
ing W to W , and creating a WꢀOH site, an in situ
generated Brønsted acid site.
6
+
5+
THFDM is dehydrated by a WꢀOH species to form
2
2
3
17
This indicates W=O is reduced to a Raman inactive speꢀ
cies under our reduction conditions in physical mixtures v.
containing Pt/TiO , but not in those containing Pt/Al O
the carbenium ion.
The carbenium ion is hydrogenated on Pt to HTO
and then desorbs.
2
2
3
or Pt/C. XPS supports this data, showing that the
5+
6+
W /W ratio increases more during reduction in physiꢀ
This process is then repeated for conversion of HTO into
cal mixtures containing Pt/TiO than it does in the mixꢀ
2
1
,6ꢀHDO.
5. Conclusions
We demonstrate that PtꢀWO /TiO catalysts
ture containing Pt/Al O under the same reducing condiꢀ
2
3
tions. One of the roles of this support is likely to transfer
H atoms across the surface from a Pt particle to a WOx
species which is not necessarily in close atomic contact.
Over Al O , this process does not occur over ~15 nm
x
2
2
3
46
produce 1,6ꢀHDO with a selectivity of 90% through a
twoꢀstep process including (I) nearly quantitative hydroꢀ
genolysis of THFDM to HTO and (II) hydrogenolysis of
HTO to 1,6ꢀHDO. A 1,6ꢀHDO yield up to 70% is obꢀ
from a Pt particle. Over TiO , on the other hand, H
atoms can diffuse up to micrometers from a Pt particle,
2
46
which explains how WO is able to be reduced in the
x
physical mixture containing Pt/TiO2.
tained in a oneꢀpot reaction over PtꢀWO /TiO . For both
x
2
4
wt% and 10 wt% PtꢀWO /TiO catalysts, the THFDM
x 2
Neither the concentration of acid sites, nor the
concentration of Pt surface sites alone correlates directly
with THFDM conversion over PtꢀWO /TiO . Brønsted
rate goes through a maximum with W loading. Characꢀ
terization data indicate that hydrogen spillover from Pt
onto WO /TiO reduces the W=O functional group on
x
2
x
2
acidity increases consistently with increasing W loading
in the range of 0ꢀ20 wt% W, likely indicating the generꢀ
5+
6+
WO and increases the W /W ratio in our catalysts.
x
3
1,33
The interaction between Pt and reduced WO plays an
x
ation of WꢀOH sites. . Lewis and total acidity do not
14
ACS Paragon Plus Environment

Page 15 of 17
ACS Catalysis
(7)
important role for THFDM conversion. Nearly the same
Werle, P.; Morawietz, M.; Lundmark, S.; Sorensen,
K.; Karvinen, E.; Lehtonen, J. in Ullmann’s
Encyclopedia of Industrial Chemistry, 2008, 263–
281.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
THFDM conversion is obtained from the physical mixꢀ
ture of Pt/TiO and WO /TiO , due to the hydrogen
2
x
2
spillover with spatially separate Pt/TiO and WO /TiO .
2
x
2
Low catalytic activity is observed with physical mixꢀ
tures of WO /TiO with Pt/C or Pt/Al O . The synergisꢀ
(
(
8)
9)
Faber, M. US 4,400,468, 1983.
x
2
2
3
tic effect between Pt and WO is required for the high
x
Buntara, T.; Noel, S.; Phua, P. H.; MelianꢀCabrera,
I.; De Vries, J. G.; Heeres, H. J. Angew. Chemie -
Int. Ed., 2011, 50, 7083–7087.
activity and selectivity of PtꢀWO /TiO and is facilitated
x
2
by the reducible support.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
10)
Daorattanachai, P.; Namuengruk, N.; Viriyaꢀ
Empikul, N.; Laosiripojana, N.; Faungnawakij, K.
J. Ind. Eng. Chem., 2012, 18, 1893–1901.
ꢀ
Supplementary Information
Catalyst Synthesis techniques, as well as synꢀ
thesis technique of THFDM from HMF, are listed in the
Supporting Information. Tables S1ꢀS3 and Figures S1ꢀ
S10 can also be found there.
(11)
(12)
Yin, S.; Pan, Y., Tan, Z. Int. J. Green Energy,
2011, 8, 234–247.
ꢀ
Acknowledgements
Kuster, B. F. M. Starch-Starke, 1990, 42, 314–321.
This material is based upon work supported by
(
13)
Buntara, T.; Noel, S.; Phua, P. H.; MelianꢀCabrera,
I.; De Vries, J. G.; Heeres, H. J. Top. Catal., 2012,
the Department of Energy, Office of Energy Efficiency
and Renewable Energy (EERE), under Award Number
DEꢀEE0006878. Any opinions, findings, and concluꢀ
sions or recommendations expressed in this material are
those of the authors and do not necessarily reflect the
views of the Department of Energy. We thank Phoebe
Wagner, Xiaoli Chen, Giulia Kitayama Canhetti Mondin
for assistance with THFDM and catalyst synthesis and
Daniel McClelland for the NMR experiments and analyꢀ
sis. The authors acknowledge use of instrumentation
supported by UW MRSEC (DMRꢀ1121288) and the
UW NSEC (DMRꢀ0832760).
5
5, 612–619.
(
14)
Gunnarsson, I.; Arnorsson, S. Geochim.
Cosmochim. Acta, 2000, 64, 2295–2307.
(15)
(16)
Siever, R. J. Geol., 1962, 70, 127–150.
Xiao, B.; Zheng, M.; Li, X.; Pang, J.; Sun, R.;
Wang, H.; Pan, X.; Wang, A.ꢀQ.; Wang, X.; Zhang,
T. Green Chem., 2016, 18, 2175–2184.
(
17)
Chia, M.; PaganꢀTorres, J.; Hibbits, D.; Tan, Q.;
Pham, H. N.; Datye, A. K.; Neurock, M.; Davis, R.
J.; Dumesic, J. A. J. Am. Chem. Soc., 2011, 133,
ꢀ References
(
1)
Werpy, T.; Petersen, G. Top Value Added
Chemicals from Biomass: Volume I- Results of
Screening for potental Candidates from Sugars and
Synthesis Gas, Golden, CO, 2004.
1
2675–12689.
(
18)
Chen, K.; Koso, S.; Kubota, T.; Nakagawa, Y.;
Tomishige, K. ChemCatChem, 2010, 2, 547–555.
(
(
2)
3)
Huber, G. W.; Iborra, S.; Corma, A. Chem. Rev.,
(19)
(20)
Allgeier, A. M.; De Silva, W. I. N.; Korovessi, E.;
Menning, C. A.; Ritter, J. C.; Sengupta, S. K.;
Stauffer, S. US 8,865,940 B2, 2014.
2
006, 106, 4044–4098.
Centi, G.; Van Santen, R. A. Catalysis for
Renewables: From Feedstock to Energy
Production, 2007.
Burt, S. P., Barnett, K. J., McClelland, D. J., Wolf,
P., Dumesic, J. A., Huber, G. W., & Hermans, I.
Green Chem., 2017, 19, 1390ꢀ1398.
(4)
De bruyn, M.; Fan, J.; Budarin, V. L.; Macquarrie,
D. J.; Gomez, L. D.; Simister, R.; Farmer, T. J.;
Raverty, W. D.; McQueenꢀMason, S. J.; Clark, J.
H. Energy Environ. Sci., 2016, 9, 2571–2574.
(
21) Sokolovskii, V.; Lavrenko, M.; Hagemeyer, A.;
Dias, E. L.; Shoemaker, J. A. W.; Murphy, V. J.
US 9,586,920 B2, 2017.
(
(
5)
6)
Cao, F.; Schwartz, T. J.; McClelland, D. J.;
Krishna, S. H.; Dumesic, J. A.; Huber, G. W.
Energy Environ. Sci., 2015, 8, 1808–1815.
(
22)
Alonso, D. M.; Hakim, S.; Zhou, S.; Won, W.;
Hosseinaei, O.; Tao, J.; GarciaꢀNegron, V.;
Motagamwala, A. H.; Mellmer, M. A.; Huang, K.;
Houtman, C. J.; Labbe, N.; Harper, D. P.;
Maravelias, C.; Runge, T.; Dumesic. J. A.
ScienceAdvances, 2017, 3(5).
MarketsandMarkets, 1,6-Hexanediol Market by
Application (Polyurethanes, Coatings, Acrylates,
Adhesives,
Plasticizers, and Others) - Trends and Forecasts to
019, 2014.
Unsaturated
Polyester
Resins,
2
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 17
Marques, J. P.; Gener, I.; Ayrault, P.; Bordado, J.
(
23)
He, J.; Liu, M.; Huang, K.; Walker, T. W.;
(37)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Maravelias, C. T.; Dumesic, J. A.; Huber, G. W.
Green Chem., 2017, 19, 3642–3653.
C.; Lopes, J. M.; Ramoa Ribeiro, F.; Guisnet, M.
Microporous Mesoporous Mater., 2003, 60, 251–
262.
(24)
Krishna, S. H.; McClelland, D. J.; Rashke, Q. A.;
Dumesic, J. A.; Huber, G. W. Green Chem., 2017,
(38)
(39)
(40)
(41)
Morterra, C.; Cerrato, G. Catal. Letters, 1991, 10,
357–363.
1
9, 1278–1285.
(
(
25)
26)
He, J.; Huang, K.; Barnett, K. J.; Krishna, S.;
Alonso, D. M.; Brentzel, Z. J.; Burt, S. P.; Walker,
T.; Banholzer, W. F.; Maravelias, C. T.; Hermans,
I.; Dumesic, J. A.; Huber, G. W. Faraday Discuss.,
Tsyganenko, A. A.; Storzheva, E. N.; Manoilova,
O. V. Catal. Today, 2001, 70, 59–71.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RossꢀMedgaarden, E. I.; Wachs, I. E. J. Phys.
Chem. C, 2007, 111, 15089–15099.
2
017, 202, 247ꢀ267.
Kiricsi, I.; Flego, C.; Pazzuconi, G.; Parker, W. O.;
Millini, R.; Perego, C.; Bellussi, G. J. Phys. Chem.,
Fan, Y.; Cheng, S.; Wang, H.; Tian, J.; Xie, S.; Pei,
Y.; Qiao, M.; Zong, B. Appl. Catal. B, Environ.,
2017, 217, 331–341.
1
994, 4627ꢀ4634.
(
(
42)
43)
Bartholomew, C. H.; Farrauto, R. J. Fundametals
of Industrial Catalytic Processes, WILEYꢀVCH
Verlag, Second Edi., 2006.
(27)
Chia, M., O’Neill, B. J.; Alamillo, R.; Dietrich, P.
J.; Ribeiro, F. H.; Miller, J. T.; Dumesic, J. A. J.
Catal., 2013, 308, 226–236.
Karanjkar, P.; Burt, S. P.; Chen, X.; Barnett, K. J.;
Ball, M.; Kumbhalkar, M.; Want, X.ꢀH.; Miller, J.
B.; Hermans, I.; Dumesic, J. A.; Huber, G. W.
Catal. Sci. Technol., 2016, 6, 7841–7851.
(
(
28)
29)
Deo, G.; Turek, A. M.; Wachs, I. E.; Machej, T.;
Haber, J.; Das, N.; Eckert, H.; Hirt, A. M. Appl.
Catal. A Gen., 1992, 91, 27–42.
Deo G.; Turek, A. M.; Wachs, I. E.; Huybrechts, R.
C.; Jacobs, P. A. Zeolites, 1993, 13, 365–373.
(
44)
Rueda, J.; Mendialdua, J.; Rodriguez, A.;
Cassanova, R.; Barbaux, Y.; Gengembre, L.;
Jalowiecki, L. J. Electron Spectros. Relat.
Phenomena, 1996, 82, 135–143.
(30)
Scepanovic, M. J.; GrujicꢀBrojcin, M.; Dohcevicꢀ
Mitrovic, Z. D.; Popovic, Z. V. Sci. Sinter., 2009,
41, 67–73.
(
(
(
45)
46)
47)
Saha, N. C.; Tompkins, H. G. J. Appl. Phys., 1992,
72, 3072–3079.
(
(
31)
32)
Wachs, I. E.; Kim, T.; Ross, E. I. Catal. Today,
2
006, 116, 162–168.
Benson, J. E.; Kohn, H. W.; Boudart, M. J. Catal.,
1966, 5, 307–313.
Ostromecki, M. M.; Burcham, L. J.; Wachs, I. E.;
Ramani, N.; Ekerdt, J. G. J. Mol. Catal. A Chem.,
Karim, W.; Spreafico, C.; Kleibert, A.; Gobrecht,
J.; VandeVodele, J.; Ekinci, Y.; van Bokhoven, J.
A. Nature, 2017, 541, 68–71.
1
998, 132, 43–57.
(
33)
Berton, D. G.; Shtein, M.; Wilson, R. D.; Soled, S.
L.; Iglesia, E. J. Phys. Chem. B, 1999, 103, 630–
6
40.
ASSOCIATED CONTENT
Supporting Information is attached
(34)
Di Iorio, J. R.; Bates, S. A.; Verma, A. A.; Delgass,
W. N.; Ribeiro, F. H.; Miller, J. T.; Gounder, R.
Top. Catal., 2015, 58, 424–434.
AUTHOR INFORMATION
Corresponding Author
(
(
35)
36)
Khaleel, A. A.; Klabunde, K. J. Chem. - A Eur. J.,
*
*
Email: gwhuber@wisc.edu
Email: jdumesic@wisc.edu
2
002, 8, 3991–3998.
Li, C.; Zhou, G.; Wang, L.; Dong, S.; Li, J.; Cheng,
T. Appl. Catal. A Gen., 2011, 400, 104–110.
Author Contributions
Jiayue He and Samuel P. Burt contributed equally.
#
16
ACS Paragon Plus Environment

Page 17 of 17
ACS Catalysis
Authors are required to submit a graphic entry for the Table of Contents (TOC) that, in conjunction with the manuscript title,
should give the reader a representative idea of one of the following: A key structure, reaction, equation, concept, or theorem,
etc., that is discussed in the manuscript. Consult the journal’s Instructions for Authors for TOC graphic specifications.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
17
ACS Paragon Plus Environment