Plasmon-enhanced reverse water gas shift reaction over oxide supported Au catalysts

Article abstract of DOI:10.1039/c4cy01183j

We show that localized surface plasmon resonance (LSPR) can enhance the catalytic activities of different oxide-supported Au catalysts for the reverse water gas shift (RWGS) reaction. Oxide-supported Au catalysts showed 30 to 1300% higher activity for RWGS under visible light compared to dark conditions. Au/TiO₂ catalyst prepared by the deposition-precipitation (DP) method with 3.5 nm average Au particle size showed the highest activity for the RWGS reaction. Visible light is converted into chemical energy for this reaction with up to a 5% overall efficiency. A shift in the apparent activation energy (from 47 kJ mol^-1 in dark to 35 kJ mol^-1 in light) and apparent reaction order with respect to CO₂ (from 0.5 in dark to 1.0 in light) occurs due to the LSPR. Our kinetic results indicate that the LSPR increases the rate of either the hydroxyl hydrogenation or carboxyl decomposition more than any other steps in the reaction network.

Full text of DOI:10.1039/c4cy01183j

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: A. Upadhye, I.
Ro, Z. Xu, H. J. Kim, M. I. Tejedor-Tejedor, M. A. Anderson, J. Dumesic and G. Huber, Catal. Sci. Technol.,
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/catalysis

Page 1 of 19
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01183J
PlasmonꢀEnhanced Reverse Water Gas Shift Reaction over Oxide
Supported Au Catalysts
Aniruddha A. Upadhyea, Insoo Roa, Xu Zenga, Hyung Ju Kima, Isabel Tejedorb, Marc A. Andersonb,
James A. Dumesica, George W. Hubera,*
Abstract
We show that localized surface plasmon resonance (LSPR) can enhance the catalytic activities of
different oxideꢀsupported Au catalysts for the reverse water gas shift (RWGS) reaction. Oxideꢀsupported
Au catalysts showed 30 to 1300% higher activity for RWGS under visible light compared to dark
conditions. Au/TiO2 catalyst prepared by depositionꢀprecipitation (DP) method with 3.5 nm average Au
particle size showed the highest activity for the RWGS reaction. Visible light is converted into chemical
energy for this reaction with up to a 5 % overall efficiency. A shift in the apparent activation energy (from
47 kJ/mol in dark to 35 kJ/mol in light) and apparent reaction order with respect to CO2 (from 0.5 in dark
to 1.0 in light) occurs due to the LSPR. Our kinetic results indicate that the LSPR increases the rate of
either the hydroxyl hydrogenation or carboxyl decomposition more than any other steps in the reaction
network.
1.0 Introduction
Environmental concerns over anthropogenic CO2 emissions combined with increased petroleum costs are
causing our society to develop new sustainable methods for production of renewable transportation fuels
and commodity chemicals. Technology breakthroughs in photovoltaics are making solarꢀdriven electricity
production more efficient1ꢀ3. This renewable electricity could be used to produce hydrogen via water
electrolysis4. There are several other approaches being developed to produce hydrogen directly from solar
photoelectrochemical water splitting1, 2, 4ꢀ8. However, hydrogen cannot power heavy machinery including
jet and diesel engines. Furthermore, using hydrogen in an internal combustion engine would require a
major change in the existing infrastructure. Liquid carbonꢀbased fuels are the preferred fuel option for the
transportation sector because they: (1) are easy to store, (2) have an established distribution infrastructure,
and (3) have higher energy density. Ideally, renewable hydrogen can be reacted with carbon dioxide to
produce liquid fuels in an economically viable process. Thus, the area of CO2 reduction to fuels and
chemicals has recently received significant attention9ꢀ12. The first step of reduction of CO2 using H2 is the
reverse water gas shift (RWGS) reaction as shown in Equation 1. This reaction is endothermic with a
standard heat of reaction as 41.27 kJ/mol.
CO2+ H2 ↔ CO + H2O
∆H0 = 41.27 kJ/mol
(1)
The carbon monoxide produced in the RWGS reaction can then be upgraded to “dropꢀin” transportation
fuels and chemicals through mature technologies such as methanol synthesis and FischerꢀTropsch
synthesis13. Both of these downstream processing options are exothermic. Thus, RWGS reaction is the
only energy intensive reaction involved in catalytic reduction of CO2 to useful fuels and chemicals.
Hence, it would be highly desirable to develop photocatalytic processes which would provide energy for
the RWGS reaction from solar energy.
aDepartment of Chemical and Biological Engineering, University of WisconsinꢀMadison, 53706
bDepartment of Civil and Environmental Engineering, University of WisconsinꢀMadison, 53706
*Corresponding Author (huber@engr.wisc.edu)

Catalysis Science & Technology
Page 2 of 19
View Article Online
DOI: 10.1039/C4CY01183J
Solar energy is the ultimate source of renewable energy. The amount of solar energy intercepted by the
earth in one hour (4.3×1020 J) is sufficient to satisfy the annual global energy consumption (4.1×1020 J)2,
14. For the past few decades researchers have been trying to harness solar energy to drive chemical
reactions for fuel production using inorganic materials. Traditionally, photocatalytic reactions have been
carried out over semiconductor based materials4, 8. Most of these materials are photocatalytically active in
the ultraviolet (UV) region. Only 4% of solar spectrum contains UV radiation, while 46% of the solar
energy is contained in the visible part of the solar spectrum. Numerous approaches including doping15,
18
quantum confinement16, dyeꢀsensitization17,
have been proposed to move semiconductor based
photocatalysis in visible range and this remains one of the most researched areas in solar energy
conversion.
Recent studies have shown that the phenomenon of localized surface plasmon resonance (LSPR) can be
applied to catalytic reactions to enhance the activity of LSPR exhibiting materials under visible light
illumination19ꢀ25. The phenomenon of LSPR occurs when the size of metal particle is in the range of the
penetration depth of the light for that metal26. For metals such as gold, copper, silver etc. the light
penetration depth is around 25 nm26. As a result of LSPR, three main effects have been reported to arise in
a plasmonic material23, 27: (1) localized heating, (2) enhanced electromagnetic fields, and (3) electronꢀhole
pair generation. All of these three effects have been proposed to enhance the activity of catalytic
materials23, 27. One of the advantages of plasmonic catalysts is that the LSPR frequency is a strong
function of type of metal, and structure (i.e. size and shape) of the metal nanoparticle28ꢀ30. This behavior
allows the engineering of catalytic materials to increase the light absorption in the visible region. Au, Ag,
and Cu are considered the most promising plasmonic materials because of their (i) optical properties (i.e.
low loss factor and LSPR frequency in the visible range) and (ii) catalytic properties for certain
reactions19.
The primary objective of this paper is to elucidate how LSPR can be used to change the kinetics of the
endothermic RWGS reaction for CO2 reduction. In addition, we will demonstrate how visible light can be
used to drive this reaction. It has previously been reported that visible light can enhance the activity for
WGS on Au/TiO2 catalysts21 and CO2 reduction on Au/ZnO catalyst31. Thus, we anticipate that the
Au/TiO2 catalyst would be active for RWGS reaction as well. However, it is unknown how the plasmons
change the intrinsic kinetics of these surface reactions30.

Page 3 of 19
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01183J
2.0 Experimental
2.1 Reactor and analytical setup
Figure 1 shows the experimental setup used to measure the activity of different catalysts for RWGS
reaction. The flow rate for the reactant gases CO2 and H2, and inert gas He were controlled using Coleꢀ
Parmer mass flow controllers (FFꢀ32907ꢀ59). Research grade CO2 (CDR200), Ultra high purity (UHP) H2
(UHP HY300) and He (UHP HE300) gas cylinders from Airgas were used. The reactor was purchased
from Harrick scientific (HVCꢀMRAꢀ5) and operated in the temperature range of room temperature to 400
°C and pressures up to 150 psi. 3ꢀ30 mg of catalyst was loaded in the heater cup of the Harrick reactor as
shown in Figure 1. The bottom part of the cup was loaded with the catalyst support which showed
minimal catalyst activity as described below. Only the top 1ꢀ2 mm of the cup were loaded with the
catalyst to ensure illumination of catalyst bed. The reactor had SiO2 windows which allowed catalyst
illumination with visible light. The temperature of the catalyst bed was controlled by a PID temperature
controller. The heating block was located at the bottom of the catalyst bed, while the window of the
reactor was exposed to room temperature as seen in Figure 1. To measure the temperature gradient, the
heater cup was filled with sand (about 5 mm depth). The window of the reactor was removed and
temperature of the sand was measured exꢀsitu with another thermocouple. This temperature gradient was
estimated as the difference between sand temperature and the set point temperature. The gradient was less
than 10% of the set point temperature as measured in °Celsius (e.g. the sand bed temperature was 366 °C
at set point temperature of 400 °C). The difference in apparent activation energy values calculated with
the temperature gradient and without the temperature gradient was only 6%. The temperature had a small
effect on the apparent reaction order, i.e. we obtained apparent rate order with respect to CO2 as 0.5 at set
point temperature of 200 °C and 0.6 at 260 °C. These results show that the temperature gradient did not
affect our measurement of apparent activation energy or apparent reaction orders. The reactor pressure
was maintained using a back pressure regulator. The product gas flow rate was measured by a bubble
flow meter.
The composition of the product gases was analyzed by an online Shimadzuꢀgas chromatograph with
barrier discharge ionization detector (GCꢀBID) system with autoꢀsampling 6ꢀport valve. The BID uses a
helium plasma to detect permanent gases such as CO2, CO, H2 with high sensitivity. The products were
separated using four different columns in the GC. First, all the gases were passed through Haysep T and Q
Bond column to separate CO2 and organics from permanent gases such as CO, H2, CH4, N2 and O2. The
permanent gases which elute early were loaded on the molecular sieve column. The remaining gases
flowing through Q Bond column were sent to the BID through the Shim Q column to detect CO2 and
other organics. After achieving elution of all the organics from the Q Bond and Shim Q columns,
permanent gases from molecular sieve column were sent to the BID to detect gases such as H2, O2, N2,
CO, and methane. The GCꢀBID system was calibrated using Scotty specialty gases (P/N 34507 and
34512).
The catalyst bed was illuminated using a DolanꢀJenner broadband visible light source (MIꢀ150) with fiber
optic cable as shown in Figure 1. The intensity and spectral distribution of the light were measured using
NISTꢀtraceable, ISOꢀ17025 calibrated spectroradiometer (ILT950) from International Light
Technologies. For this measurement, the light passed through the SiO2 window to account for losses in
transmission through the window. Using this setup the light intensity reaching the catalyst bed was

Catalysis Science & Technology
Page 4 of 19
View Article Online
DOI: 10.1039/C4CY01183J
measured as 5216 W/m2 (refer to Figure S1 for spectral distribution of the light). For dark reaction, the
window was covered with an opaque metal disc as seen in Figure 1.
Figure 1. Experimental setup for test of activity of different catalyst for RWGS reaction under light and
dark conditions
2.2 Catalyst synthesis
Catalyst investigated in this study were prepared by depositionꢀprecipitation (DP)32ꢀ34 and impregnation
(I) methods21, 35. The Au/TiO2 (DP), Au/CeO2 (DP), and Au/Al2O3 (DP) catalysts were prepared by
depositionꢀprecipitation (DP) method32ꢀ34. Degussa P25 TiO2 (Aldrich, >99.5%), CeO2 (Aldrich), and
Al2O3 (Strem Chemicals) were used as supports. HAuCl4·3H2O (Aldrich, >99.9% trace metal basis) was
used as the precursor for Au catalysts, while CuSO4·5H2O (Aldrich, 99ꢀ100.5%) was used as a precursor
for the Cu/TiO2 catalyst. For preparation of supported Au catalysts by DP method, supports (i.e. TiO2,
CeO2 and Al2O3) were dried in the air at 110 °C overnight. One hundred ml of aqueous HAuCl4 solution
(4.2×10ꢀ3 M) was heated to 80 °C and the pH was adjusted to 8 by dropꢀwise addition of NaOH (1 M).
Then, 1 g of support was dispersed in the solution, and the pH was readjusted to 8 with NaOH. The
suspension was maintained at 80 °C with continuous stirring. After 2 h the suspension was centrifuged.
The solids were then washed, dried, and calcined at 300 °C under the flow of air (30 cm3(STP)/min) with
a heating rate of 2 °C/min and maintained for 4 h. The Cu/TiO2 (I) catalyst was prepared by impregnating
1 g of TiO2 with a solution of 53 mg of CuSO4·5H2O in 10 ml of DI water21, 35. The slurry was stirred for
4 h at room temperature, then all liquid was evaporated and the solid was dried at 110 °C overnight. The
catalyst was calcined at 300 °C under flowing air (30 ml/min) with a heating rate of 2 °C/min and
maintained for 4 h. Prior to the reactions, the catalysts were reduced inꢀsitu under flowing H2 ( 10
cm3(STP)/min) with a heating rate of 1 °C/min and maintained for 2 h at 300 °C.
2.3 Catalyst characterization
The metal loadings of the catalysts were measured using Inductively Coupled Plasma (ICP) analysis. For
ICP analysis, 8ꢀ10 mg of the catalyst was dissolved in an acid solution of HF, HCl, and HNO3 in a Teflon
beaker at 110 °C until the particles dissolved. The Au ions were detected using Perkin Elmer Plasma 400
ICP emission spectrometer. Gold (2.5ꢀ25 ppm) standards were prepared from a purchased Au 1000 ppm
standard (38168, Fluka). The amount of Cu on Cu/TiO2 catalyst was estimated from the CuSO4 amount
used during impregnation as all of the Cu is deposited on the support in this procedure. The synthesized
catalysts were characterized using transmission electron microscopy (TEM) to obtain the particle size

Page 5 of 19
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01183J
distribution of the Au. A Tecnai T12 TEM at 120 kV was used for TEM analysis. The light absorption of
the catalyst was characterized using diffuse reflectance ultraꢀviolet visible spectroscopy. Thermo
Scientific Evolution 300 UVꢀVis spectrometer was used with Praying Mantis diffuse reflectance
accessory to collect UVꢀVisible absorption data.
3.0 Results
3.1 Catalyst Characterization
Table 1 shows the RWGS activity of different Au and Cu catalysts supported on different oxides such as
TiO2, CeO2, and Al2O3 with and without light. The metal loading of the catalysts was estimated using ICP
analysis postꢀcatalyst synthesis. Using TEM analysis, the Au particle size for the catalysts was estimated
to be between 3 to 11 nm as shown in Figure 2. The gold metal nanoparticles in these size ranges have
been shown to exhibit LSPR27. For example, Garcia and coꢀworkers have observed LSPR with 3ꢀ5 nm
size Au particles for water gas shift reaction21. The LSPR was further characterized by UVꢀVisible
extinction spectra as seen in Figure 3. The absorption peak around 520ꢀ580 nm is clearly seen in all the
Au catalysts corresponding to LSPR of Au nanoparticles. Difference in peak area and height can be
attributed to the different gold loading amounts. For Cu/TiO2 catalyst, a sharp LSPR peak like Auꢀbased
catalyst is not seen in Figure 3 (a). This behavior is consistent with the literature that shows Cu has a
lower light absorption compared to Au due to lower absorption coefficient of Cu19, 36, 37. As seen from the
Figure 3, all the metal supported catalysts show increased visible light absorption compared to their
supports.
Figure 2. Transmission Electron Microscopic (TEM) pictures of Au/TiO2 prepared by deposition
precipitation (DP) method

Catalysis Science & Technology
Page 6 of 19
View Article Online
DOI: 10.1039/C4CY01183J
Figure 3. Diffuse reflectance UVꢀvisible absorption spectra for (a) TiO2 supported catalysts: (i) TiO2
support, (ii) Au/TiO2 (DP), (iii) Cu/TiO2 (I); (b) CeO2 supported catalysts: (i) CeO2 support and (iii)
Au/CeO2 (DP); and (c) Al2O3 supported catalysts: (i) Al2O3 support, (ii) Au/ Al2O3 (DP)
3.2 Activity of different catalysts for plasmonꢀenhanced RWGS reaction
The activity of the catalysts was measured at 400 °C and 2:1 H2 to CO2 molar ratio. The Au/TiO2 (DP)
catalyst had the highest activity for RWGS reaction with and without light as seen in Table 1. The activity
of Au catalysts for water gas shift reaction has been reviewed extensively in the literature38ꢀ40. It is known
that metallic Au by itself is a poor catalyst for WGS reaction but when supported on oxides supports like
TiO2 and CeO2, the activity of Au catalysts increases41ꢀ44. Similar results were obtained in this study in
dark conditions, in which Au/TiO2 and Au/CeO2 showed the highest activity, while Au/Al2O3 showed low
activity. The TiO2, CeO2, and Al2O3 supports, by themselves, exhibited low activity for RWGS reaction.
All of the metal supported catalysts showed higher activity under light compared to dark. The plasmonic
enhancement of the CO2 conversion rate was described by two parameters: (i) the difference between
reaction rates under light and dark (rate difference), and (ii) the ratio between the reaction rate under light
to dark (rate enhancement). The Au/CeO2 had the maximum rate difference between the light and dark
reaction rates of 761 µmol/gmꢀcat/min at 400 °C followed by Au/TiO2 (DP). Au/Al2O3 (DP) and Cu/TiO2
(I) catalysts showed low activity in both light and dark conditions that were comparable to the supports
themselves. The Au/CeO2 catalyst showed the maximum rate enhancement of 2.2, followed by Au/Al2O3
(DP), Au/TiO2 (DP) and Cu/TiO2 (I) catalysts. All of these catalysts showed at least 30% higher rates
under light compared to dark conditions. Because of the inherent high activity of Au/TiO2 (DP) catalyst
under light and dark, the performance of Au/TiO2 catalyst was investigated further.

Page 7 of 19
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01183J
Table 1. Activity of different catalysts for plasmonꢀenhanced reverse water gas shift reaction.
Experimental conditions: P= 110 psi, T= 400 °C, PP =36.7 psi, PPꢃ = 73.3 psi, Total flow rate= 15
ꢀꢁ
ꢂ
ꢂ
cm3(STP)/min
Metal Metal particle Catalyst
CO2 conversion rate
(µmol/gmꢀcat/min)
Sr. No.
Catalyst
loadinga
(wt%)
sizeb
(nm)
amount
Enhancement
(mg)
Dark
2033
Light (LightꢀDark) (Light/Dark)
1 AuꢀTiO2 (DP)
1.4
3.5±1.0
7.4
2663
1417
118
25
630
761
42
6
1.3
2.2
1.5
1.3
0.9
1.0
1.1
2 AuꢀCeO2 (DP)
3 AuꢀAl2O3 (DP)
4 CuꢀTiO2 (I)
5 TiO2
3.3
0.7
1.3
0
5.0±1.2
10.9±4.7
15 to 30c
NA
12.8
16.6
8.6
656
77
20
21
21
67
12.2
23.9
30.0
19
ꢀ2
6 CeO2
0
NA
22
1
7 Al2O3
0
NA
74
6
aAu loading was measure by ICP analysis, Cu loading was calculated from amount of Cu precursor used
during synthesis; bParticle size of Au was estimated using TEM analysis; cParticle size Cu estimated
based on literature Ref21; NA: Not Applicable
3.3 Reaction kinetics studies on Au/TiO2 catalysts
The effect of heat and mass transport were estimated for our system using the dimensionless criteria
shown in Table S145ꢀ47. Reaction data (i.e. reaction rate, activation energy, heat of reaction, reaction
order) under dark conditions at 200 °C, 110 psi pressure and a 2:1 molar H2 to CO2 ratio were used for
these calculations. Heat and mass transfer coefficients and material properties were estimated using
correlations available in the literature47ꢀ49. A catalyst particle size of 150 µm was used for these
calculations. This particle size was estimated from sieving analysis. All of the computed values for
estimating intraparticle and interphase heat and mass transfer limitations are smaller than the controlling
regime criteria. For example, the calculated intraparticle mass transfer number or WeiszꢀPrater number
for the system was 5.9×10ꢀ8 suggesting the absence of intraparticle pore diffusion limitations46, 50. The
pore diffusion limitations are dominant for WeiszꢀPrater numbers above 0.346, 50. Thus, these reactions
were carried out in the absence of any mass or heat transport limitations.

Catalysis Science & Technology
Page 8 of 19
View Article Online
DOI: 10.1039/C4CY01183J
Figure 4: Dependence of CO2 conversion rate on total gas flow rate in plasmon enhanced RWGS reaction
over Au/TiO2 catalyst under light and dark conditions. Experimental conditions: P= 110 psia, T=200 °C,
PP =36.7 psi, PPꢃ = 73.3 psi, catalyst amounts= 7 mg and 3.3 mg.
ꢀꢁ
ꢂ
ꢂ
The reactor was operated in differential mode at low conversions, typically less than 5% and also less
than 30% of the thermodynamic limit. For example, at 200oC the conversion was about 1% which is
below the thermodynamic limit of 9%. At 300oC the conversion was about 4% which is below the
thermodynamic limit of 21%. The CO2 conversion rate was independent with respect to total gas flow rate
under both light and dark as shown in Figure 4. These experiments are consistent with our calculations
that suggest that external mass and heat transfer were negligible in these studies. To investigate light
transfer limitations, the depth of the catalyst bed was varied by loading 3.3 and 7 mg of the catalyst. As
seen by Figure 4, the reaction rates under light for 3.3 and 7 mg catalyst loading were similar. This result
indicates that even at higher 7 mg catalyst loading, most of the catalyst bed was illuminated by the light.
Thus, we ensured that the reactor was operated in absence of mass, heat and light transport limitations to
obtain intrinsic kinetic data for the reaction.
3.3.1 Effect of temperature on the activity of Au/TiO2 catalyst for RWGS
The temperature of the reaction was varied between 100 to 400 °C at constant pressure as shown in
Figure 5 (a). We observed that the reaction rate was higher with light compared to dark at all temperatures
tested. Figure 5 (b) shows that the rate enhancement decreased as the temperature increased from 100 to
400 °C. These results indicate that the plasmonic effect becomes less dominant with increasing
temperature. The rate enhancement below 175 °C cannot be calculated because the catalytic activity
under dark condition at this temperature is too low to be detected with GC. Figure 5 (c) shows that the
difference between light and dark reactions increases with temperature in the temperature range of 100 to
325 °C and decreases beyond 325 °C.

Page 9 of 19
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01183J
Figure 5. Temperature dependence of CO2 conversion rate plasmonꢀenhanced reverse water gas shift
reaction over Au/TiO2 (DP) catalyst: (a) CO2 conversion rate vs temperature under light and dark
conditions; (b) Rate enhancement i.e. ratio of CO2 conversion rate under light to CO2 conversion rate
under dark condition vs temperature; (c) Difference in CO2 conversion rate between light and dark
conditions vs temperature. Experimental conditions: P= 110 psi, PP =36.7 psi, PPꢃ = 73.3 psi, Total
ꢀꢁ
ꢂ
ꢂ
flow rate= 15 cm3(STP)/min, catalyst amount= 7.4 mg
The Au/TiO2 catalyst was characterized before and after reaction using UVꢀVis spectroscopy and TEM
imaging. Figure 6 (a) shows that the Au LSPR peak slightly redꢀshifted after the reaction. Figure 6 (b)
shows that the Au particle size distribution increases slightly after reaction. These results indicate that the
catalyst did not undergo major structural change during the reaction.
Figure 6. (a) Diffuse reflectance UVꢀvisible absorption spectra for Au/TiO2 (DP) fresh and Au/TiO2 (DP)
spent catalyst; (b) particle size distribution obtained from TEM image analysis of Au/TiO2 (DP) fresh and
spent catalysts

Catalysis Science & Technology
Page 10 of 19
View Article Online
DOI: 10.1039/C4CY01183J
The reaction rate versus temperature data were plotted in an Arrhenius plot as seen in Figure 7. The
apparent activation energy for the RWGS reaction over Au/TiO2 catalyst decreased from 47 kJ/mol under
dark conditions to 35 kJ/mol with visible light. This change in apparent activation energy suggests that
the LSPR is changing the energetics of the reaction. One theory that has been proposed to explain these
effects is that plasmons induce localized heating31, 51ꢀ53. Govorov and Richardson estimated temperature
rises of less than 1 °C for colloidal gold nanoparticle solution under similar light intensities54. In our
system, the heat dissipation from the plasmonic Au nanoparticle would be even faster due to faster
conductive heat transfer in TiO2 support compared to convective heat transfer in colloidal Au solutions.
Thus, the local temperature rise in our system should be even smaller than 1 °C. To observe such a large
increase in rate due to localized heating effect, a local temperature rise of greater than 50 °C would be
needed (see Figure 5 (a)). Hence, localized heating effects cannot explain the plasmonic enhancement of
the reaction rate. In addition, if only localized heating were responsible for this enhancement, then the
apparent activation energy for the reaction would remain the same under light and dark conditions with
such small temperature rises.
Figure 7. Arrhenius plot for CO2 conversion rate under light and dark conditions for plasmonꢀenhanced
reverse water gas shift reaction over Au/TiO2 (DP) catalyst. Experimental conditions: P= 110 psi,
PP =36.7 psi, PPꢃ = 73.3 psi, Total flow rate= 15 cm3(STP)/min, catalyst amount= 7.9 mg
ꢀꢁ
ꢂ
ꢂ
3.3.2 Effect of CO2 and H2 partial pressure on activity of Au/TiO2 catalyst for RWGS reaction
The dependence of the reaction rate on CO2 and H2 partial pressures was investigated to gain more
insights into how plasmons influence the reaction mechanism. Figure 8 (a) shows the dependence of the
reaction rate on the CO2 partial pressure under light and dark conditions at constant H2 partial pressure.
The reaction order with respect to CO2 changes from 0.5 under dark conditions to 1.0 under light. These
results can be interpreted using HougenꢀWatson type models for heterogeneous catalytic reaction based
on Langmuir adsorption47, 55. According to this model, a reaction order of less than 1 indicates the
presence of adsorbed surface species inhibiting the rate of the reaction47. Thus, under dark conditions, the
reaction rate could be inhibited by adsorbed CO2 species or intermediates produced from the CO2 on the
catalytic surface. The increase in order with respect to CO2 to 1 under light suggests that the inhibition

Page 11 of 19
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01183J
due to surface coverage of adsorbed species is reduced by the LSPR. Under dark conditions, temperature
had little effect on the apparent reaction order with respect to CO2. Similar apparent reaction orders with
respect to CO2 were obtained at 260 °C (n=0.6) and 200 °C (n=0.5) under dark conditions.
Figure 8 (b) shows the dependence of reaction rate on H2 partial pressure under light and dark conditions
at constant CO2 partial pressure. As seen from Figure 8 (b), the reaction rate is a weak function of the H2
partial pressure. The order of the reaction with respect to H2 only slightly changes from 0.07 in dark
conditions to 0.17 under light. In HougenꢀWatson models, such behavior is observed under high surface
coverage of adsorbed species when the reacting molecule is readily adsorbed on the catalyst surface47.
This observation is consistent with FTIR studies by Boccuzzi and coꢀworkers, who showed that H2
dissociatively adsorbed strongly on edge and corner Au sites56. In the study, they found that formation of
AuꢀH and AuꢀOH species due to adsorption of hydrogen on metallic Au sites and sites in contact with
oxide support56. Spillover of H atoms was also observed in their study with monotonous increase in H
adsorption with contact time and pressure56.
Figure 8. Dependence of CO2 conversion rate in plasmonꢀenhanced reverse water gas shift reaction on
(a) CO2 partial pressure and (b) H2 partial pressure under light and dark conditions over Au/TiO2 (DP)
catalyst. Experimental conditions: P= 104 psi, Total flow rate= 15 cm3(STP)/min, T=200 °C, catalyst
amount= 7.9 mg, partial pressure of the other reactant kept constant at 52 psi using He as an inert gas.
3.3 Light to chemical energy efficiency
We calculate the light to chemical energy efficiency by assuming that the difference in reaction rates
under light and dark conditions results from the light input given by Equation 2. The difference in CO2
conversion rate (LightꢀDark) is the difference between reaction rates under light and dark conditions.
∆Hreaction is the heat of the reaction (∆Hreaction = 41.27 kJ/mol), Intensity is the intensity of light input to the
reactor (5216 W/m2) and catalyst surface area is the cross sectional area of the catalyst bed exposed to
incoming light radiation. The heat of reaction changes slightly from 41.27 kJ/mol at room temperature to
38.35 kJ/mol at 400 °C. This weak dependence of heat of the reaction on temperature was neglected in
these calculations. We further validated this assumption by measuring the energy input to the heater from
the temperature controller. The energy input from the heater was, in fact, lower under light compared to
dark conditions. This result indicated that the light input did not just catalyze the reaction but also
provided the energy for the endothermic reaction.

Catalysis Science & Technology
Page 12 of 19
View Article Online
DOI: 10.1039/C4CY01183J
ꢑꢒꢓꢓꢔꢕꢔꢖꢗꢔ ꢒꢖ ꢘꢙ ꢗꢚꢖꢛꢔꢕꢜꢒꢚꢖ ꢕꢝꢞꢔ ꢎꢟꢒꢠꢡꢞ ꢢ ꢑꢝꢕꢣꢏ ꢤ ∆ꢥꢦꢧꢨꢩꢪꢫꢬꢭ
2
ꢄꢅꢆꢇꢈ ꢉꢊꢊꢅꢋꢅꢉꢌꢋꢍ ꢎ%ꢏ ꢐ
ꢤ 100%
ꢮꢖꢞꢔꢖꢜꢒꢞꢯ ꢤ ꢘꢝꢞꢝꢰꢯꢜꢞ ꢜꢱꢕꢓꢝꢗꢔ ꢝꢕꢔꢝ
(2)
In Figure 9 (a) we estimate the corresponding light to chemical efficiency. The light efficiency increases
with temperature up to 325 °C and then decreases beyond 325 °C. The maximum efficiency for the
process is obtained at 325 °C as 4.5%. This efficiency is higher than values reported in the literature for
other photocatalytic CO2 reduction processes57. Most photocatalytic CO2 reduction approaches focus on
aqueous phase conversion in an electrochemical cell using UV light. The efficiencies of these processes
are less than 1% even with UV light57. Thus, the approach in this work provides a promising alternative
towards utilization of CO2 to reduced forms using solar energy. Figure 9 (b) shows the dependence of
light to chemical efficiency on H2:CO2 molar ratio at 200 °C. Because of the higher reaction order with
respect to CO2, a lower H2:CO2 results in higher light efficiency. At low H2:CO2 ratio of 0.25, we
observed the highest light to chemical efficiency of 4.7%. At the same conditions, we obtained 13 times
enhancement of reaction rate due to LSPR as shown in Figure S2.
Figure 9. (a) Difference in CO2 conversion rate between light and dark (Conversion rate due to light) vs.
temperature in plasmonꢀenhanced reverse water gas shift reaction over Au/TiO2 catalyst (primary axis)
and Light to chemical efficiency corresponding to the conversion rate due to light (secondary axis).
Experimental conditions: P= 110 psi, PP =36.7 psi, PPꢃ = 73.3 psi, Total flow rate= 15 cm3(STP)/min,
ꢀꢁ
ꢂ
ꢂ
catalyst amount= 7.4 mg. (b) Dependence of difference in reaction rates between light and dark reactions
(primary axis) and light efficiency (secondary axis) on H2:CO2 ratio in plasmon enhanced reverse water
gas shift reaction over Au/TiO2 catalyst. Experimental conditions: P= 103 psi, T=200 °C, Total gas flow
rate= 15 cm3(STP)/min, catalyst amount= 7.9 mg.
4. Discussion
Three main effects have been reported in the literature to occur to explain how LSPR influences the
reaction chemistry as shown in Figure 10. These effect are: (1) Intense scattering of the light
(electromagnetic radiation) around the metal nanoparticle58, 59, (2) electronꢀhole pair generation through
electron excitation from the metal nanoparticle21, 30, 60ꢀ62, and (3) localized heating effect resulting from
dissipation of energy in phonons31, 51, 52
.

Page 13 of 19
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01183J
Figure 10. Effects of LSPR resulting in enhanced activity of plasmonic photocatalysts
Our reaction kinetics results show that the apparent activation energy and apparent reaction order with
respect to CO2 changes when visible light is used for Au/TiO2 catalyzed RWGS reaction. These results
suggest that the surface plasmons change the energetics of this reaction which could occur by two
different processes: 1) the surface plasmons could modify the transition state of the rate determining steps
or 2) the surface plasmons can modify the heat of adsorption of different reactants or intermediates. The
reaction mechanism of WGS and RWGS has been extensively studied under dark conditions38, 41ꢀ43, 63ꢀ67. It
is known that the WGS activity of Au catalysts is enhanced when it is supported on an oxide support like
TiO2 and CeO241ꢀ44. It has been proposed that the oxide support increases the activity of the Au catalysts
by catalyzing the water dissociation step42, 43, 68. These studies propose that the WGS reaction over
Au/TiO2 goes through carboxyl and hydroxyl intermediates42, 43, 68. Burch showed that the dominant
reactive intermediates depend upon the H2O:CO2 ratio in the reacting mixture38, 69. Under high H2O:CO2
ratio, hydroxyl species dominate the surface coverage; whereas under low H2O:CO2 ratios,
carbonate/carboxyl species dominate the surface coverage38. It has also been proposed that the WGS
reaction takes place at the interface of the gold and oxide support42, 68. Ribeiro and coꢀworkers have
shown that the activity of Au/TiO2 catalysts for WGS reaction changes with Au particle size70, 71. They
concluded that the corner and peripheral sites are the most active sites for the WGS reaction70.
Consistent with these literature studies, Burch proposed a mechanism for dry RWGS reaction (low
H2O:CO2 ratio) on Au/TiO2 catalyst under dark conditions, as shown in Figure 11 and with Equations Iꢀ
V38. The reaction proceeds by five elementary steps: (I) first, hydrogen dissociatively adsorbs on gold
sites, (II) CO2 adsorbs on oxygen vacancies on the metal oxide, (III) adsorbed H atoms react with
adsorbed CO2 molecule to form carboxyl species at sites on the interface of the metal oxide and gold, (IV)
the carboxyl species dissociate to give gaseous CO and adsorbed OH on metal oxide sites, and (V) the
adsorbed OH reacts with another adsorbed H atom to form H2O at interfacial Au and metal oxide sites.
These elementary steps are represented by Equations IꢀV, where * is a metal oxide (TiO2) site, and x is a
Au site. The equations for the surface coverage of adsorbed carbon dioxide and hydrogen are given by
Equations 4 and 5, respectively. These equations assume that adsorption Steps I and II are in quasiꢀ
equilibrium. Equating the rates of Steps III, IV, and V, a rate expression for RWGS reaction can be
derived as given by Equation 6.
ꢥꢲ ꢳ ꢴ ↔ 2 ꢤ ꢴꢥ
ꢘꢙꢲ ꢳ ∗ ↔ ꢘꢙꢲ∗
(I)
(II)
(III)
ꢘꢙꢲ∗ ꢳ ꢴꢥ → ꢘꢙꢙꢥ∗ ꢳ ꢴ

Catalysis Science & Technology
Page 14 of 19
View Article Online
DOI: 10.1039/C4CY01183J
ꢘꢙꢙꢥ∗ → ꢘꢙ ꢳ ꢙꢥ∗
(IV)
(V)
ꢙꢥ∗ ꢳ ꢴꢥ → ꢥꢲꢙ ꢳ ꢴ ꢳ ∗
ꢵꢀꢁ ꢐ ꢶꢲꢷꢀꢁ ꢵꢸ
(4)
(5)
ꢂ
ꢂ
ꢺ ꢼ
ꢹ
ꢻ
ꢽ
ꢂ
ꢴꢃ ꢐ ꢾꢿ ꢺ ꢼ
ꢹ
ꢻ
ꢽ
ꢂ
ꢺ ꣅ ꢼ
꣉
ꢂ
꣆
꣇꣈
ꢽ
ꢻ
ꢂ
ꣀꣁꣂꣃ꣄ ꢐ ꢾꢿꢺ ꢼ
(6)
꣌
꣊
꣍
ꢽ
ꢎꢾꢿꣅ ꢎ
ꢿ
ꢏꢏ
ꢂ
꣇꣈
꣆
ꢂ
꣊
꣋
Table 2. Reaction rate equations for RWGS on Au/TiO2 for various MASI assumptions
Case MASI Rate equation under dark Proposed result of LSPR Rate equation under light
ꢙꢥ∗
ꢣ꣐ ↑
ꢶꢲꢣ꣑ꢷꢀꢁ ꢴꢃ
꣉
ꢽ
ꢂ
ꣀꣁꣂꣃ꣄ ꢐ ꢺ ꣅ ꢼ
꣊ ꣏
꣆
꣇꣈
(7)
ꢂ
ꢂ
꣆
꣇꣈
1
꣎
ꢂ
ꢂ
ꢾꢿ
꣊
꣋
ꢘꢙꢙꢥ∗
ꢣ꣒ ↑
ꢶꢲꢣ꣑ꢷꢀꢁ ꢴꢃ
ꢺ ꣅ ꢼ
꣉
ꢂ
꣆
꣇꣈
꣏
ꢽ
꣌
2
ꢂ
ꣀꣁꣂꣃ꣄
ꢐ
(8)
ꢂ
꣎
ꢂ
꣊
꣆
꣇꣈
ꢽ
ꢂ
ꢾꢿ
꣊
꣍
Equation 6 can be further simplified by making assumptions about the most abundant surface
intermediate (MASI). Table 2 shows the resulting rate equations assuming that either the OH* or COOH*
is the MASI. The rate of RWGS reaction is given by the Equation 7 which assumes that OH* is the
MASI. This expression can result in fractional order with respect to CO2 if the ratio K2k3PCO2/k5≥1. The
apparent rate order with respect to CO2 can change to 1.0, under light, if the LSPR increases the rate
constants for Step V (i.e. k5) more than any other steps. In case 2, COOH* is considered to be the MASI.
With this assumption, the rate of RWGS reaction is given by Equation 8. If K2k3PCO2xH/k4≥1, then a
fractional order with respect to CO2 can be obtained. In this case, the apparent rate order with respect to
CO2 can change to 1.0, under light, if the LSPR increase the rate constant for Step IV (i.e. k4) more than
any other steps. The effect of light can be further evaluated by fitting the experimental data to a simple
two constant linear rate equation shown in Equation 9. Table 3 illustrate how the rate and equilibrium
constants in Equations 7 and 8 correspond to the slope and intercept in Equation 9.
ꢾ
ꢐ ꣗꣘ꢬ꣙ꢧ ꢳ intercept
(9)
ꣁ
ꢼ
꣇꣈
꣓꣔꣕꣖
ꢂ
Table 3. Interpretation of slope and intercept given in Equation 9 to kinetic parameters under different
MASI assumptions
OH MASI
COOH MASI
1
1
slope
ꢶꢲꢣ꣑ꢴꢃ
1
ꢶꢲꢣ꣑ꢴꢃ
1
intercept
ꢣ꣐ꢴꢃ
ꢣ꣒
The reaction kinetics data from Figure 8a at 200 °C were then fit to Equation 9. The fitting results are
shown in parity plot given in Figure S3. The slope of the fit changes from 0.236±0.015 under dark to
0.111±0.002 under light. This change in slope corresponds to 2.1 times enhancement in the product of K2,

Page 15 of 19
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01183J
k3 and xH. At the same time the intercept for the fit changed from 0.015±0.001 under dark to
0.00006±0.0002 under light. The change in the intercept corresponded to 242.9 times enhancement of
either k4 or k5xH. Thus, the LSPR affects the rates of Steps IV or V more than any other step involved in
the catalytic mechanism. This increase in rate constants with LSPR corresponds to a decrease in
activation energy of either Step IV or V of 21.6 kJ/mol.
In Figure 11 we illustrate the two possible mechanisms through which LSPR can increase these
elementary Steps, IV and V. Hot electron generated from LSPR of Au can get transferred to the adsorbed
OH or COOH species which accelerates the desorption of H2O and CO, respectively. The electron affinity
values of OH (1.827 eV)72 and COOH (1.510 eV)73 are higher than other species involved in this reaction
mechanism (e.g. electron affinity of CO2 is ꢀ0.922 eV and electron affinity of H is 0.754 eV)73, 74. Higher
electron affinities of OH and COOH intermediates suggest that these species are more likely to accept the
hot electrons generated by the LSPR and form transient negative ion (TNI). The formation of these TNI
states has been proposed as one of the possible mechanisms for plasmonic enhancement in the recent
studies20, 75. Alternatively, intense electromagnetic fields around Au nanoparticle can polarize the OꢀH or
CꢀO bonds present in the OH* and COOH* intermediates as shown in Figure 11. This effect can facilitate
the bond breaking steps and result in higher desorption rates. Both of these mechanisms could lower
surface coverage of adsorbed species and explain changes in apparent rate orders and activation energy
due to LSPR.
Figure 11. Proposed mechanism for plasmonic enhancement of reverse water gas shift reaction over
Au/TiO2 catalyst

Catalysis Science & Technology
Page 16 of 19
View Article Online
DOI: 10.1039/C4CY01183J
5. Conclusion
We demonstrate that the localized surface plasmon resonance can increase the RWGS activity of Au
supported catalysts by 30 to 1300% using visible light illumination. The enhancement was a function of
the type of oxide support used. Au/TiO2 and Au/CeO2 showed the highest activity for RWGS reaction. A
visible light to chemical efficiency of up to 5% for plasmonic RWGS reaction over Au/TiO2 catalyst was
observed. The LSPR causes a shift in the energetics of the reaction mechanism for the Au/TiO2 catalyst.
The apparent activation energy decreased from 47 kJ/mol to 35 kJ/mol with visible light for this catalyst.
The reaction order with respect to CO2 increased from 0.5 without visible light to 1.0 with visible light.
The reaction order with respect to H2 was similar with and without light. These results demonstrate that
the LSPR changes the intrinsic reaction kinetics on the catalyst surface. Our analysis suggests that the
LSPR increases the rate constants for either carboxyl decomposition or hydroxyl hydrogenation more
than any other step in the RWGS mechanism. Both of these steps are proposed to occur at the interface of
the Au and TiO2. This rate enhancement could occur via either hot electron generation mechanism or
adsorbate polarization mechanism resulting from LSPR.
6. Acknowledgements
This research was funded by the Advanced Research Projects AgencyꢀEnergy (ARPAꢀE), U.S.
Department of Energy, under Award Number DEꢀAR0000329. We thank HansꢀJoachim Freund and
Manos Mavrikakis for helpful discussions about this topic.
Abbreviations
BID: Barrierꢀdischarge ionization detector
DFT: Density functional theory
DP: Deposition precipitation
FTIR: Fourier transform infrared spectroscopy
GC: Gas chromatograph
I: Impregnation
ICP: Inductively coupled plasma
LSPR: Localized surface plasmon resonance
MASI: Most abundant surface intermediate
PID: Proportional integral derivative
RWGS: Reverse water gas shift
TEM: Transmission electron microscopy
TNI: Transient negative ion
UHP: Ultra high purity
UV: Ultraviolet
WGS: Water gas shift
Nomenclature
∆H: Heat of reaction
Biw: Biot number at the wall
Cs: Concentration of reactant at the external surface of the catalyst particle
De: Effective diffusivity
E: Activation energy of the reaction
h: Heat transfer coefficient

Page 17 of 19
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01183J
Kc: Massꢀtransfer coefficient between the catalyst and bulk phases
ki: forward rate constant for elementary step i
Ki: Equilibrium rate constant for elementary step i
ke: Effective thermal conductivity of the catalyst bed
n: Reaction order
Pi: Partial pressure of reactant i
r''': Reaction rate per catalyst volume
R: Universal gas constant
Ro: Outer radius of reactor
rp: Catalyst particle radius
Tb: Temperature of bulk phase
Ts: Catalyst surface temperature
Tw: Reactor wall temperature
xH: Surface coverage of H on Au
θi: Surface coverage of species i on the catalyst support
References
1.
2.
N. S. Lewis and G. Crabtree, Basic Research Needs for Solar Energy Utilization, 2005.
N. S. Lewis and D. G. Nocera, Proceedings of the National Academy of Sciences, 2006, 103,
15729-15735.
3.
4.
5.
NREL, Best Research-Cell Efficiencies, http://www.nrel.gov/ncpv/images/efficiency_chart.jpg,
Accessed 03/03/2014, 2014.
M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A. Santori and N. S. Lewis,
Chem. Rev., 2010, 110, 6446-6473.
K. Maeda, K. Teramura, D. L. Lu, T. Takata, N. Saito, Y. Inoue and K. Domen, Nature, 2006, 440,
295-295.
6.
7.
A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253-278.
A. Fujishima and K. Honda, Nature, 1972, 238, 37-38.
8.
M. Gratzel, Nature, 2001, 414, 338-344.
9.
G. Centi, E. A. Quadrelli and S. Perathoner, Energy Environ. Sci., 2013, 6, 1711-1731.
G. Centi, G. Iaquaniello and S. Perathoner, ChemSusChem, 2011, 4, 1265-1273.
G. Centi and S. Perathoner, Catal. Today, 2009, 148, 191-205.
A. Corma and H. Garcia, J. Catal., 2013, 308, 168-175.
C. Bartholomew and R. J. Farrauto, Fundamentals of Industrial Catalytic Processes, Wiley, New
Jersey, 2006.
10.
11.
12.
13.
14.
15.
U. N. D. Program, World Energy Assessment Report: Energy and the Challenge of Sustainability,
United Nations, New York, 2003.
X. H. Wang, J. G. Li, H. Kamiyama, Y. Moriyoshi and T. Ishigaki, Journal of Physical Chemistry B,
2006, 110, 6804-6809.
16.
17.
F. E. Osterloh, Chem. Soc. Rev., 2013, 42, 2294-2320.
A. K. L. Sajjad, S. Shamaila, B. Z. Tian, F. Chen and J. L. Zhang, Journal of Hazardous Materials,
2010, 177, 781-791.
18.
19.
20.
21.
M. Grätzel, Inorganic Chemistry, 2005, 44, 6841-6851.
S. Linic, P. Christopher and D. B. Ingram, Nature Materials, 2011, 10, 911-921.
P. Christopher, H. Xin and S. Linic, Nat Chem, 2011, 3, 467-472.
F. Sastre, M. Oteri, A. Corma and H. Garcia, Energy Environ. Sci., 2013, 6, 2211-2215.

Catalysis Science & Technology
Page 18 of 19
View Article Online
DOI: 10.1039/C4CY01183J
22.
S. Mukherjee, F. Libisch, N. Large, O. Neumann, L. V. Brown, J. Cheng, J. B. Lassiter, E. A. Carter,
P. Nordlander and N. J. Halas, Nano Letters, 2012, 13, 240-247.
W. B. Hou and S. B. Cronin, Advanced Functional Materials, 2013, 23, 1612-1619.
Z. Liu, W. Hou, P. Pavaskar, M. Aykol and S. B. Cronin, Nano Letters, 2011, 11, 1111-1116.
W. H. Hung, M. Aykol, D. Valley, W. Hou and S. B. Cronin, Nano Letters, 2010, 10, 1314-1318.
M. I. Stockman, Opt. Express, 2011, 19, 22029-22106.
P. Christopher, H. Xin, A. Marimuthu and S. Linic, Nat Mater, 2012, 11, 1044-1050.
K. A. Willets and R. P. Van Duyne, in Annual Review of Physical Chemistry, Annual Reviews, Palo
Alto, 2007, vol. 58, pp. 267-297.
23.
24.
25.
26.
27.
28.
29.
L. Wang, C. Clavero, Z. Huba, K. J. Carroll, E. E. Carpenter, D. Gu and R. A. Lukaszew, Nano
Letters, 2011, 11, 1237-1240.
30.
31.
X. M. Zhang, Y. L. Chen, R. S. Liu and D. P. Tsai, Reports on Progress in Physics, 2013, 76.
C. Wang, O. Ranasingha, S. Natesakhawat, P. R. Ohodnicki, Jr., M. Andio, J. P. Lewis and C.
Matranga, Nanoscale, 2013, 5, 6968-6974.
32.
33.
R. Zanella, S. Giorgio, C. H. Shin, C. R. Henry and C. Louis, J. Catal., 2004, 222, 357-367.
R. Zanella, S. Giorgio, C. R. Henry and C. Louis, Journal of Physical Chemistry B, 2002, 106, 7634-
7642.
34.
35.
D. L. Nguyen, S. Umbarkar, M. K. Dongare, C. Lancelot, J. S. Girardon, C. Dujardin and P. Granger,
Catalysis Communications, 2012, 26, 225-230.
F. Boccuzzi, A. Chiorino, G. Martra, M. Gargano, N. Ravasio and B. Carrozzini, J. Catal., 1997, 165,
129-139.
36.
37.
38.
39.
40.
41.
X. Zhou, G. Liu, J. Yu and W. Fan, Journal of Materials Chemistry, 2012, 22, 21337-21354.
A. Marimuthu, J. Zhang and S. Linic, Science, 2013, 339, 1590-1593.
R. Burch, Phys. Chem. Chem. Phys., 2006, 8, 5483-5500.
M. Haruta and M. Daté, Applied Catalysis A: General, 2001, 222, 427-437.
M. S. Chen and D. W. Goodman, Catalysis Today, 2006, 111, 22-33.
J. A. Rodriguez, S. D. Senanayake, D. Stacchiola, P. Liu and J. Hrbek, Accounts of Chemical
Research, 2013.
42.
J. A. Rodriguez, J. Evans, J. Graciani, J.-B. Park, P. Liu, J. Hrbek and J. Fdez Sanz, Journal of
Physical Chemistry C, 2009, 113, 7364-7370.
43.
44.
45.
J. A. Rodriguez, S. Ma, P. Liu, J. Hrbek, J. Evans and M. Perez, Science, 2007, 318, 1757-1760.
Q. Fu, H. Saltsburg and M. Flytzani-Stephanopoulos, Science, 2003, 301, 935-938.
D. E. Mears, Industrial & Engineering Chemistry Process Design and Development, 1971, 10, 541-
547.
46.
M. E. Davis and R. J. Davis, Fundamentals of Chemical Reaction Engineering, McGraw-Hill Higher
Education, 2002.
47.
48.
49.
M. Vannice, Kinetics of Catalytic Reactions, Springer New York, 2005.
, Aspen Plus v7.3, Aspen Technology, Inc., Cambridge, MA, 2011.
A. Cybulski, M. J. Van Dalen, J. W. Verkerk and P. J. Van Den Berg, Chemical Engineering Science,
1975, 30, 1015-1018.
50.
P. B. Weisz and C. D. Prater, in Advances in Catalysis, eds. V. I. K. W.G. Frankenburg and E. K.
Rideal, Academic Press, 1954, vol. Volume 6, pp. 143-196.
51.
52.
53.
54.
55.
C. W. Yen and M. A. El-Sayed, Journal of Physical Chemistry C, 2009, 113, 19585-19590.
J. R. Adleman, D. A. Boyd, D. G. Goodwin and D. Psaltis, Nano Lett., 2009, 9, 4417-4423.
C. Tabor, W. Qian and M. A. El-Sayed, Journal of Physical Chemistry C, 2007, 111, 8934-8941.
A. O. Govorov and H. H. Richardson, Nano Today, 2007, 2, 30-38.
O. Hougen and K. Watson, Chemical Process Principles. Part 3. Kinetics and Catalysis, Wiley, New
York, 1947.

Page 19 of 19
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01183J
56.
57.
58.
M. Manzoli, A. Chiorino, F. Vindigni and F. Boccuzzi, Catalysis Today, 2012, 181, 62-67.
S. C. Roy, O. K. Varghese, M. Paulose and C. A. Grimes, ACS Nano, 2010, 4, 1259-1278.
F. Le, D. W. Brandl, Y. A. Urzhumov, H. Wang, J. Kundu, N. J. Halas, J. Aizpurua and P.
Nordlander, Acs Nano, 2008, 2, 707-718.
59.
60.
S. L. Zou and G. C. Schatz, Chemical Physics Letters, 2005, 403, 62-67.
J. Sa, G. Tagliabue, P. Friedli, J. Szlachetko, M. H. Rittmann-Frank, F. G. Santomauro, C. J. Milne
and H. Sigg, Energy & Environmental Science, 2013, 6, 3584-3588.
K. H. Kim, K. Watanabe, D. Menzel and H.-J. Freund, Surf. Sci., 2012, 606, 1142-1151.
K. Watanabe, D. Menzel, N. Nilius and H.-J. Freund, Chem. Rev., 2006, 106, 4301-4320.
N. Schumacher, A. Boisen, S. Dahl, A. A. Gokhale, S. Kandoi, L. C. Grabow, J. A. Dumesic, M.
Mavrikakis and I. Chorkendorff, J. Catal., 2005, 229, 265-275.
A. A. Gokhale, J. A. Dumesic and M. Mavrikakis, Journal of the American Chemical Society, 2008,
130, 1402-1414.
R. J. Madon, D. Braden, S. Kandoi, P. Nagel, M. Mavrikakis and J. A. Dumesic, J. Catal., 2011, 281,
1-11.
L. C. Grabow, A. A. Gokhale, S. T. Evans, J. A. Dumesic and M. Mavrikakis, The Journal of Physical
Chemistry C, 2008, 112, 4608-4617.
T. Tabakova, F. B. Boccuzzi, M. Manzoli and D. Andreeva, Appl. Catal. A-Gen., 2003, 252, 385-
397.
A. Hussain, J. Gracia, B. E. Nieuwenhuys and J. W. Niemantsverdriet, ChemCatChem, 2013, 5,
2479-2488.
F. C. Meunier, D. Tibiletti, A. Goguet, D. Reid and R. Burch, Appl. Catal. A-Gen., 2005, 289, 104-
112.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
M. Shekhar, J. Wang, W. S. Lee, W. D. Williams, S. M. Kim, E. A. Stach, J. T. Miller, W. N. Delgass
and F. H. Ribeiro, Journal of the American Chemical Society, 2012, 134, 4700-4708.
W. D. Williams, M. Shekhar, W.-S. Lee, V. Kispersky, W. N. Delgass, F. H. Ribeiro, S. M. Kim, E. A.
Stach, J. T. Miller and L. F. Allard, Journal of the American Chemical Society, 2010, 132, 14018-
14020.
72.
73.
74.
J. R. Smith, J. B. Kim and W. C. Lineberger, Physical Review A, 1997, 55, 2036-2043.
NIST, NIST Chemistry Webbook, 2014.
C.-G. Zhan, J. A. Nichols and D. A. Dixon, The Journal of Physical Chemistry A, 2003, 107, 4184-
4195.
75.
76.
77.
S. Linic, P. Christopher and D. B. Ingram, Nat Mater, 2011, 10, 911-921.
D. E. Mears, J. Catal., 1971, 20, 127-131.
J. B. Anderson, Aiche J., 1962, 18, 147-148.