Surface processes occurring during aqueous phase ethanol reforming on Ru/TiO2 tracked by ATR-IR spectroscopy

Article abstract of DOI:10.1016/j.apcata.2019.05.029

Surface processes occurring at the solid-liquid interface of Ru/TiO₂ catalysts during aqueous phase reforming of ethanol at reaction conditions (200 °C and 20 bar) have been examined using attenuated total reflection infrared spectroscopy (ATR-

Full text of DOI:10.1016/j.apcata.2019.05.029

Accepted Manuscript
Title: Surface Processes Occurring During Aqueous Phase
Ethanol Reforming on Ru/TiO2 Tracked by ATR-IR
Spectroscopy
Authors: Ammara Waheed, Xianwei Wang, Nobutaka Maeda,
Shuichi Naito, Alfons Baiker
PII:
S0926-860X(19)30242-X
DOI:
Reference:
https://doi.org/10.1016/j.apcata.2019.05.029
APCATA 17095
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
7 February 2019
9 May 2019
26 May 2019
Please cite this article as: Waheed A, Wang X, Maeda N, Naito S, Baiker
A, Surface Processes Occurring During Aqueous Phase Ethanol Reforming on
Ru/TiO2 Tracked by ATR-IR Spectroscopy, Applied Catalysis A, General (2019),
https://doi.org/10.1016/j.apcata.2019.05.029
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Surface Processes Occurring During Aqueous Phase Ethanol Reforming on
Ru/TiO2 Tracked by ATR-IR Spectroscopy
Ammara Waheeda, Xianwei Wangb, Nobutaka Maedab*, Shuichi Naitoc, Alfons Baikerd*
a
Gold Catalysis Research Center, State Key Laboratory of Catalysis, Dalian Institute of
Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Dalian 116023, China
b
Key Laboratory of Industrial Ecology and Environmental Engineering, School of
Environmental Science and Technology, Dalian University of Technology, Dalian 116024,
China
c
Department of Material and Life Chemistry, Faculty of Engineering, Kanagawa University,
3-27-1, Rokkakubashi, Kanagawa-ku, Yokohama, 221-8686 Japan
d
Institute for Chemical and Bioengineering, Department of Chemistry and Applied
Biosciences, ETH Zurich, Hönggerberg, HCI, CH-8093 Zurich, Switzerland
Email: alfons.baiker@chem.ethz.ch; nobutaka.maeda@dlut.edu.cn
1

Graphical abstract
Highlights
-
Dynamic changes of surface species on Ru/TiO2 during reforming of ethanol for
hydrogen production were tracked by ATR-IR spectroscopy at 200oC and 20 bar.
Undesired methane formation was favored at high loading (5wt%) and large particle
size (~ 4 nm) of Ru, while on catalyst with low loading (0.5wt) and small particle size
(~ 1nm) it was suppressed.
-
-
-
Intermediate formation of acetylide species was observed on catalyst with high
loading and large particle size of Ru but not on catalyst with low Ru- loading and
small particle size.
Loading and particle size of Ru need to be controlled for optimizing H2 production.
2

Abstract
Surface processes occurring at the solid-liquid interface of Ru/TiO2 catalysts during aqueous
o
phase reforming of ethanol at reaction conditions (200 C and 20 bar) have been examined
using attenuated total reflection infrared spectroscopy (ATR-IR). Depending on the loading
and particle size of Ru significant differences in the population dynamics of adsorbed species
were observed. On the catalyst with high loading (5wt%) and large particle size (~ 4 nm)
formation of intermediate acetylide was detected, which is attributed to enhanced C-O bond
cleavage promoting methanation and thus leading to lower efficiency of H2 generation. In
contrast, on the catalyst with low loading (0.5wt%) and small particle size (~ 1 nm) no acetylide
was detectable, indicating that C-O bond cleavage was suppressed.
Keywords: Aqueous phase reforming; ethanol; hydrogen production; Ru/TiO2; solid-liquid
interface; in situ ATR-IR spectroscopy.
3

1. Introduction
Ever-increasing energy demand necessitates the transition from fossil fuel utilization to
renewable sources such as biomass. Compared to fossil fuels whose usage causes serious
environmental pollution, biomass is considered to be an environmentally benign and highly
efficient energy source for a large number of applications. Certain advantageous features of
biomass, such as low price, carbon neutrality, large storage capability, high commercialization
potential, and milder condition for its conversion, make it practical for fine chemicals synthesis
or hydrogen production [1]. Currently, the dominant source of energy supply in the world is
still fossil fuels. Moreover, biomass resources cover about 10–15% of the current energy
demand, which keeps increasing every year [2]. The production of ethanol from biomass is one
of the best developed technologies and has attracted much attention as a clean transportation
fuel; when ethanol is produced from renewable sources, it can improve energy security,
decrease air pollution and the balance-of-payments deficit [2]. Bio-ethanol is typically
produced by the fermentation of sugar cane, corn grains, and other starch-rich materials as a
mature technology [3].
The catalytic breakdown of cellulose and hemicelluloses chains into their component sugar
molecules results in the formation of ethanol [4]. Hydrogen derived from bio-renewable
sources such as bio-ethanol is considered as a promising future clean energy source [5]. Several
principal pathways to generate hydrogen from oxygenated hydrocarbons have been proposed
including steam reforming and aqueous phase reforming (APR) [2]. In contrast to the steam
reforming, APR, pioneered by Dumesic’s group [6,7] reduces the energy requirements for
generating hydrogen due to elimination of the need to vaporize water and oxygenated
hydrocarbons. Furthermore, the production of H2 and CO2 leads to low levels of CO in a single
reactor catalytic process. Various supported metal catalysts have been examined for APR of
biomass-derived hydrocarbons, including Pt/alumina [6,7], Ni/alumina and Raney Ni-Sn [8,9]
carbon-supported Pt, Pd and Ru [10], and more recently non-promoted and Re promoted
titania-supported Ru [11] Rh, Ir, Re and Pt [12]. Shabaker et al. proposed a mechanism for APR
of ethylene glycol, which involves the cleavage of C-C bonds as well as C-H bonds to form
adsorbed CO followed by the formation of H2 and CO2 via water-gas shift reaction [7].
4

For APR of ethanol on Ru/TiO2 the following plausible reaction steps have been proposed (1-
6) [11]:
Dehydrogenation to acetaldehyde (AcH):
CH3CH2OH → CH3CHO + H2
(1)
AcH subsequently undergoes two different reaction pathways.
Path I: Direct decomposition to form CH4 and CO2:
CH3CHO → CH4 + CO
(2)
(3)
CO + H2O → CO2 + H2
Path 2: Dehydrogenative oxidation to form acetic acid (AcOH) and H2:
CH3CHO + H2O → CH3COOH + H2 (4)
AcOH further reacts with water to form CO2 and H2:
CH3COOH + 2 H2O → 2 CO2 + 4 H2 (5)
To enhance the selectivity for H2 production and minimize the unfavorable methanation, the
APR process needs to be active in the cleavage of C-C bonds and inactive in the breaking of
C-O bonds.
In previous work [11] Ru/TiO2 showed highest catalytic activity but lowest selectivity in
APR of ethanol, acetaldehyde and acetic acid compared to similar supported Rh, Ir and Pt.
Therefore, APR on Ru/TiO2 was chosen as an interesting reaction system for studying
selectivity-controlling factors such as loading and particle size of Ru. Small Ru particles were
found to suppress unfavorable methanation reactions thereby increasing H2 selectivity.
Despite the potential of bio-ethanol APR, information about its mechanism is still limited.
In particular, knowledge about the processes occurring at the solid-liquid interface of the
catalyst under working conditions is lacking. With this in mind, we applied in situ attenuated
total reflection infrared spectroscopy (ATR-IR) to monitor the fate of surface species at the
catalytic solid-liquid interface of two Ru/TiO2 catalysts with different Ru-loadings (5 and 0.5
wt% and mean Ru particle sizes ~ 4 and ~ 1 nm) during APR of ethanol at 200 oC and 20 bar.
2. Experimental
2.1 Catalyst synthesis
5

Ru/TiO2 catalysts with different Ru-loading (0.5 wt% and 5 wt%) were prepared using
conventional impregnation with aqueous solution of ruthenium chloride hydrate (˃ 98 %,
Sigma-Aldrich) and titania (P-25, Sigma-Aldrich) as a support. The suspension of the Ru
precursor solution and TiO2 was placed in a rotary evaporator (RV 10 digital FLEX, IKA) to
remove water. The resulting samples were dried overnight at 100 oC and pretreated with pure
hydrogen in a fixed-bed catalytic microreactor at 350 oC for 5 hours.
2,2 Catalyst characterization
Specific surface areas of the pretreated catalysts were measured by nitrogen adsorption at
77 K using a Micromeritics instrument (ASAP 2020 HD88) and the BET method. Prior to each
o
measurement, the samples were degassed at 150 C for 1 h to remove the adsorbed water on
the surface and in the pores.
Transmission electron microscopy (TEM) was applied to gain information about the size
distribution of the supported Ru particles. TEM images were obtained on a FEI Tecnai G2 F30
microscope operated at 200 kV. The catalyst powders were mounted onto a carbon-coated Pt
grid, which was then placed in the TEM column. Particle size distributions were based on
analyzing 100 particles per sample.
CO chemisorption experiments were carried out using a static adsorption instrument
(Omnisorp100CX, Beckman). Ru-dispersions were estimated from the CO uptake using a
stoichiometric factor CO/Ru of 1.
2.3 Preparation of catalyst layer
For IR spectroscopic analysis, a thin film of the catalyst powder was prepared as follows:
180 mg of catalyst was added to 24 mL of ethanol and the resulting slurry was stirred overnight
to obtain a uniform suspension. 2.0 mL of the slurry was then placed over a square diamond
internal reflection element (IRE, bevel of 45o, 1 mm × 1 mm) located in an ATR-IR reactor cell
o
(Golden Gate TM ATR, SPECAC) and the ethanol was evaporated overnight at 40 C. This
procedure guaranteed that the catalyst layer adhered to the diamond IRE crystal such that no
loss of the catalyst was observed over the course of several hours under stirring conditions in
the liquid phase. The catalyst deposited onto the diamond IRE crystal amounted to 12 mg.
6

2.4 Attenuated total reflection infrared spectroscopy
Attenuated total reflection infrared spectroscopy is a powerful tool for investigating solid-
liquid interfaces [13]. ATR-IR spectra were taken using a Vertex 70V spectrometer (Bruker)
equipped with a liquid nitrogen cooled mercury-cadmium-telluride (MCT) detector (ID316,
ZnSe window) and an optical filter (F321). Spectra were recorded at 4 cm-1 spectral resolution
and 60 kHz scanning velocity and 64 scans (˂ 5 seconds) were coadded to each spectrum.
Before spectral acquisition, the beam line compartment of the spectrometer was vacuumed to
˂ 1 kPa to enhance the IR intensity. A water-cooled glober Mid-IR source was selected to
enhance the amplitude and the signal-to-noise (S/N) ratio.
Prior to each experiment, the catalyst layer on the IRE crystal was pretreated in situ with
o
pure hydrogen (100 mL/min) at 200 C for 1 h. After 1 h the reactor cell was purged with
helium (100 mL/min) for 5 min and cooled down to 25 oC using a thermostat (CD-200F, Julabo).
In the He flow, 5 mL of reaction mixture (10 vol% ethanol in water) was added to the reactor
o
o
under stirring (20 rpm) and heated to 200 C at a ramping rate of 40 C /min. When the
o
temperature reached 200 C, a spectral background was taken, followed by repeated spectral
measurements for 2 h at intervals of 5 min (in total 25 spectra including the background
spectrum recorded at 0 h). The reforming reactions of aqueous solutions of 10 vol%
acetaldehyde (AcH) and 10 vol% acetic acid (AcOH) were performed applying the same
procedure.
3. Results and Discussion
3.1 Catalyst characterization
BET surface areas of 5 wt% Ru/TiO2 and 0.5 wt% Ru/TiO2 were 60 and 69 m2/g,
respectively, indicating that the surface area decreased with higher Ru-loading. Figure 1 shows
representative TEM images and particle size distributions of the 0.5 wt% and 5 wt% Ru/TiO2
catalysts. In the image of the catalyst with low Ru-loading (0.5wt%) (Figure 1a) the crystalline
lattice of TiO2 is discernible. However, determination of the mean Ru particle size was
somewhat impaired due to the interference of dark images of TiO2 particles and relatively low
contrast. The mean Ru particle sizes were estimated to be approximately 1 nm (0.5 wt%
Ru/TiO2) and 4 nm (5 wt% Ru/TiO2), respectively. The Ru-dispersions estimated by CO
7

chemisorption were 86% for 0.5 wt% Ru/TiO2 and 21% for 5wt% Ru/TiO2, respectively,
corroborating the significantly different Ru particle sizes of the catalysts determined by TEM.
Figure 1. TEM images and Ru particle size distributions of (a) 0.5 wt% Ru/TiO2 and (b) 5 wt%
Ru/TiO2
3.2 In situ Attenuated total reflection Infrared spectroscopy
Figure 2 shows ATR-IR spectra recorded during APR of ethanol with water (10 vol%
ethanol in water; total volume of 5 mL) on 5 wt% Ru/TiO2 at 200 oC and 20 bar under stirring
o
(20 rpm). Since the background spectrum was taken when the reaction temperature (200 C)
was reached, the IR spectra only show the species that changed over time. Species that were
formed emerge as positive IR bands while those consumed or decomposed appear as negative
bands. In this way, vibrational signals originating from species that change over time can be
separated from the static IR bands of spectator species.
The spectra shown in Figure 2 clearly differ from the liquid phase spectrum (Figure 3)
taken from a probe of the reaction mixture after 2 h of reaction. Therefore, the IR bands in
Figure 2 are considered to originate from the surface species adsorbed on the catalyst surface.
The liquid phase spectrum shows O-H stretching vibrations and deformation modes of water
at around 3300 and 1640 cm-1 as well as C-H and C-O stretching vibrations of ethanol at around
2900 and 1050 cm-1. The weak bands at 1200 – 1450 cm-1 are assigned to the deformation
vibrations of C-H and O-H bonds of ethanol. Some minor bands at 1700-2150 cm-1 are due to
8

overtone and combination modes as indicated in Figure 3.
Figure 2. ATR–IR spectra measured during APR of 10 vol% ethanol (EtOH) on 5 wt% Ru/TiO2
at 200 oC and 20 bar.
Figure 3. ATR-IR spectrum of the liquid reaction mixture taken after 2h reaction over 5 wt%
Ru/TiO2 at 200 oC and 20 bar.
During the 2 hours reaction, catalytic transformation (decomposition) of ethanol occurred as
indicated by the changes in the spectra acquired in intervals of 5 min (Figure 2). In the initial
period of the reaction (0-20 min), a band at 1043 cm-1 assigned to C-O stretching vibration of
ethoxy groups emerged associated with negative bands at 2916 and 2848 cm-1 assigned to C-
9

H stretching vibrations. When measuring the background spectrum, ethoxy groups already
existed on the catalyst surface, therefore the changes in the IR bands are considered to originate
from the formation of partially dehydrogenated ethoxy species according to:
CH3CH2O(a) → C2H5-xO(a) + x H(a)
(6)
Differentiation of adsorbed species on Ru or TiO2 is a key aspect to account for the reaction
mechanism. Ethoxy species on TiO2 fall into two adsorption configurations: monodentate
(νC-O ≃ 1120 cm-1) whose oxygen atom bound to one Ti atom and bidentate (νC-O ≃ 1050 cm-
1), which bridges on two Ti atoms [14]. Normally both monodentate and bidentate are equally
o
formed on the TiO2 surface and are fairly stable even at 250 C leading to merely desorption
instead of decomposition or dehydrogenation [14]. On the other hand, ethoxy species on Ru
are always present as monodentate (νC-O ≃1040 cm-1) whose oxygen atom bound to one Ru
atom [15]. Compared to those on TiO2, ethoxy species on Ru are reactive even at around 100
oC. Since our spectra show only one C-O band at 1043 cm-1 associated with a C-C band at
≃1080 cm-1, the ethoxy species are considered to be formed on Ru in this study.
Interestingly, after this initial reaction period the C-O band at 1043 cm-1 started
disappearing gradually with further decrease of the intensity of the C-H bands at 2916 and 2848
cm-1, indicating ongoing dehydrogenation and cleavage of C-O bonds. In line with this process
new sharp bands emerged at 3016 and 1303 cm-1, which are assigned to C-H and C-C stretching
vibrations of acetylide (μ3-η2-CCH). The slight difference in the wavenumber compared to that
reported for measurements on a single crystal Ru-surface under ultrahigh vacuum (UHV) in
[16 ] is attributed to the vastly different conditions (liquid-phase, 200oC, 20bar) used in our
study. According to Greenler’s metal-surface selection rule [17, 18], vibrational modes parallel
to the surface cannot be detected by IR spectroscopy. The C-C bond of acetylide (μ3-η2-CCH)
is basically parallel to the metal surface. Because the C-C stretching vibration of acetylide was
barely detectable in this study, the carbon atom bound to a hydrogen atom would be lifted
upwards, so that the C-C bond is not perfectly parallel to the Ru surface, leading to the low
intensity of the band at 1303 cm-1. The formation of acetylide rationally accounts for the further
dehydrogenation and cleavage of C-O bonds according to the following reaction:
C2H5-xO(a) → C2H(a) + (4-x) H(a) + O(a)
(7)
10

Note that such hydrocarbonaceous species and ethoxy groups on the catalyst surfaces have so
far only been observable in ultrahigh vacuum (UHV) systems [16,19]. TiO2 is inactive for
dehydrogenation at this temperature [14] and acetylide requires two carbon-metal bonds on a
flat metallic surface [16]. Therefore, we consider that the above reaction (C-H and C-O
cleavage) occurred on the Ru surface and possibly the circumference of the Ru-particles. To
our knowledge, this is the first report on IR detection of the hydrocarbonaceous species under
real working condition of the catalyst, that is at elevated pressure (20 bar) and temperature (200
oC) in the liquid-phase. Ideally, APR of ethanol should catalyze the cleavage of C-C bonds to
enhance the selectivity to H2 and CO2 while avoiding unfavorable methanation through
cleavage of C-O bonds [7]. Therefore, we infer that the reaction path via acetylide contributes
to the formation of methane as a by-product. Figure 4 shows ATR-IR spectra in the range 1900
– 2100 cm-1 during APR of ethanol on 5 wt% Ru/TiO2 at 200 oC and 20 bar, which conditions
correspond to those used for the data presented in Figure 2. A band at 1976 cm-1 assigned to
on-top CO on Ru [11] emerged gradually over the course of 2 h. Since the intensity was
extremely low, the band shifted to the lower wavenumber due to weak dipole-dipole
interactions between neighboring CO molecules. The rather low intensity also indicates that
CO molecules rapidly reacted with H2O to form CO2. This explains the reason why no gaseous
CO was detected in APR of ethanol.
Figure 4. ATR-IR spectra during APR of 10 vol% ethanol in water on 5 wt% Ru/TiO2 at 200
oC and 20 bar.
11

In the region of 1900-2100 cm-1 in Figure 3, only a small fraction of on-top CO on Ru was
detected implying that the water gas shift reaction (reaction 3) proceeded rapidly as soon as
CO formed on the Ru surface.
As described previously, APR of ethanol involves two reaction paths through AcH or
AcOH [11]. APR of AcH showed in the initial period of reaction (0-100 min) CH4, CO2 and
H2 as main products, whereas APR of AcOH mainly produced H2 and CO2 in this period.
Figure 5a shows a similar trend as Figure 2; the dehydrogenation of AcH (cleavage of C-H
bonds at 2916 and 2848 cm-1) is accompanied by the formation of acetylide indicated by the
bands at 3016 and 1303 cm-1. This behavior accounts for the significant undesired formation
of methane at the initial stage of the reaction. In contrast, as seen in Figure 5b, IR spectra for
APR of AcOH showed a weak negative band of acetylide at 3016 and 1303 cm-1 even though
the dehydrogenation proceeded. Therefore, when the spectral background was taken at 200 oC,
already a small amount of acetylide existed but decomposed during the reaction. However,
further notable formation of acetylide was not confirmed, which is in accordance with the high
selectivity to H2 and CO2 observed in [11].
12

Figure 5. ATR–IR spectra measured during APR of (a) 10 vol% acetaldehyde (AcH), and (b)
10 vol% acetic acid (AcOH) on 5 wt% Ru/TiO2 at 200 oC and 20 bar.
The most important aspect of APR of ethanol on Ru/TiO2 is the selectivity enhancement
by controlling the loading and particle size of Ru. In the previous study [11], lowering of the
Ru-loading resulted in a significant decrease of the Ru particle size and the formation of
cationic Ru sites, which enhanced the selective production of H2 and CO2. To understand this
phenomenon, we also carried out similar ATR-IR investigations with the 0.5 wt% Ru/TiO2
catalyst (Figure 6). In the initial period of reaction (0-20 min), two characteristic bands
assigned to ethoxy groups on the Ru surface emerged at 1043 and 2978 cm-1. As the reaction
proceeded, the intensities of both C-H and C-O bands began decreasing. However, unlike the
behavior of the 5wt% Ru/TiO2 (Figure 2), no IR bands due to acetylide were observed.
Therefore, ethoxy groups simply underwent dehydrogenation to produce H2 and
decarbonylation to form CO2. This inhibition of the cleavage of C-O bonds is considered to
13

enhance the selectivity to H2 and CO2 over the low loaded catalyst with small Ru particles. No
other feasible hydrocarbonaceous species such as ethylidyne and methylidyne were detected
under reaction condition because these surface species easily undergo dehydrogenation and
decomposition already at much lower temperature [16, 19, 20]. The acetylide and ethoxy
groups appear to be the sole species that can be detected at reaction conditions (20 bar and 200
oC).
Figure 6. ATR–IR spectra during APR of 10 vol% ethanol on 0.5 wt% Ru/TiO2 at 200 oC and
20 bar.
4. Conclusions
ATR-IR spectroscopy was applied to shed some light on the surface processes occurring during
the aqueous phase reforming of ethanol on Ru/TiO2 catalysts with different Ru-loading and
particle size at 200oC and 20bar. The dynamic changes of the species adsorbed on the catalytic
solid-liquid interface revealed significant differences, depending on the Ru-loading and
particle size. Undesired methanation of the formed CO2 and CO prevalent on the catalyst with
higher Ru-loading (5wt%) and larger Ru particle size (~ 4 nm) was suppressed on the
corresponding catalyst with lower Ru-loading (0.5wt%) and smaller Ru particle size (~ 1 nm).
Intermediate formation of acetylide species identified on the 5wt% Ru/TiO2 catalyst is
attributed to C-O bond cleavage reactions, C2H5-xO(a) → C2H(a) + (4-x) H(a) + O(a), which are
considered to lead to undesired methane formation and should be avoided for efficient H2
generation from ethanol. The study of the molecular processes occurring at the solid-liquid
14

interface indicates that optimizing of the Ru-loading and particle size could provide a strong
handle to further the efficiency of hydrogen production. Key is the avoidance of C-O bond
cleavage, which results in undesired methane formation.
To date, there exist barely any knowledge about the structure sensitivity of the surface
processes involved in APR. Although the present study does not allow to differentiate between
the effects of Ru-loading and particle size, it indicates that these properties need to be properly
controlled for optimizing H2 production. Deeper insight into the structure sensitivity could be
gained by applying catalysts with the same metal loading but different particle size.
ACKNOWLEDGEMENTS
This work was financially supported by the Natural Science Foundation of China (NSFC)
(No. 21377017) and the Young Thousand Talents Program of China.
REFERENCES
[1] J. Breen, R. Burch, H. Coleman, Appl. Catal. B, 39 (2002) 65-74.
[2] H. Song, in: L. Map, N. App (Eds.), Bioethanol, InTech, 2012. pp. 255-290.
[3] S. Liu, K. Zhang, L. Fang, Y. Li, Energy & Fuels 22 (2008)1365-1370.
[4] L.V. Mattos, G. Jacobs, B.H. Davis, F.B. Noronha, Chem. Rev. 112 (2012) 4094-4123.
[5] P.R. de la Piscina, N. Homs, Chem. Soc. Rev. 37 (2008) 2459-2467.
[6] R.D. Cortright, R.R. Davda, J.A. Dumesic, Nature 418 (2002) 964-967.
[7] J.W. Shabaker, R.R. Davda, G.W. Huber, R.D. Cortright, J.A. Dumesic, J. Catal. 215
(2003) 344-352.
[8] G.W. Huber, J.W. Shabaker, J. Dumesic, Science, 300 (2003) 2075-2077.
[9] J.W. Shabaker, G.W. Huber, J.A. Dumesic, J. Catal. 222 (2004)180-191.
[10] B. Meryemoglu, A. Hesenov, S. Irmak, O.M. Atanur, O. Erbatur, Int. J. Hydrogen
Energy, 35 (2010) 12580-12587.
[11] T. Nozawa, Y. Mizukoshi, A. Yoshida, S. Naito, Appl. Catal. B, 146 (2014) 221-226.
[12] T. Nozawa, A. Yoshida, S. Hikichi, S. Naito, Int. J. Hydrogen Energy, 40 (2015) 4129-
4140.
15

[13] T. Bürgi, A. Baiker in: B.C. Gates (Ed.), Advances in Catalysis, Elsevier, 2006, Vol. 50,
pp. 227-283.
[14] W. C. Wu, C. C. Chuang, J. L. Lin, J. Phys. Chem. B, 104 (2000) 8719-8724.
[15] J. M. Sturm. C. J. Lee, F. Bijkerk, Surf. Sci., 612 (2013) 42-47.
[16] M. Hills, J. Parmeter, C.B. Mullins , W. Weinberg, J. Am. Chem. Soc. 108 (1986) 3563-
3569.
[17] R. B. Greenler, D. R. Snider, D. Witt, R. S. Sorbello, Surf. Sci., 118 (1982) 415-428.
[18] G. Ghiotti, F. Boccuzzi, A. Chiorino, Surf. Sci., 178 (1986) 553-564.
[19] P. Gazdzicki, S. Thussing, P. Jakob, J. Phys. Chem. C, 115 (2011) 25379-25388.
[20] J.A.H. Dreyer, P. Li, L. Zhang, G.K. Beh, R. Zhang, P.H.L. Sit, W.Y. Teoh, Appl. Catal.
B: Environ. 219 (2017) 715-726.
16

Figure Captions
Figure 1. TEM images and Ru particle size distributions of (a) 0.5 wt% Ru/TiO2 and (b) 5 wt%
Ru/TiO2.
Figure 2. ATR–IR spectra measured during APR of 10 vol% ethanol (EtOH) on 5 wt% Ru/TiO2
at 200 oC and 20 bar.
Figure 3. ATR-IR spectrum of the liquid reaction mixture taken after 2h reaction over 5 wt%
Ru/TiO2 at 200 oC and 20 bar.
Figure 4. ATR-IR spectra during APR of 10 vol% ethanol in water on 5 wt% Ru/TiO2 at 200
oC and 20 bar.
Figure 5. ATR–IR spectra measured during APR of (a) 10 vol% acetaldehyde (AcH), and (b)
10 vol% acetic acid (AcOH) on 5 wt% Ru/TiO2 at 200 oC and 20 bar.
Figure 6. ATR–IR spectra during APR of 10 vol% ethanol on 0.5 wt% Ru/TiO2 at 200 oC and
20 bar.
17