Hydrodeoxygenation of guaiacol over Ni@Pd and Ni@Pt bimetallic overlayer catalysts

Article abstract of DOI:10.1016/j.apcata.2016.09.009

First principle computational studies have predicted that formation of Pd or Pt overlayer on Ni base metal would cause a negative shift of the d-band center of surface Pd or Pt, resulting in reduced binding strength for adsorbed species. In this paper, silica alumina supported Ni@Pd and Ni@Pt overlayer catalysts were synthesized via directed deposition technique and tested for hydrodeoxygenation of guaiacol at atmospheric pressure. Several characterization techniques such as hydrogen chemisorption, ethylene hydrogenation descriptor reaction, XRD and TEM studies were employed to characterize the catalysts. H₂ chemisorption results showed that Ni@Pd and Ni@Pt overlayer catalysts had reduced H₂ adsorption strength compared to Pd or Pt only catalysts, which agreed with the computational prediction. As predicted, the overlayer catalysts also showed lower activity for ethylene hydrogenation than Pd or Pt only catalysts. XRD and TEM studies indicated that overlayers have been successfully deposited atop supported nickel. Guaiacol hydrodeoxygenation results showed that, especially at low reaction temperature, Pd and Pt active sites of overlayer catalysts showed significantly enhanced deoxygenation activity compared with that of Pd or Pt only catalysts. Further studies showed that guaiacol could be completely deoxygenated using silica alumina supported metal catalysts, yielding benzene, toluene, and xylenes as major products.

Full text of DOI:10.1016/j.apcata.2016.09.009

Accepted Manuscript
Title: Hydrodeoxygenation of Guaiacol over Ni@Pd and
Ni@Pt Bimetallic Overlayer Catalysts
Author: Qinghua Lai Chen Zhang Joseph H. Holles
PII:
S0926-860X(16)30473-2
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2016.09.009
APCATA 16005
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
7-6-2016
30-8-2016
17-9-2016
Please cite this article as: Qinghua Lai, Chen Zhang, Joseph H.Holles,
Hydrodeoxygenation of Guaiacol over Ni@Pd and Ni@Pt Bimetallic Overlayer
Catalysts, Applied Catalysis A, General http://dx.doi.org/10.1016/j.apcata.2016.09.009
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.

Hydrodeoxygenation of Guaiacol over Ni@Pd and Ni@Pt Bimetallic Overlayer Catalysts
Qinghua Lai, Chen Zhang, Joseph H. Holles*
Department of Chemical and Petroleum Engineering, University of Wyoming, Dept. 3295, 1000
E. University Ave, Laramie, WY 82071, United States
*Corresponding author. Tel.:+1-307-766-6772; fax: +1-307-766-6772. E-mail address:
jholles@uwyo.edu (J.H. Holles).
Graphical abstract
1

Highlights
1
. Silica alumina supported Ni@Pd and Ni@Pt overlayer catalysts were synthesized via
directed deposition technique and tested for hydrodeoxygenation of guaiacol at
atmospheric pressure.
2
3
4
. Overlayer catalysts with the reduced adsorption strength showed significantly enhanced
deoxygenation activity compared with Pd or Pt only catalyst.
. Selectivity of benzene, toluene, and xylenes could be up to 80% on Ni@Pd overlayer
catalyst at a W/F of 4 h.
. Overlayer formation using directed deposition technique provides a way to modify the
behavior of the surface metal and ultimately modify the hydrodeoxygenation activity
ABSTRACT
First principle computational studies have predicted that formation of Pd or Pt overlayer on Ni
base metal would cause a negative shift of d-band center of surface Pd or Pt, resulting in reduced
binding strength for adsorbed species. In this paper, silica alumina supported Ni@Pd and Ni@Pt
overlayer catalysts were synthesized via directed deposition technique and tested for
hydrodeoxygenation of guaiacol at atmospheric pressure. Several characterization techniques
such as hydrogen chemisorption, ethylene hydrogenation descriptor reaction, XRD and TEM
studies were employed to characterize the catalysts. H
2
chemisorption results showed that
Ni@Pd and Ni@Pt overlayer catalysts had reduced H adsorption strength compared to Pd or Pt
2
only catalyst, which agreed with the computational prediction. As predicted, the overlayer
catalysts also showed lower activity for ethylene hydrogenation than Pd or Pt only catalyst. XRD
and TEM studies indicated that overlayer have been successfully deposited atop supported
nickel. Guaiacol hydrodeoxygenation results showed that, especially at low reaction temperature,
Pd and Pt active sites of overlayer catalysts showed significantly enhanced deoxygenation
2

activity compared with that of Pd or Pt only catalyst. Further studies showed that guaiacol could
be completely deoxygenated using silica alumina supported metal catalysts, yielding benzene,
toluene, and xylenes as major products.
Keywords: guaiacol, hydrodeoxygenation, overlayer, hydrogen chemisorption, bimetallic, Pd, Pt
1
. Introduction
Currently, the major portion of the global energy supply relies on energy produced from
fossil-based resources. One of the major concerns of society today as related to energy is the
depletion of petroleum, a non-renewable fossil source of carbon. Another major society concern
is the use of fossil fuels resulting in subsequent environmental problems including greenhouse
gas emissions and acid rain. Hence, there is a need to switch to clean, renewable sources of
energy.
Lignocellulosic biomass is the most abundant class of biomass on the planet.[1] It provides an
opportunity to be used as a sustainable source of energy and organic carbon for our society.
Lignocellulosic biomass can be converted into crude bio-oil via thermochemical treatment such
as pyrolysis or liquefaction. These crude bio-oils are multicomponent mixtures of a large number
of oxygenated compounds. However, the high oxygen content of crude bio-oils, usually 20 to 50
wt%, leads to low heating value, poor stability, poor volatility, high viscosity and
corrosiveness.[2, 3] Therefore, oxygen removal from bio-oils is necessary for further application
as liquid fuels.
The lignin fraction of biomass is a three-dimensional amorphous polymer composed of
methoxylated phenylpropane structures, which contains approximately 40% of the possible
energy of the biomass.[4] Compared with the carbohydrate fraction of biomass, it is more
3

attractive to derive alternative fuels from lignin fraction because it has a lower oxygen-to-carbon
ratio. There are several approaches for upgrading bio-oils including zeolite upgrading,[3, 5-8]
aqueous phase processing[8, 9] and hydrotreating[3-5, 9, 10]. Hydrotreating, in which feedstocks
are reacted with hydrogen in the presence of a catalyst, is the most common method to upgrade
bio-oil to hydrocarbons via a hydrodeoxygenation (HDO) reaction to remove oxygen.[3, 4] HDO
studies of lignin-derived oils usually employ model compounds such as phenols, anisole and
guaiacol instead of bio-oil. Guaiacol, which has two types of C–O bond (C–OH bond and C–
OCH bond), appears to be the preferred choice as a prototype compound to represent lignin-
3
derived bio-oils.
The traditional HDS catalysts (sulfided Co-Mo/-Al
2
O
3
or Ni-Mo/-Al
2
3
O ) used in petroleum
refining are also utilized as HDO catalysts.[3, 11] However, these conventional
hydrodeoxygenation catalysts have some disadvantages such as possible sulfur contamination of
products, rapid deactivation by coke formation and water-induced catalyst deactivation.[3, 11,
12] For these reasons, non-sulfided transition metal based catalysts systems are being developed
for the HDO of lignin and related model compounds. HDO of lignin related model compounds
by base metal catalysts, such as supported Ni,[13-19] Fe,[20, 21] and Mo[22-25] have been
investigated. The supported metal phosphides (Ni
catalyst for HDO of guaiacol and anisole.[26-28] Ni
among tested metal phosphides (Ni P, Co P, MoP, WP).[26, 27] Recently, using supported noble
2
P, Fe
2
P, Co
2
P, MoP, WP) were also used as
2
P showed the highest activity for HDO
2
2
metals as effective HDO catalysts has received new attention.[29-33] These noble metal catalysts
are generally highly active for hydrogenation and hydrogenolysis, and require lower temperature
or pressure than conventional hydrotreating catalysts.[3] However, the noble metal catalysts are
4

expensive and are highly active for hydrogenation of aromatic rings causing high H
2
consumptions.
Using bimetallic catalysts is a strategy to enhance the efficiency of catalysts for the HDO, as
the interaction between metals can modify the geometric and electronic structures of metal
surface resulting in improved catalytic activity and selectivity.[3, 4, 34-37] In related work, Lobo
et al. have reported that adding Ni or Co into the Pt not only increased the meta-cresol HDO
activity, but also enhanced formation of methylcyclohexane as well.[38] Resasco et al. have
shown that a bimetallic Pt-Sn catalyst used for HDO of guaiacol at high temperatures (400°C)
had higher reactivity (80% conversion after 3 h time-on-stream) and better stability than the
monometallic catalyst.[39] They postulated that the strong adsorption of double-functionalized
guaiacol (or catechol) on the catalyst surface could block sites for deoxygenation on the Pt
catalyst, and the bimetallic Pt-Sn might be able to reduce the binding strength. Ardiyanti et al.
investigated HDO of anisole over Ni–Cu/-Al
2
3
O with various Ni/Cu ratios.[40] The bimetallic
Ni-Cu catalyst showed higher activity than monometallic nickel, which may be attributed to
more formation of active metallic nickel when the copper is present. Wang et al. also have found
that the bimetallic Pd-Fe/C catalyst exhibit a substantially enhanced activity of HDO of guaiacol,
with more than 80% yield to benzene/toluene/trimethylbenzene at 450°C.[41] The single crystal
studies of Ni/Pt overlayer system by Chen et al. have shown bimetallic Ni/Pt surface bound
hydrogen and unsaturated hydrocarbon adsorbates more weakly than corresponding
monometallic surfaces and thus enhanced the activity toward the production of butene in HDS of
thiophene.[42] The formation of weaker metal-adsorbate bonds was speculated to be favorable to
overcome the kinetic barrier for the HDS of thiophene.
5

Based on previous literature work using bimetallic catalysts for HDO, the pseudomorphic
overlayer catalyst with a monolayer of metal atop a base metal would also enhance HDO
activity. Studies have shown that adsorbate bond strengths are generally altered in overlayer
systems due to the shift of the d-band center of the overlaying metal.[43-47] Our previous
research demonstrated that pseudomorphic overlayer catalysts, such as Ni@Pt, Co@Pt or
Re@Pd (core@shell), could reduce H and CO binding energy when compared to single metal
2
and alloy counterparts, as predicted from computational work.[48-51] The decrease in hydrogen
and hydrocarbon adsorbates binding strength indicate that pseudomorphic overlayer catalysts
may demonstrate improved performance for the HDO reaction because strong bonding has been
shown to inhibit the reaction.[39] Ni@Pd and Ni@Pt overlayer have been predicted to have
negative d-band shift and decreased binding strength compared to the pure Pd or Pt.[43-46] In
this contribution, we synthesized single deposition Ni@Pd and Ni@Pt overlayer catalysts and
used a strategy of reducing binding strength to improve the efficiency of catalysts for guaiacol
HDO. H chemisorption, ethylene hydrogenation descriptor reaction, physical sorption, XRD and
2
TEM studies were conducted to characterize the catalysts. Guaiacol HDO reactivity tests were
performed in a fixed-bed reactor at atmospheric pressure to study how overlayer catalysts affect
HDO activity.
2
2
Experimental
.1. Preparation of catalysts
The monometallic Ni, Pd and Pt catalysts and non-structured bimetallic catalysts were
prepared using co-impregnation method. A silica-alumina catalyst support (Alfa Aesar, grade
35) was used as support for all catalysts. The precursors used in this paper were nickel nitrate
1
hexahydrate (Aldrich 99.999%), palladium (II) acetylacetonate (Alfa Aesar) and platinum (II)
6

acetylacetonate (Alfa Aesar). The metal loading was about 0.5 wt% for monometallic Pt and Pd
catalysts, and 5wt% for Ni parent catalysts. The non-structured bimetallic Ni-Pd and Ni-Pt
catalysts were prepared with similar metal loadings as the overlayer catalysts. After
impregnation of silica-alumina with a solution of precursors (aqueous solution for nickel nitrate
hexahydrate and toluene solution for palladium (II) acetylacetonate and platinum (II)
acetylacetonate), the mixtures were heated at around 60°C to evaporate solvent. All mixtures
were dried in an oven overnight. For bimetallic catalysts, the nickel nitrate hexahydrate was
impregnated first and then the dry mixture was used for the second impregnation to load Pd or
Pt. Then the dry catalysts were calcined in air at 400°C for 3 h using a 3°C/min ramp rate[52,
5
3]. The catalysts were then reduced under 60 sccm H
ramp rate based on suggestions in the literature[50, 52-55].
The overlayer catalysts used in this paper were synthesized using the directed deposition
method.[49, 50] Typically, the parent Ni/SiO -Al catalyst was reduced in 60 sccm H flow at
50°C for 4 h using a 3°C/min ramp rate. Then water-saturated helium gas was passed over the
catalyst for 24 h at room temperature. After that, the parent catalyst was treated with 60 sccm H
flow at 60°C for 1 h (to maximize H adsorption for the subsequent overlayer deposition
reaction). The briefly reduced sample was directly poured from reduction reactor into a toluene
solution (30 mL) of 0.1 M acetylacetone, with H gas flowing into the reduction reactor at same
2
flow at 450 ºC for 3 h using a 3°C/min
2
2
O
3
2
4
2
2
2
time. Acetylacetone was used as a surface inhibitor to minimize the platinum adsorption on the
support. After 15 minutes surface deactivation, a toluene solution of 0.05 M Pd (II)
acetylacetonate (or Pt (II) acetylacetonate) was added. The mixture was then kept at 60°C for 2
h. The amount of Pd (II) acetylacetonate added was equivalent to the 150% monolayer coverage
of parent Ni surface. The catalyst was then vacuum filtered and placed in a 110°C oven
7

overnight. The dried catalyst was calcined in near stagnant air at 400°C for 4 h using a 3°C/min
ramp rate. After calcination, the catalyst was reduced in H
2
flow at calcine temperatures for 4 h
using a 3°C/min ramp rate.
2
.2 Catalyst characterization
Inductively coupled plasma optical emission spectroscopy (ICP-OES) analysis was performed
on a PerkinElmer Optima 8300 ICP-OES spectrometer to determine actual metal weight loadings
for each catalyst. Nitrogen physisorption was carried out on a Micromeritics APAP 2020 with
surface area and porosimetry analyzer to measure the surface areas and pore volume of samples.
Powder X-ray diffraction (XRD) was employed to analyze material phases of catalysts. Before
analysis, the samples were reduced for 1 h at the same temperature used in synthesis. XRD data
were collected on a Rigaku Smart Lab automated powder diffraction system using Cu Kα
radiation operated at 40 kV and 40 mA. Transmission electron microscopy (TEM) was
performed on a FEI Tecnai G2 F20 S-Twin 200 kV transmission electron microscope. The
temperature programmed oxidation of spent samples was evaluated by using 20% O
mass spectrometer detector. The temperature was increased form room temperature to 900°C
using a 10°C/min ramp rate. NH temperature programmed desorption (NH -TPD) was
2
/N
2
with a
3
3
conducted to evaluate the acidity of catalysts. Typically, samples (200 mg) loaded in the quartz
tube were treated with pure nitrogen gas at 350°C for 1 h. After pretreatment, the samples were
exposed to 5.0% NH
3
in N
2
at 100°C for 1 h. Then, the weakly adsorbed NH
3
was removed from
samples by purging pure nitrogen gas at 100°C for 1 h. The desorption was performed by
increasing the temperature to 700°C with a ramp rate of 10°C/min and the desorbed NH
3
was
detected with a Hiden HPR20 mass spectrometer system using the signal of m/e= 16.
8

Hydrogen chemisorption studies were performed on a Micromeritics APAP 2020C instrument
using a static volumetric technique. Prior to analysis, samples were reduced at 400°C under
flowing hydrogen. The lower temperature (400°C) was used to minimize the potential for
agglomeration during the multiple isotherm measurements. Isotherms were obtained at
temperatures between 35°C and 300°C with pressures ranging from 1 mTorr to 900 Torr. Metal
dispersions were estimated assuming chemisorption stoichiometry of H/Msurf = 1. Ethylene
hydrogenation reaction was conducted as a descriptor reaction as described previously.[50]
2
.3 Guaiacol hydrodeoxygenation experiments
The HDO activity of the catalysts were studied in a fixed-bed tubular quartz reactor (i.d. = 8
mm and l = 400 mm) operated at atmosphere pressure and a temperature ranging from 350 to
50°C, with pure guaiacol (Aldrich, 99%) as feedstock. In a typical experiment, 200 mg of
4
catalyst sample was mixed with ~800 mg of silica (Alfa Aesar, 106-212 m). Before reaction,
the catalysts were reduced in the reactor with a flow of pure hydrogen (60 mL/min) at 450°C.
During catalytic tests, the pure guaiacol liquid was introduced by a HPLC pump at a rate of 0.72
mL/h and the hydrogen flow rate was 60 ml/min (the molar ratio of H
2
to guaiacol was about
25). Liquid samples collected by cold trap at 1 h intervals were analyzed off-line by gas
chromatograph (Thermo-Fisher Trace GC Ultra) equipped with a flame ionization detector (FID)
and Rt-Q-BOND column for product analysis. The 2-isopropylphenol as internal standard was
added into the collected liquid samples before analysis.
Product selectivity was calculated as S (%) =(Molproduct-i/Molreacted guaiacol)100%. Turn over
i
frequency (TOF) was calculated as described in the following. For monometallic (Ni, Pd and Pt)
and non-structured (Ni-Pd and Ni-Pt) bimetallic catalysts, total active sites for TOF calculation
were calculated based on surface metal atoms from dispersion. For overlay catalysts, we
9

assumed that all measured palladium or platinum loading is on the surface. As the Pd loading
was enough for 100% coverage of Ni@Pd overlayer catalysts, surface metal atoms calculated
from dispersion was used for TOF calculation. For Ni@Pt overlayer catalysts, Pt coverage was
less than 100% coverage, and the amount of Ni active sites on surface was calculated by
subtracting the Pt active sites from the total surface metal atoms. It was also assumed that Ni
active sites on Ni@Pt overlayer catalysts have the same activity for guaiacol conversion as that
of Ni active sites on monometallic Ni catalyst. Therefore, the guaiacol conversion by Pt active
sites could be calculated by subtracting the conversion by Ni active sites from the total guaiacol
conversion. Then the activity of Pt active sites on overlayer catalysts was calculated.
3
Results and discussion
3
.1 Physicochemical properties
The metal weight loadings, dispersions, calculated percent overlayer coverage, surface areas and
pore volumes for catalysts are summarized in Table 1. For Ni@Pd and Ni@Pt overlayer samples,
Pt or Pd was successfully loaded on the overlayer catalysts. As shown in Table 1, non-structured
bimetallic Ni-Pd and Ni-Pt catalysts were also synthesized with metal loadings similar to that of
their respective Ni@Pd or Ni@Pt counterparts for direct comparison. According to Ni-Pd and
Ni-Pt phase diagrams, at these concentrations the Pd and Pt should be completely miscible in the
Ni without forming a second phase.[56, 57] Compared with Ni parent catalyst, adding Pd or Pt
increases the metal dispersion of both non-structured bimetallic Ni-Pd and Ni-Pt catalysts.
2

1
Nitrogen physisorption results showed that all catalysts had similar surface area (410-460 m g )
3

1
and pore volume (0.57-0.68 cm g ). For overlay samples, Pd or Pt deposition only slightly
decreased the surface area and pore volume.
10

3
.2 Powder XRD
Fig. 1 shows the results of X-ray diffraction investigation of SiO
2
-Al
2
3
O supported catalysts. For
Ni and Pd system (shown in Fig. 1(a)), peaks of metallic Ni were observed in all samples
containing Ni. Metallic Pd peaks were observed in the Ni-Pd bimetallic sample, which suggests
that isolated Pd only domains exist in the randomly-structured bimetallic Ni-Pd sample. The
metallic Pd peak wasn’t observed in Ni@Pd overlayer sample in spite of the Pd weight loading
being equivalent or greater than the Ni-Pd sample. The Pd atoms in Ni@Pd overlayer appear to
be present in non-agglomerated structures that are not detectable by XRD. This also indicates
that Pd in the Ni@Pd overlayer sample is present in a different structure than the bimetallic Ni-
Pd catalyst. For Ni-Pt system (shown in Fig. 1(b)), a metallic Pt peak was not observed in Ni-Pt.
Again, no evidence of Pt only domains was observed for the Ni@Pt overlayer sample.
3
.3 TEM study
Fig. 2(a) shows TEM image of a Ni@Pd SD overlayer particle deposited on silica alumina
support. As shown in Fig. 2(b), the Fourier transform of this image indicates that it has
crystalline structure. After performing the inverse Fourier transform of different pairs of sharp
spots (from Fig 2(b)) in reciprocal space, two different types of lattice fringes (see Fig. 2(c) and
2
(d)) were observed. One type of the lattice fringes were inside the edge of particle (Fig. 2(c))
and the other type showed the lattice fringes around the particle edge (Fig. 2(d)). The lattice
spacing inside the edge of the particle (from the inset in Fig. 2(c)) is approximately 0.21 nm,
which is consistent with the planar distance spacing of Ni (111) and the fringes inside the edge of
the particle likely represent Ni. The lattice spacing around the particle edge (from the inset in
Fig. 2(d)) is approximately 0.23 nm, which is slightly larger than bulk Pd (111) planar distance
spacing (0.225 nm). Teranishi et al. observed that the interplanar spacing of Pd (111) could vary
11

from 0.225 nm to 0.229 nm among different size palladium nanoparticles.[58] The larger lattice
fringe around the particle similar to Pd spacing values indicates that the Ni@Pd overlayer was
successfully prepared. Regardless, the TEM images provide evidence for a core/shell type
structure. However, further characterizations are needed to provide additional evidence for
overlayer catalyst structure.
3
.4 Acidity of catalysts
NH -TPD was conducted to compare the acidity of catalysts. The NH
are shown in Fig. 3 and the relative peak areas are presented in Table 1. The NH
of SiO -Al support showed three major peak areas at T < 200°C, 200-400°C and 400-600°C,
which can be described as NH desorption from weak, medium and strong acid sites,
respectively.[59] All supported catalysts showed very similar NH -TPD profiles with a major
3
3
-TPD desorption profiles
3
-TPD profiles
2
2
O
3
3
3
desorption peak at around 165-170°C. All supported catalysts showed only small amounts of
desorption at 400-600°C compared to the pure support, indicating fewer strong acid sites
compared to the support. The total amount of acidity were compared by using relative desorption
peak areas (Table 1). All supported catalysts had similar total peaks area (0.47-0.60), which is
expected since the same support was used for all catalysts and the deposited metal only covers a
very small fraction of the surface. Thus, there does not appear to be any clear difference in the
support acidity resulting from the overlayer synthesis. This indicates that any observed activity
changes are likely due to changes in the metallic aspect of the catalysts. The total desorption
peak areas were obtained by integrating the peak areas from 100 to 650°C.
3
.5 Hydrogen chemisorption
Fig. 4 shows isotherms of hydrogen adsorption on Ni@Pd overlayer and bimetallic based
catalysts at 50°C in comparison to palladium and nickel only catalysts. Comparing with other
12

catalysts, the monometallic Pd showed a strong initial/low pressure (below 0.01 Torr)
adsorption. In contrast with the Pd, Ni only catalyst shows little low pressure adsorption and a
nearly linear (plotted on a semi-log scale) increase in adsorption across all pressures. Unlike the
Pd only catalyst showing a strong initial/low pressure adsorption, Ni@Pd overlayer catalyst
didn’t have this strong initial adsorption and shows a lower slope than Ni parent. Nickel platinum
system (in Fig. 5) showed similar behavior as nickel palladium based catalysts. Pt and Ni-Pt
show some low pressure adsorption, and overlayer sample didn’t have this low pressure
adsorption. These chemisorption results demonstrated that Pd and Pt behavior can be modified
via overlayer synthesis, and Ni@Pd and Ni@Pt overlayer both showed a decrease in H
2
adsorption as computationally predicted. This agrees with our previous study on alumina
supported Ni@Pt and Co@Pt catalysts.[50]
The hydrogen heat of adsorption was calculated from isotherms at multiple temperatures via
the Clausius–Clapeyron equation as described in the literature.[49] Fig. 6 and Fig. 7 show
hydrogen heat of adsorption values as a function of quantity of adsorbed hydrogen for silica-
alumina supported catalysts. The maximum heat of H
2
adsorption for Pd catalyst was 65
−
1
−1
kJ·mol , which compares favorably with the reported values of 61 and 67 kJ·mol .[60] The Pt
−1
catalyst had a maximum heat of H
2
adsorption value of 52 kJ·mol , which was similar to a
reported 58 kJ·mol for silica alumina supported Pt catalyst with a dispersion of 74%.[52, 61]
As shown in Fig. 6 and Fig. 7, heats of H adsorption were reduced by Ni@Pt and Ni@Pd
−
1
2
overlayer synthesis when compared to parent metal (Ni) and monometallic Pd or Pt only
catalysts. Decreased hydrogen adsorption strength of the Pd or Pt overlayer catalyst compared to
the pure Pd or Pt is consistent with computational studies which predict Ni@Pd or Ni@Pt
overlayer catalyst will bond hydrogen more weakly than monometallic Pd or Pt due to surface
13

metal d-band shift.[45-47] The observed decrease in heat of adsorption also provides evidence
that the desired surface structure was formed. This reduction in hydrogen and other reactant or
product adsorption strength is expected to decrease bonding strength slightly, so that fewer
surface sites are covered and thus enhance the HDO activity.
3
.6 Ethylene hydrogenation results
Ethylene hydrogenation represents a simple descriptor reaction where the rate has been shown to
be dependent on hydrogen binding strength.[47, 49] Results of ethylene hydrogenation studies
can be seen in the Arrhenius-type plots shown in Fig. 8 and 9. The Weisz–Prater criterion was
evaluated and it indicates a lack of significant intra-phase diffusion effects during the ethylene
hydrogenation reaction over all the samples. The detailed calculation process is described in the
supporting information. Ethylene hydrogenation turnover frequencies for monometallic catalysts
and non-structured bimetallic Ni-Pd catalysts were calculated based on surface active sites (from
dispersion). Overlayer catalysts’ turnover frequencies were calculated using the measured
palladium or platinum loading with an assumed dispersion of 100%. For core@shell catalysts, if
the palladium or platinum loading, with an assumed dispersion of 100%, exceeded 100% surface
coverage, the total surface active sites from dispersion were used for turnover frequency
calculation. For overlayer catalysts (shown in Fig. 8 and 9), the single sample showed an
increased activity of at least one order of magnitude when compared to the parent Ni. This
indicates that the directed deposition synthesis is adding palladium or platinum to the surface of
catalysts, thus shifting the reactivity behavior toward that of the palladium or platinum only
catalyst. The significant activity difference between the parent (Ni) and overlayer (Pt or Pd)
supports the above method in calculating TOF of the overlayer catalysts. Moreover, the
core@shell catalysts were less active than pure palladium or platinum sample, which indicates
14

that core@shell catalysts result in reduced hydrogenation reactivity. Reduced activity is in
agreement with H chemisorption results that overlayer samples exhibited a significantly lower
2
heat of adsorption than the monometallic Pd or Pt sample. For Ni and Pd based catalysts, the
non-structured Ni-Pd sample was active at the same temperature range as the Pd only catalyst,
suggesting that all activity was due to Pd only domains on the support surface, which have been
shown to exist via observed metallic Pd XRD peak in Ni-Pd sample. The low activity of the
Ni@Pd overlayer sample at these temperature suggests that the Pd is only atop Ni particles and is
likely not isolated as Pd only domains as in the non-structured Ni-Pd bimetallic catalyst. For Ni
and Pt based catalysts, Ni@Pt overlayer catalyst also showed reduced hydrogenation reactivity
compare to the Pt catalyst. In sum, overlayer samples demonstrated a reduced reactivity
compared with palladium or platinum only catalysts, which is consistent with computational
predictions and previous studies by our lab.[43, 49, 50] Based on the literature results that
suggest strong bonding inhibit the HDO reaction, we have identified Ni@Pd and Ni@Pt
catalysts with reduced adsorption strength as potential candidates for guaiacol HDO and these
catalysts were further tested for in guaiacol HDO.
Apparent activation energy for each catalyst was calculated based on the Arrhenius-type plots
and are provided in Fig. 8 and Fig. 9. The apparent activation energies for the Pd and Pt
−1
monometallic catalysts were 65 and 70 kJ·mol , respectively. These numbers were within the
range of various values reported in literature.[50, 52, 60, 62-64] Ni@Pd SD overlayer catalyst
−
1
showed an apparent activation energy of 95 kJ·mol , which is much higher than Pd catalyst.
−
1
Unlike Ni@Pd SD overlayer catalyst, Ni@Pt SD overlayer catalyst had a value of 68 kJ·mol ,
slightly lower than Pt catalyst. Both overlayer catalysts showed different apparent activation
15

energies with their non-structured bimetallic counterparts, further confirming that overlayer
catalyst show different properties compared to non-structured bimetallic catalysts.
Fig. 10 shows the correlation between the heat of H
2
adsorption at the hydrogen coverage of
.1 cm ·g and the TOF of ethylene hydrogenation at 293 K for studied palladium, platinum and
overlayer catalysts. The ethylene hydrogenation reaction has been reported to have positive
reaction order with respect to H , which suggests that reactivity and H surface coverage are
3
−1
0
2
2
positively correlated.[52, 63] Therefore, decreased hydrogen adsorption strength is anticipated to
reduce the hydrogen coverage on the catalyst surface and thus decrease the ethylene
hydrogenation reaction rate. The correlation in Fig. 10 demonstrates a positive correlation
between ethylene hydrogenation reactivity and the H adsorption strength, and also demonstrates
2
that ethylene hydrogenation can be used as a descriptor reaction for H
2
binding strength for
overlayer catalysts.
3
.7 Hydrodeoxygenation of guaiacol
For the HDO of guaiacol, both the support and metal are required for high activity. In this work,
we are examining the ability of overlayer catalysts to improve the reactivity of the metal. In an
effort to isolate the study to the metal, an identical support was used for all catalysts. BET
surface are and pore volume measurements sowed only minor changes in the support for the
different catalysts. Thus, it is likely that the support is acting similarly for all catalysts
The supported catalysts were evaluated for the HDO of guaiacol at 350°C, 400°C and 450°C
at atmospheric pressure. Blank-reactor runs showed a negligible activity for guaiacol at reaction
condition, less than 0.5% conversion (not shown). As the pure Pd and Pt catalysts showed much
higher dispersion than other catalysts, 0.5wt% metal loading Pd and Pt only catalysts were used
for HDO test to keep the number of active sites similar. The conversion and selectivity results of
16

guaiacol hydrodeoxgenation on different catalysts are presented in Table S1. The Weisz–Prater
criterion was evaluated and it indicates a lack of significant intra-phase diffusion effects during
the guaiacol HDO reaction over all the catalysts. The detailed calculation process is described in
the supporting information. Selectivity of the deoxygenation products were given separately,
while selectivity of all non-deoxygenation products (molecules with two oxygen atoms) was
reported as “others”. The support showed small guaiacol conversion with non-deoxygenation
products (“others”) as major products (see Table S1). For all tested catalysts, the main
deoxygenation product was phenol (loss of a methoxy group) under reaction condition. Some
studies have reported phenol to be either a main product or an intermediate for further
deoxygenation in guaiacol deoxygenation.[21, 26, 39, 41] Other major deoxygenation products
included cresol (conversion of the methoxy to a methyl group), while small amount of anisole
(loss of a hydroxyl group) were observed. The formation of cresol (loss of a hydroxyl) is related
to the reaction of methyl substitution in HDO of guaiacol, which was due to the acidity of the
support, because the acidity of support has been found to have a key effect on the reaction of
methyl substitution in HDO of guaiacol.[29, 65, 66] Detectable benzene was also found at some
reaction conditions. No ring saturated products (cyclohexane derivatives) were observed under
tested reaction condition. Compared with Ni parent and monometallic Pd and Pt catalysts, the
bimetallic catalysts and overlayer catalysts showed higher guaiacol conversion and
deoxygenation product selectivity. For monometallic Pd and Pt catalysts, more than 65% of
products at 350°C and 400°C were non-deoxygenation products (“others” in Table S1). Both
bimetallic catalysts and overlayer catalysts improved catalyst selectivity greatly, enhancing the
selectivity toward the production of the desired deoxygenation products. However, at the highest
temperature of 450°C, these improvements were not as remarkable as at low temperature.
17

In order to illustrate more clearly the difference in production rates of active sites on catalysts,
the turnover frequency (TOF) of deoxygenation and non-deoxygenation products were calculated
and presented in Fig. 11 and 12. As show in Fig. 11, the column in green color (second column)
shows the TOF of deoxygenation products, which is the sum of all detected deoxygenation
products including anisole, o-cresol and phenol. Though the monometallic Pd catalyst had the
highest TOF of non-deoxygenation product formation at 350°C, its TOF of deoxygenation
product formation was the second lowest. The Ni-Pd catalyst have the highest guaiacol
conversion and deoxygenation product yield among Ni, Pd, Ni-Pd and Ni@Pd overlayer catalyst
at 350°C, however its TOF of deoxygenation products was only half that of overlayer catalyst,
indicating that the high guaiacol conversion on Ni-Pd catalyst was mainly due to the high metal
dispersion. Among the tested catalysts at 350°C (Fig.11 (a)), the TOF of deoxygenation product
formation of Pd active sites on Ni@Pd overlayer catalyst was the highest, about 3 times as high
as that of the monometallic Pd catalyst (whose behavior the overlayer is designed to modify) and
5
time as high as that of the Ni catalyst. The result, that Pd active sites on Ni@Pd overlayer
catalysts had higher TOF of deoxygenation product formation than monometallic Ni and Pd
catalysts, is consistent with the postulate that Pd active sites on Ni@Pd overlayer catalyst might
have higher HDO activity than the monometallic Pd due to the decrease in hydrogen and
hydrocarbon adsorbate binding strengths. As shown in Fig. 12, the Ni@Pt, Ni-Pt and Pt catalysts
showed similar HDO activity trend compared to the Ni@Pd, Ni-Pd and Pd catalysts. The Pt
active sites on Ni@Pt overlayer catalyst showed higher TOF of deoxygenation product formation
than monometallic Ni and Pt catalysts. From the aforementioned results, it is clear that the Pd or
Pt active sites of Ni@Pd and Ni@Pt overlayer catalysts show much higher deoxygenation
activity than that of Ni, Pd, Pt, Ni-Pd and Ni-Pt samples. These result support the postulate that
18

Ni@Pd and Ni@Pt overlayers with decreased binding strength would improve performance for
the HDO reaction.
As shown in Fig. 11 and 12, reaction temperature had a strong effect on HDO of guaiacol. For
all catalysts, the guaiacol conversion and deoxygenation product selectivity increased as the
temperature was increased to 400°C and 450°C, while selectivity of non-deoxygenation
decreased with increasing temperature. This result is in good agreement with other literature
reports.[67, 68] Under our reaction condition, no ring-saturation products were observed even at
450°C. This is desirable, for aromatic products are more attractive than ring-saturation products
for further application and to reduce H consumption.[3, 21, 69] Several previous studies with
2
similar reaction conditions also did not observe ring-saturation products as well.[39, 70] Sun et
al. reported that, for carbon supported precious metals catalysts, ring-saturation products
observed at 250°C were not observed after temperature had been increased above 350°C due to
ring opening to form C1 products at high reaction temperature.[41] However, this situation does
apply to our case, where no C1 products were observed beside small amount of methanol (1%).
Recently, Tran et al. proposed that high reaction temperature (400°C) and low reaction pressure
could prevent the hydrogenation/dehydrogenation for ring saturation.[70] Again for all catalysts,
the TOFs of deoxygenation products increased with temperature increasing from 350°C to
450°C, as shown in Fig. 13. In contrast, except for the single deposition overlayer catalyst, the
TOFs of non-deoxygenation products decreased slightly from 350°C to 450°C. Above results
suggest that further deoxygenation of phenolic products is more facile at higher temperature than
at low temperature for silica aluminum supported catalysts. Overlayer catalysts showed the
highest TOFs of deoxygenation among tested catalysts with temperature from 350°C to 450°C.
19

Catalyst stability is also important for HDO applications. Some studies have reported that the
bimetallic catalyst could improve catalyst stability.[39, 71] A Au-Pt bimetallic surface with
weaker binding energies of adsorbates than Pt was also reported to inhibit carbon deposition on
the surface.[72] To gain insights into how overlayer catalysts affect the stability of the catalysts,
all catalysts were tested for guaiacol HDO reaction at 450°C for 6 h. The variations of guaiacol
conversion and production selectivity at 450°C as a function of time on stream (TOS) are
compared in Fig.14 and Fig. S1. As shown in the Fig. 14, for Pd only catalyst, the selectivity of
non-deoxygenation product decreased with TOS from 1h to 4h, and the selectivity of phenol and
o-cresol increased slightly with TOS. For Ni and Ni@Pd overlayer catalyst, selectivity of non-
deoxygenation remained stable with TOS, while selectivity of phenol decreased slightly within
first 3h and then kept stable. The selectivity of o-cresol on Ni and Ni@Pd overlayer catalyst
increased slightly with TOS as that of the Pd only catalyst. The axis on the right side of Fig. 14 is
for conversion. The scale of conversion for Ni-Pd and Ni@Pd overlayer is different from that for
Ni and Pd. Unlike selectivity, the conversion of guaiacol on Ni and Pd based catalyst decreased
with TOS. For all Ni and Pd based catalysts, the guaiacol conversion dropped to around forty
percent of initial guaiacol conversion after a reaction time of 6 hours. Severe deactivations for
guaiacol HDO caused by coking were observed for different kinds of catalyst.[20, 21, 39] As
shown in figure S1, the Ni and Pt based catalysts showed similar stabilities as Ni and Pd based
catalysts. Metal loading, acidity and pore structure of support were proposed to affect coke
deposition.[29] Since our catalyst showed similar surface area and pore volume and similar
deactivation ratio with same time, we propose that the catalysts deactivation for these catalysts is
mainly due to coking caused by the acidity of the support.[29, 73] Silica-alumina supported
Ni@Pd overlayer catalyst with a similar support acidity property to silica-alumina supported Ni,
20

Pd and Ni-Pd catalysts couldn’t enhance catalyst stability. The temperature programmed
oxidation was used to investigate the coke deposition of spent catalysts after 6 h of reaction at
4
50°C and the results are presented in the Fig S2. A broad CO
2
peak from around 250°C to
are very similar,
700°C was observed over Ni, Pd and Ni-Pd catalysts. The peak areas of CO
2
suggesting similar coke deposition amount over these three catalysts. This is expected as the
support was same for all three catalysts. The morphology of spent Ni@Pd SD catalyst was
investigated using TEM. As shown in Fig. S3, the morphology of the catalyst didn’t show any
significant change with the mean particle size slightly changed from 12.1 nm to 12.4 nm after 6 h
of reaction at 450°C. This result further supports that the deactivation was due to the coke
deposition.
The conversions of guaiacol over all tested catalysts were below 26% and the major
deoxygenation products were incomplete deoxygenation products (phenol, anisole and o-cresol)
under employed reaction conditions. This was due to the very low W/F (g of catalyst / (g of
-1
reactant h )) employed for obtaining the TOF of the active sites, which is generally used for
comparison of catalytic activity under similar reaction condition at low conversion. To achieve
high guaiacol conversion and see how overlayer catalysts affect the final deoxygenation product
distribution of the guaiacol HDO reaction, a range of W/F from 0.30 h to 4.0 h were tested via
changing the flow rate of the feedstock which was 5 vol.% guaiacol in heptane. The overall
guaiacol conversion and products distribution at different contact times and space times (W/F)
on Pd only and Ni@Pd overlayer catalyst are shown in Fig. 15. It can be seen that, at very low
W/F, both catalysts showed low guaiacol conversion and the major products were phenol, cresol,
non-deoxygenation products and anisole. As W/F is increased, xylenols and complete
deoxygenation products like benzene, toluene, and xylene were observed. The formation of
21

benzene, toluene, and xylene increase monotonically with W/F and in contrast, the selectivity of
phenol, o-cresol, anisole, xylenols and non-deoxygenation products either decreased
monotonically or went through a maximum at intermediate W/F. This implies that the phenol, o-
cresol, anisole, xylenols and non-deoxygenation products are intermediate products. In contrast
with the low conversion rate studies, the selectivity of the complete deoxygenated BTX
(
benzene, toluene, and xylene) products could be up to 82% on Ni@Pd overlayer catalyst at high
W/F condition, while the Pd catalyst showed a selectivity of 73% for BTX under the same
reaction condition. This agrees with results reported by Zhu et al. that benzene, toluene, and
xylene were major complete deoxygenation products on Pt/HBeta catalyst.[29]
Some studies have demonstrated that bimetallic catalysts can change the distribution of HDO
products.[38, 41, 71] Our Ni@Pd overlayer catalyst with slightly reduced binding strength for
reactants and intermediates also showed different deoxygenation products distribution compared
with the pure Pd catalyst. Unlike the Pd only catalyst forming much more toluene than benzene,
the Ni@Pd overlayer catalyst showed similar selectivity for toluene and benzene. One of the
major reaction pathways for producing benzene from guaiacol is that the catechol intermediate is
formed first, then phenol is produced from catechol, and finally phenol is converted into
benzene.[13, 39, 41, 74, 75] The catechol with two C–OH groups usually can be strongly bonded
to a catalyst surface.[39] It is proposed that the Ni@Pd overlayer catalyst with weaker adsorption
strength than Pd can enhance the formation of phenol from catechol and then increase final
benzene selectivity.
According to product distribution at different temperatures and product distribution with
varying W/F, major reaction pathways on SiO
2
-Al
2
3
O supported metals catalysts studied in this
work were proposed and presented in Fig. 16. Most pathways of these proposed reaction
22

pathways are consistent with reaction pathways previously proposed by others.[13, 41, 74, 75]
However, only a few of the literature proposed pathways included the formation of veratrole and
methylguaiacol. In these proposed reaction pathways, formations of catechol, 3-methylcatechol,
veratrole and methylguaiacol from guaiacol were major steps at low conversion or W/F. Since
xylenols and xylenes were observed as major intermediate and final complete deoxygenation
products, pathways involving formation of xylenols and xylenes were also proposed in this work.
A few previous proposed pathways for guaiacol HDO showed steps for formation of xylenols
and xylenes. The reaction network proposed by Runnebaum et al. gave steps for formation of
xylenols and xylenes, but didn’t included proposed steps of formation of xylenes from
xylenols.[74] Only pathways proposed by Resasco et al. for anisole HDO on Pt/HBeta included
these steps.[29] Their experience showed that xylenes were only observed during HDO on
Pt/HBeta, not on Pt/SiO2. It was speculated that the acidic support can catalyze the
transalkylation of oxygenated aromatics, and resulting in transferring methyl groups from one
aromatic molecule to another aromatic molecule instead of forming methane.[29, 76]
Combination of the proposed reaction pathways with the production distribution results from this
work under low guaiacol conversion condition tends to suggest that the catechol, 3-
methylcatechol, veratrole and methylguaiacol could be major initial intermediate products. All of
these three compounds having two functionalized groups with oxygen can have a strong bonding
with the catalyst. We proposed that overlayer catalysts with weaker bonding strength probably
enhance final HDO activity via facilitating the conversion of these intermediate products.
Conclusions
In this work, hydrodeoxygenation of guaiacol was studied using silica alumina supported Ni@Pd
and Ni@Pt overlayer catalysts under atmospheric pressure. Hydrogen chemisorption, ethylene
23

hydrogenation, XRD and TEM studies have demonstrated that palladium or platinum could be
deposited atop supported nickel catalysts and the prepared overlayer showed behavior consistent
to that of computational predictions. When normalized per active site, Pd and Pt active sites of
overlayer catalysts showed significantly enhanced deoxygenation activity compared with that of
Pd or Pt only catalyst. However, overlayer catalysts didn’t show better stability than Ni, Pd or Pt
only catalyst. Results of guaiacol HDO with different W/F showed that selectivity of benzene,
toluene, and xylenes could be up to 80% on Ni@Pd overlayer catalyst at high W/F condition. In
addition, major reaction pathways of guaiacol HDO on silica alumina supported metals catalysts
were proposed.
Reference
[
[
[
[
[
[
[
[
[
[
[
1] C.-H. Zhou, X. Xia, C.-X. Lin, D.-S. Tong, J. Beltramini, Chem. Soc. Rev., 40 (2011) 5588.
2] S. Czernik, A.V. Bridgwater, Energy Fuels, 18 (2004) 590.
3] H. Wang, J. Male, Y. Wang, ACS Catal., 3 (2013) 1047.
4] D.M. Alonso, S.G. Wettstein, J.A. Dumesic, Chem. Soc. Rev., 41 (2012) 8075.
5] G.W. Huber, J.A. Dumesic, Catal. Today, 111 (2006) 119.
6] J.D. Adjaye, N.N. Bakhshi, Fuel Process. Technol., 45 (1995) 161.
7] S. Vitolo, M. Seggiani, P. Frediani, G. Ambrosini, L. Politi, Fuel, 78 (1999) 1147.
8] T.P. Vispute, G.W. Huber, Green Chem., 11 (2009) 1433.
9] J.C. Serrano-Ruiz, J.A. Dumesic, Energy Environ. Sci., 4 (2011) 83.
10] D.C. Elliott, Energy Fuels, 21 (2007) 1792.
11] Q. Bu, H. Lei, A.H. Zacher, L. Wang, S. Ren, J. Liang, Y. Wei, Y. Liu, J. Tang, Q. Zhang,
R. Ruan, Bioresour. Technol., 124 (2012) 470.
[12] O.İ. Şenol, T.R. Viljava, A.O.I. Krause, Catal. Today, 106 (2005) 186.
[13] M.V. Bykova, D.Y. Ermakov, V.V. Kaichev, O.A. Bulavchenko, A.A. Saraev, M.Y.
Lebedev, V.A. Yakovlev, Appl. Catal. B: Environ., 113 (2012) 296.
[14] E.-J. Shin, M.A. Keane, Ind. Eng. Chem. Res, 39 (2000) 883.
[15] Y.X. Yang, C. Ochoa-Hernandez, V.A.D. O'Shea, P. Pizarro, J.M. Coronado, D.P. Serrano,
Appl. Catal. B: Environ., 145 (2014) 91.
[
[
[
[
[
16] C. Zhao, J.A. Lercher, Angew. Chem. Int. Ed., 51 (2012) 5935.
17] C. Zhao, S. Kasakov, J. He, J.A. Lercher, J. Catal., 296 (2012) 12.
18] X. Zhang, Q. Zhang, T. Wang, L. Ma, Y. Yu, L. Chen, Bioresour. Technol., 134 (2013) 73.
19] X. Zhang, T. Wang, L. Ma, Q. Zhang, X. Huang, Y. Yu, Appl. Energy, 112 (2013) 533.
20] R.N. Olcese, M. Bettahar, D. Petitjean, B. Malaman, F. Giovanella, A. Dufour, Appl. Catal.
B: Environ., 115 (2012) 63.
21] R. Olcese, M.M. Bettahar, B. Malaman, J. Ghanbaja, L. Tibavizco, D. Petitjean, A. Dufour,
Appl. Catal. B: Environ., 129 (2013) 528.
[
24

[
[
[
[
(
[
[
[
22] M.K. Huuska, Polyhedron, 5 (1986) 233.
23] Y. Romero, F. Richard, S. Brunet, Appl. Catal. B: Environ., 98 (2010) 213.
24] J. Chang, T. Danuthai, S. Dewiyanti, C. Wang, A. Borgna, ChemCatChem, 5 (2013) 3041.
25] T. Prasomsri, M. Shetty, K. Murugappan, Y. Roman-Leshkov, Energy Environ. Sci., 7
2014) 2660.
26] H.Y. Zhao, D. Li, P. Bui, S.T. Oyama, Appl. Catal. A: Gen., 391 (2011) 305.
27] K. Li, R. Wang, J. Chen, Energy Fuels, 25 (2011) 854.
28] S.-K. Wu, P.-C. Lai, Y.-C. Lin, H.-P. Wan, H.-T. Lee, Y.-H. Chang, ACS Sustainable Chem.
Eng., 1 (2013) 349.
[29] X. Zhu, L.L. Lobban, R.G. Mallinson, D.E. Resasco, J. Catal., 281 (2011) 21.
[30] A. Gutierrez, R.K. Kaila, M.L. Honkela, R. Slioor, A.O.I. Krause, Catal. Today, 147 (2009)
239.
[
[
31] D.Y. Hong, S.J. Miller, P.K. Agrawal, C.W. Jones, Chem. Commun., 46 (2010) 1038.
32] J. Wildschut, F.H. Mahfud, R.H. Venderbosch, H.J. Heeres, Ind. Eng. Chem. Res, 48 (2009)
10324.
[
[
[
[
33] J. Wildschut, I. Melián-Cabrera, H.J. Heeres, Appl. Catal. B: Environ., 99 (2010) 298.
34] C. Li, X. Zhao, A. Wang, G.W. Huber, T. Zhang, Chem. Rev., 115 (2015) 11559.
35] W. Yu, M.D. Porosoff, J.G. Chen, Chem. Rev., 112 (2012) 5780.
36] J. Gao, H. Zhao, X. Yang, B.E. Koel, S.G. Podkolzin, Angew. Chem. Int. Ed., 53 (2014)
3641.
[
[
[
[
37] J. Gao, H. Zhao, X. Yang, B.E. Koel, S.G. Podkolzin, ACS Catal., 3 (2013) 1149.
38] P.T.M. Do, A.J. Foster, J. Chen, R.F. Lobo, Green Chem., 14 (2012) 1388.
39] M.A. Gonzalez-Borja, D.E. Resasco, Energy Fuels, 25 (2011) 4155.
40] A.R. Ardiyanti, S.A. Khromova, R.H. Venderbosch, V.A. Yakovlev, H.J. Heeres, Appl.
Catal. B: Environ., 117–118 (2012) 105.
41] J.M. Sun, A.M. Karim, H. Zhang, L. Kovarik, X.H.S. Li, A.J. Hensley, J.S. McEwen, Y.
Wang, J. Catal., 306 (2013) 47.
[
[
[
[
[
(
[
42] N.A. Khan, H.H. Hwu, J.G. Chen, J. Catal., 205 (2002) 259.
43] E. Christoffersen, P. Liu, A. Ruban, H.L. Skriver, J.K. Nørskov, J. Catal., 199 (2001) 123.
44] J. Greeley, M. Mavrikakis, Nat Mater, 3 (2004) 810.
45] A. Ruban, B. Hammer, P. Stoltze, H.L. Skriver, J.K. Nørskov, J. Mol. Catal. A: Chem., 115
1997) 421.
46] V. Pallassana, M. Neurock, L.B. Hansen, B. Hammer, J.K. Norskov, Phys. Rev. B, 60 (1999)
6146.
[47] V. Pallassana, M. Neurock, J. Catal., 191 (2000) 301.
[
48] M.P. Latusek, R.M. Heimerl, B.P. Spigarelli, J.H. Holles, Appl. Catal. A: Gen., 358 (2009)
79.
[
[
49] M.P. Latusek, B.P. Spigarelli, R.M. Heimerl, J.H. Holles, J. Catal., 263 (2009) 306.
50] M.D. Skoglund, C.L. Jackson, K.J. McKim, H.J. Olson, S. Sabirzyanov, J.H. Holles, Appl.
Catal., A, 467 (2013) 355.
[
[
[
51] C. Zhang, Q. Lai, J.H. Holles, Appl. Catal., A, 526 (2016) 113.
52] C. Zhang, Q. Lai, J.H. Holles, Catal. Sci. Technol., (2016) 11.
53] V.L. Barrio, P.L. Arias, J.F. Cambra, M.B. Güemez, J.M. Campos-Martin, B. Pawelec,
J.L.G. Fierro, Appl. Catal. A: Gen., 248 (2003) 211.
54] V.L. Barrio, P.L. Arias, J.F. Cambra, M.B. Güemez, B. Pawelec, J.L.G. Fierro, Appl. Catal.
A: Gen., 242 (2003) 17.
[
25

[
[
[
[
[
[
[
[
55] J.W. Shabaker, G.W. Huber, J.A. Dumesic, J. Catal., 222 (2004) 180.
56] A. Nash, P. Nash, Bulletin of Alloy Phase Diagrams, 5 (1984) 446.
57] P. Nash, M.F. Singleton, Bulletin of Alloy Phase Diagrams, 10 (1989) 258.
58] T. Teranishi, M. Miyake, Chem. Mater., 10 (1998) 594.
59] A. Aqsha, L. Katta, N. Mahinpey, Catal. Lett., 145 (2015) 1351.
60] M.D. Skoglund, J.H. Holles, Catal. Lett., 143 (2013) 966.
61] S.-C. Chou, S.-H. Lin, C.-T. Yeh, J. Chem. Soc., Faraday Trans., 91 (1995) 949.
62] H. Zea, K. Lester, A.K. Datye, E. Rightor, R. Gulotty, W. Waterman, M. Smith, Appl. Catal.
A: Gen., 282 (2005) 237.
[
[
[
63] F. Zaera, G.A. Somorjai, J. Am. Chem. Soc., 106 (1984) 2288.
64] R.D. Cortright, S.A. Goddard, J.E. Rekoske, J.A. Dumesic, J. Catal., 127 (1991) 342.
65] T. Nimmanwudipong, R.C. Runnebaum, D.E. Block, B.C. Gates, Catal. Lett., 141 (2011)
7
[
3
79.
66] T. Nimmanwudipong, R.C. Runnebaum, D.E. Block, B.C. Gates, Energy Fuels, 25 (2011)
417.
[67] H.Y. Zhao, D. Li, P. Bui, S.T. Oyama, Appl. Catal. A: Gen., 391 (2011) 305.
[
68] Y.-K. Hong, D.-W. Lee, H.-J. Eom, K.-Y. Lee, Appl. Catal. B: Environ., 150–151 (2014)
4
[
38.
69] Y. Hong, H. Zhang, J. Sun, K.M. Ayman, A.J.R. Hensley, M. Gu, M.H. Engelhard, J.-S.
McEwen, Y. Wang, ACS Catal., 4 (2014) 3335.
[70] N.T.T. Tran, Y. Uemura, S. Chowdhury, A. Ramli, Appl. Catal. A: Gen., 512 (2016) 93.
[71] H. Ohta, B. Feng, H. Kobayashi, K. Hara, A. Fukuoka, Catal. Today, 234 (2014) 139.
[72] H. Ren, M.P. Humbert, C.A. Menning, J.G. Chen, Y. Shu, U.G. Singh, W.-C. Cheng, Appl.
Catal. A: Gen., 375 (2010) 303.
73] P.M. Mortensen, J.D. Grunwaldt, P.A. Jensen, K.G. Knudsen, A.D. Jensen, Appl. Catal. A:
Gen., 407 (2011) 1.
[
[
(
[
74] R.C. Runnebaum, T. Nimmanwudipong, D.E. Block, B.C. Gates, Catal. Sci. Technol., 2
2012) 113.
75] V.N. Bui, G. Toussaint, D. Laurenti, C. Mirodatos, C. Geantet, Catal. Today, 143 (2009)
172.
[76] A. Ausavasukhi, Y. Huang, A.T. To, T. Sooknoi, D.E. Resasco, J. Catal., 290 (2012) 90.
26

Figure Captions:
Fig. 1. XRD patterns of SiO
2
-Al
2
O supported (a) Ni and Pd based catalysts and (b) Ni and Pt
3
based catalysts.
Fig. 2. (a) HRTEM image of Ni@Pd SD overlayer particle, (b) the Fourier transform of image
a), (c) Inverse Fourier transform showing inside lattice fringes, (d) Inverse Fourier transform
showing lattice fringes around the particle.
(
Fig.3. NH -TPD profiles of SiO -Al and SiO
3
2
2
O
3
2
-Al
2
O supported (a) Ni and Pd based catalysts
3
and (b) Ni and Pt based catalysts.
Fig. 4. Log-scale H
2
2
isotherms for Pd, Ni, Ni@Pd and Ni-Pd catalysts at 50°C.
isotherms for Pt, Ni, Ni@Pt and Ni-Pt catalysts at 50°C.
Fig. 5. Log-scale H
Fig. 6. Isosteric heat of H
2
adsoption for SiO
2
-Al
2
O
3
supported Ni and Pd based catalysts as
function of volume adsorbed.
Fig. 7. Isosteric heat of H
2
adsoption for SiO
2
2
-Al O
3
supported Ni and Pt catalysts as function of
volume adsorbed.
Fig. 8. Arrhenius-type plot for ethylene hydrogenation on silica-alumina supported Ni and Pd
based catalysts.
Fig. 9. Arrhenius-type plot for ethylene hydrogenation on silica-alumina supported Ni and Pt
based catalysts.
Fig. 10. Correlation between H heat of adsorption and TOF of ethylene hydrogenation on
2
studied overlayer catalysts, pure palladium and platinum catalysts.
Fig. 11. TOFs of guaiacol deoxygenation and non-deoxygenation product formation over Ni and
Pd based catalysts at (a) 350°C, (b) 400°C, (c) 450°C.
Fig. 12. TOFs of guaiacol deoxygenation and non-deoxygenation product formation over Ni and
Pt based catalysts at (a) 350°C, (b) 400°C, (c) 450°C.
Fig. 13. TOFs of guaiacol deoxygenation products over (a) Ni and Pd based catalysts, (b) Ni and
Pt based catalysts at different reaction temperature.
Fig. 14. Conversion of guaiacol and product selectivity over Ni and Pd based catalysts as
function of time on stream.
Fig. 15. Conversion of guaiacol and product distribution over Pd and Ni@Pd SD catalysts as a
function of W/F at 30 min.
Fig. 16. Guaiacol HDO reaction pathways over silica-alumina supported metal catalyst.
Table Captions:
Table 1 Physicochemical properties of samples.
27

Fig. 1. XRD patterns of SiO
2
-Al
2
O supported (a) Ni and Pd based catalysts and (b) Ni and Pt
3
based catalysts.
2
8

Fig. 2. (a) HRTEM image of Ni@Pd SD overlayer particle, (b) the Fourier transform of image
a), (c) Inverse Fourier transform showing inside lattice fringes, (d) Inverse Fourier transform
showing lattice fringes around the particle.
(
2
9