Conversion of biomass-derived butanal into gasoline-range branched hydrocarbon over Pd-supported catalysts

Article abstract of DOI:10.1016/j.catcom.2011.09.022

For production of gasoline-range branched hydrocarbon from butanal, Pd catalysts supported on different metal oxides were applied. Among the prepared catalysts, Pd/ZrO₂ showed the complete butanal conversion with the formation of C₇-to-C₉ branched hydrocarbon (75% yield). Additionally, the ratios of O/C and straight-chain to branched hydrocarbon (n-C/br-C) were found to be 0.005 and 0.17, respectively. This indicates that an adequate combination of Pd dispersion and amphoteric ZrO₂ character promoted hydrodeoxygenation, C-C coupling and isomerization reactions. Consequently, both Pd dispersion and acid-base properties of supports are suggested to play a pivotal role in producing gasoline-range hydrocarbon at a high yield.

Full text of DOI:10.1016/j.catcom.2011.09.022

Catalysis Communications 16 (2011) 108–113
Contents lists available at SciVerse ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
Conversion of biomass-derived butanal into gasoline-range branched hydrocarbon
over Pd-supported catalysts
Sung Min Kim a, Mi Eun Lee a,b, Jae-Wook Choi a, Dong Jin Suh a, Young-Woong Suh c,
⁎
a
Clean Energy Center, Korea Institute of Science and Technology, Seoul 136–791, Republic of Korea
Department of Chemical Engineering, University of Seoul, Seoul 130–743, Republic of Korea
b
c
Department of Chemical Engineering, Hanyang University, Seoul 133–791, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 August 2011
Received in revised form 19 September 2011
Accepted 19 September 2011
Available online 28 September 2011
For production of gasoline-range branched hydrocarbon from butanal, Pd catalysts supported on different metal
oxides were applied. Among the prepared catalysts, Pd/ZrO2 showed the complete butanal conversion with the
formation of C7-to-C9 branched hydrocarbon (75% yield). Additionally, the ratios of O/C and straight-chain to
branched hydrocarbon (n-C/br-C) were found to be 0.005 and 0.17, respectively. This indicates that an adequate
combination of Pd dispersion and amphoteric ZrO2 character promoted hydrodeoxygenation, C–C coupling and
isomerization reactions. Consequently, both Pd dispersion and acid–base properties of supports are suggested to
play a pivotal role in producing gasoline-range hydrocarbon at a high yield.
Keywords:
Biomass-derived butanal
Heterogeneous catalysis
C–C coupling
© 2011 Elsevier B.V. All rights reserved.
Hydrodeoxygenation
Pd-supported catalysts
1. Introduction
catalytically converted into butanal which is further used as a raw
material for self-aldol condensation to 2-ethylhexanal (O/C and H/C
In recent years, an efficient pathway to convert renewable biomass
sources into fuels and chemicals has been studied in the field of cataly-
sis, because the development of such a catalytic process is a stringent
challenge due to complexity of the major biomass components with
higher oxygen functionalities [1]. Particularly, the practically feasible
approach is required to simultaneously achieve lower O/C and higher
H/C ratios close to those of conventional fossil fuels. However, this ulti-
mate goal cannot be accomplished only using direct fermentation into
bio-alcohols, because the product yield and the energy ratio (defined
as (heating value of alcohol+other fuel or power export)/(fossil fuels
input−process co-products)) are not satisfactory, and the O/C ratios
of alcohols (0.5 for ethanol and 0.25 for butanol) are still far from that
of current gasoline. Indirect fermentation was hence suggested to afford
about 50% improved yield of alcohols than direct fermentation [2,3].
This process consists of three steps, such as anaerobic fermentation to
carboxylic acids, their condensation to esters and subsequent hydroge-
nolysis to alcohols. However, since this strategy is also targeted at bio-
alcohols, additional endeavor to decrease the O/C ratio would be
necessary.
ratios of 0.125 and 2, respectively) and subsequent hydrodeoxygena-
tion reactions. This is in accordance with the recent focus of study
on C–C coupling (ketonization and aldol condensation) [6–8] for the
purpose of converting biomass-derived oxygenated compounds into
fuel-grade hydrocarbon. When our strategy would be successful, liquid
products obtained are comprised of a majority of hydrocarbon with
4–12 carbon atoms per molecule as well as little oxygenate compounds,
and, in turn, their O/C and H/C ratios become close to those of current
liquid fuels including gasoline. Here, it should be emphasized that the
formation of branched (br-C), rather than straight-chain (n-C), hydro-
carbon is desired, due to their high research octane number.
Since the catalytic reduction of butyric acid to butanal was well
documented in previous reports [9,10], our effort has mainly been de-
voted to demonstrating the one-pot conversion of butanal via aldol
compounds into fuel-grade products. Pd was hence chosen as an active
metal, because Pd catalysts are commonly used for condensation/
coupling reactions [7,11,12], and also promote the hydrogenation
of oxygenated compounds [13,14]. The other issue in catalyst develop-
ment is to find an appropriate support for the desired conversion,
because acidic/basic properties of supports would dominantly govern
the catalytic activity and product distribution in the aldol condensation
reaction [15,16]. Additionally, balancing the contributions given by
metal and support would be crucial in controlling the relative rates of
C–C coupling and C–O cleavage. Thus, six supports with different acidi-
ties (or basicities), such as CeO2, MgO, MgZrO3, ZrO2, Al2O3, and SiO2–
Al2O3, were utilized for Pd loading.
In this work, a series of C–C coupling and hydrodeoxygenation re-
actions in a single catalytic system has been conducted for producing
gasoline-range hydrocarbon from butyraldehyde (butanal). Typically,
butyric acid, produced through anaerobic digestion of sugars [4,5], is
⁎
Corresponding author. Tel.: +82 2 2220 2329; fax: +82 2 2298 4101.
E-mail address: ywsuh@hanyang.ac.kr (Y.-W. Suh).
1566-7367/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.catcom.2011.09.022

S.M. Kim et al. / Catalysis Communications 16 (2011) 108–113
109
2. Experimental
between reactor inlet and outlet, while the selectivity and yield for
a given carbon species (i) were calculated as follows:
2.1. Catalyst preparation
Gas− or liquid−phase carbon selectivity ð%Þ
¼ ðRi–outlet×NiÞ=ðRbutanal–reacted×NbutanalÞ×100
Pd-supported catalysts with the nominal Pd loading of 0.1 wt.%
were prepared via an incipient wetness impregnation method with
an amount of aqueous Pd(II) nitrate hydrate solution onto CeO2
(Sigma-Aldrich; BET area=42 m2/g), MgO (Sigma-Aldrich; BET
area=57 m2/g), MgZrO3 (Sigma-Aldrich; BET area=45 m2/g), ZrO2
(Sigma-Aldrich; BET area=30 m2/g), SiO2-Al2O3 (Sigma-Aldrich;
BET area=645 m2/g) or Al2O3 (Alfa Aesar; BET area=124 m2/g).
The prepared catalysts were then dried at 393 K overnight and cal-
cined in air at 773 K for 2 h.
Carbon yield ð%Þ ¼ ðRi–outlet×NiÞ=ðRbutanal–inlet×NbutanalÞ×100
O=Cðmol=moÞ and H=Cðmol=molÞ
¼ ∑ðRi–outlet×Mi;O or HÞ=∑ðRi–outlet×NiÞ
n−C=br−Cðmol=molÞ ¼ ∑ðn−Ri–outlet×NiÞ=∑ðbr−Ri–outlet×NiÞ;
where R, N and M represent the molar flow rate, the carbon number
of a given species (i), and the oxygen or hydrogen number of a given
sepecies (i), respectively. In the case of n-Ri-oulet and br-Ri-outlet, both
oxygenated hydrocarbon and hydrocarbon are included. All catalytic
data shown in this study were obtained at the reation period of 4 h,
becuase the conversion of butanal and the carbon selectivities of car-
bon species obtained over all Pd-supported catalysts were almost
unchanged up to the reaction period of 10 h. This implies that Pd-
supported catalysts are durable for the convesion of butanal into
fuel-grade compounds.
2.2. Catalyst characterization
In order to measure the Pd loading in the prepared Pd-supported
catalysts, the elemental analysis (EA) was carried out using a Varian
710–ES. Also, Pd dispersion was determined by CO chemisorption at
323 K using a BEL-CAT instrument (BEL Japan, Inc.) equipped with a
thermal conductivity detector (TCD), where the stoichiometry factor
between Pd and adsorbed CO is assumed to be 0.6 on the basis of pre-
vious reports [17,18]. Typically, the prepared catalyst (ca. 0.1 g) was
reduced in H2 at 623 K for 1 h and then purged in He at 623 K for
30 min. After cooling to 323 K, a pulse chemisorption was started.
CO2 and NH3 temperature-programmed desorption (TPD) experi-
ments were carried out using the same instrument as above. In a typical
experiment, about 50 mg of prepared catalyst was loaded and reduced
in H2 at 623 K for 1 h. The reactor was then purged in He at 623 K for
30 min and cooled to 313 K for CO2 adsorption or to 473 K for NH3
adsorption. After exposure of the reduced catalyst to a flow of 5.98%
CO2/He or 5% NH3/He (50 ml/min) for 1 h, residual gases was removed
by purging pure He for 1 h at 313 K for CO2 or at 473 K for NH3. TPD
profiles were then obtained by ramping the temperature at a heating
rate of 5 K/min under pure He (50 ml/min) to 1073 K for CO2–TPD or
to 1173 K for NH3–TPD. The quantitative amount of adsorbed CO2 or
NH3 was measured by comparison of desorption peak areas with the
calibration value calculated from a pulse of a known volume (sampling
loop=500 μl) of the studied gas. This procedure was repeated 5–10
times until the calibration value falls within 2% error. This procedure
was provided by BEL Japan, Inc., which was stated in the previous
work reported by Bravo-Suárez et al. [19].
3. Results and discussion
3.1. Catalyst characterization
Based on the Pd loading measured by ICP analysis (wt.%) and the
CO uptake by CO chemisorption (μmol/gcat), the Pd dispersion (%)
was calculated (Table 1): as a result, it decreased in the order of
Pd/CeO2 NPd/Al2O3 NPd/SiO2–Al2O3 NPd/ZrO2 NPd/MgZrO3 ≈Pd/MgO
(from 44.7% to 11.7%). In consideration of both the Pd loading and
the BET area of the supports, the Pd dispersion of Pd-supported cata-
lysts except Pd/CeO2 appeared to be fairly reasonable. In the case of
Pd/CeO2, an over-estimation of Pd dispersion is possible due to the
formation of surface carbonate species on CeO2 [20] and/or the reduc-
tion of surface Ce4+ into Ce3+ by hydrogen spillover taking place at
the interface of the metal and the support [7,21].
Based on the results reported previously [7,22–25], NH3–TPD and
CO2–TPD profiles shown in Fig. 1 were divided into three and two dis-
tinct regions, respectively: in the case of former profiles, weak and
strong NH3 adsorption sites were separated at 800 K [22,23], while
the latter profiles represented weak (340–500 K), medium (500–
700 K) and strong (above 700 K) CO2 adsorption sites [7,24,25].
A quantitative number of acid and base sites are summarized in
Table 1, where the unit is μmol/gcat. As a result, Pd catalysts supported
on MgZrO3, ZrO2, Al2O3 and SiO2–Al2O3 showed the surface acidity,
whereas those on CeO2, MgO, MgZrO3 and ZrO2 exhibited the basic
property. Particularly, the addition of ZrO2 into MgO increased
the total amount of acid sites to 163 μmol/gcat, and decreased the
total amount of base sites from 519 to 422 μmol/gcat. In the case of
Pd/ZrO2 catalyst, the total amounts of acid base sites were found to
be 198 and 273 μmol/gcat, respectively, where the strong NH3 and
CO2 adsorption sites were 117 and 0 μmol/gcat, respectively. Further-
more, Pd/Al2O3 and Pd/SiO2–Al2O3 catalysts mainly contained acid
sites, even if the former showed a weak basic character. It should be
noted here that Pd/ZrO2 and Pd/MgZrO3 catalysts contained both sur-
face acidity and basicity. This implies that the amphoteric property
of the two catalysts lead to higher activities on dehydration, isomeri-
zation and aldol condensation reactions by acid and/or base sites.
2.3. Activity test
The catalytic reaction to convert butanal into fuel-grade com-
pounds was conducted in a stainless steel reactor (10 mm i.d. and
120 mm length), where the calcined catalyst (2.54 mL in all cases)
was loaded in the middle. Prior to the activity test, the catalyst was
reduced in a flow of H2 at 623 K for 2 h. The reaction system was pres-
surized to 10 bar with H2 or N2 and stabilized for 1 h at 673 K. This
reduction is inevitable prior to the supply of the reactant for the
reaction, though Pd sintering takes place in a certain degree (the
decrease of Pd dispersion was observed to be 3–4% in our work).
Then, the reactant was fed to the catalyst bed using an HPLC pump,
where the molar ratio of H2 or N2 and reactant was 6. In order to
maintain the reactant to be a vapor phase, the flow line between
the pump and reactor inlet was kept constant at 473 K. In all activity
tests, the liquid hourly space velocity (LHSV) and gas hourly space
velocity (GHSV) were fixed at 1.2 and 2049 h−1, respectively. During
the catalytic run, gaseous product was periodically analyzed using an
online GC equipped with TCD and FID installed with Carboxen 1000
and HP-PLOT/Al2O3 columns, respectively. Finally, liquid product
collected in the separator was analyzed using a GC equipped with
FID and MS installed with HP-5 capillary column. The conversion of
butanal was determined by the difference of butanal concentration
3.2. Catalytic activity of Pd-supported catalysts
Table 2 summarizes the reaction results obtained over Pd-sup-
ported catalysts at 673 K, 10 bar H2, and 1.2 h−1 LHSV. Pd/CeO2 cata-
lyst totally converted butanal into gas-phase products including
CO and C4− (C3H8, and C3H6), due to promoted C–C cleavage (i.e.,

110
S.M. Kim et al. / Catalysis Communications 16 (2011) 108–113
Table 1
Quantification of metal (CO), acid (NH3) and base sites (CO2) of Pd-supported catalysts.
Catalyst
Pd loading
(wt.%)
CO uptake
Pd dispersiona
(%)
TPD–NH3 (μmol/gcat
)
TPD–CO2 (μmol/gcat
)
Total acid sites/total
base sites
(μmol/gcat
)
800 KNT
800 KbT
500 KNT
500bTb700 K
700 KbT
Pd/CeO2
Pd/MgO
Pd/MgZrO3
Pd/ZrO2
Pd/Al2O3
0.09
0.08
0.09
0.11
0.09
0.10
2.26b
0.55
0.60
1.05
2.24
1.45
44.7b
12.3
11.7
16.8
44.1
25.7
0
0
69
81
228
327
0
0
94
117
182
152
92
67
185
205
32
32
182
120
68
0
170
270
117
0
0
0
0/294
0/519
163/422
198/273
410/32
479/0
Pd/SiO2–Al2O3
0
0
a
The Pd dispersion is determined on the basis of the assumption that the stoichiometry factor between Pd and adsorbed CO is 0.6 [17,18].
The CO uptake and Pd dispersion for Pd/CeO2 are over-estimated because of CO chemisorption on reduced sites of CeO2 [7,20,21].
b
decarbonylation into CO and C3H8) by highly dispersed Pd particles
exhibited the lower conversions of 74% and 77%, respectively, and
the relatively higher yields toward CO. Also, these catalysts mainly
yielded butanol, indicating that the hydrogenation by Pd site is
more dominant than C–C coupling by acid or base sites. In the case
of Pd/ZrO2, Pd/Al2O3 and Pd/SiO2–Al2O3 catalysts with medium-to-
high Pd dispersion as well as acid sites, the complete conversion
with higher yields of liquid products than 70% was obtained. The for-
mation of O-free hydrocarbon was also favorable, and the carbon
number in liquid products ranged from 4 to 12, which is very close
to that of gasoline. Among the catalysts tested in this work, the lowest
n-C/br-C and O/C ratios of 0.17 and 0.005, respectively, were obtained
with Pd/ZrO2 catalyst, where the branched hydrocarbon obtained
with Pd/ZrO2 (75% yield) mainly consisted of 3-methylhexane and
2-methylheptane (Table 2c).
(Table 2a). This result is in accordance with Huber and coworkers’ re-
port wherein the higher reactivity in the decarbonylation of aldehyde
compounds was observed over metal sites [26]. In contrast, other
catalysts produced liquid hydrocarbon, n-C and br-C, in addition to
the above gases (Table 2b and c). Pd/MgO and Pd/MgZrO3 catalysts,
which are highly basic and have a low Pd dispersion of ca. 6%,
3.3. Proposed reaction pathway
For in-depth understanding of the above results, our work
was then focused on describing the reaction network depicted in
Fig. 2, which was based on the results presented in Table 2. Although
the primary reaction would be the self-aldol condensation of
butanal on acid and base sites [8], the condensed product, 2-ethyl-
3-hydroxyhexanal, was not detected in liquid products. Instead, 2-
ethylhexenal and 4-heptanone were formed by dehydration on acid
sites, and a couple of C–C cleavage and dehydrogenation on Pd
metal sites, respectively. Accordingly, the rapid consumption of 2-
ethyl-3-hydroxyhexanal prevented the reverse aldol reaction, result-
ing in a high conversion of butanal. However, water produced in the
conversion to 2-ethylhexenal led to disproportionation (cannizaro
reaction) of butanal into butyric acid (oxidation of butanal) and buta-
nol (reduction of butanal) [27,28], which further underwent not
only self-ketonization of butyric acid to produce 4-heptanone [29]
but inter-esterification between butyric acid and butanol to produce
butyl butyrate [30]. The hydrogenation of 4-heptanone and the decar-
bonylation/dehydration of 2-ethyl-3-hydroxyhexanal then yielded 3-
heptene via 4-heptanol, followed by hydrogenation to heptane and
further isomerization to 3-methylhexane taking place on acid sites.
On the other hand, 2-ethylhexenal was hydrogenated into 2-ethyl-
hexenol and 2-ethylhexanal competitively on Pd metal sites. Mean-
while, the C–C cleavage of 2-ethylhexenal into ethane and hexanal
could occur slightly. Hexanal was then hydrogenated into hexanol
or coupled with butyric acid to yield 4-nonanone that was converted
to nonane, 4-octene, 2-methylheptane and 2-methyloctane, analo-
gous with 4-heptanone. In the case of 2-ethylhexanal, the C–C
cleavage occurred to form heptane which is further isomerized to 3-
methylhexane. Finally, 2-ethylhexanol, the hydrogenated product
of 2-ethylhexanal, underwent esterification to 2-ethylhexyl butyrate.
As represented in Table 3, the experiments were conducted
over unsupported ZrO2 under H2 and over Pd/ZrO2 under N2 in
order to explicitly understand the contributions of Pd and ZrO2,
where the Pd content was 0.1 wt.%, the LHSV 1.2 h−1, the N2/butanal
or H2/butanal ratio 6 and the pressure 10 bar. Over ZrO2 under H2, 2-
ethylhexananl, 2-ethylhexenal and 2-ethylhexene were mainly
Fig. 1. CO2–TPD (a) and NH3–TPD (b) profiles of Pd-supported catalysts.

S.M. Kim et al. / Catalysis Communications 16 (2011) 108–113
111
Table 2
Product distributions of overall compounds (a), oxygenated hydrocarbon (b) and O-free hydrocarbon (c) obtained by the conversion of butanal over Pd-supported catalysts.
(a)
Catalyst
Conv. (%)
Gas-phase carbon selectivity (%)
Liquid-phase carbon selectivity (%)
Oxygenated hydrocarbon
Hydrocarbon
C8−
CO
CO2
C4−
C4
1.7
0.1
0.0
0.4
0.5
C4+
C8−
C8
C8+
C8
0.0
0.0
3.7
C8+
Pd/CeO2
Pd/MgO
100
74
20.4
17.0
11.1
4.4
7.4
1.5
2.2
1.8
0.5
0.6
1.8
0.3
73.9
16.0
16.2
3.4
2.0
2.6
0.0
0.0
0.0
0.6
0.4
9.6
0.0
19.7
26.4
2.8
27.4
5.8
0.0
7.7
0.0
0.0
11.6
13.5
0.0
16.3
8.8
0.3
0.0
0.0
5.1
0.0
0.0
0.0
5.2
0.0
7.1
Pd/MgZrO3
Pd/ZrO2
Pd/Al2O3
Pd/SiO2-Al2O3
77
15.7
60.3
39.8
14.6
100
100
100
19.3
6.7
28.4
14.7
0.0
(b)
Catalyst
Carbon yield of oxygenated compounds (%)
Butanol
4-heptanone
Butyl butyrate
3.8
2-ethylhexanal
1.7
2-ethylhexenal
2.0
2-ethyl hexylbutyrate
Pd/CeO2
Pd/MgO
Pd/MgZrO3
Pd/ZrO2
16.9
24.4
0.4
2.8
2.0
2.4
16.3
8.8
0.3
Pd/Al2O3
Pd/SiO2-Al2O3
7.1
1.2
20.3
4.6
5.8
5.9
13.5
(c)
Catalysts
Carbon yield of O-free compounds (%)
3-heptene
4-octene
Nonane
3-methylhexane
2-ethylhexene
2-methylheptane
2-methyloctane
Pd/CeO2
Pd/MgO
Pd/MgZrO3
Pd/ZrO2
Pd/Al2O3
Pd/SiO2-Al2O3
5.1
15.7
56.1
38.0
12.7
3.7
4.2
1.8
1.9
2.4
3.2
7.1
16.9
4.4
2.0
2.3
14.2
14.2
Fig. 2. Reaction network on the conversion of butanal over Pd-supported catalysts. The asterisk represents reaction intermediates undetected in product effluents.

112
S.M. Kim et al. / Catalysis Communications 16 (2011) 108–113
Table 3
Product distributions of oxygenated hydrocarbon (a) and O-free hydrocarbon (b) obtained by the conversion of butanal over Pd/ZrO2 under N2 and over unsupported ZrO2 under H2.
(a)
Catalyst
Pd/ZrO2
Conversion
(%)
Carbon yield of oxygenated compounds (%)
Butanol
4-heptanone
Butyl butyrate
2-ethylhexanal
2-ethylhexenal
4-nonanone
2-ethyl hexylbutyrate
a
98
98
7.0
0.0
41.8
0.0
6.8
1.6
10.9
25.2
0.0
11.8
2.4
6.1
0.0
2.8
b
ZrO2
(b)
Catalysts
Carbon yield of O-free compounds (%)
3-heptene
4-octene
Nonane
3-methylhexane
2-ethylhexene
2-methylheptane
2-methyloctane
a
Pd/ZrO2
ZrO2
20.0
14.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
31.3
0.0
0.0
0.0
0.0
b
a
Reaction condition: 673 K, 1.2 h−1 LHSV, 6 N2/butanal, 10 bar N2.
Reaction condition: 673 K, 1.2 h−1 LHSV, 6 H2/butanal, 10 bar H2.
b
produced, resulting from the self-aldol condensation of butanal
into 2-ethyl-3-hydroxyhexanal and its subsequent dehydration. This
means that ZrO2 support with acid and base sites plays a significant
role on C–C coupling and oxygen removal through dehydration. On
the other hand, the experiment over Pd/ZrO2 in the absence of H2
yielded the oxygenated hydrocarbon with ca. 68.9% carbon yield
(mainly, 4-heptanone and 2-ethylhexanal) but the O-free hydrocar-
bon with 20.0% carbon yield (only 3-heptene). Since 3-heptene
is formed through decarbonylation of 2-ethyl-3-hydroxyhexanal
and subsequent dehydration (explained above), it was confirmed
that Pd promoted the decarbonylation reaction to cleave the bond
between the carbonyl carbon and its nearest carbon. Additionally,
the carbon yield of oxygenated hydrocarbon was remarkably higher
under N2 than under H2 over Pd/ZrO2 catalyst, indicating that the
hydrodeoxygenation reaction occurs vigorously Pd/ZrO2 sites because
Pd activates H2 molecules.
A number of reactions described above could be divided into three
classes: Class I) coupling reactions, including aldol condensation,
ketonization and esterification on acid or base sites; Class II) hydroge-
nation, dehydrogenation and C–C cleavage on metal sites; and Class
III) isomerization and dehydration on acid sites. The results obtained
over Pd/ZrO2 strongly suggest that Pd promotes the Class II reaction
while the Class I and III reactions occurs on ZrO2 support. Thus, the
bifunctionality of catalyst is considered to be essential for effective
reduction of O/C ratio in butanal-derived liquid compounds.
reactions; however, the n-C/br-C ratio was much higher over
Pd/ZrO2 (6.18) than over ZrO2 (2.19) under N2 atmosphere. This is
due to the production of linear compounds through decarbonylation
of coupled products (e.g., 2-ethylhexenal) on metal sites.
4. Conclusions
Among the prepared Pd-supported catalysts, the outstanding yield
of C7-to-C9 branched hydrocarbon from butanal could be achieved
over Pd/ZrO2 through a series of Classes I (C–C coupling on acid or
base sites), II (hydrogenation, dehydrogenation and C–C cleavage on
metal sites) and III (isomerization and dehydration on acid sites)
reactions. In detail, ZrO2 support exhibited the amphoteric property
such as acidity and basicity, thus leading to higher activities on C–C
coupling, dehydration and isomerization reactions, while Pd with
the medium-to-high dispersion promoted the hydrodeoxygenation
reaction to obtain the higher hydrocarbon yield. Therefore, a deliber-
ate tuning of Pd dispersion and acidic-basic support property is obvi-
ously necessary in maximizing the yield of O-free hydrocarbon with
the lower n-C/br-C derived from butanal. Consequently, the present
finding will enable biomass-derived liquid products to be improved
up to a conventional gasoline fuel, thus offering new possibility for
converting cellulosic biomass into several types of hydrocarbon fuels.
3.4. Quality of liquid products obtained over Pd-supported catalysts
In order to utilize biofuels as a transportation fuel, lower ratios of
O/C and n-C/br-C were required because of lower CO2 emission and
higher octane number. Thus, the modified version of van Krevelen di-
agram (Fig. 3) was utilized to visualize the fuel quality of liquid prod-
ucts reported herein. The O/C and n-C/br-C ratios were decreased in
the order of Pd/MgONPd/MgZrO3 NPd/ZrO2 (from 0.168 to 0.005 in
the former ratio and from 0.93 to 0.17 in the latter). Such a trend is
caused by a higher acid density leading to the pronounced Class III
reaction, along with the progress of Classes I and II reactions. This is
in the same line with the decrease of both ratios between Pd/Al2O3
(less acidic) and Pd/SiO2–Al2O3 (more acidic), where the O/C and n-
C/br-C ratios were 0.066 and 0.52 on the former catalyst, and 0.036
and 0.27 on the latter, respectively. However, the two catalysts exhib-
ited the higher ratios than Pd/ZrO2, indicating that both base and
acid sites are essential to obtain lower O/C and n-C/br-C ratios. This
was confirmed by the result that the O/C ratio was lower over ZrO2,
which is well known to retain an amphoteric character [31,32],
under N2 (0.127), compared to that over basic Pd/MgO under H2
(0.168). On the other hand, the effect of Pd loading was evident in
terms of the O/C ratio due to the occurrence of hydrodeoxygenation
Fig. 3. Modified van Krevelen diagram of liquid products obtained over Pd-supported
catalysts. Red and blue circles indicate the supply of H2 and N2, respectively, for the
conversion of butanal. The n-C/br-C ratio of butanal is arbitrarily assumed to be 10.

S.M. Kim et al. / Catalysis Communications 16 (2011) 108–113
113
[13] D. Procházková, P. Zámostný, M. Bejblová, L. Cervený, J. Cejka, Applied Catalysis A
332 (2007) 56–64.
Acknowledgement
[14] D. Mei, M. Neurock, C.M. Smith, Journal of Catalysis 268 (2009) 181–195.
[15] H. Tsuji, F. Yagi, H. Hattori, H. Kita, Journal of Catalysis 148 (1994) 759–770.
[16] F. King, G.J. Kelly, Catalysis Today 73 (2002) 75–81.
[17] R.S. Monteiro, L.C. Dieguez, M. Schmal, Catalysis Today 65 (2001) 77–89.
[18] V. Höller, K. Rådevik, I. Yuranov, L. Kiwi-Minsker, A. Renken, Applied Catalysis B:
Environmental 32 (2001) 143–150.
This work was supported by the National Research Foundation
of Korea Grant funded by the Korean Government (NRF-2009-
C1AAA001-0093293).
[19] J.J. Bravo-Suarez, K.K. Bando, T. Akita, T. Fujitani, T.J. Fuhrer, S.T. Oyama, Chemical
Communications (2008) 3272–3274.
[20] T. Takeguchi, S. Manabe, R. Kikuchi, K. Eguchi, T. Kanazawa, S. Matshmoto, W.
Ueda, Applied Catalysis A 293 (2005) 91–96.
[21] J.C. Serrano-Ruiz, J. Luettich, A. Sepúlveda-Escribano, F. Rodríguez-Reinoso, Journal
of Catalysis 241 (2006) 45–55.
References
[1] A.J. Ragauskas, C.K. Williams, B.H. Davison, G. Britovsek, J. Cairney, C.A. Eckert, W.J.
Frederick, J.P. Hallett, D.J. Leak, C.L. Liotta, J.R. Mielenz, R. Murphy, R. Templer, T.
Tschaplinski, Science 311 (2006) 484–489.
[2] T. Eggeman, D. Verser, Applied Biochemistry and Biotechnology 129–132 (2006)
361–381.
[22] C. Amairia, S. Fessi, A. Ghorbei, A. Rîves, Journal of Molecular Catalysis A 332
(2010) 25–31.
[3] C.B. Granda, L. Zhu, M.T. Holtzapple, Environmental Progress 26 (2007) 233–250.
[4] J.I. Horiuchi, T. Shimizu, K. Tada, T. Kanno, M. Kobayashi, Bioresource Technology
82 (2002) 209–213.
[5] Y. Zhu, Z. Wu, S.-T. Yang, Process Biochemistry 38 (2002) 657–666.
[6] C.A. Gaertner, J.C. Serrano-Ruiz, D.J. Braden, J.A. Dumesic, Journal of Catalysis 266
(2009) 71–78.
[7] E.L. Kunkes, E.I. Gürbüz, J.A. Dumesic, Journal of Catalysis 266 (2009) 236–249.
[8] A. Gangadharan, M. Shen, T. Sooknoi, D.E. Resasco, R.G. Mallinson, Applied Catalysis
A 385 (2010) 80–91.
[23] J. Chen, Q. Zhang, Y. Wang, H. Wan, Advanced Synthesis and Catalysis 350 (2008)
453–464.
[24] D. Bianchi, T. Chafik, M. Khalfallah, S.J. Teichner, Applied Catalysis A: General 112
(1994) 219–235.
[25] J.I. Di Cosimo, G. Torres, C.R. Apesteguía, Journal of Catalysis 208 (2002) 114–123.
[26] N. Li, G.W. Huber, Journal of Catalysis 270 (2010) 48–59.
[27] X. Wang, R.Y. Saleh, U.S. Ozkan, Journal of Catalysis 231 (2005) 20–32.
[28] A. Wawrzetz, B. Peng, A. Hrabar, A. Jentys, A.A. Lemonidou, J.A. Lercher, Journal of
Catalysis 269 (2010) 411–420.
[9] T. Yokoyama, N. Yamagata, Applied Catalysis A 221 (2001) 227–239.
[10] T. Yokoyama, N. Fujita, Applied Catalysis A 276 (2004) 179–185.
[11] K.C. Nicolaou, P.G. Bulger, D. Sarlah, Angewandte Chemie International Edition 44
(2005) 4442–4489.
[12] E.I. Gürbüz, E.L. Kunkes, J.A. Dumesic, Applied Catalysis B: Environmental 94
(2010) 134–141.
[29] O. Nagashima, S. Sato, R. Takahashi, T. Sodesawa, Journal of Molecular Catalysis A:
Chemical 227 (2005) 231–239.
[30] X. Han, L. Zhou, Chemical Engineering Journal 172 (2011) 459–466.
[31] K. Tanabe, T. Yamaguchi, Catalysis Today 20 (1994) 185–197.
[32] Y. Li, D. He, Q. Zhu, X. Zhang, B. Xu, Journal of Catalysis 221 (2004) 584–593.