Catalytic decarboxylation of non-edible oils over three-dimensional, mesoporous silica-supported Pd

Article abstract of DOI:10.1016/j.molcata.2016.03.023

Deoxygenation of fatty acids (oleic and stearic acids) and non-edible oil (jatropha oil) over Pd(1-5 wt%) supported on two structurally different, three-dimensional, mesoporous silica (SBA-12 and SBA-16) catalysts was investigated. Pd/SBA-16 (cubic mesoporous structure with space group Im3ˉm) showed higher catalytic activity than Pd/SBA-12 (hexagonal mesoporous structure with space group p6₃/mmc). The influence of reaction parameters like temperature, H₂ pressure and Pd content as well as the nature of the feedstock on catalytic activity and product selectivity was studied. A temperature of above 320 °C, reaction time of 5 h and Pd content (on silica surface) of 3 wt% enabled complete conversion of the fatty compounds into diesel-range hydrocarbons. Deoxygenation proceeded through hydrodeoxygenation and decarboxylation mechanisms when a saturated (stearic) acid was used as a feed while it advanced mainly through decarboxylation route when an unsaturated (oleic) acid was employed. Higher surface hydrophobicity and smaller size particles of Pd are the possible causes for the superior catalytic activity of Pd/SBA-16.

Full text of DOI:10.1016/j.molcata.2016.03.023

Journal of Molecular Catalysis A: Chemical 417 (2016) 126–134
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Catalytic decarboxylation of non-edible oils over three-dimensional,
mesoporous silica-supported Pd
∗
Ravindra Raut, Vikram V. Banakar, Srinivas Darbha
Catalysis and Inorganic Chemistry Division, CSIR-National Chemical Laboratory, Pune 411 008, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Deoxygenation of fatty acids (oleic and stearic acids) and non-edible oil (jatropha oil) over Pd(1–5 wt%)
supported on two structurally different, three-dimensional, mesoporous silica (SBA-12 and SBA-16) cat-
alysts was investigated. Pd/SBA-16 (cubic mesoporous structure with space group Im 3¯ m) showed higher
Received 11 February 2016
Received in revised form 8 March 2016
Accepted 9 March 2016
catalytic activity than Pd/SBA-12 (hexagonal mesoporous structure with space group p63/mmc). The influ-
Available online 11 March 2016
ence of reaction parameters like temperature, H2 pressure and Pd content as well as the nature of the
◦
feedstock on catalytic activity and product selectivity was studied. A temperature of above 320 C, reaction
Keywords:
Biofuel
Vegetable oil
Deoxygenation
Mesoporous silica
Supported palladium
Diesel-range hydrocarbons
time of 5 h and Pd content (on silica surface) of 3 wt% enabled complete conversion of the fatty compounds
into diesel-range hydrocarbons. Deoxygenation proceeded through hydrodeoxygenation and decarboxy-
lation mechanisms when a saturated (stearic) acid was used as a feed while it advanced mainly through
decarboxylation route when an unsaturated (oleic) acid was employed. Higher surface hydrophobicity
and smaller size particles of Pd are the possible causes for the superior catalytic activity of Pd/SBA-16.
© 2016 Elsevier B.V. All rights reserved.
1
. Introduction
According to the world energy council, about 82% of the energy is
use of FAMEs can cause problems in vehicles due to their high
oxygen content and viscosity rendering them a less than an ideal
fuel for conventional engines. For example, the unburnt part of
biodiesel, when mixed with the lubricating oil, can promote engine
aging [9]. Further, FAMEs are associated with issues related to cold-
flow and oxidation stability. The diesel-like HCs consisting of 15–18
carbon atoms (produced by the hydroprocessing route) provide
good fuel properties [10]. Unlike FAMEs, the HC-based biofuels have
the advantage of using them as they are or as a blend with the
conventional diesel in any proportion.
currently derived from fossil resources (petroleum, natural gas and
coal). Usage of fossil fuels has been realized to cause adverse effects
to the environment [1,2]. Moreover, at the projected rate of con-
sumption, the availability of petroleum and natural gas is expected
to run out in about 70 years and coal in 200 more years. Biofuels
show the way to sustainable environment, energy independence,
employment to rural population and savings in oil import bill.
Vegetable oils such as soybean, rapeseed and palm oils comprise
triglycerides (TGs) which can be converted into diesel-like fuels
Two types of catalyst systems were reported for deoxygenation
of vegetable oils: (i) Mo and W promoted hydrotreating cata-
[
3–5]. Use of non-edible oils instead of edible ones makes the bio-
lysts (Ni-Mo/Al O , Ni-W/Al O , Co-Mo/Al O and Co-W/Al O )
2 3 2 3 2 3 2 3
fuels production more attractive as the former are cheaper and do
not lead to issues like food versus fuel controversy. Non-edible oils
contain a significant amount of free fatty acids (FAs) in their com-
position along with glycerides. One option exists in the form of
hydroprocessing for removal of oxygen from the constituent TGs
and FAs toward producing diesel-range hydrocarbons (HCs) [6–8].
Fatty acid methyl esters (FAMEs) represent the first generation
biodiesel formed via methanolysis of vegetable oils. Unfortunately,
[9,11–14] and (ii) supported noble metal catalysts [15–18]. While
the mode of deoxygenation is mainly through dehydration (H O
2
removal) over the former type catalysts, it is through decar-
bonylation/decarboxylation (CO/CO2 removal) on the latter type
catalysts. Relatively lesser amount of hydrogen is needed while
using the noble metal catalysts (5–20 bar) as against Ni-Mo/W cata-
lysts (50–70 bar). After screening several supported metal catalysts,
Snåre et al. [19] found that carbon-supported Pd converts stearic
acid completely with >98% selectivity toward heptadecane (C17).
The efficiency of deoxygenation of different metals decreased in the
order: Pd > Pt > Ni > Rh > Ir > Ru > Os. They found that decarboxyla-
tion was profound over Pd/C, while decarbonylation was prevalent
∗ Corresponding author.
E-mail address: d.srinivas@ncl.res.in (S. Darbha).
http://dx.doi.org/10.1016/j.molcata.2016.03.023
1
381-1169/© 2016 Elsevier B.V. All rights reserved.

R. Raut et al. / Journal of Molecular Catalysis A: Chemical 417 (2016) 126–134
127
◦
over a Pt/C catalyst. Alkaline Pd/C led to highest initial rates and
conversion, but formed a large quantity of aromatics which was
a drawback [20]. Catalytic activity of mesoporous Pd/C in a con-
tinuous flow reactor was studied [21]. Conversion of stearic acid
declined during a 92 h time-on-stream investigation which was
attributed for coking of the catalyst surface. Noble metals sup-
ported on zeolites and mesoporous silica were also investigated
as catalysts for this reaction. Pt-Re/H-ZSM-5 enabled conversion
of jatropha oil (to C15–C18 HCs) in yields of 67% at standard
hydrotreating conditions [22]. Ahmadi et al. [23] reported oleic acid
decarboxylation over Pt supported on small pore zeolites (SAPO-34,
DNL-6 and RHO) and hydrotalcites yielding paraffinic, branched
and aromatic HCs. They found that catalyst acidity, Pt dispersion
and pore diameter of the support influence the product selectivity.
Ping et al. [24] studied meso-cellular foam silica (MCF)-supported
Pd catalyst in hydrogen-free decarboxylation of FAs. The cata-
lyst showed deactivation. Organic deposition on the catalyst was
reported as the cause for its deactivation. Lestari et al. [25] and Lee
and Ramli [26] investigated the use of Pd/SBA-15 for converting
stearic acid and methyl oleate into HCs. Pd supported on SBA-15
having necklace-like morphology and high surface area exhibited
high catalytic activity.
We report here, for the first time, the application of three-
dimensional, mesoporous silica (SBA-12 and SBA-16)-supported Pd
as catalysts for deoxygenation of FAs (unsaturated oleic acid—OA
and saturated stearic acid—SA, as representative examples of
FAs) and non-edible jatropha oil. Three-dimensional mesoporos-
ity enables enhanced rate of transport of reactant and product
molecules and accessibility of active sites. SBA-12 and SBA-16 differ
in their structure, acidity and surface properties which could lend
differences in the behavior of the supported Pd. The present study
probes these effects on the catalytic deoxygenation activity.
gray colored solid formed was collected, calcined at 400 C (heating
◦
rate = 4 C/min) for 2 h. Just prior to reactions and characterization
studies, Pd/SBA-12, thus prepared, was reduced at 250 C for 2 h
◦
under a flow of hydrogen (30 ml/min).
Pluronic F127 block-copolymer (7.4 g) was dissolved in 384.3 g
of 2 M HCl solution at 40 C while stirring for 2 h. To it, 28.34 g of
TEOS was added drop-wise over 30 min. The stirring was continued
for 24 h. The gel formed was transferred into a Teflon-lined stainless
steel autoclave (80 ml). It was crystallised at 80 C for 48 h. The solid
(as-synthesized SBA-16) formed was filtered out, washed with dis-
tilled water (2–3 l), dried at 100 C overnight and calcined at 550 C
(heating rate = 4 C/min) for 8 h. The calcined SBA-16 [27,28] was
◦
◦
◦
◦
◦
white in color. Yield: 96%. Pd was impregnated on SBA-16 as per the
procedure reported above for Pd/SBA-12 wherein, SBA-16 instead
of SBA-12 was taken.
2.2. Characterization techniques
X-ray powder diffraction (XRD) patterns were recorded on a
X’Pert Pro Philips diffractometer equipped with Cu K␣ radiation and
a proportional counter detector. The measurements in the wide-
◦
◦
angle region (2␪ = 10–85 ) were made at a scan rate of 4 /min and
◦
◦
in the small-angle region (2␪ = 0.5–5 ) at 1 /min. High resolution
transmission electron microscopic (HRTEM) images of the catalysts
were recorded on a FEI Technai-F30 instrument fitted with a 300 kV
field emission gun. Samples were prepared by dispersing the cat-
alysts in iso-isopropanol, placing them on copper grids and drying
◦
the grids overnight at 25 C. Specific surface area of the catalysts
was determined from nitrogen adsorption-desorption isotherms
◦
recorded at −196 C using a NOVA 1200 Quanta Chrome equip-
ment. The BET method was adopted in estimation. The micropore
volume was determined from the t-plot and average pore diameter
was estimated following the Barret-Joyner-Halenda (BJH) model.
CO-chemisorption studies were conducted on a Quantachrome
Autosorb iQ instrument. In typical experiment, about 0.2 g of the
catalyst was taken in U-shaped quartz tube. It was first out-
2
. Experimental
◦
◦
2.1. Catalyst preparation
gassed at 200 C (heating rate = 20 C/min) under a flow of He
20 ml/min) for 90 min and reduced with H2 (20 ml/min) at 300 C
(heating rate = 20 C/min) for 120 min followed by evacuating the
sample at 300 C for 120 min. Temperature of the sample was low-
◦
(
◦
Brij-76 [C18H37(OCH CH ) OH, average mol. wt. ∼711, Aldrich
2
2 10
◦
Co.] and Pluronic F127 block-copolymer (EO1 PO EO , average
06
70
106
◦
mol. wt. ∼12600, Aldrich Co.) were used as structure direct-
ing agents in the synthesis of SBA-12 and SBA-16, respectively.
Tetraethylorthosilicate (TEOS, 98%, Aldrich Co.) was used as a Si
ered down to 40 C (in 1 min) under vacuum. CO was introduced
and chemisorption measurements were started (equilibrium toler-
ance = 0, equilibrium time = 3 min, adsorption points = 80, 160, 240,
◦
source. Tetraamminepalladium(II) nitrate (10 wt% in H O solution,
320, 400, 480 and 560 mmHg, analysis temperature = 40 C, thermal
2
Aldrich Co.) was used as a Pd source. All the solvents (A.R. grade)
were procured from Thomas Baker and the reagents were pur-
chased from Aldrich Co.
equilibrium time = 10 min, leak test = 1 min). Specific surface area,
percentage dispersion and average particle size of Pd were deter-
mined assuming the stoichiometric factor (S = number of moles of
f
SBA-12 was prepared by a modified procedure of Zhao et al. [27],
wherein, 8 g of Brij-76 was added to a solution of 40 g of distilled
water and 160 g of 0.1 M HCl taken in a polypropylene beaker. It
was heated to 40 C and stirred for 2 h till a homogeneous solution
was obtained. To it, 17.6 g of TEOS was added over 30 min. Stirring
was continued for another 20 h. The gel formed was transferred
CO per surface Pd) as unity. In the average particle size determi-
nation, the particle shape factor (F) was considered as 6 assuming
spherical or near spherical shaped particles. The following equa-
tions were used in determining the metal specific surface area (MS;
m /g), percentage metal dispersion (D) and average metal particle
size (d; in units of nanometers).
◦
2
into a Teflon-lined stainless steel autoclave (80 ml) and heated at
Vm × Na
MS =
◦
1
00 C for 24 h while placing in an electric oven. The solid (as-
S × S × L
f
d
synthesized SBA-12) formed was filtered, washed thoroughly with
◦
distilled water (2–3 l) and dried at 100 C for 12 h. It was calcined
◦
◦
4
at 550 C (heating rate = 4 C/min) for 8 h to remove all the organic
matter in it. The calcined SBA-12 was white in color. Yield: 94%.
Palladium was deposited on it by wet impregnation method. In a
typical procedure, the calcined SBA-12 (2 g) was suspended in dis-
tilled water (15 ml) taken in a glass round-bottom flask. A known
quantity of the palladium source (0.5611 g for 1 wt% Pd loading,
for example) dissolved in 5 ml of water was added drop-wise to
Vm × AW×10
D =
d =
%W × S
f
3
F×10
MS × Dm
Here, Vm is the CO-gas adsorbed at monolayer coverage (moles/g
of catalyst), Na is Avagadro number, Sd is metal surface density
(i.e., number of metal atoms per square meter of Pd which is
the above suspension. It was stirred over a rotary evaporator at
◦
6
0 C for 3 h. Then, water was evaporated by applying vacuum. The

1
28
R. Raut et al. / Journal of Molecular Catalysis A: Chemical 417 (2016) 126–134
0
.127 × 1020), AW is atomic weight of Pd (106.42), L is grams of
comparing with standard samples. Products were analysed also by
H NMR (Bruker 200 MHz) and FTIR (Shimadzu) spectroscopies.
1
Pd per gram of catalyst, %W is weight percent of Pd in the catalyst
(
which is equal to 100L) and Dm is density of Pd i.e., grams of Pd
3
◦
per unit volume (which is equal to 11.9 g/cm at 20 C).
3. Results and discussion
2.3. Reaction procedure
3.1. Catalyst characterization
Catalytic decarboxylation reactions are carried out in a stain-
◦
SBA-12 and Pd(1 wt%)/SBA-12 showed an intense peak at 1.68
less steel (Parr 300 ml) batch reactor equipped with a temperature
controller. To quench the reaction immediately, the reactor ves-
sel was fitted with an internal loop line cooler. Just prior to the
◦
◦
arising from (002) and two weak peaks at 2.77 and 3.0 originating
from (112) and (300) planes, respectively (Fig. 1). A weak shoulder
appeared at 1.25 due to (100) reflection. These characteristics of
SBA-12 were indexed to an ordered, three-dimensional, hexagonal
mesoporous structure with a space group of p6 /mmc [27]. Except
for some reduction in peak intensity, there was little change in the
XRD peaks pattern and position after the deposition of Pd. In the
wide-angle region, SBA-12 and Pd(1 wt%)/SBA-12 showed a broad
peak at 22.2 arising from the amorphous silica in the pore wall
structure. Two additional less intensity sharp peaks were observed
◦
◦
◦
reaction, catalyst was reduced at 250 C (heating rate = 4 C/min)
for 2 h in a flow of hydrogen (20 ml/min). In a typical reaction, a
known quantity of the substrate [oleic acid (OA), 99%; Loba Chemie;
3
2
(
g], solvent (decane, 99%; Spectrochem; 30 g) and reduced catalyst
10 wt% of the substrate) were taken in the reactor. The reactor was
flushed thoroughly with hydrogen gas to remove the traces of air
present in the vessel and then filled with hydrogen to a desired
pressure (5–30 bar). The temperature of the reactor was raised to a
◦
◦
◦
◦
◦
at 40.2 and 46.6 for Pd(1 wt%)/SBA-12. These confirm the presence
of metallic Pd, crystallised in a cubic closed-packed structure with
a space group of Fm3m (JCPDS No. 05-0681). These reflections are
desired value (300–325 C) while heating at a rate of 4 C/min and
the reaction was conducted for 5 h. After the reaction, the rector was
depressurized. The contents of the reactor were subjected to cen-
trifugation. Catalyst particles were separated. Solvent was removed
by rotary evaporator and the product was isolated. Deoxygenation
reactions were conducted also with stearic acid (SA) and non-edible
jatropha oil (JO). Fatty acid composition of jatropha oil is as fol-
lows: palmitic acid (C16.0) = 15.6%, palmitoleic acid (C16.1) = 1.0%,
stearic acid (C18.0) = 5.8%, oleic acid (C18.1) = 40.1% and linoleic acid
¯
0
originated from (111) and (200) planes of Pd , respectively. The
above stated weak peaks in low and high angle regions were broad
and diffused.
◦
SBA-16 and Pd(1 wt%)/SBA-16 showed a strong peak at 0.9
◦
◦
attributable to (110) and weak peaks at 1.21 and 1.9 due to (200)
and (211) reflections, respectively (Fig. 1). These distinct peaks of
SBA-16 correspond to an ordered, three-dimensional, cubic, meso-
(
C18.2) = 37.6%.
¯
porous structure with a space group of Im3m [28,29]. The chosen
mesoporous silica supports (SBA-12 and SBA-16) differ in their
pore structural arrangement. As is the case with SBA-12, metal
deposition did not alter the framework structure of SBA-16. In the
wide-angle region, Pd(1 wt%)/SBA-16 showed peaks due to metal-
lic Pd at 40.3 , 46.9 and 68.6 arising from (111), (200) and (220)
planes, respectively.
2
.4. Product analysis
For determination of acid values of reactants and products, 1 g of
the sample was weighed and dissolved in a solvent mixture (10 ml)
containing equal parts (v/v) of diethyl ether and ethanol; 3–5 drops
of phenolphthalein indicator (1%, Merck, India) was added. Then,
it was titrated with 0.1 N standardized sodium hydroxide solution
until the color of the solution changed to light pink. Acid value of
the sample was calculated using the formula:
◦
◦
◦
The presence of well-ordered mesopores in SBA-12 and SBA-
1
6 materials was further confirmed by N -physisorption studies
2
(Fig. 2). SBA-12 showed a type-IV adsorption isotherm with a
hysteresis loop which could be assigned to H -type indicating
1
VNaOH × N
× 56.1
ordered mesopores in the material. Adsorption isotherm of SBA-
16 also showed a type-IV isotherm, however, with a H2-type
hysteresis loop confirming ordered mesoporous architecture. The
NaOH
Acid Value =
W
sample
here, 56.1 is the molecular weight of KOH as acid value is always
reported as mg of KOH per g of sample. NNaOH = normality of NaOH,
VNaOH = volume of NaOH consumed and WSample = weight of the
sample taken for titration. Fatty acid (FA) conversion was deter-
mined from the acid values using the following formula:
H -type hysteresis loop indicates an ink bottle-neck shaped pores
2
in the material [30]. No change in shape and hysteresis of the
isotherms was observed after Pd deposition (Fig. 2). The physisorp-
tion isotherms of SBA-16 materials (Fig. 2) confirm the presence
of cubic cage-like interconnected mesopore structure. All these
materials showed narrow pore size distribution curves. Metal
impregnation did not alter the shapes of hysteresis loops of SBA-
FA conversion (%)
= (
Acid value of FA − Acid value of product sample)
× 100
1
2 and SBA-16. Textural properties of these catalysts are listed
Acid value of FA
in Table 1. The SBA-12 materials had higher specific surface area
(
SBET) than the SBA-16 materials. On loading Pd, the SBET of SBA-
2
In order to analyze by gas chromatographic (GC) technique,
the product was derivatized with N,O-bis(trimethylsilyl) triflu-
oroacetamide (BSTFA). Known quantity of the product (0.015 g)
was dissolved in pyridine (0.01 g) and 100 wt% excess of BSTFA
12 increased from 708 to 804 m /g while it decreased from 574 to
489 m /g for SBA-16. Pore volume and average pore size remained
2
nearly the same. This difference may be due to variation in Pd dis-
persion and metal-support contacts.
◦
was added to it. The mixture was then heated to 70 C
HRTEM images confirmed the long-range, three-dimensional,
mesoporous ordering of these materials (Fig. 3). While
Pd(1 wt%)/SBA-12 has hexagonally arranged pores, Pd(1 wt%)/SBA-
16 has cubic, cage-like interconnected pore arrangement. Average
particle size of Pd determined from the particle size distribution
curve (Fig. 3) pointed out that the Pd particles on SBA-12 were
larger in dimension (4.5 nm) than those on SBA-16 (2.5 nm).
SBA-12 has higher specific surface area than SBA-16. One expects
higher metal dispersion (lower particle size for Pd) on SBA-12 than
for 45 min and immediately injected in GC (Varian 450 GC;
TM
Select Biodiesel column of dimensions: 15 m-length × 0.32 mm-
diameter × 0.10 ␮m-film thickness). Helium was the cariar gas. In
the GC method, injector and detector temperatures were set at
◦
3
00 C. A split ratio of 1:50 was used. The following tempera-
◦
◦
◦
ture program was used: 80 C (start) – 280 C (at 20 C/min ramp
rate) – hold (10 min). Products were identified by GC–MS (Agilent
Technology 7890 B; HP-5 column: 30 m-long, 0.25 mm-i.d.) and by

R. Raut et al. / Journal of Molecular Catalysis A: Chemical 417 (2016) 126–134
129
Fig. 1. XRD patterns of SBA-12, Pd(1 wt%)/SBA-12, SBA-16 and Pd(1 wt%)/SBA-16 in small- (top traces) and wide-angle (bottom traces) regions. Peaks due to different planes
are marked.
Table 1
Textural properties of “bare” and Pd(1 wt%) supported on SBA-12 and SBA-16.
SBET (m /g)a
2
Pore volume (cc/g)b
Average pore size (nm)b
Average Pd particle size (nm)c
Catalyst
SBA-12
Pd(1 wt%)/SBA-12
SBA-16
708
804
575
489
1.0
1.1
0.6
0.6
2.3
2.0
1.7
1.7
–
4.5
–
Pd(1 wt%)/SBA-16
2.5
a
Calculated from N2-physisorption (BET method).
Calculated by BJH pore method.
From HRTEM analysis.
b
c
on SBA-16. However, the reverse behavior points out the impor-
tance of surface structure (pore arrangement, hydrophobicity and
acidity) on controlling the metal particle size.
cross-polarization magic-angle spinning nuclear magnetic reso-
nance (29Si MAS NMR) showed signals at around −91, −101, and
−110 ppm attributed to three chemically and magnetically inequiv-
2
3
4
CO-chemisorption studies revealed that surface area, metal dis-
persion and average particle size of Pd on SBA-12 and SBA-16 are
nearly the same within the experimental error. Among the cata-
lysts investigated, Pd(2 wt%)/SBA-16 had highest surface area and
percentage metal dispersion and lowest Pd particle size (Table 2).
The surface hydrophobicity of SBA-12 and SBA-16 was quan-
tified by thermogravimetric (TG) analysis. Calcined samples of
alent tetra-coordinated Si species (Q , Q and Q ) having molecular
formulae: Si(OSi) (OH) , Si(OSi) (OH) and Si(OSi) , respectively
2
2
3
4
[31]. SBA-12 showed a large amount of Q and Q3 species while
2
4
SBA-16 had an excess of Q species (Fig. 4 (right panel)). Higher
amount of Q4 species in case of SBA-16 materials is an indication
of their surface hydrophobicity. This conclusion agrees with the
observation made in water adsorption (TG) studies.
◦
SBA-12 and SBA-16 were exposed to water vapor at 25 C and
Total amount of acidic sites in SBA-12 and SBA-16 materials
were quantified by temperature-programmed desorption of NH3.
These studies revealed that the supports were mildly acidic with
the acid site density being 30 and 40 ␮mol/g for SBA-12 and SBA-16,
respectively [31]. Thus the nature of the support has an influence
on its surface structure and metal distribution.
the amount of water adsorbed was analysed. Fig. 4 (left panel)
shows the TG analysis of “bare” SBA-12 and SBA-16. SBA-16 showed
◦
water loss of 5 wt% below 127 C. This was almost double for
SBA-12 (10 wt%) [31]. The percentage loss of water normalised to
2
surface area also followed the same trend (0.014 g/m for SBA-
2
1
2 and 0.009 g/m for SBA-16). Thus, this experiment reveals
29
that SBA-16 is relatively more hydrophobic than SBA-12.
Si

1
30
R. Raut et al. / Journal of Molecular Catalysis A: Chemical 417 (2016) 126–134
4
50
7
6
5
4
3
2
1
00
00
00
00
00
00
00
0
SBA-16
SBA-12
400
3
3
2
2
50
00
50
00
150
00
1
5
0
0
.0
0.2
0.4
0.6
0.8
1.0
0
.0
0.2
0.4
0.6
0.8
1.0
Relative Pressure, P/P
0
Relative Pressure P/P
0
4
00
8
7
6
5
4
3
2
1
00
00
00
00
00
00
00
00
0
Pd(1 wt%)/SBA-12
3
50
Pd(1wt%)/SBA-16
3
00
50
200
50
00
2
1
1
5
0
0
.0
0.2
0.4
0.6
0.8
1.0
0
.0
0.2
0.4
0.6
0.8
1.0
Relative Pressure, P/P
0
Relative Pressure, P/P
0
Fig. 2. Nitrogen adsorption-desorption isotherms of SBA-12, Pd(1 wt%)/SBA-12, SBA-16 and Pd(1 wt%)/SBA-16.
Fig. 3. HRTEM images and Pd particle size distribution of Pd(1 wt%)/SBA-12 and Pd(1 wt%)/SBA-16.

R. Raut et al. / Journal of Molecular Catalysis A: Chemical 417 (2016) 126–134
131
Table 2
CO-chemisorption data of supported Pd catalysts.
2
Catalyst
Monolayer uptake (␮mol/g)
Active metal surface area (m /g of metal)
Metal dispersion (%)
Pd particle size (nm)
Pd(1 wt%)/SBA-12
Pd(1 wt%)/SBA-16
Pd(2 wt%)/SBA-16
Pd(3 wt%)/SBA-16
Pd(5 wt%)/SBA-16
10.13
8.90
27.68
30.02
43.52
48.0
42.2
65.6
47.5
41.3
10.8
9.4
14.7
10.6
9.2
10.1
8.9
27.7
30.0
43.5
Fig. 4. Thermogravimetric analysis (left) and 29Si MAS NMR of SBA-12 and SBA-16.
xH2
CO2
H
C
8
H
17
C
C
H
7 14
C H COOH
C17H35COOH
Stearic Acid
xH2
C17H36
Heptadecane
Oleic acid
xH2
CO
H O
2
H O
2
C H37OH
C H35CHO
17
Octadecanol Octadecanal
1
8
C H
38
1
8
Octadecane
H O
xH2
2
Scheme 1. Possible reaction pathway for deoxygenation of oleic acid.
3
.2. Catalytic activity
90
Pd(1 wt%)/SBA-16
Pd(1 wt%)/SBA-12
Scheme 1 provides possible pathways for deoxygenation of FA.
Oleic acid (OA), a representative FA, has two functional groups
80
70
60
(
C
C and COOH) for reaction with hydrogen. Hydrogenation of
C (resulting in stearic acid, SA) happens prior to the deoxy-
C
genation reaction. Decarboxylation results in heptadecane (C17
)
and hydrodeoxygenation yields octadecane (C18). Octadecanol and
octadecanal are the intermediates in the deoxygenation process. In
reactions over Pd supported on SBA-12 and SBA-16, hydrocarbons
50
40
30
(
C17 and to a small extent C18) are the major products and alcohols
and aldehydes (others) are formed in small amount. Commercial
grade OA and SA contained a small amount of palmitic acid (PA)
impurity and hence, the deoxygenated product had some portion
of C15 and C16 alkanes.
300
305
310
315
320
325
o
Temperature ( C)
Temperature has a marked effect on the deoxygenation reac-
tion. OA conversion (Fig. 5) and turnover frequency (TOF = moles
of OA converted per mole of surface exposed Pd per hour;
Table 3) increased linearly with temperature. Pd(1 wt%)/SBA-16
exhibited higher catalytic activity (OA converison and TOF) than
Fig. 5. Catalytic deoxygenation activity of Pd(1 wt%)/SBA-12 and Pd(1 wt%)/SBA-16.
Reaction conditions: Oleic acid (OA) = 2 g, n-decane (solvent) = 30 g, catalyst = 0.2 g,
H2 pressure = 10 bar, reaction time = 5 h.

1
32
R. Raut et al. / Journal of Molecular Catalysis A: Chemical 417 (2016) 126–134
Table 3
a
Kinetic parameters in decarboxylation of oleic acid over Pd(1 wt%)/SBA-12 and Pd(1 wt%)/SBA-16 .
◦
−1
2
−1
−1
)
Temperature ( C)
OA Conv. (mol%)
TOF (h
)
k × 10 (L s molPd
Ea (kJ/mol)
130.1
Pd (1 wt%)/SBA-12
300
310
320
325
37.2
53.5
62.3
80.0
260
373
435
559
5.9
9.8
12.5
20.6
Pd (1 wt%)/SBA-16
300
310
320
325
44.3
62.3
77.1
83.5
352
495
613
664
7.5
12.4
18.8
23.0
127.1
a
Reaction conditions: Oleic acid (OA) = 2 g, n-decane = 30 g, catalyst = 0.2 g, pH2 = 10 bar, reaction time = 5 h.
◦
Pd(1 wt%)/SBA-12. At 300 C, the difference in OA conversion for
Pd(1 wt%) supported on SBA-16 and SBA-12 was 7 mol%. Single
point rate constants at different temperatures for both the cata-
lysts were calculated using the following formula and assuming a
pseudo-first order reaction,
systems. Catalytic activity of the present catalysts is comparable to
that of Pd/SBA-15 [25,26].
At equal conversion of OA (62.3 mol%), the selectivity of
Pd(1 wt%)/SBA-12 and Pd(1 wt%)/SBA-16 towards C17 (92.7 and
93.9%, respectively) and C15 (3.0 and 4.7%, respectively) HC prod-
ucts was nearly the same. This points out that particle size of Pd
has little effect on the product selectivity. Temperature also had
a negligible effect on C17 selectivity over Pd(1 wt%)/SBA-12 and
Pd(1 wt%)/SBA-16 catalysts.
−
V × ln (1 − X)
k =
W × t
where, k is the pseudo-first order rate constant (L s−1 molPd ), X is
OA conversion (in moles), V is volume of the reaction mixture (in
litres), t is reaction time (in sec) and W is moles of Pd in the catalyst
taken for reaction. Table 3 lists OA conversion, TOF and reaction rate
constant at different temperatures (T) for the supported Pd(1 wt%)
catalysts. The Arrhenius equation (given below) was employed to
estimate the activation energy (Ea) and pre-exponential factor (A)
for the decarboxylation reaction.
−1
Table 4 presents the influence of hydrogen pressure and metal
◦
content on decarboxylation of OA over Pd/SBA-16 at 325 C. OA
conversion versus hydrogen pressure showed a parabolic type vari-
ation (Fig. 6). OA conversion increased with increasing hydrogen
pressure up to 10 bar and then onwards showed a decreasing trend.
Selectivity towards the hydrodeoxygenation (HDO) products (C18
and C16) showed increasing trend with H2 pressure. The selectiv-
ity for C17 decreased from 94.2% at 5 bar H2 to 89.8% at 30 bar H2.
The possible reason for the decrease in conversion of OA at very
high H2 pressure (>10 bar) is the competition between hydrogen
and OA for adsorption on the active site. Water is the byproduct
in HDO process which could be a detrimental parameter for cat-
alytic activity. An optimum pressure of 10 bar H2 is ideal in terms
of conversion and selectivity of the catalyst for the decarboxylation
of OA to diesel range hydrocarbons. Decarboxylation took place at
our reaction conditions even in the absence of external supply of
hydrogen. OA conversion of 18.6% and C17 selectivity of 65.2 mol%
were detected (Table 4). However, presence of hydrogen facilitated
OA conversion and C17 selectivity.
Pd loading on SBA-16 has a positive effect on OA conversion.
OA conversion increased from 83.5 mol% (at 1 wt% Pd loading) to
99.8 mol% (at 3 wt% Pd loading). Further increase in Pd loading (to
5 wt%) has no positive effect on OA conversion (Table 4). Pd loading
has no major influence on the selectivity of the reaction.
Internal mass transfer is responsible for transport of reactant
from entrance of catalyst pore pellet into the pore of catalyst pellet.
If the concentration of reactant inside the pellet pore is lower than
at the entrance of catalyst pore then reaction is limited by internal
mass transfer. The effect of internal mass transfer was determined
using Weisz-Prater criterion. According to this, the overall reac-
tion kinetics is free from internal mass transfer limitation if the
Ea
ln k =
+ ln A
RT
Here, R is the gas constant (8.31 J/K mol) and T is the reac-
tion temperature in Kelvin. From the plot of ln k versus 1/T,
the values of Ea and A were determined. Ea was found to be
30.1 and 127.1 kJ/mol for Pd(1 wt%)/SBA-12 and Pd(1 wt%)/SBA-
6, respectively. The value of normalised pre-exponential factor
1
1
(
4
A) for Pd(1 wt%)/SBA-12 and Pd(1 wt%)/SBA-16 was found to be
1
0
10 −1
s , respectively. As noted from the
.36 × 10 and 2.98 × 10
characterization studies SBA-16 is relatively more hydrophobic
than SBA-12 and hence, adsorption of OA (hydrophobic molecule)
is expected to be higher on the former than on the latter. Further,
the average particle size of Pd on SBA-16 is lower (2.5 nm) than on
SBA-12 (4.5 nm; HRTEM). Thus the exposed Pd atoms are higher
in the case of the former than in latter. Both these factor (surface
hydrophobicity and smaller Pd particles) are responsible for the
higher catalytic activity of Pd(1 wt%)/SBA-16 than Pd(1 wt%)/SBA-
1
2. Control experiments revealed that this reaction doesn’t occur
to a significant extent (under similar conditions) in the absence
of a catalyst (OA conversion without catalyst was 2.4 mol% only).
Also SBA-12 and SBA-16 supports do not catalyze the deoxygena-
tion reaction. Popov and Kumar [32] estimated activation energy
of 120 ± 5 kJ/mol for hydrothermal deoxygenation of OA over acti-
vated carbon (AC) in a continuous flow process. Fu et al. [33]
reported Ea of 79 and 125 kJ/mol for hydrothermal decarboxylation
of palmitic acid (PA) over AC-1 and 5% Pt/C catalysts, respec-
tively. Snåre et al. [34] studied the decarboxylation of ethyl stearate
over a Pd/C catalyst in semi-batch reactor and found Ea equals to
dimensionless Weisz number (˚ ) is lower than 1, otherwise there
I
would be influence of internal mass transport on the overall reac-
tion kinetics [36,37].
robsꢁpRp2
ꢀI =
5
7.3 kJ/mol. While producing aviation fuel through decarboxyla-
DEACAS
tion of fatty acids in microalgae oil, Yang et al. [35] determined a
value of 114 kJ/mol for the activation energy. The Ea values for the
decarboxylation of OA over Pd(1 wt%)/SBA-12 and Pd(1 wt%)/SBA-
In the above equation, r
is the observed higher reaction rate
obs
−
6
(1.1099 × 10 kg/kg s), ␳p is the density of the catalyst pel-
3
1
6 catalysts fall within the range reported for the known catalyst
let (1200 kg/m ), Rp is the radius of catalyst particle (1 ␮m),

R. Raut et al. / Journal of Molecular Catalysis A: Chemical 417 (2016) 126–134
133
Table 4
Decarboxylation of oleic acid (OA) over Pd(1 wt%)/SBA-16: Influence of reaction parameters.
Effect of reaction temperaturea
◦
OA conversion (mol%)a
Product selectivity (%)b
Reaction temperature ( C)
TOF
C18
C17
C16
C15
Others
300
310
320
325
44.3
62.3
77.1
83.5
352
495
613
664
1.2
0.5
1.0
1.1
94.4
93.9
93.5
93.3
0.2
0.2
0.2
0.1
3.5
4.7
4.5
4.7
0.7
0.6
0.8
0.7
Effect of H2 pressureb
H2 pressure (bar)
OA conversion (mol%)b
TOF
C18
C17
C16
C15
Others
0
5
0
0
0
18.6
54.7
83.5
81.0
75.8
150
435
664
644
603
0
65.2
94.2
93.3
91.1
89.8
0
0.1
4.2
4.7
4.4
4.6
34.7
1.0
0.7
0.8
1.1
0.5
1.1
3.4
4.3
0.1
0.1
0.3
0.3
1
2
3
Effect of Pd content in the catalystc
Pd content (wt%)
OA conversion (mol%)c
TOF
C18
C17
C16
C15
Others
0
1
2
3
5
2.4
83.5
86.3
99.8
99.6
–
664
221
235
162
0
40.8
93.3
92.9
94.0
93.8
0
0
59.3
0.7
0.8
0.7
0.7
1.1
1.2
0.8
1.3
0.1
0.1
0.1
0.1
4.7
4.9
4.4
4.1
a
Reaction conditions: OA = 2 g, n-decane (solvent) = 30 g, catalyst = 0.2 g, H2 pressure = 10 bar, reaction time = 5 h.
Reaction conditions: OA = 2 g, n-decane = 30 g, catalyst = 0.2 g, reaction temperature = 325 C, reaction time = 5 h.
Reaction conditions: OA = 2 g, n-decane = 30 g, catalyst = 0.2 g, H2 pressure = 10 bar, reaction temperature = 325 c, reaction time = 5 h. Others include octadecanol and
b
c
◦
◦
octadecanal.
1
00
85
80
75
70
65
60
55
50
9
5
90
8
8
7
5
0
5
C18 + C16
C17 + C15
Others
1
0
5
0
5
10
15
20
25
30
5
10
15
20
25
30
H pressure (bar)
2
H Pressure
2
◦
Fig. 6. Influence of hydrogen pressure on OA conversion and product selectivity. Reaction conditions: OA = 2 g, n-decane = 30 g, catalyst = 0.2 g, reaction temperature = 325 C,
reaction time = 5 h.
Table 5
Deoxygenation reaction over Pd(3 wt%)/SBA-16: Influence of feedstock.
Feedstock
Lipid conversion (mol%)
Product selectivity (%)
C18
C17
C16
C15
Others
OA
SA
JO
99.8
92.3
97.4
0.8
13.7
0.9
94.0
80.2
81.6
0.1
0.9
0.6
4.4
4.2
16.8
0.7
1.0
0.6
◦
Reaction conditions: Fatty compound = 2 g, n-decane = 30 g, catalyst = 0.2 g, H2 pressure = 10 bar, reaction temperature = 325 C, reaction time = 5 h. Others include octadecanol
and octadecanal. OA = oleic acid, SA = stearic acid and JO = jatropha oil.
−
10
2
DEA is the molecular diffusion coefficient (1 × 10
m /s) [38]
bulk gas phase to solid-liquid interface, (2) diffusion of hydrogen
gas into the liquid bulk phase consisting of OA and n-decane, and (3)
transfer of dissolved hydrogen and OA onto the external surface of
catalyst surface [39,40]. We have checked the possibility of external
and C
is the concentration of reactant at catalyst surface
AS
3
(
46.189 kg/m ).
External mass transfer involves transport of hydrogen gas from
bulk gas phase to external surface of the catalyst. This transport
occurs through three main steps: (1) transport of hydrogen gas from

1
34
R. Raut et al. / Journal of Molecular Catalysis A: Chemical 417 (2016) 126–134
mass transfer limitation using Mear’s criterion. Accordingly to this
criterion,
Acknowledgement
This work forms a part of the Network project “Catalysts for
Sustainable Energy (ECat; CSC 0117).”
robsnꢁpRp
ꢀE =
.
kSC
AB
References
Thiele modulus (ꢀE) should be less than 0.15 for the reaction to
be free from external mass transfer limitation. Here, n is the order
of reaction (one in the present case), kS is the external (liquid to
[1] https://www.worldenergy.org/wp-content/uploads/2013/09/Complete WER
2013 Survey.pdf.
[
2] World Energy Outlook Special Report Energy and Climate Change,
International Energy Agency, France (2015).
−4
solid) mass transfer coefficient (1.051 × 10 m/s) which can be
calculated using the formula [41]:
[3] B. Smith, C.G. Greenwell, A. Whiting, Energy Environ. Sci. 2 (2009) 262–271.
[
[
4] P.M. Mortensen, J.-D. Grunwaldt, P.A. Jensen, K.G. Knudsen, A.D. Jensen, Appl.
Catal. A: Gen. 407 (2011) 1–19.
5] J.K. Satyarthi, T. Chiranjeevi, G.D.T. Shastry, P.S. Viswanathan, PTQ Q2 (2013)
Nsh = 2 + 0.36(NRe)3/4(N )1/3
Sc
1–5.
ksdp
DEA
[6] S. Lestari, P. Mäki-Arvela, I. Simakova, J. Beltramini, G.Q.M. Lu, D.Y. Murzin,
Catal. Lett. 130 (2009) 48–51.
[7] T.V. Choudhary, C.B. Phillips, Appl. Catal. A: Gen. 397 (2011) 1–12.
[8] E. Santillan-Jimenez, M. Cooker, J. Chem. Technol. Biotechnol. 87 (2012)
where, Nsh is the Sherwood number which is defined as
wherein dp is the particle diameter (2 ␮m), NRe is the Reynolds
4
3
number which is defined as e¯ dp /(ꢂL) ; e¯ is the average specific
1041–1050.
power input per kg liquid (W/kg, assumed in the present case as 1)
[
9] T. Morgan, E. Santillan-Jimenez, A.E. Harman-Ware, Y. Ji, D. Grubb, M. Crocker,
−
5
2
and ꢂL is the kinematic viscosity of liquid phase (4.346 × 10 m /s),
Chem. Eng. J. 189–190 (2012) 346–355.
N
Sc
is the Schmidt number which is defined as ␮/␳DEA; ␮ is the
[10] M. Snåre, I. Kubi cˇ ková, P. Mäki-Arvela, K. Eränen, D.Y. Murzin, Fuel 87 (2008)
33–945.
11] J.G. Immer, M.J. Kelly, H.H. Lamb, Appl. Catal. A: Gen. 375 (2010) 134–139.
9
viscosity of liquid phase and ␳ is density of liquid phase. CAS is
the surface concentration and CAB is the bulk concentration of OA.
[
[12] T. Morgan, D. Grubb, E. Santillan-Jimenez, M. Crocker, Topics Catal. 53 (2010)
820–829.
[13] T. Kalnes, T. Marker, D.R. Shonnard, Int. J. Chem. Reactor Eng. 5 (48A) (2007)
It is assumed that C = CAB. From the above equations, we got
AS
−6
−6
˚
E = 0.274 × 10 and for ФI = 0.28 × 10 . As ˚E and ˚I are smaller
1–19.
than 0.15 and 1, respectively, there is no mass transfer limitation
on reaction kinetics. Further, the SBA materials are mesoporous
with pores in the range 1.7–2.0 nm (Table 1). The dimension of FA
molecule is smaller than the pore size of SBA and hence, its internal
transport should not be an issue. Except for the 1 wt% supported Pd
which has a higher value, the TOF of 2–5 wt% Pd samples was nearly
the same (Table 4) further confirming above conclusion related to
absence of mass transfer limitation.
[
14] A. Guzman, J.E. Torres, L.P. Prada, M.L. Nunez, Catal. Today 156 (2010) 38–43.
[15] L. Boda, G. Onyestyák, H. Solt, F. Lónyi, J. Valyon, A. Thernesz, Appl. Catal. A:
Gen. 374 (2010) 158–169.
[
[
16] J.G. Immer, M.J. Kelly, H.H. Lamb, Appl. Catal. A: Gen. 375 (2010) 134–139.
17] R.W. Gosselink, S.A.W. Hollak, S.-W. Chang, J. van Haveren, K.P. de Jong, J.H.
Bitter, D.S. van Es, ChemSusChem 6 (2013) 1576–1594.
[
[
[
[
18] L. Hermida, A.Z. Abdulla, A.R. Mohamed, Renew. Sus. Energy Rev. 42 (2015)
1223–1233.
19] M. Snåre, I. Kubi cˇ ková, P. Mäki-Arvela, K. Eränen, D.Y. Murzin, Ind. Eng. Chem.
Res. 45 (2006) 5708–5715.
20] P. Mäki-Arvela, K. Kubi cˇ ková, M. Snåre, K. Eränen, D.Y. Murzin, Energy Fuels
Pd(3 wt%)/SBA-16 was tested also for the deoxygenation of
stearic acid (C18 saturated, SA) and non-edible jatropha oil (JA;
21 (2007) 30–41.
21] S. Lestari, P. Mäki-Arvela, H. Bernas, O. Simakova, R. Sjöholm, J. Beltramini,
G.Q.M. Lu, J. Myllyoja, I. Simakova, D.Y. Murzin, Energy Fuels 23 (2009)
3842–3845.
◦
triglyceride) at 325 C for 5 h (Table 5). While near complete conver-
sion of OA was observed, conversion of SA under similar conditions
was somewhat lower (92.3 mol%). Hydrodeoxygenation (forming
C18) was more predominant with SA (selectivity = 13.7%) than with
OA (0.8%). At the same time decarboxylation forming C17 was
prevalent with OA (94%) than with SA (80.2%). In other words, the
type of FA has certain effect on reactivity and reaction pathway.
Based on the above results (differences in reactivity and product
selectivity pattern with variation in the fatty acid type), it appears
that SA is not the only intermediate to the final products formed
[
[
22] K. Marata, Y. Liu, M. Inaba, I. Takahara, Energy Fuels 24 (2010) 2404–2409.
23] M. Ahmadi, A. Nambo, J.B. Jasinski, P. Ratnasamy, M.A. Carreon, Catal. Sci.
Technol. 5 (2015) 380–388.
[24] E.W. Ping, J. Pierson, R. Wallace, J.T. Miller, T.F. Fuller, C.W. Jones, Appl. Catal.
A: Gen. 396 (2011) 85–90.
[
25] S. Lestari, P. Mäki-Arvela, K. Eränen, J. Beltramini, G.Q.M. Lu, D.Y. Murzin,
Catal. Lett. 134 (2010) 250–257.
[26] S.-P. Lee, A. Ramli, Chem. Central J. 7 (2013) 149.
[
[
[
[
27] D. Zhao, Q. Huo, J. Feng, B.F. Chmelka, G.D. Stucky, J. Am. Chem. Soc. 120
(
1998) 6024–6036.
28] C.F. Cheng, Y.C. Lin, H.H. Cheng, S.M. Liu, H.S. Sheu, Chem. Lett. 33 (2004)
262–263.
29] Y. Sakamoto, M. Kaneda, O. Terasaki, D. Zhao, J. Kim, G.D. Stucky, H. Shin, R.
Ryoo, Nature 408 (2000) 449–453.
◦
from OA. At higher temperatures (300 C and above), decarboxyla-
tion competes hydrogenation at the Pd site and the mechanism is
all together different than reported in Scheme 1. A non-edible jat-
ropha oil could be converted with significant efficiency (97.4 mol%)
into C17 and C15 products. The catalyst showed high activity for
each of these reactants portraying its efficiency in deoxygenation
of structurally different feedstocks.
30] P.I. Ravikovitch, A.V. Neimark, Langmuir 18 (2002) 9830–9837.
[31] A. Kumar, D. Srinivas, J. Catal. 293 (2012) 126–140.
[
[
[
32] S. Popov, S. Kumar, Energy Fuels 29 (2015) 3377–3384.
33] J. Fu, F. Shi, L.T. Thompson Jr., X. Lu, P.E. Savage, ACS Catal. 1 (2011) 227–231.
34] M. Snåre, I. Kubi cˇ ková, P. Mäki-Arvela, K. Eränen, J. Wärnå, D.Y. Murzin, Chem.
Eng. J. 134 (2007) 29–34.
[
[
[
35] C. Yang, R. Nie, J. Fu, Z. Hou, X. Lu, Bioresour. Technol. 146 (2013) 569–573.
36] J.A. Lopez-Ruiz, R.J. Davis, Green Chem. 16 (2014) 683–694.
37] C.A. Ferretti, R.N. Olcese, C.R. Apestegu ˘ı a, J.I. Di Cosimo, Ind. Eng. Chem. Res.
4
. Conclusions
48 (2009) 10387–10394.
[
38] K. Belkacemi, A. Boulmerka, J. Arul, S. Hamoudi, Topics Catal. 37 (2006)
113–120.
Palladium (1–5 wt%) supported on SBA-12 and SBA-16 catalyzed
[
[
39] L. Zhou, A. Lawal, Energy Fuels 29 (2015) 262–272.
40] H.S. Fogler, Elements of Chemical Reaction Engineering, 4th ed., Prentice Hall
International Series in Physical and Chemical Engineering Sciences, New
Jersey, 2011.
decarboxylation of FAs (OA and SA) and non-edible vegetable
oil (jatropha) producing diesel-range hydrocarbons (C15–C18).
Pd/SBA-16 showed higher catalytic performance than Pd/SBA-12
due to its higher surface hydrophobicity and metal dispersion.
Nature of FA affected the reactivity and reaction pathway. At
[
41] A.A.C.M. Beenackers, W.P.M. Van Swaaij, Chem. Eng. Sci. 48 (1993)
3109–3139.
◦
higher temperatures (300 C and above), decarboxylation competes
hydrogenation at the Pd site and hence, SA is not the only interme-
diate for the final products in OA decarboxylation reactions. Activity
of these catalysts is comparable or higher than the hitherto reported
catalysts for carboxylation of fatty acids.