Hydrogenation of quinolines, alkenes, and biodiesel by palladium nanoparticles supported on magnesium oxide

Article abstract of DOI:10.1039/c2dt31533e

A new catalyst composed of Pd nanoparticles supported on MgO has been prepared by the room temperature NaBH₄ reduction of Na ₂PdCl₄ in methanol in the presence of the support. TEM measurements reveal well-dispersed Pd particles of mean diameter 1.7 nm attached to the MgO surface. Further characterization was achieved by ICP-AES, XPS, XRD, H₂ pulse chemisorption and H₂-TPR. The new catalyst is efficient for the regioselective hydrogenation of the heterocyclic ring of quinolines, as well as for the mild reduction of a variety of alkenes representative of fuel components, and the partial saturation of biodiesel. The new material is considerably more reactive than commercial Pd/SiO₂ and Pd/Al₂O₃ catalysts under analogous reaction conditions.

Full text of DOI:10.1039/c2dt31533e

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 14490
www.rsc.org/dalton
PAPER
Hydrogenation of quinolines, alkenes, and biodiesel by palladium
nanoparticles supported on magnesium oxide†‡
Reena Rahi, Minfeng Fang, Atif Ahmed and Roberto A. Sánchez-Delgado*
Received 11th July 2012, Accepted 26th September 2012
DOI: 10.1039/c2dt31533e
A new catalyst composed of Pd nanoparticles supported on MgO has been prepared by the room
temperature NaBH4 reduction of Na2PdCl4 in methanol in the presence of the support. TEM
measurements reveal well-dispersed Pd particles of mean diameter 1.7 nm attached to the MgO surface.
Further characterization was achieved by ICP-AES, XPS, XRD, H2 pulse chemisorption and H2-TPR.
The new catalyst is efficient for the regioselective hydrogenation of the heterocyclic ring of quinolines, as
well as for the mild reduction of a variety of alkenes representative of fuel components, and the partial
saturation of biodiesel. The new material is considerably more reactive than commercial Pd/SiO2 and
Pd/Al2O3 catalysts under analogous reaction conditions.
common impurities. Metal nanoparticles (NPs) represent a pro-
1. Introduction
mising alternative for the development of novel catalytic systems,
due to their unique activity and selectivity features associated
with the small and generally uniform particle size.17 Palladium
nanoparticles, in particular, have been used in catalysis, in liquid
suspensions, in ionic liquids, and supported on solids.18,19 Most
reports are concerned with C–C coupling, but examples of
hydrogenation reactions are also available; they include the
reduction of simple alkenes by Pd NPs immobilized by ionic
liquids on zeolites,20 of butadiene by Pd NPs on porous alumina
membranes,21 of the CvC bond of dihydrolinalool by Pd NPs
on polystyrene-block-poly-4-vinylpyridine,22 and of cinnamalde-
hyde by Pd NPs inside multi-walled carbon nanotubes.23 Pd NPs
on polyvinylpyrrolidones have been used to selectively hydro-
genate alkynes to alkenes24 and nitrotoluenes to their corres-
ponding anilines;25 acetophenone is reduced to ethylbenzene
via phenylethanol using Pd NPs supported on polymers.26
Cascade reactions involving Heck coupling followed by CvC
bond hydrogenation of the intermediate product have been
reported with Pd/C,27 Pd/TiO2,28 and with Pd nanoparticles
stabilized by ionic liquids, either unsupported,29 or supported in
functionalized multi-walled carbon nanotubes.30 Hydrogenation
of benzene, benzylidene-acetone, phenylacetylene, diphenylace-
tylene, and quinoline was achieved by Pd NPs on hyperbranched
aramids,31 while 5-membered N/O/S-heterocycles were reduced
by Pd NPs supported on aminomethyl polystyrene modified with
hyperbranched dendritic polyamidoamines.32 Phenanthrolines
were hydrogenated by use of Pd nanoparticles supported on
silica,33 and phenol was reduced predominantly to cyclohexa-
none by use of Pd on MgO as the catalyst.34,35
Catalytic hydrogenation continues to be of great interest in
relation to a variety of industrially important processes. Partial or
total saturation of N-heteroaromatic compounds is of relevance
to hydrodenitrogenation (HDN), a key reaction routinely used
for the removal of nitrogen impurities from refinery
feedstocks.1–4 There is also a growing interest in the hydrogen-
ation of nitrogen heterocycles as a possible means to store hydro-
gen in organic liquids,5 and selective reduction of quinolines to
1,2,3,4-tetrahydroquinolines may lead to important intermediates
for the synthesis of bioactive compounds.6 Selective alkene
hydrogenation, in turn, is of interest in the conversion of fluid
catalytic cracking (FCC) naphtha into gasoline,7,8 and selective
hydrogenation of polyunsaturated fatty acid methyl esters
(FAMEs) to the corresponding monounsaturated derivatives
greatly enhances the storage stability and combustion properties
of biodiesel.9–16
Although a number of catalysts are known for all these reac-
tions, many do not display the desired activity and/or selectivity,
require drastic reaction conditions, or are susceptible to poison-
ing by the substrates, the products, or impurities present in the
feed. Therefore, there is a continuing need for new catalysts
capable of hydrogenating various unsaturated substrates under
mild reaction conditions, while being resistant to poisoning by
Chemistry Department, Brooklyn College, and The Graduate Center,
The City University of New York, 2900 Bedford Ave., Brooklyn,
NY 11210, USA. E-mail: RSdelgado@brooklyn.cuny.edu;
Fax: +1-718-951-4607; Tel: +1-718-951-5000 Ext.2827
†This paper is dedicated to Prof. David Cole-Hamilton on the occasion
of his retirement and for his outstanding contribution to transition metal
catalysis with many good memories (R.A. S.-D.) of our common time in
Geoff Wilkinson’s lab (1973–76).
In previous work on Ru NPs supported on poly(4-vinylpyri-
dine) (Ru/PVPy) we presented evidence indicating that the com-
bination of small metallic particles and basic sites on the surface
of the support leads to a nano-structure capable of promoting
heterolytic hydrogen splitting and ionic hydrogenation
‡Electronic supplementary information (ESI) available. See DOI:
10.1039/c2dt31533e
14490 | Dalton Trans., 2012, 41, 14490–14497
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 1 Heterolytic hydrogen splitting and ionic hydrogenation of
polar substrates by metal NPs on basic support.
mechanisms for N-heterocyclic substrates.36 Heterolytic hydro-
gen activation and ionic hydrogenation pathways are very
common in solution37 but extremely rare on solid surfaces. Such
outer sphere mechanisms (Scheme 1) do not require the binding
of the substrate or the products to the active metal site and there-
fore they constitute a possible route to avoid catalyst poisoning.
Here we describe a new catalyst of this series, composed of
palladium nanoparticles supported on magnesium oxide as the
basic support; this material is an efficient catalyst for the hydro-
genation of a variety of substrates of interest in relation with
fuels and energy issues.
Fig. 1 Transmission electron micrographs of a freshly prepared sample
(top) and a used sample (bottom) of 1 wt% Pd/MgO with corresponding
particle size distribution histogram.
2. Results and discussion
2.1. Catalyst preparation and characterization
2.1.1. Preparation. The 1 wt% Pd/MgO catalyst used in this
study was prepared by the room temperature reduction of
Na2PdCl4 with excess NaBH4 in methanol in the presence of
suspended MgO, which had been previously calcined at 500 °C
in the air for 2 h. Upon completion of the reaction, the solution
became colorless and the initially white solid changed to gray;
after appropriate work-up, the catalyst was stored under an inert
atmosphere. For comparison and characterization purposes, ana-
logous materials containing ca. 5 wt% and 10 wt% Pd/MgO
were prepared by using the same procedure described above,
except for the amount of Na2PdCl4 employed. The exact Pd
loading in all three catalysts, determined by inductively coupled
plasma atomic emission spectroscopy (ICP-AES), was 0.9 wt%,
4.4 wt% and 9.7 wt%, respectively. For convenience, hereafter
we use 1 wt%, 5 wt%, and 10 wt% Pd/MgO notation to rep-
resent the different solids; nevertheless, the actual metal loadings
from the ICP-AES analyses were used to calculate TOF values
in the catalytic experiments described below.
Fig. 2 Pd 3d XPS spectrum of a freshly prepared sample of 10 wt%
Pd/MgO.
distributions very similar to that of the 1 wt% material, with
average diameters of 1.5 and 1.8 nm, respectively.
2.1.2. TEM measurements. TEM analysis of the freshly pre-
pared 1 wt% Pd/MgO solid (Fig. 1) revealed the presence of
highly dispersed Pd particles with a narrow size distribution on
the surface of the support, with an average diameter of 1.7 nm,
calculated by measuring 350 particles chosen at random from
several images.
Analysis of a sample of this catalyst after being used in a qui-
noline hydrogenation run (Fig. 1) at 150 °C and 40 atm H2
shows that the average Pd particle diameter remains essentially
unchanged (1.6 nm), although the particle size distribution is
somewhat broader; these results indicate that the catalyst does
not undergo sintering or other aggregation processes to an impor-
tant extent under the harshest conditions employed for hydrogen-
ation reactions. TEM analysis of analogous fresh 5 wt% and
10 wt% Pd/MgO (Fig. S1, ESI‡) revealed particle size
2.1.3. XPS analysis. The survey XPS scan of the 1 wt% Pd/
MgO catalyst is shown in Fig. S2, ESI.‡ No peaks are observed
around 200 eV (Cl 2p), indicating that the catalyst surface is free
of Cl contamination from the preparation process, as opposed to
what is observed for a similar Pd/MgO catalyst prepared by the
impregnation method.34 The narrow scan for 1 wt% Pd/MgO in
the Pd 3d region displayed extremely weak signals because of
the very low metal loading, not allowing a reliable direct charac-
terization by this method. The 3p region is highly interfered by
large O peaks, making deconvolution invaluable. However, satis-
factory XPS data were obtained in the 3d region for the two ana-
logous solids containing 10 wt% Pd/MgO (Fig. 2) and 5 wt%
Pd/MgO (Fig. S3, ESI‡). The presence of metallic palladium on
these surfaces is clearly indicated by the signals at 335.2 and
339.3 eV, which correspond to Pd(0) 3d5/2 and Pd(0) 3d3/2
,
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 14490–14497 | 14491

View Article Online
respectively;34 no evidence for any other Pd species was
observed in the spectra of the two materials analyzed. With all
preparation parameters being identical except for the amount of
the Pd precursor, it is reasonable to conclude that the chemical
composition on the surface of the 1 wt% catalyst would be
similar. Furthermore, H2-TPR measurements (section 2.1.4
below) also support the presence of Pd(0) as the only or pre-
dominant species on the surface.
11% and 18%, respectively. The metal dispersion values
obtained from H2 chemisorption experiments were subsequently
used to calculate catalytic activities.
To provide further information concerning the reducibility of
chemical species present on the surface of the Pd/MgO materials,
H2-TPR experiments were carried out for 1 wt%, 5 wt% and
10 wt% Pd/MgO, as well as for the MgO support itself. The
TPR profiles for these four samples (Fig. 4) showed no positive
peaks indicative of hydrogen consumption. The negative peaks
observed around 360 K correspond to desorption from surface
metallic Pd of the H2 adsorbed at r.t. at the beginning of the
analysis. The relative intensities of these peaks correlate with the
amount of surface Pd in each sample, with the 10 wt% catalyst
having the largest H2 desorption peak.
These TPR profiles demonstrate that no significant reduction
processes take place within the temperature range 400–700 K.
This is in contrast to what was observed for Pd/MgO prepared
by impregnation, which showed a positive TPR peak at about
450 K ascribed to the presence of PdO.34 The TPR data for our
three Pd/MgO samples are consistent with the XRD and XPS
results presented above, which allows us to safely conclude that
the Pd in our catalysts is exclusively or predominantly metallic,
with no significant amount of PdO that can be detected by any
of the three methods employed.
2.1.4. XRD results. In order to further identify the crystalline
phase of Pd in the catalyst, powder X-ray diffraction patterns of
the three solids with different metal loadings were obtained, as
shown in Fig. 3. Similarly to what we observed in the XPS
experiments, the Pd diffraction pattern of the 1 wt% Pd/MgO
material was essentially undetectable because of the very low
metal loading; all five peaks are assigned to the MgO phase.
When the metal loading is increased to 5 wt%, a small diffuse
peak around 2θ = 40° starts to appear, which corresponds to the
Pd (111) crystalline plane; the rest of the diffraction pattern of
the 5 wt% Pd/MgO is still indistinguishable from that of MgO.
However, samples of the analogous 10 wt% Pd/MgO displayed
peaks associated with highly dispersed face centered cubic (fcc)
crystalline Pd at 2θ values of 40.3°, 46.8° and 68.3°, assigned to
the (111), (200) and (220) planes, respectively. These values are
consistent with those in the standard diffractogram of Pd (40.1°,
46.7° and 68.1°).38 The most intense peak at 2θ = 43.0° is MgO
(200),39 which was used as an internal standard to assign the
other signals.
2.2. Catalytic hydrogenation
2.2.1. Hydrogenation of quinolines. The 1 wt% Pd/MgO
catalyst is efficient and highly selective for the hydrogenation
2.1.5. BET surface area, H2 pulse chemisorption and temp-
erature programmed reduction (TPR) experiments. Table 1 con-
tains the data obtained from physisorption and chemisorption
measurements. The MgO support (after calcination at 500 °C for
2 h) has a specific surface area of 88 m2 g−1, which is essentially
retained (91 m2 g−1) after deposition of 1 wt% Pd NPs onto the
surface. H2 pulse chemisorption measurements for the 1 wt%
Pd/MgO revealed ca. 5% metal dispersion on the surface.
Similar measurements were performed on commercial 5% Pd/
SiO2 and 5% Pd/Al2O3 catalysts, which were used for catalytic
activity comparison, leading to higher metal dispersion values of
Table 1 Physical and chemical adsorption measurement results
Material
BET surface areaa (m2 g−1
)
Metal dispersion (%)
MgO
88
91
—
—
—
1% Pd/MgO
5% Pd/SiO2
5% Pd/Al2O3
4.7 0.4%
11 2%
18 2%
a Single-point measurement.
Fig. 4 H2-TPR profile of the three Pd/MgO catalysts and MgO
Fig. 3 Powder X-ray diffraction pattern of 1, 5 and 10 wt% Pd/MgO.
14492 | Dalton Trans., 2012, 41, 14490–14497
support.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 2 Hydrogenation of quinolines over 1 wt% Pd/MgOa
Entry
1
Substrate
TOF (h−1
300
)
TOFcorr (h−1
)
Product
6400
2
3
4
310
250
250
6600
5300
5400
Fig. 5 Hydrogenation of quinoline catalyzed by 1 wt% Pd/MgO
(150 °C, 40 atm H2; nQ : nPd = 1000; in THF; △: quinoline; ▲: 1,2,3,4-
tetrahydroquinoline.
a In n-hexane; nsub : nPd = 1000; 150 °C, 40 atm H2.
of a methyl group in the 3- or 8-position of quinoline results in
lower hydrogenation rates, while no effect is apparent when the
methyl group is located in the 2-position. All the substrates were
exclusively hydrogenated at the heterocycle with no evidence for
products of the hydrogenation of the carbocycle being observed
in any experiment.
For comparison, commercial 5% Pd/SiO2 and 5% Pd/Al2O3
catalysts displayed activities about four times lower than that of
our 1 wt% Pd/MgO under analogous reaction conditions and
equivalent amounts of Pd, yielding TOFcorr values of 1800 h−1
and 1600 h−1, respectively, for the hydrogenation of quinoline to
1,2,3,4-tetrahydroquinoline. To our knowledge, only one other
quinoline hydrogenation catalyst containing Pd NPs (on hyper-
branched aramids) is known.31 In that case, uncorrected TOF
values of ∼350 h−1, comparable to ours, were achieved. We are
not aware of any reported examples involving substituted quino-
lines. As mentioned in the introduction, the hydrogenation of
N-heterocycles is of interest in connection with hydrodenitro-
genation of fossil fuel feedstocks,1–4 and with hydrogen storage
in organic liquids.5 The high selectivity for saturation of the
heterocyclic ring observed could also be of use in the synthesis
of valuable tetrahydroquinoline intermediates.6
Fig. 6 Hydrogenation of quinoline over 1 wt% Pd/MgO (nsub : nPd
1000 in THF). Left: influence of pressure on the hydrogenation rate,
150 °C; right: effect of temperature on the hydrogenation rate at 40 atm.
=
of quinoline to 1,2,3,4-tetrahydroquinoline under moderate reac-
tion conditions (150 °C and 40 atm H2) using THF or hexane as
the solvent. A typical reaction profile is depicted in Fig. 5; the
conversion reaches 100% after a few hours with no evidence for
catalyst poisoning or inhibition by the substrate or the product.
The linear hydrogen uptake observed during the first 1.5–2 h of
the reaction was used to calculate an initial turnover frequency
(TOFinit) of 300 h−1
.
This value was subsequently corrected for metal dispersion to
yield a TOFcorr of 6400 h−1, which is our measure of catalytic
activity. The high selectivity of the Pd/MgO catalyst for the
hydrogenation of the heterocycle, coupled with a total lack of
activity toward toluene observed in independent experiments, is
suggestive of an ionic mechanism for quinoline reduction,
similar to the one proposed by us for the Ru/PVPy catalyst.36
As shown in Fig. 6 (left), TOFcorr values increase steadily
with H2 pressure within the range 20–40 atm, indicating that the
catalyst does not undergo metal sintering or any other deactivat-
ing processes to a significant extent at H2 pressures of up to
40 atm, in agreement with our TEM observations (Fig. 1). The
catalytic activity also increased with the temperature in the range
100–150 °C as depicted in Fig. 6 (right). Although it is likely
that hydrogenation rates can be further increased by using higher
pressures and temperatures, we consider the TOF values
obtained within this range satisfactory.
2.2.2. Hydrogenation of alkenes. FCC naphtha represents
about 36% of the gasoline pool composition in the US. It is pro-
duced with up to 40 vol% of C5–C7 alkenes of varied structures,
but specifications for the final product allow at most about 12%.
The excess of alkenes in naphtha is usually lowered by hydro-
genation that takes place during hydrodesulfurization (HDS)
over standard solid catalysts. Excessive reduction of the olefin
content, however, can result in a marked decrease in the octane
rating of the fuel.7,8 Highly substituted and branched alkenes are
responsible for most of the octane rating and therefore, it is desi-
rable to hydrogenate predominantly the linear and unsubstituted
components. In this context, we evaluated the behavior of our
1 wt% Pd/MgO catalyst in the mild hydrogenation (25 °C,
10 atm H2) of C5–C6 olefins representative of naphtha com-
ponents, with different degrees of substitution at the CvC bond.
The results collected in Table 3 (entries 1–5) show that the
The 1 wt% Pd/MgO catalyst is also efficient and highly selec-
tive for the hydrogenation of substituted quinolines under analo-
gous reaction conditions, as summarized in Table 2. Introduction
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 14490–14497 | 14493

View Article Online
Table 3 Hydrogenation of selected alkenes over 1 wt% Pd/MgOa
Entry
1
Substrate
TOF (h−1
)
TOFcorr (h−1
)
Product at 2 h
2300
50 000
2
3
2800
180
60 000
3800
4
5
6
90
70
1900
1500
1800
39 000
a nsub : nPd = 1000; r.t., 10 atm H2.
Table 4 Biodiesel composition determined by GC-MS
hydrogenation activity of our catalyst decreases sharply with
increasing congestion around the double bond.
Compounda
%
Such selectivity would leave the branched olefins essentially
untouched when mixed with linear or unsubstituted analogues.
Previous reports of C6-alkene hydrogenation using Pd NPs
supported on silica indicated TOF values ranging from
4000–38 000 h−1 under similar reaction conditions.40,41
C16 : 0
C18 : 0
C18 : 1
C18 : 2
C18 : 3
9
5
31
48
7
a In Cn : x, n indicates the number of carbons and x the number of CvC
bonds in the FAMEs.
2.2.3. Partial hydrogenation of biodiesel. Biodiesel is an
important alternative fuel most commonly produced by trans-
esterification of vegetable oils catalyzed by sodium or potassium
hydroxides or alkoxides, to yield mixtures of fatty acid methyl
esters (FAMEs). Biodiesel is an attractive alternative to pet-
roleum-based diesel because it displays similar combustion pro-
perties, viscosity, and cetane number but it is renewable, non-
toxic, and sulphur free, and it is considered to have a negative
CO2 balance. Biodiesel production is increasing steadily, to be
used in motor vehicles as a single fuel or in blends with pet-
roleum-derived diesel.9,10,13,14 However, several problems
related to ageing due to low oxidative stability, higher NOx emis-
sions in petrodiesel blends, and poor performance at low temp-
eratures have been associated with a high proportion of
polyunsaturated FAMEs; partial catalytic hydrogenation of the
polyenes to monoenes has been proposed as an efficient way to
improve the characteristics of biodiesel.14,16,42,43 Several cataly-
tic systems have been reported for the hydrogenation of vege-
table oils or biodiesel; for instance, metallic Ni suspended in oil
can be used at 170–180 °C under atmospheric pressure;11,44
silica-supported Cu is active and selective at 180 °C and 20 atm
H2.45,46 Liquid biphasic systems based on rhodium with sulfo-
nated triphenylphosphine ligands operate at 50–100 °C and
10 atm H2.42,43 Palladium supported on alumina was employed
as a catalyst in the partial hydrogenation of sunflower oil at
40 °C and 10 atm H2, as a pretreatment for biodiesel synthesis;47
other metal oxide supports, including MgO, provided lower
activities and/or selectivities. Recently, the partial hydrogenation
of biodiesel catalyzed by Pd NPs dispersed in ionic liquids48 or
Pd NPs supported on Al2O3 in the presence of ionic liquids49
has been reported; these systems are efficient at room tempera-
ture but require high H2 pressures (75 atm).
Since our 1 wt% Pd/MgO catalyst proved to be very active for
the mild hydrogenation of 1-undecene (Table 3, entry 6), it
seemed interesting to evaluate its efficacy in the partial hydro-
genation of a biodiesel prepared by transesterification of soybean
oil with methanol using KOH as the catalyst. The composition
of the biodiesel, determined by GC-MS, is shown in Table 4;
our aim was to transform the polyunsaturated FAMEs (C18 : 2
and C18 : 3), which represent over 50% of the biodiesel, into the
more desirable monounsaturated (C18 : 1) ester, without increas-
ing the proportion of the fully saturated components (C18 : 0 and
C16 : 0, initially 14%).
After preliminary screening for optimal reaction conditions in
the range 25–100 °C/1–7 atm H2, we were able to rapidly and
selectively convert the mixture of FAMEs into predominantly
(>80%) the desired C18 : 1 product, under extremely mild con-
ditions (100 °C and 1 atm H2) (Fig. 7).
No further reduction to the fully saturated species was
observed even at longer reaction times. The final composition
achieved corresponds to an upgraded first-generation biodiesel
with improved stability and performance, indicating that our
1 wt% Pd/MgO catalyst could be of practical use in this type of
application.
2.3. Catalyst recyclability
In order to ascertain the stability and recyclability of our catalyst,
three consecutive hydrogenation runs were performed using the
same sample of 1 wt% Pd/MgO, for cyclohexene and for
14494 | Dalton Trans., 2012, 41, 14490–14497
This journal is © The Royal Society of Chemistry 2012

View Article Online
3. Conclusion
We have synthesized a new catalyst composed of Pd nanoparti-
cles supported on the basic support MgO. TEM measurements
reveal well-dispersed Pd particles of about 1.7 nm. The catalyst
was further characterized by XPS, XRD, hydrogen pulse chemi-
sorption and H2-TPR measurements. The new material is
efficient for the regioselective hydrogenation of the heterocyclic
ring of quinolines under moderate reaction conditions (150 °C,
40 atm H2), as well as for the mild reduction (25 °C and 10 atm
H2) of a variety of alkenes representative of fuel components,
and the partial saturation of biodiesel under atmospheric hydro-
gen pressure at 100 °C. The catalyst is about four times more
active in the hydrogenation of quinoline than commercial Pd/
SiO2 and Pd/Al2O3 catalysts, and it is recyclable for alkene
hydrogenation for at least three cycles without significant loss of
catalytic activity, while a 30% loss of activity was observed after
three catalytic runs of quinoline hydrogenation, probably due to
inhibition by trace impurities in the substrate.
Fig. 7 Partial hydrogenation of FAMEs over 5% Pd/MgO (100 °C,
1 atm; ■: C18 : 3; ○: C18 : 0; ●: C16 : 0; △: C18 : 2; ▼: C18 : 1).
4. Experimental
4.1. Materials
Na2PdCl4 (Pressure Chemicals, Inc.), 5% Pd/SiO2 and 5% Pd/
Al2O3 (Strem Chemicals) were used as purchased. Solvents
(analytical grade, Sigma-Aldrich) were purified using a PureSolv
purification unit from Innovative Technology, Inc. and further
deoxygenated with a nitrogen flow prior to use. Substrates and
other reagents (Sigma-Aldrich) were purified by appropriate
methods prior to use as necessary. Magnesium oxide
(∼325 mesh, ≥99% trace metals basis) (Sigma-Aldrich) was cal-
cined prior to metal loading at 500 °C in the air for 2 h. Bio-
diesel was synthesized by a standard procedure: in short, a
sonicated solution of KOH (0.875 g) in methanol (50 cm3) was
added to commercial soybean oil (250 cm3) and the mixture was
stirred at 60 °C for 75 min. The biodiesel phase was separated
and then heated to 80 °C for 15 min to remove residual metha-
nol. The product was washed six times with warm water to
remove glycerol and the remaining KOH, then heated to 110 °C
for 20 min and finally dried over sodium sulfate overnight.
Fig. 8 Activities of the 1 wt% Pd/MgO catalyst (100 mg) in three con-
secutive runs of hydrogenation; left: cyclohexene, r.t., 10 atm; right: qui-
noline, 150 °C, 40 atm.
quinoline. Fig. 8 shows the corrected TOF values for each hydro-
genation cycle, together with the normalized hydrogenation
activities. The hydrogenation activities for cyclohexene dis-
played about 10% variation, which is in the order of the exper-
imental error of our TOF measurements, suggesting that the
catalyst is stable and reusable for alkene hydrogenation for at
least three cycles without significant loss of catalytic activity. In
contrast, recycling the catalyst in quinoline hydrogenation reac-
tions resulted in about 30% loss of activity after the third run.
Comparable results were obtained for commercial 5% Pd/SiO2
and 5% Pd/Al2O3 catalysts during quinoline hydrogenation re-
cycling under the same conditions.
4.2. Synthesis of Pd/MgO
ICP-AES analysis of the catalyst recovered from quinoline
recycling experiments revealed essentially no loss of Pd from the
catalyst (0.9 wt% used vs. 0.9 wt% fresh), while the liquid
organic phase contained only very small amounts of Pd (<5% of
total Pd); therefore, the observed loss of activity cannot be
ascribed to leaching of the metal into the solution. TEM images
of a used catalyst (Fig. 1) show no significant aggregation into
larger Pd NPs that could account for a decrease in the catalytic
efficiency. In independent experiments using a 50 : 50 quinoline/
1,2,3,4-tetrahydroquinoline mixture as the substrate, the rate of
hydrogenation observed was very similar to that of quinoline
alone, which rules out the possibility of inhibition by the product
1,2,3,4-tetrahydroquinoline. The observed moderate decrease in
activity is likely due to the presence of an unidentified trace
impurity in quinoline, since such a deactivation effect is not
observed during alkene hydrogenation.
To prepare the 1 wt% Pd/MgO catalyst, magnesium oxide
(2.0 g) was placed in a 3-neck round bottom flask under nitro-
gen. Dry deoxygenated methanol (10 cm3) was added and the
mixture was stirred. Two pressure-equalizing dropping funnels
were attached to the flask, one containing Na2PdCl4 (0.055 g;
0.02 g Pd) in dry deoxygenated methanol (10 cm3) and the
second containing NaBH4 (0.142 g) in the same solvent
(20 cm3). Approximately 10 cm3 of the borohydride solution
was added quickly to the flask and the mixture was stirred for a
few minutes. Subsequently, both the Na2PdCl4 and the remain-
ing NaBH4 solutions were added simultaneously to the flask at
the rate of about one drop per second. After the addition was
complete, the dark gray solution was stirred for 4 h under nitro-
gen at room temperature. The product was filtered under nitro-
gen, washed three times with methanol (10 cm3), and dried
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 14490–14497 | 14495

View Article Online
4.7. H2-TPR analysis
under vacuum at room temperature overnight. Analysis by
ICP-AES (Micro-Analysis, Inc., Wilmington) confirmed a metal
content of 0.9 wt%. Materials containing 5 wt% and 10 wt% Pd/
MgO were prepared by use of the same procedure, adjusting the
amount of Na2PdCl4 to 0.275 g and 0.550 g (0.10 g and 0.20 g
Pd), respectively; the metal content was verified by ICP-AES
analysis in both cases as 4.4% and 9.7%, respectively.
Temperature programmed reduction (TPR) of catalyst samples
was also conducted in the Chemisorb 2750 instrument equipped
with a thermal conductivity detector (TCD). About 120 mg of
the sample was first degassed at 200 °C in a flow of Ar gas
(25 mL min−1) for 1 h and then cooled down to room tempera-
ture. The carrier gas was then switched to a 9.9% H2/Ar mixture;
after 10 min when the baseline of the TCD signals was stable,
the sample was heated linearly at a rate of 10° min−1 to 700 K.
The TCD signal change was displayed in real time on the com-
puter interfaced to the instrument through the TPx system.
4.3. TEM studies
Transmission electron microscopy analysis was conducted at the
Environmental Sciences Analytical Center (ESAC) of the Earth
and Environmental Sciences Department of Brooklyn College
using a JEOL TEM-2010 high-resolution microscope, operating
at an accelerating voltage of 200 kV and providing a point-to-
point resolution of 0.19 nm. Samples for analysis were prepared
by placing a drop of a suspension of the catalyst in hexane on a
copper grid and allowing it to air-dry. Images were captured
using an AMT camera system.
4.8. Catalytic tests
4.8.1. Hydrogenation of quinolines. Hydrogenation of qui-
nolines was carried out in a glass-lined 5500 Parr reactor
(100 cm3) equipped with an internal stirrer, a thermocouple, a
dip tube, and a sampling valve, coupled to a 4843 controller.
Typically the reactor was loaded with the catalyst (100 mg) and
the desired amount of the substrate in 30 cm3 of the solvent. The
reaction mixture was deoxygenated by flushing with H2 (20 atm)
three times. The reactor was pressurized with H2 at room temp-
erature to 33 atm and then heated while stirring until the temp-
erature reached 150 °C; at that point the pressure was adjusted to
40 atm and this was taken as the zero time for each reaction. The
progress of the reaction was followed by monitoring the H2
uptake. At the end of the run, a sample of the final mixture was
analyzed by GC-MS using a Varian 3900 gas chromatograph
fitted with a FactorFour VF-5ms capillary column and a Saturn
2100T mass detector. TOFinit values were calculated from the
slope of the linear part of TON vs. t plots. Average TOF values
were obtained from at least three experiments; the dispersion
observed between individual measurements was less than 10%.
Finally, TOF values were corrected for metal dispersion
(TOFcorr), using the data obtained from chemisorption.
4.4. XRD measurements
Powder X-ray diffraction (XRD) patterns were obtained at the
Environmental Sciences Analytical Center (ESAC) of the Earth
and Environmental Sciences Department of Brooklyn College
on a Phillips X’PERT MPD diffractometer, using monochro-
matic Cu-Kα radiation at 45 kV and 40 mA and 2θ scanning
from 20° to 90°. Samples were ground in a mortar prior to analy-
sis in the air at room temperature.
4.5. XPS measurements
XPS spectra were recorded using an Omicron XPS spectrometer
equipped with a multichannel hemispherical analyzer and a dual
Al/Mg X-ray source using Mg Kα excitation (1253.6 eV). The
powdered sample was mounted on studs using a double-sided
adhesive tape in air. The data analysis was performed by decon-
voluting the XPS peaks by curve fitting using XPSPEAKS 4.1,
applying Shirley background subtraction and Lorentzian–Gaus-
sian functions (20% L, 80% G). The charging effect was
corrected based on the literature value of 529.9 eV for the O 1s
peak in MgO.50
4.8.2. Hydrogenation of alkenes. Experiments were carried
out using a 5000 Parr multireactor (75 cm3) equipped with a
magnetic stirrer and a thermocouple, and coupled to a 4871 con-
troller. The reactor was loaded with the catalyst (100 mg) and
the substrate (10 cm3), flushed with H2 (5 atm) three times, and
then pressurized to 10 atm H2 at room temperature. Stirring was
initiated at that point, which was taken as the zero time for the
reaction. At the end of the run, a sample of the mixture was ana-
lyzed by use of a Shimadzu 2010 gas chromatograph fitted with
an FID detector and an Agilent HP-Al/S capillary column. Each
experiment was repeated at least three times in order to ensure
reproducibility.
4.6. Hydrogen pulse chemisorption
Chemisorption measurements were performed using a Micro-
meritics ChemiSorb 2750 instrument, coupled to a TPx system
controller. The fresh catalyst was reduced in 10% H2/Ar at
150 °C for 30 min and then purged with Ar at 150 °C for 30 min
to remove impurities; after which the sample was allowed to
cool to room temperature. Doses of 10% H2/Ar were repeatedly
introduced into the sample tube until no further H2 uptake was
observed. The gas composition was analyzed by use of a TCD
detector. The metal dispersion was determined by built-in soft-
ware that detects the amount of hydrogen adsorbed for each
atom of palladium, using a 1 : 1 H : Pd model.51
4.8.3. Partial hydrogenation of biodiesel. Partial hydrogen-
ation of biodiesel was carried out in a low-pressure 5100 Parr
glass reactor (160 cm3) equipped with a thermocouple and a
stirrer, and coupled with a 4848 controller. The reactor was
loaded with the catalyst (100 mg) and the biodiesel (75 cm3),
and flushed three times with 1 atm H2. The mixture was heated
while stirring until the desired temperature was reached, which
was taken as zero time for the reaction. The mixture was kept
under a constant H2 pressure of 1 atm and the hydrogenation pro-
gress was followed by taking samples every 10 min and
14496 | Dalton Trans., 2012, 41, 14490–14497
This journal is © The Royal Society of Chemistry 2012

View Article Online
17 D. Astruc, F. Lu and J. R. Aranzaes, Angew. Chem., Int. Ed., 2005, 44,
7852–7872.
18 J. Durand, E. Teuma and M. Gómez, Eur. J. Inorg. Chem., 2008, 2008,
3577–3586.
immediately analyzing them by GC-MS on a Supelco SP-2330
capillary column.
4.8.4. Recycling experiments. A quinoline hydrogenation
run was performed as described in Section 4.8.1, using 1 mL of
quinoline and 30 cm3 hexane as the solvent. At the end of the
reaction, the reactor was cooled to room temperature and the
catalyst was allowed to settle down; the supernatant liquid was
carefully withdrawn as much as possible and 1 mL of fresh
quinoline and 30 cm3 of solvent were added. A second hydro-
genation run was then started following the same procedure;
further recycling steps were carried out by repeating the same
procedure.
When using cyclohexene as the substrate, 10 mL of neat
cyclohexene was used in the first run and the experiments were
carried out at room temperature. After 3 h of reaction, another
10 mL of fresh cyclohexene was added and a second run was
started. The hydrogenation was carried out for three runs in total.
19 I. Favier, D. Madec, E. Teuma and M. Gomez, Curr. Org. Chem., 2011,
15, 3127–3174.
20 J. Huang, T. Jiang, H. Gao, B. Han, Z. Liu, W. Wu, Y. Chang and
G. Zhao, Angew. Chem., Int. Ed., 2004, 43, 1397–1399.
21 H.-P. Kormann, G. Schmid, K. Pelzer, K. Philippot and B. Chaudret,
Z. Anorg. Allg. Chem., 2004, 630, 1913–1918.
22 L. M. Bronstein, D. M. Chernyshov, I. O. Volkov, M. G. Ezernitskaya,
P. M. Valetsky, V. G. Matveeva and E. M. Sulman, J. Catal., 2000, 196,
302–314.
23 J.-P. Tessonnier, L. Pesant, G. Ehret, M. J. Ledoux and C. Pham-Huu,
Appl. Catal., A, 2005, 288, 203–210.
24 C. Evangelisti, N. Panziera, A. D’Alessio, L. Bertinetti, M. Botavina and
G. Vitulli, J. Catal., 2010, 272, 246–252.
25 J.-L. Pellegatta, C. Blandy, R. Choukroun, C. Lorber, B. Chaudret,
P. Lecante and E. Snoeck, New J. Chem., 2003, 27, 1528–1532.
26 A. Drelinkiewicza, A. Waksmundzka, W. Makowski, J. W. Sobczak,
A. Król and A. Zieba, Catal. Lett., 2004, 94, 143–156.
27 M. Gruber, S. Chouzier, K. Koehler and L. Djakovitch, Appl. Catal., A,
2004, 265, 161–169.
28 M. J. Climent, A. Corma, S. Iborra and M. Mifsud, Adv. Synth. Catal.,
2007, 349, 1949–1954.
29 S. Jansat, J. Durand, I. Favier, F. Malbosc, C. Pradel, E. Teuma and
M. Gómez, ChemCatChem, 2009, 1, 244–246.
Acknowledgements
30 L. Rodríguez-Pérez, C. Pradel, P. Serp, M. Gómez and E. Teuma, Chem-
CatChem, 2011, 3, 749–754.
31 D. Tabuani, O. Monticelli, A. Chincarini, C. Bianchini, F. Vizza,
S. Moneti and S. Russo, Macromolecules, 2003, 36, 4294–4301.
32 Z. Guo, H. Feng, H.-c. Ma, Q.-X. Kang and Z.-W. Yang, Polym. Adv.
Technol., 2004, 15, 100–104.
33 C. Bianchini, V. Dal Santo, A. Meli, S. Moneti, R. Psaro, L. Sordelli and
F. Vizza, Inorg. Chim. Acta, 2008, 361, 3677–3680.
34 P. Claus, H. Berndt, C. Mohr, J. Radnik, E.-J. Shin and M. A. Keane,
J. Catal., 2000, 192, 88–97.
We thank Dr Zhongqi Cheng and the Environmental Sciences
Analytical Center (ESAC) of Brooklyn College for use of the
X-ray powder diffractometer. We also thank Dr Alan Lyons of
the Chemistry Department of the College of Staten Island for
access to the XPS spectrometer. Financial support from the
Donors of the American Chemical Society Petroleum Research
Fund (Grant # 47472-AC3) and the US Department of Energy
(Grant # DEEE0003129) is gratefully acknowledged.
35 N. Mahata, K. V. Raghavan, V. Vishwanathan, C. Park and M. A. Keane,
Phys. Chem. Chem. Phys., 2001, 3, 2712–2719.
36 M. Fang, N. Machalaba and R. A. Sánchez-Delgado, Dalton Trans.,
2011, 10621–10632.
37 C. A. Sandoval, T. Ohkuma, K. Muniz and R. Noyori, J. Am. Chem.
Soc., 2003, 125, 13490–13503.
38 Powder Diffraction File, Card # 46-1043, PDF-2/Release 2004; Inter-
national Center for Diffraction Data, Newton Square, PA, USA.
39 Powder Diffraction File, Card # 65-0476, PDF-2/Release 2004; Inter-
national Center for Diffraction Data, Newton Square, PA, USA.
40 A. V. Biradar, A. A. Biradar and T. Asefa, Langmuir, 2011, 27, 14408–
14418.
Notes and references
1 H. Topsøe, B. S. Clausen and F. E. Massoth, Hydrotreating Catalysis
Science and Technology, Springer-Verlag, New York, 1996.
2 The Desulfurization of Heavy Oils and Residua, ed. J. G. Speight, Marcel
Dekker, New York, 2000.
3 R. A. Sánchez-Delgado, Organometallic Modeling of the Hydrodesulfuri-
zation and Hydrodenitrogenation Reactions, Kluwer Academic Publish-
ers, Dordrecht, 2002.
41 O. Domínguez-Quintero, S. Martínez, Y. Henríquez, L. D’Ornelas,
H. Krentzien and J. Osuna, J. Mol. Catal. A: Chem., 2003, 197,
185–191.
42 A. Bouriazos, K. Mouratidis, N. Psaroudakis and G. Papadogianakis,
Catal. Lett., 2008, 121, 158–164.
4 K. Toshiaki, I. Atsushi and Q. Weihua, Hydrodesulfurization and Hydro-
denitrogenation: Chemistry and Engineering, Wiley-VCH, Tokyo, 1999.
5 R. H. Crabtree, Energy Environ. Sci., 2008, 1, 134–138.
6 S.-I. Murahashi, Y. Imada and Y. Hirai, Bull. Chem. Soc. Jpn., 1989, 62,
2968–2976.
43 A. Bouriazos, S. Sotiriou, C. Vangelis and G. Papadogianakis, J. Organo-
met. Chem., 2010, 695, 327–337.
7 S. Brunet, D. Mey, G. Pérot, C. Bouchy and F. Diehl, Appl. Catal., A,
2005, 278, 143–172.
44 B. R. Moser, M. J. Haas, J. K. Winkler, M. A. Jackson, S. Z. Erhan and
G. R. List, Eur. J. Lipid Sci. Technol., 2007, 109, 17–24.
45 N. Ravasio, F. Zaccheria, M. Gargano, S. Recchia, A. Fusi, N. Poli and
R. Psaro, Appl. Catal., A, 2002, 233, 1–6.
46 F. Zaccheria, R. Psaro and N. Ravasio, Green Chem., 2009, 11, 462–465.
47 B. Nohair, C. Especel, G. Lafaye, P. Marécot, L. C. Hoang and J. Barbier,
J. Mol. Catal. A: Chem., 2005, 229, 117–126.
8 P. J. Baricelli, E. Lujano, M. Modroño, A. C. Marrero, Y. M. García,
A. Fuentes and R. A. Sánchez-Delgado, J. Organomet. Chem., 2004,
689, 3782–3792.
9 M. Di Serio, M. Cozzolino, M. Giordano, R. Tesser, P. Patrono and
E. Santacesaria, Ind. Eng. Chem. Res., 2007, 46, 6379–6384.
10 M. Di Serio, R. Tesser, L. Pengmei and E. Santacesaria, Energy Fuels,
2007, 22, 207–217.
11 O. Falk and R. Meyer-Pittroff, Eur. J. Lipid Sci. Technol., 2004, 106,
837–843.
12 E. Lotero, Y. Liu, D. E. Lopez, K. Suwannakarn, D. A. Bruce and
J. G. Goodwin, Ind. Eng. Chem. Res., 2005, 44, 5353–5363.
13 J. A. Melero, J. Iglesias and G. Morales, Green Chem., 2009, 1285–
1308.
14 S. Pinzi, I. L. Garcia, F. J. Lopez-Gimenez, M. D. Luque de Castro,
G. Dorado and M. P. Dorado, Energy Fuels, 2009, 23, 2325–2341.
15 Y. C. Sharma, B. Singh and J. Korstad, Fuel, 2011, 90, 1309–1324.
16 R. Yang, M. Su, M. Li, J. Zhang, X. Hao and H. Zhang, Bioresour.
Technol., 2010, 101, 5903–5909.
48 M. S. Carvalho, R. A. Lacerda, J. P. B. Leao, J. D. Scholten,
B. A. D. Neto and P. A. Z. Suarez, Catal. Sci. Technol., 2011, 1, 480–
488.
49 B. S. Souza, D. M. M. Pinho, E. C. Leopoldino, P. A. Z. Suarez and
F. Nome, Appl. Catal. A., 2012, 433–434, 109–114.
50 J. F. Moulder, W. F. Stickle, P. E. Sobol and K. D. Bomben, in Handbook
of X-ray Photoelectron Spectroscopy: A Reference Book of Standard
Spectra for Identification and Interpretation of XPS Data, ed. J. Chastain,
Perkin-Elmer, Eden Prairie, MN, 1992.
51 C. Amorim and M. A. Keane, J. Colloid Interface Sci., 2008, 322,
196–208.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 14490–14497 | 14497