Highly dispersed palladium nanoclusters incorporated in amino-functionalized silica spheres for the selective hydrogenation of succinic acid to γ-butyrolactone

Article abstract of DOI:10.1002/aoc.3342

Highly dispersed palladium nanoclusters incorporated on amino-functionalized silica sphere surfaces (Pd/SiO₂-NH₂) were fabricated by a simple one-pot synthesis utilizing 3-(2-aminoethylamino)propyltrimethoxysilane (AAPTS) as coordinating agent. Uniform palladium nanoclusters with an average size of 1.1 nm can be obtained during the co-condensation of tetraethyl orthosilicate and AAPTS owing to the strong interaction between palladium species and amino groups in AAPTS. The palladium particle size can be controlled by addition of AAPTS and plays a significant role in the catalytic performance. The Pd/SiO₂-NH₂ catalyst exhibits high catalytic activity for succinic acid hydrogenation with 100% conversion and 94% selectivity towards γ-butyrolactone using 1,4-dioxane as solvent at 240°C and 60 bar for 4 h. Moreover, the Pd/SiO₂-NH₂ catalyst is robust and readily reusable without loss of its catalytic activity.

Full text of DOI:10.1002/aoc.3342

Full paper
Received: 17 May 2015
Accepted: 25 May 2015
Published online in Wiley Online Library: 3 August 2015
(wileyonlinelibrary.com) DOI 10.1002/aoc.3342
Highly dispersed palladium nanoclusters
incorporated in amino-functionalized silica
spheres for the selective hydrogenation of
succinic acid to γ-butyrolactone
Chenjia You, Chi Zhang, Lifang Chen* and Zhiwen Qi*
2 2
Highly dispersed palladium nanoclusters incorporated on amino-functionalized silica sphere surfaces (Pd/SiO -NH ) were
fabricated by a simple one-pot synthesis utilizing 3-(2-aminoethylamino)propyltrimethoxysilane (AAPTS) as coordinating agent.
Uniform palladium nanoclusters with an average size of 1.1 nm can be obtained during the co-condensation of tetraethyl
orthosilicate and AAPTS owing to the strong interaction between palladium species and amino groups in AAPTS. The palladium
particle size can be controlled by addition of AAPTS and plays a significant role in the catalytic performance. The Pd/SiO -NH
2
2
catalyst exhibits high catalytic activity for succinic acid hydrogenation with 100% conversion and 94% selectivity towards
γ-butyrolactone using 1,4-dioxane as solvent at 240°C and 60 bar for 4 h. Moreover, the Pd/SiO -NH catalyst is robust and readily
reusable without loss of its catalytic activity. Copyright © 2015 John Wiley & Sons, Ltd.
2
2
Additional supporting information may be found in the online version of this article at the publisher’s web-site.
Keywords: palladium nanoclusters; amino-functionalized silica spheres; hydrogenation; succinic acid; γ-butyrolactone
Much attention has been focused on controlling the size of palla-
dium particles to improve catalytic efficiency and stability of sup-
Introduction
[7]
γ-Butyrolactone (GBL) is a widely used solvent in polymer, pharma-
ceutical and paint industries. In particular, as a high value-added
chemical intermediate, it can be converted into 1,4-butanediol (BDO),
tetrahydrofuran (THF), N-methyl-2-pyrrolidone and 2-pyrrolidone.
Conventionally, GBL is produced via direct hydrogenation of maleic
ported palladium catalysts. Some essential factors such as nature
of supports, palladium loadings and preparation methods have
considerable influence on the fabricated palladium particle size. As
a result, research is ongoing for facilitating the dispersion of palla-
dium particles by finding different methods to bind catalysts to
suitable supports. Silica is one of the most suitable support materials
[
1]
[
2]
[
3]
anhydride obtained from fossil fuels, which results in several
challenges such as unstable price and environmental issues. As a
consequence, a renewable alternative to the fossil fuel route is of
[8]
with wide accessibility, excellent stability and porosity. Moreover,
silicified organic groups can robustly anchor precious metal ions at
[4]
[9]
much interest. Recently, succinic acid (SUC) has attracted much
attention as a bio-derived platform chemical that can be produced
the surface of silica. For example, P-, S- and N-containing organic
groups covalently linked to siloxane can significantly increase the
interaction between Pd(II) and the silica matrix leading to Pd(II)
[5]
by the fermentation of polysaccharides. Therefore, the hydroge-
nation of SUC to GBL is a promising alternative to the current maleic
anhydride-based method with the increased SUC production in the
biorefinery process.
[10]
immobilized on the support by co-condensation. Then palladium
nanoparticles can be incorporated in the silica matrix by a calcination
or reduction procedure.
Currently, one of main challenges for SUC hydrogenation lies in
the control of catalytic activity to improve the selectivity towards
GBL and avoid consecutive hydrogenation of GBL to THF and
BDO. It is known that palladium-based catalysts have moderate
activity for selective formation of GBL through hydrogenation of
SUC. In general, the catalytic properties of heterogeneous palladium
catalysts strongly depend on the particle size, stability and the
amount of palladium nanoparticles. Nevertheless, palladium nano-
particles supported on inert supports (carbon, alumina, etc.) reported
recently were mostly produced with high loadings of palladium of
approximately 3–5 wt%, which resulted in large sizes and poor
In this article, we report a facile synthesis of aminoethyl- and
aminopropyl-functionalized silica sphere-supported palladium
nanoclusters (Pd/SiO -NH ) with high dispersion. 3-(2-Aminoethy-
2
2
lamino)propyltrimethoxysilane (AAPTS) was used to anchor palla-
dium ions and covalently bond to the silica matrix via siloxane
groups during the sol–gel synthesis. Liquid-phase hydrogenation
of SUC to GBL was carried out over the Pd/SiO -NH catalyst in a
2
2
* Correspondence to: Lifang Chen and Zhiwen Qi, State Key Laboratory of Chemical
Engineering, School of Chemical Engineering, East China University of Science and
Technology, Shanghai 200237, China.
[1,6]
dispersion of the palladium particles.
The excessive price and
high weight percentage of palladium limit its large-scale industrial
applications. Consequently, it remains a continuing challenge to de-
velop palladium-based catalysts with high efficiency and low weight
percentage of metal.
E-mail: lchen@ecust.edu.cn; zwqi@ecust.edu.cn
State Key Laboratory of Chemical Engineering, School of Chemical Engineering,
East China University of Science and Technology, Shanghai 200237, China
Appl. Organometal. Chem. 2015, 29, 653–660
Copyright © 2015 John Wiley & Sons, Ltd.

C. You et al.
batch reactor. The effect of adding AAPTS on the palladium particle
size and catalytic performance was examined. Furthermore, the
energies were calibrated by utilizing the C 1s line as a reference
with an energy of 284.6 eV. The actual palladium contents in the
prepared catalysts were determined using inductively coupled
plasma atomic emission spectrometry (ICP-AES) with an IRIS In-
trepid II XSP instrument (Thermo Electron Corporation). Specific sur-
2 2
reusability and stability of the Pd/SiO -NH catalyst were also
investigated.
2
face area of the samples was measured by N adsorption–
Experimental
desorption at À196°C (Micromeritics ASAP 2010C) using the
Brunauer–Emmett–Teller (BET) method. Before analysis, samples
were previously degassed overnight at 250°C. The dispersion of
Synthesis of catalyst
All chemicals used in this investigation were of analytical grade and
used without further purification. The Pd/SiO -NH catalyst was-
supported palladium nanoparticles was examined by H chemi-
2
sorption with a Micromeritics Autochem 2920 at 70°C. Each sample
2
2
À1
fabricated using a simple one-pot synthesis (Scheme 1). The silica
(about 0.1 g) was reduced in H
2
flow (30 ml min ) at 300°C for 1 h,
[
11]
À1
core was prepared according to the Stöber method. Typically,
ml of deionized water, 1.5 ml of aqueous ammonia and 2 ml of
then flushed with an Ar flow (30 ml min ) at the same temperature
5
for 2 h, and finally cooled to 70°C before H pulses. A H /Pd average
2 2
tetraethyl orthosilicate (TEOS) were successively dissolved in 50
ml of ethanol with vigorous magnetic stirring at room temperature
for 24 h in order to synthesize the silica core. The obtained white
suspension was then heated to 60°C, followed by addition of a mix-
ture of 5 ml of ethanol, 0.2 ml of AAPTS, 1 ml of TEOS and pre-
stoichiometry of 1:2 was used for calculations. Transmission elec-
tronic microscopy (TEM) was performed with a JEM-2100 operated
at 200 kV. Aberration-corrected high-angle annular dark-field scan-
ning transmission electron microscopy (AC-HAADF-STEM) images
were obtained using an FEI Titan Chemi STEM 80-200 at 200 kV with
an incident angle of 22 mrad. Before microscopy examination, each
sample was suspended in ethanol with ultrasonic dispersion for
5–10 min and then a drop of the resulting solution was placed on
a holey carbon film supported by a molybdenum TEM grid.
2 4
designated amounts of potassium tetrachloropalladate (K PdCl ).
After stirring for 1 h, 0.9 g of urea was added to accelerate
co-condensation of TEOS and AAPTS. The suspension was main-
tained under stirring at 60°C for 24 h. After cooling the slurry to
room temperature, the solid was then centrifuged, and washed
with deionized water and ethanol three times each. The final
material was air-dried at 60°C overnight and calcined in air in a muf-
fle oven at 300°C for 4 h. Catalysts with addition of AAPTS (0.1 ml,
Hydrogenation of SUC
Liquid-phase hydrogenation of SUC was performed in a 100 ml
batch reactor (Parr 4590) at 240°C and 60 bar. Amounts of 0.4 g of
SUC, 0.2 g of reduced catalyst and 360 μl of dodecane (internal
standard) were charged into the reaction vessel. 1,4-Dioxane
0.4 ml) and without addition of AAPTS were also fabricated for com-
parison according to the procedure described above (denoted as
Pd/SiO
2
-NH
2
(0.1), Pd/SiO
2
-NH
2
(0.4) and Pd/SiO
2
, respectively). All
À1
the catalysts were reduced in a tube furnace in H
2
(40 ml min
)
(30 ml) was used as solvent for the reaction. The reactor was
atmosphere at 300°C for 3 h and stored under vacuum in the dark
no longer than two weeks before testing. Palladium loading in all
catalysts was fixed at 1 wt% unless otherwise mentioned.
purged five times with hydrogen to remove residual air in the
autoclave, and then pressurized to 30 bar using hydrogen. After
heating the reactor to reaction temperature (240°C), the hydro-
gen pressure was raised to 60 bar. During the reaction process,
the mixture was stirred at 600 rpm in order to avoid mass trans-
fer limitation.
The liquid samples taken from the reactor at regular intervals
were analysed using both a GC instrument (Agilent 7890A)
equipped with a flame ionization detector (SE-54 column, 30 m ×
Catalyst characterization
Thermogravimetry (TG) and derivative TG (DTG) of as-synthesized
samples were carried out with a TA SDT-Q600 thermobalance-TG
À1
analyser under an air flow of 50 ml min with a heating rate of
À1
1
0°C min from room temperature to 800°C. Fourier transform in-
530 mm, thickness 0.5 μm) and an HPLC instrument (Shimadzu
frared (FTIR) spectroscopy was used to qualitatively identify the
chemical functionality of the synthesized catalyst. The transmission
spectra were recorded using a KBr pellet containing 0.1 wt% of
LC-2010AHT) equipped with UV and refractive index detection
(
ODS-SP column at 45°C, 20% methanol and 80% phosphate buffer
À1
solution as mobile phase at a flow rate of 0.4 ml min ). The main
À1
sample, which was measured from 4000 to 400 cm using a Bruker
1
13
product, GBL, was confirmed using GC-MS, H NMR and C NMR
analyses of the crude or purified reaction mixtures (Figs S1 and S2
in the supporting information). By-products consisted of BDO,
THF, γ-hydroxybutyric acid (GHB), propionic acid (PA), butyric acid
Tensor 27 spectrometer. X-ray photoelectron spectroscopy (XPS)
was performed with a Multilab 2000 spectrometer (Thermo VH Sci-
entific) using Al K radiation (1486.6 eV). The aluminium anode was
operated at an accelerating voltage of 15 kV. The pressure in the
(BA), 1-butanol (BuOH) and 1-propanol (ProOH). All the experiments
À9
analysis chamber was maintained at 5 × 10 bar. All binding
were performed twice and the conversion of SUC and the selectivity
to GBL were reproducible.
Results and discussion
TG–DTG
A number of analytical studies have been performed to obtain in-
formation on AAPTS-stabilized palladium nanoclusters incorpo-
rated in core–shell silica spheres. Thermogravimetric analysis was
carried out to investigate the thermal behaviours of AAPTS intro-
duced onto the silica spheres.
Scheme 1. One-pot synthesis of Pd/SiO
ethanol, AAPTS, TEOS, K PdCl , urea, 60°C, 24 h; (ii) calcination and
reduction. (R : –CH CH CH –; R : –CH CH –.)
2 2
-NH catalyst: (i) mixture of
2
4
1
2
2
2
2
2
2
wileyonlinelibrary.com/journal/aoc
Copyright © 2015 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2015, 29, 653–660

Amino-functionalized Pd catalysts for succinic acid hydrogenation
2 2
The TG–DTG curves of the Pd/SiO -NH catalyst are illustrated in
Fig. 1. The weight loss of as-synthesized sample at a temperature
lower than 150°C is around 5–6%, which is attributed to the desorp-
[12]
tion of physically adsorbed water. The second step is assigned to
the elimination of alcohol and chemically adsorbed water at a tem-
perature from 150 to 300°C, which gives rise to a weight loss of ap-
proximately 4%. Also, the weight loss of around 7–9% at 300–450°C
probably results from the decomposition of functional groups
[
13]
incorporated in SiO spheres. The DTG profile with two peaks
2
in this temperature region indicates the stepwise decomposition
of aminoethyl and aminopropyl groups in the Pd/SiO -NH catalyst
2
2
through heating in continuous-flow air. However, the peaks of
aminoethyl and aminopropyl groups are absent in this region for
the Pd/SiO catalyst (Fig. S3). In addition, the as-prepared sample
2
exhibits a weight loss of 2–3% above 450°C, which can be ascribed
[
14]
to the dehydroxylation of the silicate networks. According to the
TG–DTG results and hydrogenated reaction temperature, the func-
tional aminoethyl and aminopropyl groups of the Pd/SiO -NH2
2
2 2 2 2
Figure 2. FTIR spectra of (a) Pd/SiO -NH and (b) used Pd/SiO -NH
catalysts.
catalyst precursors treated at 300°C can efficiently improve the dis-
persion and inhibit the agglomeration of palladium species during
the calcining process.
XPS analysis
The surface composition and chemical state of the Pd/SiO -NH2
2
catalyst are revealed using XPS characterization (Fig. 3) and
substantiating the FTIR results, demonstrating the formation of
chemical bonds among surface C, N, Si, O and Pd atoms of
FTIR spectra
2 2
The functional groups contained in Pd/SiO -NH samples were fur-
ther identified using the FTIR technique (Fig. 2(a)). The main bands
at around 1095 and 950 cm are attributed to Si–O–Si and Si–O–H
the silica shells. The C 1s peak at 284.5 eV is attributed to the
À1
[20]
carbon from C–Si bond.
The other two peaks at 285.2 and
stretching vibrations, respectively. The bands at roughly 787 and
286.2 eV are assigned to the carbon in the neutral and aliphatic
À1
[15]
4
66 cm result from Si–O vibrations. However, only 0.2 ml of
chain (C–C) and C–N bond, which arise from aminoethyl and
AAPTS (AAPTS/TEOS = 1/15, v/v) was added in the catalyst’s
preparation. The N–H stretching vibration at 3280–3370 cm is
overlapped with O–H stretching of the surface silanol groups and
adsorbed water in the range 3200–3600 cm
vibration at 1635 cm is overlapped with the deformational vibra-
tions of adsorbed water molecules at 1638 cm
with the spectrum of the Pd/SiO
C–H bands at 2942, 2873 and 1457 cm associated with a weak
aminopropyl groups and are in agreement with previous
À1
[21]
literature. From the deconvolution of N 1s peaks, the binding
energies located at 399.9 and 400.7 eV indicate the existence of pri-
À1 [16]
.
The N–H bending
2
mary amine groups (–NH ) and secondary amine groups (–NH–),
À1
[22]
respectively.
À1 [17]
.
Compared
The Si 2p spectrum is dominated by a characteristic peak at 103.1
2
catalyst (Fig. S4), the aliphatic
eV ascribed to Si–O bond in SiO
2
and a small peak at 104.1 eV corre-
À1
[20,23]
sponding to Si–C species.
As for the O 1s spectrum of the cata-
À1
N–H band at 1389 cm
are observed for the Pd/SiO -NH2
2
lyst, one peak located at 533.1 eV is obviously related to Si–O–Si
[
18]
catalyst.
1
The peak of C–N stretching vibration at around
[24]
environments in SiO
2
.
The other peak situated at 532.4 eV may in-
À1
097 cm cannot generally be resolved due to its overlap with
clude contributions from Si–OH groups and adsorbed water on the
the strong Si–O–Si stretching band in the range 1000–1200
[25]
catalyst surface. Additionally, a weak peak with binding energy of
À1 [19]
cm
.
These characteristic peaks further confirm the incorpora-
[26]
5
31.6 eV is the characteristic feature from O–Pd bond, which im-
tion of organic species in the shell of the Pd/SiO -NH catalyst.
2
2
plies the existence of a small amount of Pd(II) species (PdO) on the
catalyst surface. The above assignments are consistent with the FTIR
data and the network formation from the co-condensation of TEOS
and AAPTS.
Taking into account that the valence state of oxygen involves
O–Pd bonds on the catalyst surface, it is important to identify
the palladium characteristics. The XPS spectrum for palladium
element presents a doublet of Pd 3d5/2 and Pd 3d at 335.9
3/2
[27]
0
and 341.1 eV, which are assigned to Pd species.
The other
doublet peaks positioned at 337.6 and 342.7 eV can be attrib-
[
26,28]
uted to ‘PdO’ species.
The presence of Pd–O bond in the
sample is attributed to reduced palladium species bonding with
the surface O in the support, which is observed to be atomically
dispersed and catalytically active for aerobic oxidation of allylic
[
29]
alcohols.
As can be seen from the XPS results, it is further
borne out that the aminoethyl and aminopropyl organic groups
are successfully introduced in the catalyst and the palladium
species maintain highly dispersed state during calcination and
reduction.
2 2
Figure 1. TG–DTG curves of Pd/SiO -NH catalyst.
Appl. Organometal. Chem. 2015, 29, 653–660
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

C. You et al.
Physicochemical properties
The physicochemical properties of palladium-based silica catalysts
are summarized in Table 1. The actual palladium loading in the
supported catalysts identified using ICP-AES is close to the nominal
value (1 wt%), which indicates that all palladium precursors are
incorporated in the silica support. The specific surface area of the
2 2 2
Pd/SiO -NH catalyst is more than three times that of the Pd/SiO
catalyst, which is probably because the AAPTS containing
aminoethyl and aminopropyl organic groups affects the physical
properties of the catalysts during the catalyst preparation process.
AAPTS can co-condense with the silica matrix via siloxane groups
to form a three-dimensional netting structure, which results in the
2
increased BET surface area. It is also found from H chemisorption
that the Pd/SiO -NH catalyst possesses higher palladium disper-
2
2
sion than the Pd/SiO catalyst, which results from the enhanced
2
interaction between palladium precursors and supports through
the addition of AAPTS. Meanwhile, the palladium nanoparticles are
not aggregated during subsequent calcination and reduction pro-
cesses. Moreover, as evident from Table 1, the palladium particle size
calculated using H chemisorption is consistent with the average pal-
2
ladium particle size determined using TEM and AC-HAADF-STEM.
The smaller the palladium nanoparticles, the higher should be the
dispersion of the palladium particles on the support. Therefore, the
size of palladium nanoparticles can be efficiently controlled through
introducing AAPTS during the preparation process.
AC-HAADF-STEM and TEM analyses
The morphology of the support and the size of the palladium nano-
particles were investigated using TEM and AC-HAADF-STEM (Figs. 4
and 5). All the silica spheres are approximately 100 nm in diameter
with palladium particles dispersed on their surfaces. It is notable
that the addition of AAPTS has a strong influence on palladium par-
ticle size distributions because palladium nanoclusters in the
Pd/SiO -NH catalyst (Fig. 4(a)) cannot be detected using regular
2
2
TEM. Thus the AC-HAADF-STEM image of the Pd/SiO -NH catalyst
2
2
confirms that palladium nanoclusters are highly dispersed on the
silica spheres with diameters of 0.5–2.0 nm and the average palla-
dium particle size is 1.1 nm from measurement of 200 particles
(
Figs. 4(b) and 6(a)). In contrast, the Pd/SiO catalyst shows a com-
2
paratively larger average size (3.0 nm) and wider size distribution
of palladium nanoparticles. Aggregated palladium nanoparticles
with diameter up to 9.0 nm, as well as palladium nanoclusters
smaller than 2.0 nm in diameter can be observed for the Pd/SiO2
catalyst (Figs. 5 and 6(b)). Obviously, the addition of AAPTS results
in smaller palladium particle size due to the improved palladium
dispersion on the Pd/SiO -NH catalyst surface. Also, the uniform
2
2
Table 1. Physicochemical properties of palladium-loaded silica
catalysts
Sample
Pd
S
BET
Pd
Pd
Pd
2
À1
content (m (g·cat.) ) dispersion particleb particlec
a
(
wt%)
(%)
size (nm) size (nm)
Pd/SiO
Pd/SiO
2
-NH
2
0.95
1.16
120.0
35.8
58.4
37.0
1.9
3.1
1.1
3.0
2
a
Measured by ICP-AES.
b
2
Determined by H chemisorption.
c
Figure 3. XPS binding energy spectra of Pd/SiO
s, (c) Si 2p, (d) O 1s and (e) Pd 3d.
2
-NH
2
catalyst: (a) C 1s, (b) N
Determined from AC-HAADF-STEM and TEM images.
1
wileyonlinelibrary.com/journal/aoc
Copyright © 2015 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2015, 29, 653–660

Amino-functionalized Pd catalysts for succinic acid hydrogenation
2
Figure 5. TEM images of Pd/SiO catalyst.
and the selectivity to GBL increase with increasing amount of
AAPTS (entries 1–3). When the amount of AAPTS increases to 0.4
ml, SUC conversion and GBL selectivity remain at constant values.
However, structural transformation of the catalysts may occur if
[
31]
coordinating agent increases to an excessive amount. Therefore,
the optimized amount of AAPTS is 0.2 ml for the Pd/SiO -NH cata-
2
2
lyst. Furthermore, through optimizing the reaction conditions
entries 2, 4 and 5), the catalytic activity of the Pd/SiO -NH catalyst
(
2
2
leads to a high conversion of SUC of up to 100% and 94% selectivity
towards GBL at 240°C and 60 bar. In addition, the Pd/SiO -NH cat-
2 2
Figure 4. (a) TEM and (b) AC-HAADF-STEM images of Pd/SiO -NH catalyst.
2
2
alyst exhibits higher hydrogenation activity with a turnover fre-
À1
quency (TOF) of 47 h in comparison with the Pd/SiO catalyst
2
palladium nanoclusters with a narrow size distribution in the
Pd/SiO -NH catalyst may be ascribed to the interconnected struc-
À1
(
TOF of 17 h , entries 2 and 6) which can be explained by the
different reaction rate related to the amount of active phase.
[32]
2
2
ture formed from the co-condensation of AAPTS and TEOS, which is
suitable for stabilizing highly dispersed palladium nanoclusters and
preventing their agglomeration during the annealing and reducing
process. Therefore, the well-dispersed palladium nanoclusters in
the Pd/SiO -NH catalyst with more active sites may contribute to
Considering H chemisorption, TEM and AC-HAADF-STEM results,
the dispersion of palladium nanoparticles is markedly improved
2
through introducing AAPTS in the support. Therefore the Pd/SiO
2
-
NH catalyst contains more active sites per mole of palladium avail-
2
2
2
able to participate in the reaction. These energetic sites such as
defects, edges and corners can give rise to efficient dissociation
the hydrogenation of SUC to GBL.
[
33]
of H
SUC using the Pd/SiO
as that using the Pd/SiO
selectivity to GBL rises from 37 to 94% probably due to the basic
amino groups in the Pd/SiO -NH catalyst, which suppress the forma-
tion of by-product PA, as well as facilitating the reactant adsorption
2
from palladium surfaces.
-NH catalyst is more than twice as high
catalyst with 45% SUC conversion. The
As a result, the conversion of
2
2
Catalytic activity
2
The selective hydrogenation of SUC to GBL is a particularly attrac-
[
30]
tive approach over palladium-based catalysts.
The palladium-
2
2
loaded silica catalysts fabricated for hydrogenation of SUC as a
model reaction were used to elucidate the marked effect of adding
AAPTS (Table 2). It is interesting to find that the conversion of SUC
[34]
to and product desorption from the palladium active sites.
Also, compared with 5.0% Pd/A450 and 5.0% Pd/AX850 catalysts,
[6]
Appl. Organometal. Chem. 2015, 29, 653–660
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

C. You et al.
the Pd/SiO
2
-NH
2
catalyst also displays the highest hydrogenation ac-
tivity with respect to TOF value under similar reaction conditions,
demonstrating that it is a competitive candidate for hydrogenation
of SUC to GBL.
The conversion of SUC and the selectivity towards main products
2 2
over the Pd/SiO -NH catalyst as a function of reaction time are il-
lustrated in Fig. 7. It can be seen that the conversion of SUC reaches
approximately 98% after 2 h, yielding mainly GBL (86% selectivity)
together with GHB (11%) and PA (3%) in solution. When the reac-
tion time reaches 4 h, SUC can be totally converted and the highest
yield for GBL is observed (94%; Table 2). Moreover, the selectivity to
GHB and PA is 2 and 4%, respectively. The decline in GHB selectivity
indicates that GHB is an unstable intermediate and is further
dehydrated to form target product GBL. Then with an increase of
reaction time, the selectivity to GBL decreases slowly and GBL is
partially hydrogenated to BDO and THF. Meanwhile, more by-
products, such as BA, BuOH, PA and ProOH, through hydrogenolysis
of GBL are also detected. After 30 h of reaction, GBL is still the main
product and the yield for GBL, BDO and THF reaches 77, 9 and 2%,
respectively. Therefore, a reaction route for SUC hydrogenation can
be proposed, as depicted in Scheme 2. SUC is partially hydroge-
nated to give GHB, along with the formation of PA and ProOH,
[
35]
while GHB is then dehydrated to the target GBL. Further hydro-
genation and hydrogenolysis reactions may occur with increased
reaction time. The final products are formed through competitive
adsorption and dissociation of various intermediates on the surface
[
36]
of the Pd/SiO
To investigate the durability and reusability of the Pd/SiO
catalyst, recycle tests for hydrogenation of SUC were carried out.
The used Pd/SiO -NH catalyst was filtered after reaction and then
washed with distilled water and ethanol, followed by drying at
0°C for 12 h and reducing at 300°C for 2 h. Catalyst reusability
2 2
-NH catalyst.
-NH
2 2
2
2
6
was investigated by using the recovered solid catalyst under the
optimized reaction conditions, identical to those for the reaction
with fresh sample. As shown in Fig. 8, the conversion of SUC is
still more than 95% after three times of recycling. Moreover, the
selectivity to GBL is essentially constant within experiential errors.
2 2
The Pd/SiO -NH catalyst can be reused without significant loss of
its catalytic activity because palladium species are anchored on
Figure 6. (a) Pd/SiO
distributions determined from measurement of 200 particles.
2 2 2
-NH catalyst and (b) Pd/SiO catalyst particle size
a
Table 2. Catalytic activity of palladium-loaded silica catalysts for hydrogenation of succinic acid to γ-butyrolactone
Entry
Catalyst
Reaction
temperature (°C)
Reaction
pressure (bar)
Conversion
of SUC (mol%)
Selectivity
to GBL (mol%)
Yield
of GBL (mol%)
TOF
(h )
À1 b
c
1
2
3
4
5
6
7
8
9
Pd/SiO
Pd/SiO
Pd/SiO
2
-NH
-NH
-NH
2
(0.1)
240
240
240
220
240
240
240
240
240
60
60
60
60
40
60
60
60
60
75
100
100
90
80
94
95
90
84
37
0
60
94
95
81
74
17
0
34
47
45
41
40
17
—
10
9
2
2
d
2
2
(0.4)
Pd/SiO
Pd/SiO
2
-NH
-NH
2
2
2
88
Pd/SiO
SiO -NH
5.0% Pd/A450-com
2
45
2
2
0
e
85
80
65
68
51
e
5.0% Pd/AX850-com
79
a
Reaction conditions: 0.2 g catalyst, 0.4 g succinic acid, 360 μl dodecane, 30 ml 1,4-dioxane, 4 h.
b
Calculated as moles of succinic acid converted per mole of palladium per hour.
Addition of AAPTS: 0.1 ml.
c
d
Addition of AAPTS: 0.4 ml.
e
Reaction conditions: 0.2 g catalyst, 0.5 g succinic acid, 50 ml 1,4-dioxane, 4 h.
wileyonlinelibrary.com/journal/aoc
Copyright © 2015 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2015, 29, 653–660

Amino-functionalized Pd catalysts for succinic acid hydrogenation
2 2
Figure 8. Conversion of SUC and selectivity to GBL over Pd/SiO -NH
catalyst versus the number of recycling runs at 240°C and 60 bar for 4 h.
Error bars represent one standard deviation.
Conclusions
Palladium nanoclusters incorporated on AAPTS-functionalized
core–shell silica nanospheres (Pd/SiO -NH ) were fabricated using
2 2
a simple one-pot synthesis, and showed high activity and selectivity
for liquid-phase hydrogenation of SUC to GBL. Palladium particle
size has a significant effect on catalytic performance and can be
controlled by addition of AAPTS, which greatly strengthens the in-
teraction between palladium species and supports, leading to a
Figure 7. Hydrogenation of SUC in the presence of Pd/SiO
2 2
-NH catalyst at
240°C and 60 bar. Evolution of conversion of reactant and selectivity to
high dispersion of palladium nanoclusters. Thus the Pd/SiO -NH2
2
products as a function of time. Error bars represent one standard deviation.
catalyst contains smaller palladium nanoclusters (1.1 nm) with
more active sites than the Pd/SiO catalyst and can efficiently im-
2
prove the reaction rate of SUC hydrogenation. It leads to complete
conversion of SUC and a high selectivity (94%) to GBL. Furthermore,
the Pd/SiO -NH catalyst can be recovered and reused without
2
2
loss of catalytic activity, demonstrating its potential in industrial
applications.
Acknowledgments
This work was supported by the National Natural Science Founda-
tion of China (NSFC 21476084), the National High Technology
Research and Development (863 Program 2012AA061601), the
Specialized Research Fund for the Doctoral Program of Higher Edu-
cation (SRFDP20120074110008), the Open Project of Shanghai Key
Laboratory of Molecular Catalysis and Innovative Materials
(2014MCIMKF02) and the Fundamental Research Funds for the
Central Universities of China.
Scheme 2. Reaction route for hydrogenation of succinic acid in the
2 2
presence of Pd/SiO -NH catalyst.
References
silica spheres through aminoethyl and aminopropyl groups. Con-
sidering the FTIR spectrum of used Pd/SiO -NH catalyst, the
amino-functionalized structure is still maintained in the catalyst
after reaction (Fig. 2(b)). In the XRD patterns of Pd/SiO -NH and
Pd/SiO catalysts before and after reaction (Fig. S5), the lack of
[
1] a) R. Luque, J. H. Clark, K. Yoshida, P. L. Gai, Chem. Commun. 2009, 35,
305; b) U. R. Pillai, E. Sahle-Demessie, D. Young, Appl. Catal. B 2003,
3, 131.
2
2
5
4
2
2
[
2] a) P. Lanzafame, G. Centi, S. Perathoner, Chem. Soc. Rev. 2014, 43, 7562;
b) T. M. Lammens, M. C. R. Franssen, E. L. Scott, J. P. M. Sanders, Green
Chem. 2010, 12, 1430; c) G. Budroni, A. Corma, J. Catal. 2008, 257, 403.
3] U. R. Pillai, E. Sahle-Demessie, Chem. Commun. 2002, 422.
4] I. Bechthold, K. Bretz, S. Kabasci, R. Kopitzky, A. Springer, Chem. Eng.
Technol. 2008, 31, 647.
2
palladium phase reveals that palladium nanoclusters and nanopar-
ticles are still uniformly dispersed on the silica spheres after SUC hy-
drogenation. Therefore, the Pd/SiO -NH catalyst is demonstrated
[
[
2
2
to be stable and reusable in the liquid-phase hydrogenation of
SUC to GBL.
[
5] B. Tapin, F. Epron, C. Especel, B. K. Ly, C. Pinel, M. Besson, ACS Catal.
2013, 3, 2327.
Appl. Organometal. Chem. 2015, 29, 653–660
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

C. You et al.
[
[
6] a) U. G. Hong, S. Hwang, J. G. Seo, J. Lee, I. K. Song, J. Ind. Eng. Chem.
[22] Y. G. Ko, H. J. Lee, H. C. Oh, U. S. Choi, J. Hazard. Mater. 2013, 250–251, 53.
[23] S. W. Lin, D. H. Chen, Small 2009, 5, 72.
[24] J. Zhang, L. R. Grabstanowicz, S. Gao, N. S. Hosmane, B. Huang, Y. Dai,
D. Liu, T. Xu, Catal. Sci. Technol. 2012, 2, 390.
2
2
011, 17, 316; b) U. G. Hong, J. Lee, S. Hwang, I. K. Song, Catal. Lett.
011, 141, 332.
7] a) B. Karimi, S. Abedi, J. H. Clark, V. Budarin. Angew. Chem. Int. Ed. 2006,
5, 4776; b) F. Durap, Ö. Metin, M. Aydemir, S. Özkar, Appl. Organometal.
4
[25] S. Ntais, A. Siokou, J. Mol. Catal. A 2006, 245, 87.
Chem. 2009, 23, 498.
8] M. Opanasenko, P. Štěpnička, J. Čejka, RSC Adv. 2014, 4, 65137.
9] C. Wang, L. Chen, Z. Qi, Catal. Sci. Technol. 2013, 3, 1123.
10] a) Z. Yinghuai, S. C. Peng, A. Emi, S. Zhenshun, Monalisa, R. A. Kemp,
Adv. Synth. Catal. 2007, 349, 1917; b) A. Mori, T. Mizusaki, M. Kawase,
T. Maegawa, Y. Monguchi, S. Takao, Y. Takagi, H. Sajiki, Adv. Synth.
Catal. 2008, 350, 406; c) H. Zhao, G. Ding, L. Xu, M. Cai, Appl.
Organometal. Chem. 2011, 25, 871.
[26] M. C. Militello, S. J. Simko, Surf. Sci. Spectra 1994, 3, 395.
[27] a) K. R. Priolkar, P. Bera, P. R. Sarode, M. S. Hegde, S. Emura,
R. Kumashiro, N. P. Lalla, Chem. Mater. 2002, 14, 2120; b) Y. Wang,
J. Yao, H. Li, D. Su, M. Antonietti, J. Am. Chem. Soc. 2011, 133, 2362.
[28] V. Z. Radkevich, T. L. Senko, K. Wilson, L. M. Grishenko, A. N. Zaderko,
V. Y. Diyuk, Appl. Catal. A 2008, 335, 241.
[29] a) S. F. J. Hackett, R. M. Brydson, M. H. Gass, I. Harvey, A. D. Newman,
K. Wilson, A. F. Lee, Angew. Chem. Int. Ed. 2007, 46, 8593; b)
E. J. Peterson, A. T. DeLaRiva, S. Lin, R. S. Johnson, H. Guo, J. T. Miller,
J. H. Kwak, C. H. F. Peden, B. Kiefer, L. F. Allard, F. H. Ribeiro,
A. K. Datye, Nat. Commun. 2014, 5, 4885.
[30] Z. Shao, C. Li, X. Di, Z. Xiao, C. Liang, Ind. Eng. Chem. Res. 2014, 53, 9638.
[31] X. Wang, L. Chen, M. Shang, F. Lin, J. Hu, R. M. Richards, Nanotechnology
2012, 23, 294010.
[32] C. Franch, R. G. H. Lammertink, L. Lefferts, Appl. Catal. B 2014, 156–157,
166.
[
[
[
[
[
11] W. Stöber, A. Fink, E. Bohn, J. Colloid Interface Sci. 1968, 26, 62.
12] L. Chen, J. Hu, Z. Qi, Y. Fang, R. Richards, Ind. Eng. Chem. Res. 2011, 50,
1
3642.
13] X. Wang, K. S. K. Lin, J. C. C. Chan, S. Cheng, J. Phys. Chem. B 2005, 109,
763.
[
[
1
14] S. F. Lee, X. M. Zhu, Y. X. J. Wang, S. H. Xuan, Q. You, W. H. Chan,
C. H. Wong, F. Wang, J. C. Yu, C. H. K. Cheng, K. C. F. Leung, ACS Appl.
Mater. Interfaces 2013, 5, 1566.
[
[
15] T. Yu, R. Yang, S. Xia, G. Li, C. Hu, Catal. Sci. Technol. 2014, 4, 3159.
16] a) D. Coutinho, S. Madhugiri, K. J. Balkus Jr, J. Porous Mater. 2004, 11,
[33] E. Ghedini, F. Menegazzo, M. Signoretto, M. Manzoli, F. Pinna, G. Strukul,
J. Catal. 2010, 273, 266.
2
39; bb) L. Chen, J. Hu, R. Richards, J. Am. Chem. Soc. 2009, 131, 914.
17] a) X. Wang, J. C. C. Chan, Y. H. Tseng, S. Cheng, Micropor. Mesopor. Mater.
006, 95, 57; b) E. Borfecchia, K. A. Lomachenko, F. Giordanino,
H. Falsig, P. Beato, A. V. Soldatov, S. Bordiga, C. Lamberti, Chem. Sci.
015, 6, 548.
[34] Y. Chen, Z. Guo, T. Chen, Y. Yang, J. Catal. 2010, 275, 11.
[35] L. Corbel-Demailly, B. K. Ly, D. P. Minh, B. Tapin, C. Especel, F. Epron,
A. Cabiac, E. Guillon, M. Besson, C. Pinel, ChemSusChem 2013, 6, 2388.
[36] R. M. Deshpande, V. V. Buwa, C. V. Rode, R. V. Chaudhari, P. L. Mills, Catal.
Commun. 2002, 3, 269.
[
2
2
[
18] a) Q. Wei, Z. R. Nie, Y. L. Hao, L. Liu, Z. X. Chen, J. X. Zou, J. Sol-Gel Sci.
Technol. 2006, 39, 103; b) Y. K. Lu, X. P. Yan, Anal. Chem. 2004, 76, 453.
19] P. Wang, Q. Lu, J. Li, Mater. Res. Bull. 2010, 45, 129.
20] P. Laoharojanaphand, T. J. Lin, J. O. Stoffer, J. Appl. Polym. Sci. 1990, 40, 369.
21] a) A. Durdureanu-Angheluta, A. Dascalu, A. Fifere, A. Coroaba, L. Pricop,
H. Chiriac, V. Tura, M. Pinteala, B. C. Simionescu, J. Magn. Magn. Mater.
[
[
[
Supporting information
Additional supporting information may be found in the online
2012, 324, 1679; b) S. Karthikeyan, C. Anandan, J. Subramanian,
version of this article at the publisher’s web site.
G. Sekaran, RSC Adv. 2013, 3, 15044.
wileyonlinelibrary.com/journal/aoc
Copyright © 2015 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2015, 29, 653–660