Palladium-nanoparticles on end-functionalized poly(lactic acid)-based stereocomplexes for the chemoselective cinnamaldehyde hydrogenation: Effect of the end-group

Article abstract of DOI:10.1016/j.jcat.2015.07.012

Differently carboxylic acid end-functionalized poly(lactic acid) (PLA)-based stereocomplexes were used as polymer support to stabilize Pd-nanoparticles (NPs) generated by the metal vapor synthesis technique. The dispersion of Pd was strongly dependent on

Full text of DOI:10.1016/j.jcat.2015.07.012

Journal of Catalysis 330 (2015) 187–196
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Palladium-nanoparticles on end-functionalized poly(lactic acid)-based
stereocomplexes for the chemoselective cinnamaldehyde
hydrogenation: Effect of the end-group
b,
b
c
c
Werner Oberhauser a, , Claudio Evangelisti , Ravindra P. Jumde , Giorgio Petrucci , Mattia Bartoli ,
⇑
⇑
Marco Frediani c, Matteo Mannini d, Laura Capozzoli e, Elisa Passaglia f, Luca Rosi c
a Istituto di Chimica dei Composti Organometallici (CNR-ICCOM), Via Madonna del Piano 10, 50019 Sesto Fiorentino, Italy
b Istituto di Scienze e Tecnologie Molecolari (CNR-ISTM), Via Fantoli 16/15, 20138 Milano, Italy
c Dipartimento di Chimica, Università Degli Studi di Firenze, via della Lastruccia 13, 50019 Sesto Fiorentino, Italy
d Laboratory of Molecular Magnetism, Dipartimento di Chimica Ugo Sciff, Università Degli Studi di Firenze and INSTM RU of Firenze, Via della Lastruccia 3,
50019 Sesto Fiorentino, Italy
e Centro di Microscopie Elettroniche (Ce.M.E.), Area di Ricerca CNR di Firenze, Via Madonna del Piano 10, 50019 Sesto Fiorentino, Italy
f CNR-ICCOM, UOS Pisa, Area della Ricerca, Via Moruzzi 1, 56124 Pisa, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Differently carboxylic acid end-functionalized poly(lactic acid) (PLA)-based stereocomplexes were used
as polymer support to stabilize Pd-nanoparticles (NPs) generated by the metal vapor synthesis technique.
The dispersion of Pd was strongly dependent on the end-group present in the polymer structure, as
shown by HRTEM measurements. 2,20-Bipyridine- and pyridine-based stereocomplexes showed high
metal dispersion (i.e. well-separated Pd-NP size of 2.0 0.6 nm), whereas stereocomplexes bearing ben-
zyl and carboxylic acid end groups exhibited strong Pd-NPs aggregation. The heterogenous catalysts were
employed to chemoselectively hydrogenate the C@C double bond in cinnamaldehyde, showing for
Pd-NPs, stabilized by the 2,20-bipyridine-modified polymer support, the best performance in terms of
chemoselectivity (99%) and recyclability in THF solution. Even under bulk cinnamaldehyde hydrogena-
tion conditions, chemoselectivity for 3-phenylpropanal of 90% at 88% conversion was obtained.
Ó 2015 Elsevier Inc. All rights reserved.
Received 13 March 2015
Revised 1 July 2015
Accepted 4 July 2015
Keywords:
Poly(lactic acid)
Stereocomplex
Pd-nanoparticles
Hydrogenation
Cinnamaldehyde
1. Introduction
the favored back-bonding interaction of the Pd-NPs with the
p⁄CO-orbital [8].
The selective hydrogenation of the C@C double bond in
,b-unsaturated carbonyl compounds is an important organic
Alternatively to carbon supports, organic polymers [9,10], such
as polyamines [11] and polyketones [12], endowed with functional
groups that interact with the metal-NPs’ surface, have shown to be
a promising alternative support material for metal Pd-NPs, used in
the hydrogenation of the structure-sensitive cinnamaldehyde to
3-phenylpropanal. Chemoselectivity in the range between 84%
and 88% has been observed with the latter catalysts [11,12].
Recently we introduced the stereocomplex of 2,20-bipyridi
ne-functionalized poly(lactic acid) PLA as organic support for
Pd-NPs, which selectively catalyzed the partial hydrogenation of
phenylacetylene and diphenylacetylene [13]. The stereocomplex,
which is formed upon hydrogen bond interactions between l-
and d-PLA [14], is much more resistant against hydrolytic [15,16]
and thermal degradation [17] compared to l- or d-PLA.
a
transformation leading to the saturated counterparts which are
widely applied in synthetic and pharmaceutical chemistry [1–3].
Metallic palladium embedded in a carbon support is generally used
to catalyze the latter substrate conversion, which is not always
chemoselective [4–7]. The support, which functions as macroli-
gand, notably influences the electron density of the surface atoms
of the Pd nanoparticles (NPs), thus allowing to promote a specific
chemoselectivity. The graphitic structure of carbon, for instance,
increases the charge density on the NP-surface [8]. As a conse-
quence, the binding energy of the C@C bond in
a,b-unsaturated
carbonyl compounds decreases due to repulsive four-electron
interactions, whereas the C@O bond hydrogenation is fostered by
Herein we report the synthesis of differently end-functionalized
PLA-based stereocomplexes, which were subsequently employed
to stabilize Pd-NPs generated by metal vapor synthesis (MVS) tech-
nique [18,19]. The obtained heterogeneous polymer-based
⇑
Corresponding authors.
E-mail addresses: werner.oberhauser@iccom.cnr.it (W. Oberhauser), claudio.
evangelisti@istm.cnr.it (C. Evangelisti).
http://dx.doi.org/10.1016/j.jcat.2015.07.012
0021-9517/Ó 2015 Elsevier Inc. All rights reserved.

188
W. Oberhauser et al. / Journal of Catalysis 330 (2015) 187–196
catalysts were used in the chemoselective C@C double bond hydro-
genation of cinnamaldehyde to 3-phenylpropanal. Pd-NP disper-
sion and chemoselectivity of the catalytic reactions were notably
influenced by the nature of the end-group present in the polymer
back bond.
atoms (SMA) solution was heated over a heating plate in a porce-
lain crucible in the presence of aqua regia (2.0 mL) for six times fol-
lowed by dissolving the solid residue in 0.5 M aqueous HCl.
Powder X-ray diffraction (PXRD) experiments were carried out
at room temperature with a PANalytical X’PERT PRO powder
diffractometer, employing Cu K
a (k = 1.5418 Å) radiation and a
parabolic MPD-mirror. Diffractograms were acquired in the 2
H
2. Experimental
range from 5.0° to 80.0°, applying a step size of 0.1050° and a
counting time of 441 s.
2.1. Materials and apparatus
Thermogravimetric (TG) analyses were carried out under nitro-
gen atmosphere using a Seiko EXSTAR 7200 TG/DTA instrument.
TG spectra were collected on samples of 5–10 mg in the tempera-
ture range from 30 to 700 °C (N2 flow = 200 mL/min) with a heat-
ing rate of 10 °C/min. The onset temperature of degradation
(Tonset) is defined as the temperature that corresponds to a mass
loss of 5%. The rate inflection temperature for the different degra-
dation steps was extracted from derivative TG (DTG) curves.
X-ray photoelectron spectroscopy (XPS) analyses were
performed in a UHV chamber equipped with an X-ray source
Tin octanoate (Sn(Oct)2) 4-hydroxymethylpyridine, benzylalco-
hol, 4-methyl-40-(hydroxymethyl)-2,20-bipyridine and l-lactide
and d-lactide were purchased from Aldrich. l-lactide and d-lactide
were sublimated before utilization and stored thereafter at 4 °C
under nitrogen atmosphere. Solvents such as n-hexane, CHCl3,
CDCl3 and HPLC-grade THF were purchased from Aldrich and used
without further purification. Mesitylene and n-hexene were puri-
fied by conventional methods, distilled and stored under nitrogen.
The co-condensation of palladium and the appropriate solvent was
carried out in a previously described static reactor [18,20].
1H NMR spectra were recorded on a Bruker Avance DPX 300
spectrometer, measuring at 300.13 MHz, while 13C{1H} NMR spec-
tra were obtained with a Bruker Avance DRX-400 spectrometer,
acquiring at 100.62 MHz.
(nonmonochromatized Mg K
a
source, 1253.6 eV) and an
hemispherical analyzer by VSW mounting a 16-channel detector.
The X-ray source, mounted at 54.44° with respect to the analyzer,
was operated at a power of 100 W (10 kV and 10 mA). The analyzer
has been equipped with a differential pumping stage in order to
operate up to a pressure of 5 ꢁ 10ꢀ8 mbar. XPS spectra were
measured at normal emission with a fixed pass energy of 22 eV
for the high resolution Pd3d region spectra and 44 eV in the survey
scans. All spectra were referenced to the C1s peak due to adventi-
tious carbon at 284.5 eV. The inelastic background in the spectra
was subtracted by means of the Shirley method [21]. Data analysis
was based on a standard method for deconvolution using mixed
Gaussian and Lorentzian line shapes for each component in the
spectra.
ATR-IR spectra were recorded on a Shimadzu model IR-Affinity
apparatus, equipped with a Golden Gate single reflection diamond
ATR accessory.
Gel-Permeation Chromatography (GPC) was carried out with a
Waters Binary HPLC 1525 pump, a manual injector with a six
way valve and a 200 lL loop, three Shodex KF-802, KF-803 and
KF-804 columns connected in series (length: 300 mm each, inner
diameter: 8.0 mm, 24,500 theoretical plates, exclusion limit for
polystyrene (PS) up to 400,000 g molꢀ1; a refraction index (RI)
detector (Optilab T-rEXTM, Wyatt Technology) and a UV detector
(Waters mod. 2489)). HPLC-grade THF with a water content of
maximal 0.1 vol% was used as eluent at a constant flow of
1.0 mL minꢀ1, keeping the columns at 30.0 °C with a thermostat.
The GPC system was calibrated using PS as standard. Samples were
prepared by dissolving 5.0 mg of analyte in 1.0 mL of eluent. The
Environmental scanning electron microscopy (ESEM) analyses
were carried out on a FEI ESEM QUANTA 200 apparatus with a
tungsten source using a gaseous secondary electron detector
(GSED) with a 500 lm aperture. The images were collected with
a magnification of 3000ꢁ, applying 6 kV, a chamber pressure of
3 Torr and a working distance of 7.0 mm.
obtained solution was filtered through a 0.2
injected.
l
m PTFE filter and
2.2. Synthesis of end-functionalized macroligands l-PLAR and d-PLAR
GC analyses were performed with a Shimadzu 2010 gas chro-
matograph equipped with a flame ionization detector and a 30 m
A general synthetic procedure for the synthesis of l-PLAR and
d-PLAR (R = 4-methyl-40-(hydroxymethyl)-2,20-bipyridine (BiPy),
4-(hydroxymethyl)-pyridine (Py) and benzyl (Bn)) is described as
follows. In a Schlenk tube, l-lactide (d-lactide) (4.00 g, 28.00 mmol)
was heated at 135 °C under a nitrogen atmosphere in the presence
of Sn(Oct)2 (56.3 mg, 0.139 mmol) and ROH (0.400 mmol) for 3 h.
Afterward, the reaction mixture was allowed to cool to room tem-
perature and residual crystalline l- or d-lactide (ca. 1.0%), which
sublimated during reaction, was removed mechanically from the
Schlenk tube. The crude reaction product was then dissolved in
CHCl3 (20.0 mL) and precipitated upon addition of n-hexane
(30.0 mL), giving an off-white solid powder, which was separated
from solution by filtration, washed several times with n-hexane
and dried by vacuum at room temperature for 12 h.
(0.25 mm i.d., 0.25
column.
lm film thickness) VF-WAXms capillary
GC–MS analyses were performed with a Shimadzu QP5000
apparatus, equipped with a 30 m (0.32 mm i.d., 0.50
ness) CP-WAX 52CB WCOT-fused silica column.
lm film thick-
High resolution transmission electron microscopy (HRTEM)
analyses of the supported Pd-NPs were carried out with a ZEISS
LIBRA 200FE HRTEM instrument, equipped with a FEG source oper-
ating at 200 kV, in column second-generation omega filter for
energy selective spectroscopy (EELS) and energy selective imaging
(ESI), HAADF STEM facility, EDS probe for chemical analysis, inte-
grated tomographic HW and SW. The samples were dispersed by
sonication in a 1:1 solvent mixture of isopropanol/CHCl3 and a
drop of the obtained solution was deposited on a holey-carbon film
supported on a copper TEM grid of 300 mesh. Histograms of the
particle size distribution were obtained by counting at least 500
particles. The mean particle diameter (dm) was calculated by using
In case R = H (i.e. carboxylic acid end group) the synthetic pro-
tocol was modified as follows. To a two necked flask was added
l-lactide (d-lactide) (2.50 g, 18.00 mmol), Sn(Oct)2 (90.00 mg,
0.222 mmol), water (5.00 lL, 0.270 mmol) and anhydrous toluene
(20.0 mL) (i.e. in order to control the amount of water present in
the reaction mixture). The reaction mixture was then refluxed
under a nitrogen atmosphere for 24 h. Afterward, the obtained
solution was cooled to room temperature causing the precipitation
of a white solid. The crude product was dissolved in CH2CL2
(20.0 mL), passed through a paper filter and to the clear solution
P
P
the formula dm
with diameter di.
Inductively coupled plasma–optical emission spectrometry
(ICP–OES) was carried out with an iCAP 6200 Duo upgrade,
Thermofisher instrument. A sample (0.5 mL) of Pd-solvated metal
=
dini/ ni, where ni is the number of particles

W. Oberhauser et al. / Journal of Catalysis 330 (2015) 187–196
189
was added n-hexane (30.0 mL), causing the precipitation of an
off-white solid which was successively separated, washed with
n-hexane (3 ꢁ 10.0 mL) and dried by vacuum at room temperature
for 12 h.
same synthesis procedure as reported above and by using
10.7 mL of Pd SMA (15.9 mg).
2.5. Synthesis of Pd(OAc)2(d-PLABiPy
)
Yield: l-PLABiPy (3.70 g, 90%), d-PLABiPy (3.80 g, 95%); l-PLAPy
(3.80 g, 95%), d-PLAPy (3.70 g, 93%); l-PLABn (3.80 g, 96%), d-PLABn
(3.40 g, 90%); l-PLAH (2.40 g, 96%), d-PLAH (2.40 g, 96%). The molar
weight of the polymers (Mn) and the poly-dispersity index (PDI)
were determined by GPC, using a refraction index (RI) detector.
Mn (PDI): l-PLABiPy, 10,570 g/mol (1.81); d-PLABiPy, 10,600 g/mol
(1.90); l-PLAPy, 9615 g/mol (1.22); d-PLAPy, 10,556 g/mol (1.37);
l-PLABn, 10,800 g/mol (1.62); d-PLABn, 10,590 g/mol (1.90); l-PLAH,
9460 g/mol (1.87); d-PLAH, 8170 g/mol (1.90).
To a solution of d-PLABiPy (300.0 mg, 0.0284 mmol) in CH2Cl2
(10.0 mL) was added Pd(OAc)2 (OAc = acetate) (6.46 mg,
0.0288 mmol) and the resulting solution was stirred under a nitro-
gen atmosphere at room temperature for 6 h. Afterward the sol-
vent was removed completely by vacuum and the slightly yellow
solid was washed with diethyl ether (2 ꢁ 5.0 mL). Yield:
275.6 mg, 90%. 1H NMR (300.13 MHz, CD2Cl2, ppm) = d 1.58 (d,
3JHH = 7.0 Hz, 442H, CH3), 2.01 (br s, 6H, OAc), 2.53 (s, 3H,
3
1H NMR data for l/d-PLAPy [22] and l/d-PLABn [23] correspond to
those reported in the literature.
BiPyCH3CH2O), 4.38 (q, JHH = 7.0 Hz, 1H, CH(terminal)), 5.19 (q,
3JHH = 7.0 Hz, 147H, CH), 5.24 (s, 2H, BiPyCH3CH2O), 7.31 (br. s.
1H, ArH), 7.38 (br. s, 1H, ArH), 7.90 (br s, 2H, ArH), 8.22 (br. s.
1H, ArH), 8.39 (br. s. 1H, ArH).
NMR data for l/d-PLABiPy: 1H (300.13 MHz, CDCl3, ppm) = d 1.58
3
(d, JHH = 7.0 Hz, 442H, CH3), 2.47 (s, 3H, BiPyCH3CH2O), 4.38 (q,
3
3JHH = 7.0 Hz, 1H, CH(terminal)), 5.17 (q, JHH = 7.0 Hz, 147H, CH),
5.26 (s, 2H, BiPyCH3CH2O), 7.18 (br. s. 1H, ArH), 7.28 (1H, ArH,
overlapped with CHCl3), 8.25 (s, 1H, ArH), 8.35 (s, 1H, ArH), 8.57
(br. s. 1H, ArH), 8.70 (br. s. 1H, ArH). 13C{1H} (100.62 MHz, CDCl3,
ppm) = d 16.64 (s, CH3), 20.52 (s, CH3(terminal)), 21.20 (s,
BiPyCH3CH2O), 65.37 (s, BiPyCH3CH2O), 66.71 (s, CH(terminal)),
69.01 (s, CH), 119.42 (s, ArC), 121.70 (s, ArC), 122.07 (s, ArC),
125.02 (s, ArC), 145.05 (s, ArC), 149.01 (s, ArC), 149.50 (s, ArC),
169.60 (s, COOC), 175.34 (s, COOR).
2.6. Reduction of Pd(OAc)2(d-PLABiPy) and trans-[Pd(OAc)2(d-PLAPy)2]
with hydrogen
Deaerated
solutions
of
Pd(OAc)2(d-PLABiPy
)
and
trans-[Pd(OAc)2(d-PLAPy)2] [24] (300.0 mg) in CH2Cl2 (7.0 ml) were
transferred to Teflon-coated stainless steel autoclaves (80.0 mL)
which were sealed and pressurized with hydrogen (15.0 bar). The
autoclaves were then stirred for 12 h at room temperature,
followed by releasing the excess hydrogen and transferring the
black solutions to round-bottom flasks. The solvent was in both
cases completely evaporated, and the dark brown solids were
washed with diethyl ether (2 ꢁ 5.0 mL) and vacuum-dried. Yield:
85–90%.
NMR data for l/d-PLAH: 1H (400.13 MHz, CDCl3, ppm) = d 1.57 (d,
3JHH = 7.1 Hz, 390H, CH3), 4.38 (br s, 1H, CH(terminal)), 5.18 (q,
3JHH = 7.1 Hz, 130H, CH). 13C{1H} (100.62 MHz, CDCl3, ppm) = d
16.64 (s, CH3), 20.26 (s, CH3(terminal)), 65.85 (s, CH (terminal)),
66.72 (s, CH(terminal)), 69.01 (s, CH), 169.61 (s, COOC), 172.98 (s,
COOH), 175,15 (s, COOC(terminal)).
2.7. Catalytic cinnamaldehyde hydrogenation reactions
2.3. Synthesis of the stereocomplexes LR
2.7.1. Hydrogenation reactions in THF solution
Pd@LR (21.0 mg, 0.0012 mmol Pd) was added to a Teflon-coated
stainless steel autoclave (80.0 mL) equipped with magnetic stirrer,
temperature and pressure controller, which was sealed and evacu-
Solutions of l-PLAR and d-PLAR (800.0 mg) in CH2Cl2 (8.0 mL)
with identical R, were mixed at room temperature under vigorous
stirring for half an hour. Afterward the obtained clear solution was
concentrated to dryness by means of a vacuum pump at room tem-
perature. The obtained off-white solids were used without further
purification. Yield: 90–95%.
ated. A solution of cinnamaldehyde (141.0 lL, 1.12 mmol) in
deaerated THF (10.0 mL) was introduced in the autoclave by
suction. Then the autoclave was placed into a stirred oil bath which
was heated to 60 °C. Once the latter temperature was reached, the
autoclave was pressurized with hydrogen (150 psi) and stirred at
700 rpm for the stabilized reaction time. The autoclave was then
successively cooled to 10 °C by a water/ice bath, the hydrogen
gas vented off and the heterogeneous catalyst separated from
THF solution by decantation. The solid catalyst was washed with
fresh THF (5.0 mL) and the combined THF solutions were analyzed
by GC and GC–MS. Recycling experiments were carried out with
the latter recovered, THF-washed and vacuum-dried catalyst
following the same experimental procedure as described above.
2.4. Synthesis of supported Pd-NPs Pd@LR and Pd@C by MVS technique
In
a typical synthesis procedure, Pd vapor generated at
1.5 ꢁ 10ꢀ6 psi by resistive heating of the metal (500.0 mg) in an
alumina-coated tungsten crucible was co-condensed with a 1:1
mixture of mesitylene (30.0 mL) and 1-hexene (30.0 mL) in a glass
reactor at liquid nitrogen temperature. The reactor chamber was
heated to the melting point of the solid matrix (ca. ꢀ40 °C), and
the resulting brown solution was siphoned and handled at low
temperature (ꢀ20 °C) with the Schlenk tube technique. The
Pd-content of the obtained Pd solvated metal atoms (SMA) deter-
mined by ICP–OES analysis was 1.4 mg of Pd/mL. In a Schlenk tube
a portion of the Pd SMA (3.0 mg, 2.2 mL) was added to LR
(500.0 mg) or Vulcan XC-72 (C) in deaerated CHCl3 (10.0 mL) under
nitrogen atmosphere and the resulting suspension was allowed to
stir at 25 °C for 6 h. The addition of an excess of diethyl ether
(50.0 mL) to the latter solutions caused the precipitation of a gray
(Pd@LR) or black (Pd@C) powder, which was isolated upon
centrifugation of the suspension and successive decanting of the
colorless supernatant. The obtained powder was dried under
reduced pressure. All isolated samples contained 0.6 w/w% of Pd
as determined by ICP–OES analysis. The sample containing a
3.0 w/w% Pd loading (Pd0@LBiPy) was prepared by applying the
2.7.2. Solventless hydrogenation reactions
Pd0@LBiPy (40.0 mg, 0.014 mmol, 3.0 w/w% Pd loading) was
placed into a Teflon-coated stainless steel autoclave (80.0 mL)
together with the substrate (3.0 mL, 24.0 mmol). The autoclave
was then sealed and flushed with nitrogen for 2 min. Afterward
the autoclave was pressurized with hydrogen gas, heated at 60 °C
by means of an oil bath and stirred at 700 rpm. The autoclave
was continuously fed with hydrogen gas from a reservoir in order
to maintain a constant gas pressure (150 psi) during the catalytic
reaction. After the desired reaction time, the autoclave was cooled
to room temperature and the excess of hydrogen vented off. To the
dark suspension was added diethyl ether (8.0 mL) causing the
separation of the catalyst, which settled at the bottom of the
autoclave. The supernatant was decanted and the catalyst was

190
W. Oberhauser et al. / Journal of Catalysis 330 (2015) 187–196
washed with diethyl ether (2 ꢁ 3.0 mL). The combined diethyl
ether phases were analyzed by GC.
3. Results and discussion
3.1. Synthesis and characterization of end-functionalized
stereocomplexes
PLA containing
a functional end-group R
(l/d-PLAR) with
R = 4-methyl, 40-methylene-2,20-bipyridine (BiPy), 4-methylenepyridine
(Py) and benzyl (Bn) attached covalently via an ester group to the
polymer backbone was synthesized by Sn(Oct)2-catalyzed ring
opening polymerization (ROP) [23] of bulk l- and d-lactide in the
presence of the polymerization initiator ROH (Scheme 1) at 135 °C.
In contrast, carboxylic acid end-functionalized l-PLA and d-PLA
(i.e. l-PLA and d-PLAH) were obtained upon the same catalytic
reaction carried out in refluxing toluene instead of molten lactide
and in the presence of a defined amount of water as polymerization
initiator. The end-functionalized polymers obtained were white
solids, featured by a molar weight (Mn) in the range between 9000
and 11,000 g/mol (i.e. determined by GPC in THF) which corresponds
to a polymerization degree of 125–150, due to the chosen molar
ratio between lactide and the polymerization initiator ROH of 70.
XPS analyses of LR proved the presence of Sn(II) (i.e. binding energy
for Sn3d5/2 of 486.5 eV) on the polymer surface, which stems from
the Sn-catalyst used for the ROP [25].
Fig. 1. PXRD diffractograms of LR (upper traces) and Pd@LR.
Room temperature PXRD spectra of LR exhibited the character-
istic Bragg reflexes of the PLA-stereocomplex L centered at 12.048°,
20.783°, 23.982° and 32.781° (2H) [27,28] regardless of the nature
of terminal group present in the PLA backbone (Fig. 1, upper
traces).
The thermal stability of LR was studied by TG analyses carried
out under a nitrogen atmosphere. As a result LBiPy/Py/Bn showed a
Tonset (i.e. temperature at 5% weight loss) between 259.9 and
239.7 °C, whereas LH (i.e. carboxylic acid polymer termination)
exhibited the lowest Tonset of 214.5 °C. This experimental result
clearly indicates an increase of the thermal stability of the
polyester-based polymer when the polymer chain ends with an
ester functionality instead of an carboxylic acid group. The former
functional group notably retards the back-biting of terminal lactate
unit, shortening hence the polymer chain under the concomitant
release of lactide [14].
PXRD spectra showed a typical diffraction pattern of crystalline
PLA (Fig. S1), while the successful incorporation of the functional
groups (R) in the PLA backbone was proved by acquiring 13C{1H}
NMR spectra in CDCl3 showing for R = Bn, Py and BiPy a singlet
at ca. 65.4 ppm which corresponds to the methylene unit of the
functional groups attached via ester functionality to PLA, whereas,
a broad 13C{1H} NMR singlet centered at 172.97 ppm for l/d-PLAH
proved the presence of an carboxylic acid terminal group in the
latter polymer [26].
ESEM micrographs acquired for LBiPy/Py/Bn/H under identical
experimental conditions (i.e. magnification of 3000ꢁ, working
distance of 6.9 mm) showed for LH a significantly rougher surface
compared to LBiPy/Py/Bn (Fig. S2). In fact, a similar surface roughness
was observed after hydrolytic degradation of PLA characterized by
a terminal ester functionality [16].
Equimolar solutions of l- and d-PLAR in CH2CH2 with identical R
were mixed at room temperature, obtaining a clear solution which
upon solvent evaporation gave the corresponding stereocomplex
LR [14] (Scheme 1) as an off white solid in 90–95% yield.
Scheme 1. Synthesis of l/d-PLAR, LR and Pd@LR.

W. Oberhauser et al. / Journal of Catalysis 330 (2015) 187–196
191
3.2. Synthesis and characterization of Pd-NPs onto end-functionalized
stereocomplexes
which was dried under vacuum. The Pd-content of Pd@LR, deter-
mined by ICP–OES, was 0.6 w/w%.
Isolated samples of Pd@LR were analyzed by HRTEM, PXRD, XPS,
IR spectroscopy and TGA. HRTEM micrographs of Pd@LBiPy and
Pd@LPy (Fig. 2) showed well-dispersed NPs characterized by the
same average size (dm) of 2.0 0.6 nm.
Pd-NPs supported onto LR were obtained by means of the MVS
technique [12,18,19]. According to this latter synthesis approach a
Pd-vapor was co-condensed with vapors of mesitylene and
1-hexene at ꢀ196 °C. Upon warming of the latter matrix to
ꢀ20 °C, a brown solution of Pd-solvated atoms was obtained to
which was added LR dispersed in CHCl3 (Scheme 1). The LR
supported Pd-NPs (i.e. Pd@LR) were isolated upon addition of
diethyl ether, causing the precipitation of Pd@LR as a gray powder
In contrast, HRTEM micrographs of Pd@LBn/H (Fig. 3) exhibited
Pd-NPs of comparable size, which formed large aggregates, leading
hence to a poor Pd dispersion on the polymer surface. In fact,
bipyridine and pyridine end-functionalized polymers revealed
notable Pd-NP stabilizing properties [13,19,24,29,30]. Even on
Fig. 2. HRTEM micrographs and histograms of Pd@LBiPy (left) and Pd@LPy (right).
Fig. 3. HRTEM micrographs of Pd@LBn (left) and Pd@LH (right).

192
W. Oberhauser et al. / Journal of Catalysis 330 (2015) 187–196
increasing the Pd loading from 0.6 to 3.0 w/w% as in Pd0@LBiPy, a
similar Pd dispersion was obtained (i.e. Pd-NPs of 2.3 0.7 nm,
Fig. S3).
PXRD spectra of Pd@LR (Fig. 1, lower traces) showed in all cases
the same diffraction pattern as for the polymer support (LR) prov-
ing that no alteration of the support occurred in the course of the
Pd-NP generation by MVS technique.
XPS analysis of Pd@LR showed the presence of Pd(0) (65%) and
Pd(II) (35%) regardless of the polymer support employed. The
rather small Pd-NPs found in Pd@LR are reactive and can hence
be easily be oxidized to PdO even under a careful treatment of
the Pd-NP sample after MVS (i.e. samples stored under nitrogen
atmosphere). In fact, the Pd(0)/Pd(II) molar ration found is in
accordance with Pd-NPs covered by a monolayer of PdO [31,32].
A comparison of the ATR-IR spectra of LBiPy/H with those of
Pd@LBiPy/H showed at a first glance the typical IR absorption bands
of the PLA stereocomplex (Fig. S5) [33]. However, in case of LBiPy we
observed the characteristic H-out of plane bending motion of
2,20-bipyridine at 738 cmꢀ1 [34] (Fig. 4, trace c, peak labeled with
asterisk), while the one at 757 cmꢀ1 was overlapped by a bending
motion of PLA [33]. In contrast, Pd@LBiPy (Fig. 4, trace d) showed
the absence of the absorption band at 738 cmꢀ1, which might be
the result of interactions of the aromatic rings of 2,20-bipyridine
with the Pd-NPs’ surface. To verify this hypothesis, we reduced
the model compound Pd(OAc)2(2,20-bipyridine) (OAc = acetate)
[35] with hydrogen pressure to 2,20-bipyridine stabilized Pd-NPs.
As a result, the IR spectrum of 2,20-bipyridine-stabilized Pd-NPs
showed a notable intensity decrease of the absorption band at
757 and 738 cmꢀ1 (Fig. S6).
Fig. 5. PXRD spectra acquired at room temperature of: Pd@LH as-synthesized (a);
after melting and cooling to room temperature (b); Pd@LBiPy as-synthesized (c) and
after melting and cooling to room temperature (d). Bragg reflexes labeled with an
asterisk are assigned to the sample holder material (Al2O3).
polymer upon cooling of the melt, as shown by PXRD for
Pd@LBiPy (Fig. 5, trace d vs c). A similar enhanced thermal stability
of polystyrene and polypropylene in the presence of Pd-NPs was
rationalized by a reduced mobility of the polymer chains in the
presence of NPs, avoiding the intermolecular free radical transfer
[36]. In contrast Pd@LH decomposed upon melting. Accordingly,
after melting and cooling of Pd@LH only Bragg reflexes stemming
from sample holder material (Fig. 5, trace b) were observed.
TG analyses carried out on Pd@LBiPy/Py/Bn/H in an nitrogen
atmosphere clearly showed an increased thermal stability of the
functionalized stereocomplexes in the presence of Pd-NPs.
Accordingly, the Tonset of degradation of the polymer structure of
Pd@LBiPy/Py/Bn/H shifted to higher values compared to LBiPy/Py/Bn/H
(i.e. 273.9 vs 259.9 °C (LBiPy), 285.6 vs 250.5 °C (LPy), 272.3 vs
239.7 °C (LBn) and 232.5 vs 214.5 °C (LH)). It is furthermore impor-
tant to stress on the fact that LR decompose upon melting (ca.
225 °C), regardless of the nature of the terminal group. The pres-
ence of Pd-NPs onto LBiPy/Py/Bn led to recrystallization of the
3.3. Catalytic hydrogenation reactions
The catalysts Pd@LBiPy/Py/Bn/H were employed to hydrogenate
cinnamaldehyde which is a structure sensitive model compound
for a,b-unsaturated carbonyl compounds in THF at 60 °C. The cat-
alysts are completely insoluble in THF [13].
Hydrogen pressure (150 psi) and stirring rate (700 rpm) were
chosen to guarantee a substrate conversion without notable
mass-transport effects. Pd@C (i.e. C = Vulcan XC-72 (254 m2/g), Pd
loading of 0.6 w/w% and NPs size of 2.6 0.7 nm (Fig. S4)) were
chosen as catalytic reference system. The metal dispersion for
Pd@LBiPy/Py (0.6 w/w% Pd loading), Pd0@LBiPy (3.0 w/w% Pd loading)
and Pd@C (0.6 w/w% Pd loading) was calculated from HRTEM data
applying the equation: dVS/dat = 3.32/(FE)1.23 with dVS = mean size
of Pd crystallites, dat = atomic diameter of Pd (0.275 nm) and FE
the exposed fraction of Pd [37]. As a result, Pd@LBiPy/Py and Pd@C
exhibited a metal dispersion of 53% and 43%, respectively. The
results obtained from the catalysts’ screening are compiled in
Table
1 and normalized TOF values are reported only for
Pd@LBiPy/Py which showed uniformly dispersed Pd-NPs.
A blank reaction conducted with LBiPy under catalytic conditions
reported in Table 1 for 2 h gave a substrate conversion of 1.4%,
which is indicating a negligible contribution of tin residues (i.e.
stemming from ROP of lactide) to the observed catalytic activity
of Pd@LR.
A comparison of the catalytic performance of Pd@LR showed
under identical catalytic conditions the following trend of activity:
Pd@LBiPy > Pd@LPy > Pd@LBn > Pd@LH. The much higher catalytic
activity of Pd@LBiPy/Py compared to Pd@LBn/H is due to the higher
Pd-dispersion found for the former catalysts, as proved by the cor-
responding TEM-images acquired for the as-synthesized catalysts
(Figs. 2 and 3). Consequently, more substrate accessible metal sites
on the NP surface are available for the catalytic hydrogenation
reaction. The chemoselectivity for C@C bond hydrogenation of cin-
namaldehyde to give 3-phenylpropanal followed the same trend as
Fig. 4. ATR-IR spectra of LH (a), Pd@LH (b), LBiPy (c) and Pd@LBiPy (d).
observed for the catalytic activity, showing for Pd@LBiPy
a

W. Oberhauser et al. / Journal of Catalysis 330 (2015) 187–196
193
Table 1
a substrate to Pd molar ratio of 1714. Since the hydrogenation of
3.0 mL of cinnamaldehyde consumed a notable amount of hydro-
gen, the autoclave was continuously fed with hydrogen from a
reservoir in order to guarantee a constant gas pressure in the
course of the catalytic reactions. The substrate conversion reached
88% after a reaction time of 6 h, along with a chemoselectivity of
90%.
Hydrogenation of cinnamaldehyde by Pd@C, Pd@LBiPy/Py/Bn/H and Pd0@LBiPy
.
Entrya
Catalyst
t (h)
Conv.(%)/TOF (hꢀ1
)
Sel.(%)c
b
1
2
3
Pd@LBiPy
Pd@LBiPy
Pd@LBiPy
Pd@LBiPy
Pd@LBiPy
Pd0@LBiPy
Pd0@LBiPy
Pd@LPy
1
2
3
2
2
2
2
2
2
2
2
2
2
38/682
74/664
100/n.d.
72/646
69/619
65/645
55/n.d.
54/485
42/377
30/269
30/n.d.
14/n.d.
99/921
99
97
95
96
94
95
90
88
75
68
65
60
55
4d
5e
6f
7d,f
8
3.4. Analysis of recovered catalysts
9d
10e
11
12
13
Pd@LPy
Pd@LBiPy/Py was easily recovered from catalytic reactions carried
out in THF by simply decanting the supernatant and washing the
solid with THF. Pd0@LBiPy was recovered from solventless catalytic
reactions by adding diethyl ether to the reaction mixture, causing
the separation of the polymer from organic substrate. The super-
natant was then simply decanted and the catalyst washed with
diethyl ether. ICP–OES analysis of the catalytic THF solutions stem-
ming from Pd@LBiPy-catalyzed reactions showed the presence of a
very low amount of Pd (i.e. <0.5 ppm), confirming an efficient
anchoring of Pd on the polymer support.
Pd@LPy
Pd@LBn
Pd@LH
Pd@C
Solventless hydrogenationg
14
15
16
Pd0@LBiPy
Pd0@LBiPy
Pd0@LBiPy
2
4
6
30/547
58/n.d.
88/n.d.
95
91
90
a
Catalytic conditions: Pd (0.0012 mmol), cinnamaldehyde (141.2 lL,
1.12 mmol), THF (10.0 mL), p(H2) = 150 psi, T = 60 °C.
b
Normalized TOF (i.e. mmol substrate (converted) ꢁ (mmol Pd(surface) ꢁ h)ꢀ1
.
HRTEM micrographs obtained from recovered Pd@LBiPy/Py after
hydrogenation of cinnamaldehyde for 2 h (Fig. 6) showed a slight
increase of the particle size in both cases compared to the corre-
sponding as-synthesized catalysts (i.e. 2.8 0.7 (Pd@LBiPy) and
3.1 0.8 nm (Pd@LPy) vs 2.0 0.5 nm (as-synthesized catalysts)).
STEM images of both latter recovered catalysts (Fig. 6) clearly
c
Selectivity for 3-phenylpropanal.
d
2nd recycling experiment.
e
4th recycling experiment.
f
The same amount of Pd exposed as Pd@LBiPy
.
g
Pd0 (3.0 w/w% metal loading, 0.014 mmol), cinnamaldehyde (3.0 mL,
24.0 mmol).
exhibited
a much more pronounced Pd-NP aggregation for
Pd@LPy, which is in accordance with a significantly higher drop
of the catalytic activity and chemoselectivity found for Pd@LPy
chemoselectivity of 99% (Table 1, entry 1), which decreased to 95%
at complete substrate conversion (Table 1, entry 3). This latter
chemoselectivity is significantly higher compared to that observed
for Pd@C (55%) at the same substrate conversion (Table 1, entry 3
vs 13).
compared to Pd@LBiPy
.
PXRD spectra of Pd0@LBiPy, recovered after solventless catalytic
reactions conducted at different reaction times showed for the
support the same diffraction pattern as the as-synthesized catalyst
(Fig. S7) proving the stability of the support even under
non-diluted catalytic conditions. A slow acquisition of the same
Pd@LBn/H which is characterized by the presence of
NPs-aggregates on the polymer surface showed much higher
C@O hydrogenation activity compared to Pd@LBiPy/Py. This latter
experimental fact corroborates the known observation that the
formation of cinnamyl alcohol is strictly related to the shape of
the Pd-NP surface. Large particles, which have a small surface
curvature trigger the repulsion of the phenyl ring from the metal
surface [38,39]. As a consequence, the interaction of C@O is favored
over C@C with the NP surface, fostering hence the hydrogenation
of the C@O bond.
spectra in the 2
an intensity change of the characteristic 111 Bragg reflex of fcc Pd
centered at ca. 39.0° (2 ). As a result, the Pd-NP size, determined
H range between 36.5° and 42.5° (Fig. 7) exhibited
H
by the Debye–Scherrer method [43] using the Pd (111) Bragg
reflex, increased during the first 2 h from 2.0 to 3.9 nm, whereas
the Pd-NP size detected after 4 h is almost identical to that after
2 h.
XPS spectra acquired in the Pd3d region for recovered Pd0@LBiPy
after a solventless hydrogenation reaction lasting 2 h (Fig. 8)
Interestingly, on varying the molar ratio between the anchored
bipyridine functionalities and the surface Pd atoms (i.e. Pd@LBiPy
:
showed 69.4% Pd(0) (i.e. Pd 3d5/2, 334.8 eV and Pd 3d3/2,
BiPy to Pdsurface of 6 vs Pd0@LBiPy: BiPy to Pdsurface of 0.7) the
340.6 eV) [44] and 30.6% Pd(II) (i.e. Pd 3d5/2, 337.2 eV and Pd
3d3/2, 342.4 eV) [45] on the NPs’ surface, which is similar to those
observed for the as-synthesized catalyst (i.e. 68.4% Pd(0) and 31.6%
chemoselectivity was notably altered, even when the initial parti-
cle size was comparable (i.e. 2.0 nm (Pd@LBiPy
) and 2.3 nm
(Pd0@LBiPy)). As a result, Pd0@LBiPy exhibited under identical exper-
imental conditions a chemoselectivity and substrate conversion
which was lower compared to Pd@LBiPy (Table 1, entry 6 vs 2). In
addition, a recycling experiment with Pd0@LBiPy clearly confirmed
a drop of chemoselectivity to 90% (Table 1, entry 7 vs 6), which
is a direct consequence of the scarce stabilization of the initially
small NPs, in case of an almost 1:1 M ratio between
2,20-bipyridine functional groups and surface Pd atoms [40–42].
Recycling experiments with Pd@LBiPy/Py were carried out,
recovering the catalyst by a simple filtration and washing with
THF in air atmosphere. Pd@LBiPy showed a comparable catalytic
performance in the different recycling experiments (Table 1,
entry 2 vs 4 and 5). In contrast, Pd@LPy exhibited a significant
drop of catalytic activity as well as chemoselectivity (i.e. from
88% to 68% after the fourth catalytic cycle) (Table 1, entry 8 vs
9 and 10).
Pd(II)). This experimental result is rationalized by a fast
re-oxidation of Pd(0) located at the Pd-NP surface to PdO [31].
In order to underscore the significantly higher Pd-NP-stabilizing
properties of 2,20-bipyridine compared to pyridine, we synthesized
the model Pd(OAc)2-macrocomplexes (OAc = acetate) of the for-
mula Pd(OAc)2(d-PLABiPy
)
and trans-[Pd(OAc)2(d-PLAPy)2] [24]
(Scheme 2) which share the same nitrogen to Pd atom ratio.
Upon reduction of CH2Cl2 solutions of the latter
Pd-macrocomplexes by means of hydrogen, d-PLABiPy and
d-PLAPy-stabilized Pd-NPs were obtained and analyzed by TEM mea-
surements. As a result, only Pd@d-PLABiPy showed well-dispersed
Pd-NPs with an average size of 3.8 0.6 nm (Fig. S14), while
Pd@d-PLAPy exhibited worm-like structures of Pd-NP aggregates,
which resembled those obtained after the reduction of
trans-[Pd(OAc)2(PEGPy)2] (PEG = poly(ethyleneglycol)) under identi-
cal experimental conditions [24]. Pd@PLABiPy is characterized by a
Pd content of 1.0 wt% and a molar ratio between 2,20-bipyridine
units and surface Pd atoms of 3. A parallel catalytic reaction with
We tested the catalytic performance of Pd0@LBiPy also under
solventless catalytic conditions (Table 1, entries 14–16) applying

194
W. Oberhauser et al. / Journal of Catalysis 330 (2015) 187–196
Fig. 6. HRTEM (upper) and HRSTEM (middle) micrographs and histograms of recovered Pd@LBiPy (left) and Pd@LPy (right) after catalysis conducted at 60 °C.
Pd@d-PLABiPy under experimental conditions identical to that
reported in Table 1, gave a chemoselectivity for 3-phenylpropanal
of 91% along with a substrate conversion of 57% for a reaction time
of 2 h. This experimental result corroborates again the drop of
chemoselectivity for 3-phenylpropanal by increasing the size of
the Pd-NP.
Based on experimental data stemming from TEM, XPS and
IR-measurements along with catalytic data collected for Pd@LBiPy
,
we propose the following catalytic cycle operative for the
Pd@LBiPy-catalyzed cinnamaldehyde hydrogenation as sketched in
Fig. 9.
Upon the reaction of the as-synthesized Pd-NPs of Pd@LBIPy the
oxide layer is removed and the naked Pd-surface activates hydro-
gen that reacts with the C@C coordinated cinnamaldehyde to give
3-phenylpropanal. Anchored 2,20-bipyridine controls catalytic
activity and the size of the Pd-NPs by reversible discoordination
(i.e. making accessible catalytically active sites) and coordination
to the surface Pd atoms [42].
Fig. 7. PXRD diffractograms of Pd0@LBiPy acquired between 36.5° and 42.5° (2
H):
as-synthesized (lower trace); after solventless catalysis conducted for 2 h and 4 h
(upper traces).

W. Oberhauser et al. / Journal of Catalysis 330 (2015) 187–196
195
4. Conclusions
Poly(lactic acid) (PLA)-based stereocomplexes functionalized at
the carboxylic acid end were used to support Pd nanoparticles
(NPs), obtained by means of the metal vapor synthesis technique.
The nature of the functional end group notably influenced the dis-
persion of Pd-NPs on the polymer support. Pd-NPs supported onto
2,20-bipyridine
end-functionalized
PLA-stereocomplex
(i.e.
Pd@LBiPY) exhibited the best chemoselectivity for the C@C bond
reduction in cinnamaldehyde (i.e. 95% at complete substrate con-
version). In contrast, other heterogeneous polymer-based
Pd-catalysts showed significantly lower chemoselectivity for
3-phenylpropanal (i.e. <90%) [11,12]. The role of the
polymer-anchored 2,20-bipyridine is best described as an efficient
Pd-NP stabilizer in the course of the catalytic hydrogenation reac-
tion, provided the molar ratio between the 2,20-bipyridine units
and the Pd surface is high enough. Indeed, when the latter ratio
was 6, as found for Pd@LBiPY, small Pd-NPs of 2.0 nm were effi-
ciently stabilized. Consequently, the high chemoselectivity
observed for 3-phenylpropanal with Pd@LBiPy is particle size driven
[46,47] (i.e. small particle sizes foster C@C bond coordination to the
Pd surface and hence formation of 3-phenylpropanal). LBiPy stabi-
lized Pd-NPs were found robust enough to be used as recyclable
catalyst.
Fig. 8. Pd3d XPS spectra of Pd0@LBiPy before (A) and after (2 h) solventless catalysis
(B).
Acknowledgments
C.E. and R.P.J. thank MIUR-Italy (FIRB 2010; contract
RBFR10BF5V) for financial support. Ente Cassa di Risparmio di
Firenze is acknowledged for financial support and B. Cortigiani of
the CeTeCS center for the technical assistance during XPS
measurements.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2015.07.012.
References
Scheme 2. Pd(OAc)2-macrocomplexes with d-PLABiPy and d-PLAPy
.
[1] B. Chen, U. Dingerdissen, J.G.E. Krauter, H.G.J. Lansink Rotgerink, K. Möbus, D.J.
Ostgard, P. Panster, T.H. Riermeier, S. Seebald, T. Tacke, H. Trauthwein, Appl.
Catal. A: Gen. 280 (2005) 17–46.
[2] H.U. Blaser, C. Malan, B. Pugin, F. Spindler, H. Steiner, M. Studer, Adv. Synth.
Catal. 345 (2003) 103–151.
[3] O.S. Chambyal, S. Paul, T. Shamim, M. Gupta, R. Gupta, A. Loupy, Synth.
Commun. 43 (2013) 656–667.
[4] P. Gallezot, D. Richard, Catal. Rev. Sci. Eng. 40 (1998) 81–126.
[5] X. Ji, X. Niu, B. Li, Q. Han, F. Yuan, F. Zaera, Y. Zhu, H. Fu, ChemCatChem 6 (2014)
3246–3253.
[6] M. Steffan, F. Klasovsky, J. Arras, C. Roth, J. Radnik, H. Hofmeister, P. Claus, Adv.
Synth. Catal. 350 (2008) 1337–1348.
[7] P. Claus, Top. Catal. 5 (1998) 51–62.
[8] F. Delbecq, P. Sautet, J. Catal. 152 (1995) 217–236.
[9] P. Kaur, J.T. Hupp, S.T. Nguyen, ACS Catal. 1 (2011) 819–835.
[10] M. Králik, A. Biffis, J. Mol. Catal. A: Chem. 177 (2001) 113–138.
[11] Y. Gao, C.-A. Chen, H.-M. Gau, J.A. Bailey, E. Akhadov, D. Williams, H.-L. Wang,
Chem. Mater. 20 (2008) 2839–2844.
[12] A.M. Raspolli Galletti, L. Toniolo, C. Antonetti, C. Evangelisti, C. Forte, Appl.
Catal. A: Gen. 447–448 (2012) 49–59.
[13] G. Petrucci, W. Oberhauser, M. Bartoli, G. Giachi, M. Frediani, E. Passaglia, L.
Capozzoli, L. Rosi, Appl. Catal. A: Gen. 469 (2014) 132–138.
[14] H. Tsuji, Macromol. Biosci. 5 (2005) 569–597.
[15] S. Regnell Andersson, M. Hakkarainen, S. Inkinen, A. Södergard, A.-C.
Albertsson, Biomacromolecules 11 (2010) 1067–1073.
[16] S. Regnell Andersson, M. Hakkarainen, S. Inkinen, A. Södergård, A.-C.
Albertsson, Biomacromolecules 13 (2012) 1212–1222.
[17] H. Yamane, K. Sasai, Polymer 44 (2003) 2569–2575.
[18] G. Vitulli, C. Evangelisti, A.M. Caporusso, P. Pertici, N. Panziera, S. Bertozzi, P.
Salvadori, in: B. Corain, G. Schmid, N. Toshima (Eds.), Metal Nanoclusters in
Catalysis and Materials Science: The Issue of Size Control, Elsevier,
Amsterdam, 2008 (Chapter 32).
Fig. 9. Proposed catalytic cycle operative for Pd@LBiPy-catalyzed cinnamaldehyde
hydrogenation.

196
W. Oberhauser et al. / Journal of Catalysis 330 (2015) 187–196
[19] C. Evangelisti, N. Panziera, A. D’Alessio, L. Bertinetti, M. Botavina, G. Vitulli, J.
Catal. 272 (2010) 246–252.
[20] K.J. Klabunde, Free Atoms, Clusters and Nanoscale Particles, Academic Press,
San Diego, 1994.
[21] D.A. Shirley, Phys. Rev. B 5 (1972) 4709–4714.
[22] G. Giachi, M. Frediani, W. Oberhauser, E. Passaglia, J. Polym. Sci. Part A: Polym.
Chem. 49 (2011) 4708–4713.
[23] H.R. Kricheldorf, I. Kreiser-Saunders, C. Boettcher, Polymer 36 (1995) 1253–
1259.
[31] E.H. Voogt, A.J.M. Mens, O.L.J. Gijzeman, J.W. Geus, Surf. Sci. 350 (1996) 21–31.
[32] M. Brun, A. Berthet, J.C. Bertolini, J. Electron Spectrosc. Relat. Phenom. 104
(1999) 55–60.
[33] J. Zhang, H. Sato, H. Tsuji, I. Noda, Y. Ozaki, Macromolecules 38 (2005) 1822–
1828.
[34] J.S. Strukl, J.L. Walter, Spectrochim. Acta 27A (1971) 209–221.
[35] G. Villain, A. Gaset, Ph. Kalck, J. Mol. Catal. A: Chem. 12 (1981) 103–111.
[36] J.-Y. Lee, Y. Liao, R. Nagahata, S. Horiuchi, Polymer 47 (2006) 7970–7979.
´
[37] A. Borodzinski, M. Bonarowska, Langmuir 13 (1997) 5613–5620.
[24] G. Giachi, W. Oberhauser, M. Frediani, E. Passaglia, L. Capozzoli, L. Rosi, J.
Polym. Sci. Part A: Polym. Sci. 51 (2013) 2518–2526.
[25] A. Södergård, M. Niemi, J.-F. Selin, J.H. Näsman, Ind. Eng. Chem. Res. 34 (1995)
1203–1207.
[26] J.L. Espartero, I. Rashkov, S.M. Li, N. Manolova, M. Vert, Macromolecules 29
(1996) 3535–3539.
[38] A. Giroir-Fendler, D. Richard, P. Gallezot, Catal. Lett. 5 (1990) 175–182.
[39] C. Minot, P. Gallezot, J. Catal. 123 (1990) 341–348.
[40] N. Yan, C. Xiao, Y. Kou, Coord. Chem. Rev. 254 (2010) 1179–1218.
[41] N. Yan, Y. Yuan, P.J. Dyson, Dalton Trans. 42 (2013) 13294–13304.
[42] A. Roucoux, J. Schulz, H. Patin, Chem. Rev. 102 (2002) 3757–3778.
[43] J.I. Langford, A.J.C. Wilson, J. Appl. Cryst. 11 (1978) 102–113.
[44] S. Miao, C. Zhang, Z. Liu, B. han, Y. Xie, S. Ding, Z. Yang, J. Phys. Chem. C 112
(2008) 774–780.
[27] D. Brizzolara, H.-J. Cantow, K. Diederichs, E. Keller, A.J. Domb, Macromolecules
29 (1996) 191–197.
[28] D. Sawai, Y. Tsugane, M. Tamada, T. Kanamoto, M. Sungil, S.-H. Hyon, J. Polym.
Sci. Part B: Polym. Phys. 45 (2007) 2632–2639.
[45] B.M. Choudary, S. Madhi, N.S. Chowdari, M.L. Kantam, B. Sreedhar, J. Am. Chem.
Soc. 124 (2002) 14127–14136.
[29] J. Lu, P.H. Toy, Chem. Rev. 109 (2009) 815–838.
[30] Y. Uozumi, R. Nakao, Angew. Chem. Int. Ed. 42 (2003) 194–197.
[46] K. Zhou, Y. Li, Angew. Chem. Int. Ed. 51 (2012) 602–613.
[47] K. An, G.A. Somorjai, ChemCatChem 4 (2012) 1512–1524.