Selective hydrogenation over Pd nanoparticles supported on a pore-flow-through silica monolith microreactor with hierarchical porosity

Article abstract of DOI:10.1039/c2dt31690k

Well-dispersed Pd nanoparticles have been synthesized inside the mesoporosity of a silica monolith featuring hierarchical porosity of homogeneous interconnected macropores (4 microns) and mesopores (11 nm). These monoliths have been implemented as microreactors for selective hydrogenation reactions. Conversion and selectivity can be tuned by adjusting the flow rates of hydrogen and substrates. In the selective hydrogenation of cyclooctadiene, a conversion of 95% and a selectivity of 90% in the monohydrogenated product, constant over a period of 70 h, have been reached. These figures correspond to a productivity of 4.2 mmol s^-1 g^-1_MonoSil (or 0.32 mol s ^-1 g^-1_Pd). In the stereoselective hydrogenation of 3-hexyn-1-ol a constant conversion of 85% was observed, with however moderate selectivity into the cis isomer, over a test period of 7 h. These results open the route to the synthesis of important chemicals and intermediates via safe and green processes. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c2dt31690k

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Selective hydrogenation over Pd nanoparticles
supported on a pore-flow-through silica monolith
microreactor with hierarchical porosity†‡
Cite this: Dalton Trans., 2013, 42, 1378
Alexander Sachse,a Noemi Linares,b Pierluigi Barbaro,*b François Fajulaa and
Anne Galarneau*a
Well-dispersed Pd nanoparticles have been synthesized inside the mesoporosity of a silica monolith fea-
turing hierarchical porosity of homogeneous interconnected macropores (4 microns) and mesopores
(11 nm). These monoliths have been implemented as microreactors for selective hydrogenation reactions.
Conversion and selectivity can be tuned by adjusting the flow rates of hydrogen and substrates. In the
selective hydrogenation of cyclooctadiene, a conversion of 95% and a selectivity of 90% in the monohy-
drogenated product, constant over a period of 70 h, have been reached. These figures correspond to a
productivity of 4.2 mmol s−1 g−1MonoSil (or 0.32 mol s−1 g−1Pd). In the stereoselective hydrogenation of 3-
hexyn-1-ol a constant conversion of 85% was observed, with however moderate selectivity into the cis
isomer, over a test period of 7 h. These results open the route to the synthesis of important chemicals
and intermediates via safe and green processes.
Received 26th July 2012,
Accepted 27th September 2012
DOI: 10.1039/c2dt31690k
www.rsc.org/dalton
been reported.5 The very first reactors were developed by the
group of Fréchet and Svec6 and were based on polymers. These
1. Introduction
Chemistry in flow is a promising option for the implemen- reactors have then been further developed by the group of
tation of catalytic reactions enabling process intensification Kunz and Kirschning7 to obtain hybrid glass/polymer reactors
and simplified workups. Continuous flow chemistry allows in that have been used for a large variety of organic transform-
addition safer process operation through better control of re- ations. To overcome swelling problems of polymer monoliths a
action parameters such as temperature, contact time, flow rate, new generation of unconventional monolithic inorganic
minimization of solvent use, etc.1 Better control of contact (silica) microreactors for catalysis has appeared,8 which are
time leads in turn to improved product selectivity and yield in formed by a combination of spinodal decomposition and a
comparison to batch reactors and represents a valuable tool sol–gel process featuring an interconnected and homogeneous
for process intensification.1 Recently, efforts have been made system of macropores (2–4 microns) and comprising within
for the design of efficient microreactors comprising either their skeleton homogeneously distributed mesopores adjusta-
microstructured multichannel reactors2 or macroporous ble in size (3–40 nm).9 This novel family of structured micro-
monoliths.3 Macroporous monoliths present the advantages of reactors features high chemical and physical stability with a
an interconnected and isotropic network of macropores allow- high condensation state of silica (76% of Q4 and 24% of Q3)9c
ing high surface-to-volume ratios and efficient mixing of and can be readily functionalized by several catalytic active
fluids.4
species: organic functions by organosilane grafting (e.g.
Several approaches for the implementation of macroporous aminosilane),8 inorganic species by surface deposition (e.g.
monoliths as pore-flow-through catalytic microreactors have alumina lining),10a in situ synthesis of metal nanoparticles as
Pd0,11 in situ synthesis of hybrid materials as metal–organic
frameworks (MOFs, e.g. CuBTC),10b pseudomorphic transform-
aInstitut Charles Gerhardt Montpellier, UMR 5253 CNRS/UM2/ENSCM/UM1,
ation of the silica skeleton into zeolites,3b,c or direct incorpor-
ENSCM, 8 rue de l’Ecole Normale, 34296 Montpellier Cedex 5, France.
ation of Al12 or Ni13 species during monolith synthesis. Ni
E-mail: anne.galarneau@enscm.fr
bIstituto di Chimica dei Composti Organo Metallici, Consiglio Nazionale delle
species have also been incorporated by direct synthesis in
similar titania monoliths.14 Sato et al.11b,12 have demonstrated
Ricerche, 50019 Sesto Fiorentino, Italy. E-mail: pierluigi.barbaro@iccom.cnr.it
†This work is dedicated to Professor David Cole-Hamilton on the occasion of his
first through Knudsen diffusion calculations in batch and in
retirement and for his outstanding contribution to transition metal catalysis.
packed-bed reactors filled with crushed monoliths that
bimodal porous silica bodies were more active than purely
‡Electronic supplementary information (ESI) available. See DOI: 10.1039/
c2dt31690k
1378 | Dalton Trans., 2013, 42, 1378–1384
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
mesoporous silica because the interconnected network of alumina membrane reactor demonstrated very promising
macropores allows for rapid diffusion resulting in improved activity for the selective hydrogenation of 1,5-cycloocta-
reaction kinetics. They have also shown that too small meso- diene.22a The alumina membrane reactor proved to be more
pores (<10 nm) restrict mass diffusion. The productivity in fine productive than batch and packed-bed reactors.22a However it
chemicals transformations was increased by using higher flow suffered from several drawbacks, among them a low surface
rates until the point where contact times were not enough to area (8 m2 g−1), a poor dispersion of the Pd NPs at the surface,
ensure high conversion.10,11 Previously we have shown that the formation of oligomers during the reaction due to residual
these new kinds of pore-flow-through silica monolith micro- acidity of the support which imposed intermediate reactivation
reactors induced high productivity and high stability for base steps affecting long term productivity, and the need for a rela-
and acid catalyzed reactions of fine chemicals in the liquid tively complex and large set-up to operate such a reactor.
phase under continuous flow.3b,8,10 The superiority of this new
Herein we present an alternative to membrane reactors for
kind of monolith, used as a single piece microreactor, was selective hydrogenation reactions using Pd-particles supported
demonstrated by the higher conversions and productivities on a pore-flow-through monolith featuring a bimodal porosity
achieved compared to packed-bed or batch reactors filled with and a high surface area (600 m2 g−1) easy to operate under soft
crushed monolith.8,10 This result is most probably due to the conditions (25 °C, 2–3 bars H2) under continuous flow.
more uniform contact time arising for the homogeneity of the This unconventional Pd-silica monolith microreactor was
macroporous network in the single piece monolith in compa- used for the selective hydrogenation of 1,5-cyclooctadiene and
rison to the dual macroporosity (inter- and intra-particle) 3-hexyn-1-ol.
encountered in a packed-bed system.
In the present study, we analyze the effectiveness of these
promising pore-flow-through silica monolith microreactors in
a triphasic liquid/gas/solid system for selective hydrogenation
2.1 Synthesis of the macro-/mesoporous silica monolith
reactions under conditions allowing a fine control of gas and
2. Experimental section
The synthesis of the macro-/mesoporous silica monolith has
been adapted from the original synthesis developed by
Nakanishi.9a In a flask 46.3 g distilled H2O and 4.6 g HNO3
(68%, Aldrich) were mixed for 15 min at 0 °C and 4.79 g of
PEO (20 kDa, Aldrich 99%) were added and stirred for 1 h.
Then 37.7 g TEOS (Aldrich 99%) were added and the mixture
was stirred for 1 h. The resulting solution was poured into PVC
tubes of 10 cm length and 8 mm internal diameter and kept at
40 °C for 3 days. The monoliths were then washed in water
and treated in an ammonia solution (0.1 M) at 40 °C for 20 h.
The monoliths were then dried at 40 °C for 24 h and calcined
at 550 °C for 8 h. These silica monoliths (10 cm length and
6 mm diameter, 0.08 g cm−1) were cut into pieces of desired
size (2 cm) and are hereafter named as parent monoliths or
MonoSils.
liquid contact times on Pd nanoparticles (NPs) catalytic sites
immobilized onto the silica network.
Different groups have reported on the catalytic behavior of
Pd NPs supported on polymeric matrixes, dendrimers, zeolites
and dispersed silica in selective hydrogenation.15 Selective
hydrogenation of hydrocarbons with multiple CvC and/or
CuC and/or CvO bonds to achieve selective hydrogenation
products (e.g. monoalkenes) is a highly desired and challen-
ging process in the pharmaceutical, agrochemical and petro-
chemical industry sectors.16 Under comparable conditions,
hydrogenation rates roughly followed the order CuC > CvC >
CvO. This behavior can be justified in terms of greater affinity
of Pd for CvC than for CvO bonds in aliphatic compounds.17
Similar behavior (CvC > CvO) has been noticed using Ni-
titania crushed monolith in packed-bed reactors.14 Particu-
larly, the stereo- and chemo-selective hydrogenation of alkenes
and alkynes in the presence of other functional groups is of
fundamental importance in the synthesis of food additives,
2.2 Synthesis of Pd-nanoparticles in the silica monolith
(Pd-MonoSil)
flavors and fragrances.18 For instance, there is growing interest A solution containing Pd(NH3)4(NO3)2 (0.141 mmol, 0.042 g,
in new hydrogenation processes under soft conditions that Aldrich 98%) in water (10 mL) was prepared. The pH of the Pd
reduce the amount of double bonds in vegetable oils to complex solution was adjusted to 8, 9 or 10 using a 0.1 M
product oils with higher consistency, stability and oxidation ammonia solution. The 2 cm length silica monoliths were
resistance, but also to avoid trans fatty acid esters as trans immerged into these solutions. The amount of Pd in the
isomers have negative effects on health.19 The formation of cis- initial solution corresponded to 1.5 Pd wt% with respect to
alkenes through stereo-selective addition of H2 on alkynes is silica. Monoliths were impregnated for 24 h at room tempera-
also an important step in industrial processes as in fragrance ture under gentle agitation at 100 rpm. The monoliths were
production (e.g. cis-3-hexen-1-ol leaf alcohol with a production then subsequently washed with distillated water (1 L), dried at
of 400 tons per year).20 Selective hydrogenation reactions are 40 °C for 72 h and calcined at 550 °C for 8 h to form Pd oxide
also crucial in polymerization processes to achieve the com- nanoparticles inside the silica monoliths. The reduction of
plete elimination of alkynes and dienes from alkene feed- PdO NPs was then performed with diluted hydrogen (3 vol%
stocks.21 Flow processes could be a solution to finely tune the H2 in He) at 280 °C for 2 h. The resulting monoliths are here-
selectivity of these reactions by limiting the hydrogenation to after named Pd-MonoSils. For use as flow microreactors, the
one double bond. In that respect Pd NPs deposited on an monoliths ended with two glass tubes and were cladded with a
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 1378–1384 | 1379

View Article Online
Paper
Dalton Transactions
heat-shrinkable DERAY® PTFE (polytetrafluoroethylene) clad was collected for GC analysis and the excess amount of the
(DSG-CANUSA) at 280 °C for 2 h as previously described.3b
hydrogen gas released at atmospheric pressure. Commercially
available H2 (99.995%) was used as received.
2.3. Materials characterization
In a typical experiment, the Pd-MonoSil catalyst (0.154 g,
The physico-chemical properties of the monoliths were charac- 1.3 Pd wt%) was first washed under flow with deaerated
terized by nitrogen sorption at 77 K, scanning electron methanol at a flow rate of 0.15 mL min−1 at room temperature.
microscopy (SEM), and transmission electron microscopy No leaching of Pd was observed. A deaerated solution of the
(TEM). Supernatants of Pd complex solutions after impreg- substrate (cyclooctadiene or 3-hexyn-1-ol) in methanol (0.1 and
nation were analyzed with UV/Vis spectroscopy.
0.2 M, respectively) was then allowed to flow through the
The adsorption–desorption isotherms of nitrogen at 77 K cladded Pd-MonoSil catalyst (0.154 g, 1.3 Pd wt%) at a constant
were determined using a Micromeritics ASAP 2020 instrument flow rate of ca. 0.12 mL min−1, together with a constant H2
after outgassing the samples (30 mg) at 250 °C under high flow rate of ca. 0.5 mL min−1 at room temperature, while the
vacuum (0.5 Pa) for 8 h with pressures between 0.001 < p/p0 < hydrogen gas was released at atmospheric pressure at the
0.9959 with an equilibrium interval of 5 s and 36 points in outlet of the reactor. For cyclooctadiene the substrate solution
adsorption and 30 points in desorption. The mesopore volume flow rate was varied between 0.05 and 0.25 mL min−1 and the
was taken at the end of the filling of the mesopores. Surface H2 flow rate between 0.3 and 0.55 mL min−1. This resulted in a
area was determined by the BET (Brunauer Emmett Teller) H2 pressure at the reactor inlet between 2.2 and 2.4 bars
equation considering the Rouquerol law to determine the corresponding to a molar ratio 2.2 < H2/cyclooctadiene < 4.5.
highest relative pressure to use in the BET method.23 The For 3-hexyn-1-ol the substrate solution flow rate was varied
mesopore diameter was determined using the desorption between 0.05 and 0.25 mL min−1 and the H2 flow rate between
branch of the nitrogen isotherm and the Broekhoff and De 0.2 and 1.3 mL min−1. This resulted in a H2 pressure at the
Boer method as it has been demonstrated to be one of the reactor inlet between 2.4 and 3.1 bar corresponding to a molar
more accurate methods for mesoporous silica materials.24 The ratio 0.6 < H2/hexynol < 2.5. Therefore, the pressure drops gen-
morphology of the monoliths was examined using a Hitachi erated by monolithic reactors were between 1 and 2 bars. The
S-4500 I SEM. The chemical analysis of the monolith was per- attainment of the steady state conditions (ca. 1–3 h) was taken
formed by EDX on different parts of cut pieces of the Pd- as the reaction start time. The product solution was period-
MonoSils using a FEI instrument Quanta 200F (15 kV) in retro- ically analyzed for conversion by GC, while 7.2 mL h−1 aliquots
diffusion. TEM images have been obtained with a JEOL 1200 were continuously sampled for subsequent Pd leaching
EX II at 100 kV after grinding the monoliths, dispersing the analysis.
powder in ethanol and deposition of the carbon grid. UV/Vis The content of palladium leached in the catalysis solutions
absorption spectra were recorded on a Hewlett-Packard 8452 A was determined by Inductively Coupled Plasma Atomic Emis-
diode array spectrophotometer.
sion Spectroscopy (ICP-OES) with a Varian 720ES instrument
at a sensitivity of 500 ppb. Each sample (7.2 mL) was treated in
a microwave-heated digestion bomb (Milestone, MLS-200) with
concentrated HNO3 (1.5 mL), 98% H2SO4 (2 mL), 37% HCl
2.4 Selective hydrogenation of 1,5-cyclooctadiene and
3-hexyn-1-ol
For the implementation of the flow experiments a system was (0.5 mL) and a pellet of a digestion aid reagent (0.5 g, 0.1% Se
designed to allow for simultaneous flow of a substrate solution in K2SO4). The solutions were analyzed directly after 1 : 5
and hydrogen gas (up to 40 bars pressure) through a reactor dilution in 0.1 M hydrochloric acid.
tube containing the Pd-MonoSil catalysts. The reactor was
completely inert, as all wet parts were made of PEEK (poly-
ether ether ketone), PFA (perfluoroalkoxy) or PTFE (polytetra-
fluoroethylene). The flow of the substrate solution was
regulated by an Alltech® model 426 HPLC pump in PEEK. A
3. Results and discussion
3.1 Characterization of Pd-MonoSil
constant flow of hydrogen gas was adjusted by a flow controller The functionalization of the parent silica monoliths with Pd
BRONKHORST HI-TEC model F200CV-002-RGD-11-V-MFC. nanoparticles was carried out by impregnation of the monoliths
The hydrogen pressure in the reactor was monitored by a with an aqueous solution of Pd(NH3)4(NO3)2. The initial Pd
BRONKHORST HI-TEC P502C-AGD-11-V-6K0R-EPC meter. The amount corresponds to 1.5 wt% with respect to silica. The
concurrent flows of gas and liquid were driven through a increase of the pH of the impregnation solution leads to an
T-shaped PEEK mixer to ensure efficient gas dispersion. The increase of the amount of Pd incorporated within the silica
mixed hydrogen–substrate solution stream was introduced into monolith due to an increase of the negative charge of the
the reactor through a 6-port Rheodyne mod. 9060 switching surface of the silica support. Pd(NH3)42+ ions are introduced on
valve in PEEK. The Pd-MonoSils cladded into a heat-shrinkable the silica surface by ion-exchange with the protons on the silica
PTFE tube together with two glass tubes (4 mm inner diameter) surface, which compensate the negative charge of the silica.
at each end were connected in a top-down arrangement to the Impregnations at pH 8 (Pd-MonoSil-1), pH 9 (Pd-MonoSil-2) and
system by PFA Swagelok fittings and Chemraz® O-rings (ESI pH 10 (Pd-MonoSil-3) have been carried out. The amount of
Scheme S1‡). At the outlet of the reactor, the product solution impregnated Pd complex was monitored by comparing
1380 | Dalton Trans., 2013, 42, 1378–1384
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
adsorption maxima at λ = 300 nm (adsorption band of Pd
(NH3)4(NO3)2) of the initial solution and the resulting super-
natant solution after the impregnation with the monolith.
The percentage of Pd incorporated within the silica mono-
liths increases linearly from 4 to 89% with the increase of pH
from 8 to 10 and corresponds therefore to a loading of 0.06,
0.66 and 1.33 Pd wt% with respect to silica for Pd-MonoSil-1,
-2 and -3, respectively (Fig. 1). Pd-MonoSil-3 with a Pd content
of 1.33 Pd wt% (UV/Vis) was subsequently calcined at 550 °C
and the Pd complexes in electrostatic interaction with the
silica surface are changed by Pd migration into nanoparticles
of PdO at the surface of the silica. Then the PdO NPs are trans-
formed into metallic Pd0 NPs by reduction under H2 at 280 °C.
The reduction process changes the color of the monolith from
yellow (PdO NPs) to brown (Pd0 NPs in the monolith) (Fig. 2).
EDX analysis carried out on a cross section of the entire mono-
lith confirms this observation with evenly distributed Pd
atoms with a calculated amount of 1.28 wt%, in accordance
with previous UV/Vis analysis of the supernatant solutions
during impregnation.
Fig. 3 Nitrogen sorption isotherms at 77 K of (black) the parent silica monolith
and (grey) Pd-MonoSil-3.
The nitrogen sorption isotherm at 77 K of Pd-MonoSil-3
(after reduction) shows that the silica mesopores network of the
monolith was kept intact throughout the synthesis of Pd NPs in
the monolith (Fig. 3). In comparison to the parent silica mono-
lith the surface area decreases slightly from 523 to 490 m2 g−1
,
the mesoporous volume from 1.19 to 1.18 mL g−1 and the meso-
pore diameter increases slightly from 11 to 12 nm.
Fig. 4 SEM (a) and TEM (b) images of Pd-MonoSil-3. Arrows indicate the Pd
NPs in the skeleton of the silica monolith. (c) Particle size distribution of Pd NPs
in Pd-MonoSil.
From TEM images it can be clearly seen that Pd-MonSil-3
contains well-dispersed Pd NPs featuring particle sizes with
average diameters of 6–7 nm within the silica monolith
(Fig. 4). No particles larger than 11 nm have been detected,
which indicates that the Pd NPs were formed within the meso-
pores (11–12 nm) of the silica monolith. The SEM image of the
Pd-MonoSil-3 shows that the homogeneous and isotropic inter-
connected network of macropores (ca. 4 microns) was kept
intact during the formation of Pd NPs within the monolith
(Fig. 4).
3.2. Selective hydrogenation under continuous flow with
Pd-MonoSil
Fig. 1 Amount of impregnated Pd(NH3)4(NO3)2 as a function of pH of the
impregnation solution.
Before performing catalytic reactions, Pd-MonoSil-3 was
washed under a flow of 0.15 mL min−1 of methanol. No Pd
leaching was evidenced by ICP-OES analysis of the solutions.
Pd-MonoSil-3 was used as a catalyst in the hydrogenation
under continuous flow at room temperature and low H2
pressure (2–3 bars) of 1,5-cyclooctadiene (COD) and 3-hexyn-
1-ol using methanol as a solvent (Scheme 1).
The efficiency of Pd-MonoSil-3 (1.3 Pd wt%) was first tested
for the selective hydrogenation of COD in cyclooctene (COE).
As shown in Fig. 5, under the test conditions a constant con-
version of 95% and a constant selectivity in COE of 90% were
obtained after ca. 2 hours and a half under flow. The conver-
sion and the selectivity are controlled by tuning H2 and sub-
strate flow rates (Fig. 6). Increasing H2 flow rate increases the
Fig. 2 Pd-MonoSil-3 after calcination at 550 °C (a) and after reduction under
H2 at 280 °C (b).
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 1378–1384 | 1381

View Article Online
Paper
Dalton Transactions
content was below the detection limit of the ICP-OES instru-
ment, i.e. 0.006 ppm, thus indicating that the Pd loss is negli-
gible even after 70 h time-on-stream.
In that time interval only a slight decrease in conversion
from 95 to 90% was noticed. Similar conversion and selectivity
levels (95–100 and 92–94%, respectively) were obtained using a
Pd@alumina membrane reactor for 1,5-cyclooctadiene selec-
tive hydrogenation in flow at 50 °C and 10 bars of hydrogen
pressure.23 The pore-flow-through monolithic reactor loaded
with Pd NPs allows one to reach the same performance under
softer conditions (25 °C, 2.6 bars) and, above all, demonstrates
unique stability with time on stream definitely enabling con-
tinuous process operation.
Scheme 1 Model reactions used to evaluate the effectiveness of the pore-
flow-through silica monolith in selective hydrogenation: (up) 1,5-cyclooctadiene
and (down) 3-hexyn-1-ol.
The second reaction investigated using Pd-MonoSil is the
stereo-selective hydrogenation of 3-hexyn-1-ol, a reaction of
prime importance carried out very efficiently in industry in
batch mode using the Lindlar catalyst (Pd NPs deposited on
calcium carbonate) (89% selectivity in alkene). A main short-
coming in this reaction is the use of toxic additives as Pb,23
which can further be found in food products. There is there-
fore a need for greener catalysts for this reaction. Different
additives have been tested on Pd supported catalysts such as
quinoline, carbon monoxide, organic bases, sulfides, metal
ions such as copper, which are necessary to reach high stereo-
selectivity in this reaction, namely 99% cis-selectivity and 97%
yield in 3-hexen-1-ol in batch mode.21a
Fig. 5 Hydrogenation of 1,5-cyclooctadiene (0.1 M in methanol) over Pd-
MonoSil-3 (1.3 wt Pd%) at room temperature at a H2 pressure of 2.6 bars under
continuous flow (H2 flow rate 0.5 mL min−1, substrate flow rate 0.12 mL min−1).
In order to determine whether this reaction could be
implemented in continuous mode, we have evaluated the per-
formance of the Pd-MonoSil (1.3 wt% Pd) microreactor in the
selective hydrogenation of 3-hexenyl-1-ol at room temperature
with a hydrogen flow rate of 0.7 mL min−1 and a substrate flow
rate of 0.15 mL min−1. Under these conditions a constant con-
version level equal to ca. 85% was reached after 3 h under flow
while the selectivity into 3-hexen-1-ol (cis + trans) increased
steadily from 70 till 80% during the 7 h of the test (Fig. 7). The
substrate conversion and the selectivity in alkene products can
be tuned by adjusting the flow rates of the substrate solution
Fig. 6 Hydrogenation of 1,5-cyclooctadiene (0.1 M in methanol) over Pd-
MonoSil-3 (1.3 wt Pd%) at room temperature: (a) substrate flow rate 0.12 mL and of H2 gas. Complete conversion of 3-hexyn-1-ol was
min−1 and variable H2 flow rate (variable H2/substrate); (b) constant H2/sub-
achieved by increasing H2 flow rate, but the selectivity in
alkenes decreased to 40%. As expected increasing H2 flow rate
increases the conversion by adding more H2 in contact with
strate molar ratio = 4 and variable substrate and H2 flow rates; (c) selectivity in
the cyclooctene product as a function of conversion of cyclooctadiene in the
same experimental conditions as in (a).
the substrate and Pd NPs but decreases the selectivity by
forming the 3-hexanol dihydrogenated product. The cis-
conversion and changes marginally the selectivity, with a
slight decrease noticed from 94 to 89%. Increasing substrate
flow rate (at the same H2 to substrate ratio of 4) decreases
slightly the conversion from 92 to 80% due to a decrease in
contact time of cyclooctadiene with the catalyst, whilst the
selectivity increases from 82 to 94%. The best compromise
between conversion and selectivity was reached with a flow
rate of substrate of 0.12 mL min−1 and a H2 flow rate of 0.5 mL
min−1 giving 95% conversion and 90% selectivity. The pro-
ductivity of the reaction was 4.2 mmol s−1 g−1
or
MonoSil
0.32 mol s−1 g−1Pd. The stability of the catalyst proved remark-
able as conversion and selectivity were maintained nearly con-
stant over a period of 70 h. During the whole reaction Pd
Fig. 7 Hydrogenation of 3-hexyn-1-ol (0.2 M in methanol) over Pd-MonoSil-3
(1.3 wt Pd%) at room temperature and 2.8 bars H2 under continuous flow (H2
flow rate 0.7 mL min−1, substrate flow rate 0.15 mL min−1
.
1382 | Dalton Trans., 2013, 42, 1378–1384
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
microreactors for the continuous selective hydrogenation at
room temperature of 1,5-cyclooctadiene and 3-hexyn-1-ol
model substrates. In both reactions, the catalytic microreactor
demonstrates remarkable stability of the activity with time on
stream. The substrate conversion level as well as the selectivity
into the target product can be readily controlled by indepen-
dently adjusting the hydrogen and substrate flow rates. In the
selective hydrogenation of 1,5-cyclooctadiene, a conversion of
95% and a selectivity of 90% in monohydrogenated product,
constant over a period of 70 h, have been reached. These
figures correspond to a productivity of 4.2 mmol s−1 g−1
MonoSil
(or 0.32 mol s−1 g−1Pd) and compare well with benchmark
membrane based catalysts. In the stereoselective hydrogen-
ation of 3-hexyn-1-ol a constant conversion of 85% and a selec-
tivity of 80% into the cis isomer have been obtained over a
period of 7 h, corresponding to a 10 times higher productivity
in comparison to batch using the Lindlar catalyst. The versati-
lity and flexibility of the monolithic microreactor in triphasic
systems have been demonstrated and open new perspectives
for the implementation of fine chemicals synthesis in continu-
ous flow mode.
Fig. 8 Hydrogenation of 3-hexyn-1-ol (0.2 M in methanol) over Pd-MonoSil-3
at room temperature under continuous flow with (a) substrate flow rate
0.15 mL min−1 and variable H2 flow rate (variable H2/substrate); (b) constant
H2/substrate = 2.7 molar ratio and variable substrate and H2 flow rates; (c)
alkene selectivity and cis-selectivity (cis-3-hexen-1-ol) as a function of conversion
of 3-hexyn-1-ol hydrogenation (same experimental conditions as in (a)).
selectivity decreases from 88 to 64% upon increasing the
hydrogen flow rate from 0.2 to 1.3 mL min−1 (Fig. 8c). Higher
selectivity in alkene products was achieved, up to 80%, by
increasing substrate flow rate at a constant H2/substrate molar
ratio with a minimal decrease of the conversion (72%) (Fig. 8).
An increase in the solution flow rate (i.e. a decrease of the resi-
dence time from 420 to 140 s while keeping a constant H2/sub-
strate molar ratio) invariably led to an increase in partial
hydrogenation. For a constant conversion, the cis-selectivity
remains constant over time. The best conversion and selectiv-
ity in cis-3-hexen-1-ol was achieved at a H2 flow rate of 0.7 mL
min−1 and a substrate flow rate of 0.15 mL min−1 and gives a
Acknowledgements
The research leading to these results has received funding
from the European Community’s Seventh Framework Program
through the Marie-Curie Initial Training Network NANO-HOST
(Grant Agreement ITN 215193-2). The authors thank Thomas
Cacciaguerra for SEM and TEM pictures.
Notes and references
productivity of 9.6 mmol s−1 g−1MonoSil or 0.75 mol s−1 g−1
.
Pd
Although the performances in terms of product selectivity
stand below those of the Lindlar catalyst these preliminary
results highlight the high potential and the flexibility of the
Pd-MonoSil operated in continuous flow for a rather complex
selective hydrogenation reaction. Pd-MonoSil can be used
under flow conditions, with enormous benefits from an indus-
trial point of view in terms of long-term productivity and
space-to-time yield (STY): e.g. for 3-hexyn-1-ol hydrogenation,
the STY (calculated from ref. 20e) of the Lindlar catalyst is 26 g
l−1 h−1 vs. 270 g l−1 h−1 for Pd-MonoSil. Furthermore, the
increase of selectivity with time on stream with Pd-MonoSil
suggests some negative influence of surface sites (silanol
groups), which can be amended by passivation treatments.
The effect of green additives can be readily investigated. Work
in these directions is in progress.
1 A. Kirschning, C. Altwicker, G. Dräger, J. Harders,
N. Hoffmann, U. Hoffmann, H. Schönfeld, W. Solodenko
and U. Kunz, Angew. Chem., Int. Ed., 2001, 40, 3995.
2 (a) A. Cybulski and J. A. Moulijn, Catal. Rev. Sci. Eng., 1994,
36, 179; (b) P. Avila, M. Montes and E. Miró, Chem. Eng. J.,
2005, 109, 11; (c) J. L. Williams, Catal. Today, 2001, 69, 3.
3 (a) M. V. Twigg and J. T. Richardson, Ind. Eng. Chem. Res.,
2007, 46, 4166; (b) A. Sachse, A. Galarneau, F. Fajula, F. Di
Renzo, F. Fajula and B. Coq, Microporous. Mesoporous.
Mater., 2011, 140, 58; (c) A. Sachse, A. Galarneau, F. Di
Renzo, F. Fajula, F. Fajula and B. Coq, Chem. Mater., 2010,
22, 4123; (d) N. Brun, A. Babeau Garcia, M.-F. Achard,
C. Sanchez, F. Durand, G. Laurent, M. Birot, H. Deleuze
and R. Backov, Energy Environ. Sci., 2011, 4, 2840.
4 (a) L. Kiwi-Minsker, Chimia, 2002, 56, 143; (b) R. M. Heck,
S. Gulati and R. J. Farrauto, Chem. Eng. J., 2001, 82, 149.
5 See as review: A. Sachse, A. Galarneau, B. Coq and
F. Fajula, New J. Chem., 2011, 35, 259.
6 (a) F. Svec and J. M. J. Fréchet, Anal. Chem., 1992, 64, 820;
(b) C. Viklund, F. Svec, J. M. J. Fréchet and K. Ignum, Chem.
Mater., 1996, 8, 744; (c) E. C. Peters, F. Svec and
J. M. J. Fréchet, Adv. Mater., 1999, 11, 1169; (d) N. Hird,
4. Conclusion
Pore-flow-through silica monoliths featuring a bimodal distri-
bution of macropores (4–5 μm) and mesopores (11 nm) have
been loaded with Pd nanoparticles and implemented as
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 1378–1384 | 1383

View Article Online
Paper
I. Hughes, D. Hunter, M. G. J. T. Morrison,
D. C. Sherrington and L. Stevenson, Tetrahedron, 1999, 55,
9575; (e) J. A. Tripp, J. A. Stein, F. Svek and J. M. J. Fréchet,
Org. Lett., 2000, 2, 195; (f) J. A. Tripp, F. Svec and
J. M. J. Fréchet, J. Comb. Chem., 2001, 3, 216.
7 (a) U. Kunz, A. Kirschning, H.-L. Wen, W. Solodenko,
R. Cecilia, C. O. Kappe and T. Turek, Catal. Today, 2005,
105, 318; (b) K. Mennecke, R. Cecilia, T. N. Glasnov,
Dalton Transactions
6840; (c) P. Barbaro, C. Bianchini, V. Dal Santo, A. Meli,
S. Moneti, R. Pasaro, A. Scaffidi, L. Sordelli and F. Vizza,
J. Am. Chem. Soc., 2006, 128, 7065; (d) J. Mandal, D. Roy,
R. V. Chaudhari and M. Sastry, Chem. Mater., 2004, 16, 3714.
16 (a) G. J. Hutchings, J. Mater. Chem., 2009, 19, 1222; (b) The
Handbook of Homogeneous Hydrogenation, ed. J. G. de Vries
and C. J. Elsevier, Wiley-VCH, Weinheim, Germany, 2007,
Vol. 1.
S. Gruhl, C. Vogt, A. Feldhoff, M. A. L. Vargas, C. O. Kappe, 17 See. e.g.: (a) E. Sulman, V. Matveeva, V. Doluda,
U. Kunz and A. Kirschning, Adv. Synth. Catal., 2008, 350,
717; (c) W. Solodenko, U. Kunz, G. Jas and A. Kirschning,
Bioorg. Med. Chem. Lett., 2002, 12, 1833; (d) W. Solodenko,
H. Wen, S. Leue, F. Stuhmann, G. Sourkouni-Argirusi,
L. Nicoshvili, L. Bronstein, P. Valetsky and I. Tsvetkova,
Top. Catal., 2006, 39, 187; (b) N. Marín-Astorga, G. Pecchi,
T. J. Pinnavaia, G. Alvez-Manoli and P. Reyes, J. Mol. Catal.
A: Chem., 2006, 247, 145.
G. Jas, H. Schönefeld, U. Kunz and A. Kirschning, 18 (a) G. Alvez-Manoli, T. J. Pinnavaia, Z. Zhang, D. K. Lee,
Eur. J. Org. Chem., 2004, 3601; (e) K. Mennecke and
A. Kirschning, Beilstein J. Org. Chem., 2009, 5, 1.
8 A. El Kadib, R. Chimenton, A. Sachse, F. Fajula,
A. Galarneau and B. Coq, Angew. Chem., Int. Ed., 2009, 48,
4967.
9 (a) K. Nakanishi and N. Soga, J. Non-Cryst. Solids, 1992,
139, 1; (b) J. Babin, J. Iapichella, B. Lefevre, C. Biolley,
J.-P. Bellat, F. Fajula and A. Galarneau, New J. Chem., 2007,
K. Marín-Astorga, P. Rodriguez, F. Imbert, P. Reyes and
N. Marín-Astorga, Appl. Catal., A: Gen., 2010, 387, 26;
(b) F. Zaera, Acc. Chem. Res., 2009, 42, 1152;
(c) L. A. Saudan, Acc. Chem. Res., 2007, 40, 1309.
19 (a) J. F. Sanchez, M. D. E. Boldrine, G. M. Tonetto and
D. E. Damiani, Chem. Eng. J., 2011, 167, 355; (b) A. F. Perez-
Cadenas, F. Kapteijn, M. M. P. Zieverink and J. A. Moulijn,
Catal. Today, 2007, 128, 13.
31, 1907; (c) A. Galarneau, J. Iapichella, D. Brunel, 20 (a) M. P. R. Spee, J. Boersma, M. D. Meijer, M. Q. Slagt,
F. Fajula, Z. Bayram-Hahn, K. K. Unger, G. Puy,
C. Demesmay and J. L. Rocca, J. Sep. Sci., 2006, 29, 844.
10 (a) A. Sachse, V. Hulea, A. Finiels, B. Coq, F. Fajula and
A. Galarneau, J. Catal., 2012, 287, 62; (b) A. Sachse,
R. Ameloot, B. Coq, F. Fajula, B. Coasne, D. De Vos and
A. Galarneau, Chem. Commun., 2012, 48, 4749.
G. van Koten and J. W. Geus, J. Org. Chem., 2001, 66, 1647;
(b) J. Warr, S. Fraser and O. Gouault (Takasago Perfumery
Co., Ltd) Pat. EP, 1964542, 2008; (c) H. Lindlar, Helv. Chim.
Acta, 1952, 35, 447; (d) H. Bönnemann, W. Brijoux,
A. Schultze Tilling and K. Siepen, Top. Catal., 1997, 4, 217;
(e) H. Bönnemann, W. Brijoux, K. Siepen, J. Hormes,
R. Franke, J. Pollmann and J. Rothe, Appl. Org. Chem.,
1997, 11, 783.
11 (a) P. He, S. J. Haswell, P. D. I. Fletcher, S. M. Kelly and
A. Mansfield, Beilstein J. Org. Chem., 2011, 7, 1150;
(b) S. Sato, R. Takahashi, T. Sodesawa and M. Koubata, 21 (a) A. Molnar, A. Sarkany and M. Varga, J. Mol. Catal. A:
Appl. Catal., A: Gen., 2005, 284, 247.
12 R. Takahashi, S. Sato, T. Sodesawa, K. Arai and M. Yabuki,
J. Catal., 2005, 229, 24.
13 R. Takahashi, S. Sato, S. Tomiyama, T. Ohashi and
N. Nakamura, Microporous Mesoporous. Mater., 2007, 98,
107.
14 M. Numata, R. Takahashi, I. Yamada, K. Nakanishi and
S. Sato, Appl. Catal., A: Gen., 2010, 383, 66.
Chem., 2001, 173, 185; (b) C. Godínez, A. L. Cabanes and
G. Víllora, Chem. Eng. Proc., 1995, 34, 459.
22 (a) A. Schmidt, A. Wolf, R. Warsitz, R. Dittmeyer,
D. Urbanczyk, I. Voigt, G. Fischer and R. Schomäcker,
AIChE J., 2008, 54, 258; (b) J. Caro, et al., Ind. Eng. Chem.
Res., 2007, 46, 2286; (c) N. Numata, R. Takahashi,
I. Yamada, K. Nakanishi and S. Sato, Appl. Catal., A: Gen.,
2010, 383, 66.
15 (a) O. Dominguez-Quintero, J. Martinez, Y. Henriquez, 23 F. Rouquerol, J. Rouquerol and K. Sing, Adsorption by
L. D’Ornelas, H. Krenzien and J. Osuna, J. Mol. Catal. A: Powders & Solids, Academic Press, 1999.
Chem., 2003, 197, 185; and ref herein; (b) Y. Niu, 24 A. Galarneau, D. Desplantier, R. Dutartre and F. Di Renzo,
L. K. Yeung and R. M. Crooks, J. Am. Chem. Soc., 2001, 123,
Microporous Mesoporous Mater., 1999, 27, 297.
1384 | Dalton Trans., 2013, 42, 1378–1384
This journal is © The Royal Society of Chemistry 2013