PdAlqEn: A novel upgraded version of the PdEnCat family of polyurea encapsulated catalysts

Article abstract of DOI:10.1002/adsc.200800055

Palladium nanoparticles are stabilized and encapsulated in a combination of polyurea and an ionic liquid (Aliquat 336) to obtain the PdAlqEn hybrid catalyst. The novel composite material is applied as a recyclable and robust catalyst in hydrogenation reactions.

Full text of DOI:10.1002/adsc.200800055

COMMUNICATIONS
DOI: 10.1002/adsc.200800055
TM
PdAlqEn: A Novel Upgraded Version ofthe PdEnCat
ofPolyurea Encapsulated Catalysts
Family
Judith Toubiana,a Mandan Chidambaram,a Asaf Santo,a and Yoel Sassona,*
a
Casali Institute of Applied Chemistry, The Hebrew University of Jerusalem, Jerusalem 91904, Israel
Phone: (+972)-2-652-9626; fax: (+972)-2-652-9626; e-mail: ysasson@huji.ac.il
Received: January 24, 2008; Published online: May 13, 2008
Abstract: Palladium nanoparticles are stabilized
and encapsulated in a combination of polyurea and
an ionic liquid (Aliquat 336) to obtain the PdAlqEn
hybrid catalyst. The novel composite material is ap-
plied as a recyclable and robust catalyst in hydroge-
nation reactions.
thors verified that swift catalytic reaction took place
with the resin bound aryl iodide unquestionably sub-
stantiating that a sufficient amount of active catalyst
leached out of the polymeric matrix under the reac-
tion conditions. An even stronger argument was put
forward by Richardson and Jones[3] who tested PdEn-
Cat40TM in the Heck reaction where specific polymer-
ic palladium catalyst poisons, namely poly(4-vinylpyri-
dine) and Quadrapure TU (a polymer-bound thiourea
derivative) were added to the reaction mixtures under
various conditions. These polymers inhibit only solu-
ble catalytic species and cannot affect the encapsulat-
ed catalyst. Since no reaction could be detected in the
presence of each of these resins it was unequivocally
Keywords: Aliquat 336; hydrogenation; microen-
capsulation; PdAlqEn catalyst; PdEnCatTM; poly-
urea
Microencapsulation is one of the most promising concluded that the polyurea encapsulated palladium
techniques for catalyst immobilization. The group of is inactive in the Heck reaction and that the observed
commercial catalysts based on palladium(II) acetate activity was a result of palladium leaching to the solu-
entrapped in a polyurea (PU) matrix, reported by tion.
Ley and co-workers[1] (marketed under the trade
We have now examined Pd(0)EnCat30TM (pur-
name PdEnCatTM), has been described as versatile, chased from Aldrich) as catalyst in the simple hydro-
active, robust and recycleable catalyst. These catalysts genation reaction of 4-phenyl-3-butene-2-one to 4-
À
were successfully applied in various cross-coupling C
C bond forming reactions such as the Suzuki, Stille
and the Heck reactions. Some of the latter processes
were carried out in supercritical CO2 as solvent and
were also demonstrated in a laboratory scale continu-
ous flow system. In an alternative version, the encap-
sulated catalyst was reduced to Pd(0) nanoparticles
and this hybrid catalyst showed activity in hydrogena-
phenyl-2-butanone [(Eq. (1)] at 708C in toluene and
tion and hydrogen transfer reactions. Only a negliga- in methanol under 1 bar hydrogen pressure. A
ble amount of catalyst leaching to solution was re- smooth hydrogenation reaction was realized (GC
ported in these reactions.
analysis) with a complete conversion of the substrate
The robustness and recyclability, as well as the after 40 min with 100%. The measured TOF was
mode of action of the PdEnCatTM were recently chal- 0.027 secÀ1. Upon addition of poly(4-vinylpyridine) or
lenged in two independent studies using different ex- Quadrapure TU to the reaction mixture (poisons that
perimental
methodologies.
Broadwater
and inhibit soluble catalyst only) the hydrogenation rate
McQuade[2] utilized the “three-phase” test to assess fell to 30% of its original level. This clearly suggests
the activity of PdEnCat in the Suzuki coupling and that, similar to the observations made in the Heck
the Heck reactions where one of the substrates (aryl and Suzuki reactions, the major part of the catalytic
iodide) was covalently attached to a resin. The latter activity in this system is taking place in solution and
polymeric substrate could obviously react only when not within the polyurea microcapsules.
a soluble catalyst is present in the system. Testing
three different commercial PdEnCat catalysts, the au- energy dispersive X-ray spectroscopy (EDS) com-
Examining particles of the PdEnCatTM catalyst by
1230
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1230 – 1234

PdAlqEn: A Novel Upgraded Version of the PdEnCatTM Family
COMMUNICATIONS
Figure 1. Morphology of Pd(0)EnCat30TM. Left, fresh catalyst (”400); middle and right, after hydrogenation reaction of Eq.
(1) (”200) (”1000).
bined with scanning electron micrography (SEM) in the actual catalyst in all the reactions where PdEnCat
backscattered mode, before and after the catalytic was used is Pd(0) and not Pd(II). Consequently we
process, revealed that a dramatic change in the mani- have deliberately reduced (using sodium borohydride)
festation of the polyurea spheres is taking place in the the palladium(II) to stabilized Pd(0) nanoparticles
course of the reaction. This is shown in Figure 1 concurrently with the interfacial polymerization.
where Pd(0)EnCat30 is exposed prior to the hydroge-
Using the protocol described below in the Experi-
nation reaction (left, magnification ”400) and after mental Section we have prepared a novel version of
filtration and separation from the hydrogenation mix- the polyurea encapsulated Pd(0) catalyst which we
ture (middle, magnification ”200 and right, magnifica- entitle PdAlqEn. The latter was characterized to have
tion ”1000). It is evident that the spherical capsules average particle diameter of 4 mm and high surface
are cracked and fractured during the reaction and area (BET 1.50 m2 g) (compared, respectively, to
that the palladium particles (appearing as white 250 mm and 0.07 m2 g for PdEnCat30TM). The porosity
stains) are exposed to the exterior. XPS analyses re- of PdAlqEn was determined to be 5.45 mm3 g. TEM
vealed that the surface mass concentration of Pd on analysis of the polyurea matrix showed evenly distrib-
the original catalyst particles was 4.7% while on the uted palladium nanoparticles with a uniform diameter
surface of the catalyst recovered from the hydrogena- of 2 nm. The palladium content was found to be
tion reaction the surface mass concentration of Pd tri- 0.15 mmol/g of catalyst by ICP analysis.
pled to 11.2%. Our immediate conclusion was that
PdAlqEn was tested in the reaction of Eq. (1) and
the polyurea matrix utilized for the fabrication of the was confirmed to perform better than the PdEnCat
PdEnCatTM is not robust and stable as originally catalyst, achieving complete conversion after 10 min
claimed and that the material breaks down almost in- under
stantly upon exposure to the very mild catalytic con- 0.11 secÀ1). The performance of our catalyst was ac-
ditions. tually similar to that of 10% Pd/C catalyst containing
identical
reaction
conditions
(TOF=
We alleged that the main drawback of the PdEnCat the same number of Pd atoms. PdAlqEn could be
stems from the interfacial encapsulation process simply and swiftly filtered from the reaction mixture
which evidently creates a weak and fragile shell with and recycled to a new hydrogenation reaction of Eq.
low porosity and small surface area.[4] Assessing that (1) without any apparent loss in catalytic activity even
the latter is a result of excess of surface active agents after five consecutive cycles (Figure 2).
remaining in the particlesꢁ crust, we have removed all
Moreover, SEM analysis of PdAlqEn before and
the three emulsifiers and dispersing agents advocated after reaction (Figure 3), in methanol or toluene as
by Ley et al. and replaced them with a single reagent solvents, revealed that the catalyst is stable under the
– tricaprylmethylammonium chloride (Aliquat 336) reaction conditions and its physical appearance and
which, as will be shortly described, not only served as morphology do not change at all even after five con-
effective emulsifier in the encapsulation process but secutive runs in hydrogenation of 4-pheynl-3-buten-2-
also played a critical and multidimensional role in the one [Eq. (1)]. No change in the initial surface concen-
preparation and performance of the immobilized cat- tration of Pd (0.05%) was monitored after the reac-
alyst.
In addition, in view of the recent critical reviews by
tion, based on XPS analysis.
Only a negligible decline (1–3%) in the catalytic ac-
Jones,[5] by DeVries[6] and by Finke[7] we inferred that tivity of PdAlqEn in the reaction of Eq. (1) was ob-
Adv. Synth. Catal. 2008, 350, 1230 – 1234
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1231

Judith Toubiana et al.
COMMUNICATIONS
ions[12] and H2O molecules[13] into the oil phase (see
Experimental SecÀtion). We perceived that the pres-
ence of both BH4 and water in the organic phase is
essential for the rapid conversion of the isocyanate
function into amine thus allowing the polymerization
to take place. In the absence of either Aliquat 336 or
of borohydride the polymerization is very sluggish
and ineffective.
A
of PdCl4 and of BH4 leads to rapid reduction of the
Pd(II) species to Pd(0) nanoparticles. The lipophilic
ammonium salt then stabilizes[14] the palladium nano-
particles in the polyurea matrix. NMR analysis, ele-
mentary analysis and mass balance calculations con-
firmed that the Aliquat 336 completely remains
within the polymeric matrix after the preparation and
after several consecutive catalytic runs.
Figure 2. PdAlqEn catalyst recycle experiments in the reac-
tion of Eq. (1).
For further understanding of the nature of palladi-
um metal and its distribution in the matrix of PdAl-
qEn, we carried out a transmission electron microsco-
served when either of the poisons PVP, Quadrapure py (TEM) study (Figure 4, A) and energy dispersive
TU or elemental mercury[8] was added to the reacting X-ray spectroscopy (EDAX) coupled with scanning
mixture. This clearly suggests that the catalytic activi- tunneling electron microscopy (STEM) (Figure 4, B).
ty is limited to the bulk of the polymeric matrix and These analyses identified the presence of the palladi-
no part of the catalyst is active in solution.
um metal, the nature of the nanoparticles and the
The crucial component in the new proposed formu- average particle size as 2 nm and also revealed the
lation of polyurea microencapsulation is the quaterna- uniform distribution of palladium nanoparticles
ry ammonium salt, Aliquat 336[9] which is also an within the polymer matrix.
ionic liquid. We explain the unique multifunctional
role of this ingredient as follows.
Moreover, the analysis by TEM and by electron dif-
fraction showed a 1D fringe with a d spacing of about
(i) A surface active agent[10] that supports the for- 0.23 nm (Figure 4, C). Since the common structure of
mation of water/toluene emulsion in which the inter- Pd metal is cubic close packed with the unit cell pa-
facial polymerization is performed. We have mea- rameter equal to 0.389nm and d spacing equal to
sured a dramatic decrease in the interfacial water/tol- 2.24 Š corresponding to the crystal plane,[1e] this
uene pressure in the presence of Aliquat 336.
(ii) ÀA phase-transfer vehicle that initially extractÀs palladium(II) to palladium(0). Further, we visualize
PdCl4 ions[11] and in the second step extracts BH4
the microstructure of the catalyst as an array of Ali-
result confirms the effective reduction of the initial
Figure 3. Morphology of PdAlqEn catalyst. Left, fresh catalyst (”9316); right, filtered and dried catalyst after five consecu-
tive hydrogenation runs [Eq. (1)] (”9600).
1232
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1230 – 1234

PdAlqEn: A Novel Upgraded Version of the PdEnCatTM Family
COMMUNICATIONS
Figure 4. (A) TEM image of PdAlqEn shows the uniform distribution of palladium nanoparticles of 2 nm. (B) STEM image
shows the uniform distribution of palladium in the spherical microcapsule. White dots represent the palladium nanoparticles.
(C) TEM image through electron diffraction reveal the fundamental state of palladium [Pd(0)].
quat 336 (actually an ionic liquid) globules evenly
Other transition metals such as cobalt, ruthenium,
scattered within the polyurea matrix containing the rhodium and gold could be similarly encapsulated as
immobile (thus stabilized) Pd particles (Figure 4, C). stabilized nanoparticles in polyurea matrix. Their ap-
The matrix is permeable to the substrates and the plication as catalysts will be published separately.
products but impassable to the Aliquat/Pd complex.
In conclusion, we have demonstrated that polyurea
PdAlqEn was successfully tested in several hydro- resin combined with ionic liquid (Aliquat 336) is an
genation and hydrogenolysis reactions. Typical exam- excellent and robust support for catalytic nanoparti-
ples are shown below [(Eq. (2) and Eq. (3)]. In all cles of palladium and other transition metals.
Experimental Section
Synthesis ofPdAlqEn Ccatalyst
Aliquat 336 (0.5 g, 1 mmol) and 5 g of toluene were mixed
in a 50-mL flask. To this solution 0.07 g (0.25 mmol) of
Na2PdCl4 was added and the mixture was stirred at 508C for
10 min. Then 1.2 g (5 mmol) of PMDI were added to the
above solution which was stirred for 5 min. This mixture was
slowly added to 100 g (5.5 mol) of water containing 0.05 g
(1 mmol) of sodium borohydride. Instant and rapid polymer-
ization was observed and stirring was continued for 2 h
using an overhead mechanical stirrer at 200 rpm at room
temperature. The resulting microcapsules were filtered
through a polyethylene frit (20 micron porosity), washed by
de-ionized water and by dichloromethane and dried at
1008C for 4 h to obtain the PdAlqEn(0) catalyst, a brown
solid powder.
these examples no leaching of metal was detected and
the catalyst could be recycled five times without any
apparent loss in activity.
To understand the recycle efficiency of catalyst and
palladium loss in the reaction of Eq. (2), under identi-
cal reaction conditions, a 100% yield of aniline was
obtained from the first cycle to the fifth cycle and
Representative Procedure for Hydrogenation of
Nitroaromatics using PdAlqEn Catalyst
also we found no loss of palladium from the matrix, Toluene (5 g) was charged in a round-bottom flask and
1 mmol of substrate and 100 mg (0.015 mmol Pd content) of
catalyst were added and heated to 708C in a preheated oil
bath under stirring. Hydrogen was bubbled in to the mixture
through a tube which was connected to a hydrogen cylinder.
After stirring the whole mixture for 20 min, a small sample
was removed, diluted with dichloromethane and injected in
a gas chromatograph [HP 5890 GC, a 95%-dimethyl-5%-di-
phenyl packed column (25 m length and 0.53 mm ID)]. The
GC injection program was: initial temperature 908C,
(1 min), ramp at 158C/min, final temperature 2808C
(5 min). Helium was used as carrier gas at a pressure of
which was confirmed by ICP analysis of the catalyst
after every cycle. To substantiate this observation fur-
ther, the reaction solution (filtrate after removal of
catalyst) was subjected to the same reaction by the
addition of a second batch of substrate (nitroben-
zene). We found that no reaction took place under
these conditions, which undoubtedly further confirm
that there is no palladium leaching from the polyurea
matrix. The same observation was seen in the reaction
of Eq. (3).
Adv. Synth. Catal. 2008, 350, 1230 – 1234
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1233

Judith Toubiana et al.
COMMUNICATIONS
50 kPa. The GC was calibrated using biphenyl as internal
standard. The yields of the products were calculated based
on the area percent of the products and confirmed by inject-
ing authentic samples.
[2] S. J. Broadwater, D. T. McQuade, J. Org. Chem. 2006,
71, 2131.
[3] J. M. Richardson, C. W. Jones, Adv. Synth. Catal. 2006,
348, 1207.
For the recycle study, prior to each cycle, the catalyst was
separated from the reaction mixture by filtration using
Whatman 40 and washed with an excess amount of water
and acetone and dried at 1008C for 2 h. The obtained cata-
lyst was subjected to SEM and ICP analyses to confirm the
catalyst morphology and concentration of palladium subse-
quently used for next cycle.
[4] The origin of the PU interfacial microencapsulation
technique is in the agrochemical industry where this
methodology is used for controlled release of insecti-
cides. In this particular application one would indeed
requires a certain rate of discharge of the active ingre-
dient through the capsule walls; see: G. J. Mars, H. B.
Scher, Controlled delivery of crop protecting agents,
(Ed.: R. M. Wilkens), Taylor and Francis, London,
1990, p 65.
Acknowledgements
[5] a) N. T. S. Phan, M. Van der Sluys, C. W. Jones, Adv.
Synth. Catal. 2006, 348, 609; b) M. Weck, C. W. Jones,
Inorg. Chem. 2007, 46, 1865.
[6] J. G. DeVries, Dalton Trans. 2006, 421.
[7] J. A. Widegren, R. G. Finke, J. Mol. Catal A: Chem.
2003, 198, 317; E. E. Finney, R. G. Finke, Inorg. Chim.
Acta 2006, 359, 2879.
MC is grateful to the Pikovski-Valazzy Fund for financial
support.
References
[8] Y. Lyn, R. G. Finke, Inorg. Chem. 1994, 33, 4891.
[9] In the preparation of Pd(0)EnCatTM tetraoctylammoni-
um bromide (TOAB) has been used to stabilize the Pd
nanoparticles prepared prior to the polymerization
step. However, the latter was carefully washed out
from the encapsulated catalyst before it was used. Note
that the surface activity of TOAB, according to our
measurements, is an order of magnitude lower than
that of Aliquat 336.
[10] D. Mason, S. Magdassi, Y. Sasson, J. Org. Chem. 1990,
55, 2714.
[11] Y. Sasson, A. Zoran, J. Blum, J. Mol. Catal. 1981, 11,
293.
[1] a) C. Ramaro, S. V. Ley, S. C. Smith, I. M. Shirley, N.
DeAlmeida, Chem. Commun. 2002, 1132; b) S. V. Ley,
C. Ramarao, R. S. Gordon, A. B. Holmes, A. J. Morri-
son, I. F. McConvey, I. M. Shirley, S. S. Smith, M. D.
Smith, Chem. Commun. 2002, 1134; c) N. Bremeyer,
S. V. Ley, C. Ramarao, I. M. Shirley, S. C. Smith, Synlett
2002, 1834; d) J-Q. Yu, H-C. Wu, C. Ramarao, J. B.
Spencer, S. V. Ley, Chem. Commun. 2003, 678; e) S. V.
Ley, C. Mitchell, D. Pears, C. Ramarao, J-Q. Yu, W.
Zhou, Org. Lett. 2003, 5, 4665; f) C. Ramarao, D. J. Ta-
polezay, I. M. Shirley, S. C. Smith, S. V. Ley, US Patent
Application 20040254066, 2004; g) C. K. Y. Lee, A. B.
Holmes, S. V. Ley, I. F. McConvey, B. Al-Duri, G. A.
Leeke, R. C. D. Santos, J. P. K. Seville, Chem. Commun.
2005, 2175; h) I. R. Baxendale, C. M. Griffin-Jones,
S. V. Ley, G. K. Tranmer, Chem. Eur. J. 2006, 12, 4407;
i) S. V. Ley, A. J. P. Stewart-Liddon, D. Pears, R. H.
Perni, K. Treacher, Beilstein J. Org. Chem. 2006, 2, 15;
j) D. A. Pears, S. C. Smith, Aldrichimica Acta 2005, 38,
23.
[12] G. D. Yadav, S. V. Lande, J. Mol. Catal. A: Chem. 2006,
247, 253.
[13] a) O. Arrad, Y. Sasson, J. Chem. Soc. Chem. Commun.
1988, 148; b) O. Arrad, Y. Sasson, J. Am. Chem. Soc.
1988, 110, 185.
[14] M. T. Reetz, M. Masse, Adv. Mater. 1999, 11, 9 .
1234
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1230 – 1234