Ossification: A new approach to immobilize metal complex catalysts-applications to carbonylation and Suzuki coupling reactions

Article abstract of DOI:10.1016/j.jcat.2006.05.032

A simple approach for immobilization of transition metal complexes is reported here based on the transformation of the complex into its intrinsically insoluble counterpart, thus generating solid molecular catalysts. This approach that we call "ossification" is based on a principle, in which the water-soluble analogues of the metal complexes are precipitated out from aqueous solutions as insoluble ionic ensembles having catalytically active metal-centered coordination environments and robust framework. The approach has been illustrated for Pd complex catalyzed carbonylation and Suzuki coupling reactions. "Ossification" was found to be an economically and environmentally attractive alternative to other exotic immobilization methodologies.

Full text of DOI:10.1016/j.jcat.2006.05.032

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Journal of Catalysis 242 (2006) 231–238
www.elsevier.com/locate/jcat
Research Note
Ossification: A new approach to immobilize metal complex catalysts—
applications to carbonylation and Suzuki coupling reactions
Bibhas R. Sarkar, Raghunath V. Chaudhari ∗
Homogeneous Catalysis Division, National Chemical Laboratory, Dr. Homi Bhabha Road, Pune, 411008, India
Received 25 March 2006; revised 27 May 2006; accepted 30 May 2006
Available online 11 July 2006
Abstract
A simple approach for immobilization of transition metal complexes is reported here based on the transformation of the complex into its
intrinsically insoluble counterpart, thus generating solid molecular catalysts. This approach that we call “ossification” is based on a principle,
in which the water-soluble analogues of the metal complexes are precipitated out from aqueous solutions as insoluble ionic ensembles having
catalytically active metal-centered coordination environments and robust framework. The approach has been illustrated for Pd complex catalyzed
carbonylation and Suzuki coupling reactions. “Ossification” was found to be an economically and environmentally attractive alternative to other
exotic immobilization methodologies.
© 2006 Elsevier Inc. All rights reserved.
Keywords: Immobilization; Heterogeneous catalysts; Palladium complex; Ossification; Carbonylation; Suzuki coupling
1. Introduction
hybrid solids composed of modified zirconium phosphonates
with pendant chiral ligands holding Ru catalysts for the asym-
metric hydrogenation reactions. These catalysts are efficient,
but the handling of highly sensitive chiral phosphonates, the
longer reaction times (∼20 h), and the tedious and intricate
synthetic methodology of the catalysts are some of the most
important drawbacks of the system. In this report, we present a
very simple and efficient immobilization approach in which the
metal complex is first converted to its water-soluble analogue
by known methods and then transformed to almost completely
insoluble forms.
The homogeneous metal complex catalysts are well known
for their high activity and selectivity for a wide variety of reac-
tions [1,2]. Despite several discoveries of novel metal complex
catalysts and their applications for efficient synthesis of a vari-
ety of products, their utility has been limited due to the difficul-
ties in catalyst–product separation and economic utilization [3].
An obvious solution is to develop an appropriate methodology
for immobilization of the soluble catalysts to give active, selec-
tive, stable, and non-leaching catalysts. Over the years, several
attempts have been made to immobilize homogeneous catalysts
based on the principles of phase differentiation, which led to
the development of biphasic catalysis [4], immobilization using
anchoring, tethering, encapsulation techniques, and precipita-
tion of products/catalysts by solvent selection/manipulation of
physical parameters [3]. In only a few cases have truly hetero-
genized catalysts with commercially acceptable performance
been developed, and the search for new but simple methodolo-
gies is open to investigation. Recently, Lin et al. [5,6] reported
Our new approach of immobilization called ossification in-
volves the preparation of inherently insoluble catalytically ac-
tive molecules by a process, resembling the self-assembly of
robust insoluble biomaterials (e.g., shells, bones, coral reefs)
from simpler soluble precursor molecules and ions. In our two-
step protocol of synthesis, the metal complexes are first made
aqueous-soluble by introducing non-coordinating anionic func-
tionalities such as sulphonate (–SO−3 ), carboxylate (–CO2−), and
others, to any of the ligands by well-established methods [7,8].
This derivatization of the ligand and thus of the complex is
followed by their transformation to solid materials/precipitates
*
Corresponding author. Fax: +91 20 2590 2621.
E-mail address: rv.chaudhari@ncl.res.in (R.V. Chaudhari).
by treatment with the heavier group IIA cations, such as Sr2+
,
0021-9517/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2006.05.032

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
232
B.R. Sarkar, R.V. Chaudhari / Journal of Catalysis 242 (2006) 231–238
Ba2+, and others. These group IIA metal sulphonates are inher-
ently insoluble in water and most of the organic solvents, and
are high-melting and non-subliming solids. Thus, we can ob-
tain molecular solids with intact catalytically active transition-
metal-centered environments (see the schematic in Fig. 1) with
a solid appendage, and thereby a whole new range of mole-
cularly heterogenized metal complex catalyst entities, by an
extremely simple synthesis approach. To prove the general ap-
plicability of the protocol, we have chosen different palladium
complex catalysts for application to different classes of reac-
tions with diverse reaction conditions. We have selected two
important classes of C–C bond formation reactions: hydrocar-
boxylation and Suzuki cross-coupling reactions.
We selected a class of palladium complex catalysts hav-
ing one hemilabile chelating ligand as a benchmark. The N∩O
and N∩N chelating ligands selected were pyca [2-picolinate],
acpy [2-acetylpyridine], pycald [pyridine-2-carboxaldehyde],
and bipy [2,2ꢀ-bipyridyl]. The Pd(pyca)(PPh3)(OTs) and other
similar complexes were synthesized as described previously
[9,10]. The complexes were derivatized by simply replacing
the PPh3 ligand with its –SO3H derivative, triphenylphos-
phine trisulphonate (TPPTS), thus forming aqueous-soluble
counterparts [11,12] [e.g., Pd(pyca)(TPPTS) complex] by a
partitioning technique. The regulated addition of these aque-
ous solutions to a solution of Ba2+ or Sr2+ produced a
pale-yellowish precipitate. The ossified complexes were des-
ignated as ossified-Pd(pyca)(TPPTS) (catalyst 1A), ossified-
Pd(acpy)(TPPTS) (catalyst 1B), ossified-Pd(pycald)(TPPTS)
(catalyst 1C), and ossified-Pd(bipy)(TPPTS) (catalyst 1D).
The catalytic materials may be empirically represented as
Ba3x[Pd(L)(TPPTS)]2y·zH2O, where x, y, z ꢀ 1 (integer) and
L is the bidentate chelating ligand as mentioned earlier. The
characterization of these catalysts and their performances for
carbonylation and Suzuki coupling reactions are discussed be-
low.
another 4 h at 333 K, and then filtered. Residue was subjected
to Soxhlet extraction treatment with water and acetone for 12 h
each and dried in air. All of the other catalysts (catalysts 1B, 1C,
and 1D) were synthesized in the same fashion. The synthesized
catalysts were characterized using various techniques, includ-
ing powder XRD and 31P CP MAS NMR using a Bruker MSL
300-MHz instrument. The 31P NMR spectra were recorded at
202.456 MHz using 85% H3PO4 as an external standard. X-ray
photoelectron spectra (XPS) of the catalysts were recorded on
a VG-Microtech ESCA 3000 spectrometer using the MgKα
emission (E = 1253 eV) under a vacuum of ∼10−9 Torr. Scan-
ning electron microscopy (SEM) was performed using a Philips
XL 30 instrument, with the catalysts suspended in isopropanol
and cast on gold-plated discs, followed by drying under vacuum
and coating with a conducting material.
2.2. Carbonylation reactions
Hydrocarboxylation1 reactions were carried out in a 50-mL
Parr autoclave reactor composed of Hastelloy C-276 material
following a procedure similar to that described previously [9].
For a typical reaction, the solid catalyst and the liquid-phase re-
actant/solvent were charged in the reactor, sealed, and flushed.
The contents were heated to the desired temperature, and then
the reactor was filled with CO to the desired pressure The
progress of the reaction was followed by observing the pres-
sure drop in the CO reservoir as a function of time, and the
reaction was stopped on completion of CO absorption. Af-
terward, the reactor was brought to ambient temperature, its
content gas was released and flushed three times with N2, and
liquid-phase samples were analyzed for reactant/product com-
position using a HP GC 6890 gas chromatograph fitted with a
HP-FFAP capillary column (25 m × 0.33 mm × 0.2 µm). Gas
chromatography–mass spectroscopy (GC-MS) analyses of the
samples were performed using an Agilent 5973 N Mass Selec-
tive Detector attachment.
2. Experimental
2.3. Suzuki cross-coupling reactions
2.1. Ossification of Pd complexes
For a typical Suzuki reaction experiment, the necessary
liquid reactants and solid catalyst were charged in a round-
bottomed flask fitted with a condenser and refluxed for a spec-
ified time. The intermediate samples were collected, and reac-
tion progress was monitored by GC. The final reaction mixture
after cooling was analyzed for the quantitative composition of
reactants and products by GC using a HP GC 6890 device fitted
with a HP-5 MS capillary column (30 m×0.25 mm×0.25 µm).
Products were identified by GC-MS using a Agilent 5973 N
Mass Selective Detector, using a similar capillary column.
A typical two-step ossification process for synthesis of im-
mobilized Pd complexes is described here, using Pd(pyca)-
(PPh3)(OTs) complex as a representative example. In the first
step, Pd(pyca)(PPh3)(OTs) complex (0.357 g, 0.5 mmol) was
dissolved in CHCl3 (7 mL), then partitioned with an aque-
ous solution of TPPTS-Na salt (0.286 g, 0.5 mmol in 10 mL
H2O) to get a yellow solution. Only stoichiometric amount of
TPPTS was used for synthesis of a pure aqueous-soluble com-
plex to avoid any excess of free TPPTS in aqueous solution of
the complex. Washing the aqueous phase with organic solvent
is necessary to remove any adhering organic-soluble precursor
complex; thus, this aqueous phase was washed thoroughly with
CHCl3 to remove all organic-soluble precursors. In the second
step, regulated addition of this aqueous solution to a saturated
aqueous solution of Ba(NO3)2 (10 mL) with vigorous stirring
under argon blanket immediately yielded pale yellow precip-
itate. The slurry was stirred for 2 h at room temperature, for
3. Results and discussion
Catalysts 1A, 1B, 1C, and 1D were characterized to gain in-
sight into the nature of ossified catalysts before evaluation of
1
Hazardous CO involved, authors suggest safe and expert handling during
experiments.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
B.R. Sarkar, R.V. Chaudhari / Journal of Catalysis 242 (2006) 231–238
233
Fig. 1. Schematic representation of ossification methodology showing the characterization strategies. Part 1: Powder XRD pattern of Ba-salt of Pd(pyca)(TPPTS)
31
complex [catalyst 1A]. The numbers inside the graph are the Miller indices of the respective Bragg’s planes. Part 2: P CP MAS NMR spectra of catalysts.
*
(A) Pd(pyca)(PPh )(OTs), (B) ossified Pd(pyca)(TPPTS) complex using barium [catalyst 1A], (C) catalyst 1A after 4 recycles. in NMR spectra represent sidebands
3
at 7 kHz.
their catalytic performance. The powder XRD pattern of the
ossified-Pd(pyca)(TPPTS) (catalyst 1A) complex is shown in
Fig. 1, Part 1. The formation of the Ba-sulphonate catalyst com-
posite (Ba salt of the sulphonated triphenylphosphine coordi-
nated to the Pd center) can be confirmed from the characteristic
d-values, which are similar to those in the literature for inor-
ganic BaSO4, with very slight variation in the intensities of the
principle peaks. The powder XRD patterns of the Ba-ossified
forms of other complexes (catalysts 1B, 1C, and 1D) showed
similar results (Fig. 2, Part I), proving the formation of the Ba–
sulphonate ion pair and generality of the synthesis protocol.
Solid-state 31P CP-MAS NMR experiments were used to inves-
tigate the Pd–P linkage after ossification. The spectrum of the
pure Pd(pyca)(PPh3)(OTs) complex (Fig. 1, Part 2, curve A)
showed an intense peak at δiso 33.44 ppm, corresponding to
the single P atom in the complex. The spectrum for the ossi-
fied counterpart (catalyst 1A) showed a chemical shift of δiso
into the phenyl rings of PPh3, which makes the P atom more
electron-deficient, causing a downfield shift compared with the
organic-soluble counterpart. For the ossified complex, forma-
tion of the solid Ba–sulphonate ion pair and the lattice effects
quench the electron-withdrawing effect of the –SO−3 group; in
addition, the reduced mobility in the solid state creates an up-
field shift of the 31P signal. The spectrum of Ba salt of free
TPPTS shows a singlet at δiso of −4.3 ppm. The absence of
any such signal for the ossified complex shows that the Pd
complex maintained its integrity after ossification. The 31P CP-
MAS NMR spectra of the other ossified complexes (catalysts
1B, 1C, and 1D) also showed formation of the Ba–sulphonate
ion pair and retention of integrity of the complexes (Fig. 2,
Part II). The appearance of the peak positions at slightly dif-
ferent positions for the different complexes were in compliance
with the stereo-electronic outplay of the respective chelating
ligands. A variation of nucleophilicity of ligands in the order
as pyca > acpy > pycald > bipy was also evident from the
31P signals of their respective ossified complexes. The FTIR
data for ossified Pd-pyca complex (catalyst 1A) showed Pd–N,
O=C–O, and C=O stretching vibrations at 568, 1327, and
1667 cm−1, respectively, which agree nicely with those of the
29.83 ppm (singlet) (Fig. 1, Part 2, curve B). However, the 31
P
NMR of the water-soluble complex Pd(pyca)(TPPTS)(OTs) in
D2O showed two signals at δ 36.13 ppm (N cis to TPPTS,
weak) and δ 35.31 ppm (N trans to TPPTS, strong) [12].
This can be attributed to the introduction of the –SO3H group

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
234
B.R. Sarkar, R.V. Chaudhari / Journal of Catalysis 242 (2006) 231–238
Part I
Fig. 2. Characterization of the ossified Pd-complexes using Ba
Part II
2+
ions. Part I: Powder XRD patterns of (A) Pd(acpy)(TPPTS) complex [catalyst 1B],
31
(B) Pd(pycald)(TPPTS) complex [catalyst 1C], (C) Pd(bipy)(TPPTS) complex [catalyst 1D]. Part II: P CP-MAS NMR spectra of the ossified catalysts using
2+
Ba ions; (A) ossified Pd(pyca)(TPPTS) complex [catalyst 1A], (B) ossified Pd(acpy)(TPPTS) complex [catalyst 1B], (C) ossified Pd(pycald)(TPPTS) complex
*
[catalyst 1C], (D) ossified Pd(bipy)(TPPTS) complex [catalyst 1D]. in the spectra indicate the side bands at 7 kHz.
pure complex (568, 1330, and 1667 cm−1, respectively). The
scanning electron microscopy (SEM) images of the ossified
complex 1A showed irregular polyhedral macrocrystallites with
an average size of 450 nm (see supplementary information,
Fig. S1). The XPS results of representative ossified catalyst
1A are as follows (see supplementary information, Fig. S2).
The peaks at 342.9 and 337.6 eV indicate the 3d3/2 and 3d5/2
states of Pd(II), with the characteristic band gap of 5.3 eV. Sim-
ilarly, peaks at 804.7 and 789.2 eV are indicative of the 3d3/2
and 3d5/2 states of Ba(II), respectively, with the characteristic
15.5 eV band gap. The other elements also matched in the bind-
ing energy values of their respective oxidation states. The other
ossified Pd-complexes (catalysts 1B, 1C, and 1D) showed sim-
ilar results in XPS. The synthesis methodology also involved
Soxhlet extraction of the catalyst materials using water and ace-
tone for 12 h each, which showed no leaching of the Pd or Ba
in to the waste solvents. These confirmed our immobilization
strategy on stability aspect and the immobilized Pd-complex
catalysts (otherwise, the derivatized Ba sulphonates) were no
less stable than our model BaSO4 molecule.
tivity and regioselectivity to the branched isomer products for
different substrates, as shown in Table 1. It was observed that
the TONs for all of the substrates were >1000 with regioselec-
tivity of >99% to the branched isomer and conversions >92%.
Compared with the previous cases, the ossified catalyst (cata-
lyst 1A) had a higher TON than the biphasic catalyst [11,12]
but was inferior to the Pd/C [13] and anchored Pd-complex cat-
alysts [14]. However, the regioselectivity was better than all of
them. The stability of the catalyst was tested through leaching
experiments. A typical reaction was stopped intermittently, and
the hot-filtered reaction mixture was analyzed for Pd-content by
atomic absorption spectroscopy (AAS). The analysis showed
almost no loss of Pd from the catalyst (8.8 × 10−5% loss of to-
tal Pd content). It is noteworthy that the ossified catalyst showed
greater stability than any of the heterogeneous Pd catalysts (Ta-
ble 1). Further, the ossified catalyst 1A was recycled four times
(Fig. 3, Part I) successfully (cumulative TON ∼5780 after 4
recycles for styrene), with only marginal variation in activity
(<1%) and practically the same regioselectivity. The recycled
catalyst 1A was characterized using 31P CP-MAS NMR (δiso
29.54 ppm), which showed no variation compared to the orig-
inal catalyst (Fig. 1, Part II, curve C). The XPS data of the
used catalyst 1A show retention of oxidation states of Pd, Ba,
Hydrocarboxylation of aryl olefins and alcohols were per-
formed using an ossified Pd-complex catalyst (catalyst 1A). The
catalyst showed excellent performance with respect to high ac-