Anchored Pd-complexes in mesoporous supports: Synthesis, characterization and catalysis studies for carbonylation reactions

Catalysis Today 198 (2012) 154–173

Contents lists available at SciVerse ScienceDirect

Catalysis Today

journal homepage: w ww.elsevier.com/locate/cattod

Anchored Pd-complexes in mesoporous supports: Synthesis, characterization

and catalysis studies for carbonylation reactions

Bibhas R. Sarkar^a,b, Raghunath V. Chaudhari^b,∗,1

^aCatalysis, Reactors and Separation Technology (CReST) Unit, Chemical Engineering Division, National Chemical Laboratory, Pune 411008, India

^bCenter for Environmentally Beneﬁcial Catalysis, Department of Chemical and Petroleum Engineering, University of Kansas, Lawrence, KS 66047, USA

a r t i c l e i n f o

a b s t r a c t

Article history:

Pd(pyca)(PPh₃)(OTs) [pyca = 2-picolinate] complex is efﬁciently anchored inside different mesoporous

matrices, such as MCM-41, MCM-48, SBA-15 using a molecular aminopropyl tether moiety employing

different synthesis strategies. Thorough characterization of the materials using powder XRD, multinu-

clear (¹³C, ²⁹Si, ³¹P) CP-MAS NMR, XPS, SEM, N₂-sorption studies etc. conﬁrmed the successful anchoring

of the palladium complex to the walls of the support matrices thus establishing the synthesis protocols

unambiguously. The catalysts were found to be highly active and selective for the carbonylation of dif-

ferent aryl oleﬁns and alcohols. Consecutive recycling and successful reuse proved the stability and true

heterogeneous nature of all the anchored catalysts, which is a substantial advancement over the existing

heterogeneous catalysts for carbonylation.

Received 10 January 2012

Received in revised form 8 April 2012

Accepted 9 April 2012

Available online 9 June 2012

Keywords:

Immobilization

Anchored catalysts

Mesoporous materials

Palladium complex

Carbonylation

1. Introduction

the very high regioselectivity (>97%) under lower operating con-

ditions (P_CO∼ 5.4 MPa). Pd(pyca)(PPh₃)(OTs) complex catalyst [5]

Carbonylation of aryl oleﬁns and alcohols is one of the exem-

plary cases of using environmentally benign “green” catalytic

routes to obtain a variety of specialty and pharmaceutical prod-

ucts replacing the classical multi-step organic synthesis routes.

The synthesis of 2-arylpropionic acids is an important application

of carbonylation, for synthesis of a useful class of non-steroidal

anti-inﬂammatory agents [1,2]. Although established by the com-

mercial success of the Hoechst-Celanese process for Ibuprofen [3],

this carbonylation method using a PdCl₂(PPh₃)₂catalyst and 10%

aqueous HCl as a promoter gave the very high regioselectivity

to the desired branched isomeric product only at high pressures

(16–35 MPa), the intrinsic activity (TOF ∼ 50–70 h⁻¹) being quite

poor. The regioselectivity is pressure dependent and reduced to

∼60% under lower operating pressures of ∼6–7 MPa [4]. Subse-

quent developments of carbonylation catalysts were thus focused

toward achieving higher activity and regioselectivity operating

under low-pressure conditions. It was shown that the use of pro-

moters like p-toluenesulfonic acid, TsOH and lithium chloride,

LiCl [4] facilitated a quantum jump in the carbonylation activity

(TOF ∼ 2250 h⁻¹, for styrene) of Pd complex catalysts maintaining

improved the activity further (TOF ∼ 2600 h⁻¹) and this is the

best known homogeneous catalyst for such carbonylation reac-

tions. Recent studies on kinetics of hydroxycarbonylation of styrene

using homogeneous Pd(pyca)(PPh₃)(OTs) complex [6] has opened

up new directions in exploring the use of this complex in het-

erogenized form to facilitate efﬁcient catalyst–product separation

to minimize costs associated with energy intensive downstream

processing. “Efﬁcient” and “Economic” catalyst–product separa-

tion and reuse being the key issues for “green” technology, the

research focus was mainly centered on the design of heterogeneous

catalysts. We have previously reported novel anchored palladium

complex catalysts inside mesoporous MCM-41 and MCM-48 mate-

rials for the carbonylation reactions, which exhibited good activity,

selectivity and excellent recyclability of the solid catalysts [7].

Recently, we have reported another method for immobilization

of metal complexes – ‘ossiﬁcation’, and applications of these cat-

alysts for carbonylation showed lower activity compared to the

anchored catalysts [8]. In this paper, we present a detailed study on

the design of heterogeneous carbonylation catalysts by anchoring

palladium complexes inside MCM-41, MCM-48 and SBA-15 mate-

rials along with their characterization followed by their catalytic

applications. A comparative analysis is also presented to under-

stand the correlation between parameters affecting the activity,

regioselectivity, stability and recyclability of the immobilized cat-

alysts.

∗

Corresponding author. Tel.: +1 785 864 1634; fax: +1 785 864 6051.

E-mail address: rvc1948@ku.edu (R.V. Chaudhari).

1

A part of the work was performed at NCL, India.

http://dx.doi.org/10.1016/j.cattod.2012.04.047

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

155

2. Experimental

batch synthesis, 8 g of the nonionic block copolymer Pluronic P123

(Aldrich) was taken in a polypropylene beaker and 60 mL distilled

deionized water was added to it. The mixture was stirred thor-

oughly using an overhead stirrer (with Teﬂon blades) for 4 h to

dissolve the gel completely. To this solution, 240 mL of 2N HCl was

added drop wise and allowed to stir further for 1 h at ambient tem-

perature (∼301–303 K). The temperature of the mixture was then

raised to 313 K and stirred for 1 h again. To this, 18 g tetraethy-

lorthosilicate (TEOS) was added drop wise with vigorous stirring.

After the completion of addition, the mixture was allowed to stir for

24 h, maintaining the temperature at 313 K. The synthesis gel thus

formed was then loaded in a 500 mL polypropylene bottle, sealed

using Teﬂon tape and kept at 373 K for 48 h under static conditions.

Afterwards, the bottle was allowed to cool, the contents were ﬁl-

tered and residue was washed with water. The white solid was

allowed to dry in air. Calcination of the as-synthesized solid mate-

rial was done at 773 K for 6 h in air, using a slow heating ramp of

1^◦min⁻¹. This was designated as S15 for all future use.

2.1. Syntheses

2.1.1. Synthesis of Pd(pyca)(PPh₃)(OTs) complex

The Pd(pyca)(PPh₃)(OTs) complex was prepared by the lit-

erature procedure [5]. In

a typical synthesis 0.5 g Pd(OAc)₂

(2.225 mmol), 0.274 g pyridine-2-carboxylic acid (pycaH)

(2.225 mmol), 0.846 g p-toluenesulfonic acid (4.45 mmol) and

1.168 g triphenylphosphine (4.45 mmol) were taken in minimum

volume (∼6 mL) of chloroform in a round-bottomed ﬂask and

the mixture vigorously shaken at room temperature for a few

minutes until all the components were completely dissolved and

the solution turned bright yellow from the brownish yellow color

of the Pd(OAc)₂solution. The product was isolated as yellow

oily mass by addition of n-hexane to the chloroform solution

and vigorous shaking for 10 min. The oily material was washed

with n-hexane and then with diethyl ether a number of times.

Applying vacuum led to the formation of a yellow ﬂuffy solid. For

elemental and spectroscopic analysis, the complex was puriﬁed

by re-precipitation from chloroform four times. The synthesized

complex was preserved under argon and used hereafter as such.

Calculated C 55.18, H 4.23, N 1.98, P 4.56, S 4.71; analyzed C 55.08,

H 4.23, N 1.91, P 4.08, S 4.68

2.1.5. Functionalization of mesoporous matrices

The mesoporous matrices were functionalized following dif-

ferent strategies. For MCM-41 and MCM-48, only post-synthesis

grafting method was applied but for SBA-15, both post-synthesis

grafting and one-pot co-condensation methods were applied.

2.1.2. Synthesis of MCM-41

MCM-41 was synthesized following the procedure reported

by Mukhopadhyay et al. [9] with slight modiﬁcations to obtain

reproducible results. No promoters were used in the synthesis of

MCM-41. The stoichiometric composition of the synthesis gel was:

1 SiO₂/0.32 NaOH/0.2 CTABr/125 H₂O. For a typical synthesis batch,

3 g of fumed silica was added to a solution of 0.64 g NaOH in 50 mL

H₂O, and stirred for 1 h. To this mixture, a solution of 3.64 g of CTABr

in 50 mL H₂O was added drop wise under vigorous stirring. After

this, 12 mL H₂O was also added drop wise, and the mixture was

stirred vigorously for another 90 min, after which the gel was auto-

claved at 373 K for 16 h. After cooling to ambient temperature, the

white precipitate was ﬁltered, washed with water and was air-

dried. The white solid was then calcined at 813 K for 6 h in air

using a slow heating rate of 1^◦min⁻¹. After cooling, the free ﬂowing

powder was preserved, sealed under argon atmosphere. This was

designated as M41 for further use.

2.1.5.1. Post-synthesis grafting. Functionalization of MCM-41 and

MCM-48 were done using the method described by Shephard

et al. [11]. In this method, consecutive application of an inert

graft and an active graft molecule is used to obtain a material

with a passivated outer surface and the active-functionalized inner

one. In a typical preparation method, 1 g of calcined MCM-41/48

was taken in a round-bottomed ﬂask and stirred well with 30 mL

dichloromethane (DCM) under argon atmosphere to get uniform

slurry. To it, 0.03 mL of dichlorodiphenylsilane was added and

stirred for 1 h. The reaction mixture was cooled to 195 K and 1 mL

of 3-aminopropyl trimethoxysilane (APTS) was added to the slurry.

The mixture was stirred for 24 h at 313 K, ﬁltered. The residue was

washed with DCM and dried in vacuo. This resulted in MCM-41/48

with passivated outer silanol groups by diphenylsilyl groups while

the inner silanol groups were functionalized to 3-aminopropyl end-

groups. The functionalized MCM-41 and MCM-48 samples were

designated as M41-NH₂and M48-NH₂respectively.

2.1.3. Synthesis of MCM-48

Post-synthesis functionalization of SBA-15 was done following

a method based on that described by Asefa et al. [12]. This method

employed the strategic passivation of the external surface of the

as-synthesized material using a passive graft before the removal

of the template. Washing out the template revealed the inner sur-

face, which was then functionalized using an active graft molecule.

In a typical batch of preparation, 5 g of the as-synthesized SBA-

15 was treated with 5 mL hexamethyldisilazane (HMDS) in 60 mL

toluene under inert atmosphere. Filtering and washing the residue

yielded a white powder. Removal of the template was done by

stirring the passivated SBA-15 with 1:1 diethyl ether and ethanol

(100 mL per 0.5 g of solid) at ambient temperature for 6 h. The solid

was ﬁltered, washed and dried. To introduce the dangling amine

function onto the inner surface, 1 g of the dried solid suspended

in 60 mL DCM was cooled to 250 K and to it 1 mL of APTS was

added very slowly with vigorous stirring. After the completion of

addition, the reaction mixture was brought to ambient tempera-

ture and then heated to 313 K and stirred for 24 h. The solid was

ﬁltered out and washed thoroughly using more DCM, and then

dried. Thus, we obtained SBA-15 having external silanol groups

passivated with trimethylsilyl groups and inner ones functional-

ized with 3-aminopropyl end-groups. The material was designated

as S15-NH₂-(G).

MCM-48 was prepared following the literature procedure

reported by Mukhopadhyay et al. [9] using a promoter-mediated

synthesis method. The initial stoichiometric composition of gel

was: 1 SiO₂/0.4 NaOH/0.21 CTAB/120 H₂O/0.0033 Promoter. For a

typical synthesis batch, 3 g of fumed silica was added to a solu-

tion of 0.8 g of NaOH in 50 mL H₂O and stirred for 1 h. To this

mixture, drop wise addition of a solution of 3.82 g CTABr in 50 mL

H₂O was followed by the addition of a solution of 0.472 g phospho-

tungstic acid in 8 mL H₂O with vigorous stirring. The mixture was

then stirred for 90 min and then the gel was autoclaved at 423 K

for 12 h. After cooling, the contents of the autoclave were ﬁltered,

washed thoroughly with generous amount of water to remove any

residual water-soluble promoter. The white residue was air-dried

and then calcined at 813 K for 8 h in air following a very slow heat-

ing ramp of 1^◦min⁻¹. After cooling, the free-ﬂowing white powder

was preserved under argon atmosphere for further use. This was

designated as M48 for all further use.

2.1.4. Synthesis of SBA-15

SBA-15 material was synthesized following the procedure as

reported by Zhao et al. [10], with slight modiﬁcations in the

methodology as required to obtain reproducible results. In a typical

156

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

2.1.5.2. One-pot co-condensation. For the functionalization of SBA-

15 using the one-pot co-condensation technique, the typical

synthesis of SBA-15 as described previously (Section 2.1.4) was

modiﬁed following the strategies described by Wang et al. [13].

The synthesis was downscaled to half for this purpose. The molar

recipe of the synthesis gel was (1 − x)TEOS:xAPTS:6.1 HCl:0.017

P123:165 H₂O (where x = 0.05–0.2, typically 0.1). In a typical syn-

thesis batch, 4 g of Pluronic P123 was dissolved in 30 mL H₂O for 4 h,

to this 120 mL of 2N HCl was added and stirred for 1 h at ambient

temperature. The temperature was raised to 313 K and the mix-

ture was stirred at 313 K for another 2 h and then 9(1 − x) g TEOS

was added drop wise. The mixture was allowed to prehydrolyze

for 0–3 h (typically 1 h) before the addition of requisite amount of

APTS, after which it was allowed to stir for a total of 24 h (including

prehydrolysis) at 313 K. The mixture was transferred to a 250 mL

polypropylene vessel and heated at 373 K for 48 h under static con-

ditions. The contents were cooled, ﬁltered, and washed to water

and dried. The white solid was then reﬂuxed in EtOH for 24 h

(200 mL EtOH per 1.5 g of solid), ﬁltered and dried to get the amine-

functionalized SBA-15. This functionalized SBA-15 was designated

as S15-NH₂-(C).

2.1.7.2. Synthesis of NaY zeolite supported Pd(pyca)(PPh₃)(OTs) cat-

alyst. To slurry of 3 g NaY zeolite in 30 mL ethanol, an ethanolic

solution of Pd(pyca)(PPh₃)(OTs) complex was added and reﬂuxed

for 18 h. After bringing to the ambient temperature, the contents

were ﬁltered, washed and dried at 353 K and was stored under

argon as a free-ﬂowing pale yellow solid. This was designated as

Pd-Y.

2.2. Carbonylation reactions

Carbonylation²reactions were carried out in a 50 mL Parr

autoclave reactor made of Hastelloy C-276 material following a

procedure similar to that described previously [8]. For a typical

reaction, the solid catalyst and the liquid phase reactants/solvent

were charged in the reactor and sealed. The contents were heated to

the desired temperature, then CO gas was introduced to the reac-

tor to the desired pressure and the reaction started by switching

on the stirring. The consumed CO was made up from a reservoir

through a constant pressure regulator. The progress of the reac-

tion was followed by observing the pressure drop in a CO reservoir

as a function of time and the reaction was stopped at completion

of CO absorption. Afterwards, the reactor was brought to ambient

temperature, its content-gas was released and ﬂushed thrice with

N₂and liquid phase samples were analyzed for reactant/products

composition using HP GC 6890 ﬁtted with a HP-FFAP capillary

column (25 m × 0.33 mm × 0.2 ␮m). GC–MS analyses of the sam-

ples were performed using Agilent 5973 N Mass Selective Detector

attachment.

2.1.6. Immobilization of Pd-complex in the functionalized

mesoporous matrices MCM-41 and MCM-48 (using M41-NH₂and

M48-NH₂)

To immobilize the Pd-complexes inside the mesoporous chan-

nels of the functionalized MCM-41 and MCM-48, these materials

were treated with the solutions of the metal complexes under inert

atmosphere. For a typical synthesis batch, 1 g of the functionalized

support matrix, namely M41-NH₂and M48-NH₂, was suspended

in 30 mL dichloromethane (DCM). To it, a 50 mL solution contain-

ing 0.33 g of Pd(pyca)(PPh₃)(OTs) complex in dichloromethane was

added, and was stirred for 24 h at room temperature. The pale

yellow solid obtained after ﬁltration, washing and subsequent dry-

ing in air was given a Soxhlet extraction in DCM for 10 h and the

paler yellow solid was preserved under argon for further use. The

catalysts were designated as M41-NH₂-Pd and M48-NH₂-Pd respec-

tively following the respective supports used.

2.3. Recycle studies and leaching experiments

For the recycle studies using the anchored catalysts, the follow-

ing procedure was followed. After the carbonylation reaction run

was complete, the reactor was allowed to rest, to settle the solid

catalyst and the supernatant organic phase was decanted slowly to

separate it from the solid catalyst. The solid residue was washed

with the reaction solvent i.e. methyl ethyl ketone for a couple of

times in the same manner. A fresh charge of reactants and promot-

ers was introduced and the reaction was carried on in the same

fashion as described previously, after withdrawing an initial sam-

ple before starting the recycle run. This was repeated for all the

recycle reactions. The reaction mixture thus obtained was retained

for further analyses.

Metal leaching experiments with the anchored Pd-complex cat-

alysts were performed by ﬁltration of the hot reaction mixture at

388 K by stopping the reaction abruptly at ∼30% conversion and

subsequently, testing the catalytic activity of the ﬁltrate for hydro-

carboxylation without the addition of any fresh heterogeneous

catalyst. The ﬁltrate and the catalyst thus recovered were ana-

lyzed for Pd-content by ICP–AES (Inductively Coupled Plasma for

Atomic Emission Spectroscopy) analyses to contemplate the pal-

ladium content in solution and loss of palladium from the parent

catalyst during the reaction.

2.1.6.1. SBA-15 (using S15-NH₂-(G) and S15-NH₂-(C)). To a sus-

pension of 1.5 g of the functionalized SBA-15 matrix in 100 mL

dichloromethane, a solution of 100 mg of Pd(pyca)(PPh₃)(OTs)

complex in 25 mL in DCM was added and stirred for 24 h at room

temperature under argon atmosphere. The contents were ﬁltered,

washed with DCM and dried in air. The dried pale yellowish solid

was given Soxhlet extraction with dichloromethane for 8 h, to

obtain a faint yellow colored free-ﬂowing solid. This process was

used for both the amine – functionalized SBA-15 matrices, and the

corresponding immobilized catalysts were designated as S15-NH₂-

(G)-Pd and S15-NH₂-(C)-Pd respectively.

2.1.7. Synthesis of supported catalysts

Two different supported catalysts were prepared for the car-

bonylation reactions. An aluminosilicate microporous material

having a well-ordered deﬁnite structure was used i.e. NaY zeo-

lite, and pure silica (catalyst grade from Davisil^®) were used for

supporting the Pd-complex by simple adsorption on the surface.

2.4. Analytical and characterization methods

For characterization of the synthesized materials at different

stages of synthesis and after the reactions etc., numerous tech-

niques have been employed to identify, understand and conﬁrm

our postulates. Powder XRD analysis of the mesoporous materials

at room temperature were done using two instruments for dif-

ferent 2ꢀ ranges. Philips XL machine was used for the low-angle

2.1.7.1. Synthesis of SiO₂supported Pd(pyca)(PPh₃)(OTs) catalyst. 3 g

of Davisil^®silica, 100–200 mesh, was suspended in ethanol, and

stirred for 15 min to obtain a uniform slurry, to it an ethanolic solu-

tion of 0.135 mol of Pd(pyca)(PPh₃)(OTs) complex was added and

the mixture was reﬂuxed for 18 h. The solid was ﬁltered off, washed

and dried at 353 K. The faint yellow solid was stored under argon

and was designated as Pd-S.

2

Hazardous CO involved, authors suggest safe and expert handling during exper-

iments.

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

157

scans (2ꢀ range of 0.5–5.0^◦) required for SBA-15 samples, while

MCM-41, MCM-48 were analyzed using a Rigaku Miniﬂex instru-

ment for 2ꢀ range of 1.5–10^◦at a slow scan rate of 1^◦/min using

Cu–K_˛(ꢁ = 0.15404 nm) radiation. FT-IR spectra were obtained

using a Perkin Elmer Spectrum-2000 in transmission mode using

KBr pellets as well as in Shimadzu Hyper IR in DRS (diffused

reﬂectance spectroscopy) mode by mixing samples with KBr. NMR

was obtained from a Bruker MSL-300 and Bruker DRX-500 spec-

trometers. The ³¹P NMR spectra were recorded at 202.456 MHz by

using 85% H₃PO₄as an external standard and data were collected at

spectral width of 20 kHz with a ﬂip angle of 45^◦, with ∼6000 actual

data points and 5 s relaxation delay. Solid-state analyses were done

using a 3 mm CP-MAS probe. X-ray photoelectron spectra (XPS)

were recorded on a VG – Microtech ESCA 3000 spectrometer using

the Mg-K_˛emission (E = 1253 eV) under a vacuum of ∼10⁻⁹Torr.

Scanning electron microscopy (SEM) were performed using a

Philips XL 30 instrument – the catalyst materials were suspended in

isopropanol, cast on gold plated discs followed by drying under vac-

uum and coating with a conducting material and then were imaged.

For the catalytic carbonylation reactions, the reaction mixtures

(liquids) were analyzed by gas chromatography, using a Hewlett-

Packard 6890 Series GC instrument, controlled by HP ChemStation

software, and HP-FFAP capillary column (25 m × 0.33 mm × 0.2 ␮m

ﬁlm-thickness, on a polyethylene glycol stationary phase). Cali-

bration of the method was performed using standard materials in

the concentration range similar to that of the catalytic reactions

for each component of substrate(s) and products in the reaction

mixture.

the microporous materials, these mesoporous materials might lose

the organometallic complex if they are just entrapped physically.

Therefore, the complexes need to be chemically ﬁxed to the channel-

walls to prevent leaching. The large pore dimensions (unit cell

˚

parameter, a₀∼ 35–300 A) would allow easy access of the sub-

strate(s) and reagent(s) to the catalytic sites and hence a good

catalytic activity, along with the separation advantage. In case of

MCM-41 and MCM-48, the selective passivation of the outer sur-

˚

face might work due to the smaller size of the channels (∼35 A),

which enables kinetically rapid silylation (passivation) of the outer

surface only. However, for SBA-15 materials, two different tech-

niques of immobilization of the metal complex were used, namely

the post-synthesis grafting (different from that of M41s) and co-

condensation technique. This was primarily because of the larger

˚

pore size of the SBA-15 material (∼80–100 A), where the access of

the soluble graft molecule to the internal as well as the external

surface is almost equal. Thus, if a selective passivation of the exter-

nal surface is attempted, it may so happen that the graft molecule is

distributed over the inner surface, in addition to the outer surface

also – resulting in a randomly passivated (rather randomly func-

tionalized) support matrix. To immobilize Pd-complex catalysts,

3-aminopropyltrimethoxysilane (APTS) was used as the molecu-

lar graft to hold the metal complex, while dichlorodiphenylsilane

and hexamethyldisilazane (HMDS) were used as the passive graft

to block the surface silanol groups on the external mesopore wall. A

schematic representation of the anchored palladium complex cata-

lysts for catalytic carbonylation is given in Scheme 1. To describe the

general morphology of these hybrid composites a hexagonal motif

of the support matrix is chosen. The selectively passivated external

surface ensures that the catalytically active palladium complex is

anchored only to the internal wall of the mesoporous channels, thus

minimizing the chances of the degradation and leaching of the pal-

ladium complex from the catalyst-composite under the stringent

reaction conditions as required for the carbonylation reactions.

3. Results and discussions

3.1. Rationale and strategy of immobilization

In order to select the supports from a wide range of available

materials, a few criteria were employed, such as the matrix should

3.2. Powder XRD analyses

(i) be chemically non-invasive to the organometallic chemical

environment of the anchored complex.

(ii) accommodate the complex comfortably without distorting the

conformational requirements for catalysis.

(iii) be easy to synthesize and subsequently easy to functionalize

using choice of functional materials for tailor-made purposes.

(iv) have good stability to sustain the functionalization and appli-

cation conditions.

(v) have distinguishable deterministic properties different from

those of the complex, which could be used for characterization

of the ensemble.

Fig. 1 represents a comparison of the powder XRD patterns at dif-

ferent stages of synthesis of the immobilized Pd(pyca)(PPh₃)(OTs)

complex inside the mesoporous M-41 (Part A) and M-48 (Part B),

respectively. For both the parts (A and B), the powder XRD patterns

of the sequential passivation of the exterior surface ((b)-curves),

amine-functionalization of the already passivated mesoporous

matrices ((c)-curves), and Pd-complex anchored to the internal sur-

faces of the mesoporous matrices ((d)-curves) have been presented.

It is clearly indicated that the characteristic diffraction peaks of the

mesoporous matrices are present in all the samples (strong 100

and feeble 110, 200, 210 peaks at 2ꢀ values of 2.08^◦, 3.67^◦, 4.27^◦,

5.57^◦respectively for MCM-41, and sharp 211, 220 and weak 420,

322, peaks at 2ꢀ values of 2.37^◦, 2.67^◦, 4.37^◦, 4.59^◦respectively

for MCM-48). Interestingly, both the mesoporous matrices, MCM-

41 and MCM-48 retain their respective characteristic properties

during the process of anchoring of the Pd-complex to the interior

surface of these matrices. Practically, no changes in the peak posi-

tions were observed during the step of passivation of the exterior

silanol groups ((b)-curves) for both the M-41 and M-48 matrices. A

slight shifting of the principle peaks (100 for M41-NH₂-Pd, and 211

for M48-NH₂-Pd) to the higher 2ꢀ values was observed indicating

the incorporation of the Pd-complex inside the mesoporous cavity

leading to the shrinkage of the unit cell dimension parameters (a₀).

A detailed analysis of the physical characteristics of the different

MCMs at different stages of immobilization of the Pd-complexes is

provided in Table 1. It is necessary to mention here that the prefer-

ential passivation of the exterior silanol groups in both the MCM-41

and MCM-48 occurred by the relatively faster silylation kinet-

ics to the most accessible external silanols only. However, it was

A siliceous support matrix would be the most effective for the

previously mentioned purpose, as it would not interfere with the

palladium complex during synthesis as well as the carbonylation

reactions. Microporous material such as ZSM-5 was not suitable

due to the smaller size of the 3-dimensional cavity in its frame-

work to accommodate the relatively larger sized Pd-complexes,

which might actually ﬁt inside the cavity but with distortion, thus

tampering the organometallic atmosphere conducive to catalysis.

Mesoporous materials have the characteristic pore dimensions of

˚

20–300 A, with very high surface areas and hydrothermal stability

and hence could be better option. These materials (e.g. MCM-41,

MCM-48, etc.) unlike simple silica are high surface area porous

substances having long-range periodicity and tunable pore struc-

tures. In addition, the SBA class of materials has emerged lately

as good candidates for support matrices. These SBAs are meso-

porous materials with high porosity and surface areas, having larger

pore-wall thickness than the MCM materials. Being larger than

158

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

Scheme 1. Hydrocarboxylation of alkenes or alcohols to acids with anchored Pd-complex catalysts.

observed elsewhere [14] that when the treatment with Ph₂SiCl₂

was omitted, random uncontrollable amine-functionalization and

subsequent anchoring of the Pd-complex was obtained, which con-

sequently may weaken the orderedness of the matrices.

Sequential powder XRD experiments were carried out for dif-

ferent SBA-15 materials during the course of the immobilization

of the Pd-complex and the results are presented in Fig. 2. It was

clear from the plot that the SBA-15 did not change during the

process of anchoring of the Pd-complex. The principle peaks at

1 0 0, 1 1 0 and 2 0 0 were clearly visible for all of the SBA-15 sam-

ples. The intensities of these peaks were almost equal. However,

for the NH₂-functionalized SBA-15, prepared by grafting method

(S15-NH₂-G), and the subsequent Pd-complex anchored SBA-15

(S15-NH₂-G-Pd) showed shift of the (1 0 0) peak toward higher

2ꢀ value. This decrease in the d-spacing might be due to the

incorporation of the NH₂group and the Pd-complex inside the

mesopores. For the in situ functionalized SBA-15 (S15-NH₂-C), the

(1 0 0) peak appeared at slightly higher 2ꢀ than its grafted counter-

part (S15-NH₂-G). This might be due to the initially larger number

of anchoring agent molecules (APTS) entering the building units

4−

before the condensation of the SiO₄

monomers in the synthe-

sis gel, during the in situ synthesis-cum-functionalization step. The

resulting smaller d-values are self-implicative of the space required

by the larger number of the anchoring agent moieties, leaving out

only a smaller void space. The d-spacing further decreased (d_{1 0 0}

9.24 nm) when the Pd-complex was incorporated into the channels.

Fig. 1. Comparative powder XRD patterns of M-41 (Part A) and M-48 (Part B) (a) curves = pristine mesoporous matrices [M-41 and M-48]; (b) curves = exterior surface-

passivated mesoporous matrices using Ph₂SiCl₂; (c) curves = NH₂functionalized mesoporous matrices [M41-NH₂and M48-NH₂]; (d) curves = Pd(pyca)(PPh₃)(OTs) complex

anchored inside mesoporous matrices [M41-NH₂-Pd and M48-NH₂-Pd] respectively.

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

159

Table 1

Comparative physical characteristics of the different surface-modiﬁed MCM-41, MCM-48 and SBA-15 matrices.

Sample designation

d_{h k l}^a, nm

a₀^b, nm

Pore volume, cm³/g

Pore diameter, nm

Wall thickness^c, nm

Surface area, m²/g

M41

4.232 (1 0 0)

4.122 (1 0 0)

4.164 (1 0 0)

3.711 (2 1 1)

3.708 (2 1 1)

3.681 (2 1 1)

10.100 (1 0 0)

10.084 (1 0 0)

9.346 (1 0 0)

10.090 (1 0 0)

9.242 (1 0 0)

–

4.89

4.76

4.81

9.09

9.08

8.96

11.65

11.64

10.78

11.65

10.67

–

1.203

1.101

1.057

1.399

1.243

1.176

1.189

1.134

0.976

1.087

1.002

–

3.668

3.022

2.896

3.040

2.760

2.530

7.746

7.224

6.251

7.246

6.151

–

1.222

1.738

1.914

1.420

1.560

1.632

3.904

4.416

4.529

4.404

4.519

–

1322

1152

1035

1632

1510

1401

886

798

734

774

721

M41-NH₂

M41-NH₂-Pd

M48

M48-NH₂

M48-NH₂-Pd

S15

S15-NH₂-(G)

S15-NH₂-(G)-Pd

S15-NH₂-(C)

S15-NH₂-(C)-Pd

Pd-S

189

Pd-Y

–

785

a

Calculated from powder XRD patterns, as nꢁ=2d sin ꢀ, where, n = 1, ꢁ = 0.15404 nm; values in parentheses indicate the Miller indices.

√

6 (for MCM-48).

b

c

Calculated as a₀= 2d_{1 0 0}

/

3 (for MCM-41 and SBA-15), a₀= d_{2 1 1}

×

Calculated as a₀-PD (for MCM-41 and SBA-15), a₀/3.092 – PD/2 (for MCM-48), where PD pore diameter.

Thus, the powder XRD illustrates successfully, the incorporation of

the graft moiety by both the methods, and the anchoring of the

Pd(pyca)(PPh₃)(OTs) complex to the mesoporous matrices.

other except for a very narrow dissimilar region indicating a

very narrow pore size distribution, and a slight capillary con-

densation. For MCM-41 and MCM-48, the adsorption isotherms

were classiﬁed as Type IV of the IUPAC conventions [15] – char-

acteristic of the M41S type materials; [16] and the inﬂection

point near P/P₀∼ 0.3–0.4 indicated the sharp capillary conden-

sation region, showing a very narrow pore size distribution. A

very small hysterisis loop was observed for both the materi-

als around P/P₀∼ 0.25–0.4. However, for MCM-48 the pore-size

distribution curve showed the maximum at a lower pore diam-

eter (between 2 and 3 nm) than that for MCM-41 (between 3

and 4 nm) indicating smaller pores. The N₂adsorption–desorption

isotherm of calcined SBA-15 material is shown in Fig. 5. The

adsorption arm of the plot clearly showed three distinguished

regions corresponding to (a) monolayer–multilayer adsorption,

(b) capillary condensation inside the porous cavities and walls,

and (c) multilayer adsorption on the external surface of the

material. A clear but small hysterisis loop of Type I is also

observed in the adsorption–desorption isotherm, showing the

occurrence of capillary condensation at higher relative pressures

(P/P₀∼ 0.65–0.8). The surface area obtained from single point

calculations at P/P₀= 0.303 using the Brunauer–Emmett–Teller

(BET) equation for adsorption isotherm was 871.25 m²/g. However,

multi-point calculations and averaging at lower relative pressures

(P/P₀< 0.31, hence, considering monolayer formation only) gave

us another surface area based on unimolecular layer formation

during adsorption, as 885.94 m²/g. Further calculations using the

BJH method gave us the pore dimensions from the same N₂-

adsorption isotherm. Using the adsorption branch of the isotherm,

the pore diameter was determined as 7.746 nm, having an aver-

age pore volume of 1.1892 cm³/g. The pore-size distribution curve

showed a very narrow distribution pattern having a maximum

at ∼7.75 nm corresponding to the average pore size. Consider-

ing the d_{1 0 0}spacing from the powder XRD data and the unit

cell volume, a₀, the pore wall thickness was found to be nearly

3.9 nm, which is almost the same as the pore size of the MCM-

41 material. Thus, the SBA-15 was proved to be having very high

surface area with larger pore size and thicker mesoporous frame-

walls than MCM-41, which might be of better mechanical strength.

A comparative presentation of the different surface parameters

of all three mesoporous matrices as well as their functionalized

and metal-complex loaded counterparts is given in Table 1. From

the data, it is observed that for all of the functionalized matrices

and the Pd-complex loaded mesoporous materials, the long-range

order is maintained along with the consistent variation of certain

parameters. As observed from the powder XRD patterns already,

the shifting of the position of the principle peak toward the

3.3. Adsorption–desorption studies

The nitrogen adsorption–desorption isotherms along with the

pore size distribution of calcined MCM-41 and MCM-48 are pre-

sented in the Figs. 3 and 4 respectively. Both the isotherms show

one sharp inﬂection point. For MCM-41 and MCM-48, the adsorp-

tion and desorption regions of the isotherms almost overlap each

F

E

D

C

B

A

1

2

3

4

5

2θ, degrees

Fig. 2. Comparative powder XRD data of different SBA-15 materials. (A)

As-synthesized SBA-15 (S15); (B) outer surface passivated SBA-15; (C) NH₂-

functionalized on the internal surface of SBA-15 (S15-NH₂-G); (D) Pd-complex

anchored inside SBA-15 (S15-NH₂-G-Pd); (E) in situ NH₂-functionalized SBA-15

(S15-NH₂-C), (F) Pd-complex anchored to in situ NH₂-functionalized SBA-15 (S15-

NH₂-C-Pd).

160

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

v = Pore Volume (cm³g^-1)

d = Pore Diameter (Å)

Adsorption

Desorption

800

700

600

500

400

300

200

100

0.10

0.08

0.06

0.04

0.02

0.00

0.0

0.2

0.4

0.6

0.8

1.0

20

40

60

80

100

P/Po

Pore diameter (Å)

Fig. 3. N₂adsorption–desorption isotherm and BJH pore size distribution of calcined MCM-41.

higher 2ꢀ values during the NH₂-functionalization and subsequent

loading of the palladium complex is being reinforced by the surface

analyses.

SEM micrographs of the different samples of catalysts made using

SBA-15 as the mesoporous support are presented in Fig. 6. The char-

acteristic entwined worm-like morphology is clearly evident from

the SEM images of the as synthesized as well as the calcined SBA-

15 samples. This shows that the external morphology of SBA-15

did not change upon calcination at 813 K, which removed the sur-

factant micelles from the internal channels of SBA-15. For all the

samples of SBA-15 matrix at different stages of immobilization of

the palladium complex inside the mesoporous channels, the simi-

lar worm-like morphology was observed. This stable morphological

pattern as maintained throughout the immobilization process def-

initely refers to the stability and robust nature of the solid catalyst

materials. These infer about the mechanical stability of the SBA-15

support matrix and also for the synthesized catalysts.

3.4. Scanning electron microscopy analysis

The SEM micrographs of the different catalysts based on MCM-

41 and MCM-48 show regular spherical morphology was observed

for the pristine support matrices (i.e. M41 and M48) and also for the

mesoporous matrices with Pd(pyca)(PPh₃)(OTs) complex anchored

inside (i.e. M41-NH₂-Pd and M48-NH₂-Pd). The anchored catalysts

did not appear much different from the pristine support matrices

in terms of the general morphology. The average sizes of the par-

ticles were around 500 nm in diameter, for all the samples. The

0.14

v = Pore Volume (cm³g^-1)

d = Pore Diameter (Å)

900

800

700

600

500

400

300

200

100

Adsorption

Desorption

0.12

0.10

0.08

0.06

0.04

0.02

0.00

0.0

0.2

0.4

0.6

0.8

1.0

20

40

60

80

100

P/Po

Pore diameter (Å)

Fig. 4. N₂adsorption–desorption isotherm and BJH pore size distribution of calcined MCM-48.

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

161

900

800

700

600

500

400

300

200

0.030

0.025

0.020

0.015

0.010

0.005

0.000

Adsorption

Desorption

0.0

0.2

0.4

0.6

0.8

1.0

50

100

150

200

250

300

Relative pressure, P/P

Pore diameter, A

0

Fig. 5. N₂adsorption–desorption isotherm and BJH pore size distribution of calcined SBA-15.

3.5. Transmission electron microscopy analysis

concluded that the Pd complex retained its original oxidation

state of Pd(II) during the synthesis of the anchored catalysis each

by using a different methodology and reagents as required by

the different support matrix. Thus, the catalysts were stable and

robust throughout the synthesis process. The data of the catalysts

after carbonylation of styrene is also presented here. No signiﬁ-

cant change in the oxidation states of the elements are observed.

The stability of the synthesized catalysts is hinted from this

characterization.

The mesoporous materials were characterized using TEM at dif-

ferent stages of the synthesis of the anchored Pd-complex catalysts.

The comparative data are presented in Fig. 7. The pristine MCM-41,

MCM-48 and SBA-15 are shown as images A D and G respectively.

Distinct hexagonal and cubic mesophases were clearly visible for

MCM-41, SBA-15 and MCM-48 respectively, expecially with very

clear pore-channels for SBA-15. After passivation of the exterior

surface (with Ph₂SiCl₂for MCM-41 and MCM-48 and HMDS for

SBA-15) and functionalization of the interior walls of the matrices

with an aminopropyl group, the materials namely, M41-NH₂, M48-

NH₂and S15-NH₂-(G) were also imaged using TEM (images B, E and

H in Fig. 7). Not much difference of the patterns was observed in the

images. However, after the anchoring of the Pd(pyca)(PPh₃)(OTs)

complex inside the mesoporous channels (M41-NH₂-Pd, M48-NH₂-

Pd and S15-NH₂-(G)-Pd), the images (C, F and I in Fig. 7 respectively)

showed distinct deep contrasting furrows with respect to the light

shaded surface. This might be interpreted as due to the presence

of the metal complexes inside the matrix, but not on the surface. If

the Pd-complex was anchored on the surface of the functionalized

matrices, then the high-contrast dark furrows would have appeared

along the boundary of the visualized matrices and not inside the

porous body as observed previously by Shephard et al. [11]. Thus,

the immobilization of the Pd-complexes inside the pore-channels,

by anchoring to the interior walls of these porous channels may be

supported by TEM.

3.7. FT-IR studies

FT-IR analysis of all the materials showed two distinctive sig-

nals at 1329 cm⁻¹(ꢂ_{O O}) and 1669 cm⁻¹(ꢂ_{C O}) as similar to the

C

signals at 1330 cm⁻¹and 1668 cm⁻¹for the Pd(pyca)(PPh₃)(OTs)

complex [5]. The Pd N stretching frequency signal at 564 cm⁻¹

(ꢂ_{N Pd}) was observed for all the anchored catalysts. This was dif-

ferent from that of the complex at 568 cm⁻¹showing a possible

coordination of N_APTSto Pd center. These are in agreement to our

previously observed results [7].

3.8. Solid state CP-MAS NMR spectroscopy

Multi-nuclear NMR studies were used for the characterization

of the anchored Pd-complex catalysts. To understand the changes

occurring under the synthesis as well as during the course of

reaction conditions, ¹³C, ²⁹Si and ³¹P solid-state CP MAS NMR

experiments were performed at every step of synthesis of the SBA-

15 based anchored catalysts (for both post-synthesis grafting and

co-condensation methods) and also after the carbonylation reac-

tions as a test the stability of the catalysts.

3.6. X-ray photoelectron spectroscopy

The oxidation state of the elements present in the catalyst is

typically analyzed using the X-ray photoelectron spectroscopy. The

elements of major concern are Pd, N, C, P etc., which might prove

decisive in the determination of the actual stereo-electronic condi-

tions arising out of the anchoring process. The data of the different

other catalysts namely, M48-NH₂-Pd, S15-NH₂-(C)-Pd, Pd-S and

Pd-Y are presented in Table 2. As observed form the XPS data of

the M41-NH₂-Pd sample, the binding energies of palladium was

337.3 eV and 342.5 eV respectively, which were assigned for its

3d_5/2and 3d_3/2states respectively of the Pd(II) oxidation states.

The characteristic band gap of 5.2 eV was also indicating to the same

conclusion. No peaks in the vicinity of 335 eV were observed con-

ﬁrming the absence of any Pd(0) state in the catalyst. The other

elements were also complying with the expected oxidation states

in the anchored catalyst, e.g. Si 2p (101.4 eV), N 1s (399.8 eV), P

2p(132.5 eV). The observed values were in good agreement with

standard oxidation states as reported elsewhere [17]. All these

3.8.1. ²⁹Si CP-MAS NMR studies

The SBA-15 material is mainly a siliceous support matrix and

therefore, ²⁹Si NMR studies is an important tool to track and iden-

tify the structural and functional changes in the matrix during the

various stages of anchoring process (e.g. selective surface function-

alization, anchoring of Pd-complex) and carbonylation reaction, if

any.

3.8.1.1. For the post synthesis grafted material [S15-NH₂-(G)-Pd].

The results of stepwise ²⁹Si CP-MAS NMR experiments of S15-NH₂-

(G)-Pd sample are presented in Fig. 8. The calcined SBA-15 and

the as-synthesized sample (curves i and ii respectively) showed

almost identical chemical shift values, with peaks at ı = −93.14,

−102.32, −112.33 ppm. The occurrence of these multiple peaks

is attributed to the different types of the Si-sites namely Q²

162

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

Fig. 6. SEM of different SBA-15 samples.

[(SiO)₂Si(OH)₂], Q³[(SiO)₃Si(OH)] and Q⁴[(SiO)₄Si], based on the

number of free silanol groups, Si OH on the Si-atom – 2, 1,

0 respectively. The spectra clearly showed that the proportion

of Q³was the highest in the SBA-15 sample as expected for a

well-ordered siliceous matrix having a high degree of cross-linking

in the silica [18]. The occurrence of identical peak patterns in the

as-synthesized and the calcined samples conclude that the SBA-

15 did not undergo any structural change due to the calcination,

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

163

Fig. 7. TEM of different mesoporous samples.

and the long-range mesoporosity order is maintained after the cal-

cination process. The curve iii of Fig. 8 presents the spectrum for

the external surface passivated SBA-15, which shows an additional

peak at ı 11.85 ppm and the disappearance of the Q²peak with

respect to curve ii. The peak at ı 11.85 ppm is attributed to the

presence of the SiMe₃moiety as the passive graft onto the surface

of SBA-15. The disappearance of the Q²peak might be due to the

absence of any dihydroxylsilyl unit, in the matrix, as those might

have been consumed in the silylation process using HMDS, leav-

ing only monohydroxylated silyl moiety, Q³and non-hydroxylated

silyl moiety, Q⁴in the matrix, with a higher proportion of the for-

mer. This might be the expected material having the outer surface

passivated by SiMe₃groups and the internal silanol groups made

available after the removal of the template. It may be mentioned

here that the passivation using HMDS was performed before the

template removal so that most of the external Si OH groups are

methylated, without affecting the internal silanol groups. The next

spectrum, curve iv represents the spectrum for the aminopropyl

functionalized SBA-15, i.e. S15-NH₂-(G), which shows appearance

of two new peaks at ı −62.28 ppm and ı −70.47 ppm, along with the

reversal of the proportions of the Q³and Q⁴peaks in comparison to

group, H₂N CH₂CH₂CH₂. These values of chemical shifts are

in good agreement of the literature [19,20]. The other peaks due

to Q³, Q⁴and ı 10 ppm however remain unchanged in comparison

to the previous spectrum i.e. curve iii. The appearance of the T^m

(m = 2 and 3) peaks conﬁrm that the organosilyl graft moiety i.e.

3-aminopropylsilyl groups are incorporated as a part of the silica

wall structure. It may be mentioned here that the proportion of

the T³is higher than that of T², which also indicates that after the

incorporation of both the passive and the active graft, only a small

amount of silanol groups are left free, and they occur only as single

silanol groups on Si-centers in the functionalized SBA-15 matrix

(as T²and Q³only). The next spectrum in Fig. 8 i.e. curve v, rep-

resents the spectrum of Pd(pyca)(PPh₃)(OTs) complex anchored to

the NH₂group of the aminopropyl moiety of the functionalized

SBA-15, i.e. the S15-NH₂-(G)-Pd material. This spectrum is very sim-

ilar to the curve iv with respect to the peak positions as well as the

relative intensities. It may be concluded from this spectrum that,

the incorporation of the Pd-complex does not affect the siliceous

mesostructure of SBA-15 at all during the anchoring process. This

is in agreement with the powder XRD patterns as seen previously

(Fig. 2). The curve vi represents the spectrum of the anchored Pd-

complex material S15-NH₂-(G)-Pd after carbonylation reaction. The

spectrum shows no change in the characteristic peaks in position

curve iii. The new peaks may be assigned to the T²

[ SiR(OSi)₂(OH)]

and T³[ SiR(OSi)₃] moieties, in which R group is the 3-aminopropyl

Table 2

XPS data for different anchored and supported catalysts (values in eV).

Elements Catalysts

Si 2p

Pd 3d_5/2; 3d_3/2

N 1s

P 2p

M41-NH₂-Pd

M48-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

Pd-Y

Pd-S

M41-NH₂-Pd (after carbonylation)

S15-NH₂-(G)-Pd (after carbonylation)

S15-NH₂-(C)-Pd (after carbonylation)

103.8

103.7

103.8

102.8

103.7

103.78

337.3; 342.5

337.1; 342.3

337.1; 342.4

337.4; 342.8

337.1; 342.4

337.4; 342.8

337.35; 342.5

337.4; 342

400

132.6

132.5

132.4

132.5

132.6

132.5

132.6

399.9

400.8

400.6

400.3

400.2

400

400.6

400.5

337.4; 342

164

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

vi

v

iii

ii

i

iv

iii

50

0

-50

ppm

-100

-150

Fig. 9. ²⁹Si CP MAS NMR spectra of S15-NH₂-(C)-Pd material. (i) Solvent extracted

S15-NH₂-(C); (ii) Pd-complex loaded [S15-NH₂-(C)-Pd], (iii) after carbonylation

reaction.

NMR spectra again conﬁrm that the functionalization of the meso-

porous matrix was complete as speculated, by the one-pot in situ

co-condensation of the precursors inside the synthesis gel, and as

a result, the functional groups have been incorporated within the

walls of the mesoporous matrix. Also, the removal of the template

after the one-pot synthesis by solvent extraction method left the

functionalized matrix S15-NH₂-(C) unchanged. The peaks resem-

ble those of the material synthesized using the grafting method

in position and intensity. The next spectrum, i.e. curve ii repre-

sents the spectrum of the Pd(pyca)(PPh₃)(OTs) complex anchored

inside the functionalized matrix, i.e. S15-NH₂-(C)-Pd. The peaks due

to the Q², Q³and Q⁴species appear at the same positions as that

in curve i, so also the peaks for T²and T³species of the function-

alized SBA-15. These imply that the mesoporous matrix did not

undergo any change in structure during the anchoring. The curve

iii in the Fig. 2.14 represents the spectrum of the S15-NH₂-(C)-Pd

after a carbonylation reaction. The observation of the peaks due to

Q², Q³and Q⁴as well as for the T²and T³in the same position with

the same intensity reﬂected on the stability of the anchored cata-

lyst even after treatment with strong acid promoter (TsOH) at high

temperature (388 K) and pressures (∼5.4 MPa). Thus, ²⁹Si CP MAS

NMR data gave important information regarding the nature of the

anchored catalyst, its formation and stability during the synthesis

as well as carbonylation reaction.

ii

i

50

0

-50

ppm

-100

-150

Fig. 8. ²⁹Si CP MAS NMR spectra of S15-NH₂-(G)-Pd material. (i) Calcined; (ii) as-

synthesized; (iii) outer silanol passivated; (iv) NH₂-functionalized; (v) palladium-

complex anchored; (vi) after carbonylation reaction.

or intensity, reinstating the stability of the anchored Pd-complex

catalyst for carbonylation.

3.8.1.2. For the one-pot co-condensed material [S15-NH₂-(C)-Pd].

The results of the sequential ²⁹Si CP MAS NMR experiments of

the catalyst material synthesized using the direct one-pot co-

condensation method are presented in Fig. 9. For this material, the

matrix synthesis was performed in the presence of the precursor

of the active graft molecule, and hence the as-synthesized material

itself held the functional group in it, so calcination was not possi-

ble. The solvent extracted material was the ﬁrst sample obtained.

Curve i shows the spectrum of the solvent extracted material. The

spectrum shows a bunch of three peaks around ı −93.6, −102.6

and −112.4 ppm, and another bunch at ı −64.07 and −71.6 ppm

respectively. The ﬁrst group of peaks is attributed to the Q², Q³

and Q⁴species of the SBA-15 matrix respectively. It may be men-

tioned here that the relative proportion of Q³is the highest for

this material showing good cross-linking of siliceous matrix. The

other two peaks at ı −64.07 and −71.6 ppm are assigned to the T²

and T³moieties respectively. The occurrence of the T^mpeaks in the

3.8.2. ¹³C CP-MAS NMR studies

The anchored catalysts consist of a molecular anchor hold-

ing the catalyst molecule at one end and itself being attached to

the support matrix. Essentially, these are organic molecules hav-

ing desired functional groups. So ¹³C CP MAS NMR studies are

important from viewpoint of understanding of the molecular envi-

ronment of the anchored catalysts. With this target, sequential ¹³C

CP MAS NMR experiments were carried out for all steps of synthe-

sis of the anchored catalysts to learn about the chemical pathway

of anchoring process, its stabilization and reactivity.

3.8.2.1. For the post synthesis grafted material [S15-NH₂-(G)-Pd].

The ¹³C CP-MAS NMR experimental results are presented in Fig. 10.

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

165

iii

iv

c

b

a

d

ii

iii

h

e

g

f

i

ii

200

150

100

50

0

-50

-100

Fig. 11. ¹³C CP MAS spectra of S-15-NH₂-(C)-Pd material. (i) NH₂-functionalized

SBA-15 by co-condensation method; (ii) Pd-complex anchored; (iii) after carbony-

lation reaction.

H₂N CH₂CH₂CH₂Si of the molecular anchor, in the left-to-

right order respectively, as similar to the literature data [22]. The

presence of these peaks directly implied the incorporation of the 3-

aminopropylsilyl anchor moiety in the walls of the SBA-15 matrix

and also proved that the aminopropyl group was not decomposed

under the preparation procedure. The curve iii of the ﬁgure rep-

resents the spectrum of Pd(pyca)(PPh₃)(OTs) complex anchored to

the NH₂-functionalized SBA-15 i.e. S15-NH₂-(G)-Pd material. The

appearance of a number of peaks in the spectrum may be attributed

to the different types of C-atoms present in the complex, in addition

to those already in the solid matrix. The assignment of each of these

peaks is given in the adjacent model, though the peaks of many C-

atoms may easily overlap for some of the atoms in the aromatic

region i.e. ∼130–170 ppm. However, the data are in satisfactory

matching with those in the literature [23]. Thus, the Pd complex

was proved to be chemically anchored to the functionalized walls of

the SBA-15 matrix using an aminopropylsilyl bridge. The next spec-

trum, i.e. curve iv represents the ¹³C spectrum of the used catalyst

after a carbonylation reaction of styrene. The spectrum is almost

similar to that of the fresh catalyst, (curve iii) and all the principle

peaks as marked are observed with negligible change in intensity

and position, and hence it may be concluded from the comparison

that the catalyst material did not change chemically during the car-

bonylation reaction. Furthermore, the Pd-complex and the other

organic fragments were intact after the reaction, proving the sta-

bility of the catalyst under the stringent carbonylation conditions.

Therefore, the post-synthetic grafting method is considered as a

truly efﬁcient methodology for the anchoring of Pd complexes to

generate very stable heterogeneous catalysts.

i

200

150

100

50

0

-50

-100

Model of Anchored Pd-complex

g

O

h

N

Me

O

b

d

O

a

Pd

O

Si

NH₂

Si

c

O

f

P

O

S

O

e

Fig. 10. ¹³C CP MAS NMR spectra of S15-NH₂-(G)-Pd material. (i) Exterior Si OH

passivated; (ii) NH₂-propyl-functionalized, (iii) Pd-complex anchored, (iv) after car-

bonylation reaction.

The purely siliceous SBA-15 matrix used for anchoring the Pd-

complex has no C-atoms in it, so the spectrum of calcined SBA-15

showed no ¹³C signals. However, for the post synthesis grafting

method, the as-synthesized material was ﬁrst treated with HMDS

to ensure that all the exterior silanol groups were passivated by

the SiMe₃moieties. This outer-surface passivated material was

solvent extracted, and the NMR result is given in curve i. The

curve showed a single peak at ı 0.00 ppm, which is attributed to

the three methyl carbons of the SiMe₃moiety. The absence of

any resonance around ı 62–70 ppm conﬁrmed that all the sur-

factants (Pluronic P123) were successfully removed during the

template removal extraction step [21]. The next spectrum, curve

ii is the spectrum of the amine-functionalized SBA-15 i.e. [S15-

NH₂-(G)]. It is notable that three distinct peaks at ı 42.7, 21.9

and 10.8 ppm are observed along with the 0 ppm peak for SiMe₃.

These correspond to the three C-atoms in the 3-aminopropyl group,

3.8.2.2. For the one-pot co-condensed material [S15-NH₂-(C)-Pd].

The ¹³C CP MAS NMR experimental results of the anchored Pd-

complex catalyst synthesized by the one-pot co-condensation

method are presented in Fig. 11. No passivation of the exterior

silanol groups was necessary for the one-pot synthesis and hence

this step was not present in the synthesis process. The ﬁrst spec-

trum in this ﬁgure, i.e. curve i showed three distinct peaks at ı 9.2,

166

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

21.1 and 42.7 ppm. These are attributed to the carbon atoms of the

3-aminopropyl group present in the molecular anchor as similar to

the previously described material i.e. S15-NH₂-(G)-Pd. The desig-

nations of the three peaks are the same as for Fig. 10. The absence

of the 0.0 ppm peak in the spectrum of S15-NH₂-(C)-Pd was also

justiﬁed, as the material was synthesized using one-pot synthesis,

without using any passivation step. The curve ii of Fig. 11 represents

the spectrum of the Pd-complex anchored material. The occurrence

of the several peaks at ∼ı 128, 140, 167, 174 ppm in the spectrum

indicate the presence of the Pd(pyca)(PPh₃)(OTs) complex attached

to the walls of the functionalized SBA-15 solid matrix. The assign-

ing of the peaks is similar to that of the spectra in Fig. 10, which

further conﬁrms that the Pd-complex was attached to the NH₂-

functionalized SBA-15 formed by the co-condensation method, in

the same manner as the post synthetic grafted material. The next

spectrum, i.e. curve iii is of the used catalyst after a run of carbony-

lation of styrene. It is noticeable here that the principle peaks in

the spectrum did not change at all (in intensity or position) after

the successful carbonylation run. Therefore, the anchored catalyst

prepared by one-pot co-condensation method was equally stable

as the previously described material.

3.8.3. ³¹P CP-MAS NMR studies

³¹P CP MAS NMR experiments of the solid catalysts is essential to

understand the coordination environment of the central Pd-atom

in the anchored catalyst. This is necessary to explain the behavior

of the anchored Pd-complex involved in catalysis in, light of the

observed change in the stereo-electronic properties in the com-

plex itself as a result of immobilization. A comparison of the ³¹P

CP MAS NMR spectra of the different anchored catalysts, namely

M41-NH₂-Pd, S15-NH₂-(G)-Pd and S15-NH₂-(C)-Pd along with that

of the Pd(pyca)(PPh₃)(OTs) complex are presented in Fig. 12. As

observed from the spectra, there is considerable change in the peak

positions in the different samples. The pure Pd-complex shows a

signal at ı 33.13 ppm (curve i). This value of the chemical shift is

slightly different from the liquid state spectrum of the complex,

which shows two signals at ∼ı 36.45 and 35 ppm respectively, cor-

responding to the isomers having N cis or trans to P existing in

the solution phase. In the solid-state spectrum such isomerization

equilibrium is not existent and also the width is much higher due to

the anisotropic effects. Each of the anchored complexes showed a

strong single peak (curves ii, iii and iv) at an upﬁeld position to

the pure complex. The peaks for the M41-NH₂-Pd, M48-NH₂-Pd,

S15-NH₂-(G)-Pd, and S15-NH₂-(C)-Pd occur at ı 17.97, 17.86 (not

shown in Fig. 12), 25.06, 25.01 ppm respectively. The occurrence

of the single peaks in all the anchored complexes is indicative of

the stability and integrity of the Pd-complex after immobilization,

as any degradation in the complex would have been reﬂected in

the ³¹P NMR spectra. The shifting of the peaks to an upﬁeld posi-

tion deﬁnitely points to some sort of interaction of the complex

with the functionalized support matrix. A probable coordination

of the N_APTS→ Pd may be visualized as the cause of such shifts.

The N-atom of the 3-aminopropyl group attached to the walls of

the siliceous matrices coordinates to the Pd-center, which in turn

thus becomes more electron rich. This electron density may how-

ever be transferred to the P-atom of the triphenyl phosphine ligand

via a probable d␲–d␲ back-bonding interaction, as the PPh₃pos-

sesses good ␲-acid properties. This enhancement in the electron

density over the P-atom will cause the upﬁeld shift of the reso-

nance peaks as observed in the spectra. The degree of variation of

this phenomenon may be different for the different matrices (MCM-

41, SBA-15 etc.), which might depend on the unit cell dimensions

and hence vary the molecular constraints and mobility of the Pd-

complex inside the porous channels. For example, the narrower

channels in MCM-41 (diameter ∼ 3.5 nm) might have very different

set of stereo-electronic constraints than for the larger pore SBA-15

Fig. 12. ³¹P CP MAS spectra of various anchored Pd-complex catalyst materials.

(i) Pd(pyca)(PPh₃)(OTs) complex; (ii) M41-NH₂-Pd; (iii) S15-NH₂-(G)-Pd; (iv) S15-

NH₂-(C)-Pd.

matrix (diameter ∼ 10 nm). The ³¹P CP MAS NMR of the used cata-

lysts M41-NH₂-Pd, S15-NH₂-(G)-Pd, and S15-NH₂-(C)-Pd occur at ı

17.93, 25.1, 25.0 ppm respectively. These are in good agreement in

comparison to the pristine catalysts, emphasizing upon the intact-

ness of the catalysts and Pd-center precisely, after the carbonylation

of styrene.

Therefore, ³¹P CP MAS NMR experiments supported the pro-

posed chemical anchoring of the Pd-complexes to the solid matrices

via a N_APTS→ Pd coordination. The stability of the complexes dur-

ing the immobilization process is successfully proven by these NMR

studies. The stability of the anchored complexes under the carbony-

lation environment was also appreciable.

3.9. Palladium content analysis

All the four anchored palladium complexes were analyzed

for Pd-content using the atomic absorption spectroscopic (AAS)

method. The samples of a ﬁxed weight were dissolved and digested

in concentrated HNO₃and were diluted to a ﬁxed volume. AAS

experiments were conducted using these solutions. The other

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

167

Fig. 13. (A) A typical concentration-time proﬁle of carbonylation of styrene using M41-NH₂-Pd (B) major-product (2-PPA) concentration vs. time proﬁles for different

anchored catalysts. Reaction conditions: catalyst: 100 mg; styrene: 9.6 mmol; LiCl: 0.5 mmol; TsOH: 0.5 mmol; PPh₃: 0.095 mmol; H₂O: 50 mmol; solvent: MEK; P_CO: 5.4 MPa;

temperature: 388 K, volume: 25 mL.

supported Pd-complex catalysts i.e. Pd-S and Pd-Y were also ana-

lyzed similarly. The results of AAS analysis are presented in Table 3.

It was generally observed that the palladium content of the differ-

ent anchored Pd-complex catalysts were roughly in the range of

0.1–0.3% by weight for the anchored catalysts while, the Pd-content

was relatively higher for the supported catalysts. This may be due

to the fact that for the supported catalysts the complex is simply

adsorbed at random to the surface and voids, while for the anchored

catalysts, the complex are attached to only the speciﬁc anchoring

sites, which are much less in number.

maximum and then reduced at the end of the reaction. A typi-

cal concentration-time proﬁle for carbonylation of styrene using

M41-NH₂-Pd is presented in Fig. 13A. It is noticeable that the reac-

tion has an initial period of ∼3 h as the induction period of the

reaction. After the induction period, the reaction rate increases

signiﬁcantly. There was nearly ∼100% mass balance of the reac-

tants consumed and products formed. The substrate and the major

product (2-phenylpropionic acid, 2-PPA) are plotted on the left Y-

axis while the minor product (3-phenylpropionic acid, 3-PPA) and

the chloro intermediate (1-chloroethylbenzene, 1-CEB) are plot-

ted on the right Y-axis for better clarity. Upon completion of 12 h,

a high conversion of 98.12% of styrene was observed. A turnover

number of 5568 was observed with a corresponding high turnover

frequency of 464 h⁻¹. The regioselectivity of >99.31% to the desired

branched acid (2-PPA) was obtained.

To compare the carbonylation activities of the other anchored

catalysts, the standard reaction was repeated using M48-NH₂-Pd,

S15-NH₂-(G)-Pd and S15-NH₂-(C)-Pd, and the results are presented

in Fig. 13B. An important observation was that the reaction using

the SBA-15 based catalyst [S15-NH₂-(G)-Pd] was found to be

the most active anchored Pd-complex catalyst. This reaction got

over (98.25% conversion) in 12 h with a very high regioselectivity

(99.54%) to the branched-acid product. The TOFs were calculated as

464, 417, 760 and 401 h⁻¹respectively for the catalysts M41-NH₂-

Pd, M48-NH₂-Pd, S15-NH₂-(G)-Pd and S15-NH₂-(C)-Pd. Thus, the

anchored Pd-complex catalysts proved to be efﬁcient carbonylation

catalysts. But the observed highest TOF for the S15-NH₂-(G)-

Pd is distinctly different from the other SBA-15 based catalyst

[S15-NH₂-(C)-Pd]. The probable reason might be that during the

one-pot co-condensation synthesis of the catalyst [S15-NH₂-(C)-

Pd], many of the dangling active NH₂-moieties might be well

embeded in the framework as well as in the micropores of the

mesoporous channels, which would also anchor the Pd-complex

but cannot take part in catalytic reaction, probably due to con-

formational constraints. Hence, even though the co-condensed

catalyst shows a higher total Pd-content, active number of Pd-sites

may be less leading to a lower TOF.

3.10. Carbonylation of oleﬁns and alcohols

Carbonylation of aryl oleﬁns and alcohols were performed using

the anchored Pd(pyca)(PPh₃)OTs) complex inside the mesoporous

channels of MCM-41, MCM-48 and SBA-15. Following the homoge-

neous counterpart [5] as the standard benchmark, all the reactions

were carried out in the presence of an acidic and an alkali metal

halide promoter, with a little amount of free triphenyl phos-

phine (5–25 mg) with a target to obtain highest catalytic activity,

selectivity and a truly heterogeneous catalyst for the carbonyla-

tion reactions. Styrene was used as the model substrate for the

standardization reactions, however, different substrates were also

tried, the results of which will be discussed in a later section.

The reactions using all the four anchored Pd-complex catalysts

i.e. M41-NH₂-Pd, M48-NH₂-Pd, S15-NH₂-(G)-Pd and S15-NH₂-(C)-

Pd required long durations of ∼12 h to complete. >98% conversion

was observed for all the catalysts with high chemo- and regiose-

lectivity of >99%. Similar to the homogeneous Pd(pyca)(PPh₃)OTs)

complex catalyst, the carbonylation of styrene proceeds via

intermediate chloro-derivative which might be the active sub-

strate in the catalytic cycle. The amount of the chloro-derivative,

i.e. 1-chloroethylbenzene (1-CEB) gradually increased, reached a

Table 3

Pd-content of different anchored and supported Pd-complex catalysts.

Entry

Material

Pd-content, wt%

1

2

3

4

5

6

M41-NH₂-Pd

M48-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

Pd-S

0.18

0.20

0.11

0.19

0.62

0.81

3.11. Recycle and stability aspects

With the demonstration of the high activity of the anchored

catalysts, it was important to know the recyclability and stability

of the anchored catalysts. The comparative recycle studies of the

Pd-Y

168

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

Fig. 15. Effect of substrate concentration on styrene carbonylation. Reaction con-

ditions: catalyst: 100 mg; LiCl: 0.5 mmol; TsOH: 0.5 mmol; PPh₃: 0.095 mmol; H₂O:

50 mmol; solvent: MEK; P_CO: 5.4 MPa; T: 388 K, duration: 12 h, volume: 25 mL.

Fig. 14. Recycle studies using different anchored Pd-complex catalysts. Reaction

conditions: catalyst: 100 mg; styrene: 9.6 mmol; LiCl: 0.5 mmol; TsOH: 0.5 mmol;

PPh₃: 0.095 mmol; H₂O: 50 mmol; solvent: MEK; P_CO: 5.4 MPa; temperature: 388 K,

duration: 12 h, volume: 25 mL.

in Fig. 15. It was observed that the conversion of styrene changed

with its concentration for all the catalysts, with marginal variations

considering a ﬁxed reaction time of 12 h. When the concentration

was half of that of the standard charge, (∼0.197 kmol m⁻³) the

reactions went to complete conversion (>99%) for all the catalysts

in 12 h. The regioselectivity to the branched acid product was

found to be >99.5% for all the catalysts. When the substrate con-

centration was doubled to 0.768 kmol m⁻³the reaction rate was

almost similar and as a result, the conversion at the end of 12 h was

much less (∼63%) compared to the standard reaction. On further

increase in the substrate concentration to four times the standard

(∼1.536 kmol m⁻³), the conversion of styrene decreased, with only

∼35% conversion for all the catalysts. However, in all the cases, the

regioselectivity to 2-PPA was as high as 99.5% for all the catalysts.

The constancy of the regioselectivity values prove that the catalytic

cycle is unhindered by the changes in styrene concentrations.

four anchored Pd-complex catalysts are presented in Fig. 14. All the

reactions were carried out up to a ﬁxed duration of 12 h and each

catalyst was recycled for three successive cycles. It is clearly evident

from the results in Fig. 14 that the anchored complex catalysts in

mesoporous matrices recycled efﬁciently for at least three cycles,

without any signiﬁcant variation in the activity and regioselectivity.

The cumulative turnover numbers considering recycle experiments

were observed to be >22,000 for all the anchored catalysts with the

highest value of 36,500 for S15-NH₂-(G)-Pd catalyst. The regiose-

lectivity to the branched-acid product (2-PPA) was >99.4% for all

the catalysts and recycle experiments.

The anchored catalysts showed excellent stability and even

though a very negligible amount of Pd (∼3.5 × 10⁻⁴% for M41-NH₂-

Pd; 3.26 × 10⁻⁴% for M48-NH₂-Pd; 2.16 × 10⁻⁴% for S15-NH₂(G)-Pd;

and 2.38 × 10⁻⁴% for S15-NH₂(C)-Pd) was lost during each run as

observed from the ICP-OES analysis, the change in the TOF was

quite insigniﬁcant. To understand the importance of “chemically”

anchoring Pd(pyca)(PPh₃)(OTs) complex to the mesoporous matri-

ces, the recycle performances were also compared with Pd-Y and

Pd-S catalysts were employed for carbonylation of styrene. Both

Pd-Y and Pd-S showed extensive leaching of Pd during carbonyla-

tion and as much as 20% of the Pd-loaded was lost during the ﬁrst

reactions. Even when NaY zeolite was used as a support matrix

(in Pd-Y) having a regular deﬁned structure, ∼18% of the total pal-

ladium charged in a particular reaction was leached out. Also, the

minute black precipitate observed in the latter cases on standing for

a few hours after the reaction, indicates degradation of the detached

Pd-complex under the reaction conditions to Pd(0). For the four

anchored catalysts, no such degradation was observed indicating

their excellent stability during carbonylation.

3.12.2. Effect of catalyst charge

The catalyst charge was varied in the range of 50 to 600 mg,

alongwith the standard charge of other components in the carbony-

lation reactions, for all the four anchored Pd-complex catalysts,

namely, M41-NH₂-Pd, M48-NH₂-Pd, S15-NH₂-(G)-Pd and S15-NH₂-

(C)-Pd. The observed variation in the catalyst activity, with the

varying catalyst charge is shown in Fig. 16. The trends show that

the activity (expressed as the TOF) remained same with variation

in the catalyst charge. The S15-NH₂-(G)-Pd catalyst showed highest

activity among the other catalysts. The lowest TOFs were obtained

for the S15-NH₂-(C)-Pd catalyst due to the higher Pd-content in it.

The catalyst concentration however did not have any effect on the

regioselectivity of the reaction also, and >99% regioselectivity was

obtained for the branched acid, 2-PPA for all of the reactions using

the four anchored catalysts. When the catalyst concentration was

increased, the actual number of the active catalytic sites increased,

but the TOF remains constant due to the proportional increase in

the catalyst-weight.

3.12. Parametric effects

The effect of variation of the reaction parameters such as sub-

strate concentration, catalyst loading, promoter concentrations and

ratios of H⁺and Cl⁻, CO-partial pressure, temperature etc. was stud-

ied and the results are discussed below.

3.12.3. Effect of CO partial pressure

Reactions were carried out using the four anchored catalysts

for different carbon monoxide partial pressures. The results are

presented in Fig. 17. It is clearly noticeable from the graphs that

the catalytic activity was inﬂuenced directly by the CO partial

pressure. The activity (expressed in terms of TOF of the reaction)

seemed to vary almost linearly with respect to the partial pressure

of carbon monoxide, in the range of the CO-pressures selected. The

3.12.1. Effect of substrate concentration

Substrate concentration effect was investigated using three dif-

ferent styrene concentrations keeping all other reaction conditions

constant. The results using the different catalysts i.e. M41-NH₂-Pd,

M48-NH₂-Pd, S15-NH₂-(G)-Pd and S15-NH₂-(C)-Pd are presented

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

169

S15-NH₂-(C)-Pd

S15-NH₂-(G)-Pd

M48-NH₂-Pd

900

800

700

600

500

400

300

200

1000

500

0

M41-NH₂-Pd

M48-NH -Pd

S15-NH₂²-(G)-Pd

S15-NH₂-(C)-Pd

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Factor of Promoter Conc. w.r.t. standard

100

200

300

400

500

600

Catalyst charge, mg

Fig. 18. Effect of (LiCl + TsOH) promoter amount on carbonylation of styrene.

Reaction conditions: catalyst: 100 mg; styrene: 9.6 mmol; LiCl-to-TsOH ratio = 1:1

(standard amount: 0.5 mmol each); PPh₃: 0.095 mmol; H₂O: 50 mmol; solvent:

MEK; P_CO: 5.4 MPa; temperature: 388 K, volume: 25 mL.

Fig. 16. Effect of catalyst charge on styrene carbonylation. Reaction conditions:

styrene: 9.6 mmol; LiCl: 0.5 mmol; TsOH: 0.5 mmol; PPh₃: 0.095 mmol; H₂O:

50 mmol; solvent: MEK; P_CO: 5.4 MPa; T: 388 K, volume: 25 mL.

direct dependence of the activity of all the four anchored catalysts

with the partial pressure of CO reﬂects the involvement of carbon

monoxide in the catalytic cycle as an important step. The regiose-

lectivity was however, constant at > 99.5% at different CO pressures.

of the chloro-derivative as the active substrate around 1:1 ratio of

promoters. The regioselectivity to the branched isomeric product

did not alter at all due to such variations as mentioned for any of

the four different anchored catalysts. Therefore, a mutual ratio of

1:1 proved to be the best for the carbonylation of styrene.

3.12.4. Effect of promoters

However, after obtaining the optimum ratio of TsOH and LiCl

as 1:1 the issue of total concentration of the promoters becomes

important. The concentrations of the promoters were varied at a

constant ratio of 1:1 and the results are presented in Fig. 18. The

graph shows that when the promoter amount was reduced to half,

the TOF for all the four catalysts became nearly half when compared

to the standard runs. In addition, when the promoter charge was

doubled or quadrupled, increase in the TOF was not proportional.

The use of free triphenylphosphine as promoter was necessary

for the carbonylation of styrene using the anchored Pd-complex

catalysts. A speculative explanation can be presented as that,

PPh₃can stabilize the Pd(0) species which might be formed as

an intermediate in the catalytic cycle. Different concentrations

of PPh₃were used to study this effect for all the four catalysts.

The results of these studies are presented in Fig. 19. It is clearly

observed from the graphs that the reaction progress is dependent

upon the concentration of the excess triphenyl phosphine only at

As already demonstrated for the homogeneous catalysis using

the Pd(pyca)(PPh₃)(OTs) complex, TsOH and LiCl have been used

as the sources of H⁺and Cl⁻ions, which promote the carbonyla-

tion reactions achieving the highest turnovers reported [5]. Two

different aspects were recognized to affect the reactions, such as

the ratio of the promoters and the total quantity of the promot-

ers used. The effect of the ratio of promoters on the performance

of the carbonylation using all the catalysts is presented in Table 4.

The data represents that the activities of the reactions did not vary

signiﬁcantly when the ratio of the two promoters was changed.

While using the 1:1 ratio of TsOH to LiCl in the reaction mixture,

the reactions showed slightly better performances for all the four

anchored catalysts. Even though we had observed the involvement

of the chloro-derivative (1-CEB) as the intermediate in the carbony-

lation of styrene, formed by the reaction of styrene and Cl⁻, the

conversions and turnovers for the reactions using TsOH: LiCl as 1:2

were similar to reactions using 1:1 ratio of the respective promot-

ers. The reason behind this might be the optimum concentration

800

700

600

500

400

1000

900

800

700

600

500

300

M41-NH -Pd

M48-NH²₂-Pd

400

200

M41-NH₂-Pd

M48-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

S15-NH -(G)-Pd

S15-NH²₂-(C)-Pd

300

200

100

0

1

2

3

4

5

6

7

8

3

4

5

6

7

Conc. of PPh₃x 10³, kmol.m^-3

CO partial pressure, MPa

Fig. 19. Effect of PPh₃concentration on styrene carbonylation. Reaction condi-

tions: catalyst: 100 mg; styrene: 9.6 mmol; LiCl: 0.5 mmol; TsOH: 0.5 mmol; H₂O:

50 mmol; solvent: MEK; P_CO: 5.4 MPa; temperature: 388 K, volume: 25 mL.

Fig. 17. Effect of CO partial pressure on styrene carbonylation. Reaction condi-

tions: catalyst: 100 mg; styrene: 9.6 mmol; LiCl: 0.5 mmol; TsOH: 0.5 mmol; PPh₃:

0.095 mmol; H₂O: 50 mmol; solvent: MEK; T: 388 K„ volume: 25 mL.

170

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

Table 4

Effect of ratio of (TsOH, LiCl) promoters on styrene carbonylation for different anchored Pd-complex catalysts.

Sl. no.

Catalyst

Ratio of TsOH:LiCl

Conv., mol%

Time, h

Selectivity, %

iso

TOF, h⁻¹

n

1

2

3

4

5

6

7

8

M41-NH₂-Pd

M48-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

2:1

1:1

1:2

2:1

1:1

1:2

2:1

1:1

1:2

2:1

1:1

1:2

96.5

98.1

97.5

97.2

98.4

97.6

98.2

96.8

98.2

97.6

12

99.23

99.31

99.32

99.10

99.20

99.40

99.54

99.50

99.21

99.46

99.35

0.1

0.68

0.6

0.1

456

464

461

412

417

414

755

760

758

395

401

398

0.1

0.50

0.44

0.46

0.76

0.52

0.64

9

10

11

12

Reaction conditions: catalyst: 100 mg; styrene: 9.6 mmol; LiCl: 0.5 mmol; TsOH: 0.5 mmol; PPh₃: 0.095 mmol; H₂O: 50 mmol; solvent: MEK; P_CO: 5.4 MPa; temperature:

388 K, duration: 12 h, volume: 25 mL.

lower concentrations. The activity of the catalysts increase almost

linearly with the concentration of PPh₃until a speciﬁc quantity,

and beyond that the reaction rates did not alter showing a plateau

in the graph. The nature of variation of the catalytic activity

was same for all the four anchored catalysts, but the turnover

frequencies are different for each of them, due to their intrinsic

different activities. The physical signiﬁcance of the enhancement

in the activity may be directly linked to the number of active sites,

and in each case the PPh₃present will just aggravate the activity

until a certain concentration of it only. Further to this limiting

value, the activity remains constant.

3.12.5. Effect of water

Water has multiple roles to play in the carbonylation of oleﬁns

and alcohols. It is not only a reactant for the hydroxycarbonyla-

tion of oleﬁns, but also helps to solubilize the polar promoters

(LiCl particularly) in the reaction mixture (MEK solvent), during

carbonylation. As because the miscibility of the water is ∼15% in

Table 5

Carbonylation of different substrates using different anchored Pd-complex catalysts.

Sl. no.

Catalysts

Substrate

Conversion, mol%

Selectivity, %

Iso

TON

Time, h

TOF, h⁻¹

n

1

2

3

4

5

6

7

8

9

M41-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

M41-NH₂-Pd

Styrene

98.1

98.4

98.2

85.8

86.2

83.1

87.3

93.4

95.1

95.0

94.8

91.3

94.4

94.1

94.0

90.5

93.9

93.5

94.7

92.8

95.2

91.6

97.3

96.3

97.4

96.4

95.6

96.3

95.7

95.2

95.0

96.1

96.7

99.31

99.10

99.54

99.46

98.60

99.10

99.66

99.53

99.31

99.23

99.32

99.29

98.27

98.33

98.65

98.47

99.01

98.86

99.03

99.10

98.42

98.16

99.06

98.89

98.35

98.67

99.06

98.94

99.60

99.50

99.60

99.46

97.50

97.90

98.10

97.89

0.68

0.90

0.44

0.52

1.38

0.88

0.44

0.46

0.67

0.75

0.67

0.68

1.71

1.65

1.12

0.68

0.90

0.12

0.95

0.86

1.55

1.82

0.82

1.10

1.56

1.30

0.91

1.05

0.38

0.49

0.40

0.52

2.45

1.18

1.90

1.11

5568

5004

9120

4812

4872

4404

7716

4248

5300

4858

8822

4604

5184

4824

8736

4572

5136

4800

8688

4548

5376

4740

8844

4452

5520

4920

9048

4740

5472

4884

8940

4656

5400

4853

8928

4704

12

464

417

760

401

406

367

643

354

441

405

735

383

432

402

728

381

428

400

724

379

448

395

737

371

460

410

754

395

456

407

745

388

450

405

744

392

4-Methylstyrene

4-^tButylstyrene

4-Acetylstyrene

4-Methoxystyrene

4-Chlorostyrene

4-Isobutylstyrene^a

4-Isobutylstyrene

1-Phenylethanol

IBPE^b

M41-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

M41-NH₂-Pd

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

M41-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

M41-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

M41-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

M41-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

M41-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

M41-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

M41-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

IBPE

Reaction conditions: catalyst: 100 mg; substrate: 9.6 mmol; LiCl: 0.5 mmol; TsOH: 0.5 mmol; PPh₃: 0.095 mmol; H₂O: 50 mmol; solvent: MEK; P_CO: 5.4 MPa; temperature:

388 K, agitation: 16.7 Hz, volume: 25 mL.

a

10 ppm t-butylcatechol was used as polymerization inhibitor for 4-isobutylstyrene.

IBPE = 1-(4-isobutylphenyl)ethanol.

b

Table 6

Comparative study of anchored Pd-complex catalysts with principle examples from literature for carbonylation of oleﬁns and alcohols.

Entry

Substrate

Catalyst system

Conv., mol%

Selectivity, %

TON

Time, h

TOF, h⁻¹

Pd-loss in

reaction

Remarks and reference

Iso

99.01

n

1

2

Styrene

Homogeneous Pd(pyca)(PPh₃)(OTs)

Pd(OAc)₂–Montmorillonite–PPh₃/HCl

97

0.98

–

468

0.18

24

2600

–

Tedious catalyst–product

separation and reuse

Hydroesteriﬁcation, long duration,

used C₆H₆–MeOH as solvent – Lee

et al. [24]

Substituted

styrenes

88–100

100

–

20–30

3

4

Styrene and IBS

Styrene

PdCl₂/TPPTS/TsOH and other

biphasic systems

Montmorillonite–dipenyl

phosphine–PdCl₂–HCl

62–100

73–80

56–90

33–10

–

240–1200

–

6

40–200

16–25

–

Low activity [26–29]

50–100

92

24

Hydroesteriﬁcation, long duration,

used C₆H₆–MeOH as solvent –

Nozaki et al. [25]

Catalyst deactivates in air, recycles

only in CO atmosphere – Jayasree

et al. [30]

Heavy leaching obtained, high

activity due to leached Pd –

Jayasree et al. [31]

5

6

Styrene

Aq. biphasic Pd(pyca) (TPPTS)

Pd/C

94.5

92

7.2

1.9

428

1.42

4.6

302

0.225%

37.5%

Oleﬁns and

alcohols

98

13,340

2900

7

8

Substituted

styrenes

Styrene

Silica–Chitosan–Pd complex

M41-NH₂-Pd

86–97

98.1

88–95

99.3

10–3

0.68

–

10

12

Poor selectivity, use of hazardous

co-catalysts – Zhang et al. [32]

Our work [7] highly stable

recyclable and truly heterogeneous

catalyst with negligible leaching

5568

464

3.5 × 10⁻⁴

%

9

10

11

Styrene

M48-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

98.4

98.2

99.1

99.5

99.4

0.90

0.44

0.52

5004

9120

4812

12

417

760

401

–

2.16 × 10⁻⁴

%

–

172

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

800

700

600

500

400

recycled only under CO-atmosphere and lost activity after 2–3

recycles only. Pd-Charcoal catalyst as demonstrated by Jayasree

et al. [31] showed extremely high TOFs for IBS carbonylation but

high leaching of Pd under reaction conditions was also observed.

Use of SiO₂-supported Chitosan–Pd complex catalyst [32] showed

poor selectivity; also the presence of another metal (Co/Ni/Cr/Fe-

chlorides and other Lewis acids) is another obstacle for making a

recyclable catalyst (Conversion ∼80% and iso-selectivity ∼60% for

3rd recycle run). Our previously reported ossiﬁed catalysts showed

good performance but with the limitation of lower TOFs. [8] Also,

TOF observed for S15-NH₂-(G)-Pd catalyst is better than our previ-

ously reported anchored catalyst M41-NH₂-Pd [7]. In comparison

to these literature results, the current anchored catalysts showed

high activity (TOF > 750 h⁻¹for the S15-NH₂-(G)-Pd catalyst), very

high regioselectivity (>99.5%), in addition to the extremely low Pd-

loss by leaching makes the anchored catalysts the most-promising

candidates as heterogeneous carbonylation catalyst.

M41-NH₂-Pd

M48-NH₂-Pd

S15-NH₂-(G)-Pd

S15-NH₂-(C)-Pd

300

0

1

2

3

4

5

6

H₂O Conc., kmol.m^-3

Fig. 20. Effect of water on styrene carbonylation. Reaction conditions: catalyst:

100 mg; substrate: 9.6 mmol; LiCl: 0.5 mmol; TsOH: 0.5 mmol; PPh₃: 0.095 mmol;

solvent: MEK; P_CO: 5.4 MPa; temp: 388 K, agitation: 16.7 Hz., volume: 25 mL.

4. Conclusion

Immobilized Pd-complexes inside mesoporous materials have

been successfully synthesized and characterized unambiguously

using different techniques. All the characterizations conﬁrmed

that the anchoring of the Pd-complex occurred as designed. The

materials showed excellent activity, regioselectivity, stability and

recyclebility for the low-pressure carbonylation of oleﬁns and

alcohols. From the comparative study of activity of the four

different catalysts, the catalyst based on SBA-15 synthsized by post-

synthetic grafting method showed the best activity. The cumulative

turnovers (∼36,000 till 3rd recycle of styrene using S15-NH₂(G)-

Pd) of the recycle runs, coupled with the excellent stability and

low-leaching indicate that the anchored catalysts reported here are

truly heterogeneous carbonylation catalysts and useful in practical

applications.

MEK, the variation in the quantity of water in the reaction mixture

is important to avoid liquid–liquid phase separation and related

added transport limitations. The concentration of water was var-

ied within the soluble range. The detailed results are presented in

Fig. 20. An important observation was that the activity of the cata-

lyst had little variation in the lower concentration region of water,

but this variation gradually diminished when larger amounts of

water was used during the reaction. The reason of such an obser-

vation may be that the amount of water added was more than

the amount as required for the chemical reaction, and the excess

water might facilitate in the solvation and other similar phenom-

ena, which indirectly vary the activity of the catalysts. Reactions

were also carried out without any added water, but then the con-

versions were within only ∼2.5–3.2% range. The water present in

the solvent and the other reagents added during the reaction might

be responsible for such an observation.

Acknowledgment

BRS would like to thank CSIR, India for Research fellowship for

the part of the work performed at NCL India.

3.13. Different substrates

References

The results of carbonylation of different substrates using the four

different anchored Pd-complex catalysts are presented in Table 5.

It was observed that for almost all the substrates, the activity for

the SBA-15 based catalyst (S15-NH₂-(G)-Pd) was the highest (TOF

in the range of >700 h⁻¹). The trends in catalytic activity and regios-

electivity were found to be similar for the substartes studied.

[1] H.M. Colquhoun, D.J. Thompson, M.V. Twigg, Carbonylation-Direct Synthesis of

Carbonyl Compounds, Plenum Press, New York, 1991.

[2] L. Weisse, Specialty Chemicals Magazine 22 (3) (2002) 36.

[3] V. Elango, M.A. Murphy, G.N. Mott, E.G. Zey, B.L. Smith, G.L. Moss, EP 400,892

(1990).

[4] A. Seayad, S. Jayasree, R.V. Chaudhari, Organic Letters 1 (1999) 459.

[5] S. Jayasree, A. Seayad, R.V. Chaudhari, Organic Letters 2 (2000) 203.

[6] Y. Li, R.V. Chaudhari, Industrial and Engineering Chemistry Research 50 (2011)

9677.

3.14. Comparative advancement studies

[7] K. Mukhopadhyay, B.R. Sarkar, R.V. Chaudhari, Journal of the American Chem-

ical Society 124 (2002) 9692.

[8] B.R. Sarkar, R.V. Chaudhari, Journal of Catalysis 242 (2006) 231.

[9] K. Mukhopadhyay, A. Ghosh, R. Kumar, Chemical Communications (2002) 2404.

[10] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky,

Science 279 (1998) 548.

[11] D.S. Shephard, W. Zhou, T. Maschmeyer, J.M. Matters, C.L. Roper, S. Parsons,

B.F.G. Johnson, M.J. Duer, Angewandte Chemie International Edition 37 (19)

(1998) 2719.

[12] T. Asefa, R.B. Lennox, Chemistry of Materials 17 (10) (2005) 2483.

[13] X. Wang, K.S.K. Lin, J.C.C. Chen, S. Cheng, Journal of Physical Chemistry B 109

(2005) 1763.

[14] A. Ghosh, Ph.D. Thesis, University of Pune, 2004, p. 63 (Chapter 2).

[15] S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and Porosity, Academic Press,

London, 1967, p. 121.

[16] C.-Y. Chen, H.-X. Li, M.E. Davis, Microporous Materials 2 (1993) 17.

[17] C.D. Wagner, W.M. Riggs, L.E. Davis, J.E. Moulder, G.E. Muilenberg (Eds.), Hand-

book of X-ray Photoelectron Spectroscopy, Perkin-Elmer Corporation, USA,

1979.

[18] C.Y. Chen, H.X. Li, M.E. Davis, Microporous Materials 2 (1993) 27.

[19] M. Llusar, G. Monros, C. Roux, J.L. Pozzo, C. Sanchez, Journal of Materials Chem-

istry 13 (2003) 2505.

A comparative study of the anchored Pd(pyca)(PPh₃)(OTs) com-

plex catalysts on MCM-41, MCM-48 and SBA-15, with the other

heterogeneous Pd-based catalysts for the carbonylation of oleﬁns

and alcohols is essential to understand the advancement made in

the development of these catalysts. The results are compared with

previous work on carbonylation of aryl substituted oleﬁns/alcohols

in Table 6. As evident form the comparison table, the data of Lee and

Alper [24] and Nozaki et al. [25], using Montmorillonite as support

gave good conversion (88–100%) and branched acid selectivity but

the duration was too long (24 h) and involved benzene as solvent,

thus these catalyst systems had lower activity (TOF = 20–30 h⁻¹).

Biphasic systems developed by Papadogianakis et al. [26], Tilloy

et al. [27]; Bertoux et al. [28] and Goedheijt et al. [29] were effec-

tive for styrene carbonylation but with only lower conversions

(∼50%). The Biphasic catalyst system developed by Jayasree et al.

[30] showed good conversions and selectivities but the catalyst

B.R. Sarkar, R.V. Chaudhari / Catalysis Today 198 (2012) 154–173

173

[20] Y.Q. Wang, C.M. Yang, B. Zibrowius, B. Spliethoff, M. Linden, F. Schuth, Chem-

istry of Materials 15 (2003) 5029.

[26] G. Papadogianakis, G. Verspui, L. Matt, R.A. Sheldon, Catalysis Letters 47 (1997)

43.

[21] C. Yang, B. Zibrowius, F. Schuth, Chemical Communications (2003)

1772.

[27] S. Tilloy, E. Monﬂier, B. Bertoux, Y. Castanet, A. Mortreux, New Journal of Chem-

istry 21 (1997) 529.

[22] H.H.P. Yiu, P.A. Wright, N.P. Botting, Journal of Molecular Catalysis B: Enzymatic

15 (2001) 81.

[28] F. Bertoux, E. Monﬂier, Y. Castanet, A. Mortreux, Journal of Molecular Catalysis

A: Chemical 143 (1999) 23.

[23] C. Zapilko, M. Widenmeyer, I. Nagl, F. Estler, R. Anwander, G. Raudaschl-Sieber,

O. Groeger, G. Engelhardt, Journal of the American Chemical Society 128 (50)

(2006) 16266.

[24] C.W. Lee, H. Alper, Journal of Organic Chemistry 60 (1995) 250.

[25] K. Nozaki, M.L. Kantam, T. Horiuchi, H.J. Takaya, Journal of Molecular Catalysis

A: Chemical 118 (1997) 247.

[29] M.S. Goedheijt, J.N.H. Reek, P.C.J. Kamer, P.W.N.M. van Leeuwen, Chemical Com-

munications (1998) 2431.

[30] S. Jayasree, A. Seayad, R.V. Chaudhari, Chemical Communications (2000) 1239.

[31] S. Jayasree, A. Seayad, R.V. Chaudhari, Chemical Communications 12 (1999)

1067.

[32] J. Zhang, C. Xia, Journal of Molecular Catalysis A: Chemical 206 (1–2) (2003) 59.

Article Doi

Anchored Pd-complexes in mesoporous supports: Synthesis, characterization and catalysis studies for carbonylation reactions

DOI: 10.1016/j.cattod.2012.04.047

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.cattod.2012.04.047

Full text of DOI:10.1016/j.cattod.2012.04.047

Products guided by the article

R&D Labs maybe for 228575-92-6

Relevant to this article

Hot Product