Solventless triarylmethane synthesis via hydroxyalkylation of anisole with benzaldehyde by modified heteropoly acid on mesocellular foam silica (MCF)

Article abstract of DOI:10.1016/j.mcat.2018.06.003

Triarylmethane (TRAM) compounds have wide applications such as leuco dyes for sensing tumors and other biological activities. Hydroxyalkylation of arenes with benzaldehyde results in formation of triarylmethane compounds. In the present study, 20 (wt.%) Cs_2.5H_0.5PW₁₂O₄₀ (Cs-DTP) supported on mesocellular foam (MCF) silica was prepared, characterized and tested for its activity in hydroxyalkylation reaction of anisole with benzaldehyde. Its activity was compared with commercial catalysts like Amberlyst-15, montmorillonite clay K-10, H₃PW₁₂O₄₀ and unsupported Cs_2.5H_0.5PW₁₂O₄₀.The prepared catalyst showed the best activity compared to others with advantage of separation of catalyst and reusability. Reaction parameters were studied in detail and kinetic study was carried out for the said reaction. 20 (wt. %) Cs-DTP/MCF was found to be the best, robust and reusable catalyst. Reaction mechanism and kinetics were also studied. The results are new.

Full text of DOI:10.1016/j.mcat.2018.06.003

MolecularCatalysis455(2018)150–158
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Solventless triarylmethane synthesis via hydroxyalkylation of anisole with
benzaldehyde by modified heteropoly acid on mesocellular foam silica
(MCF)
Kalpesh H. Bhadra, Ganapati D. Yadav⁎
Department of Chemical Engineering, Institute of Chemical Technology, Nathalal Parekh Marg, Matunga, Mumbai, 400 019, India
A R T I C L E I N F O
A B S T R A C T
Keywords:
Triarylmethane (TRAM) compounds have wide applications such as leuco dyes for sensing tumors and other
biological activities. Hydroxyalkylation of arenes with benzaldehyde results in formation of triarylmethane
compounds. In the present study, 20 (wt.%) Cs2.5H0.5PW12O40 (Cs-DTP) supported on mesocellular foam (MCF)
silica was prepared, characterized and tested for its activity in hydroxyalkylation reaction of anisole with
benzaldehyde. Its activity was compared with commercial catalysts like Amberlyst-15, montmorillonite clay K-
10, H3PW12O40 and unsupported Cs2.5H0.5PW12O40.The prepared catalyst showed the best activity compared to
others with advantage of separation of catalyst and reusability. Reaction parameters were studied in detail and
kinetic study was carried out for the said reaction. 20 (wt. %) Cs-DTP/MCF was found to be the best, robust and
reusable catalyst. Reaction mechanism and kinetics were also studied. The results are new.
Hydroxyalkylation
Mesocellular foam silica (MCF)
Heteropoly acid
Anisole
Cs2.5H0.5PW12O40
Benzaldehyde
1. Introduction
catalysts like AlCl3 and BF3-H2O [11,12], and other metal based Lewis
acid like AuCl3/AgOTf [13], [Ir(COD)Cl]2-SnCl4 [14], FeCl3/Ac2O [15],
Solventless reactions and use of heterogeneous catalysts are
amongst the twelve important principles of green chemistry. Both
principles improve the commercial feasibility of organic reactions.
Friedel-Crafts alkylation and acylation are ubiquitous in a spectrum of
industries and used, for instance, to make pharmaceuticals, fine che-
micals and specialities [1,2]; however, a majority of these processes
involves the use of AlCl3, ZnCl2 and H2SO4 types of homogeneous
catalysts which are corrosive in nature and environmentally harmful
[3,4]. Such processes involve tedious work up procedure to separate
products from the reaction mass and beset with retention of deleterious
impurities in the product. Heterogeneous catalysts are preferred to
circumvent the demerits of homogeneous catalysts and thus newer cost-
effective, active and selective catalysts are needed.
Triarylmethanes (TRAM) are an important class of compounds.
Different derivatives of TRAM have wide biological activities such as
antimicrobial [5], antifungal [6] and antitubercular agents [7]. Amino
derivatives of TRAM form a class of dyes which are used as leuco dyes
for detection of tumors [8]. Various synthetic strategies have been
pursued for synthesis of TRAM network which involves reaction of
chloroform or triethylorthoformate with arenes [9], Pd catalyzed re-
action of aryl halide with (aza aryl) methanes [10] and Friedel-Crafts
hydroxylalkylation of arens with aromatic aldehydes using Lewis acid
and ZnBr2/SiO2 [16] have been reported. Molecular iodine [17] and o-
benzenedisulfonimide [18] are also used for hydroxyalkylation. All of
the foregoing processes involve use of non-reusable catalysts, harmful
and voluminous solvents, and require longer reaction times.
Recently, Wang et al. [19] reported use of BAIL (Brønsted acid ionic
liquids) to carry out hydroxyalkylation of electron rich arenes with
aromatic aldehyde. Ionic liquid acts as a solvent as well as catalyst in
these reactions and hence used in more quantity to achieve better
conversion. However, reaction of moderately activated arenes like
phenol and anisole with aldehydes requires longer time. Similarly, deep
eutectic solvents have been also reported in synthesis of TRAM deri-
vatives [20]. Hayden et al. [21] reported nafion–H catalysed con-
tinuous synthesis of TRAM using conventional as well as microwave
heating.
Keggin anion based heteropoly acids (HPA) and their salts are
proven effective catalysts to carry out a variety of organic reactions like
alkylation, acylation, esterification, oligomerisation, etherification and
many more [22]. Udayakumar et al. [23] have reported the use of
dodecatungstophosphoric acid (DTP) and DTP supported on MCM-41 in
TRAM synthesis in a solvent.. Hamidian et al. [24] used different types
of HPA in ultrasonic condition with a solvent. Substantial solubilty of
heteropoly acids in common polar organic solvents and water is a major
⁎
Corresponding author.
E-mail address: gd.yadav@ictmumbai.edu.in (G.D. Yadav).
https://doi.org/10.1016/j.mcat.2018.06.003
Received 22 March 2018; Received in revised form 15 May 2018; Accepted 3 June 2018
2468-8231/©2018ElsevierB.V.Allrightsreserved.

K.H. Bhadra, G.D. Yadav
MolecularCatalysis455(2018)150–158
drawback. Hence any reaction in which the byproduct is water or al-
cohol type moieties, the catalyst gets dissolved in the reaction mass and
hence it is difficult to separate from it. Alkali metal salts of HPA convert
homogeneous HPA into heterogeneous catalyst. Amongst group I alkali
metals, Cs salt of DTP particularly, Cs2.5H0.5PW12O40 was found better
acid catalyst than DTP [25,26]. However, the problem encountered
with Cs2.5H0.5PW12O40 (Cs-DTP) salt catalyst is separation from reac-
tion mass because of ultrafine particle size [27]. Hence a technique was
used to impregnate Cs-DTP catalyst on high surface area supports in our
laboratory. Reports have shown successful impreganation of
Cs2.5H0.5PW12O40 on mesoporous silicas like HMS [28], MCM-41 [29]
and SBA-15 [30]. Nano-particles of Cs2.5H0.5PW12O40 generated into
the porous matrix of acidic montmorillonite clay K-10 have been
proved to be versatile catalysts for several industrially important re-
actions [31–36]. Mesocellular foam (MCF) is a modified version of SBA-
15 with 3D ultracage spherical particles and higher pore volume than
SBA-15 itself. This ultracage 3D spherical particles make MCF a good
choice as a support for enzymes [37], metal oxide [38] and heteroply
acid [39,40] catalysts as well.
magnifications using SEM. Acidity of catalyst samples determined by
recording ammonia desorption thermograms on AutoChem II 2920
TPD/TPR instrument (Micromeritics, USA) by using 10% v/v NH3 in
He. Approximately 0.1 g of catalyst was loaded in the sample holder,
heated to 573 K in helium, and subsequently cooled to 298 K. After
being dosed with 10 % v/v ammonia in helium at a flow rate of 30 ml
min-1 for 30 min; the system was purged with helium for 1 h at the same
temperature. The physisorbed ammonia was removed by passing he-
lium gas at room temperature for 30 min. After cooling to 298 K, the
temperature was raised to 1023 K at 20 K/min, and the outlet gases
were analyzed by a thermal conductivity detector. TGA/DSC analysis
was carried out on NETZSCH instrument (model STA 449 F3 Jupiter).
Approximately, 5–10 mg sample quantity was used. TGA analysis was
done under flow of nitrogen gas. Nitrogen flow of 50 ml/min was
maintained during heating as well as cooling. Results were obtained
with downward exotherm mode. Since catalyst is well characterized
and discussed in our previous work [40] only a few results are discussed
in the present study.
In the present work, incipient wetness technique was adopted to
prepare 20% (wt.%) Cs2.5H0.5PW12O40 supported on MCF which was
characterized by different analytical techniques. As per our knowledge,
there is no literature on the reported catalyst to be used in solventless
hydroxyalkylation of anisole with benzaldehyde with a detailed kinetic
study.
2.3. Reaction procedure and analysis
Experiments were conducted in a thermostatic oil bath containing a
flat bottom glass reactor of 50 ml capacity provided with baffles and six
bladed turbine impeller. In a typical experiment, 0.07 mol anisole,
0.014 mol benzaldehyde and 0.6 ml n-undecane as internal standard
were charged in the reactor. Total volume of the reaction mixture was
9.6 ml Catalyst (0.048 g) was added (which corresponds to 0.005 g/ml
of liquid volume) to the reaction mass which was maintained at 130 °C.
In the case of 20 (wt.%) Cs-DTP/MCF, the weight corresponds to 20%
active catalyst. The speed of agitation was maintained at 1000 rpm. At
403 K, a zero minute sample was withdrawn immediately after addition
of catalyst and agitation started. Samples were then withdrawn peri-
odically and analyzed b gas chromatography (Chemito 8610 GC)
equipped with FID and 5% SE-30 packed column. Conversions were
calculated on the basis of limiting reactant benzaldehyde (Supporting
Information (SI), Figure S1 and S2). Confirmations of products were
done by GC–MS (Figure S3).The major product, p,p’-isomer was isolated
by column chromatography and its NMR analysis was done (Figure S4,
S5)
2. Experimental
2.1. Catalyst synthesis
The catalysts used in this study were prepared by methods devel-
oped by our laboratory.
2.1.1. Mesocellular foam silica (MCF) support and 20 (wt.%)
Cs2.5H0.5PW12O40/MCF
MCF support and 20 (wt.%) Cs2.5H0.5PW12O40/MCF (abbreviated as
Cs-DTP/MCF) were prepared by the detailed procedure mentioned in
our previous report [40].
2.1.2. Bulk Cs2.5H0.5PW12O40 (Cs-DTP)
Bulk Cs-DTP catalyst was prepared according to the procedure re-
ported in literature [41]. 0.2808 g (1.671 × 10−3 mol) of CsCl was
dissolved in 10 ml of 50% aq. methanol and added drop wise to 2 g
(6.688 × 10-4 mol) of dodecatungstophosphoric acid (DTP) in 10 ml of
methanol. With progress in addition of CsCl solution, a white pre-
cipitate of Cs2.5H0.5PW12O40 catalyst was formed. After complete ad-
dition, resulting mixture was stirred further for 1 h. Solvent was re-
moved by heating in rotary evaporator. The dried catalyst was calcined
at 300 °C for 3 h. This catalyst will be called bulk Cs-DTP henceforth.
3. Results andesults and discussion
3.1. Catalyst characterization
The catalysts were systematically characterized by different tech-
niques and are described in detail in our previous report [40]. Only a
few points related to surface area, acidity measurements and TGA/DSC
analysis are discussed here.
3.1.1. Surface area
2.2. Catalyst characterization techniques
Cs-DTP is insoluble in water and solvents making it heterogeneous
[41]. The adsorption of active Cs-DTP catalyst on surface of MCF sup-
port results in decrease in surface are and pore volume. The detailed
comparison of surface area of MCF and 20% Cs-DTP/MCF is already
discussed in our previous publication [40]. Here only nitrogen ad-
sorption-desorption isotherm results are summarized in Table 1. MCF
has large pore volume and high surface area. Decrease in the value of
surface area and pore volume of MCF after impregnating Cs-DTP in-
dicates proper distribution of active catalyst on silica surface.
Bruker D8 advances X-Ray diffractometer was used for determina-
tion of XRD patterns of prepared catalysts using Cu Kα radiation
(λ = 0.1540562 nm). Samples were step scanned from 5 to 80° in
0.045 steps with a stepping time of 0.5 s. Surface area, pore volume and
pore size distribution were done by liquid nitrogen adsorption-deso-
rption at 77 K on Micromeritics ASAP 2010. The analysis was carried
out after preheating of sample at 300 °C for 4. Infrared spectra of the
samples pressed in KBr pellets were obtained at a resolution of 2 cm−1
between 4000 and 400 cm−1. Spectra were collected with a Perkin-
Elmer instrument and in each case the sample was referenced against a
blank KBr pellet. The SEM images and elemental composition of MCF
and 20% (wt%) CS-DTP on MCF were obtained on FEI quanta 200 in-
strument operated at 20 kV with a working distance of 2–10 mm. Dried
samples were mounted on specimen studs and scanned at various
3.1.2. Acidity measurement
Acidity of catalysts was measured by ammonia-TPD method (SI,
Figure S6). MCF silica does not show any typical acidic behaviuor be-
cause of its neutral nature. Bulk Cs2.5H0.5PW12O40 showed weak acidity
peak at around 115 °C and strong acidity at higher temperature range.
20% (wt.%) Cs2.5H0.5PW12O40- MCF showed better acidity at lower as
151

K.H. Bhadra, G.D. Yadav
MolecularCatalysis455(2018)150–158
Table 1
N2 adsorption-desorption textural analysis.
Catalyst
SBET[a] (m2/g)
Pore volume[b] (cm3/
g)
Pore size[c]
(nm)
MCF
606.7
534.7
1.90
1.32
9.1
7.7
20% (w%/w%) Cs-
DTP/MCF
[a] BET surface area [b] BJH desorption pore volume [c] BJH desorption avg.
pore size.
Table 2
Ammonia-TPD results of catalysts.
Catalyst
NH3 desorption
temp. (°C)
NH3 desorbed
(mmol/g)
Total NH3 desorbed
(mmol/g)
MCF
130.4
115.2, 643.1
102.8, 535.4
0.097
0.14, 0.16
0.23, 0.25
0.097
0.3
0.48
Bulk Cs-DTP
20 (wt.%) Cs-
DTP/MCF
well as higher temperature range. The quantitative results are depicted
in Table 2. The higher acidity of supported catalyst confirms proper
distribution of acidic sites on support and is responsible for maintaining
the catalytic activity in reaction.
3.1.3. SEM/EDXS and TEM
Surface morphology of catalysts was studied by using SEM and TEM
techniques (Figure S7 in SI). Bulk Cs-DTP particles are smaller in size
which can be observed in SEM image (a1) and as black dots in TEM
image (a2). This ultrafine size of Cs-DTP particles makes it inseparable
from the reaction mass. Mesoporous MCF silica particles help in
maintaining its activity as well as proper separation from the reaction
mass and makes it a perfect heterogeneous catalyst. MCF particles are
spherical in nature (2–10 μm) as shown in SEM image (b1). TEM image
of MCF (b2) shows strut like structure with uniform sized spherical
cells. This strut like structure is a characteristic feature of MCF silica
[42]. With impregnation of Cs-DTP on MCF, the strut is slightly affected
as seen in TEM image (c2) of 20 (wt.%) Cs-DTP. The active catalyst
species is observed as black dark spots. Being ultrafine size particles, Cs-
DTP is seen as agglomerates in TEM image (c2). SEM image (c1) of 20
(wt.%) Cs-DTP on MCF does not show much distortion in the spherical
shape of MCF particles. All these observations suggest that the 3D ultra-
cage like spherical shape of MCF support after impregnation of the
active acidic catalyst is maintained.
Energy dispersive X-ray (EDAX) spectroscopy results of 20 (wt.%)
Cs-DTP-MCF catalyst for elemental compositions are shown in Table 3.
Theoretical and observed values of each element are tabulated. Each
element detected as a peak in EDX spectrum and its corresponding
weight and atomic percentages were observed (Figure S3 in SI). Since
EDX technique detects elements on surface of catalyst and hence per-
centage of oxygen (O) was found more than its expected value. Oxygen
has high content in catalyst. However, content of caesium (Cs) and
tungsten (W) was found similar to the expected one. Percentage of P
found was slightly higher than theoretical value. Difference was
Table 3
Comparison of elemental composition.
Elements
Theoretical weight % in 20 (wt.
%) Cs-DTP-MCF
Observed weight % in 20(wt.%)
Cs-DTP/MCF by EDXS
Fig. 1. (a). TGA/DSC comparison of DTP and Cs-DTP catalysts. (b). TGA
comparison of MCF and 20 (wt.%) Cs-DTP/MCF. (c). DSC comparison of MCF
and 20 (wt.%) Cs-DTP/MCF. (d). DSC comparison of bulk Cs-DTP and 20 (wt.
%) Cs-DTP/MCF.
O
Si
P
Cs
W
46.6
37.4
0.2
2.1
13.7
60.42
23.85
0.54
1.99
13.21
152

K.H. Bhadra, G.D. Yadav
MolecularCatalysis455(2018)150–158
due to low surface area of Amberlyst-15 and less available catalytic
sites. Dodecatungstophosphoric acid (DTP) gave the highest conversion
of benzaldehyde (∼87%) with ∼88% selectivity of p,p’- triarylmethane
isomer.
DTP contains 3 replaceable acidic protons which impart high cata-
lytic activity but on the contrary, DTP shows solubility in polar solvents
as well as leaches in reaction which was observed in this reaction as
well. DTP acts as a homogeneous catalyst because during the course of
reaction water is formed as co-product and DTP has solubility in water
and hence it behaves as homogeneous catalyst. DTP was not selected for
further studies because of its homogeneous nature in the reaction
system. Unsupported and supported Cs-DTP catalysts were compared
on equal loading of active catalyst used in the reaction. Bulk Cs-DTP
and 20 (wt.%) Cs-DTP/MCF gave almost similar conversions, namely,
65% and 64.5% respectively. Selectivity to p,p’- triarylmethane isomer
was also found similar in both the cases (85% with bulk Cs-DTP and
85.2% with 20 (wt.%) Cs-DTP/MCF. Since almost equal conversions
and selectivity to desired product are obtained with both the catalysts,
it would mean that all active Cs-DTP particles are well supported on
MCF and available for reaction. Cs-DTP particles are occluded in MCF
porous matrix and hence there is no intra-particle diffusion and no
leaching of the active catalyst whereas unsupported Cs-DTP catalyst has
very small particle size (SI, Fig. S2 SEM and TEM) and thus is difficult to
separate it from the reaction mixture. Hence further reaction para-
meters were optimized with 20% (wt.%) Cs-DTP/MCF catalyst.
Fig. 2. Catalyst screening for conversion of benzaldehyde and selectivity to p, p’
isomer. Reaction conditions: anisole (0.07 mol); benzaldehyde (0.014 mol), n-
undecane (0.6 ml), total reaction volume (9.6 ml), catalyst (0.005 g/ml), tem-
perature (403 K), speed of agitation (1000 rpm).
3.2.2. Effect of speed of agitation
observed in the value of O and Si.
Effect of agitation speed was studied between 800 and 1200 rpm (SI,
Figure S5) keeping other reaction parameters unchanged as in section
3.2.1. At 800 rpm, benzaldehyde conversion was 54.7% whereas at
1000 and 1200 rpm, the final conversions were found practically the
same during the course of reaction (∼64%) thereby suggesting the
mass transfer resistance was eliminated. Hence further parameter op-
timizations were carried out at 1000 rpm.
3.1.4. Thermo-gravimetric (TGA)/differential scanning calorimetry (DSC)
analysis
TGA/DSC analysis was carried out to confirm thermal stability of
prepared catalysts. Fig. 1 (a) represents TGA and DSC comparison of
DTP and Cs-DTP catalysts. In the case of DTP, three stages of weight loss
are observed. Choi et al. [43] studied thermal stability of DTP and bulk
Cs-DTP in detail. During initial heating, at less than 300 °C, physically
adsorbed water and water of crystallization is lost (first two stages of
weight loss) because of which weight loss was observed [44]. These two
weight losses in the case of DTP are endothermic processes which are
shown by upward peak in Fig. 2(a) whereas at 600 °C, DTP decomposes
in to corresponding metal oxide as per following reaction [44]. This
decomposition is an exothermic process (downward peak).
3.2.3. Effect of catalyst loading
The catalyst loading studies were carried out between 0.005 and
0.015 g/ml. At low catalyst loading of 0.005 g/ml, the final conversion
was ∼65% in 180 min. Further increase in catalyst loading to 0.01 g/
mL, conversion of benzaldehyde reached to 82% and 100% conversion
was achieved with catalyst loading of 0.015 g/ml in 180 min (Fig. 3).
From these results, it can be concluded that increased catalyst loading
helps in providing more acidic sites and increase the rate of reaction.
To understand the exact effect of catalyst loading, initial reaction
rates were plotted against catalyst loading (Fig. 4). Initial reaction rate
increased linearly with catalyst loading which proves that the increase
in catalyst loading increases the number of catalytic active sites pro-
portionately which in turn increases the rate of reaction. Also there is
no mass transfer resistance and intra-particle diffusion resistance.
Theoretical calculations for external mass of transfer resistance and
intra-particle diffusion resistance (Weisz-Prater modulus < < 1) were
also performed to validate these results according our earlier work
[45,46].
1
2
3
2
H3PW O40
→
P2O5 + 12WO3 + H2O
12
Bulk Cs-DTP shows only one endothermic peak near 373 K which
corresponds to removal of physisorbed water molecule. Later on,
weight decrease is almost nearly constant which proves that the addi-
tion of Cs imparts thermal stability to Keggin structure of DTP. Fig. 1(b)
and (c) represents TGA and DSC comparison of MCF and 20% (wt.%)
Cs-DTP/MCF. Both TGA and DSC trends of MCF support and Cs-DTP
supported on MCF are comparable becasue there are no major changes
observed. Furthermore, the DSC trends of bulk Cs-DTP and supported
Cs-DTP are also comparable (Fig. 1(d)). These comparisons prove that
the thermal stability of bulk Cs-DTP is maintained after supporting it on
MCF.
3.2.4. Effect of mole ratio
Mole ratio studies were carried out between 3:1 and 5:1 (anisole:-
benzaldehyde). Catalyst loading were kept constant (i.e. 0.015 g/mL) in
all experiments. With low mole ratio of 3:1, conversion was ∼84% in
180 min (Fig. 5). With increase in mol ratio to 4:1 and 5:1, the con-
version was found 100% in both the cases. The selectivity of products
was found nearly the same in all cases (∼84% for p,p’- isomer). Hence
for temperature studies, 4:1 mol ratio was selected.
3.2. Optimization of reaction parameters
3.2.1. Catalyst screening
Hydroxyalkylation of anisole with benzaldehyde was carried out in
solventless condition (Scheme 1). Different catalysts were screened for
hydroxyalkylation (Fig. 2). Commercial catalysts Amberlyst-15 and
montmorillonite clay- K10 gave 24% and 8% conversion of benzalde-
hyde respectively. Althogh Amberlyst-15 is highly active Brønsted
acidic catalyst still the conversion was found very low which may be
3.2.5. Effect of temperature
Effect of temperature is very important to provide further proof of
intrinsic kinetic rate equation. Hence under all optimized reaction
153

K.H. Bhadra, G.D. Yadav
MolecularCatalysis455(2018)150–158
Scheme 1. Hydroxyalkylation of anisole with benzaldehyde.
Fig. 5. Effect of mole ratio on conversion of benzaldehyde. Reaction conditions:
anisole: benzaldehyde (mole ratio 3:1, 4:1 and 5:1), n-undecane (0.6 mL),
catalyst loading (0.015 g/mL), temperature (403 K), speed of agitation
(1000 rpm).
Fig. 3. Effect of catalyst loading on conversion of benzaldehyde. Reaction
conditions: anisole (0.07 mol); benzaldehyde (0.014 mol), n-undecane (0.6 ml),
total reaction volume (9.6 ml), temperature (403 K), speed of agitation
(1000 rpm).
parameters, reaction was tested at different temperatures. At 373 K, the
final conversion was 78% (Fig. 6). With further 10 K rise in tempera-
ture, final conversion reached to ∼97%. Complete conversion of ben-
zaldehyde was observed in case of 393 K and 403 K. Selectivity of
products was almost nearly same in all cases (∼ 84% for p,p’ isomer).
Hence optimum temperature for the said reaction was selected as
393 K.
3.2.6. Catalyst stability and reusability study
To confirm heterogeneous nature of catalyst, a leaching test was
carried out. Reaction was carried out at 4:1 mol ratio of anisole: ben-
zaldehyde, catalyst loading 0.015 g/ml and speed of agitation 1000 rpm
and temperature 403 K. The experiment was conducted first 30 min and
a sample withdrawn for analysis. The conversion was found as 34.2%.
Thereafter the reactor was removed from oil bath and the reaction mass
filtered to remove catalyst. It was placed again in the bath. The liquid
reaction mass without the catalyst was further stirred at 403 K at
1000 rpm, and sampling was done at 60, 120 and 180 min and ana-
lysed. There was no major increase in the conversion even after 180 min
Fig. 4. Initial reaction rate vs. catalyst loading.
154

K.H. Bhadra, G.D. Yadav
MolecularCatalysis455(2018)150–158
Fig. 8. Reusability of catalyst. Reaction conditions: anisole (0.07 mol), ben-
zaldehyde (0.0175 mol), n-undecane (0.6 ml), total reaction volume (9.9 ml),
catalyst loading (0.015 g/ml), temperature (403 K), speed of agitation
(1000 rpm).
Fig. 6. Effect of temperature on conversion of benzaldehyde. Reaction condi-
tions: anisole (0.07 mol), benzaldehyde (0.0175 mol), n-undecane (0.6 ml),
total reaction volume (9.9 ml), catalyst loading (0.015 g/mL), speed of agitation
(1000 rpm).
Fig. 9. FTIR comparison of fresh 20 (wt.%) Cs-DTP/MCF and reused catalyst
after reaction.
Fig. 7. Catalyst stability study.
(35.9%). Thus, within experimental results it is concluded that the
catalyst is stable and heterogeneous (Fig. 7).
Table 4
Comparison of fresh and reused catalyst.
The robustness of catalyst in terms of activity and selectivity was
studied by the reusability test. The reaction was carried out under op-
timal conditions of mole ratio of anisole: benzaldehyde, (4:1), speed of
agitation (1000 rpm), catalyst loading (0.015 g/ml) and temperature
(393 K) for 180 min. After the reaction was over, the catalyst was fil-
tered and separated from reaction mass. The separated catalyst was
taken in cyclohexane and refluxed for 3 h to remove any adsorbed or
trapped reactants and products from its surface and pore space. It was
filtered, dried and weighed. There was some loss during filtration (∼4-
5%) due to attrition of catalyst which was made up in the next ex-
periment with fresh one keeping the same reaction conditions. There
was a marginal loss in activity (Fig. 8) thereby proving heterogeneous
nature of catalyst and its reusability.
The third reused catalyst was characterised and compared with the
freshly prepared. FTIR comparison (Fig. 9) showed no specific change
in IR bands in finger print region (400-1400 cm−1).
However, a slight variation was found in surface area and acidity of
reused catalyst which could be due to some occlusion of product in
some junctions of porous network which could be removed by solvent
at reflux. Comparison of fresh and reused catalyst is shown in Table 4.
Catalyst
Surface area (m2/g)
Total acidity (mmol/g)
Fresh Catalyst
After 3rd reuse
534.7
518.4
0.48
0.37
3.2.7. Substrate study as extension of catalyst application
Different arenes as well as aldehydes were tested under the similar
reaction conditions in order to understand the effect of each of them.
Table 5 shows effect on conversion and selectivity based on different
substrates.
During substrate study, phenol (entry 1) being activated arene gave
100% conversion when reacted with benzaldehyde but with addition of
methyl group at para position it reduced the conversion to 89% (entry
2). This observation indicates that presence of methyl group does not
activate an arene up to a considerable extent. The difference in re-
activity of phenol and p-cresol was also observed by Wang et al [19]
while using ionic liquid as a catalyst. Veratrole (entry 3) with two
methoxy groups at ortho position to each other gave 100% conversion
which supports the activating effect of methoxy group. When anisole
155

K.H. Bhadra, G.D. Yadav
MolecularCatalysis455(2018)150–158
Table 5
Extension of catalyst 20 (wt.%) CsDTP-MCF to other arenes and aldehydes.
Entry
1
Arenes
Aldehydes
Products (% GC Selectivity)
% (GC) Conversion
100
2
3
89
100
4
30
5
52
Reaction conditions: Arene: aldehyde, 4:1, catalyst 0.015 g/mL, temperature 403 K, n-undecane 0.6 ml, time 3 h. Conversion and selectivity calculated by GC
results.
reacted with different aldehydes, namely, salicylaldehyde (entry 4) and
p- methoxy benzaldehyde (entry 5) low conversions were obtained,
namely, 30% and 52%. These observations conclude that aromatic al-
dehydes with electron donating groups reduce conversion. Similar re-
sults are obtained by Wang et al. [19] while using BAIL (Brønsted acid
ionic liquids).
both the species occurs resulting in formation of p,p‘-isomer (C) and
o,o‘- isomer (D). The complete derivation of the model is given in
supplementary information (Eqs. (25) and (26)).
When CA > > CB or M > > 1, the final equation is as follows:
0
0
−dCB
= k1CB w
(1)
dt
In terms of fractional conversion, it becomes a pseudo-first order
4. Reaction mechanism and kinetic model
equation to get the following integrated form:
4.1. Reaction mechanism
− ln(1 − XB) = k1wt
(2)
Hence a plot of ln(1−XB) vs. time t would be a straight line with
slope equal to k1w from which the rate constant k1 could be evaluated.
Moreover, the linearity of this plot would confirm the pseudo first order
behavior of the reaction.
A probable reaction mechanism is shown in Scheme 2, where the
first step is adsorption of reactants on the catalyst surface followed by
surface reaction in the second step. Finally products are desorbed from
the surface of catalyst in the third step and the cycle continues.
It was found that equation fitted that data very well. Fig. 10 shows
the validity of this theory at different temperatures.
From the slopes of lines in Fig. 10, the rate constants, k1 at different
temperatures were calculated.
4.2. Kinetic modeling
A variety of models have been discussed [45,46] for solid-liquid
catalytic reaction including adsorption, surface reaction and deso-
rption. As explained earlier, in the absence of both external mass
transfer and intra-particle resistances, the Langmuir-Hinshelwood-
Hougen-Watson (LHHW) model was tested. According to LHHW me-
chanism, adsorption of A (anisole) and B (benzaldehyde) takes place on
two adjacent vacant sites of surfaces (S) and surface reaction between
k1 (100 °C) = 0.01s−1
k1 (110 °C) = 0.0144 s−1
k1 (120 °C) = 0.01666 s−1
k1 (130 °C) = 0.0355 s−1
156

K.H. Bhadra, G.D. Yadav
MolecularCatalysis455(2018)150–158
Scheme 2. Mechanism of hydroxyalkylation of anisole.
From, Arrhenius plot (Fig. 11), apparent activation energy was
calculated and found to be 11.74 kcal/mol.
In the absence of both external mass transfer resistance and intra-
particle diffusion limitation, and adsorption and desorption resistance,
intrinsic kinetics should control and hence effect of temperature will be
pronounced on initial rate of reaction which is confirmed by activation
energy values (> 8 kcal/mol) [28,45–47]. In the current work, Ar-
rhenius plot was made to get the apparent energy of activation as 11.74
kcal/mol to prove that the reaction was intrinsically kinetically con-
trolled.
5. Conclusion
Solventless hydroxyalkylation of anisole was studied in detail with
several catalysts including commercial ones. 20 (wt.%) Cs-DTP sup-
ported on MCF silica was the best catalyst. The MCF supported Cs-DTP
showed better catalytic activity than commercial catalysts like
Amberlyst-15 and montmorillonite clay K-10 and equivalent activity to
the bulk Cs-DTP catalyst with the advantage of easy recovery from the
reaction mass. The catalyst is thermally stable and reusable. The op-
timum temperature was 120 °C and 100% conversion was obtained
Fig. 10. Pseudo first order kinetic plot at different temperatures.
′
with 84% selectivity to the p, p - triarylmethaane isomer. The LHHW
mechanism was used to determine the mechanism and kinetics of the
reaction. All species were found to be weakly adsorbed thereby leading
to a pseudo-first rate equation. Overall reaction is intrinsically kineti-
cally controlled and follows second order rate equation. The apparent
activation energy was found to 11.74 kcal/mol. The process is green.
Conflict of interest statement
The authors declare no conflict of interest.
Acknowledgements
KHB is grateful to UGC New Delhi for BSR fellowship. GDY ac-
knowledges support from the R.T. Mody Distinguished Professor
Fig. 11. Arrhenius plot.
157

K.H. Bhadra, G.D. Yadav
MolecularCatalysis455(2018)150–158
endowment and J.C. Bose National Fellowship of DST- GOI.
59022–59026.
[21] S. Hayden, T. Glasnov, C.O. Kappe, Chem. Eng. Technol. 38 (2015) 1743–1748.
[22] G.D. Yadav, Catal. Surveys Asia 9 (2) (2005) 117–137.
[23] S. Udayakumar, S. Ajaikumar, A. Pandurangan, Catal. Comm. 8 (2007) 366–374.
[24] H. Hamidian, Org. Chem. Inter. 2013 (2013) 1–5.
Appendix A. Supplementary data
[25] Y. Izumi, M. Ogawa, K. Urabe, Appl. Catal. A Gen. 132 (1995) 127–140.
[26] J.A. Dias, E. Caliman, S.C.L. Dias, Microporous Mesoporous Mater. 76 (2004)
221–232.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2018.06.003.
[27] Y. Izumi, M. Ono, M. Kitagawa, M. Yoshida, K. Urabe, Microporous Mater. 5 (1995)
255–262.
References
[28] G.D. Yadav, G. George, Catal. Today 141 (2009) 130–137.
[29] K.M. Parida, S. Rana, S. Mallick, D. Rath, J. Coll. Inter. Sci. 350 (2010) 132–139.
[30] P.M. Rao, M.V. Landau, A. Wolfson, A.M. Shapira-Tchelet, M. Herskowitz,
Microporous Mesopoous. Mater. 80 (2005) 43–55.
[31] G.D. Yadav, S.B. Kamble, Ind. Eng. Chem. Res. 48 (2009) 9383–9393.
[32] G.D. Yadav, S.S. Salgaonkar, N.S. Asthana, Appl. Catal. A: Gen. 265 (2004)
153–159.
[1] Ullman, Encyclopedia of Industrial Chemistry, 6th edn., Wiley-VCH Verlag GmbH,
Weinheim, 2002, p. A127.
[2] V.P. Reddy, G.K.S. Prakash, Kirk-Othmer Encyclopedia of Chemical Technology,
Wiley-Blackwell, Hoboken, NJ, 2013, pp. 1–49.
[3] G.A. Olah, Friedel–Crafts Chemistry, Wiley–Interscience, New York, 1973.
[4] G.A. Olah, R. Krishnamuri, G.K. Suryaprakash, Comprehensive Organic Synthesis,
Pergamon, Oxford, 1991, p. 3 chap 1.8.
[33] G.D. Yadav, N.S. Asthana, S.S. Salgaonkar, Clean. Tech. Environ. Policy 6 (2004)
105–113.
[5] D.F. Duxbury, Chem. Rev. 93 (1993) 381–433.
[34] G.D. Yadav, N.S. Asthana, V.S. Kamble, J. Catal. 217 (2003) 88–99.
[35] G.D. Yadav, N.S. Asthana, Appl. Catal. A: Gen. 244 (2003) 341–357.
[36] G.D. Yadav, N.S. Asthana, V.S. Kamble, Appl. Catal. A: Gen. 240 (2003) 53–69.
[37] M. Sridhar, G.K. Reddy, N. Hu, A. Motahari, Dale W. Schaefer, S. Thiel,
P. Smirniotis, Microporous Mesoporous Mater. 190 (2014) 215–226.
[38] Y. Su, Y.M. Liu, L. Wang, M. Chen, Y. Cao, W.L. Dai, H.Y. He, K.N. Fan, Appl. Catal.
A: Gen. 315 (2006) 91–100.
[6] S.J. Culp, F.A. Beland, J. Am. Coll. Toxicol. 15 (1996) 219–238.
[7] M.K. Parai, G. Panda, V. Chaturvedi, Y.K. Manju, S. Sinha, Bioorg. Med. Chem. Lett.
18 (2008) 289–292.
[8] I.K. Kandela, J.A. Bartlett, G.L. Indig, Photochem. Photobiol. Sci. 1 (2002) 309–314.
[9] G. Casiraghi, G. Casnati, M. Cornia, Tet. Lett. 14 (1973) 679–682.
[10] T. Niwa, H. Yorimitsu, K. Oshima, Org. Lett. 9 (2007) 2373–2375.
[11] G.K. Surya Prakash, C. Panja, A. Shakhmin, E. Shah, T. Mathew, G.A. Olah, J. Org.
Chem. 74 (2009) 8659–8668.
[39] S. Park, J.H. Choi, T. Kim, Y.M. Chung, S.H. Oh, I.K. Song, Catal. Today. (185)
(2012) 162–167.
[12] X. Wang, Y. Wang, D.M. Du, J. Xu, J. Mol. Catal. A: Chem. 255 (2006) 31–35.
[13] V. Nair, K.G. Abhilash, N. Vidya, Org. Lett. 7 (2005) 5857–5859.
[14] S. Podder, J. Choudhury, S. Roy, U.K. Roy, J. Org. Chem. 72 (2007) 3100–3103.
[15] Z. Li, Z. Duan, J. Kang, H. Wang, L. Yu, Y. Wu, Tetrahedron 64 (2008) 1924–1930.
[16] M. Kodomari, M. Nagamatsu, M. Akaike, T. Aoyama, Tet. Lett. 49 (2008)
2537–2540.
[40] K.H. Bhadra, G.D. Yadav, Microporous Mesoporous Mater. 263 (2018) 190–200.
[41] T. Okuhara, T. Nishimura, H. Watanabe, M. Misono, J. Mol. Catal. 74 (1992)
247–256.
[42] P. Schmidt-Winkel, W.W. Lukens Jr., D. Zhao, P. Yang, B.F. Chmelka, G.D. Stucky,
J. Am. Chem. Soc. 121 (1999) 254–255.
[43] S. Choi, Y. Wang, Z. Nie, J. Liu, C.H.F. Peden, Cata. Today 55 (2000) 117–124.
[44] M. Misono, Catal. Rev. Sci. Eng. 29 (1987) 269–275.
[45] P.S. Kumbhar, G.D. Yadav, Chem. Eng. Sci. 44 (1989) 2535–2544.
[46] G.D. Yadav, M.S. Krishnan, Ind. Eng. Chem. Res. 37 (1998) 3358–3365.
[47] H.S. Fogler, Elements of Chemical Reaction Engineering, 4th ed., Prentice-Hall,
New Delhi, 2005.
[17] J. Jaratjaroonphong, S. Sathalalai, P. Techasauvapak, V. Reutrakul, Tet. Lett. 50
(2009) 6012–6015.
[18] M. Barbero, S. Cadamuro, S. Dugghera, C. Magistris, P. Venturello, Org. Biomol.
Chem. 9 (2011) 8393–8399.
[19] A. Wang, X. Zheng, Z. Zhao, C. Li, Y. Cui, X. Zheng, J. Yin, G. Yang, Appl. Catal. A:
Gen. 482 (2014) 198–204.
[20] A. Wang, P. Xing, X. Zheng, H. Cao, G. Yang, X. Zheng, RSC Adv. 5 (2015)
158