TTAB mediated synthesis of Meso-H-BEA and its application in the production of n-butyl levulinate

Article abstract of DOI:10.1016/j.cattod.2019.10.009

The present work is focusing on the improvement of catalytic characteristics of zeolite. In the present endeavor, zeolite H-BEA has been modified using sol-gel route by making use of tetradecyltrimethyl ammonium bromide (TTAB) surfactant as a structure directing agent. The synthesized mesozeolite was characterized by SEM, HRTEM, ²⁹Si and ²⁷A1 MAS-NMR, small and wide-angle XRD, FT-IR, TGA, NH₃-TPD, ICP-OES and BET surface area analysis. The synthesized material was found to exhibit mesoporosity along with microporosity of parent H-BEA. The utility of modified H-BEA as heterogeneous catalyst was demonstrated in the production of value added chemical, n-butyl levulinate from biomass derived feedstock, levulinic acid (LA) via esterification reaction. n-butyl levulinate is one of the significant industrial chemical and has wide use as plasticizing agent, solvent and odorous substance. Various process variables such as molar ratio (LA to n–butanol), catalyst concentration (%) and reaction time (h) were optimized by response surface methodology (RSM), using the Box–Behnken Design (BBD). Analysis of variance (ANOVA) has been performed to investigate the suitability and significance of the quadratic model.

Full text of DOI:10.1016/j.cattod.2019.10.009

Journal Pre-proof
TTAB mediated synthesis of Meso-H-BEA and its application in the
production of n-butyl levulinate
Dhara H. Morawala, Dharmesh R. Lathiya, Ajay K. Dalai, Kalpana
C. Maheria
PII:
S0920-5861(19)30563-2
DOI:
https://doi.org/10.1016/j.cattod.2019.10.009
CATTOD 12514
Reference:
To appear in:
Catalysis Today
Received Date:
Revised Date:
Accepted Date:
5 June 2019
30 August 2019
8 October 2019
Please cite this article as: Morawala DH, Lathiya DR, Dalai AK, Maheria KC, TTAB mediated
synthesis of Meso-H-BEA and its application in the production of n-butyl levulinate, Catalysis
Today (2019), doi: https://doi.org/10.1016/j.cattod.2019.10.009
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
©
2019 Published by Elsevier.

TTAB mediated synthesis of Meso-H-BEA and its application in the
production of n-butyl levulinate
a
a
b
a
Dhara H. Morawala , Dharmesh R. Lathiya , Ajay K. Dalai , Kalpana C. Maheria *
aApplied Chemistry Department, Sardar Vallabhbhai National Institute of Technology, Surat
-
395007, Gujarat, India.
Department of Chemical and Biological Engineering, University of Saskatchewan,
b
S7N 5A9, SK, Saskatchewan, Canada.
*
Corresponding author’s E-mail:kcmaheria@gmail.com
Graphical abstract
Silica from desilication
Zeolite
Surfactant
NaOH
Surfactant
Desilication
Self organization
O
O
Template
removal
O
O
OH
+
HO
O
Mesozeolites with introduction of zeolitic
Mesozeolite
particles on or inside mesopore walls
MTCK catalysed LA esterification
Highlights


Synthesis of micromeso zeolite H-BEA using desilication post synthetic approach
Use of tetradecyltrimethyl ammonium bromide (TTAB) surfactant for meso-zeolite BEA
synthesis

An assessment of catalytic activity of MTCK towards synthesis of n-butyl levulinate via LA
esterification
1
| P a g e


Optimization of reaction parameters using BBD
Abstract
The present work is focusing on the improvement of catalytic characteristics of zeolite. In the
present endeavor, zeolite H-BEA has been modified using sol-gel route by making use of
tetradecyltrimethyl ammonium bromide (TTAB) surfactant as a structure directing agent. The
2
9
27
synthesized mesozeolite was characterized by SEM, HRTEM, Si and A1 MAS-NMR,
small and wide-angle XRD, FT-IR, TGA, NH -TPD, ICP-OES and BET surface area
3
analysis. The synthesized materials were found to exhibit mesoporosity along with
microporosity of parent H-BEA. The utility of modified H-BEA as heterogeneous catalyst
was demonstrated in the production of value added chemical, n-butyl levulinate from biomass
derived feedstock, levulinic acid (LA) via esterification reaction. n-butyl levulinate is one of
the significant industrial chemical and has wide use as plasticizing agent, solvent and odorous
substance. Various process variables such as molar ratio (LA to n–butanol), catalyst
concentration (%) and reaction time (h) were optimized by response surface methodology
(RSM), using the Box–Behnken Design (BBD). Analysis of variance (ANOVA) has been
performed to investigate the suitability and significance of the quadratic model.
Keywords: Zeolite H-BEA, TTAB, Sol-Gel method, Esterification, Box-Behnken Design,
Response surface methodology
1
. Introduction
Porous materials are one of the prominent advanced materials and are well known for their
characteristics uses for chemical processes and house hold purposes and advanced uses in
medicines, optics and electronics. These materials are classified as: (i) microporous, with
2
| P a g e

pores less than 2 nm; (ii) mesoporous, with pores from 2 to 50 nm; and (iii) macroporous,
with pores between 50 and 1000 nm [1,2]. Amongst several microporous materials, zeolites
are gaining increasing attention due to their wide use as solid acid catalysts in varied
catalysed organic transformations, gas storage materials and sorbents / ion-exchangers in
water softening. Responsible salient features for such wide use of zeolites include, their
composition, tuneable acidity and well-defined uniform pore structure of their cavities and
channel system [3]. However, the processing of heavy feedstock and bulkier molecules over
zeolites is restricted due to their pore diffusion limitations which is imposed by complex
nature of their pores and cavities. In order to overcome these limitations, there is increasing
research interest towards development of synthesis strategies for the preparation of micro-
meso zeolites (mesozeolites) having inter-connected pores. These materials allow facile pore
access and fast transport of bulky molecules through such interconnecting pores, which
ultimately leads to applicability of these materials to wide range of catalysed organic
transformations involving bulkier and complex entities such as, esterification, condensation,
aromatization, isomerization, alkylation, cracking, pyrolysis, hydro-desulfurization etc. These
are the key reactions used to synthesize relatively bulky and complex molecules for the
petrochemical, pharmaceutical, fragrance, and agrochemical industries [3,4]. Various top-
down and bottom-up approaches are exists to synthesize such micromeso materials or
hierarchical materials. Amongst several post synthetic routes, desilication method using
alkaline aqueous solution is one of the prominent method to prepare mesozeolites [5]. Groen
et al. (2008) [6] reported about mesoporous beta zeolite and demonstrated its catalytic
activity in benzene alkylation reaction. Patil et al. (2014) [5] have demonstrated applicability
of bimodal H-BEA zeolite for the synthesis of ethyl levulinate. Wang et al. (2016) [7] have
reported selective catalytic cracking of n-hexane to produce propylene over desilicated and
3
| P a g e

dealuminated Beta zeolite. Sonar et al. (2017) [8] have synthesized hierarchical BEA zeolite
and reported its utility for acetalization reaction of glycerol and formaldehyde.
Recently, processing of renewable raw materials derived from ligno-cellulosic biomass into
value added chemicals and fuels have gained much attention amongst researchers and
scientists [9]. The key challenge is to develop green route for conversion of renewable
resources (agricultural waste biomass, forest residues, etc.) in to value added chemicals such
as glucose, xylose, alkyl levulinates etc [10]. Amongst these, levulinic esters are extensively
used for varied applications as fuel additives, reaction intermediates and odouring substances.
n-butyl levulinate is one such levulinic ester. As per the literature survey, there are few
reports available on the use of biomass derived LA for the synthesis of n-butyl levulinate
[
11]. Dharane et al. (2010) [12] have studied the catalytic performance of heteropolyacid
supported on acid-treated clay in esterification of renewable LA with n-butanol. Maheria et
al.(2012) [11] have studied esterification of LA to synthesize n-butyl levulinate over H-BEA
zeolite as a heterogeneous catalyst.
In this work, an attempt has been made to prepare micro-mesoporous zeolite BEA with
bimodal porosity via sol-gel route from alkali treated zeolite BEA precursor using
tetradecyltrimethyl ammonium bromide (TTAB) as structure directing agent. So far, not
much attention has been paid to the synthesis of micro – meso zeolite, an advanced catalytic
material using desilication synthetic approach along with use of TTAB. This synthesized
material was characterized by various characterization techniques such as SEM, HRTEM,
2
9
27
Si and A1 MAS-NMR, small and wide angle XRD, FT-IR, TGA, NH -TPD, ICP-OES
3
and BET analysis. The utility of micro-meso H-BEA as a heterogeneous catalyst was
demonstrated for the production of n-butyl levulinate via LA esterification with n-butanol.
The effects of the reaction conditions such as, reactant molar ratio, catalyst loading and
reaction time on LA conversion have been investigated using Box-Behnken method (BBD).
4
| P a g e

BBD is a part of response surface methodology (RSM), which is a combination of
mathematical and statistical methods. RSM has been successfully applied for optimization of
reaction parameters for effective catalytic synthesis of biodiesel and esters [10,13]. Lathiya et
al.(2018) [13] has reported synthesis of sulfonated carbon catalyst from waste orange peel for
cost effective biodiesel production via BBD method. In the present study also, RSM is used
for the optimization of process parameters for esterification of LA with n-butanol over
modified H-BEA catalyst with three important variables such as, LA to n-butanol molar ratio
(X
1
), catalyst concentration (%) (X
2
) and reaction time (h) (X ). The effects of these variables
3
on the conversion of LA was studied using BBD. The study is also emphasized on the
identification of crucial process parameter which is responsible for maximizing conversion of
LA. The study was also extended to validation of theoretical and experimental data.
2
. Experimental section
2
.1 Materials
Zeolite H-BEA (Si/Al ratio: 18) was obtained from Sud - Chemie India Pvt. Ltd., Vadodara,
Gujarat, India. Ethanol (99%) and n-butanol (99%) were procured from Merck, Mumbai,
India. LA (99%) was procured from TCI Chemicals (India) Pvt. Ltd., Tamilnadu. All
chemicals were used without further purification and of analytical grade.
2
.2 Post modification treatment of H-BEA using TTAB
This material is synthesized as per earlier reported methods with slight modification [14]a
b
[
14] . In a typical procedure, 1.5 gm of H-BEA zeolite (Si/Al ratio: 18) was stirred with an
o
aqueous solution of 0.1M NaOH for 4 h at 80 C and then filtered. In the filtrate, 0.5 gm
amount of TTAB was added with continuous stirring and then the pH of the solution was
adjusted to 12.3 - 12.5. The resultant gel was kept for ageing for 24 h, at room temperature
5
| P a g e

o
o
and then filtered, dried and calcined at 550 C (heating rate of 1 C/min) for 5.5 h; in a muffle
furnace. The obtained micro-meso material was converted in to proton form by treating with
a
3
0% ammonia solution in reflux condition as per reported method [14] . The final material is
designated as sample MTCK.
2
.3 Catalyst characterization and instrumentation
The catalyst was characterized by different techniques such as scanning electron microscopy
2
9
27
(
SEM), high resolution transmission electron microscopy (HRTEM), Si and A1 Magic
2
9
27
Angle Spinning - Nuclear Magnetic Resonance ( Si and A1 MAS-NMR), X-ray diffraction
analysis (XRD), Fourier transform infra-red spectroscopy (FT-IR), thermogravimetric
analysis (TGA),ammonia temperature program desorption (NH -TPD), inductively coupled
3
plasma- optical emission spectrometry (ICP-OES) and Brunauer–Emmett–Teller (BET)
surface area analysis.
SEM images were obtained using JEOL microscope (Model JSM 7100F, USA). HRTEM
2
9
images were taken using JEOL microscope (Model JEM 2100, USA).Solid state Si and
27A1 MAS-NMR measurements were performed using a Brucker make spectrometer (Model
2
9
Avance II (500MHz), Germany). Si MAS NMR measurements were performed at a
resonance frequency of 500 MHz. Spectra were recorded using a 5 mm probe head at a
2
7
spinning rate of 8 KHz and a delay time of 24 s. A1 MAS NMR spectra were recorded at a
resonance frequency of 500 MHz and at a spinning rate of 8 KHz, using a 5 mm probe head.
The delay time was 1 s. The XRD studies were performed on Rigaku Miniflex, instrument by
using Cu-K
α
radiation (K=1.5418Å) and nickel filter. FT-IR spectra were recorded on
Shimadzu spectrometer (Model FT-IR 8400S, Japan) using KBr pellet between the spectral
-1
range 4000 - 400 cm . TGA curves of the catalysts were obtained using a TA instruments
thermal analyzer (Model SDTQ 600, USA) by heating the samples in the range, 50–600 °C,
6
| P a g e

−
1
at heating rate of 10 °C min , under N
2
atmosphere. NH
3
-TPD profile was recorded on
-TPD profile.
BELCAT II instrument, Japan. Total acidity was calculated from the NH
3
Acidity of the samples was determined from the ammonia adsorbed on the sample, and is
expressed in term of mili-moles of ammonia per g of sample. In a typical experiment for the
determination of acidity, 200 mg of each sample is loaded in sample tube, and placed in a
o
vertical furnace, and then heated in a flow of He to 600 C, held for an h and cooled down to
o
1
00 C. Then, the sample is saturated. Then, the sample is saturated with 1% NH
3
/N (v/v)
2
mixture at a flow rate of 30 mL/min for 120 min. Then the spectra is recorded between 100
C to 750° C, with a temperature ramp of 10° C/min. The obtained spectra is curve fitted
using ASiQwin software and the QC fit analysis is generated, which provides the amount of
NH desorbed per g of sample in m.moles/g. The amount of ammonia desorbed (m.moles/g)
°
3
gives the concentration of the active site. Si/Al ratio of samples was determined by ICP-OES
analysis using Perkin Elmer spectrometer (Model Optima 3300 RL, USA). The BET surface
area and pore size of catalysts were measured using Micromeritics instrument (Model ASAP
2
020, USA) by multipoint N
2
adsorption-desorption technique using liquid N at -196°C.
2
2
.4 Procedure for the Esterification of LA with n-Butanol
In a typical procedure, the reactants are taken in a 50 ml Round Bottom Flask (RBF) (Scheme
). The RBF is attached to a condenser and then the whole setup is kept in an oil bath under
1
atmospheric pressure while maintaining the temperature at 120°C (having ± 0.5°C of
accuracy) (equipped with magnetic stirrer (400 rpm)). Molar ratio amount of LA and n-
butanol were calculated based on the desired reaction conditions. The reaction was monitored
by thin layer chromatography (TLC). The products were collected after separation of the
catalyst. After the completion of the reaction, the spent catalysts were separated and then
washed with ethanol. To obtain the desired product, the filtrate was distilled. The percentage
7
| P a g e

conversion was calculated by titration method using 0.1 M alcoholic KOH solution [11,15].
The % conversion of LA was determined using following equation (Eq. 1). The product was
analysed with gas chromatography to measure product % yield and % selectivity (Eq. 2)
(Thermofisher scientific Inc., Model- Trace-1110) with an FID detector and a TG-5
(Thermofisher scientific Inc.) column (30.0 m × 0.25 mm × 0.25 µm). The injection volume
was 3 µL and N
2
was used as a carrier gas at a flow rate of 1 ml/min. The injector and
detector temperature was at 250 °C and 280 °C respectively.
%
Conversion = [(A-B)/A] x 100……(1)
;
Where, A = Acid value of the sample before reaction,
B = Acid value of sample withdrawn after reaction
푴풐풍풆풔 풐풇 풏−풃풖풕풚풍 풍풆풗풖풍풊풏풂풕풆 풇풐풓풎풆풅
푴풐풍풆풔 풐풇 푳푨 풄풐풏풗풆풓풕풆풅
%
Selectivity =
X 100……(2)
O
O
O
MTCK
+
HO
OH
-
H O
2
O
O
n-Butyl alcohol
n-Butyl levulinate
Levulinic acid
Scheme 1. Esterification reaction of LA and n-butanol
3
. Result and discussion
3
.1 SEM
The SEM images of surface morphology of parent H-BEA (A, B, C) and MTCK (D, E, F)
shown in Fig. 1. The parent H-BEA particles were found of uniform size. In the SEM image
of MTCK, the particles are appeared as agglomeration with flat surface due to mesophase
formation. At higher magnification, the particles of MTCK appeared to be assemblies of very
small Beta zeolite crystals.
8
| P a g e

C
A
B
D
E
F
Fig.1 SEM of parent H-BEA (A, B, C) and MTCK (D, E, F)
.2 HRTEM
3
The high resolution TEM images of parent H-BEA and MTCK are shown in Fig. 2 (A, B, C)
and in Fig. 2 (D, E, F), respectively, at different magnifications. Both the samples showed
characteristic stripped channels corresponding to microporous zeolite H-BEA having 12
membered ring structure. However, in case of HRTEM image of MTCK, interconnected
pores appeared (formed due to alkali treatment) (Fig. 2E) and there is also discontinuity in
original microporous structure. Thus, in the present case, using desilication post modification
route with micelle forming cationic surfactant, mesoporous zeolite is formed, and this
resultant material feature preserved crystallinity i.e. no distortion of crystalline lattice fringes
is observed (Fig. 2D). The similar observation also made elsewhere [16] for the similar kind
of material.
9
| P a g e

B
C
A
0.30 nm
D
E
F
0.33 nm
0.31 nm
Fig.2 HR-TEM of parent H-BEA (A, B, C) and MTCK (D, E, F)
2
9
3
.3 Si MAS NMR
2
9
The Si MAS NMR spectrum of H-BEA and MTCK (Fig. 3) shows signal at -110 ppm,
4
which is attributed to the presence of Si in Si[4Si] (Q ) environment. The signal observed at
ca.-102 ppm is assigned to Si(OH)(OSi) species at defect sites or it may be attributed to
3
3
29
Si[3Si,1Al] (Q ) sites [17,18]. In Si MAS NMR spectrum, widening of the band at -102
ppm is observed which indicates higher exposure of surface SiOH groups in the desilicated
zeolite.
2
7
3
.4 Al MAS NMR
2
7
The Al MAS NMR spectrum was recorded to know the distribution of octahedral
coordinated aluminum atoms [as extra-framework aluminium (EFAL)] and tetrahedral
coordinated Al atoms [as framework sites] in the material upon desilication. A signal is
2
7
observed at ca.0 ppm, Al MAS NMR spectra (Fig. 3) indicate existence of octahedral Al.
The broadening of the peak at around 0 ppm is a result of decrease symmetry of some Al sites
[
19]. A strong peak at ca. 53 ppm ascribed to framework Al atoms in tetrahedral
1
0 | P a g e

coordination. The results showed that desilication has not affected the framework Al content
of MTCK material. As per reported literature, if the loss of Si upon desilication is negligible
or less than 40% of its original content, the tetrahedral Al sites are almost remain unaffected
by desilication [20].
29Si MAS NMR
27Al MAS NMR
Parent H-BEA
Parent H-BEA
MTCK
MTCK
2
9
27
Fig. 3 Si MAS NMR and Al MAS NMR of parent H-BEA and MTCK
3
.5 XRD
The wide angle XRD patterns of parent H-BEA, As-MTCK (As synthesized) and MTCK
have been presented in Fig. 4A. The main characteristic peaks of parent H-BEA is located
between 2θ range, 7° - 10° and 23° – 25° [21]. The XRD patterns of As-MTCK (as
synthesized, non-protonic form) and MTCK are found to exhibit typical diffraction peaks of
parent zeolite H-BEA, indicating retention of crystallinity of zeolite H-BEA, after ammonium
ion exchange and even upon desilication. The appearance of highly intense peak in 2θ range
~
0.5°–1°, in XRD pattern of MTCK (Fig. 4B) indicates generation of highly ordered
mesostructure with long range order and hexagonal symmetry, due to desilication [15].
1
1 | P a g e

(A)
(B)
MTCK
As-MTCK
MTCK
Parent H-BEA
0
1
2
3
4
5
1
0
20
30
40
50
60
70
80
2
θ (Degree)
2
Theta (degree)
Fig.4 (A) Wide angle XRD of parent H-BEA, As-MTCK and MTCK, 4(B) Small angle
XRD of MTCK
3
.6 FT-IR
The FT-IR spectrum of parent H-BEA and MTCK are depicted in Fig. 5. The parent zeolite
-
1
H-BEA showed a broad band in the wavelength range of 3248 to 3650 cm which may be
attributed to the –OH stretching of hydrogen bonded internal silanol groups and O-H
-1
stretching and the band at 1637 cm was assigned to the -OH bending vibrations due to
-1
-1
adsorbed water molecules. The broad bands at around 1086 cm and 1223 cm , corresponds
to asymmetric stretching of internal and external vibration bands, corresponding to SiO and
AlO of siliceous materials, respectively [22]. From elemental analysis, the Si/Al ratio of
desilicated zeolite BEA was found lower (Si/Al ratio = 12) than that of parent zeolite BEA
Si/Al ratio = 18). As a result, the vibration band shifts to lower side. The shifting of the band
4
4
(
-1
-1
from 1086 cm in case of BEA to 1076 cm in MTCK, of the present work reveals alteration
of framework Si/Al via desilication. This leads to corresponding changes in the acidic
-1
characteristic of zeolite BEA. The stretching bands around 798 cm is attributed to the
-1
symmetric stretching vibrations of SiO
4
and AlO
4
.The band at 572 cm is a characteristic
band of zeolite H-BEA due to the presence of five and six membered rings in the structure
-1
[
23]. The small band observed at around 467 cm corresponding to Si-O-Si bending modes.
1
2 | P a g e

A
MTCK
797
3
635
1
637
1227
1
086
461
72
5
Parent H-BEA
4
000
3000
2000
Wavenumber (cm )
1000
-
1
Fig.5 FT-IR of parent H-BEA and MTCK
3
.7 TGA
In case of parent H-BEA and MTCK, the % Wt. loss upto 110°C were found to be 10.8 %
and 18.1 %, respectively. The % Wt. losses in the temperature region, 110-250°C were found
to be 2.07 % and 1.80 % and beyond 250°C the % Wt. losses were found to be 1.31 % and
1
.35 %, respectively (Fig. 6). The % Wt. loss is found higher in the case of MTCK as
compared to parent H-BEA due to large amount of adsorbed moisture. The % Wt. loss in the
temperature region of 110-250°C is attributed to the removal of some of the organic matter.
In both cases, negligible % Wt. loss is observed, beyond 250 °C.
1
3 | P a g e

Fig. 6TGA of parent H-BEA and MTCK
.8 NH - TPD
Figure 7 shows the NH
parent H-BEA and MTCK zeolites. The acidity of the samples was determined from these
3
3
3
-TPD profiles obtained from ammonia adsorbed on the samples of
NH
3
3
-TPD profiles and expressed in terms of mili-moles of ammonia per g of sample. The
-TPD profiles of both samples exhibited two desorption peaks at higher temperature
NH
(above 300 °C) and at lower temperature, in the range ~150–250 °C corresponding to
presence of strong and weak acid sites, respectively [24]. MTCK material was found to
exhibit enhanced both strong and weak acidity. The value of total acidity of MTCK found to
be 1.34 mmol/g. This value is slightly higher as compared to parent H-BEA material (1.31
mmol/g) (Table 1), which, may be ascribed to introduction of zeolite particles inside or on the
walls of mesopore of MTCK material. Similar observations are also made elsewhere
[
25][14]a.
1
4 | P a g e

Parent H-BEA
MTCK
1
00
200
300
400
500
600
700
Temperature ꢀC
Fig.7 NH – TPD of parent H-BEA and MTCK
3
3
.9 BET Surface area
The specific surface area of parent H-BEA and MTCK were determined by BET (Brunauer–
Emmett–Teller) method. The pore size and pore volume were calculated using desorption
branches of nitrogen isotherms and BJH (Barrett Joyner Halenda) method. The data are
2
depicted in Table 1. The BET surface area of parent H-BEA (490.1 m /g) was found lower
2
than MTCK (657.2 m /g).The external surface area is found to be increased significantly in
the case of MTCK due to mesopore formation [26].In the case of MTCK, a typical type - IV
adsorption branch consisting of H4 hysteresis loop, is observed. H4 loop are generally found
with aggregated crystals of zeolites and mesozeolites. A sharp step-down of desorption
branch is observed in narrow p/p
0
range = 0.4 – 0.5, indicating filling of micropores. The
~ 0.4 (Fig. 8A) indicates the
presence of a very small hysteresis loop which closes at p/p
0
presence of inter-crystalline void with bimodal porosity (mesopores and micropores) [5,27].
In MTCK, a larger hysteresis loop appears at relative pressure of p/p0 = 0.8 - 1.0 in the
1
5 | P a g e

isotherm which indicates presence of secondary mesopores. The pore size distribution studies
Fig.8B) reveals existence of uniform secondary mesopores with a narrow pore size
(
distribution centred at ~2.3 nm (MTCK). BJH pore size distribution curve strongly supports
the formation of hierarchical zeolite structure with pronounced mesoporosity [28]. The figure
depicts the existence of a step wise distributed pore structure with pore sizes of about 2 to 2.5
and 25 nm. However, the primary pores with a diameter of 2.5 nm are dominant. The pores
with diameters 2 – 2.5 nm may be mostly of framework confined (primary pores) whereas
pores with larger diameters are more likely the secondary pores with the contribution of the
filling of interparticle spaces [19]. This fact is also evidenced by a t-plot data (Table 1). It is
3
-1
estimated that the beta structure in the material contributes a pore volume of 0.44 cm .g and
2
-1
that mesopore surface area around 404 m .g . Thus, the results depicted in Table 1, clearly
revealed that the MTCK material contain a bimodal porosity.
The Si / Al ratio in case of MTCK was found lower as compared to parent H-BEA (Table 1),
as expected, due to successful desilication process. The hexagonal unit cell parameter a
be calculated using the equation, a = 2d100/√3, and t is the wall thickness of the mesopore
structure in the material, which can be obtained from the equation t = a – D, where; D is a
o
, can
o
o
pore diameter. The values of wall thickness are found quite distinguishable, in case of parent
H-BEA and MTCK, the values are being, 4.60 and 10.43 nm, respectively. The wall of
MTCK assembled from the filtrate of alkali-treated microporous zeolite BEA and hence
found thicker than that of parent H-BEA. The study clearly reveals that the introduction of
zeolite building units in to the mesopore walls of the materials are responsible for thicker
walls. This observation is also found in accordance with the reported literature [14]a.
1
6 | P a g e

A
4
3
3
2
2
1
1
00
50
00
50
00
50
00
MTCK
Parent H-BEA
0
.0
0.2
0.4
0.6
0.8
1.0
Relative pressure P/P
0
B
Parent H-BEA
MTCK
0
50
100
150
200
250
300
350
400
Pore diameter (Å)
Fig. 8(A): N
2
adsorption-desorption isotherm of parent H-BEA and MTCK and
Fig. 8(B): Pore diameter of parent H-BEA and MTCK
1
7 | P a g e

Table 1. Physicochemical properties of materials
a
cSext
2
b
3
d
e
f
Materials
S
BET
2
S
micro
2
Pore volume cm /g
Pore
size
(Å)
Wall
Si / Al Total
(
m /g) (m /g) (m /g)
thickness ratio
(t)
acidity
(mmol/g
)
b
V
micro
V
meso
V
total
(nm)
Parent H-BEA 490.1 424
66
0.21
0.13
0.40
0.31
0.61
0.44
50.05 4.60
18
12
1.31
MTCK
657.2 253
404
27.25 10.43
1.34
a
b
c
e
f
BET method, BJH method, Sext =
S
BET -Smicro, ICP-OES method, Ammonia TPD method
3
.10 Optimization study using BBD
BBD was used to find optimized reaction parameters for obtaining maximum % conversion
of LA, using MTCK catalyst. An experimental design was performed using Minitab 18.0
), catalyst concentration
) were chosen as independent variables. Each variable is
®
(
trial version) software. In this study, LA to n-butanol molar ratio (X
1
(
%) (X ) and reaction time (h) (X
2
3
coded as the three level, -1, 0, +1 by representing low, centre and high level, respectively
Table 2). Total 15 experiments are found to be required for development of BBD and their
(
results are presented in Table 3 [29,30]. The predicted conversion of LA was calculated using
following regression equations (Eq.2 and Eq.3)
Table 2. Selected variables and coded levels used in the BBD
Variables
Symbol
Coded levels
-1
0
+1
Molar ratio (LA to n-
butanol)
Catalyst concentration (%)
X
1
X
2
X
3
5
1
1
7.5
10
7
13
6
Reaction time (h)
3.5
ퟐ
ퟑ
ퟑ
ퟑ
ퟐ
풀 = 휷 + ∑ 휷 푿 + ∑ 휷 푿 + ∑ ∑
휷 푿 푿 … … . (2)
ퟎ
풊
풊
풊풊
풊
풊풋
풊
풋
풊ꢀퟏ
풊ꢀퟏ
풋ꢀ풊ꢁퟏ
풊ꢀퟏ
Y = 97.9 – 3.43 X
0.0430 (X
……(3)
1
- 0.006 X
2
– 2.77 X
3
+ 0.289 (X
1
)2 + 0.0256 (X
3
2
)2 + 0.473 (X )2
3
-
…
1
X
2
) - 0.013 (X
1
X
3
) + 0.1588 (X
2
X )
1
8 | P a g e

Where, 97.9 is the intercept and Y is the predicted LA conversion. X
1
, X2
and X are input
3
variables associated with, LA to n-butanol molar ratio, catalyst concentration and reaction
time, respectively. -3.43, -0.006, -2.77 are linear coefficient, 0.289, 0.0256, 0.473 are square
coefficient, whereas; -0.0430, -0.013, 0.1588 are interaction coefficient, respectively. In the
Eq. 3, negative sign in front of the terms, indicates antagonistic effect while a positive sign
indicates synergistic effect [31].
Table 3. The BBD matrix of the three variables in coded units and the response values
Standard
run
Experimental variables
LA % Conversion
X
1
X
2
X
3
Experimental Predicted
1
2
3
4
5
6
7
8
-1
-1
-1
+1
+1
0
0
0
0
0
82.65
86.34
94.46
95.57
85.42
95.01
88.74
98.00
84.20
88.27
92.52
94.01
86.41
94.38
89.36
97.00
+1
-1
+1
-1
-1
+1
-1
+1
-1
-1
+1
+1
0
-1
0
+1
+1
0
0
9
0
0
0
0
0
0
0
-1
+1
-1
+1
0
88.43
87.20
89.85
98.15
87.45
87.26
86.34
85.87
88.13
88.91
100.70
87.02
87.02
88.27
1
1
1
1
1
1
0
1
2
3
4
5
0
0
0
0
2
The model was found to have correlation value (R ) 0.91, i.e. the results show 8.51% of total
deviation in the experiments as compared to predicted conversion which is depicted in Fig.
2
9
.The ideal value of R is 1.00 indicating a complete agreement between predicted and
2
experimental responses [10]. Hence, the value of R is close to 1 which indicates a good
agreement between predicted and experimental conversion of LA to n-butyl levulinate in this
BBD model. The relationship between the response and experimental levels of each factor
1
9 | P a g e

was established by polynomial equation and the optimum conditions were obtained by 3D
response surface plots. The statistical analysis of the response surface model is shown in
Table 4. The value of P for the model is < 0.05 which indicates that it is a significant and
desirable model [32]. Besides, the value of P < 0.038, indicates that there is only 3.8% chance
of noise exists in the experiments. Apart from that, the ‘Lack of Fit F-value’ of 27.28 implies
that Lack of Fit is not significant relative to pure error. Hence, there is only a mere 3.6 %
chance that a ‘Lack of fit F-value’ this may be due to noise factors such as human errors
(experimental) or instrumental errors such as speed of stirrer, fluctuation in temperature etc.
2
R = 0.91
Experimental % conversion of n-butyl levulinate
Fig.9 Predicted Vs. experimental values for the synthesis n-butyl levulinate over MTCK
catalyst
Table 4. Analysis of variance for response surface quadratic model
Source
Degree Adjusted
of sum of
freedom square
Adjusted
mean
square
F-value
P-value
Model
Linear
9
3
1
1
1
3
303.288
236.037
15.429
98.772
121.836
42.855
33.699
78.679
15.429
98.772
121.836
14.285
5.50
0.038
0.009
0.174
0.010
0.007
0.191
12.83
2.52
X
1
X
2
X
3
16.11
19.87
2.33
Square
2
0 | P a g e

(
(
(
X
X
X
1
2
3
)2
)2
)2
1
1
1
3
1
1
1
5
3
2
14
12.052
3.137
12.052
3.137
32.332
8.132
1.664
0.027
22.705
6.132
9.976
0.366
-
1.97
0.51
5.27
1.33
0.27
0.00
3.70
-
0.220
0.506
0.070
0.364
0.625
0.949
0.112
-
32.332
24.397
1.664
2
-Way Interaction
(
(
(
X
1
1
2
X
2
3
3
)
)
)
X
X
X
X
0.027
22.705
30.660
29.929
0.731
Error
Lack-of-Fit
Pure Error
Total
27.28
-
0.036
-
333.948
-
-
F -mean of the within group variances. P – probability
Table 5. Optimum process parameters and validation of model adequacy
Parameters
Molar
ratio
Catalyst
concentration
(%)
Reaction time
(h)
% conversion
(LA : n-
butanol)
1:10
1:10
Predicted
Experimental
13
13
6
6
103.17
99.44
(MTCK)
The response of each factor was examined by analysis of variance (ANOVA). Further, the
Minitab-18.0 software was used to find optimization of individual response (Predicted
response). The predicted conversion under optimization was found to be 99.44% (Table 5).
The experiments have been conducted thrice under the optimized reaction conditions and
obtained higher LA conversion than the higher conversion found in the 15 experiments
(Table 3, experiment-12) which authenticates the applicability of the BBD method for n-butyl
levulinate synthesis using MTCK as a solid acid catalyst.
3
.10.1 Influences of process variables on % conversion of LA to n-butyl levulinate
The whole model can be presented in the form of three dimensional response surface plots
3D) for the conversion of LA to n-butyl levulinate for various values of reaction time to
(
2
1 | P a g e

catalyst concentration (Fig. 10), reaction time and molar ratio (Fig. 11) and molar ratio and
catalyst concentration (Fig. 12). Any two variables were kept constant for all these cases.
Fig. 10 shows the relation of different reaction time and the molar ratio of LA to n-butanol on
the n-butyl levulinate synthesis in 3D surface response plot. It is observed that, as molar ratio
increased from 5 to 10 and reaction time increased from 2 to 6 h, the LA conversion
increased proportionally. It suggests that the reaction time has slightly positive effect on LA
conversion [33,34].
1
00
9
5
%
Conversion
9
0
6
4
8
5
Reaction time
5
.0
2
7
.5
1
0.0
Molar ratio
Fig. 10 Surface plot of experimental conversion of LA Vs. reaction time and molar ratio
The effects of catalyst concentration and reaction time on LA conversion is shown in Fig. 11.
The % LA conversion was influenced significantly by the catalyst concentration and reaction
time. At low catalyst concentration (at moderate level of catalyst concentration), the % LA
conversion slightly increases with reaction time. This suggests that the reaction time has
larger positive effect on LA conversion and thus, the reaction time is more significant
parameter in comparison with other parameters.
2
2 | P a g e

1
00
9
5
%
Conversion
9
0
5
6
8
4
Reaction time
0
2
4
8
1
2
Catalyst concentration
Fig. 11 Surface plot of experimental conversion of LA Vs. reaction time and catalyst
concentration
Fig. 12 represents the effects of interaction between catalyst concentration and molar ratio on
the % LA conversion in 3D response surface. High % LA conversion was obtained when
high catalyst concentration and molar ratio is used. In case of high catalyst concentration, LA
conversion was found initially steady. And then, the conversion was found to be increased
with increasing molar ratio and catalyst concentration due to presence of acid sites on the
surface of the catalyst.
9
9
2.5
0.0
%
Conversion
8
8
7.5
5.0
1
0.0
7
.5
Molar ratio
0
4
8
5.0
1
2
Catalyst concentration
Fig. 12 Surface plot of experimental conversion of LA Vs. molar ratio and catalyst
concentration
2
3 | P a g e

3
.10.2 Catalyst reusability study
To examine the reusability of the MTCK catalyst (Fig. 13), the catalyst was filtered and
o
washed with ethanol, followed by, drying at 120°C for 2 h in an oven and then 5.5 h at 550 C
for its reuse in the subsequent experiment with fresh reactants. The catalytic reusability study
was carried out upto 4 catalytic runs. The catalyst was found robust and reusable without loss
in its significant activity. The wide angle XRD patterns (Fig. S1) of the spent MTCK was
found to be similar to the fresh catalyst, which indicates retention of the catalyst framework
structure throughout the course of the reaction.
1
00
8
6
4
2
0
0
0
0
0
1
st
2nd
3rd
4th
Catalytic run
S
Fig. 13. Catalyst reusability study
.10.3 Comparison between present and reported protocols
The literature is exits on LA esterification for synthesis of n-butyl levulinate. Yang et al.
2018) [35] have reported 92 % of yield of n-butyl levulinate using 10% GC400 catalyst
dosage and 1 : 5 LA to n-butanol. Enumula et al. (2017) [36] have reported 94 % yield of n-
butyl levulinate using 20% Al /SBA-15 catalyst, 1 : 7, LA : n-butanol molar ratio, and 250
C reaction temperature. Yadav et al. (2008) [37] achieved 89% yield of n-butyl levulinate
3
(
2
O
3
°
using bioenzyme, Novozym 435 (Candida antarctica lipase). All these protocols are
associated with certain drawbacks such as higher catalyst loading, high cost and poor
2
4 | P a g e

o
recyclability. In the present case, 99.4 % LA conversion is observed at 120 C with 91.51%
yield of n-butyl levulinate and 99 % selectivity using an efficient reusable 13% of micromeso
MTCK catalyst.
4
Conclusion
In the present research, an attempt has been made to modify parent zeolite H-BEA zeolite
using TTAB as structure directing agent via desilication post synthetic route. The resultant
mesozeolite BEA material (MTCK) is obtained with improved catalytic characteristics
(higher surface area and acidity) and bimodal porosity. Low angle XRD and BET surface
area analysis confirms the creation of mesopores within micropores. Further, in order to
evaluate the catalytic activity of MTCK, an esterification reaction of LA with n-butanol has
been carried out to synthesize n-butyl levulinate over MTCK. The optimization of process
parameters has been successfully carried out using BBD and an optimum of 99.44 % LA
conversion was obtained under optimized reaction conditions. RSM analysis predicted that
the reaction time is most significant and molar ratio of LA to n-butanol is least significant
parameter in esterification reaction of LA with n-butanol. The quadratic model developed by
the software was well validated and proven to be statistically adequate and accurate to predict
the optimal conversion of LA. The catalytic process developed with MTCK for levulinate
ester synthesis can also be further explored for the environmentally benign conversion of
other biomass derived compounds into value added chemicals using various organic
transformations.
Acknowledgment
The authors wish to thank Director, Sardar Vallabhbhai National Institute of Technology,
Surat, Gujarat, India for providing research facilities and Rajiv Gandhi National Fellowship,
2
5 | P a g e

UGC, New Delhi, India, for financial assistance. The authors would like to thank Sud -
Chemie India Pvt. Ltd., Vadodara, India, for providing zeolite samples. The authors
acknowledge Department of Science and Technology (DST), New Delhi, Government of
INDIA (Grant No.: DST/TM/WTI/2K14/191(G), dated: 23/02/2015) awarded to Dr. Kalpana
C. Maheria for equipment / instrument support, needed for the catalysts’ synthesis.
2
6 | P a g e

References:
[
[
[
[
[
[
[
[
1] V. Valtchev, G. Majano, S. Mintova, J. Pérez-Ramírez, Tailored crystalline
microporous materials by post-synthesis modification, Chem. Soc. Rev. 42 (2013)
2
63–290. doi:10.1039/c2cs35196j.
2] N. Linares, A.M. Silvestre-Albero, E. Serrano, J. Silvestre-Albero, J. García-Martínez,
Mesoporous materials for clean energy technologies, Chem. Soc. Rev. 43 (2014)
7
681–7717. doi:10.1039/C3CS60435G.
3] D.X. Liu, L. Emdadi, Y.Q. Wu, G.H. Zhu, Dual template synthesis of meso /
microporous zeolite with tailored activity in catalytic reactions, Chem. Matreials. 247
(2014) 1345−1355.
4] A.F. Lee, J.A. Bennett, J.C. Manayil, K. Wilson, Heterogeneous catalysis for
sustainable biodiesel production via esterification and transesterification, Chem. Soc.
Rev. 43 (2014) 7887–7916. doi:10.1039/C4CS00189C.
5] C.R. Patil, P.S. Niphadkar, V. V. Bokade, P.N. Joshi, Esterification of levulinic acid to
ethyl levulinate over bimodal micro-mesoporous H / BEA zeolite derivatives, Catal.
Commun. 43 (2014) 188–191. doi:10.1016/j.catcom.2013.10.006.
6] J.C. Groen, S. Abelló, L.A. Villaescusa, J. Pérez-Ramírez, Mesoporous beta zeolite
obtained by desilication, Microporous Mesoporous Mater. 114 (2008) 93–102.
doi:10.1016/j.micromeso.2007.12.025.
7] Y. Wang, T. Yokoi, S. Namba, T. Tatsumi, Effects of dealumination and desilication
of beta zeolite on catalytic performance in n-Hexane cracking, Catalysts. 6 (2016) 8.
doi:10.3390/catal6010008.
8] S.K. Sonar, A.S. Shinde, A. Asok, P.S. Niphadkar, S. Mayadevi, P.N. Joshi, V. V.
2
7 | P a g e

Bokade, Solvent free acetalization of glycerol with formaldehyde over hierarchical
zeolite of BEA topology, Environ. Prog. Sustain. Energy. 00 (2017) 1–11.
doi:10.1002/ep.12742.
[
[
9] K.Y. Nandiwale, V. V Bokade, Selective synthesis of propofol ( 2 , 6-
Diisopropylphenol ) an intravenous anesthetic drug by isopropylation of phenol over
H-Beta and H-Mordenite, RSC Adv. 4 (2014) 32467–32474.
doi:10.1039/C4RA03004D.1.
10] K.Y. Nandiwale, V. V Bokade, Optimization by Box – Behnken experimental design
for synthesis of n -Hexyl levulinate biolubricant over hierarchical H-ZSM-5ꢀ: An effort
towards agricultural waste minimization, Process Saf. Environ. Prot. 99 (2015) 159–
1
66. doi:10.1016/j.psep.2015.11.003.
[
[
[
[
11] K.C. Maheria, J. Kozinski, A. Dalai, Esterification of levulinic acid to n-butyl
levulinate over various acidic zeolites, Catal. Letters. 143 (2013) 1220–1225.
doi:10.1007/s10562-013-1041-3.
12] S. Dharne, V. V Bokade, Esterification of levulinic acid to n -butyl levulinate over
heteropolyacid supported on acid-treated clay, J. Nat. Gas Chem. 20 (2011) 18–24.
doi:10.1016/S1003-9953(10)60147-8.
13] D.R. Lathiya, D. V Bhatt, K.C. Maheria, Synthesis of sulfonated carbon catalyst from
waste orange peel for cost effective biodiesel production, Bioresour. Technol. Reports.
2
(2018) 69–76. doi:https://doi.org/10.1016/j.biteb.2018.04.007.
14] (a) S.R. Mistry, K.C. Maheria, Synthesis , characterization of mesoporous zeolite BEA
and its application in multicomponent reactions, J. Catal. Catal. 1 (2014) 1–21.
(b) Unpublished work: K. Maheria, S. Mistry, J. Kozinski and A. Dalai, Development
2
8 | P a g e

of novel mesozeolite BEA and their application in biomass derived value added
chemical synthesis.
[
[
[
[
[
[
15] D. Morawala, A. Dalai, K. Maheria, Rice husk mediated synthesis of meso-ZSM-5 and
its application in the synthesis of n-butyl levulinate, J. Porous Mater. 26 (2018) 677–
6
86. doi:10.1007/s10934-018-0664-6.
16] D. Verboekend, M. Milina, S. Mitchell, J. Pérez-Ramírez, Hierarchical zeolites by
desilication: Occurrence and catalytic impact of recrystallization and restructuring,
Cryst. Growth Des. 13 (2013) 5025–5035. doi:10.1021/cg4010483.
17] B. Dupuy, S. Laforge, C. Morais, C. Bachmann, P. Magnoux, F. Richard, Alkylation
of 3-methylthiophene by 2-methyl-1-pentene over HY , HBEA and HMCM-22 acidic
zeolites, Appl. Catal. A Gen. 414 (2012) 192–204. doi:10.1016/j.apcata.2011.11.009.
18] B. Tang, W. Dai, X. Sun, N. Guan, L. Li, M. Hunger, A procedure for the preparation
of Ti-Beta zeolites for catalytic epoxidation with hydrogen peroxide, Green Chem. 16
(2014) 2281–2291. doi:10.1039/c3gc42534g.
19] W. Guo, L. Huang, P. Deng, Z. Xue, Q. Li, Characterization of Beta / MCM-41
composite molecular sieve compared with the mechanical mixture, Microporous
Mesoporous Mater. 45 (2001) 427–434.
20] C. Doremieux-Morin, A. Ramsaran, R. Le Van Mao, P. Batamack, L. Heeribout, V.
Semmer, G. Denes, J. Fraissard, 1H broad-line and MAS NMR: application to the
study of acid sites of desilicated zeolite ZSM-5, Catal. Letters. 34 (1995) 139–149.
doi:10.1007/BF00808330.
[
21] J. Gabla, S. Mistry, M. Kalpana, Catalysis Science & Technology tetra-substituted
imidazoles catalyzed by zeolite, Catal. Sci. Technol. 7 (2017) 5154–5167.
doi:10.1039/C7CY01398A.
2
9 | P a g e