Suppression of coke formation and enhancement of aromatic hydrocarbon production in catalytic fast pyrolysis of cellulose over different zeolites: Effects of pore structure and acidity

Article abstract of DOI:10.1039/c5ra11332f

Rapid deactivation of zeolites caused by high formation and deposition of coke is a great challenge in catalytic conversion of biomass materials into value-added chemicals and fuels. The main purpose of this work was to reduce the formation of both types

Full text of DOI:10.1039/c5ra11332f

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Suppression of coke formation and enhancement
of aromatic hydrocarbon production in catalytic
fast pyrolysis of cellulose over different zeolites:
effects of pore structure and acidity
Cite this: RSC Adv., 2015, 5, 65408
Pouya Sirous Rezaei, Hoda Shafaghat and Wan Mohd Ashri Wan Daud*
Rapid deactivation of zeolites caused by high formation and deposition of coke is a great challenge in
catalytic conversion of biomass materials into value-added chemicals and fuels. The main purpose of
this work was to reduce the formation of both types of thermal and catalytic coke over zeolites. It was
revealed that there is a significant interaction between zeolite pore structure and the density of acid sites
which could be optimized for reduced coke formation. In this study, catalytic pyrolysis of cellulose was
conducted using HZSM-5 (Si/Al: 30), HY (Si/Al: 30) and physically mixed catalysts of HZSM-5 (Si/Al: 30)
and dealuminated HY (Si/Al: 327). Coke formation over the physically mixed catalysts was remarkably
lower than that over HZSM-5 and HY; the coke contents of HZSM-5, HY and physically mixed catalysts
of HZSM-5 and dealuminated HY with a ratio of 70 : 30 wt% were 7.01, 11.47 and 4.82 wt%, respectively.
The aromatic hydrocarbon yield was also considerably enhanced over the physically mixed catalysts
compared to HZSM-5 and HY; the aromatic hydrocarbon yields achieved over HZSM-5, HY and the
mixture of HZSM-5 and dealuminated HY (70 : 30 wt%) were 20.31, 8.91 and 27.01 wt%, respectively. This
study shows that the interactive effects of zeolite characteristics such as pore structure and acidity could
be taken into account for designing more efficient catalysts to achieve lower coke formation and higher
production of desired products.
Received 14th June 2015
Accepted 27th July 2015
DOI: 10.1039/c5ra11332f
www.rsc.org/advances
of pyrolysis vapors in catalytic pyrolysis of biomass feed-
stocks. The major challenge in this process is high formation
1
. Introduction
The depletion of fossil fuels and the negative impacts of their and deposition of coke on zeolite which causes high deacti-
1
0
utilization on the environment such as global warming and air vation of catalyst. The reason for high yield of coke is low
pollution necessitate the discovery of renewable, sustainable hydrogen to carbon effective ratio of biomass which leads to
1
–3
and environmentally friendly alternative sources of energy.
Lignocellulosic biomass is of great potential to be considered Coke deposited on catalyst is divided into two types of thermal
low hydrogen content in hydrocarbon pool inside catalyst.
1
1
as such an alternative since it is abundantly available, and the and catalytic origin. Thermal coke is produced by homoge-
fuel obtained from biomass is carbon dioxide neutral.
Biomass pyrolysis which leads to a liquid product called bio- and is mainly deposited on outer surface of catalyst. Cata-
4
neous thermal polymerization of compounds in gas phase,
1
2
oil has gained extensive attention as an economically lytic coke is formed in the internal channels of catalyst as a
5
feasible technique for large-scale exploitation of biomass.
result of heterogeneous transformation of oxygenate
However, bio-oil is highly oxygenated and has poor fuel compounds over zeolite acid sites through reactions of olig-
properties such as high viscosity, poor heating value, corro- omerization, cyclization, aromatization and condensa-
6
–9
13–15
siveness, and chemical and thermal instability. It is also tion.
immiscible with conventional fossil fuels, and therefore needs through poisoning zeolite acid sites and pore blockage. In
to be upgraded. Catalytic pyrolysis is considered an efficient addition to catalyst deactivation, coke formation is
Coke deposition results in catalyst deactivation
a
technology since it includes both steps of pyrolysis and cata- competing reaction with production of desired products.
lytic upgrading in one unit. Zeolites have been the most Therefore, it is essential to optimize catalyst properties in
common catalysts used for atmospheric pressure upgrading order to lower coke formation.
Zeolite pore structure and acidity are the two signicant
catalyst properties which greatly affect the amount of coke
Department of Chemical Engineering, Faculty of Engineering, University of Malaya, formation. In catalytic upgrading of pine wood pyrolysis vapors,
5
0603 Kuala Lumpur, Malaysia. E-mail: pouya.sr@gmail.com; h.shafaghat@gmail.
it was revealed that among beta, Y and ferrierite zeolites,
com; ashri@um.edu.my; Fax: +60 3 79675319; Tel: +60 3 79675297
65408 | RSC Adv., 2015, 5, 65408–65414
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
1
6
ferrierite with too narrow pores led to less coke yield.
2
. Experimental
Similarly, in catalytic pyrolysis of pine wood, it was reported
that higher amount of coke was formed over the catalyst with
2.1. Catalyst preparation
1
7
higher pore size. Coke content of spent zeolites decreased in HZSM-5, HY (Zeolyst, CBV 720, SiO /Al O molar ratio: 30) and
2
2 3
the order H–Y > H-beta > H-mordenite > HZSM-5. In another mixtures of HZSM-5 and dealuminated HY were used for cata-
work for catalytic pyrolysis of wood biomass, catalysts with lytic pyrolysis of cellulose (Acros Organics) in this work. HZSM-5
larger pore size resulted in higher coke yields; Y zeolite and was obtained by calcination of the ammonium form of ZSM-5
ꢀ
activated alumina which contain larger pore size produced (Zeolyst, CBV 3024E, SiO /Al O molar ratio: 30) at 550
C
2
2
ꢁ1
3
1
8
ꢀ
higher amount of coke compared to ZSM-5. It was declared (with heating rate of 3 C min ) for 12 h. The dealuminated HY
that large coke precursors could diffuse into pore structure of was obtained by treatment of HY in 2 M aqueous HCl solution at
ꢀ
catalysts with larger pore size and involve in coke formation. 80 C for 12 h using 15 ml acid solution/gzeolite. Then, the
In catalytic pyrolysis of glucose using various zeolites with sample was ltered, washed with distilled water, and dried at
ꢀ
pores of different size and shape, the least amount of coke 100 C for 12 h. Aerward, the dealuminated sample was con-
formation was observed over ZSM-5 and ZSM-11 with medium verted to the protonic form by ion exchange in 0.1 M aqueous
ꢀ
pore size, and the highest coke yield was obtained using beta NH Cl solution at 60 C for 12 h using 50 ml NH Cl solution/
4
4
1
9
ꢀ
zeolite, SSZ-55 and Y zeolite which had the largest pores. It gzeolite, followed by calcination at 550 C for 12 h.
was also shown that coke yield is a great function of the space
inside zeolite channels; MCM-22, TNU-9 and IM-5 with
medium pore size caused high amounts of coke due to their
high internal pore space which provides enough space for
coke formation. However, in contrast with what is mentioned
2
.2. Catalyst characterization
The crystallinity of zeolites was determined by X-ray diffraction
(XRD) on a Rigaku Miniex diffractometer using Cu Ka radia-
˚
tion (l ¼ 1.54443 A) at 45 kV and 40 mA. Data were recorded in
2
0
above, Zhiqiang Ma et al. concluded that lower yield of coke
is formed over the catalyst with larger pore size, since larger
pore size allows larger molecules to enter catalyst and react
and not to be converted to thermal coke. In addition to pore
size, other characteristics of pore structure of zeolites such as
total porosity, pore shape, the amount of intercrystalline
pores and connectivity of zeolite channels have also signi-
cant impact on the amount of coke formation. Pore shape
could cause steric constraints for formation of the transition
ꢀ
ꢀ
the 2q range of 5–80 with a step size of 0.026 and scan rate of
ꢀ
ꢁ1
0.05 s . X-ray uorescence (XRF) instrument (PANalytical
AxiosmAX) was used for chemical analysis of catalysts. The
surface area and pore size distribution of catalysts were deter-
ꢀ
mined by N isothermal (ꢁ196 C) adsorption–desorption using
2
Micromeritics ASAP 2020 surface area and porosity analyzer.
Prior to the analysis, the samples were degassed in vacuum at
ꢀ
180 C for 4 h. Characterization of acid site distribution of the
catalysts was conducted by temperature programmed desorp-
tion of ammonia (NH -TPD) using Micromeritics ChemiSorb
720 instrument. Using He as carrier gas (20 ml min ), 200 mg
of each sample set in TPD cell was heated from ambient
2
1
states which are involved in production of coke precursors.
Meso- and macropores between zeolite crystals allow high
3
ꢁ
1
2
2
2,23
degree of polymerization resulting in the growth of coke.
Three-dimensional porous structure could reduce coke
formation due to high connectivity of channels which results
in enhanced movement of coke precursor intermediates to
ꢀ
ꢀ
ꢁ1
temperature to 600 C at a heating rate of 20 C min and
maintained at 600 C for 1 h. Aerward, the sample was cooled
to 210 C and purged with 10% NH /90% He mixture (20 ml
min ) for 30 min. Then, the sample was ushed with He gas
for 30 min for removal of physisorbed ammonia. Aer cooling
ꢀ
ꢀ
2
4
3
the outside of zeolite crystals. Apart from pore structure,
zeolite acidity is also inuential on the amount of coke
formation. Since catalytic coke is formed over zeolite acid
sites, its yield is dependent on strength distribution and
density of acid sites. Catalytic coke content of zeolite is
expected to be increased by increase in strength and number
of acid sites.
ꢁ
1
ꢀ
down to 70 C and when the thermal conductivity detector
(TCD) signal was stable, ammonia desorption measurement
ꢀ
ꢀ
was performed by heating the sample to 600 C at a rate of 10 C
ꢁ
1
ꢁ1
min under He ow (20 ml min ). The coke content of spent
catalysts was measured by thermogravimetric analysis using a
PerkinElmer STA 6000 Simultaneous Thermal Analyzer. Under
synthetic air ow (100 ml min ), samples were heated from 30
to 750 C at the rate of 5 C min and kept at nal temperature
In this work, the interactive effects of zeolite pore structure
and density of acid sites on coke formation was investigated.
Cellulose which is the most abundant organic polymer in
nature was used as feedstock. The main purpose of this study
was to enhance the yield of aromatic hydrocarbons and to
suppress coke formation in catalytic pyrolysis of cellulose using
a physically mixed catalyst system. Since aromatic hydrocar-
ꢁ
1
ꢀ
ꢀ
ꢁ1
for 30 min.
2
.3. Catalytic activity measurement
bons have wide range of applications and are the building Catalytic pyrolysis was carried out in a continuous, down-ow,
blocks for petrochemical industry, these highly desirable xed-bed tubular reactor (ID: 7.5 cm; height: 60 cm) made of
compounds are usually considered as target products in cata- stainless steel 316L which was placed coaxially in a two-zone
10,25,26
lytic conversion of biomass derived feedstocks.
HZSM-5, furnace. A stainless steel cylindrical cup with screen of 400
HY and mixtures of HZSM-5 and dealuminated HY were used mesh at the bottom side was set in each zone of the reactor. In
as catalyst.
each run, 5 g calcined catalyst was loaded in the second cup (ID:
2.5 cm; height: 5 cm), and cellulose was fed into the rst cup
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 65408–65414 | 65409

View Article Online
RSC Advances
Paper
through a feed hopper on the top of the reactor. The rst zone
was used for pyrolysis, and the pyrolysis vapors were passed
through catalyst bed in the second cup for catalytic upgrading.
All runs were conducted at atmospheric pressure. Temperature
of each zone was measured by a K-type thermocouple. Both
ꢀ
zones were heated to 500 C and nitrogen (Linde Malaysia Sdn.
ꢁ
1
Bhd.) was purged to the reactor at ow rate of 2 L min for
0 min. Aerward, 30 g cellulose was introduced to the reactor
3
ꢁ
1
2
with weight hourly space velocity (WHSV) of 6 h and N ow
rate was kept at 2 L min . The liquid products were collected by
ꢁ1
ꢀ
two condensers maintained at ꢁ10 C. All lines were heated to
avoid any condensation. Aer each run, the catalyst bed was
ꢁ
1
purged by N
2
ow (2 L min ) at the reaction temperature for
30 min in order to eliminate the components which might
remain on the catalyst. The organic phase of liquid product was
separated from the aqueous phase with ethyl acetate (R&M
chemicals) before injection to GC/MS. Qualitative and quanti-
Fig. 1 NH
3
-TPD profiles of HZSM-5 and the parent and dealuminated
tative analysis of liquid products was carried out by GC/MS forms of HY.
(
Shimadzu QP 2010, DB-5 30 m ꢂ 0.25 mm ꢂ 0.25 mm),
equipped with ame ionization and mass spectrometry detec-
tion. The GC oven temperature program was as follows:
ꢀ
ꢀ
temperature was held at 50 C for 5 min, ramped to 300 C at
ꢀ
ꢁ1
ꢀ
10
C min , and kept at 300 C for 10 min. The injector
temperature was 290 C and a split ratio of 50 : 1 was employed.
Helium (Linde Malaysia Sdn. Bhd) was used as carrier gas with
ꢀ
ꢁ
1

ow rate of 1.26 ml min . NIST (National Institute of Standards
and Technology) mass spectrum library was used for peak
identication. 2-Isopropylphenol (Sigma-Aldrich) was used as
internal standard for quantitative analysis of products. Yield
and selectivity of products were calculated as follows: yield ¼
(
weight of a certain product/total weight of feed) ꢂ 100; selec-
tivity ¼ (weight of a certain product/total weight of organic
phase in liquid product) ꢂ 100.
3. Results and discussion
3
.1. Physicochemical characteristics of catalysts
Fig. 2 X-ray diffraction patterns of the parent and dealuminated forms
of HY.
The acidity of catalysts determined by NH -TPD analysis is
3
shown in Fig. 1. The lower peak area of acid-treated HY
compared to that of parent HY demonstrates the reduction in
the number of acid sites caused by leaching of Al from zeolite
displayed H4-type hysteresis loop associated with narrow slit-
shaped pores, and dealuminated HY exhibited H3-type hyster-
esis loop associated with slit-shaped pores.
2 2 3
structure. SiO /Al O molar ratio of the parent and deal-
27
uminated forms of HY were 31.3 and 326.7, respectively. As
depicted in Fig. 2, both parent and dealuminated forms of HY
displayed the typical diffraction lines of Y zeolite. It can be seen
from XRD patterns that crystallinity of acid-treated HY had a
3.2. Catalytic pyrolysis of cellulose over HZSM-5 and HY
slight reduction, and crystalline structure of HY was not Table 2 presents the yields of gas, liquid and solid products
signicantly affected by dealumination. Table 1 presents the obtained from catalytic pyrolysis of cellulose using different
textural properties of catalysts evaluated from nitrogen zeolites. HZSM-5 resulted in lower yield of oil as well as higher
isothermal adsorption–desorption. BET surface area of deal- yields of gas and water compared to HY due to higher amount of
uminated HY was 13% lower than that of parent HY. Micropo- deoxygenation taken place over HZSM-5. As shown in Table 2,
rous surface area and volume of HY were reduced, while surface the aromatic hydrocarbons yield achieved from catalytic pyrol-
area and volume of mesopores were increased. This indicates ysis of cellulose over HZSM-5 and HY were 20.31 and 8.91 wt%,
that a portion of micropores was changed to mesopores due to respectively. The higher aromatic hydrocarbons yield of HZSM-
extraction of aluminium from zeolite microporous channels 5 compared to HY is due to the different pore structures of these
and creation of mesoporous space. As shown in Fig. 3, all catalysts. ZSM-5 is a zeolite with three-dimensional framework
zeolites displayed type IV isotherm. HZSM-5 and parent HY formed of 10-membered ring pores with dimensions of 0.51 ꢂ
65410 | RSC Adv., 2015, 5, 65408–65414
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Table 1 Chemical and textural properties of catalysts
a
3
b
2
ꢁ1
c
2
ꢁ1
d
3
ꢁ1
e
3
ꢁ1
f
3
ꢁ1
Sample
SiO
2
/Al
2
O
S
BET (m g
)
S
meso (m g
)
SBET/Smeso
V
total (cm g
)
V
micro (cm g
)
V
meso (cm g )
HZSM-5
Parent HY
Dealuminated HY
32.3
31.3
326.7
291
645
563
99
158
256
2.94
4.08
2.19
0.191
0.429
0.431
0.094
0.238
0.149
0.097
0.191
0.282
a
b
c
d
Determined by XRF analysis. Calculated in the range of relative pressure (P/P
0
) ¼ 0.05 ꢁ 0.25. Evaluated by t-plot method. Total pore volume
e
f
evaluated at P/P
0
¼ 0.99. Evaluated by t-plot method.
V
meso ¼ Vtotal ꢁ Vmicro
.
which act as coke precursors, resulting in lower deposition of
catalytic coke on HZSM-5 acid sites, and in turn, less deactiva-
28
tion of catalyst and higher yield of aromatic hydrocarbons.
The main aromatic hydrocarbons produced from cellulose
pyrolysis over HZSM-5 were toluene, xylene, trimethylbenzene
and ethyl-methylbenzene. The dominant oxygenated
compounds detected in the organic phase of liquid product
obtained using HZSM-5 and HY were furfural, benzofuran, 5-
hydroxymethyl furfural, phenol, cresol and benzenediol. The
reaction pathway for conversion of cellulose into aromatic
hydrocarbons is as follows: pyrolysis of cellulose to volatile
organics which are dehydrated to furans, followed by decar-
bonylation of furans to allene, and subsequent oligomerization
of the allene to olens which react with furans to form
29,30
aromatics.
bons, HY resulted in higher coke formation compared to HZSM-
; the content of coke deposited on HZSM-5 and HY were 7.01
In addition to lower yield of aromatic hydrocar-
Fig. 3 Nitrogen adsorption–desorption isotherms of HZSM-5 and the
parent and dealuminated forms of HY.
5
and 11.47 wt%, respectively. Besides, the main cause of coke
formation over these two zeolites was different. The results
obtained by thermogravimetric analysis of spent catalysts
shown in Fig. 4a and Table 3 depict that the coke deposited on
HZSM-5 is mostly of thermal origin and the coke formed over
0
.55 and 0.53 ꢂ 0.56 nm, while Y zeolite contains 12-membered
19
ring channels of 0.74 ꢂ 0.74 nm. The smaller pore size of
HZSM-5 prevents from formation of polyaromatic compounds
Table 2 Product yields and selectivities (wt%) obtained from catalytic pyrolysis of cellulose over different zeolites. Reaction conditions: WHSV, 6
ꢁ1
ꢀ
h
; reaction temperature, 500 C; pressure, 1 atm
HZSM-5/dealuminated HY
Catalyst
Yield
HZSM-5
HY
Dealuminated HY
70 : 30 wt%
50 : 50 wt%
30 : 70 wt%
%
Oil
27.96
31.85
37.59
28.98
30.28
31.22
Gas
33.74
29.73
27.17
33.46
32.86
32.26
Aqueous fraction
Char/coke
17.46
19.57/1.27
15.95
20.30/2.17
14.65
19.86/0.73
17.30
19.43/0.83
16.71
19.38/0.77
16.49
19.30/0.73
%
Yield of aromatic hydrocarbons
0.31
2
8.91
0.48
27.01
22.18
15.71
%
Selectivity in organic phase of liquid product
Benzene
Toluene
Xylene
Ethyl-methylbenzene
Trimethylbenzene
Tetramethylbenzene
Naphthalenes
Other hydrocarbons
Oxygenated compounds
1.36
19.47
17.08
7.53
9.91
2.86
9.24
5.19
27.36
0.30
5.72
6.82
3.07
4.22
1.46
4.84
1.54
72.03
0.09
0.23
0.27
0.13
0.16
0.11
0.19
0.10
98.72
2.01
26.28
24.8
11.64
13.75
1.97
8.40
4.35
6.80
1.17
0.71
11.49
10.65
6.65
8.06
1.21
6.5
18.45
16.85
8.85
11.24
2.18
9.68
4.83
26.75
5.05
49.68
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 65408–65414 | 65411

View Article Online
RSC Advances
Paper
HY is mostly of catalytic origin; HZSM-5 and HY resulted in on catalyst surface as thermal coke. However, HY which
catalytic coke content of 2.28 and 10.21 wt%, and thermal coke contains larger channels allows larger molecules enter the
content of 4.73 and 1.26 wt%, respectively. The two weight loss catalyst and react over zeolite acid sites inside catalyst channels.
ꢀ
regions in temperature range of 300–500 and 500–750 C were On the other hand, HY leads to higher yield of catalytic coke
considered as the amounts of thermal and catalytic coke since this zeolite with larger pore diameter provides larger space
ꢀ
deposited on catalyst, respectively. The weight loss below 300 C for polymerization of coke precursors and formation of the
20
was assigned to desorption of water and volatile components.
intermediates and transition states which are involved in coke
Differential thermogravimetry (DTG) shown in Fig. 4b indicates production resulting in higher amount of carbonaceous resi-
that maximum combustion of thermal and catalytic coke was dues deposited on zeolite acid sites. In contrast to HY, smaller
ꢀ
ꢀ
occurred at 440 and 580 C for HZSM-5, and 410 and 650 C for channels of HZSM-5 limit the degree of polymerization inside
HY, respectively. The difference in catalytic and thermal coke catalyst and cause lower yield of catalytic coke.
contents of HZSM-5 and HY is caused by the different pore
structures of these two zeolites. In fact, coke formation is a
shape selective reaction. Large molecules formed by thermal
cracking in homogeneous gas phase outside catalyst could not
3
.3. Catalytic pyrolysis of cellulose over physically mixed
catalysts of HZSM-5 and dealuminated HY
enter the narrow channels of HZSM-5 and undergo repolyme- The amount of catalytic coke formed over HY was remarkably
rization and condensation outside catalyst, and are deposited reduced by dealumination. The catalytic coke contents of HY
and dealuminated HY were 10.21 and 3.08 wt%, respectively.
27
Huifu Xue et al. also reported that dealumination resulted in
less deposition of catalytic coke on mordenite zeolite. Deal-
uminated HY contains lower density of acid sites which leads to
lower yield of catalytic coke. High molecular coke is formed
through several reaction steps, and since lower density of acid
sites leads to reduction in the number of acid sites which a
reactant encounters, the possibility for converting into coke
over dealuminated HY is attenuated. However, the aromatics
yield achieved over dealuminated HY was very low (below
0.5 wt%). Use of mixtures of HZSM-5 and dealuminated HY
showed to be efficient to achieve high yield of aromatic hydro-
carbons with low content of coke deposited on catalyst. The
coke contents of HZSM-5 and HY were 7.01 and 11.47 wt%,
while the coke contents of mixtures of HZSM-5 and deal-
uminated HY with ratios of 70 : 30, 50 : 50 and 30 : 70 wt% were
4
5
.82, 4.38 and 4.17 wt%, respectively. In the case of using HZSM-
as catalyst, the compounds with molecular size larger than
pore diameter of HZSM-5 could not enter catalyst and are con-
verted to thermal coke and deposited on HZSM-5 outer surface.
However, presence of HY with larger channels in catalyst
mixture allows the compounds in wider range of molecular size
to diffuse into catalyst and react over HY acid sites. In fact, the
dealuminated HY showed to be a suitable catalyst for initial
cracking of the compounds derived from pyrolysis of cellulose.
Therefore, some compounds which could not enter HZSM-5 are
Fig. 4 TGA (a) and DTG (b) of the spent catalysts used for cellulose
ꢁ1

rstly cracked over HY acid sites and converted to smaller
pyrolysis (WHSV, 6 h ; time on stream, 60 min; reaction temperature,
ꢀ
5
00 C).
compounds which could diffuse inside HZSM-5 channels and
Table 3 Content of total coke, thermal coke and catalytic coke deposited on the catalysts used for cellulose pyrolysis. Reaction conditions:
ꢁ1
ꢀ
WHSV, 6 h ; reaction temperature, 500 C; pressure, 1 atm; time on stream, 60 min
HZSM-5/dealuminated HY
Catalyst
HZSM-5
HY
Dealuminated HY
70 : 30 wt%
50 : 50 wt%
30 : 70 wt%
% gcoke/gcatalyst
Thermal coke
Catalytic coke
Total coke
4.73
2.28
7.01
1.26
10.21
11.47
1.07
3.08
4.15
2.36
2.46
4.82
1.73
2.65
4.38
1.38
2.79
4.17
65412 | RSC Adv., 2015, 5, 65408–65414
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
be transformed to aromatic hydrocarbons over HZSM-5 acid production of aromatic hydrocarbons in catalytic pyrolysis of
sites. Considering the thermal coke contents of the mixtures of biomass feedstocks. Therefore, catalyst properties should be
HZSM-5 and dealuminated HY shown in Table 3 as well as the optimized in order to have minimum selectivity towards coke
thermal coke content of dealuminated HY which is 1.07 wt% and maximum selectivity towards desired products. Deposition
(thermal coke content of dealuminated HY is supposed to be of both types of coke is needed to be suppressed in an efficient
unchanged in presence of HZSM-5), the thermal coke content of catalyst system. Pore structure and location of acid sites in
HZSM-5 in mixtures of HZSM-5 and dealuminated HY with catalyst structure are the two signicant properties which
ratios of 70 : 30, 50 : 50 and 30 : 70 wt% are estimated to be should be taken into account for reducing coke formation.
2.91, 2.39 and 2.10 wt%, respectively. This clearly shows that Presence of channels with larger dimensions facilitates the
thermal coke content of HZSM-5 is reduced by increase in the diffusion of larger molecules inside catalyst leading to reduc-
amount of dealuminated HY in catalyst mixture. Meanwhile, tion in the yield of thermal coke. Meanwhile, acid sites with
dealuminated HY does not contain enough number of acid sites high density should be located in small spaces which restrict
for high conversion of reactants to coke. Therefore, mixture of the degree of polymerization and catalytic coke deposition.
HZSM-5 and dealuminated HY results in less coke formation However, there should be enough space in the vicinity of acid
compared to HZSM-5 and HY due to the reduction in thermal sites in order to allow formation of transition states for desired
coke deposited on HZSM-5 and the decrease in catalytic coke products.
formed inside the channels of dealuminated HY. It could be
inferred from the results obtained in this study that there is a
4
. Conclusions
signicant interaction between pore space and density of zeolite
acid sites which should be taken into account in designing an
efficient catalyst. The steric constraints caused by limited space
in the vicinity of acid sites could prevent from coke formation.
Therefore, high density of acid sites located in small spaces
could not lead to high formation of coke. However in larger
channels, coke formation could be effectively suppressed by
decrease in the density of acid sites which leads to reduction in
the degree of polymerization and condensation. The aromatic
hydrocarbons production is also improved over the physically
mixed catalyst system. The aromatics yield achieved over HZSM-
The results obtained in this study revealed that there is a
signicant interaction between zeolite pore structure (pore size
and shape) and density of acid sites which greatly affects the
amount of coke formation and deposition on zeolite in
conversion of biomass feedstocks. It was also shown that these
zeolite properties could be optimized in order to suppress coke
formation and to enhance the yield of desired products. In
catalytic pyrolysis of cellulose, lower formation of coke as well as
higher yield of aromatic hydrocarbons were achieved over
physically mixed catalysts of HZSM-5 and dealuminated HY
compared to HZSM-5 and HY. Addition of HY to HZSM-5 results
in lower deposition of thermal coke over HZSM-5 due to larger
pores of HY which allow larger molecules diffuse into zeolite
and react. Besides, formation of catalytic coke over the physi-
cally mixed catalysts was suppressed by small space inside
HZSM-5 pores and low density of acid sites inside dealuminated
HY pores which both restrict the degree of polymerization of
coke precursors. The coke contents of HZSM-5 and HY were 7.01
and 11.47 wt%, respectively, while the coke content of physically
mixed catalysts of HZSM-5 and dealuminated HY with ratio of
5
was 20.31 wt% which was enhanced to 22.18 and 27.01 wt%
over mixtures of HZSM-5 and dealuminated HY with ratios of
0 : 50 and 70 : 30 wt%, respectively. This increase in aromatics
5
yield is due to the possibility for a higher fraction of compounds
to react over HZSM-5 since the compounds with molecular size
larger than pore diameter of HZSM-5 could undergo cracking in
larger channels of HY and diffuse inside HZSM-5 channels.
However, the mixture of HZSM-5 and dealuminated HY with
ratio of 30 : 70 wt% resulted in less aromatics yield compared to
HZSM-5 illustrating that the amount of HZSM-5 in the catalyst
mixture should be adequate in order to proceed the reactions
required for aromatics production; as mentioned in the
previous section, formation of polyaromatic compounds as
coke precursors is restricted by the steric constraints caused by
smaller pore size of HZSM-5 resulting in higher catalytic activity
of HZSM-5 and higher yield of aromatic hydrocarbons produced
70 : 30 wt% was 4.82 wt%. Meanwhile, The aromatic hydrocar-
bons yield achieved over HZSM-5 and HY was 20.31 and 8.91
wt%, respectively, which was enhanced to 27.01 wt% over
mixture of HZSM-5 and dealuminated HY (70 : 30 wt%).
over this catalyst compared to HY. Therefore in the mixture of Acknowledgements
HZSM-5 and dealuminated HY, the amount of dealuminated
This work was conducted with the nancial support of the
Faculty of Engineering at University of Malaya through the HIR
Grant (D000011-16001).
HY should be sufficient for initial cracking of pyrolysis-derived
compounds and effective reduction of thermal coke deposited
on HZSM-5. Besides, the amount of HZSM-5 in the mixture
should be high enough for efficient conversion of reactants. In
the physically mixed catalyst system studied in this work, the
lowest coke formation and the highest yield of aromatic
hydrocarbons were observed at HZSM-5 to dealuminated HY
ratios of 30 : 70 and 70 : 30 wt%, respectively.
References
1 C. Perego and A. Bosetti, Microporous Mesoporous Mater.,
2011, 144, 28–39.
It could be concluded from this study that formation of coke
as an undesired product is a competing reaction with
2 K. C. Kwon, H. Mayeld, T. Marolla, B. Nichols and
M. Mashburn, Renewable Energy, 2011, 36, 907–915.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 65408–65414 | 65413

View Article Online
RSC Advances
Paper
3
4
5
6
7
8
9
J. C. Serrano-Ruiz and J. A. Dumesic, Energy Environ. Sci., 17 A. Aho, N. Kumar, K. Er ¨a nen, T. Salmi, M. Hupa and
011, 4, 83–99. D. Y. Murzin, Fuel, 2008, 87, 2493–2501.
Q. Zhang, J. Chang, T. Wang and Y. Xu, Energy Fuels, 2006, 18 P. T. Williams and P. A. Horne, J. Anal. Appl. Pyrolysis, 1995,
0, 2717–2720. 31, 39–61.
Y. Zhang, T. R. Brown, G. Hu and R. C. Brown, Chem. Eng. J., 19 J. Jae, G. A. Tompsett, A. J. Foster, K. D. Hammond,
2
2
2
013, 225, 895–904.
S. M. Auerbach, R. F. Lobo and G. W. Huber, J. Catal.,
2011, 279, 257–268.
20 Z. Ma, E. Troussard and J. A. van Bokhoven, Appl. Catal., A,
2012, 423–424, 130–136.
21 H. Zhang, Y.-T. Cheng, T. P. Vispute, R. Xiao and
G. W. Huber, Energy Environ. Sci., 2011, 4, 2297–2307.
22 B. Valle, P. Casta n˜ o, M. Olazar, J. Bilbao and A. G. Gayubo,
J. Catal., 2012, 285, 304–314.
23 P. M. Mortensen, J. D. Grunwaldt, P. A. Jensen,
K. G. Knudsen and A. D. Jensen, Appl. Catal., A, 2011, 407,
1–19.
W. Yu, Y. Tang, L. Mo, P. Chen, H. Lou and X. Zheng,
Bioresour. Technol., 2011, 102, 8241–8246.
M. A. Peralta, T. Sooknoi, T. Danuthai and D. E. Resasco,
J. Mol. Catal. A: Chem., 2009, 312, 78–86.
M. Song, Z. Zhong and J. Dai, J. Anal. Appl. Pyrolysis, 2010, 89,
1
66–170.
H. Zhang, R. Xiao, H. Huang and G. Xiao, Bioresour. Technol.,
009, 100, 1428–1434.
2
1
1
1
1
1
1
1
0 P. S. Rezaei, H. Shafaghat and W. M. A. W. Daud, Appl. Catal.,
A, 2014, 469, 490–511.
1 A. G. Gayubo, B. Valle, A. T. Aguayo, M. Olazar and J. Bilbao, 24 M. Ib ´a n˜ ez, B. Valle, J. Bilbao, A. G. Gayubo and P. Casta n˜ o,
J. Chem. Technol. Biotechnol., 2010, 85, 132–144. Catal. Today, 2012, 195, 106–113.
2 T. R. Carlson, T. P. Vispute and G. W. Huber, ChemSusChem, 25 Z. Du, X. Ma, Y. Li, P. Chen, Y. Liu, X. Lin, H. Lei and
008, 1, 397–400. R. Ruan, Bioresour. Technol., 2013, 139, 397–401.
3 A. G. Gayubo, B. Valle, A. T. Aguayo, M. Olazar and J. Bilbao, 26 B. Zhang, Z. Zhong, M. Min, K. Ding, Q. Xie and R. Ruan,
Ind. Eng. Chem. Res., 2010, 49, 123–131. Bioresour. Technol., 2015, 189, 30–35.
4 A. G. Gayubo, B. Valle, A. T. Aguayo, M. Olazar and J. Bilbao, 27 H. Xue, X. Huang, E. Zhan, M. Ma and W. Shen, Catal.
Energy Fuels, 2009, 23, 4129–4136. Commun., 2013, 37, 75–79.
5 A. G. Gayubo, A. T. Aguayo, A. Atutxa, R. Prieto and J. Bilbao, 28 A. Corma, G. Huber, L. Sauvanaud and P. Oconnor, J. Catal.,
Energy Fuels, 2004, 18, 1640–1647. 2007, 247, 307–327.
6 A. Aho, N. Kumar, A. V. Lashkul, K. Er ¨a nen, M. Ziolek, 29 T. R. Carlson, G. A. Tompsett, W. C. Conner and
2
P. Decyk, T. Salmi, B. Holmbom, M. Hupa and
D. Y. Murzin, Fuel, 2010, 89, 1992–2000.
G. W. Huber, Top. Catal., 2009, 52, 241–252.
30 Y. T. Cheng, J. Jae, J. Shi, W. Fan and G. W. Huber, Angew.
Chem., Int. Ed., 2012, 51, 1387–1390.
65414 | RSC Adv., 2015, 5, 65408–65414
This journal is © The Royal Society of Chemistry 2015