Microporous and Mesoporous Materials 166 (2013) 161166

Contents lists available at SciVerse ScienceDirect

Microporous and Mesoporous Materials

journal homepage: www.elsevier.com/locate/micromeso

BEA zeolite nanocrystals dispersed over alumina for n-hexadecane

hydroisomerization
N. Batalha a,b, S. Morisset a, L. Pinard a,, I. Maupin a, J.L. Lemberton a, F. Lemos b, Y. Pouilloux a
a

Laboratoire de Catalyse en Chimie Organique, UMR 6503 CNRS, Universit de Poitiers, 40 avenue du Recteur Pineau, 86022 Poitiers Cedex, France
Institute for Biotechnology and Bioengineering (IBB), Centre for Biological and Chemical Engineering, Instituto Superior Tcnico, Universidade Tcnica de Lisboa,
Av. Rovisco Pais, 1049-001 Lisboa, Portugal
b

a r t i c l e

i n f o

Article history:
Available online 3 May 2012
Keywords:
BEA aggregation
Diffusion limitations
Nanocrystals
Hydroisomerization
Composite materials

a b s t r a c t
The direct synthesis of BEA nanocrystals was performed on an a-Al2O3 surface, in order to avoid the typical aggregation of the zeolite, which causes diffusion limitations. Two different composite samples, with
different zeolite loadings were synthesized and compared with a pure BEA nanocrystals sample (similar
synthesis gel composition). The characterizations showed that both the BEA zeolite sample and the zeolite nanocrystals on the composite samples had similar textural and acidic properties. However, through
TEM and SEM characterizations it was possible to conrm not only the smaller occurrence of zeolite
aggregates on the composite samples, but also the existence of completely isolated BEA nanocrystals,
indicating the success of the synthesis procedure.
The catalytic test results of n-hexadecane hydroisomerization, at 220 C and 30 bar, showed a positive
effect of the decrease in nanocrystal agglomeration on both activity and isomers selectivity. The composite catalysts were 3.5 times more active than the pure zeolite sample and the maximum isomers yield
increased from 35 to 80 wt.%.
2012 Published by Elsevier Inc.

1. Introduction
The FischerTropsch waxes composed, mainly, of linear parafns must be upgraded for liquid fuel or lubricant base oil applications [1,2]. The catalytic dewaxing process, through skeletal
isomerization of the parafns, reduces the pour points, enhancing
the low temperature properties of diesel and lubricating oils [3].
The skeletal branching of n-alkanes is achieved using bifunctional
catalysts containing both metallic sites, for the dehydrogenation
and hydrogenation reactions, and acidic sites for the isomerization
(and cracking) reactions [3]. Zeolites loaded with platinum or palladium have been widely used for this purpose [4,5]. The olens
formed from the parafns through dehydrogenation on the metallic sites are protonated on the acidic sites into an alkylcarbenium
ion. This ion undergoes skeletal rearrangement and eventually,
cracking through b-scission [6,7]. The later is favored as the
branching degree of the carbon chain increases. Consequently, high
isomers selectivity is directly linked to low cracking rates.
When the hydrogenating function is highly active, the activity
and selectivity will depend on the number, the strength and the
location of the acidic function [810]. Therefore, a decrease in acidity will lessen not only cracking but also the global activity, since
Corresponding author. Tel.: +33 (0)5 49 45 39 05; fax: +33 (0)5 49 45 37 79.
E-mail address: ludovic.pinard@univ-poitiers.fr (L. Pinard).
1387-1811/$ - see front matter 2012 Published by Elsevier Inc.
http://dx.doi.org/10.1016/j.micromeso.2012.04.041

the acid step is rate limiting. The zeolite pore structure can also
modify the activity and the selectivity of the bifunctional catalyst
[1,5,1114]. If the geometry and the dimensions of the zeolites
pores are such that cracking is suppressed by molecular shapeselectivity, the yield in isomerization is improved substantially
[1,5,15]. For instance, PtHZSM-22 (10 MR zeolite, TON framework
structure) is very selective for the production of monobranched
isomers [1517]. Indeed, since the reaction intermediates are too
bulky to be formed within the ZSM-22 channels the reaction only
takes place at the pore entrance (pore mouth catalysis) limiting
the formation of monobranched isomers [1517]. Nonetheless,
the consequence of this small penetration is a rather low activity
of the catalyst [18]. The molecular shape-selectivity obtained with
the ZSM-22 zeolite is lost on similar pore size zeolites with two- or
three-dimensional pore systems, i.e. ZSM-5 and MCM-22. On these
supports the reaction occurs inside the zeolite channels resulting
in high cracking yields. Indeed, the reaction intermediates are
blocked inside the channels where they crack after multi-isomerization steps [19]. In order to avoid this phenomenon, Christensen
et al. introduced, through disilication, a complementary mesoporous system in ZSM-5 zeolites reducing the isomers residence time
inside the zeolite framework and, consequently, limiting cracking
[20]. As a result, better acid site accessibility and rapid isomers
desorption outside the zeolite crystal improved both hydroisomerization activity and selectivity.

162

N. Batalha et al. / Microporous and Mesoporous Materials 166 (2013) 161166

Soualah et al. have shown that PtHBEA (12MR, BEA framework

structure) was more active and more selective for the hydroisomerization of long n-alkanes (n-C14, n-C16) than catalysts based
on MCM-22 or ZSM-5 zeolites [19]. This was explained by a rapid
diffusion of the reactant and reaction intermediates inside the
large channels of the BEA zeolite. Yet, the PtHBEA catalyst yielded
signicant amounts of cracked products due to rather strong protonic acidity. Thus, decreasing the zeolite acidity would decrease
the cracking reactions rate. This was conrmed by Merabti et al.,
using partially sodium exchanged BEA zeolites, which showed an
isomerization selectivity similar to that of ZSM-22 [18]. Nonetheless, this was achieved with a 75% loss of the BEA zeolite activity,
1
even if the PtNaHBEA (0:7 gn-C16 g1
cat g ) catalyst remained two
1
times more active than PtHZSM-22 (0:3 gn-C16 g1
cat g ).
Like for the ZSM-5 zeolite, the reduction of the isomers residence time inside the BEA zeolite crystal would positively enhance
the selectivity towards isomerization. Moreover, this approach
would not cause an activity loss. Zeolite nanocrystals are often
used for this purpose [2124]. However, the BEA nanocrystallites
aggregate into micrometric clusters, more than 1 lm, creating an
intercrystalline disordered porous system, which alters the molecular diffusion [25,26]. Lima et al. showed that the diffusivity values
of n-alkanes obtained from zeolite BEA nanocrystals were extremely low when compared to those of zeolites with similar pore
aperture, e.g. FAU [26]. The authors suggested a possible inuence
of extracrystallite diffusion. Thus, in order to improve the catalytic
properties, it is essential to avoid the BEA crystallite agglomeration
in order to minimize the isomers residence time inside the zeolite
catalyst. This can be obtained by the direct BEA zeolite germination
on a support, such as alumina [27]. Lovallo et al. have shown that
in the case of nanocrystal zeolites (<100 nm), the matrix impedes
growth and aggregation [28].
The purpose of this work was to study the structural and the morphological properties of a zeolitealumina composite obtained by
mixing a zeolite synthesis gel with alumina particles. n-Hexadecane
hydroisomerization was chosen as the model reaction to check if a
self-organized zeolite nanocrystal allowed to obtain a high activity
together with a high selectivity into isomers. For this purpose, the
catalytic performance of a BEA nanocrystalline zeolite was compared
with those of composite materials with different zeolite loadings.
2. Experimental
2.1. BEA coated alumina synthesis
Particles of a-Al2O3 were used as support for the BEA nanocrystals germination, due to their high stability. They were obtained
from extrudates supplied by PREMICAT (University of Poitiers),
which were crushed and sieved to 0.20.4 mm particles (particle
size required for the catalytic testing).
The supported zeolite synthesis was carried out using the following precursor gel molar composition: 23.6SiO2:1.0Al2O3:1.9NaO2:1.9TEA2O:235H2O. The alumina particles were then mixed
with the gel in a PTFE lined autoclave and the synthesis was performed under hydrothermal conditions at 155 C during 7 days
without stirring. Two samples were synthesized with a weight ratio between the synthesis gel and alumina of 5/1 and 5/3. Consequently, the two resulting samples were called HL (ratio 5/1) and
LL (ratio 5/3), respectively, for high loading and low loading of
BEA zeolite over the alumina support.
In addition, a synthesis without alumina particles was also done,
in order to be used as reference (noted BEA). After crystallization,
the samples were ltered and rinsed several times with distillated
water. The resulting solids were subsequently dried at 90 C overnight. Then, the organic template was removed through calcination
at 550 C for 12 h under air ow. The zeolites ammonium forms

were prepared by ion exchange of the raw materials with a 2 M

ammonium nitrate solution, 50 mL g1 of catalyst, under reux
during 1 h at 80 C. This procedure was performed twice. The protonic form of the materials was obtained by calcination in a mufe
furnace at 450 C. The alumina particles coated with zeolite were
recovered by sieving. In order to remove the weakly anchored
zeolite particles, the materials were sonicated in distillated water
during 10 min. Then these wet materials were dried at 90 C and
re-sieved to obtain 0.20.4 mm particles.
Platinum was introduced into the samples (BEA, LL and HL)
through ion exchange by [Pt(NH3)4]2+ in competition with NH4+
(NH4+/Pt = 100), followed by calcination under a dry air ow at
450 C for 4 h. The amount of platinum in solution was calculated
to reach 1 wt.% on the bifunctional catalyst.
2.2. Physicochemical characterization
The structural characterization of the zeolite and composite
samples was carried out by X-ray powder diffraction (XRD). The
XRD patterns were obtained on a D5005 BRUKER
AXS diffractom0
eter using a Cu-Ka radiation (k = 1.5406
A) as incident beam. The
zeolite crystal sizes were calculated through the Scherrer equation.
Nitrogen adsorption and desorption measurements were carried out at 196 C over a Micromeritics ASAP 2010 apparatus.
Prior to analysis, the samples were pretreated at 350 C under vacuum for 8 h. The specic surface area was determined by the BET
method [29] and the micropore volume (pore diameter < 2 nm)
was estimated through the t-plot method applied to a layer thickness between 5 to 7 [30,31] and the DubininRaduskevitch
method [32], which enabled to distinguish between ultra-micropores (pore diameter < 0.8 nm) and super-micropores (pore diameter between 0.8 and 2 nm). Consequently, the mesopore volume
was determined by the difference between the total pore volume,
determined at 0.97 of relative pressure, and the micropore volume.
The zeolite framework aluminum (AlFram) content was determined by infrared spectroscopy measurements on a FT-IR Magna
550 Nicolet spectrometer. The position of the zeolite structure
bands (4501250 cm1) and especially that of the asymmetric
stretching vibration (mTOT) at 10801200 cm1 allowed to calculate the AlFram from the correlation given in literature [33]. The
TOT bands were determined using KBr wafers containing 2 wt.%
of sample. The acidity of the samples was measured by infrared
spectroscopy of adsorbed pyridine. The concentration of the
Brnsted and the Lewis acid sites were calculated from the integrated area, after pyridine adsorption at 150 C, of the PyH+ and
PyL bands at 1545 and 1450 cm1, respectively. The experimental
methods were described in details in a previous paper [34].
The catalyst morphology was observed through scanning electron microscopy (SEM) performed on a FEG-SEMJEOL microscope
(JSM 5600-LV model). The mapping of the silicium element on
the coated catalysts was performed on an energy dispersive Xray (EDX) scanning Brucker apparatus.
The bifunctional catalysts were characterized by transmission
electronic microscopy (TEM) using a Philips CM 120 microscope
equipped with a LaB6 lament. To prepare the sample, a small drop
of a zeolite suspension in ethanol was put on an Au grid and the
solvent was evaporated. TEM observations allowed determining
the zeolite crystallite size and the platinum particles average size,
as well as the metal location on the zeolite and on the a-Al2O3. The
platinum size determination was done on the average of 500 particles, assuming that metallic particles were sphere-like.
2.3. Hydroisomerization reaction
The transformation of n-C16 (Aldrich, >99.9% purity) was carried
out in a xed-bed stainless steel reactor under the following

N. Batalha et al. / Microporous and Mesoporous Materials 166 (2013) 161166

conditions: temperature 220 C; total pressure 30 bar; H2/n-C16

molar ratio 20; and WHSV (weight hourly space velocity) between
2 and 100 h1. WHSV was changed by modifying the reactant ow
rates in order to obtain different conversion values. The reaction
products were analyzed online by GC equipped with a 50 m
CPSil-5 capillary column from Chrompack, with hydrogen as carrier gas (13 psi) and a FID detector. To avoid product condensation
problems it was necessary to dilute n-C16 in an inert solvent. nHexane (Aldrich, >99.9% purity) was chosen as solvent and the feed
composition was 10 mol.% n-C16 and 90 mol.% n-C6. During reaction, none of the catalysts presented deactivation. Before reaction
all the catalysts were reduced under hydrogen ow at 450 C for
6 h.
3. Results and discussion
The samples resulting from the direct synthesis of BEA nanocrystals over an a-Al2O3 surface (HL and LL catalysts) will be compared to a pure BEA catalyst in terms of textural, morphological
and acidic properties. Likewise, the catalytic performances of these
samples will be evaluated through the n-C16 hydroisomerization.
3.1. Textural and morphological properties
The X-ray diffractograms (XRD) of a-Al2O3, BEA, HL and LL catalysts are shown on Fig. 1. The a-Al2O3 and BEA samples present
the typical patterns of the corresponding phases. On the other
hand, the HL and LL catalyst XRD patterns showed the presence
of both a-Al2O3 and BEA phases. No traces of other compounds,
such as c- or h-Al2O3, were observed. Concerning the BEA zeolite
phase, a broad peak in the range of 6.58.5 2h indicated the presence of the two isomorphs. The BEA zeolite crystal sizes were calculated through the Scherrer equation using the most intense peak,
22.4 2h (Miller index 3 0 2). The BEA crystallites changed from
30 nm on the BEA to 43 nm on both composite catalysts. Finally,
through the catalysts XRD patterns it was possible to estimate
the zeolite content on the coated catalysts, by using the X-ray mass
attenuation coefcient of the pure BEA zeolite and of a-Al2O3. The
zeolite content on the LL and the HL samples was, respectively, 13
and 40 wt.%. The zeolite loading observed on the HL was three
times higher than on the LL and, thus, in accordance with the proportion of alumina introduced upon zeolite synthesis, e.g. three
times lower for the HL sample.
The textural properties of the samples deriving from the N2
adsorption are reported in Table 1. The isotherms of all the samples
were quite similar to those of purely microporous materials, with a
plateau from very low relative pressure values (Fig. 2). The mesopore volume created by the intercrystalline voids was quite low,
only 0.08 cm3 g1, which indicated that the nanocrystallites were
strongly agglomerated and that the ultra-micropore volume was

-Al2O3

LL

HL

BEA

15

25
2

35

Fig. 1. XDR patterns of the samples.

45

163

equal to the micropore volume. On the other hand, the N2 adsorption results indicated that there was no signicant change in the
zeolite and alumina textural properties between the pure and composite samples, since when considering a zeolite basis the HL and
LL samples presented characteristics similar to those of the pure
zeolite. Likewise, the pore size distribution of the BEA, HL and LL
samples were similar (Fig. 3). Thus, the catalysts textural properties obtained through N2 adsorption conrmed the assumption
made for the zeolite composition estimation, which considered
that there was no change in the zeolite nor in the alumina structures between the pure and composite samples, which therefore,
supported the composite catalysts zeolite compositions.
Scanning electronic microscopy (SEM) showed differences between the samples textures. As expected according to literature
[26], the SEM image of the bare BEA zeolite (Fig. 4A) showed a
high agglomeration of the zeolite crystals which, despite of
having a nanometric size, formed 2.4 lm average size spherical
clusters. On the other hand, on the HL sample, which contained
40 wt.% zeolite, the alumina surface was completely covered
with agglomerated BEA crystallites (Fig. 4B). By opposition, on
the LL sample (13 wt.% of zeolite) only a few BEA clusters could
be observed (Fig. 4C). Nevertheless, the alumina surface was
fully covered by zeolite crystals, as shown by the Si element
mapping performed on the catalyst surface (Fig. 5), suggesting
the presence of BEA nanocrystals attached to the alumina
surface.
Fig. 6 shows the TEM micrographs of the two composite samples and of the pure BEA sample. The later consisted on BEA nanocrystallites, which, as expected, were strongly agglomerated
(Fig. 6A). Moreover, the strong aggregation was already suggested
by the low mesopore volume measured through N2 adsorption.
Likewise, the nanometric size of the BEA crystals was in accordance
with the 30 nm determined by XRD. On the composite samples, the
occurrence of isolated nanocrystals could be observed. However, it
was more frequent on the LL sample. The crystallite shapes were
cubic, and the crystal sizes were similar to those determined by
XRD (43 nm). Moreover, lattice fringes and sharp edges can be seen
in Fig. 6B and C, indicating a high degree of crystallinity. The TEM
images also showed that the alumina matrix in contact with the
zeolite crystal suffered a restructuration in order to match the zeolite pattern. This suggests not only the alumina partial degradation
by the zeolite synthesis gel, but also explains why the zeolite crystals remained attached to the alumina surface after the ultrasound
treatment.
3.2. Acidic and metallic properties
The three samples were also characterized by FT-IR spectroscopy, in order to obtain further insight on their acidic and structural properties (Table 2). The framework Si/Al ratios, estimated
from the zeolite structure bands (4501250 cm1) [33], were
rather low for a BEA zeolite, e.g. 1113 (Table 2). In the OH region
of IR spectra, the BEA zeolite presented four bands (Fig. 7). The
most intense band, at 3740 cm1, corresponded to the external
and internal silanol groups [35] and was, consequently, in agreement with the small crystallite size of the zeolite (30 nm). The
three other less intense bands were ascribed, respectively, to bridging OH groups (3608 cm1) [35], to hydroxylated monomeric and
polymeric extraframework Al (EFAL) species (3663 cm1) [35,36]
and to hydroxylated monomeric EFAL species and framework defects with Lewis acidity (3782 cm1) [37]. These bands were also
present both on the HL and LL catalysts (Fig. 7), and their intensities (normalized at the zeolite content) were close to those of the
zeolite alone. Nonetheless, the appearance of one additional band
at 3510 cm1 (Fig. 7) was observed on the composite samples. This
band, not typical of the BEA zeolite or any alumina phase, was

164

N. Batalha et al. / Microporous and Mesoporous Materials 166 (2013) 161166

Table 1
Textural properties of BEA, a-Al2O3 and composite catalysts.
Catalyst

Zeolite loadinga (wt.%)

Crystal sizeb (nm)

SBET (m2/gcat)

Sext (m2/gcat)

Vmesopore (cm3/gcat)

Vmicroc (cm3/gcat)

VUltra-microd (cm3/gcat)

BEA
HL
LL
a-Al2O3

100
40
13
0

30
43
43
94

671
270 (666*)
92 (663*)
6

56
21
8
5

0.08
0.03
0.01
0.01

0.26
0.11 (0.27*)
0.04 (0.28*)
0

0.262
0.11 (0.26*)
0.04 (0.27*)
0

Estimated from the XRD pattern.

Calculated through the Scherrer equation.
Obtained from the DubininRaduskevitch method.
d
Obtained from the t-plot method.
per gzeolite.
b

400
BEA
HL
LL

350

Vads (cm 3.g-1)

300
250
200
150
100
50
0
0

0.2

0.4
0.6
Relative pressure

0.8

Fig. 2. Nitrogen adsorption isotherms of the samples.

Pore volume (cm3.g-1)

0.02
BEA
HL
LL

0.015

0.01

0.005

0
0

4
6
Pore radius (nm)

10

Fig. 3. Pore size distribution of the samples.

ascribed to a hydroxyl group formed on alumina during synthesis.

However, the acidic properties of this hydroxyl were weak since no
interaction with pyridine was observed.
On the three catalysts, the number of Brnsted and Lewis acid
sites was determined by pyridine desorption at 150 C. When considering the composite catalysts on a zeolite content basis, the
Brnsted sites density was in the same range as that of the BEA
zeolite, around 600 lmol g1. These rather high values are in
agreement with the low Si/Al ratios observed on the samples. On
the other hand, the number of Lewis acid sites, on a zeolite basis,
was 3.5 higher on the LL and HL catalysts than on BEA zeolite. This
suggests that during the zeolite maturation the high crystalline
alumina was partially decomposed by the high pH synthesis gel
solution (pH = 13), which led to the formation of both Lewis acid
sites and the new hydroxyls group, observed on the IR spectra.

Fig. 4. SEM images of the samples: (A) BEA zeolite; (B) HL; (C) LL.

On the HL and LL catalysts, the platinum introduced by ion exchange was located both on the alumina support and on the zeolite, even if, the TEM micrographs showed that the major part of
metal was located on the zeolite. The average platinum particle
size, however, was similar on all the samples (Table 2).
In conclusion, the acidic and the metallic properties were similar on the HL, LL and BEA bifunctional catalysts. Consequently, the
main difference between the samples concerned the agglomeration of the zeolite nanocrystals: strong on the BEA, absent on the
LL catalyst.

165

N. Batalha et al. / Microporous and Mesoporous Materials 166 (2013) 161166

BEA

HL

LL
-Al2O3

3800

3750

3700

3650
3600
3550
Wavenumber (cm-1)

3500

3450

Fig. 7. IR spectra of the samples, in the hydroxyl band range (3800 to 3450 cm1).

Fig. 5. EDX mapping of the Si element on the LL sample surface.

Fig. 6. TEM images of the samples: (A) BEA zeolite (30,000); (B) HL (120,000);
(C) LL (200,000).

Table 2
Catalyst acidity and platinum particle size.
Si/
AlFrama

Acidityb (lmol/gcat)
Brnsted

Lewis

BEA
HL

11
13

LL

11

604
229
(579*)
86
(658*)

286
421
(1052*)
121
(931*)

Catalyst

Platinum particle average sizec

(nm)
3.1
2.9
2.6

a
Calculated from the mTOT band at 10801200 cm1 by using the correlation
given in literature [33].
b
Drawn from pyridine adsorption at 150 C followed by FT-IR.
c
Drawn from transmission electron microscopy.
*
per gzeolite.

3.3. n-Hexadecane hydroisomerization

n-Hexadecane hydroisomerization was carried out on the three
bifunctional catalysts at 220 C and 30 bar. In all cases, the n-hexadecane molecules were transformed into isomers (M mono-

branched isomers; B multibranched isomers) and cracking

products. Moreover, all the catalysts were stable, since no deactivation was observed under the reaction conditions that were used.
The absence of deactivation indicates the presence of a strong
hydrogenating function, which avoids the transformation of the
olenic intermediates into coke. Consequently, the catalyst activity
is directly connected to the density of the acid sites, since the acid
support is similar on all catalysts [18]. Thus, as the composite samples possess similar BEA zeolite acidity (around 600 lmol gzeo1),
the catalysts activity should be proportional to the zeolite content
on the catalysts. Indeed, the LL catalyst (13 wt.% zeolite) was 3.1
times less active than HL catalyst (40 wt.% zeolite) (Table 3). The
pure zeolite catalyst, however, did not follow the tendency and
was 50% less active than the HL catalyst (Table 3). The turnover frequency (TOF), i.e. the activity per acid site, of the composite catalysts was, as expected, similar (HL-83 h1; LL-72 h1) and high in
comparison with the agglomerated BEA catalyst (22 h1). The relative low activity and TOF of the BEA catalyst probably results from
the BEA nanocrystallite (30 nm) aggregation into clusters (2.4 lm),
which could provoke reactant diffusion limitations. Consequently,
it is possible to conclude that even though the BEA catalyst is composed of nanocrystals, the crystallites aggregation phenomenon
approached the catalyst performance to that of a higher crystal size
BEA zeolite.
Fig. 8 clearly shows a signicant improvement in the isomers
yield when using the composite catalysts. The maximum isomers
yield improved from 35 to 80 wt.% when comparing the BEA zeolite with the LL catalyst. Meanwhile, the HL sample presented an
intermediate behavior with a maximum isomers yield of 65 wt.%.
Moreover, the M/B molar ratio was lower on the BEA catalyst than
on the LL catalyst, and slightly higher on the LL catalyst than on the
HL catalyst (Fig. 9). These results indicate that, when the crystallites were isolated from each other as on the LL catalyst, the residence time of the monobranched isomer inside the crystallite
was reduced, limiting their transformation and increasing the isomer yield. On the other hand, when the nanocrystallite were
agglomerated, as was the case on the BEA catalyst, the residence
time of the monobranched isomers was longer, which favored
the formation of multibranched isomers and consequently cracking. Similar results were obtained by Win et al. for the benzoylation of anisol [38], where BEA zeolite supported on silicon carbide
exhibited high catalytic activity. Moreover, the composite catalyst
was extremely stable in comparison with the BEA zeolite. The
authors have justied the results by the improved product escaping rate, which reduced the secondary acid catalyzed condensation
reactions, responsible for the carbonaceous species formation inside the zeolite porosity.
Finally, the reduction of the zeolite agglomeration by performing the direct germination of a BEA zeolite over an alumina surface,

166

N. Batalha et al. / Microporous and Mesoporous Materials 166 (2013) 161166

Table 3
Catalyst activity.

Catalyst

Activity gn-C16 g1

cat h

TOFa (h1)

BEA
HL
LL

3.0
4.4
1.4

22
85
72

1

Turnover frequency activity per Brnsted acid site.

the zeolitic phase. By isolating the nanocrystals, the isomers residence time inside the zeolite framework is reduced, thus avoiding
secondary reactions, including cracking and, consequently increasing the yield in monobranched isomers. In conclusion, the reduction in the nanocrystal agglomeration enhanced the diffusion
properties of the BEA zeolite, resulting in high reaction activity
and selectivity.
Acknowledgement

Isomerization yield (wt.%)

100

Nuno Batalha thanks Fundao para a Cincia e a Tecnologia

(FCT) for the nancial support (Ref. SFRH/BD/43551/2008).

BEA
LL
HL

80

References
60
40
20
0
0

20

40
60
Conversion (%)

80

100

Fig. 8. Isomerization yield as a function of n-hexadecane conversion.

1.5
BEA
LL
HL

M/B

0.5

0
0

20

40
60
Conversion (%)

80

100

Fig. 9. Monobranched (M)/multibranched (B) ratio as a function of n-hexadecane

conversion.

not only enhanced the catalyst activity, but also enabled to obtain a
substantial increase in the isomerization selectivity, despite the
rather high acidity of the BEA zeolite.
4. Conclusion
The direct germination of BEA zeolite nanocrystallites on an aAl2O3 surface prevented their agglomeration without signicantly
changing the acidic and textural properties of the zeolite, as well
as the size of the metallic particles deposited. The composite catalysts were more active and more selective for isomerization than
the BEA sample. These results were attributed to the BEA germination of the zeolite on the a-Al2O3, which increased the ability of the
reactant and products to diffuse in and out of the porous system of

[1] D.A. Bell, B.F. Towler, M. Fan, Coal Gasication and its Applications, rst ed.,
Elsevier, 2011.
[2] V. Calemma, C. Gambaro, W.O. Parker Jr., R. Carbone, R. Giardino, P. Scorletti,
Catal. Today 149 (2010) 4046.
[3] C. Bouchy, G. Hastoy, E. Guillon, J.A. Martens, Oil Gas Sci. Technol. Rev. IFP 64
(2009) 91112.
[4] A. Corma, Catal. Lett. 22 (1993) 3352.
[5] H. Deldari, Appl. Catal. A 293 (2005) 110.
[6] P.B. Weisz, Science 123 (1956) 887888.
[7] P.B. Weisz, E.W. Swegler, Science 126 (1957) 3132.
[8] F. Alvarez, F.R. Ribeiro, G. Perot, C. Thomazeau, M. Guisnet, J. Catal. 162 (1996)
179189.
[9] M. Guisnet, F. Alvarez, G. Gianneto, G. Perot, Catal. Today 1 (1987) 415433.
[10] F. Alvarez, G. Gianneto, M. Guisnet, G. Perot, Appl. Catal. 34 (1987) 353365.
[11] G. Kinger, H. Vinek, Appl. Catal. A 218 (2001) 139149.
[12] P. Raybaud, A. Patrigeon, H. Toulhoat, J. Catal. 197 (2001) 98112.
[13] A. Lugstein, A. Jentys, H. Vinek, Appl. Catal. A 176 (1999) 119128.
[14] J.K. Lee, H.K. Rhee, Korean J. Chem. Eng. 14 (1997) 451458.
[15] M.C. Claude, J.A. Martens, J. Catal. 190 (2000) 3948.
[16] W. Souverijns, J.A. Martens, G.F. Froment, P.A. Jacobs, J. Catal. 174 (1998) 177.
[17] J.A. Martens, G. Vanbutsele, P.A. Jacobs, J. Denayer, R. Ocakoglu, G. Baron, J.A.
Muoz Arroyo, J. Thybaut, G.B. Marin, Catal. Today 65 (2001) 111116.
[18] R. Merabti, L. Pinard, J.L. Lemberton, P. Magnoux, A. Barama, K. Moljord, React.
Kinet. Mech. Cat. 100 (2010) 19.
[19] A. Soualah, J.L. Lemberton, L. Pinard, M. Chater, P. Magnoux, K. Moljord, Appl.
Catal. A 336 (2008) 2328.
[20] C.H. Christensen, I. Schmidt, C.H. Christensen, Catal. Commun. 5 (2004) 543
546.
[21] A. Corma, J. Catal. 216 (2003) 298312.
[22] M.V. Landau, L. Vradman, V. Valtchev, J. Lezervant, E. Liubich, M. Talianker, Ind.
Eng. Chem. Res. 42 (2003) 27732782.
[23] S.C. Larsen, J. Phys. Chem. C 111 (2007) 1846418474.
[24] J. Prez-Ramrez, C.H. Christensen, K. Egeblad, C.H. Christensen, J.C. Groen
Chem, Soc. Rev. 37 (2008) 25302542.
[25] M.V. Landau, D. Tavor, O. Regev, M.L. Kaliya, M. Herskowitz, V. Valtchev, S.
Mintova, Chem. Mater. 11 (1999) 20302037.
[26] P.L. Lima, C.V. Gonalves, C.L. Cavalcante Jr., D. Cardoso, Micropor. Mesopor.
Mater. 116 (2008) 352357.
[27] N. van der Puil, F.M. Dautzenberg, H. van Bekkum, J.C. Jasen, Micropor.
Mesopor. Mater. 27 (1999) 95106.
[28] M.C. Lovallo, M. Tsapatsis, in: Advanced Catalysis and Nanostructured
Materials: Modern Synthesis Method, Academic Press, New York, 1996.
[29] S. Brunauer, P.H. Emmett, E. Teller, J. Am. Chem. Soc. 60 (1938) 309319.
[30] B.C. Lippens, J.H. de Boer, J. Catal. 4 (1965) 319323.
[31] J.H. de Boer, B.C. Lippens, B.G. Lisen, B.C.P. Broekhoff, A. van den Heuvel, T.J.
Osinga, J. Interface Sci. 21 (1966) 405414.
[32] J. Lynch, F. Raatz, P. Dufresne, Zeolites 7 (1987) 333340.
[33] C. Coutanceau, J.M. Da Silva, M.F. Alvarez, F.R. Ribeiro, M. Guisnet, J. Chim.
Phys. 94 (1997) 765781.
[34] D. Meloni, S. Laforge, D. Martin, M. Guisnet, E. Rombi, V. Solinas, Appl. Catal. A
215 (2001) 5566.
[35] M. Maache, A. Janin, J.C. Lavalley, J.F. Joly, E. Benazzi, Zeolites 13 (1993) 419
426.
[36] E. Loefer, U. Lohse, C. Peuker, G. Oehlmann, L.M. Kustov, V.L. Zholobenko, V.B.
Kazansky, Zeolites 10 (1990) 266271.
[37] A. Vimont, F. Thibault-Starzyk, J.C. Lavalley, J. Phys. Chem. B 104 (2000) 286
291.
[38] G. Win, Z. El Benichi, C. Pan-Huu, J. Mol. Catal. A 278 (2007) 6471.