Impact of cationic surfactant chain length during SAPO-11 molecular sieve synthesis on structure, acidity, and n-octane isomerization to di-methyl hexanes

Article abstract of DOI:10.1016/j.jcat.2012.07.016

The structure, acidity, and n-octane di-branched isomerization of SAPO-11 synthesized with surfactants of different chain lengths in water-propanol system were investigated. The results showed that with the increasing chain length of surfactants, the crys

Full text of DOI:10.1016/j.jcat.2012.07.016

Journal of Catalysis 294 (2012) 161–170
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Impact of cationic surfactant chain length during SAPO-11 molecular sieve
synthesis on structure, acidity, and n-octane isomerization to di-methyl hexanes
b
b
c
Lin Guo a,b, Xiaojun Bao a,b, Yu Fan a,b, , Gang Shi , Haiyan Liu , Danjiang Bai
⇑
a State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Beijing 102249, PR China
b The Key Laboratory of Catalysis, China National Petroleum Corporation (CNPC), China University of Petroleum, Beijing 102249, PR China
c Technology Department, Jinao (Hubei) Science & Technology Chemical Industry Co. LTD., PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
The structure, acidity, and n-octane di-branched isomerization of SAPO-11 synthesized with surfactants
of different chain lengths in water–propanol system were investigated. The results showed that with the
increasing chain length of surfactants, the crystallite size of SAPO-11 firstly decreased and then increased,
but the external surface area and the active B acid sites of SAPO-11 firstly increased and then decreased.
The smallest crystallite, the highest external surface area, and the most B acid sites were obtained simul-
taneously with the surfactant of 12 carbon atoms. The isomerization results of n-octane indicated that
among the series Pt/SAPO-11 catalysts, the catalyst with DoTAB-directed SAPO-11 as support had the
highest selectivity to di-branched isomers and the lowest cracking selectivity due to the largest external
surface area that created more active sites to promote the formation of di-branched isomers by the key-
lock mechanism and improve the diffusion of intermediates and products for reducing cracking reactions.
Ó 2012 Elsevier Inc. All rights reserved.
Received 30 March 2012
Revised 25 June 2012
Accepted 17 July 2012
Available online 20 August 2012
Keywords:
SAPO-11
Cationic surfactants
Chain length
External surface area
Di-branched isomerization
1. Introduction
contribute to the formation of di-branched isomers and the number
of pore mouths is closely related to the external surface area of
Hydroisomerization of n-alkanes to their branched isomersis one
of the important processes in the modern petroleum refining indus-
try [1] that has been widely applied in producing high-octane gaso-
line [2], reducing the pour point of diesel, and improving the
viscosity–temperature property of lubricant basestocks [3,4]. In par-
ticular, with the ever-increasing demand for high-octane gasoline
due to the rapid development of the automobile industry, searching
for catalysts to isomerize C5–C8 n-alkanes of low octane rating into
multi-branched isomers of high octane rating has been the subject
of many investigations around the world [5,6]. Among the numerous
catalytic materials explored for n-alkane isomerization, SAPO-11
molecular sieve as an acidic carrier of bifunctional catalysts has at-
tracted much attention due to its outstanding isomerization perfor-
mance [7–9]. As a member of the silicoaluminophosphate molecular
sieve family, SAPO-11 with a one-dimensional 10-membered-ring
channel system of 0.39 nm ꢀ 0.63 nm [10] has shown its advantages
in n-alkane isomerization [11,12] due to its distinct pore structure
and mild acidity [13–15]. However, it is also noted that the channels
of conventional SAPO-11 can only produce mono-branched isomers
[16–18] but hardly allow the formation of di-branched isomers with
higher research octane numbers (RONs) [19]. Sinha et al. [20] has
pointed out that the active sites near the pore mouths of SAPO-11
SAPO-11. Recently, in the presence of hexadecyltrimethyl ammo-
nium bromide (CTAB) and alcohol, Franco et al. [21] obtained a SAPO
molecular sieve with larger external surface area and more acid sites
compared with those of conventional SAPO-11, similar to the results
obtained by Blasco et al. [22] and Huang et al. [23], demonstrating
the enhancing effect of surfactant introduction into the water–
alcoholsynthesissystemontheexternalsurfacearea of theresultant
SAPO-11. However, the effects of chain length of surfactants on the
structure and acidity of resulting SAPO-11 molecular sieve as well
as its di-branched isomerization performance have not been
reported to our best knowledge, to which this work is addressed.
Herein, a series of SAPO-11 were hydrothermally synthesized
with cationic surfactants of different chain lengths in the water–
propanol system. The chain length of surfactants was correlated
with the structure and acidity of the SAPO-11 synthesized and
the corresponding Pt/SAPO-11 catalysts were assessed with the
di-branched isomerization of n-octane as a model reaction. The
optimal surfactant was determined according to the discussed
relationship between the physicochemical properties and the
isomerization performances.
2. Experimental
2.1. Materials
⇑
Corresponding author at: State Key Laboratory of Heavy Oil Processing, China
University of Petroleum, Beijing 102249, PR China. Fax: +86 (0) 10 89734979.
Tetrapropyloxy silane (C12H28O4Si, Aladdin), pseudoboehmite
(containing 73% Al2O3, Tianjin Hengmeilin Chemical Corporation),
E-mail address: fanyu@cup.edu.cn (Y. Fan).
0021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2012.07.016

162
L. Guo et al. / Journal of Catalysis 294 (2012) 161–170
and ortho-phosphoric acid (85%, Beijing Chemical Reagents Com-
pany) were used as the sources of Si, Al and P, respectively. Di-n-
propylamine (DPA, C6H15N, Beijing Chemical Reagents Company)
was used as the template, and n-propanol (C3H7OH, Beijing Chem-
ical Reagents Company) was used as the co-solvent. The cationic
surfactants used include decyltrimethyl ammonium bromide
(DeTAB, C10H21(CH3)3NBr, Aladdin), dodecyltrimethyl ammonium
bromide (DoTAB, C12H25(CH3)3NBr, Aladdin), tetradecyltrimethyl
ammonium bromide (TTAB, C14H29(CH3)3NBr, Aladdin), and hex-
according to 0.5 wt% Pt loading and dried at 383 K for 2 h and cal-
cined in air at 723 K for 2 h to yield a series of Pt/SAPO-11 catalysts.
2.4. Characterizations
The phase structure of the SAPO-11 samples was characterized
by powder X-ray diffraction (XRD) conducted on a Shimadzu 6000
diffractometer (Kyoto, Japan) that used CuK
a radiation and was
adecyltrimethyl ammonium bromide (CTAB,
C16H33(CH3)3NBr,
operated at 40 kV and 30 mA with 2h in 5–40° and scanning speed
at 5°/min. The morphology and size of the SAPO-11 samples were
determined by means of scanning electron microscopy (SEM) con-
ducted on a Cambridge S-360 instrument (England). Nitrogen
adsorption–desorption measurements were performed on an ASAP
2020 instrument (Micromeritics, USA). The surface area and pore
volume of the SAPO-11 samples were determined according to
the Brunauer–Emmett–Teller (BET) method and the t-plot method,
respectively, and the most probable pore sizes in the ranges of 0–
2 nm and 2–50 nm were calculated by the Horvath–Kawazoe (HK)
method and the Barret–Joyner–Halenda (BJH) method, respec-
tively. The surface zeta potentials of the SAPO-11 samples were
determined by a zeta-sizer instrument (Malvern, England). The
acidic properties of the SAPO-11 samples were measured by the
pyridine-adsorbed infrared (Py-IR) spectra and 2,6-dimethylpyri-
dine-adsorbed infrared (2,6-DMPy-IR) spectra on a MAGNA-IR
560 instrument (Nicolet Co., USA). The samples were dehydrated
at 773 K for 4 h under a vacuum of 1.33 ꢀ 10ꢁ3 Pa and the adsorp-
tion of pure pyridine or 2,6-dimethylpyridine vapor at room tem-
perature for 20 min was followed. After reaching equilibrium, the
pyridine-adsorbed system was evacuated at 473 K and 573 K,
respectively, and the 2,6-dimethylpyridine-adsorbed system was
evacuated at 423 K and 573 K, and finally, the IR spectra were re-
corded. 29Si solid-state magic-angle-spinning nuclear magnetic
resonance (MAS NMR) spectra were recorded on a Bruker MSL-
300 NMR spectrometer with frequency at 59.6 MHz, pulse width
at 4.5 ms, and delay at 30 s.
Aladdin), which have different hydrophobic chain lengths. All of
the above chemicals were used as purchased without any further
purification.
2.2. Synthesis of SAPO-11 without and with different cationic
surfactants
First, the SAPO-11 was synthesized by using n-propanol as co-
solvent without adding any surfactant, as shown by Route 1 in
Fig. 1. A typical synthesis procedure was detailed as follows: first,
13.6 g of phosphoric acid and 40.0 g of deionized water were well
mixed, and then, 9.0 g of pseudoboehmite was slowly added under
stirring; second, after the formation of homogeneous gel, 5.0 g of
tetrapropyloxy silane and 11.3 g of n-propanol were added drop-
wise, and the resulting mixture was stirred vigorously for 4 h in or-
der to dissolve the silica species and yield a homogeneous gel;
third, 7.6 g of DPA was slowly added and the synthesis system
was stirred for 1 h; fourth, the resulting gel with the molar compo-
sition of 1.0 Al2O3:0.95 P2O5:0.3 SiO2:3.0 n-propanol:1.2 DPA:40
H2O was transferred into a 150 mL stainless steel autoclave lined
with polytetrafluorethylene (PTFE) and kept in an oven at 458 K
for 24 h; finally, the as-synthesized SAPO-11 was collected by cen-
trifugation, washing with deionized water, and drying at 393 K
overnight. To remove the organic template, the as-synthesized
SAPO-11 was calcined in flowing air at 873 K for 6 h. The synthe-
sized sample was named as SAPO-11-P, where P stands for the
co-solvent n-propanol.
The synthesis of SAPO-11 with different cationic surfactants
was the same as Route 1 in Fig. 1 used for synthesizing sample
SAPO-11-P except for adding different cationic surfactants of the
identical amount with the template DPA together, as shown by
Route 2 in Fig. 1. The molar ratio of the surfactants to SiO2 was
0.01. The resulting samples were denoted as SAPO-11-P-De,
SAPO-11-P-Do, SAPO-11-P-T, and SAPO-11-P-C, respectively,
where De, Do, T, and C stand for cationic surfactants DeTAB, DoTAB,
TTAB and CTAB, respectively.
2.5. Catalytic performance assessment
Isomerization of n-octane (n-C8) was carried out in a flowing-
type microreactor under the conditions of 573 K, 1.5 MPa, volu-
metric H2/n-octane ratio 400, and different weight hour space
velocities (WHSVs). In each run, 3.0 g of the catalyst to be assessed
was loaded into the reactor, and the n-octane was fed into the reac-
tor by a syringe pump at a predetermined flow rate. Before the
reaction, each catalyst was reduced in hydrogen flowing at 673 K
for 4 h and then the temperature was lowered to the reaction tem-
perature. After steady state was achieved, the products were ana-
lyzed by an Agilent 1790 gas chromatograph installed with a
2.3. Preparation of Pt/SAPO-11 catalysts
The above SAPO-11 samples were extruded, crushed, and sieved
to obtain the particles of 20–40 mesh in size. Then, these SAPO-11
particles were impregnated with an aqueous solution of H2PtCl6
flame ionization detector and a HP-PONA capillary column
(50 m ꢀ 0.2 mm), and the data were processed by a GC99 software
(Research Institute of Petroleum Processing, SINOPEC, China).
Fig. 1. Schematic diagram for the synthesis procedure of SAPO-11. Route 1: in water–propanol system; Route 2: in water–propanol–surfactant system.

L. Guo et al. / Journal of Catalysis 294 (2012) 161–170
163
3.2. Morphology
3. Results and discussion
The SEM images of the samples at different magnifications are
presented in Fig. 3. All the SAPO-11 samples display the spherical
particles (Fig. 3A), and the single spherical aggregate of each sam-
ple consists of many crystallites assembled layer by layer (Fig. 3B).
The SAPO-11 samples synthesized with surfactants have smaller
particle and crystallite sizes than those without surfactants. For
the SAPO-11 samples synthesized with surfactants, the particle
and crystallite sizes decrease in the order of SAPO-11-P-
C > SAPO-11-P-T > SAPO-11-P-De > SAPO-11-P-Do, indicating that
the surfactants with different chain lengths reduce the particle
and crystallite sizes of the resulting SAPO-11 samples to different
degrees and SAPO-11-P-Do with the 12-carbon surfactant has the
smallest crystallite size of about 150 nm ꢀ 200 nm among the dif-
ferent samples. Generally, these small crystallites are initially
formed during the crystallization of SAPO-11, and then, they aggre-
gate into large spheres among which intercrystal pores are gener-
ated [26,27].
3.1. Phase structure
Fig. 2 shows the XRD patterns of the SAPO-11 samples synthe-
sized by the different methods. All the samples synthesized with
and without the cationic surfactants present the diffraction peaks
at 2h = 8.1, 9.4, 13.1, 15.6, 20.3, and 22.1–23.2° attributed to the
typical SAPO-11 phase with AEL structure [24,25]. These results
demonstrate that the addition of cationic surfactants in the synthe-
sis system well preserves the crystal structure of SAPO-11.
3.3. Pore structure
The nitrogen adsorption–desorption isotherms of the SAPO-11
samples are shown in Fig. 4. For SAPO-11-P-x series (x = De, Do,
T, C), the adsorption–desorption isotherm of each sample has a
large hysteresis loop at the high relative pressure attributed to
mesopores in addition to the characteristics of microporous mate-
rials at the low relative pressure [28]. For SAPO-11-P, however,
there exists an isotherm typical for pure microporous materials.
These results clearly reveal that the involvement of different
Fig. 2. XRD patterns of the SAPO-11 samples.
Fig. 3. SEM images of the SAPO-11 samples at low (A) and high (B) magnifications.

164
L. Guo et al. / Journal of Catalysis 294 (2012) 161–170
the SAPO-11-P-x samples in Fig. 6, there are two maxima: one is
centered at ca. 4.0 nm, usually attributed to the so-called tensile
strength effect [29]; the other is centered at ca. 10.0 nm represent-
ing real mesopores. The above results indicate that the SAPO-11-P-
x series have a hierarchical pore structure, with the micropores
coming from microcrystallines and the mesopores from the inter-
crystal voids. Furthermore, in view of the similar mesopore distri-
butions of the different SAPO-11-P-x samples, it can be concluded
that the length of hydrophobic chain in the cationic surfactants
hardly changes the mesopore size of the resulting SAPO-11. The
SAPO-11-P sample has only a so-called tensile peak at ca. 4.0 nm
in its mesoporous distribution curve, indicating its property of pure
micropores.
Table 1 gives the textural parameters of all the SAPO-11 sam-
ples. Compared with the SAPO-11 synthesized without any surfac-
tant, the SAPO-11-P-x samples have much higher external surface
area and mesopore volumes, showing the promoting effect of cat-
ionic surfactants on the mesopore formation. By combining the
pore structure results with the above morphology results, it can
be inferred that the greatly increased external surface area and
mesopore volume result from the decreased crystallite size accom-
panied with the formation of more intercrystal mesopores. There-
fore, SAPO-11-P-Do with the smallest crystallite has the highest
mesopore structure among all the samples.
Fig. 4. Nitrogen adsorption–desorption isotherm curves of the SAPO-11 samples.
cationic surfactants in the SAPO-11 synthesis system can generate
bimodal pore structure.
The microporous and mesoporous size distributions of the sam-
ples are shown in Figs. 5 and 6, respectively. The micropore diam-
eters of all the samples are narrowly centered at about 0.50 nm
that is comparable to the diameter of a crystallographic 10-mem-
bered-oxygen ring (Fig. 5). In the mesopore distribution curves of
3.4. Surface zeta potential
It is well known that in zeolite synthesis, the aggregation of
zeolitic precursors is a spontaneous process owing to their colloi-
dal nature [30]. Therefore, conventional SAPO-11 usually has large
crystallite size. In this study, we found that the presence of the cat-
ionic surfactants can significantly reduce the crystallite size of the
obtained SAPO-11 samples. According to the DLVO (Derjaguin–
Landau–Verwey–Overbeek) theory [31], the dispersion behavior
of crystallites is in good coincidence with their zeta potential, so
the higher absolute value of zeta potential on crystallite surface
leads to the better dispersion of crystallites. To clarify the effect
of different surfactants on the crystallite size of SAPO-11, the sur-
face zeta potentials of the synthetic gel system at different crystal-
lization times were measured and the results are shown in Fig. 7.
The zeta potentials for all the samples are firstly negative and then
increase from a negative value to a positive one with the increasing
crystallization time. These phenomena can be explained as fol-
lows: as shown in Fig. 8, at first, SAPO-11 precursors are negatively
charged because Si atoms are introduced into the neutral AlPO4
framework, so the cationic surfactants can be easily adsorbed on
the surface of the SAPO-11 precursors due to the electrostatic
attraction between them, which gradually decreases the negative
charge number on the precursor surface and thus weakens the
electrostatic repulsion between the precursor crystallites. Hence,
the absorption of cationic surfactants on the precursor surface
leads to the crystallite flocculation. After the negative charges of
precursors are neutralized completely by the cations of surfactants,
the cationic surfactant in solution will be further bridged onto the
precursor surface through the formed bilayer micelles resulting
Fig. 5. Micropore size distributions of the SAPO-11 samples.
Table 1
Pore structure parameters of the SAPO-11 samples.
Sample
SBET
Sexternal
Vmicropore
Vmesopore
(m2/g)
(m2/g)
(cm3/g)
(cm3/g)
SAPO-11-P
218
290
330
270
259
64
217
261
180
170
0.09
0.06
0.05
0.07
0.07
0.12
0.42
0.51
0.27
0.25
SAPO-11-P-De
SAPO-11- P-Do
SAPO-11- P-T
SAPO-11- P-C
Fig. 6. Mesopore size distributions of the SAPO-11 samples.

L. Guo et al. / Journal of Catalysis 294 (2012) 161–170
165
with its increasing concentration, but reaches the adsorption equi-
librium at the critical micelle concentration (cmc) of the surfactant.
As shown in Fig. 7, the SAPO-11 samples synthesized with four dif-
ferent surfactants present different zeta potentials, suggesting that
the chain length is of critical importance in determining the
adsorption behavior of a surfactant. Indeed, the surfactant with
longer hydrocarbon chain has a much lower cmc [33], which af-
fects its adsorption behavior dramatically. Herein, the concentra-
tion of each cationic surfactant is 2.75 ꢀ 10ꢁ3 mol/L, which is
higher
than
cmcTTAB
(2.1 ꢀ 10ꢁ3 mol/L)
and
cmcCTAB
(9.2 ꢀ 10ꢁ4 mol/L), but lower than cmcDeTAB (6.5 ꢀ 10ꢁ2 mol/L)
and cmcDoTAB (1.6 ꢀ 10ꢁ2 mol/L) [34]. As a result, the effective
adsorptive concentrations for the four surfactants are in the order
of DeTAB (2.75 ꢀ 10ꢁ3 mol/L) = DoTAB (2.75 ꢀ 10ꢁ3 mol/L) > TTAB
(2.1 ꢀ 10ꢁ3 mol/L) > CTAB (9.2 ꢀ 10ꢁ4 mol/L). Hence, it is reason-
able to believe that DeTAB and DoTAB with the higher effective
adsorptive concentrations have the higher adsorption capacity
than the other two surfactants. According to the surfactant adsorp-
tion theory that the adsorption capacity of the long-chain surfac-
tant is higher than that of the short-chain surfactant under the
same conditions [33], it can be concluded that DoTAB with longer
hydrophobic chain has the higher adsorption capacity compared
with DeTAB in the case of the same effective adsorptive concentra-
tions and crystallization time.
Fig. 7. The surface zeta potential of the synthetic gel system with different
crystallization time.
from the association of its hydrophobic groups, which continu-
ously increases the positive charge number on the precursor sur-
face until reaching stability. Therefore, the electrostatic repulsion
between positive charges gradually increases and thus the disper-
sion effect from the cationic surfactant rather than the flocculation
effect is predominant.
In line with the above analyses, the zeta potentials of the SAPO-
11 samples synthesized with the different surfactants decrease
in the order of SAPO-11-P-Do > SAPO-11-P-De > SAPO-11-P-
T > SAPO-11-P-C (Fig. 7) when the stable adsorption state is
On the basis of the surfactant adsorption model on the solid sur-
face [32], the adsorption capacity of a specified surfactant increases
Fig. 8. The dispersion and flocculation effects of cationic surfactants on the negatively charged SAPO-11 precursors.

166
L. Guo et al. / Journal of Catalysis 294 (2012) 161–170
reached. The SAPO-11 crystallites adsorbed by DoTAB shows the
highest positive potential among all the samples, so their disper-
sion is optimal due to the greatly enhanced electrostatic repulsion
between them. Therefore, SAPO-11-P-Do has much smaller crystal-
lites and much higher external surface area than the other samples
with the weak electrostatic repulsion, as evidenced by the SEM and
nitrogen adsorption–desorption results.
3.5. Acidity
The Py-IR spectra of all the SAPO-11 samples are shown in
Fig. 9. For Py-IR spectra, the bands located at 1540 cmꢁ1 and
1450 cmꢁ1 can be assigned to pyridine adsorbed on Brönsted (B)
and Lewis (L) acid sites, respectively [35,36]. Total B acid sites
and total L acid sites, and medium and strong B acid sites and med-
ium and strong L acid sites can be obtained from the Py-IR mea-
surement results at 473 and 573 K, respectively. The quantitative
calculation of B and L acidity by Py-IR analysis is based on the inte-
grated Lambert–Beer Law [25]:
Csw ¼ AS=m
e
ð1Þ
where Csw
(l
mol/g) is the concentration of B or L acid sites in refer-
ence to a unit weight of dry sample, A (cmꢁ1) is the integrated
absorbance, S (cm2) is cross sectional area of the sample wafer, m
(g) is the weight of the dry sample, and
extinction coefficient. According to the literature reported [37], for
acid sites, B = 0.059 0.004 (cm/ mol); for acid sites,
L = 0.084 0.003 (cm/ mol).
The 2,6-DMPy-IR spectra of all the SAPO-11 samples are shown
e is the integrated molar
B
e
l
L
e
l
in Fig. 10. For 2,6-DMPy-IR spectra, the bands located at 1630–
1650 cmꢁ1 can be assigned to 2,6-dimethylpyridine adsorbed on
Fig. 10. IR spectra of 2,6-dimethylpyridine adsorbed on the SAPO-11 samples at
423 K (A) and 573 K (B).
B acid sites [38,39]. The kinetic diameter of 2,6-dimethylpyridine
(0.67 nm) [40] is too large to enter the pore channel of SAPO-11
(0.39 nm ꢀ 0.63 nm) [10], so only B acid sites at the pore mouths
of SAPO-11 are accessible for 2,6-dimethylpyridine [41]. Similar
to the quantitative calculation of B acid sites by Py-IR analysis,
the total B acid sites and the medium and strong B acid sites at
the pore mouths of SAPO-11 can be obtained from the
2,6-DMPy-IR measurement results at 423 K and 573 K [42],
respectively. The integrated molar extinction coefficient of 2,6-
dimethylpyridine given by Onfroy et al. [43] was used here.
Table 2 gives the quantitative results about the acid properties
of the different samples from Py-IR and 2,6-DMPy-IR. Compared
with SAPO-11-P, the SAPO-11-P-x series have similar amount of
L acid sites but more amount of B acid sites, indicating the promo-
tion effect of cationic surfactants on the exposure of B acid sites
[22]. For the SAPO-11-P-x series, the amounts of medium and
strong B acid sites at the pore mouths decrease in the order of
SAPO-11-P-Do > SAPO-11-P-De > SAPO-11-P-T > SAPO-11-P-C,
showing that the suitable chain length of cationic surfactants is
necessary for more detected B acid sites. Combining the acidity re-
sults and the above SEM and nitrogen adsorption–desorption re-
sults, it can be inferred that the smaller crystallites of SAPO-11
with higher external surface area can produce more amount of
medium and strong B acid sites at the pore mouths of SAPO-11.
It has been well recognized that the acidity of a SAPO-11 molec-
ular sieve originates from Si atoms incorporated into the AlPO4-11
framework, so the manners of Si substitution directly determine
the acidity properties of SAPO-11 [44]. Generally, the incorporation
of Si atoms into the AlPO4-11 structure follows two different mech-
anisms [22]: the one is the so-called SM2 mechanism, that is, one
Si atom substitutes for one P atom to form a Si(4Al) site; the
other is the so-called SM3 mechanism, that is, two Si atoms
Fig. 9. IR spectra of pyridine adsorbed on the SAPO-11 samples at 473 K (A) and
573 K (B).

L. Guo et al. / Journal of Catalysis 294 (2012) 161–170
167
Table 2
Acidity properties of the SAPO-11 samples determined by Py-IR and 2,6-DMPy-IR characterizations.
Sample
Acidity (
lmol/g)
Weak acid sites (Py-IR)
Medium and strong sites (Py-IR)
Weak acid sites (2,6-DMPy-IR)
B
Medium and strong sites (2,6-DMPy-IR)
L
B
L
B
B
SAPO-11-P
43.3
45.2
45.9
44.2
43.8
50.9
68.7
85.9
66.7
61.5
25.5
26.2
26.0
25.4
25.1
72.3
12.4
49.6
48.1
37.4
38.0
15.7
57.6
70.4
38.9
35.2
SAPO-11-P-De
SAPO-11- P-Do
SAPO-11- P-T
SAPO-11- P-C
123.6
130.8
96.4
85.9
simultaneously substitute adjacent one P atom and one Al atom.
According to the theory of Martens et al. [45], the SAPO domains
originate from Si incorporated into AlPO4 framework by SM2
mechanism with the Si(4Al) environment of each Si atom, and all
the second nearest neighbors of Si are P atoms. Different from
SAPO domains, the aluminosilicate (SA) domains formed by SM3
and SM2 mechanisms do not contain P atoms. To understand the
chemical surroundings of Si in the different SAPO-11 samples syn-
thesized, the 29Si MAS NMR spectroscopy characterization was car-
ried out and the results are shown in Fig. 11. The 29Si MAS NMR
spectra of both SAPO-11-P and SAPO-11-P-Do show a broad band
in the range of ꢁ80 to ꢁ120 ppm. The resonance peak at around
ꢁ90 ppm is attributed to Si(4Al) species in SAPO domains, and
those at ꢁ86, ꢁ95, ꢁ99, ꢁ106, and ꢁ112 ppm represent Si(4Al),
Si(3Al), Si(2Al), Si(1Al), and Si(4Si) in SA domains, respectively
[14,45]. The experimentally determined 29Si MAS NMR spectra
were deconvolved into six distinct peaks by using Gaussian curves,
and the results are shown in Table 3. Si(1Al) and Si(2Al) in SA do-
mains, and Si(3Al) located at the interface between SA and SAPO
domains have medium and strong B acidity that promotes the skel-
etal isomerization of hydrocarbons, so the SAPO-11-P-Do sample
with more Si(nAl)(0 < n < 4) has more amounts of medium and
strong B acid sites [46] than the SAPO-11-P sample as confirmed
by the above Py-IR characterization results.
3.6. Catalytic performance
Höchto et al. [44] pointed out that the loading of 0.5 wt% Pt on
an acidic support could meet the requirement for the hydrogena-
tion–dehydrogenation function of bifunctional catalysts. Thereby,
a series of 0.5 wt% Pt/SAPO-11 catalysts were prepared with the
above SAPO-11 samples as supports and assessed with n-octane
isomerization as a model reaction herein. n-Octane isomerization
results are shown in Fig. 12. Both of selectivity to di-branched iso-
mers and cracking selectivity over different Pt/SAPO-11 catalysts
increase with the increasing n-C8 conversion, suggesting that the
increase in n-octane conversion not only promotes the formation
of di-branched isomers but also enhances the undesirable cracking
reactions. At the same conversion of n-octane, the Pt/SAPO-11-P-x
catalysts are always found to exhibit higher selectivity to di-
branched isomers and lower cracking selectivity than the Pt/
SAPO-11-P catalyst, indicating the improving effect of the cationic
surfactants on the di-branched isomerization of SAPO-11-based
catalysts. As to the Pt/SAPO-11-P-x catalysts, at the same conver-
sion of n-octane, the selectivity to di-branched isomers decreases
in the order of Pt/SAPO-11-P-Do > Pt/SAPO-11-P-De > Pt/SAPO-
11-P-T > Pt/SAPO-11-P-C while the cracking selectivity increases
in the order of Pt/SAPO-11-P-Do < Pt/SAPO-11-P-De < Pt/SAPO-
11-P-T < Pt/SAPO-11-P-C, demonstrating that the suitable chain
length of cationic surfactants is critical for improving the di-
branched isomerization and restraining the cracking reactions over
SAPO-11-based catalysts.
The detailed results of n-octane isomerization over all the cat-
alysts are listed in Table 4. 2-methyl heptane (2-MC7) and 3-
methyl heptane (3-MC7) are the major mono-branched products
over all the catalysts, in good agreement with the product shape
selectivity in SAPO-11 [47], while the selectivity to di-branched
isomers decreases in the order of 2,5-dimethylhexane (2,5
DMC6) > 2,4-dimethylhexane (2,4 DMC6) > 2,3-dimethylhexane
(2,3 DMC6) > 3,4-dimethylhexane (3,4 DMC6) > 2,2-dimethylhex-
ane (2,2 DMC6) over all the catalysts, implying the prior isomeri-
zation of terminal alkyls of n-octane over the above SAPO-11
catalysts. Among the Pt/SAPO-11-P-x catalysts, the highest activ-
ity for n-octane conversion reflected by the k and TOF values,
the highest selectivity to di-branched isomers, and the lowest
cracking selectivity are realized over Pt/SAPO-11-P-Do, displaying
the remarkable advantage of DoTAB in balancing high-efficient
isomerization and minimal cracking.
Fig. 11. 29Si MAS NMR spectra of the selected SAPO-11 samples.
Table 3
Deconvolution results of the 29Si MAS NMR spectra of SAPO-11-P and SAPO-11-Do based on the normalized peak areas of the different Si species.
Sample
SAPO domain
SA domain
Si (4Al) (%)
Si (4Al) (%)
Si (3Al, 1Si) (%)
Si (2Al, 2Si) (%)
Si (1Al,3Si) (%)
Si (4Si) (%)
ꢁ90 ppm
ꢁ86 ppm
ꢁ95 ppm
ꢁ99 ppm
ꢁ106 ppm
ꢁ112 ppm
SAPO-11-P
SAPO-11-P-Do
3.3
15.8
2.3
8.0
7.2
21.5
16.4
26.8
25.1
15.5
45.7
12.4

168
L. Guo et al. / Journal of Catalysis 294 (2012) 161–170
Pt/SAPO-11-P with the largest crystallites among all the catalysts
(Fig. 13), its longest pore channels lead to the greatly increased res-
idence time of the intermediates of mono-branched C8 isomers and
thus result in their further cracking on medium and strong acid
sites, so Pt/SAPO-11-P presents the highest cracking selectivity
among all the catalysts. Due to the larger size of di-branched C8
isomers (0.71 nm) [19] than the pore diameter (0.63 nm) of
SAPO-11 [19], the di-branched C8 isomers can be formed only near
the pore mouths rather than in the microporous channels accord-
ing to the ‘‘pore mouth and key-lock’’ concept [49]. As a result,
SAPO-11 with more amounts of medium and strong B acid sites
at the pore mouths should be beneficial to produce more
di-branched isomers. In view of the direct proportion relationship
between the amount of B acid sites at the pore mouths and the
external surface area of SAPO-11 [50], SAPO-11-P with the lowest
external surface area has the least B acid sites at the pore mouths
among all the SAPO-11 samples, so the corresponding catalyst Pt/
SAPO-11-P produces the least di-branched C8 isomers.
The Zeta potential analysis results have clarified that the
suitable chain length in cationic surfactants can maximize the
dispersion of SAPO-11 crystallites. The SEM, nitrogen adsorption–
desorption, and acidity results have confirmed that incorporating
the cationic surfactant with the optimal chain length in the synthe-
sis system can produce the SAPO-11 with the smallest crystallite,
the highest external surface area and the most amounts of active
B acid sites at the pore mouths among the SAPO-11 samples
synthesized with different surfactants. DoTAB has just the
above-mentioned suitable chain length and thus the corresponding
catalyst Pt/SAPO-11-P-Do presents the best di-branched isomeri-
zation performance among all the catalysts.
As illustrated in Fig. 13, Pt/SAPO-11-P-Do with smaller crystal-
lites and higher external surface area has abundant and accessible
B acid sites at the pore mouths that guarantee the efficient mono-
branched isomerization of two terminal alkyls of n-octane near the
two adjacent pore mouths, so more di-branched C8 isomers are
produced over it. However, the other catalysts with larger crystal-
lites and lower external surface area have less B acid sites at
the pore mouths that reduce the di-branched isomerization of
n-octane near pore mouths, so these catalysts show lower
di-branched isomer selectivity.
Fig. 12. Selectivity to di-branched isomers (A) and cracking selectivity (B) versus n-
octane conversion over Pt/SAPO-11-P (a), Pt/SAPO-11-P-De (b), Pt/SAPO-11-P-Do
(c), Pt/SAPO-11-P-T (d), and Pt/SAPO-11-P-C (e).
On the basis of the isomerization mechanism of n-alkanes over
bifunctional catalysts [48], for Pt/SAPO-11 catalysts, n-octane is
first dehydrogenated into n-octene on Pt centers, and the resulting
n-octene molecules with a critical molecular diameter of 0.56 nm
[6] diffuse onto the B acid sites of SAPO-11 where the intermedi-
ates of mono-branched C8 isomers are formed; then, the mono-
branched C8 intermediates can be further converted into the
di-branched C8 intermediates on the B acid sites in the case of no
steric hindrance; finally, the di-branched C8 isomers of n-octane
are generated by the hydrogenation on Pt centers. For microporous
The results of the 29Si MAS NMR and acidity characterizations
have demonstrated that the SAPO-11-P-Do sample has more bor-
ders of Si domains and thus more amount of medium and strong
B acid sites compared with the other samples. For isomerization
Table 4
Results of n-octane isomerization over the different Pt/SAPO-11 catalysts.
Pt/SAPO-11-P
Pt/SAPO-11-P-De
Pt/SAPO-11-P-Do
Pt/SAPO-11-P-T
Pt/SAPO-11-P-C
ka (10ꢁ6 mol gꢁ1sꢁ1
)
3.9
16.6
84.7
1.7
19.4
22.7
78.1
15.5
26.5
30.5
74.7
23.1
12.0
21.8
82.6
8.9
9.1
18.8
83.1
8.1
TOFb (10ꢁ2 sꢁ1
)
c
SMB (%)
c
SDB (%)
PSc (%)
2-MC7
3-MC7
4-MC7
40.3
35.4
9.0
1.3
0.2
0.1
0.1
0.0
13.6
33.7
35.5
8.9
8.0
5.4
1.5
0.4
0.2
6.4
30.2
36.9
7.6
10.8
9.2
2.1
0.8
0.2
2.2
38.5
35.6
8.5
6.1
2.2
0.5
0.1
0.0
38.1
36.4
8.6
5.9
2.3
0.5
0.1
0.0
2,5-DMC6
2,4-DMC6
2,3-DMC6
3,4-DMC6
2,2-DMC6
c
Sc (%)
8.5
8.8
a
b
c
Reaction rate constant obtained by assuming a pseudo-first-order reaction for n-octane isomerization at 573 K, 1.5 MPa, H2/octane volumetric ratio of 400, and 7.0 WHSV.
Number of reacted n-octane molecules per second and per medium and strong B acid site at the pore mouths.
SMB, SDB, PS, and Sc correspond to the total selectivity to mono-branched C8 isomers, the total selectivity to di-branched C8 isomers, the product selectivity, and the
cracking selectivity, respectively, at 54% conversion of n-octane.

L. Guo et al. / Journal of Catalysis 294 (2012) 161–170
169
Fig. 13. Reaction locations and diffusion pathways for n-octane isomerization over the Pt/SAPO-11 catalysts based on large-sized and small-sized SAPO-11 crystallites,
respectively.
(Table 4). In summary, the superior di-branched isomerization of
n-octane can be accomplished over the Pt/SAPO-11 catalyst with
small crystallite and high external surface area.
4. Conclusions
A series of SAPO-11 molecular sieves with different crystallite
sizes, pore structures, and acidities were successfully synthesized
via incorporating the cationic surfactants of different chain lengths
in water–propanol system, and their di-branched isomerization
performances were assessed. The results showed that increasing
chain length of surfactants could firstly decrease and then increase
the crystallite size of SAPO-11, while the external surface area and
the active B acid sites of SAPO-11 firstly increased and then de-
creased with the increasing chain length. Dodecyltrimethyl ammo-
nium bromide (DoTAB) could maximize the dispersion of SAPO-11
crystallites and thus produced the smallest crystallite, the highest
Fig. 14. External surface area versus crystallite size and selectivity to di-branched
external surface area, and the most amounts of medium and strong
isomers over the Pt/SAPO-11 catalysts with the conversion of n-octane at 54%.
B acid sites at the pore mouths of SAPO-11 among the four cationic
surfactants investigated, endowing the corresponding catalyst
with superior di-branched isomer selectivity and minimal cracking
selectivity. The findings obtained herein highlight the development
of novel catalysts for boosting high octane-rating gasoline via the
di-branched isomerization of long n-alkanes with low octane
rating.
of n-alkanes, the medium and strong B acid sites are considered as
the active sites for the skeletal isomerization and C–C bond crack-
ing of hydrocarbons [19], so the increase in their number can effi-
ciently promote not only the mono- and di-branched sequential
isomerization of n-octane but also accelerate the cracking of n-oc-
tane. However, Pt/SAPO-11-P-Do with more amounts of medium
and strong B acid sites has not only better di-branched isomer
selectivity, but also lower cracking selectivity compared with the
other catalysts at the same conversion of n-octane (Fig. 12 and
Table 4). This distinct feature can be attributed to the shortened
diffusion paths because of the greatly decreased crystallite size of
Pt/SAPO-11-P-Do, accelerating mass transport and thus suppress-
ing the cracking of n-octane.
To better understand the relationship between the physico-
chemical properties of the SAPO-11 samples and their isomeriza-
tion performances, the external surface areas of the SAPO-11
samples were correlated with their crystallite sizes and di-
branched isomer selectivities, and the results are shown in
Fig. 14. Decreasing the crystallite size of the SAPO-11 samples
can increase their external surface areas [50], and the increased
external surface area of the SAPO-11 samples can increase the
medium and strong B acid sites at the pore mouths (Tables 1 and
2) [51], which enhances the di-branched isomerization of n-octane
Acknowledgments
The authors gratefully acknowledge the financial support of the
National Basic Research Program (Grant No. 2010CB226905), the
Program for New Century Excellent Talents in University (Grant
No. NCET-09-0763), the National Natural Science Foundation of
China (Grant No. 21076228), and the Science Foundation of China
University of Petroleum, Beijing (Grant No. KYJJ2012-03-04).
References
[1] V.M. Akhmedov, S.H. Al-Khowaiter, Catal. Rev. 49 (2007) 33.
[2] Y. Rezgui, M. Guemini, Energy Fuels 21 (2007) 602.
[3] V. Calemma, S. Peratello, C. Perego, Appl. Catal. A – Gen. 190 (2000) 207.
[4] S.J. Miller, Stud. Surf. Sci. Catal. 84 (1994) 2319.
[5] A. Chica, A. Corma, J. Catal. 187 (1999) 167.
[6] P. Mériaudeau, V.A. Tuan, V.T. Nghiem, S.Y. Lai, L.N. Hung, C. Naccache, J. Catal.
169 (1997) 55.
[7] K.C. Park, S.K. Ihm, Appl. Catal. A – Gen. 203 (2000) 201.

170
L. Guo et al. / Journal of Catalysis 294 (2012) 161–170
[8] Y. Liu, C. Liu, C. Liu, Z. Tian, L. Lin, Energy Fuels 18 (2004) 1266.
[9] V.M. Benitez, J.C. Yori, J.M. Grau, C.L. Pieck, C.R. Vera, Energy Fuels 20 (2006)
422.
[32] S. Paria, K.C. Khilar, Adv. Colloid Interface Sci. 110 (2004) 75.
[33] R. Atkin, V.S.J. Craig, E.J. Wanless, S. Biggs, Adv. Colloid Interface Sci. 103 (2003)
219.
[10] N. Venkatathri, R. Srivastava, Stud. Surf. Sci. Catal. 154 (2004) 978.
[11] J. Walendziewski, B. Pniak, Appl. Catal. A – Gen. 250 (2003) 39.
[12] J.M. Campelo, F. Lafont, J.M. Marinas, J. Catal. 156 (1995) 11.
[13] A.K. Sinha, S. Sivasanker, Catal. Today 49 (1999) 293.
[14] P. Mériaudeau, V.A. Tuan, F. Lefebvre, V.T. Nghiem, C. Naccache, Micropor.
Mesopor. Mater. 22 (1998) 435.
[34] T.F. Tadros, Applied Surfactants: Principles and Applications, Wiley-VCH,
Weinheim, 2005. p. 21.
[35] J.A.Z. Pieterse, S. Veefkind-Reyes, K. Seshan, L. Domokos, J.A. Lercher, J. Catal.
187 (1999) 518.
[36] Y. Fan, H. Xiao, G. Shi, H. Liu, Y. Qian, T. Wang, G. Gong, X. Bao, J. Catal. 279
(2011) 27.
[15] P. Liu, J. Ren, Y.H. Sun, Catal. Commun. 9 (2008) 1804.
[16] J.M. Campelo, F. Lafont, J.M. Marinas, Zeolites 15 (1995) 97.
[17] M. Höchtl, A. Jentys, H. Vinek, Catal. Today 65 (2001) 171.
[18] B. Parlitz, E. Schreier, H.L. Zubowa, R. Eckelt, E. Lieske, G. Liscke, R. Fricke, J.
Catal. 155 (1995) 1.
[19] Y. Fan, H. Xiao, G. Shi, H. Liu, X. Bao, J. Catal. 285 (2012) 251.
[20] A.K. Sinha, S. Sivasanker, P. Ratnasamy, Ind. Eng. Chem. Res. 37 (1998) 2208.
[21] M.J. Franco, A. Mifsud, J. Perez-Pariente, Zeolites 15 (1995) 117.
[22] T. Blasco, A. Chica, A. Corma, W.J. Murphy, J. Agúndez-Rodríguez, J. Pérez-
Pariente, J. Catal. 242 (2006) 153.
[23] X. Huang, L. Wang, L. Kong, Q. Li, Appl. Catal. A – Gen. 253 (2003) 461.
[24] S. Zhang, S. Chen, P. Dong, G. Yuan, K.Q. Xu, Appl. Catal. A – Gen. 332 (2007) 46.
[25] Y. Fan, D. Lei, G. Shi, X. Bao, Catal. Today 114 (2006) 388.
[26] M. Brunsteiner, A.G. Jones, F. Pratola, S.L. Price, S.J.R. Simons, Cryst. Growth
Des. 5 (2005) 3.
[37] S. Zeng, J. Blanchard, M. Breysse, Y. Shi, X. Shu, H. Nie, D. Li, Micropor. Mesopor.
Mater. 85 (2005) 297.
[38] J.A. Lercher, A. Jentys, Stud. Surf. Sci. Catal. 168 (2007) 435.
[39] C. Morterra, G. Cerrato, G. Meligrana, Langmuir 17 (2001) 7053.
[40] F. Thibault-Starzyk, I. Stan, S. Abelló, A. Bonilla, K. Thomas, C. Fernandez, J.
Gilson, J. Pérez-Ramírez, J. Catal. 264 (2009) 11.
[41] W. Huang, D. Li, X. Kang, Y. Shi, H. Nie, Stud. Surf. Sci. Catal. 154 (2004) 2253.
[42] M. Moreno, A. Rosas, J. Alcaraz, M. Hernández, S. Toppi, P.D. Costa, Appl. Catal.
A – Gen. 251 (2003) 369.
[43] T. Onfroy, G. Clet, M. Houalla, Micropor. Mesopor. Mater. 82 (2005) 99.
[44] M. Höchtl, A. Jentys, H. Vinek, J. Catal. 190 (2000) 419.
[45] J.A. Martens, P.J. Grobet, P.A. Jacobs, J. Catal. 126 (1990) 299.
[46] D. Barthomeuf, Zeolites 14 (1994) 394.
[47] T.L.M. Maesen, M. Schenk, T.J.H. Vlugt, J.P.D. Jonge, B. Smit, J. Catal. 188 (1999)
403.
[27] H. Zhu, Z. Liu, D. Kong, Y. Wang, X. Yuan, Z. Xie, J. Colloid Interface Sci. 331
(2009) 432–438.
[28] S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and Porosity, Academic Press,
London, 1982. p. 154.
[29] N. Danilina, F. Krumeich, J.A. van Bokhoven, J. Catal. 272 (2010) 37.
[30] B.J. Schoeman, J. Sterte, J.E. Otterstedt, Zeolites 14 (1994) 110.
[31] J.P. Valleau, R. Lvkov, G.M. Torrie, J. Chem. Phys. 95 (1991) 520.
[48] Y. Ono, Catal. Today 81 (2003) 3.
[49] J.A. Martens, W. Souverijns, W. Verrelst, R. Parton, G.F. Froment, P.A. Jacobs,
Angew. Chem., Int. Ed. 34 (1995) 2528.
[50] P. Mériaudeau, V.A. Tuan, G. Sapaly, V.T. Nghiem, C. Naccache, Catal. Today 49
(1999) 285.
[51] S. Zhang, S. Chen, P. Dong, Z. Ji, J. Zhao, K. Xu, Catal. Lett. 118 (2007) 109.