Synthesis, characterization, and structure solution of CIT-5, a new, high-silica, extra-large-pore molecular sieve

Article abstract of DOI:10.1021/jp982169t

The synthesis, structure solution, and characterization of the high-silica molecular sieve CIT-5 (California Institute of Technology Number 5) is described. CIT-5 is synthesized at hydrothermal conditions in the presence of N(16)-methylsparteinium and preferrably lithium cations. The structural solution of CIT-5 shows that it contains one-dimensional pores circumscribed by 14 tetrahedral atoms (14 MR). Rietveld refinement of the synchrotron X-ray powder data gives a symmetry and space group assignment for the structure of Pmn2₁ (no. 31) with refined unit cell parameters of a = 13.6738(8) ?, b = 5.0216(3) ?, and c = 25.4883(7) ? (V = 1750.1 ?³). Electron diffraction and transmission electron microscopy confirm the space group and the topology of the structure viewed along the [010] direction. Solid-state ²⁹Si NMR spectroscopy results are consistent with the space group assignment. The thermal/hydrothermal stability of CIT-5 compares well to that of other large- and extra-large-pore, high-silica molecular sieves. The acid form of CIT-5 is able to perform hydrocarbon reactions such as cracking and alkylation and shows behaviors that are different from other zeolites.

Full text of DOI:10.1021/jp982169t

J. Phys. Chem. B 1998, 102, 7139-7147
7139
Synthesis, Characterization, and Structure Solution of CIT-5, a New, High-Silica,
Extra-Large-Pore Molecular Sieve
Masahito Yoshikawa,† Paul Wagner, Mark Lovallo, Katsuyuki Tsuji, Takahiko Takewaki,
,‡
†
§
†
†
|
†
†
§
|
Cong-Yan Chen, Larry W. Beck, Chris Jones, Michael Tsapatsis, Stacey I. Zones, and
Mark E. Davis*,†
DiVision of Chemistry and Chemical Engineering, California Institute of Technology,
Pasadena, California 91125, Department of Chemical Engineering, UniVersity of Massachusetts,
Amherst, Massachusetts, 01003, and CheVron Research and Technology Co., Richmond, California
ReceiVed: May 7, 1998; In Final Form: June 23, 1998
The synthesis, structure solution, and characterization of the high-silica molecular sieve CIT-5 (California
Institute of Technology Number 5) is described. CIT-5 is synthesized at hydrothermal conditions in the
presence of N(16)-methylsparteinium and preferrably lithium cations. The structural solution of CIT-5 shows
that it contains one-dimensional pores circumscribed by 14 tetrahedral atoms (14 MR). Rietveld refinement
of the synchrotron X-ray powder data gives a symmetry and space group assignment for the structure of
Pmn2
1
(no. 31) with refined unit cell parameters of a ) 13.6738(8) Å, b ) 5.0216(3) Å, and c ) 25.4883(7)
3
Å (V ) 1750.1 Å ). Electron diffraction and transmission electron microscopy confirm the space group and
2
9
the topology of the structure viewed along the [010] direction. Solid-state Si NMR spectroscopy results
are consistent with the space group assignment. The thermal/hydrothermal stability of CIT-5 compares well
to that of other large- and extra-large-pore, high-silica molecular sieves. The acid form of CIT-5 is able to
perform hydrocarbon reactions such as cracking and alkylation and shows behaviors that are different from
other zeolites.
Introduction
hydrothermal stability characteristics of other high-silica mo-
lecular sieves and is the first high-silica molecular sieve to
Molecular sieves with 12 tetrahedral atoms (T-atoms) cir-
cumscribing the pores have been denoted as large-pore materials
and have pore diameters of approximately 6-8 Å. Com-
mercially useful zeolites such as NaY, L, and beta are large-
pore materials. Pore systems that contain rings circumscribed
by more than 12 T-atoms are classified as extra-large pores.
For many years, extra-large-pore materials have been aggres-
sively pursued both in industry and in academia.
demand for these materials is led by a desire to perform shape-
selective adsorption and catalysis on molecules that are greater
than 8 Å in size. Provided these extra-large-pore materials can
retain their pore structures under the thermal or hydrothermal
conditions required, they should find applications as catalysts
in the petrochemical industry and show potential for use in the
fine chemicals and pharmaceuticals industries as well.
The first extra-large-pore material, VPI-5 (an aluminophos-
phate material), was synthesized in 1987, and since that time
numerous other phosphate-based extra-large-pore materials have
contain pores circumscribed by 14 tetrahedral atoms (14 MR).
The synthesis of UTD-1 requires an organometallic structure-
directing agent that can only be removed from the pores through
calcination and acid treatment. Additionally, aluminum frame-
work substitution (needed for acid catalysis) requires postsyn-
thetic treatments.
1
Here, we report on the syntheses, characterization, and
structural solution of CIT-5, a new, high-silica molecular sieve
with 14-ring pores that is synthesized using an organic structure-
directing agent. A preliminary report on the topology of CIT-5
has appeared, and here we provide the complete structure
2
-8
The
17
solution. CIT-5 and UTD-1 are the only high-silica molecular
sieves to contain these extra-large pores. The high thermal and
hydrothermal stabilities of UTD-1 and CIT-5 confirm that extra-
large-pore materials can have the stability required for industrial
use. The direct incorporation of heteroatom substitutes such
as aluminum into the framework of CIT-5 and the ease of
removal of the structure-directing agent through calcination
make this material an appealing catalyst for large molecules.
9
3
,10-14
been prepared.
These phosphate-based materials suffer
from poor thermal and hydrothermal stability that limits their
usefulness.
Recently, the high-silica molecular sieve UTD-1 has been
characterized in detail and was shown to possess large, one-
Experimental Section
Synthesis. CIT-5 was synthesized at hydrothermal conditions
within a temperature range of 423-448 K. The syntheses were
carried out in both sealed quartz tubes (75 × 15 mm i.d.) and
Teflon-lined Parr reaction vessels. CIT-5 can be prepared from
the following composition:
15,16
dimensional pores.
UTD-1 has the high thermal and
*
Address correspondence to this author at Division of Chemistry and
Chemical Engineering, 210-41, California Institute of Technology, Pasadena,
CA 91125 (818) 395-4251.
†
California Institute of Technology.
Current address: Toray Industries Inc., Nagoya 455, Japan.
University of Massachusetts.
Chevron Research and Technology Co.
1
SiO :0.1 MOH:0.2 MeSPAOH:x W:40 H O
2 2
‡
§
|
where W can include the following: Ga(NO3)3 (Aldrich), H3-
S1089-5647(98)02169-5 CCC: $15.00 © 1998 American Chemical Society
Published on Web 08/21/1998

7140 J. Phys. Chem. B, Vol. 102, No. 37, 1998
Yoshikawa et al.
6
BO3,(J. T. Baker), or Al(NO3)3 (J. T. Baker) (x can be varied
from 0 to 0.02), and MOH can be either LiOH (Fisher) alone
or a mixture of LiOH together with either NaOH (Aldrich) or
KOH (Aldrich). As discussed below, CIT-5 can be formed in
under vacuum (<10- Torr) at 573 K for 2 h. Approximately
1000 data points were obtained from the continuous flow
-
6
experiments with argon pressures ranging between 10 and
550 Torr. Sample sizes were ∼100 mg, and the argon flow
rate was low enough to require 1-2 h to admit enough argon
to fill the sample micropores. The equilibration times for the
static adsorption experiments were 10 min for each data point.
Doses were small enough so that 5-10 h were required to admit
enough argon to fill the micropores. Micropore volumes and
external areas were estimated from R-plot analyses of the
adsorption isotherms, using R values obtained from a silica
standard (CPG-75). The adsorption capacities of zeolites for
vapor-phase hydrocarbons were measured at room temperature
using a Cahn C-2000 balance coupled with a computer via an
ATI-Cahn digital interface. The adsorbates studied were
n-hexane, 2,2-dimethylbutane, cyclohexane, and 1,3,5-triiso-
propylbenzene. The vapor of the adsorbate was delivered from
the liquid phase. The relative vapor pressure P/P0 was
maintained at ∼0.3 by controlling the temperature of the liquid
adsorbate using a cooling circulator. Prior to the adsorption
experiments, the calcined zeolites were dehydrated at 623 K
+
+
+
the absence of Li cations if the alkali metal cation (K or Na )
is present in very low concentration ([MOH]/[SiO2] less than
-
0
.05) when the total [OH ]/[SiO2] ratio is held constant at 0.3.
The synthesis of the organic structure-directing agent, N(16)-
methylsparteinium hydroxide (I) (MeSPAOH), has been previ-
ously described.18
In a typical synthesis preparation of pure-silica CIT-5, 0.019
g of LiOH is added to a solution of 4.21 g of distilled, deionized
water and 0.85 g of a 50.0 wt % solution of N(16) methyl-
sparteinium hydroxide. Then, 1.60 g of the colloidal silica
source (HS-30 DuPont) is added, and the mixture is stirred for
-3
under a vacuum of 10 Torr for 5 h. The adsorption capacities
are reported in milliliters of liquid per gram of dry zeolite,
assuming bulk liquid density for the adsorbate in the micropores.
30 min. The resulting synthesis mixture is charged into quartz
tubes that are heat sealed and placed in a convection oven at
Catalytic Studies. Three catalytic test reactions are reported
using the acid form of CIT-5: cracking of C6 isomers, m-xylene
isomerization, and hydrocracking of n-hexadecane. The CIT-5
material was synthesized as previously described, with the
exception that highly dealuminated Y zeolite (TOSOH HUA
90) was used as the silica source. The syntheses were carried
out in Teflon-lined Parr reactors heated to 433 K while rotating,
for a period of 6-10 days.
The active CIT-5 catalyst was prepared by calcining the as-
synthesized CIT-5 to 873 K in air, followed by ion exchange
of the calcined material with ammonium nitrate. The ion
exchange was carried out by heating 1 g of the calcined zeolite
with 1 g of ammonium nitrate in 50 mL of water at 363 K for
1
75 °C for 3-7 days. The recovered solid is washed with
distilled, deionized water and allowed to air-dry.
Analytical. X-ray Diffraction. Synchrotron powder X-ray
diffraction (SPXRD) patterns were collected on the X7A
synchrotron beam line at Brookhaven National Laboratory. The
CIT-5 sample was prepared for data collection by first calcining
the material at 923 K (heating from room temperature at 5
K/min). The calcined sample was then packed into a 1-mm
glass capillary, and the synchrotron data were collected at
ambient conditions with a step size of 0.005° from 4° to 50°
19
3
2
-θ at a wavelength of 1.147 98 Å. Powder X-ray diffraction
patterns were also obtained on a Scintag XDS 2000 diffracto-
meter using CuKR radiation (λ ) 1.541 84 Å) with a liquid-
nitrogen-cooled solid-state germanium detector in a Bragg-
Brentano configuration. The diffraction data were collected in
continuous scanning mode over the range 2° < 2-θ < 51° with
a scanning rate of 0.2°/s. National Institute of Standards and
Technology Standard Reference Material 675 (fluorophologopite
mica) was used as the external standard.
2
h, followed by filtration, washing with water, and finally air-
drying. The ammonium-exchanged zeolite was then packed into
the catalytic reactor and converted to the acid form by heating
to 813 K in situ. The acid form of CIT-5 was used for the
cracking of C6 isomers and for the isomerization of m-xylene.
The test reaction involving the hydrocracking of n-hexadecane
(
a model compound) utilized an acid/palladium form of the
The thermal and hydrothermal stability of CIT-5 were studied
by in situ, temperature-programmed powder X-ray diffraction.
The samples were mounted as thin films on a platinum-rhodium
alloy heater filament inside an Edmund Buhler high-temperature
X-ray diffraction chamber. The sample thickness of the calcined
CIT-5 powder on the filament varied from 0.5 to 1 mm. The
chamber was evacuated and refilled with the flow gas, and a
steady flow of 0.5 L/min was then maintained during data
collection. The flow gas for the thermal stability studies was
dry nitrogen, while dry nitrogen bubbled through distilled water
held at room temperature was used as the flow gas for the
hydrothermal stability studies. The samples were heated at a
rate of 10 K/min and held at the designated temperature for 15
min prior to data collection. The X-ray diffraction scans were
collected as continuous scans from 2° to 51° 2-θ at a scan rate
of 5°/min.
zeolite (as described below).
The cracking of C6 isomers was carried out as follows: 500
mg of ammonium-exchanged CIT-5 (20-40 mesh after pelleting
at 3 kpsi and breaking the pellet) was calcined to 813 K for 4
h and cooled under a flow of nitrogen. The catalyst was then
placed in a 3/8 in. stainless steel reactor with a bed of alundum
packed on either side of the catalyst. Helium was passed
3
through the reactor at 10 cm /min while the reactor tube was
heated by a Lindburg furnace. At a reactor tube temperature
of 645 K, a 50/50 (w/w) feed of n-hexane and 3-methylpentane
was introduced by a downflow (controlled using a Brownlee
pump) of 8 µL/min. On-line sampling via a Valco 6-way valve
was started after 10 min of feed delivery. The data were
collected on a Hewlett-Packard 5880 gas chromatograph with
product peaks assigned using standards.
Physisorption Studies. Argon physisorption measurements
were carried out on a Coulter Omnisorp 100CX adsorption
instrument. Argon adsorption isotherms were measured at 87
K by both static and continuous-flow techniques. Prior to the
adsorption experiments, the calcined samples were degassed
The m-xylene isomerization test reactions were conducted
using two CIT-5 samples as catalysts. Sample A had a silica-
to-alumina ratio (SAR) of 200, and sample B a silica-to-galia
ratio (SGR) of 100. Sample A was syntheszied using a
dealuminated Y zeolite as the silicon source, while sample B

Synthesis, Characterization, and Structure Solution of CIT-5
J. Phys. Chem. B, Vol. 102, No. 37, 1998 7141
TABLE 1: Synthesis of CIT-5a
phase
(silicon/Y)
b
SiO
2
MOH
MeSPAOH
W
H
2
0
temp (°C)
time (days)
1
1
1
1
1
1
1
1
1
1
0.1 LiOH
0.05 KOH
0.05 NaOH
0.2
0.25
0.25
0.2
0.2
0.2
0.2
0.2
0.2
0.2
40
40
40
40
40
40
40
40
40
40
175
175
175
175
175
175
175
175
175
175
5
60
18
7
5
5
5
7
11
7
CIT-5
CIT-5 + AFI
CIT-5
AFI
CIT-5
AFI
CIT-5
CIT-5 (63)
CIT-5 (95)
CIT-5 (109)
0.1 NaOH
0.075 LiOH, 0.025 NaOH
0.050 LiOH, 0.050 NaOH
0.075 LiOH, 0.025 KOH
0.1 LiOH
0.1 LiOH
0.1 LiOH
0.01 H
3
3
3
3
BO
3
3
3
3
0.01 H
0.01 H
0.01 H
0.01 Al(NO
0.01 Ga(NO )
BO
BO
BO
3
)
3 3
3
a
Values for reactants in table are given as mole ratios. b Values in parentheses are the molar Si/Y ratios (Y: Ga, Al, or B) measured by elemental
analysis.
was synthesized using colloidal silica as the silicon source. After
conversion of the CIT-5 material to the acid form (mesh size
5-70), 100-400 mg of the catalyst was added to a downflow
reactor. Under a helium flow of 50 mL/min, the following
temperature program was carried out: the reactor was heated
to 448 K over 2 h and held at this temperature for an additional
order quadrupolar effects. 13C NMR spectra (75.45 MHz) were
measured using cross-polarization with a 1 H-90° pulse of 6.5
1
3
3
ms with a recycle delay of 1 s. C NMR chemical shifts are
referenced to adamantane (downfield resonance at 38.4 relative
to TMS). Spectral deconvolution and simulation were per-
formed using both the Bruker Linesim and the MacFID software
packages.
2
h, the temperature was then increased from 448 to 623 K
over 1.5 h and held at 623 K for 3 h, and finally, the temperature
of the reactor was reduced to 590 K over 25 min. Once the
helium flow rate had been reduced to 20 mL/min, the feed was
introduced by passing the helium flow gas through a saturator
with m-xylene at 283 K. The products were analyzed on-line
as previously described.
TEM Studies. Samples for transmission electron microscopy
(TEM) were prepared both by microtomy and by dispersion.
The samples prepared by dispersion were first crushed and then
sonicated in ethanol. A drop of this dispersion was transferred
to a carbon-coated Cu grid and allowed to dry. The samples
prepared by microtomy were embedded in an epoxy (LR White)
which was subsequently cured at 363 K for 2 h. Slices from
the hardened epoxy sample were prepared using a diamond knife
mounted on an ultra-microtome. The samples were dried in a
desiccator overnight prior to mounting in the TEM specimen
holder. The TEM, ED, and HREM images were taken on a
JEOL 2000 FX operating at 200 kV, a JEOL 100CX operating
at 100 kV, and a Phillips 430 operating at 200 kV, respectively.
Image analysis was performed using the “Image” software for
Macintosh.
Additional Characterization Studies. Scanning electron
micrographs were collected on a Camscan Series 2-LV electron
microscope operating at an acceleration voltage of 15 kV. A
DuPont 951 thermogravimetric analyzer was used to carry out
the thermogravimetric analysis using a constant heating rate of
5 K/min. The elemental analyses were performed by Galbraith
Laboratories (Knoxville, TN).
The catalyst used in the hydrocracking experiments was
prepared by ion-exchange of the palladium cations into the
ammonium-exchanged material. Palladium tetraamine dinitrate
was dissolved in a solution containing 2 mL of 0.156 N NH4-
OH in 9 mL of H2O (the quantity of palladium tetraamine
dinitrate dissolved was such that if all the palladium were to
exchange into the zeolite there would be 0.5 wt % Pd in the
catalyst). Next, 990 mg of the zeolite was introduced into the
solution, which was then aged statically at room temperature
for 3 days. The zeolite was filtered, washed with 200-300
mL of water, and then calcined to 755 K in air using the
following program: the catalyst was heated to 393 K and held
at this temperature for 2 h and the temperature was then ramped
to 755 K at 1 K/min and held at 755 K for 3 h. The material
was then pelleted, broken and meshed to 20-40, and jacketed
by quartz chips of the same mesh size. The packed zeolite
catalyst was dried in situ, and the reactor temperature was
reduced to 588 K. The inert carrier gas was replaced with a
hydrogen atmosphere at 1200 psi. The n-hexadecane was
introduced at a flow rate of 1 µL/min. The temperature of the
catalyst was gradually increased to promote a higher conversion
of the feed. After 167 h on-stream at 622 K, a 96% conversion
of n-hexadecane to isomers and lighter products was achieved.
NMR Studies. Solid-state NMR spectra were collected on a
Bruker AM 300 spectrometer. Samples were packed into 7 or
Results and Discussion
Synthesis. Pure-silica CIT-5 is formed from the following
synthesis mixture (mole ratio):
1 SiO :0.1 LiOH:0.2 MeSPAOH:40 H O
2
2
at a temperature of 448 K upon heating for 5 days. The optimal
+
synthesis conditions include Li cations ([LiOH]/[ SiO2] ) 0.1);
29
+
+
+
+
4
mm ZrO2 rotors and spun in air. Si NMR spectra (59.63
however, ratios of Li /M (where M can be either Na or
+
MHz) were obtained using magic angle spinning (MAS) at
spinning rates of 3-4 kHz, pulse widths of 6 ms, and recycle
delays of 30-60 s. Tetrakis(trimethylsilyl)silane was used as
the external reference material for 29Si and C NMR chemical
shift determination. All chemical shifts are reported in ppm
K ) as low as 3/1 do not hinder the formation of CIT-5 (Table
1). As the Li /M ratio is decreased to 1/1, CIT-5 is not formed
+
+
and the synthesis results in the production of a previously
13
18,20,21
reported, high-silica phase, SSZ-24
(International Zeolite
2
2
+
Association code: AFI). CIT-5 can be formed without Li
present in the synthesis mixture at much longer crystallization
times (>20 days), provided an alkali cation (K or Na ) is
present in low concentrations ([MOH]/[SiO2] ratios less than
27
relative to TMS.
Al NMR spectra were recorded at a
+
+
frequency of 78.2 MHz on samples packed into a 4-mm ZrO2
rotor spinning at 8-9 kHz. A pulse width of 4 ms correspond-
ing to a flip angle of π/2 along with a recycle delay of 1 s were
used. 27Al NMR chemical shifts are referenced to a 1.0 M Al-
-
0.05) and the ratio of [OH ]/ [SiO2] is held constant at 0.3
(synthesis results are summarized in Table 1). As the concen-
+
-
(NO3)3 solution () 0.00 ppm) and are not corrected for second-
tration of Na is increased at constant [OH ] ([NaOH]/[SiO2]

7142 J. Phys. Chem. B, Vol. 102, No. 37, 1998
Yoshikawa et al.
Figure 1. Scanning electron micrograph of CIT-5 crystallites.
ratio is increased from 0.05 to 0.1), SSZ-24 is formed to the
exclusion of CIT-5.
These synthesis results indicate that the organic methyl-
sparteinium cation weakly structure-directs CIT-5. The presence
+
of Li cations serves to accelerate the crystallization times, while
+
+
the presence of other alkali metal cations (Na or K ) at low
concentrations also weakly assists in the crystallization of CIT-
+
5
. At higher concentrations, the presence of the Na cations
dominates the structure direction toward SSZ-24, presumably
by affecting the silica dissolution kinetics or stabilizing species
in solution prior to nucleation, leading to the stabilization of
SSZ-24 growth units. The organic methylsparteinium cations
may then serve to further stabilize the forming SSZ-24 structure
by filling the pore system.
CIT-5 has also been synthesized in the presence of various
atoms that can substitute for silicon, e.g., gallium, boron, and
aluminum. The elemental analyses results for the aluminum-
containing CIT-5 (Al-CIT-5) indicate a silicon-to-aluminum ratio
27
27
of 95. The chemical shift of the Al resonance in the Al
MAS NMR spectrum from the as-synthesized Al-CIT-5 and the
catalytic data for the Al-CIT-5 material both confirm the
presence of tetrahedrally coordinated aluminum and the Bronsted
acid site formed from the framework aluminum (both results
will be discussed in greater detail below).
Scanning electron micrographs (SEMs) show the CIT-5
crystallites to have a thin flat plate morphology (Figure 1). TEM/
ED analysis of the flat plate crystals reveal the (301) lattice
fringes (Figure 2) running along the long axis of the crystals
and indicates that the b-axis (parallel to the channel direction)
coincides with the long axis of the crystals.
Figure 2. Transmission electron micrograph of pure-silica CIT-5
crystallite showing (301) lattice fringes indicting pore direction parallel
to the long axis of the crystallite.
the systematic absences. However, initial Rietveld refinement
of the starting model in this space group indicated an unac-
ceptably large variation in the Si-O bond distances and the
Structure Solution of CIT-5. Synchrotron powder X-ray
data (SPXRD) were collected from a sample of the calcined
pure-silica CIT-5. Indexation of the SPXRD revealed an
orthorhombic crystal class for the pure-silica CIT-5 sample with
refined unit cell parameters of a ) 13.6738(8) Å, b ) 5.0216-
1
7
O-Si-O bond angles.
To obtain a reasonable refinement of the SPXRD data, it was
necessary to reduce the topological symmetry of the space group
within the orthorhombic crystal class. Distance least squares
(DLS) refinement was used as a guide for indicating reasonable
structural parameters for the starting model in a given space
group. The DLS residual values for the six orthorhombic
3
23
(
3) Å, and c ) 25.4883(7) Å (V ) 1750.1 Å ). Initial analysis
of the systematic absences in the SPXRD data indicate body
centering consistent with six possible space groups.
The starting model for Rietveld refinement24 was obtained
by an iterative process of model building, distance least squares
1 1 1
subgroups of Imma (Imm2, I2 2 2 , Pmma, Pnma, Pnna, and
Pmna) are summarized in Table 2. Although the space group
Pmna gives the lowest DLS R value (0.0050), it is high in
comparison to other high-silica zeolite and indicates distortions
in the structural parameters. As a result, the four orthorhombic
subgroups of Pmna were also evaluated using DLS as a guide.
Of these subgroups, space group Pmn21 was found to give the
lowest R value (0.0032).
2
5
(
DLS) refinement of the atomic positions for the model, and
26
comparison of the simulated powder X-ray pattern (CERIUS)
to the experimental pattern. The model with the closest match
between the simulated X-ray powder pattern and the experi-
mental XRD was found to have maximal topological symmetry
of space group Imma (no. 74, standard setting), consistent with

Synthesis, Characterization, and Structure Solution of CIT-5
J. Phys. Chem. B, Vol. 102, No. 37, 1998 7143
TABLE 2: CIT-5 Space Group Analysis
preliminary Rietveld refinement
DLS
a
reduced ø2
space group
R-value
wR
p
R
p
Imma
0.0171
orthorhombic subgroups of Imma
Imm2
1 1 1
I2 2 2
Pmma
Pnma
Pnna
0.0150
0.0088
0.0153
0.0096
0.0076
0.0050
Pmna
0.1107
0.0853
91.23
orthorhombic subgroups of Pmna
Pnc2
P222
0.0047
0.0049
0.0036
0.0032
0.1040
0.1186
0.1051
0.1028
0.0771
0.0859
0.0784
0.0762
80.08
102.90
80.95
1
Pma2
Pmn2
1
77.33
a
DLS refinements carried out using the following orthorhombic unit
cell parameters (Å): 5.021, 13.697, 25.497.
Initial Rietveld refinements were carried out in the space
group Pmna and in each of its four orthorhombic subgroups
(Pnc2, P2221, Pma2, and Pmn21). The atomic coordinates
obtained from the DLS refinements were input as starting
models for the Rietveld refinements. The initial Rietveld
refinement procedures were carried out similarly for each of
the space groups, and the results are summarized in Table 2.
Examination of the DLS R values and the initial Rietveld wRp,
2
Rp and reduced ø values led to the selection of space group
Pmn21 for further refinements of the CIT-5 starting model.
The initial atomic positions for the Rietveld refinement of
CIT-5 in space group Pmn21 (no. 31, standard setting) were
obtained from the DLS refinement. During the initial stages
of the refinement, the scale factor and zero shift were refined
2
4
along with a 15-parameter shifted Chebyshev function for
Figure 3. Comparison of experimental synchrotron pattern (upper)
calculated (middle) and difference profile (lower) for the Rietveld
refinement of CIT-5 (λ ) 1.147 98 Å).
background subtraction. The lattice parameters and the peak
shape function parameters27 were then refined until convergence
of the R values was obtained. Soft geometric constraints for
the Si-O bond distances (d(Si-O) ) 1.610(10) Å) and
interatomic O-O distances (d(O-O) ) 2.610(10) Å) were
employed for the initial stages of the atomic position refinement.
The weighting factor for the soft geometric constraints was
gradually reduced. However, complete elimination of the soft
constraints caused slight distortions in the Si-O bond distances
and O-Si-O angles.
TABLE 3: Crystallographic and Refinement Data
space group
a
b
1
Pmn2 (no. 31)
13.67388(40) Å
5.02163(12) Å
25.48837(95) Å
1.14798 Å
298 K
4.0-50.0°
9200
0.005°
131
449
c
wavelength
data collection temp
profile range
no. of observables
step scan increments
no. of variables
no. of reflections
Table 3 provides a summary of the refinement details. All
structural parameters are within reasonable ranges for silicate
2
8
materials. The average bond distance, d(Si-O), is 1.591 Å
with a range of 1.555-1.641 Å. The average Si-O-Si angle
is 149.4° with a maximum of 168.7° and a minimum of 140.6°,
and the average O-Si-O angle is 109.4° with a maximum value
of 113.8° and a minimum value of 104.2°. The low Rietveld
residuals values (wRp ) 7.09%, Rp ) 5.40%) indicate a close
fit between the calculated and the experimental pattern (Figure
wR
p
7.09%
5.40%
R
p
av d(Si-O)
1.591(3) Å
1.555(0) Å
1.641(0) Å
149.4(5)°
140.6(0)°
168.7(0)°
109.4(5)°
104.2(0)°
113.8(0)
min
max
av Si-O-Si angle
min
max
av O-Si-O angle
min
max
3). This solution has been accepted by the International Zeolite
Association Structure Commision and the code CFI has been
established for CIT-5.
The structure of CIT-5 viewed down the b-axis (Figure 4) is
composed of one-dimensional, extra-large pores of nearly
circular cross section (center of oxygen to center of oxygen
TEM/ED Studies. The simulated, experimental, and image-
processed experimental high-resolution transmission electron
micrographs (HRTEM) of CIT-5 viewed down the 14-MR pores
are shown in Figure 5. Comparison of the experimental image
to the simulated image confirms the proposed topology. The
CIT-5 topology consists of zigzag ladders of 4-rings with
pendant 5-rings. These units are interconnected through single
zigzag chains to produce the pseudo-body-centered structure.
2
distance 9.91 × 9.87 Å ) circumscribed by 14 T-atoms. The
asymmetric unit contains 10 T-atoms and 19 oxygen atoms,
resulting in a unit cell content of [Si32O64], a framework density
3
3
of 18.3 T-atoms/1000 Å , and a density of 1.821 g/cm . Table
contains the final atomic positional parameters obtained from
the Rietveld refinement.
4

7144 J. Phys. Chem. B, Vol. 102, No. 37, 1998
Yoshikawa et al.
Figure 4. CIT-5 framework structure (bridging oxygen atoms omitted)
viewed down 14 MR pores ([010] zone).
TABLE 4: Refined Atomic Positional Parameters
(
1
Fractional Coordinates) in Space Group Pmn2 (no. 31)
and Isothermal Temperature Factors (Esd in Parentheses)
atom
x
y
z
U
iso(Å2)
Si1
Si2
Si3
Si4
Si5
Si6
Si7
Si8
0.1100(10)
0.000000(0)
0.2868(9)
0.000000(0)
0.1956(20)
0.3884(9)
0.3001(12)
0.500000(0)
0.500000(0)
0.2064(12)
0.1258(22)
0.1834(7)
0.1391(20)
0.000000(0)
0.000000(0)
0.0926(6)
0.000000(0) -0.071(4)
0.2742(13)
0.4029(5)
0.2337(14)
0.2433(16)
0.9053(6)
0.3164(13)
0.500000(0)
0.2333(14)
0.4043(6)
0.2980(4)
0.7300(5)
0.733(33)
0.2360(4)
0.250(4)
0.2082(34)
0.300(4)
0.764(4)
0.264(5)
0.789(4)
0.091(32)
0.242(4)
0.5933(32)
0.303(9)
0.425(4)
0.782(4)
0.0318(6)
0.2702(7)
0.1883(5)
0.2055(6)
0.1414(5)
0.4634(5)
0.3523(6)
0.2248(5)
0.2912(7)
0.3076(6)
-0.0139(6)
0.019(1)
0.048(9)
0.010(3)
0.041(2)
0.017(2)
0.027(5)
0.034(8)
0.016(7)
0.009(6)
0.038(8)
0.027(8)
Si9
Si10
O11
O12
O13
O14
O15
O16
O17
O18
O19
O20
O21
O22
O23
O24
O25
O26
O27
O28
O29
0.0794(33) -0.025(6)
0.0092(7)
0.0518(10)
0.2541(7)
0.3075(7)
0.2216(7)
0.1581(8)
0.1906(6)
0.1584(8)
0.2476(5)
0.1694(5)
0.4141(5)
0.4453(10)
0.3260(8)
0.3256(6)
0.3468(7)
0.2716(8)
0.2446(8)
0.013(5)
0.041(9)
0.039(5)
0.049(7)
0.005(5)
0.029(2)
0.003(5)
0.008(4)
0.023(6)
0.006(6)
0.065(5)
0.017(9)
0.037(2)
0.044(0)
0.039(3)
0.057(7)
0.003(1)
0.465(5)
0.796(4)
-0.036(4)
0.728(5)
0.3024(34)
0.238(5)
0.241(8)
0.083(4)
0.303(5)
0.584(4)
Figure 5. Transmission electron micrographs (TEMs) of CIT-5: (a)
experimental image, (b) experimental averaged image, (c) simulated
TEM image of CIT-5 topology, and (d) CIT-5 topology.
material has been deconvoluted into five peaks: -107.8 (11.4%),
-
108.3 (25.7%), -112.9 (26.7%), -113.2 (25.9%), and -114.9
(10.3%) (Figure 8c).
The basis set for the CIT-5 model in space group Pmn21
contains 10 T-atoms. However, half of the T-atoms in the
asymmetric unit are related by a pseudo-2-fold symmetry
operation at (1/4,1/4,z) that exists in the highest maximal
topology. The similarities in the chemical environments of these
T-atoms and the resolution limitations in the NMR intrumen-
tation reasonably account for the observed deconvolution of the
0.2529(15)
0.500000(0) -0.037(5)
0.500000(0) 0.464(4)
Electron diffraction patterns from the principal axes (Figure 6)
of pure-silica CIT-5 crystals contain considerable dynamical
diffraction effects. However, they show no signs of streaking,
indicating that the crystals are neither faulted nor disordered.
2
9
Si MAS NMR of the calcined Si-CIT-5 sample. The site
13
NMR Studies. Comparison of the liquid C NMR of N(16)-
methylsparteinium hydroxide with the MAS 13C NMR of the
as-synthesized CIT-5 (shown in Figure 7) indicates that the
structure-directing agent occluded in the pores of CIT-5 is intact
and has not decomposed under the synthesis conditions. The
populations for the CIT-5 model in space group Pmn21 [Si2 +
Si8 (4), Si1 + Si6 (8), Si3 + Si10 (8), Si5 + Si7 (8), Si4 +
Si9 (4)] corresponds to a relative intensity ratio of 1:2:2:2:1
consistent with the observed deconvolution.
27
The Al MAS NMR spectrum of the as-synthesized Al-CIT-5
shows a single peak at 54 ppm. This peak is assigned to
tetrahedrally coordinated aluminum and is a strong indication
that the aluminum present in the material has been incorporated
into the framework.
2
9
Si MAS NMR spectrum of the as-made and calcined CIT-5
29
are shown in Figures 8a and 8b. The Si NMR resonances
4
from -107 to -115 ppm lie within the typical Q range reported
for other high-silica zeolites.29 The NMR spectra of the calcined

Synthesis, Characterization, and Structure Solution of CIT-5
J. Phys. Chem. B, Vol. 102, No. 37, 1998 7145
Figure 7. 13C NMR of organic SDA, N(16)-methylspartienium: (a)
1
3
C MAS NMR of as-synthesized pure-silica CIT-5 with SDA occluded
in pores and (b) liquid 13C NMR of N(16)-methylspartienium bromide.
Figure 6. Electron diffraction patterns from pure-silica CIT-5: (a)
[010], (b) [001], and (c) [111]. The reciprocal lattice vectors are shown
for the principal zone axes [010] and [001].
Adsorption and Thermal Gravimetric Studies. The ad-
sorption capacities of various organic molecules in CIT-5 are
reported in Table 5. The adsorption capacity of 1,3,5-triiso-
propylbenzene is consistent with the presence of extra-large,
one-dimensional pores in CIT-5. The adsorption capacities of
the other organic molecules listed in Table 5 for CIT-5 are
similar to those reported for other large- and extra-large-pore
zeolites, e.g., UTD-116 and SSZ-24.
19,20
While the pore-mouth
openings of CIT-5 and SSZ-24 are similar, the CIT-5 topology
contains side pockets within the pore system (parallel to the c
direction) that results in a maximum free pore diameter of 10.72
Å and explains the differences in the uptake of large organic
molecules such as 1,3,5-triisopropylbenzene (∼8.5 Å kinetic
diamter). The calculated geometric void volume for CIT-5 is
Figure 8. 29Si MAS NMR of (a) as-synthesized pure-silica CIT-5, (b)
calcined pure-silica CIT-5, and (c) simulated spectra.
0.13 mL/g and compares well to the argon physisorption results
reporting 0.11 mL/g. In addition, the thermal gravimetric
analysis shows an 11.2 wt % loss from the SDA, N(16)-
methylsparteinium. This amount of organic corresponds to 1
SDA molecule per unit cell, resulting in a pore volume of 0.13
mL/g, consistent with the argon physisorption and the calculated
geometric void volume.
Thermal/Hydrothermal Stability. Thermal and hydrother-
mal studies were conducted using in situ X-ray diffraction to
determine the crystallinity of pure-silica CIT-5 samples as a
function of temperature. For investigation of the thermal
stability of CIT-5, a dry nitrogen atmosphere was maintained
inside the diffraction chamber while the sample temperature was

7146 J. Phys. Chem. B, Vol. 102, No. 37, 1998
Yoshikawa et al.
TABLE 5: Comparison of Adsorption Capacities for CIT-5
and Other Large and Extra-Large-Pore Molecular Sieves
TABLE 7: m-Xylene Isomerization for CIT-5 (Selectivities
Extrapolated to Zero Time On-Stream)
adsorption capacity (mL/g)
kinetic
molecular
sieve
initial p-xylene/o-xylene
diam (Å) CIT-5 SSZ-24a UTD-1a
pore size (nm)a
0.73
0.73
0.875
0.715
0.57
0.74 + mesopores
0.545
selectivity
adsorbate
CIT-5
0.60
0.77
0.66
0.75
1.27
1.12
2.68
argon
n-hexane
cyclohexane
3.0
4.4
6.0
6.2
8.5
0.108
0.092
0.090
0.093
0.041
0.109
0.101
0.114
0.128
0.011
0.110
0.121
0.111
0.119
0.111
SSZ-24
UTD-1
SSZ-31
ZSM-12
USY
2
1
,2-dimethyl butane
,3,5-triisopropyl benzene
a
Data from ref 16.
ZSM-5
a
Pore diameter of the largest pore. If pore is not circular, the average
TABLE 6: Constraint Index Conversion Data (C6 cracking)
for CIT-5 (SAR ) 200) at 427 °C
is used.
convn after
0 min
on-stream
convn after
430 min
on-stream
TABLE 8: Trimethylbenzene Selectivities for CIT-5 and
Other Zeolites
1
trimethylbenzene isomer (%)
feed conversion
-methylpentane
n-hexane
constraint index
product distribution
C6 isomerization
38%
52%
24.3%
0.38
19.7%
27%
12.5%
0.42
zeolite
CIT-5
ZSM-12
UTD-1
USY
SSZ-24
SSZ-31
ZSM-5
equilib at 350 °C
1,2,3
1,2,4
1,3,5
3
8
0
7
6.5
9
7
0
8
83.5
100
71
66
83.5
76
8.5
0
22
27.5
7.5
17
0
a
10.1%
84.1%
3.2%
12.9%
83.3%
1.1%
b
C5 minus cracking
c
aromatics
d
dehydrogenation
1.4%
1.3%
100
68
a
b
24
C6 products other than hexane and 3-methylpentane. C5 and lower
hydrocarbons from cracking (paraffins and olefins). For example,
benzene, toluene, etc. C6 olefins.
c
d
by the low ratio of p- to o-xylene isomers (p/o ratio). The p/o
ratio from CIT-5 is similar to the ratio of products from reaction
over one-dimensional large- and extra-large-pore zeolites SSZ-
increased from room temperature to 973 K in 100 K increments
as previously described. A water/nitrogen atmosphere was
maintained in the chamber during the hydrothermal stability
studies.
31, SSZ-24, and UTD-1. Zeolites with smaller pore sizes, such
as ZSM-12, show a higher p/o ratio, indicating some shape-
selectivity. As with C6 cracking, CIT-5 steadily deactivates
when accomplishing the m-xylene conversions, but the p/o ratio
does not change with time on-stream. In contrast, the p/o ratio
from reaction over UTD-1 steadily increases with time on-
stream, indicative of a selective narrowing of the pores by
coking.
The pure-silica CIT-5 samples are found to be both thermally
and hydrothermally stable to over 923 K. These results are
consistent with the thermal/hydrothermal stabilities of other
large- and extra-large-pore high-silica zeolites and, together with
the previously reported stability studies of UTD-1,16 indicate
that extra-large-pore materials can have the thermal and hy-
drothermal stability required for industrial application.
Catalytic Studies. Table 6 summarizes the results from the
cracking of C6 isomers (n-hexane and 3-methylpentane). The
values for the constraint index (CI ) log(fraction n-hexane
remaining)/log(fraction 3-methylpentane remaining)) calcu-
lated for the C6 cracking over CIT-5 indicate that the cracking
activity is consistent with the extra-large pores of CIT-5 (CI <
, indicating no shape-selective preference for the linear
hydrocarbon). Other large-pore zeolites such as SSZ-24 (CI )
.35, 73% conversion), SSZ-31 (CI ) 1.20, 50% conversion),
USY (CI ) 0.25, 50% conversion), and UTD-1 (CI ) 0.35,
8% conversion) show similar results. ZSM-5 (CI ) 5.0, 70%
conversion) and ZSM-12 (CI ) 1.8, 20% conversion), which
both have smaller pores, show some shape-selectivity (CI >
). As can be observed from the data shown in Table 6, the
catalyst does steadily deactivate. However, during the decline
in activity, the ratios reported are unaffected by the gradual
coking. These results indicate that there exist acid sites within
the material that are strong enough to crack alkanes and,
additionally, that there is no selective narrowing of the pores
or change in the reaction site as has been seen for zeolites beta
In addition to isomerization, m-xylene can disproportionate
into trimethylbenzenes (TMBs) and toluene. The distribution
of trimethylbenzene isomers reflects important structural features
32
such as pore size and shape. Larger-pore zeolites tend to have
a trimethylbenzene distribution that approaches equilibrium.
However, structures with smaller-sized void spaces tend to
suppress the formation of the bulky 1,2,3 and 1,3,5 isomers.
The data in Table 8 illustrate the TMB selectivity of CIT-5 and
several other zeolites. Reaction of m-xylene over CIT-5
produces trimethylbenzenes with a low abundance of the bulky
1,3,5 isomer relative to equilibrium or reaction over the larger-
pore UTD-1. Similar results are observed for the reaction over
SSZ-24, as might be expected.
3
0
1
0
7
For the experimental modeling of hydrocracking reactions,
the n-hexadecane feed was carried out at 96% conversion (for
comparison to other catalysts tested previously). Santilli and
1
33
co-workers described an inVerse shape-selective phenomenon
in which isomerized cracked products are preferred over straight-
chain hydrocarbons as the pore wall dimensions more closely
match the kinetic diameters of the potential products. This
behavior was shown to be particularly effective in the promotion
of dimethylbutane products from n-hexadecane hydrocracking
via the SSZ-24 catalyst with a pore diameter near 7.3 Å.33 Upon
comparison of the ratio of the formed dimethylbutanes to
n-hexane versus pore diameter, a bell-shaped curve is observed
where the ratio is low for pore systems below 6 Å (the
dimethylbutanes would be hindered from diffusing out of the
constricted pores); higher values are achieved for 12-ring
3
1
and MCM-22.
The acidic forms of Al-substituted (sample A) and Ga-
substituted (sample B) CIT-5 were used in m-xylene reactions.
Both samples showed similar product selectivities and activities.
The results described below are from sample B.
m-Xylene isomerization selectivity of CIT-5 and other zeolites
is shown Table 7. The large pore diameter of CIT-5 is evident

Synthesis, Characterization, and Structure Solution of CIT-5
J. Phys. Chem. B, Vol. 102, No. 37, 1998 7147
TABLE 9: Hydrocracking of n-Hexadecane. Comparison of
Dimethylbutanes to n-Hexane Product Ratios with Pore
Diameters
for their assistance in collecting the synchrotron powder XRD
data. The data were collected at the X7A beamline at the
National Synchrotron Light Source at Brookhaven National
Laboratory (Upton, NY), which is supported by the Department
of Energy, Division of Material Science and Division of
Chemical Sciences. Dr. Chuck Kibby of Chevron is thanked
for the collection of argon isotherms, Dr. Ronald C. Medrud of
Chevron is thanked for synchrotron powder XRD data collec-
tion, and Drs. Tom Harris and Bowman Lee also of Chevron
are thanked for the experimental hydrocracking and cracking
data. M.T. and M.L. acknowledge support from the David and
Lucile Packard Foundation. P.W. thanks Air Products and
Chemicals for financial support. Additional financial support
for this work was provided by Chevron.
molecular
sieve
diameter
(Å)
dimethylbutane/
dimensionality
n-hexane
ZSM-12
SSZ-31
SSZ-24
CIT-5
UTD-1
LTL
1
1
1
1
1
1
3
6.0
0.08
0.50
0.80
0.33
0.20
0.20
0.15
8.8 × 5.5
7.3
b
7.3
10.0 × 7.7
a
10
a
FAU
13
a
Multidimensional channel system. b Small side pockets within pores
give maximum free pore diameter of 10.7 Å.
zeolites, and then for larger pores the effect of the inverse shape-
selectivity is lost and the values drop again for this ratio. This
ratio for several zeolites and CIT-5 is presented in Table 9 along
References and Notes
(
(
1) Davis, M. E. Chem. Eur. J. 1997, 3, 1745-1750.
2) Davis, M. E. Chem. Ind. (London) 1992, 4, 137.
with the nominal pore-size data. A particularly interesting case
(3) Estermann, M.; McCusker, L. B.; Baerlocher, Ch.; Merrouche, A.;
is SSZ-3134 which has an elliptical, one-dimensional pore that
Kessler, H. Nature 1991, 352, 320-323.
(
(
(
4) Davis, M. E. Nature 1991, 352, 281.
5) Davis, M. E. Nature 1989, 337, 117.
6) Barrer, R. M.; Villiger, H. Z. Kristallografiga 1963, 128, 352.
reaches 8.8 Å in diameter across the major axis of the ellipse
but is limited to 5.5 Å in diameter across the minor axis. In
this case, the smaller cross section determines the product
distribution. SAPO-11, another molecular sieve with elliptical
pores and a diameter across the major axis greater than 6 Å,
also shows this restrictive behavior. Although the pore aperature
is 7.3 Å, CIT-5 does not give a ratio near the maximum of the
bell-shaped curve, as SSZ-24 does. Instead, the hydrocracking
over CIT-5 yields a lower ratio, which can be attributed to small
side pockets that give a maximum free pore diameter of 10.7
Å. The reaction data when taken in total show that CIT-5
provides for unique reaction behavior when compared to other
zeolites. This is not unexpected from the fact that it contains
a unique pore system.
(7) Smith, J. V.; Dytrych, W. J. Nature 1984, 309, 607.
(
(
8) Brunner, G. O.; Meier, W. M. Nature 1989, 337, 147.
9) Davis, M. E.; Saldarriaga, C.; Montes, C.; Garces, J.; Crowder,
C. Nature 1988, 331, 698.
(10) Jones, R. H.; Thomas, J. M.; Chen, J. S.; Xu, R. R.; Hou, Q. S.;
Li, S. G.; Ma, Z. J. Solid State Chem. 1993, 102, 202-208.
(
(
11) Loiseau, T.; Ferrey, G. J. Solid State Chem. 1994, 111, 403-415.
12) Loiseau, T.; Ferrey, G. J. Mater. Chem. 1996, 6, 1073-1074.
(13) Kahn, M. I.; Meyer, L. M.; Haushalter, R. C. Chem. Mater. 1996,
8, 43-53.
(
14) Schindler, M.; Joswig, W.; Bauer, W. H. Z. Anorg. Allg. Chem.
1
997, 623, 45-54.
(15) Freyhardt, C. C.; Tsapatsis, M.; Lobo, R. F.; Balkus, K. J.; Davis,
M. E. Nature 1996, 381, 295.
16) Lobo, R. F.; Tsapatsis, M.; Freyhardt, C. C.; Khodabandeh, S.;
(
Wagner, P.; Chen, C. Y.; Balkus K. J.; Zones, S.; Davis, M. E. J. Am.
Chem. Soc. 1997, 119, 8474-8484.
Conclusions
(17) Wagner, P.; Yoshikawa, M.; Lovallo, M.; Tsuji, K.; Tsapatsis, M.;
The structure of CIT-5, a new high-silica molecular sieve,
has been determined, and the material is shown to contain extra-
large pores circumscribed by 14 T-atoms. Rietveld refinement
of the synchrotron X-ray powder data gives the symmetry and
space group assignment for the structure of Pmn21 (no. 31) with
Davis, M. E. Chem. Commun. 1997, 2179-2180.
18) Lobo, R. F.; Davis, M. E. Microporous Mater. 1994, 3, 61.
(19) Zones, S. I.; Nakagawa, Y. U.S. Patent 5,225,179; 1993.
20) Bialek, R.; Meier, W. M.; Davis. M. E.; Annen, M. J. Zeolites 1991,
1, 438-442.
21) Richardson, J. W.; Smith, J. V.; Han, S. X. J. Chem. Soc., Faraday
(
(
1
(
refined unit cell parameters of a ) 13.6738(8) Å, b ) 5.0216-
Trans. 1990, 86, 2341.
(22) Meier, W. M.; Olson, D. H.; Baerlocher, Ch. Atlas of Zeolite
Structure Types, 4th ed.; Elsevier: London, 1996.
3
(
3) Å, and c ) 25.4883(7) Å (V ) 1750.1 Å ). All Si-O-Si
and O-Si-O bond angles and Si-O bond distances calculated
from the final atomic coordinates are within reasonable ranges
for silicate materials. TEM/ED indicates that the one-
dimensional pore material does not contain faulting defects and
that the pores run along the long axis of the crystals. Both the
adsorption (1,3,5-triisopropylbenzene uptake) and the catalytic
results (p/o ratio from m-xylene isomerization and the dimeth-
ylbutane/n-hexane product ratio from the hydrocracking of
n-hexadecane) support the structure assignment. CIT-5 is
synthesized hydrothermally in the presence of the organic,
N(16)-methylsparteinium, and lithium cations. 13C MAS NMR
indicates the organic is intact in the pores of the as-synthesized
material and has not degraded under synthesis conditions.
Stability studies reveal that CIT-5 retains its pore structure under
both high thermal and hydrothermal conditions. CIT-5 together
with UTD-1 are the only extra-large-pore, high-silica molecular
sieves that have high thermal/hydrothermal stability.
(
23) Werner, P. E.; Erikson, L.; Westdahl, M. J. Appl. Crystallogr. 1985,
18, 367.
(24) Larson A. C.; Von Dreele, R. B. Los Alamos Laboratory Report,
987, No. LA-UR-86-748.
25) Baerlocher, Ch.; Hepp, A.; Meier, W. M. DLS-76: A Fortran
1
(
Program for the Simulation of Crystal Structures by Geometric Refinement,
Institute fur Kristallographie, ETH: Zurich, Switzerland, 1977.
(
26) CERIUS Molecular Simulations, version 3.2; Cambridge, U.K.,
993.
27) Thompson, P.; Cox, D. E.; Hastings. J. B. J. Appl. Crystallogr.
1987, 15, 615-620.
28) Lewis, J.; Freyhardt, C. F.; Davis, M. E. J. Phys. Chem. 1996, 100,
1
(
(
5045-5049.
(
29) Englehardt, G.; Michel, D. High-resolution solid-state NMR of
silicates and zeolites; Wiley: New York, 1987.
(30) van Bekkum, H.; Flanigen, E. M.; Jansen, J. C. Introduction to
Zeolite Science and Practice; Elsevier: Amsterdam, 1991.
(31) Yuen, L. T.; Geilfuss, J.; Zones, S. I. Microporous Mater. 1997,
1
2, 229.
(32) Martens, J. A.; Perez-Pariente, J.; Sastre, E.; Corma, A.; Jacobs,
P. A. Appl. Catal. A 1988, 45, 85-101.
33) Santilli, D. S.; Harris, T. V.; Zones, S. I. Microporous Mater. 1993,
, 329-341.
34) Lobo, R. F.; Tsapatsis, M.; Freyhardt, C. F.; Chan, I. Y.; Chen, C.
Y.; Zones, S. I.; Davis, M. E. J. Am. Chem. Soc. 1997, 119, 3732-3744.
(
1
Acknowledgment. We gratefully acknowledge Dr. John
Higgins of Air Products and Chemicals Inc. and Dr. D. E. Cox
(