www.sciencemag.org/cgi/content/full/334/6062/1533/DC1
Supporting Online Material for
Growth of Uniformly Oriented Silica MFI and BEA Zeolite Films on
Substrates
Tung Cao Thanh Pham, Hyun Sung Kim, Kyung Byung Yoon*
*To whom correspondence should be addressed. E-mail: yoonkb@sogang.ac.kr
Published 16 December 2011, Science 334, 1533 (2011)
DOI: 10.1126/science.1212472
This PDF file includes:
Materials and Methods
SOM Text
Figs. S1 to S25
Tables S1 to S5
References (39–42)
Table of Contents
Materials
Methods
Preparation of leaflet shaped SL seed crystals
Preparation of rounded coffin shaped SL seed crystals in four different sizes
Preparation of pure silica beta zeolite seed (Si-BEA) seed crystals
Preparation of glass plates coated with a-oriented monolayer of silicalite-1 crystals (aSLm/g plates)
Preparation of glass plates coated with b-oriented monolayer of silicalite-1 crystals (bSLm/g plates)
Preparation of glass plates coated with a-oriented monolayer of Si-BEA crystals (a-SiBEAm/g)
Secondary growth of b-SLm/g in a TPA gel
Secondary growth of b-SLm/g in a t-TPA gel
Secondary growth of a-SLm/g plates in Gel-1
Secondary growth of b-SLm/g plates in Gel-2
Secondary growth of a-Si-BEAm/g plates in Gel-3
Laser Scan Confocal Microscope (LSCM) measurement
Inclusion of 1-bromododecane into b-SLf/g/b-SLf plates and analyses of the included
amounts
Inclusion of hemicyanine dyes (HC-n) into b-SLf/g/b-SLf plates and analyses of the
included amounts
Preparation of porous silica substrates
Assembly of SL monolayers on porous silica substrates
Secondary growth of b-SLm/p-SiO2 plates in Gel-2 (Preparation of perfect b-oriented SL
film on porous SiO2)
Separation of the o-/p-xylene mixture with b-SLf/p-SiO2
Instrumentation
SOM Text
Tables S1 to S5
Figs. S1 to S25
2
Materials
Aqueous tetramethylammonium hydroxide (TMAOH, 25%, Sigma-Aldrich), aqueous
tetraethyl ammonium hydroxide (TEAOH, 35%, Alfa), aqueous tetrapropyl ammonium
hydroxide (TPAOH, 1M, Sigma-Aldrich), tetra-n-butylammonium hydroxide (TBAOH,
40%, Sigma-Aldrich), ammonium hexafluoro silicate [(NH4)2SiF6, 98 %, SigmaAldrich)], tetraethylorthosilicate (TEOS, 98 %, Acros-Organic), and n-hexane (HPLC
grade, ≥ 95%, Sigma-Aldrich), Fluorescein free acid [2-(6-Hydroxy-3-oxo-(3H)-xanthen9-yl) benzoic acid 95 %, Aldrich], HF (48-51 %, A. C. S. reagent, J. T. Baker) were
purchased and used as received. Slide glass plates with the sizes of 75  25  1 mm3 were
purchased from Marienfeld. bis-N,N-(tripropylammoniumhexamethylene) di-N,Npropylammoniumtrihydroxide (trimer-TPAOH) was synthesized according to the
reported procedure. (6)
Methods
Preparation of leaflet shape SL seed crystals
Leaflet shape SL crystals with the average size of 0.3 × 1.3 × 1.5 μm3 were synthesized
according to the literature procedure. (6) For this, a gel consisting of TEOS, trimerTPAOH, KOH and distilled deionized water (DDW) was prepared, where the molar ratio
of the gel in terms of TEOS:trimer-TPAOH:H2O:KOH was 4.0:0.5:950.0:0.8. The rest of
the procedure is the same with the procedure for the secondary growth of a-SLM/g in
TPAOH gels. The SL powder sedimented at the bottom of the autoclave was collected by
centrifugation, washed with DDW to remove the mother liquor. The washed leaflet shape
SL crystals (fig. S1) were dried at 100 C by placing them in an oven.
Preparation of rounded coffin shape SL seed crystals in four different sizes
Rounded coffin shape SL crystals with the average size of 0.35 × 0.12 × 0.7 μm3 were
synthesized from a gel composed of TEOS, TPAOH, and H2O with the mole ratio of
6:3:330. The SL crystals with the average size of 1.0 × 0.5 × 1.4 μm3 were synthesized
from a gel composed of TEOS, TPAOH, and H2O with the mole ratio of 6:1.28:620. The
SL crystals with the average size of 1.5 × 0.6 × 1.9 μm3 were synthesized from a gel
composed of TEOS, TPAOH, and H2O with the mole ratio of 6:0.9:620. The above
synthesis gels were prepared by introducing TEOS into the solution containing TPAOH
and H2O. The mixture transformed into a clear gel after stirring for 24 h at room
temperature. The clear gel was filtered through a filter paper (Whatman® No.5) and
charged into a Teflon-lined autoclave. The hydrothermal reaction was carried out at 150
ºC for 12 h with vigorous stirring with the aid of a magnetic stirrer.
The rounded coffin shape SL crystals with the average size of 2.8 × 1.1 × 4.8 μm3 were
synthesized from a gel composed of TEOS, TPAOH, TEAOH, ethylene glycol (EG), and
H2O with the mole ratio of 6:0.9:0.6:24:600. The gel was prepared by first adding 31.8 g
of TEOS to the solution containing TPAOH (22.5 mL), H2O (247 mL), and EG (37.2 g).
The mixture was stirred for 24 h at room temperature, and TEAOH (6.17 mL) was added
into the mixture. The clear gel was aged for 12 h more at room temperature. The obtained
clear gel was filtered through a filter paper (Whatman® No.5) and charged into a Teflon3
lined autoclave containing a magnetic stirring bar. The hydrothermal reaction was carried
out at 150 ºC for 12 h with stirring at the spin rate of 490 rpm.
The obtained crystals were thoroughly washed with copious amounts of DDW to remove
the mother liquor. The crystals were then re-dispersed into a 25 % TMAOH solution and
shaken for 6 h to remove amorphous nanoparticles adsorbed on the crystals. The
TMAOH-treated crystals were washed with DDW until the supernatant solution became
neutral. The washed crystals were dried at 100 ºC by placing them in an oven overnight.
Preparation of pure silica beta zeolite seed (Si-BEA) seed crystals
Si-BEA zeolite was synthesized as the procedure described in the literature (25) with
some modifications. The gel consisting of fumed silica, TEAOH, (NH4)2SiF6, KOH and
DDW was prepared, where the molar ratio of the gel in terms of fumed
silica:TEAOH:(NH4)2SiF6:KOH:H2O was 4.00:1.92:0.36:0.40:31.20. The gel was
prepared as follows.
(I) Preparation of the fumed silica/TEAOH solution (solution I): TEAOH (35 %, 12.62 g),
DDW (0.91 g) and KOH (95 %, 0.60 g) were sequentially added into a plastic beaker and
continuously stirred by magnetic stirrer. Fumed silica (SiO2, 6.01 g) was slowly added
within 30 min with continuously stirring until all SiO2 became dissolved. This mixture
was stirred for additional 10 min until it became clear yellow and viscous solution.
(II) Preparation of the TEAOH/(NH4)2SiF6 solution (solution II): TEAOH (35 %, 7.57 g),
(NH4)2SiF6 (1.64 g) were introduced into a plastic beaker and stirred until it became a
homogeneous gel.
Solution II was transferred into solution I with vigorously stirring. The mixture was
stirred for additional 1 h until it solidified. The solidified mixture was aged under a static
condition for 24 h. After aging, the solid gel was ground using a food mixer until it
became pale yellow dry powder. It was transferred and packed into a Teflon-lined
autoclave. The hydrothermal reaction was carried out at 165 C. After 7 days of reaction,
the autoclave was removed from the oven and quickly cooled to room temperature by
running tap water onto them. The obtained crystals were thoroughly washed with copious
amounts of DDW and dried at 100 C by placing them in an oven overnight.
Preparation of glass plates coated with a-oriented monolayer of silicalite-1 crystals
(a-SLm/g plates)
Slide glass plates were washed by first placing them in a Piranha solution for 45 min
followed by rinsing them with copious amounts of DDW. The rinsed glass plates were
dried by blowing N2. The clean glass plates were coated with a thin layer of PDMS
(polydimethylsilane) by spin coating a PDMS solution (0.1 % in Hexane) at spin rate of
2,500 rpm for 15 sec. The PDMS layer was cured at 80 C for 1 h and etched with O2
plasma etching for 30 sec to make the surface hydrophilic. On the hydrophilic PDMScoated glass plates a solution of PEI in ethanol (0.1 % PEI) was coated by spin coating at
the rate of 2,500 rpm for 15 sec. Onto the PEI-coated glass plates leaflet shape SL
crystals were rubbed using a flat PDMS mold. The glass plates coated with the monolayer
4
of leaflet shape SL crystals were (denoted as a-SLm/g) calcined at 500 C for 10 h. The
rate of temperature increase from the room temperature to 500 C was 1.2 C/min and
after calcination the temperature of the furnace was decreased back to room temperature
with the same rate 1.2 C/min. After calcination, the supported seed crystals were washed
by placing the glass plates in a magnetically stirred NH4F solution (0.2 M) for 4 h. After
treatment with the NH4F solution, the a-SLM/g monolayers were washed with copious
amounts of DDW, and dried by blowing N2 gas.
Preparation of glass plates coated with b-oriented monolayer of silicalite-1 crystals
(b-SLm/g plates)
Monolayers of rounded coffin shape SL crystals were assembled on clean slide glass
plates (denoted as b-SLm/g) by directly rubbing the seed crystals on glass plates with a
finger, without coating the glass plates with any polymer glue. The monolayer coating
was conducted on both sides of the glass plates (denoted as b-SLm/g/b-SLm) when the
glass plates coated with continuous films of perfect b-oriented SL films were used as the
host materials for the production of 2-NLO films. Otherwise, the monolayer coating was
conducted only one side of each glass plate. The b-SLm/g and b-SLm/g/b-SLm plates were
calcined in a furnace at 550 C for 10 h. The rate of temperature increase from the room
temperature to 550 C was 1.2 C/min and after calcinations the temperature of the
furnace was decreased back to room temperature with the same rate 1.2 C/min.
Preparation of glass plates coated with a-oriented monolayer of Si-BEA crystals (aSi-BEAm/g)
A PEI solution in ethanol (0.3 % PEI) was coated on clean slide glass plates by spin
coating at the spin rate of 2,500 rpm for 15 sec. Monolayers of truncated bipyramid shape
Si-BEA crystals on the PEI-coated slide glass plates were prepared by rubbing Si-BEA
crystals with a latex glove wearing finger. The glass plates coated with the monolayer of
Si-BEA crystals (denoted as a-Si-BEAm/g) were calcined in a furnace at 550 C for 24 h.
The rate of temperature increase from the room temperature to 550 C was 1.2 C/min
and after calcinations the temperature of the furnace was decreased back to room
temperature with the same rate 1.2 C/min.
Secondary growth of b-SLm/g in a TPA gel
For the secondary growth of b-SLm/g plates in a TPA gel, a gel consisting of TEOS,
TPAOH, and DDW was prepared, where the molar ratio of the gel in terms of
TEOS:TPAOH:H2O was 4:1:600. TEOS (10.6 g) was added into the solution containing
TPAOH (8.4 mL) and DDW (84 g). This mixture was stirred at room temperature for 12
h. The obtained clear gel was poured into a Teflon-lined autoclave containing a Teflon
support having several b-SLm/g plates. The b-SLm/g plates were tilted by ~30 with the
SL monolayer side tilted down. The secondary growth was carried out under a static
condition in an oven preheated at 165 ºC for desired periods of time. After reaction, the
autoclave was removed from the oven and the reaction was quenched by quickly cooling
the autoclave with running tap water. The glass plates coated with randomly oriented
silicalite-1 crystals were removed from the autoclave and subsequently washed with
copious amounts of DDW. The SL powder sedimented at the bottom of the autoclave was
collected by centrifugation and washed several times with fresh DDW. The washed SL
5
crystals were dried at 100 C by placing them in an oven. The dried silicaite-1 powder
was weighed.
Secondary growth of b-SLm/g in a t-TPA gel
The secondary growth of b-SLm/g plates in a t-TPA gel was carried out according to the
literature procedure. (6) For this, a gel consisting of TEOS, trimer-TPAOH, and DDW
was prepared, where the molar ratio of the gel in terms of TEOS:trimerTPAOH:H2O:KOH was 4.0:0.5:950.0:0.8. The rest of the procedure is the same with the
procedure for the secondary growth of b-SLm/g in a TPA gels. The SL powder sedimented
at the bottom of the autoclave was collected by centrifugation washed with copious
amounts of DDW to remove the mother liquor. The washed silicaite-1 crystals were dried
at 100 C by placing them in an oven. The dried silicaite-1 powder was weighed.
Secondary growth of a-SLm/g plates in Gel-1
Gel-1 consisting of fumed silica, TEAOH, (NH4)2SiF6, KOH, and H2O with a mole ratio
of 4.00:1.92:0.36:0.40:n1, where n1 = 30 to 80 was prepared as follows. (A typical
procedure)
(I) Preparation of the fumed silica/TEAOH solution (solution I): TEAOH (35 %, 12.62 g),
DDW (0.91 g) and KOH (95 %, 0.60 g) were sequentially added into a plastic beaker and
continuously stirred with the help of a magnetic stirrer. Fumed silica (SiO2, 6.01 g) was
slowly added into the above solution during the period of 30 min with continuous stirring
until all SiO2 became dissolved. This mixture was stirred for additional 2 min until it
became clear yellow and viscous.
(II) Preparation of the TEAOH/(NH4)2SiF6 solution (solution II): TEAOH (35 %, 7.57 g),
(NH4)2SiF6 (1.64 g) were introduced into a plastic beaker and stirred until it became
homogeneous.
Solution II was transferred into solution I with vigorously stirring. The mixture was
stirred for additional 1 h until it solidified. The solidified mixture was aged under a static
condition for 12 h. After aging, the solid gel was ground using a food mixer until it
became pale yellow dry powder. It was transferred into a Teflon-lined autoclave. The aSLm/g plates were placed vertically inside the solid gel. To induce good contacts between
a-SLm/g plates and the solid gel, the autoclave was tapped by gently hitting the bench top.
The sealed autoclaves were placed in an oven preheated 150 ˚C. After desired periods of
time, the autoclaves were removed from the oven and quickly cooled to room
temperature by running tap water onto them. The perfect a-oriented films supported on
glass (denoted as a-SLf/g) were removed from autoclaves and washed with DDW to clean
the surface, and dried by blowing N2 gas.
Secondary growth of b-SLm/g plates in Gel-2
A gel consisting of TEOS, TEAOH, (NH4)2SiF6, and H2O (denoted Gel-2) was prepared,
where the molar ratio of the gel was 4.00:1.92:0.36:n2, where n2 = 40-80. The gel was
prepared as follows. (A typical example).
6
(I) Preparation of the TEOS/TEAOH solution (solution I): TEAOH (35 %, 20.2 g) and
DDW (22.2 g) were sequentially added into a plastic beaker containing 31.8 g of TEOS
(98 %). This beaker containing the above solution was tightly covered using plastic wrap
and magnetically stirred for about 30 min until the solution became clear.
(II) Preparation of the TEAOH/(NH4)2SiF6 solution (solution II): TEAOH (35 %, 10.1 g),
(NH4)2SiF6 (2.45 g), and DDW (11.1 g) were introduced into a plastic beaker and stirred
until all (NH4)2SiF6 became dissolved.
Solution II was quickly poured into the solution I with vigorously stirring. The mixture
solidified immediately. The solidified mixture was stirred for additional 2 min using a
plastic rod, and aged under a static condition for 6 h. After aging, the semisolid gel was
ground using a food mixer and transferred into a Teflon-lined autoclave. The b-SLm/g
plates were placed vertically within the semisolid gel. The sealed autoclaves were placed
in an oven preheated at desired temperatures. After desired periods of time, the autoclaves
were removed from the oven and quickly cooled to room temperature by running tap
water onto them. The perfect b-oriented SL films supported on glass (denoted as b-SLf/g)
were removed from autoclaves and washed with DDW to clean the surface. After
cleaning the films by sonication in a sonic bath charged with DDW for 1 min, they were
washed with DDW, and dried under the stream of N2 gas. The obtained b-SLf/g plates
were calcined at 550 C for 15 h under the oxygen flow to remove TEA+ ions from the
channels. The rate of temperature increase from the room temperature to 550 C was 1
C/min and after calcination the temperature of the furnace was decreased back to room
temperature with the same rate. The calcined films were washed with the 0.2 M NH4F
solution for 6 h and subsequently with copious amounts of water to remove the
amorphous silica layers or particles from film surfaces to expose channel openings to the
atmosphere.
Secondary growth of a-Si-BEAm/g plates in Gel-3
The a-oriented continuous Si-BEA film on glass plates (denoted as a-Si-BEAf/g) was
prepared from Gel-3 (TEOS:TEAOH:HF:H2O = 4.00:2.20:2.20:n3, where n3 = 30-40).
The gel was prepared as follows. (39)
TEAOH (35 %, 23.14 g) and TEOS (98 %, 21.2 g) were introduced into a plastic beaker.
This beaker containing the above solution was tightly covered using plastic wrap and
magnetically stirred for about 30 min until the solution became clear. HF (50 %, 2.20 g)
was added drop-wise into the above clear solution with vigorously stirring. The mixture
solidified immediately. The solidified mixture was stirred for additional 2 min using a
plastic rod, and aged under a static condition for 5 h. After aging, the semisolid gel was
ground using a food mixer and transferred into a Teflon-lined autoclave. The a-SiBEAm/g plates were placed vertically within the semisolid gel. The sealed autoclaves
were placed in an oven preheated at 150 C. After 4 day of reaction time, the autoclaves
were removed from the oven and quickly cooled to room temperature by running tap
water onto them. a-Si-BEAf/g plates were removed from autoclaves, washed with DDW
to clean the surface and dried by blowing N2 gas.
7
Laser Scan Confocal Microscope (LSCM) measurement
The LSCM measurements were carried out with two type of membrane including random
oriented silicalite-1 films supported on porous silica substrates (denoted as r-SLF/p-SiO2)
and b-SLf/p-SiO2. The calcined membranes were mounted on a home-made permeance
cell. The zeolite site was contacted to pure MeOH while the support site was contacted to
fluorescin (see below) solution 0.1 M in MeOH. The contact areas were sealed by O-ring.
After 4 days for dye inclusion at room temperature, the membranes were removed and
washed with copious amount of MeOH, dried by blowing N2 gas, and kept at room
temperature for 12 h.
The LSCM measurements were conducted using LSM-710 (Carl Zeiss) with Argon laser
source (488 nm) and z-stack scan mode. The r-SLF/p-SiO2 membrane was measured at
laser power of 3.5 % using Plan-Apochromat 40/0.95 Korr M27 objective lens with the
zoom value of 0.6 and the master gain value of 547. The b-SLF/p-SiO2 membrane was
measured at laser power of 6.5 % using Plan-Apochromat 40/0.95 Korr M27 objective
lens with the zoom value of 2.0 and the master gain value of 700. The 3D images were
built using ZEN 2009 Light Edition software (Carl Zeiss).
Molecular structure of fluorescein [2-(6-hydroxy-3-oxo-(3H)-xanthen-9-yl) benzoic
acid].
O
Absorption maximum: 496 nm
O
OH
O
OH
Inclusion of 1-bromododecane into b-SLf/g/b-SLf plates and analyses of the included
amounts
The inclusion of 1-bromododecane (1-Br-C12) into b-SLf/g/b-SLf plates was carried out
by immersing them in neat 1-Br-C12 under vacuum. Four b-SLf/g/b-SLf plates which were
calcined calcined and washed with NH4F (18  25  1 mm3) were first evacuated at 300
ºC for 24 h to dehydrate the films. The dehydrated b-SLf/g/b-SLf plates were transferred
into a Schlenk flask in a glove box charged with dry Ar. 1-Br-C12 (5 mL) was added into
the Schlenk flask containing dry b-SLf/g/b-SLf plates and the Ar gas residing inside the
Schlenk flask was removed by briefly applying vacuum through the connection of the
side arm of the Schlenk tube to a vacuum line connected to the external (outside the glove
box) vacuum system. After disconnection from the vacuum line the tightly capped
Schlenk tube was inserted into an aluminum block whose temperature was maintained at
50 C. After 3, 5, and 7 days, one (after 3 and 5 days) or two (after 7 days) b-SLf/g/b-SLf
plates were removed from the Schlenk tube and washed the surface-coating 1-Br-C12
molecules off the b-SLf/g/b-SLf plate by flowing 15 mL of n-hexane onto the plate.
The profiles of the relative concentrations of 1-Br-C12 in SL channels along the film depth
8
were obtained from a 1-Br-C12-incorporating b-SLf/g/b-SLf plate by energy dispersive Xray spectroscopic (EDX) analyses of Br and Si. From the two 7-day 1-Br-C12incorporating b-SLf/g/b-SLf plates 1-Br-C12 was also extracted as follows. The 7-day
included plates were introduced into a plastic beaker containing 3 mL of HF solution (3
M). After gentle shaking for 5 min, the glass plates were removed from the solution and
washed them with additional 1 mL of HF solution (3 M) in the plastic beaker. The
collected HF solution was cooled to ~0 C by placing the plastic beaker in an ice bath. An
aqueous NaOH solution (3 M, 4 mL) was added drop wise into the cold HF solution.
After warming up the aqueous mixture to room temperature 8 mL of n-hexane was added
into the aqueous mixture. The mixture was shaken for 1 min and subsequently transferred
into a separatory funnel. After standing still for 10 min the upper organic phase was
transferred into a 25-mL volumetric flask. The lower aqueous phase was transferred back
to the separatory funnel and 8 mL of fresh n-hexane was added into the funnel. After the
cycle of shaking, standing still, and separation, the n-hexane layer was transferred into
the 25-mL volumetric flask already having ~8 mL of the first cycle extract. The extraction
procedure was repeated one more time using 8 mL of n-hexane. Into the 25-mL
volumetric flask fresh n-hexane was added until the total volume of the n-hexane solution
became 25 mL. The concentration of 1-Br-C12 was analyzed from the area of the
chromatogram after injecting 5 μL of the solution into a FID-GC equipped with a HPINNO Wax column. A calibration column was independently made for the concentration
analysis.
To check the accuracy of our analytical procedure we carried out the following simulated
experiment using a known amount of 1-Br-C12. In a plastic beaker, 5 mg of freshly
calcined SL powder and two slide glass plates with the same size (18 × 25 × 1 mm3) were
introduced. Into the plastic beaker 3 mL of HF solution (3 M) were added. After gentle
shaking until silicalite-1 powder was completely dissolved, these glass substrates were
removed from solution and washed with 1 mL of HF solution (3 M). The HF solution was
first cooled to ~0 C by placing the plastic beaker in an ice bath. Into the HF solution 4
mL of NaOH (3 M) was added to neutralize the solution. After warming up, 5 mL of an
n-hexane solution of 1-bromododecane (200 mM) was added into the neutral solution.
The rest of the extraction and analysis procedure was the same. The obtained recovery
was 99.3 %.
The key notes used to calculate the incorporated amount of 1-Br-C12 in an SL channel are
as follows.
Film type: b-SLf/g/b-SLf
Thickness of the film on each side = 3 m
Area of the film in one side = 17 × 25.8 mm2 (slide glass)
Number of channels in one side = 3.25  1014 channels.
Number of channels in two sides = 3.25  1014  2 channels.
Total number of 1-bromododecane molecule = 9.4322x1017
Exp. Nc = 1448.063
Corrected total Nc = 1448.063  (extraction factor: 100 / 99.25) = 1,459
9
Molecular length = 18.051 Å
Total length = 2.634 μm
Occupied = 87.8 %
Inclusion of hemicyanine dyes (HC-n) into b-SLf/g/b-SLf plates and analyses of the
included amounts
The synthesis of hemicyanine dyes with different alkyl chain lengths (HC-n) and their
inclusion into SL films supported on glass plates are well described in our previous report.
(14) For this, each NH4F washed b-SLf/g/b-SLf plate was divided into 6 pieces with the
size of 12 × 25 × 1 mm3. The glass supported small b-SLf/g/b-SLf plates were calcined at
385 C for 15 h under the oxygen flow and used immediately after calcination. Into each
vial containing a methanol solution of difference HC-n (n = 6, 9, 12, 15, 18, 22), three
pieces of b-SLf/g/b-SLf plates were added. These vials were capped and kept at room
temperature for 1 week. The films were removed from each solution and washed with
fresh methanol, and dried by blowing N2 gas. The analyses of the number of HC-n
molecules incorporated in each channel (NC) and the second harmonic (SH) intensity
measurements were carried out according to the procedures described in our previous
report. (14)
Preparation of porous silica substrates
Porous silica substrates were prepared from 50-550 nm sized silica beads which were
synthesized according to the Stöber method. (40) For this, 10 g of 350-nm SiO2 beads and
10 g of 550-nm SiO2 beads were mixed together using food mixer. Into the mixed silica
beads 0.6 mL of aqueous solution of Na2SiO3 (0.5% in DDW) was added drop wise and
the silica bead mixture was ground for 10 min in food mixer. Porous silica supports were
prepared by placing 1.8 g of the above mixture in a home-made stainless steel mold and
pressing at the pressure of 150 kgf/cm2. The resulting silica dishes were calcined at 1,020
C for 2 h with the heating rate of 100 C/h. After cooling to room temperature, both
sides of the porous silica disc were polished using a SiC sandpaper (Presi, grit size P800).
To make the surface smooth, one side was polished again using a SiC sandpaper (Presi,
grit size P1200). The diameter and thickness of the porous silica disc were 20 and 3 mm,
respectively. The porosity measured by a mercury porosimeter is 45.5 % with the average
pore size of 250 nm.
One drop of DDW was dropped onto a porous silica support. Independently, 70-nm silica
beads were prepared and calcined at 550 ºC for 24 h. The calcinced 70-nm silica beads
were gently rubbed on the porous silica supports until the surface became shiny. The
shiny porous silica supports were dried overnight at room temperature and sintered at 550
C for 8 h on a muffle furnace. The temperature was increased to 550 C during the 8 h
period and cooled to room temperature during the period of 4 h. An acetone solution of
epoxy resin (10 wt%) was spin coated onto the porous silica at the speed of 3,000 rpm for
15 sec and cured at 80 ºC for 30 min.
Assembly of SL monolayers on porous silica substrates
Onto the epoxy-coated porous silica supports an ethanol solution of polyethyleneimine
(PEI, 0.1 %) was spin-coated with the spin rate of 2,500 rpm for 15 sec. Perfect boriented SL crystals (1.0 × 0.5 × 1.4 μm3) were assembled on the porous supports by
10
rubbing them onto the supports using a finger. The SL crystal monolayer supported on
porous silica is denoted as b-SLm/p-SiO2. The b-SLm/p-SiO2 plates were calcined at 550
ºC for 24 h in air on a tubular furnace to remove the organic polymer layers as well as to
fix the SL monolayers on the silica supports through the formation of Si-O-Si bonding.
The rate of temperature increase was 65 ºC/h. The rate of temperature decrease was 100
ºC/h.
The calcined b-SLm/p-SiO2 plates were kept in a constant humidity chamber overnight to
allow the plates to absorb H2O. The hydrated b-SLm/p-SiO2 plates were then immersed
into an aqueous NH4F solution (0.2 M) for 5 h. The NH4F-treated b-SLm/p-SiO2 plates
were immersed in fresh DDW for 1 h and dried at room temperature for 24 h.
Secondary growth of b-SLm/p-SiO2 plates in Gel-2 (Preparation of perfect b-oriented
SL film on porous SiO2)
b-SLm/p-SiO2 plates were placed vertically in Gel-2. The hydrothermal reactions were
carried out at 165 C for 18 h. After the reaction, the produced perfect b-oriented SL films
supported on porous SiO2 substrates (denoted as b-SLf/p-SiO2) was removed and washed
with copious amounts of DDW. To remove the alkali in the porous SiO2 support, the bSLf/p-SiO2 membranes were immersed in DDW for 2 h and subsequently in a NH4F
solution (0.2 M) for 4 h. The membranes were then washed with DDW, dried by blowing
N2 gas, and kept at room temperature for 24 h. Finally they were calcined at 440 ºC for 8
h in air to remove TEAOH template. The heating rate was 60 ºC/h and the cooling rate
was 90 ºC/h. The calcined membranes were kept in a desiccator for permeation test.
Separation of the o-/p-xylene mixture with b-SLf/p-SiO2
The separation of the xylene mixture was carried out according to the Wicke-Kallenbach
method (41) (fig. S25). A b-SLf/p-SiO2 membrane was mounted on a home-made
stainless steel cell. AS-568A O-rings (Kalrez® , DuPont Performance Elastomers) were
used as the sealing materials. The active area was 2.0 cm2. Helium was passed through
the xylene mixture placed in a container whose temperature was kept at 25 ºC. This vapor
stream was mixed with a second He stream in a mixer. This xylene vapor was fed into the
feed side of the membrane. The total flow rate in the feed side was maintained at 60
mL/min. The p- and o-xylene vapor pressures in the feed side were 0.32, and 0.31 kPa,
respectively. Helium with the flow rate of 15 mL/min was used to sweep the permeate
side. The total pressure on both sides was atmospheric pressure. The separation cell was
mounted in a convection oven. To prevent the condensation, all the lines of system were
kept at 110 ºC by tape heater. The permeance tests were conducted at a desired
temperature to which the temperature was increased slowly at the rate of 1 ºC/min from
the room temperature. A fresh membrane was used for each test at different temperatures.
During the temperature increase, pure He gas was passed to both sides of membrane.
For permeance measurements, the gas stream of the permeate side was passed to a GC
through a 6-port valve. The concentrations of the components (p- and o-xylene) were
analyzed by the GC chromatogram areas. The area-concentration curve was obtained
before the membrane tests for each component by passing reference streams of He with
different concentrations of each component.
11
The permeance (P in mole s-1 m-2 Pa-1) is defined as the flux (F in mole s-1 m-2) of a
component M over the difference in the partial pressure of M between the feed and
permeate sides (eq. 1).
P = F/p
(1)
The separation factor (αp/o) is defined as the ratio of the mole fractions of the para isomer
(fp) with respect to the ortho isomer (fo) at the feed and permeate sides (eq 2).
αp/o = [(fp/fo)]permeate/[(fp/fo)]feed
Instrumentation
Scanning electron microscopy (SEM) images were obtained using a field-emission
scanning electron microscope (Hitachi S-4300) operating at an acceleration voltage of 20
kV and JEOL (JSM-7600F) operating at an acceleration voltage of 15 kV. Elemental
analyses of the samples were carried out by analyzing the energy-dispersive X-ray (EDX)
spectra of the samples using a Horiba EMAX 6853-H EDX spectrometer. Transmission
electron microscopy (TEM) images were collected on a JEOL JEM 4010 microscope.
Powder X-ray diffraction (XRD) patterns were obtained using a Rigaku D/MAX-2500/pc
diffraction meter. Gas chromatographic analyses were carried out on a HP 6890 GC
equipped with a HP INNO Wax Polyethylene Glycol (HP-19091N-136) column. The
laser-scanning confocal microscope (LSCM) images were obtained using a LSM-710
(Carl Zeiss) equipped with an argon ion laser source (488 nm) and z-stack scan mode.
12
SOM Text
This result shows that the fast in-plane growth along the c direction significantly
contributes to interconnection between the b-oriented SL crystals. Furthermore, as
Tsapatsis and coworkers pointed out (6, 8) the higher growth rate along the b (out-ofplane) direction than the a (in-plane) direction seems to help preserving the b-orientation.
Moreover, it appears that during growth, the growth along the c-axis is effectively
suppressed yielding grains with a isometric cross section which further points to a
balanced in-plane vs. out-of-plane growth.
13
Table S1-S5
Table S1. Thermal expansion coefficients of SL in several temperature ranges.
thermal
expansion
coefficients (10-6 °C-1)
αa
αb
αc
αV
*
25 - 150 °C
-13.518
+8.853
+7.180
+2.647
Temperature range
150 - 600 °C
-4.204
-1.339
-2.159
-7.602
600 - 750 °C
-3.990
-5.698
-2.493
-12.080
Data from (32, 33)
The SL crystals have complex temperature-dependent anisotropic thermal expansion
coefficients, αa, αb, αc, and αV, where the subscripts a, b, c, and V denote the principal
axes and volume, respectively. Accordingly, each crystal undergoes anisotropic thermal
expansion during the initial stage and anisotropic thermal contraction during the later
stage of calcination, that is, in the 25-150 and 150-550 °C regions, respectively.
14
Table S2. Performances of the glass places coated with HC-n-incorporating uniformly boriented SL films on both sides.
Thickness (nm)*
HC-n
130
2,400
2,700
3,000
Nc†
I2w‡
d33§
Nc†
I2w‡
d33§
Nc†
I2w‡
d33§
Nc†
I2w‡
d33§
6
10.3
2.1
6.78
10.2
0.17
0.10
11.4
0.8
0.21
12.9
1.0
0.15
9
13.4
8.7
13.65
16.2
20.6
1.09
31.9
20.1
1.00
33.9
19.2
0.88
12
16.5
51.1
32.95
42.9
104.1
2.97
50.0
116.9
2.43
60.5
113.8
2.18
15
15.9
58.9
35.42
50.0
144.5
2.93
66.2
158.7
2.86
71.9
174.5
2.68
18
12.6
41.1
29.72
40.4
126.9
2.72
51.5
134.5
2.68
53.3
170.9
2.64
22
9.90
43.6
30.48
38.5
108.0
2.53
43.9
119.4
2.47
47.5
159.3
2.55
SL film thickness in nm. †Number of HC-n dyes in each channel. ‡Relative second
harmonic intensity of the HC-n-including SL film with respect to that of a 3-mm y-cut
quartz as the reference . §A corresponding polarizability tensor component.
*
15
Table S3. Values of tensor components for the quadratic nonlinear susceptibility of
a 2-NLO materials.
*
Materials
dnm (pm/V)*
Quartz
0.364 (d11)
LiNbO3
2.76 (d22)
BBO
2.22 (d22)
0.16 (d31)
KTP
6.5 (d31)
5.0 (d32)
13.7 (d33)
COANP
10.0 (d33)
Data from (42)
H
N
CH2OH
NO2
COANP: 2-cyclooctylamino-5-nitropyridine
16
Table S4. Performances of the glass plates coated with HC-n-incorporating randomly
oriented SL films on both sides prepared from glass plates coated with SL films which
were prepared by secondary growth of b-SLm/g plates in TPA and t-TPA gels, respectively.
From TPA gel (400 nm)
From t-TPA gel (1,300 nm)
HC-n
*
Nc*
I2ω†
d33‡
Nc*
I2ω†
d33‡
6
23.1
0.1
0.50
5.6
0.23
0.22
9
15.4
1.6
2.25
7.6
0.23
0.22
12
8.20
3.8
3.59
8.0
2.44
0.69
15
5.70
7.0
4.99
12.6
7.08
1.21
18
3.50
7.9
5.30
8.8
2.83
0.76
22
0.90
0.9
1.71
8.0
2.72
0.75
†
Number of HC-n dyes in each channel. Relative second harmonic intensity of the HC-nincluding SL film with respect to that of a 3-mm y-cut quartz as the reference. ‡A corresponding
polarizability tensor component.
17
Table S5. Comparison of the characteristics and performances of the uniformly b-oriented
SL membranes prepared by our method with the SL membranes prepared by other groups.
Orientation
p-xylene permeance
[10-10 mol s-1 m-2 Pa-1]
SF*
Temp.
[C]
Calcination
method
Ref
(in text)
random
0.5
2,700
17
400
C†
5
b
1.0
2,460
378
150
C†
6
b
1.0
1,960
483
200
C†
6
60
150
C†
7
~5,000
200
-
12
~1,000
150
C†
This work
random
-
270
random
-
-
b
*
Thickness
[μm]
2,100-500
1.0
†
Separation factor. Conventional slow temperature rising and slow temperature cooling.
‡
Rapid thermal processing.
18
Figure S1-S25
2 µm
Fig. S1. A SEM image of leaflet shaped SL crystals used in this work.
19
A
B
5 µm
5 µm
C
D
5 µm
10 µm
Fig. S2. SEM images of SL crystals with the sizes of 0.35 × 0.12 × 0.7 (A), 1.0 × 0.5 ×
1.4 (B), 1.5 × 0.6 × 1.9 (C), and 2.8 × 1.1 × 4.8 mm3 (D).
20
25 µm
25 µm
Fig. S3. SEM images of Si-BEA crystals with the sizes of 14 × 14 × 19 m3 in two
different magnifications.
21
A
B
C
10 µm
25 µm
25 µm
5 µm
10 µm
10 µm
Fig. S4. SEM images (top and side views) of SL films supported on glass which were pre
pared from b-SLm/g plates by secondary growth in a TPA gel (TEOS:TPAOH:H2O = 4:1:
600) at 165 ⁰C for 3 (A), 6 (B), and 24 h (C).
22
A
D
5 µm
B
E
25 µm
C
25 µm
F
1 µm
1 µm
Fig. S5. SEM images of SL films prepared by secondary growth of b-SLm/g plates in a T
PA gel which was pre-heated at 150 °C for 2 h to consume most of the nutrients (in other
words to dilute the concentration of the nutrients) according to the method of (21) after 3
(A-C) and 6 h (D-F).
23
A
B
C
5 µm
5 µm
5 µm
5 µm
5 µm
5 µm
Fig. S6. SEM images (top and side views) of SL films prepared from b-SLM/g by seconda
ry growths in a t-TPA gel at 175 ⁰C for 3 (A), 12 (B), and 24 h (C).
24
A
B
5 µm
5 µm
C
D
5 µm
5 µm
E
F
5 µm
25 µm
G
H
25 µm
10 µm
Fig. S7. SEM images (top and side views) of SL films prepared from a-SLm/g plates by s
econdary growths in the TPA gel at 175 ⁰C for 3 (A, B), 6 (C, D), 12 (E, F), and 24 h (G,
H).
25
A
B
2 µm
2 µm
C
D
2 µm
2 µm
Fig. S8. SEM images of SL films prepared from a-SLm/g plates by secondary growths in
the t-TPA gel at 175 ⁰C for 3 (A), 6 (B), 12 (C), and 24 h (D).
26
A
25 µm
BEA seed gel
165 ⁰C – 3 d
B
MFI gel / TPAOH
165 ⁰C - 24 h
D
F
25 µm
50 µm
C
E
25 µm
Zeolite Y gel
100 ⁰C - 24 h
50 µm
G
20 µm
5 µm
Fig. S9. SEM images of Si-BEA films prepared from a-Si-BEA/g plates (A) by secondar
y growths in the Si-BEA seed gel at 175 ⁰C for 3 d (B, C), in the TPA gel at 175 ⁰C for 2
4 h (D, E), and in a gel for the synthesis of zeolite Y (Y gel) at 100 ⁰C for 24 h (F, G).
27
B
A
10 µm
25 µm
D
C
10 µm
10 µm
Fig. S10. SEM images of Si-BEA films prepared from randomly oriented-Si-BEA/g plate
s (A) by secondary growths in Gel-3 at 165 C for 4 d. Top views with different magnific
ations (B, C) and side view (D).
28
Selectively to film (Sf), %
110
100
90
80
70
60
50
40
30
20
10
0
Gel-2
TPA gel
3
6
9
12
TPA gel
15
18
21
24
Reaction time (h)
Fig. S11. Plot of Sf values with respect to time for the cases of secondary growth of b-SL
m/g plates in Gel-2, TPA gel, and t-TPA gel.
29
8
1.1 (m)
1.1
1.1
1.1
0.6
0.6
0.6
Thickness (m)
7
6
5
4
0
180 ( C)
165
150
140
180
165
150
3
2
1
0
0
1
2
3
4
5
6
7
Reaction time (d)
Fig. S12. Plot of the thickness of the uniformly b-oriented SL film with respect to time du
ring the secondary growth of b-SLm/g plates having different initial thicknesses of SL see
d crystals in Gel-2 at different reaction temperatures
30
A
B
5 µm
5 µm
C
6
D
nd
c
5
t (m)
Intensity (arb.)
MFI powder 2 -growth
MFI seed powder
4
3
2
b
1
a
0
5 10 15 20 25 30 35 40 45 50 55 60
2 theta (deg.)
0
10
20 30 40 50
Reaction time (h)
60
70
E
Fig. S13. SEM images of leaflet shaped SL seed crystals (A) and the SL crystals grown fr
om the SL seed crystals by secondary growths in Gel-2 (B). (C) XRD diffraction patterns
of leaflet shaped SL seed crystals and the SL crystals grown from the seed crystals by se
condary growth in Gel-2. (D) The plots of the average length increases of the SL crystals
vs. reaction time during the secondary growths of leaflet SL crystals in Gel-2. (E) Illustra
tion of the morphology change of SL seed crystals after secondary growth
31
2 μm
2 µm
10 µm
25 µm
2600
(040)
(080)
(0100)
ka1
ka2
2400
Thickness (nm)
Intensity (arb.)
(020)
(060)
o
180 C
o
165 C
o
150 C
o
140 C
2200
2000
1800
1600
1400
1200
1000
5
10 15 20 25 30 35 40 45 50
2 theta (degree)
0
1
2 3 4 5
Reaction time (d)
6
7
Fig. S14. SEM images of a b-SLm/g plate (A) and the continuous b-SLf/g plate prepared b
y secondary growth of b-SLm/g plates in a TMAOH analog of Gel-2 (B). The reaction wa
s carried out at 165 °C for 7 d. (C) The corresponding XRD pattern of a b-SLf/g plate.
(D). Plot of film thickness vs. reaction time at different reaction temperatures. The insets i
n (A) and (B) show the corresponding side views.
32
5h
10 h
500 0C
4h
RT
Gel-2
A
RT
TPA gel
B
t-TPA gel
C
Fig. S15. Comparison of the crack formation tendencies after calcination at 500 C. (A)
A uniformly b-oriented SL film supported on glass prepared from b-SLm/g plates by seco
ndary growth in Gel-2. (B) A randomly oriented SL film supported on glass prepared fro
m b-SLm/g plates by secondary growth in a TPA gel. (C) A randomly oriented SL film su
pported on glass prepared from b-SLm/g plates by secondary growth in a t-TPA gel. They
were calcined simultaneously under the same condition. The calcination condition was a
5-h increase of the temperature to 500 C with the heating rate of 100 C/h (1.67 C/min),
calcination at 500 C for 10 h at the temperature, and cooling with the rate of 125 C/h.
33
A
D
10 µm
B
C
Laser power = 3.5%
Gain = 547
Zoom = 0.6
5 µm
E
Laser power = 6.5%
Gain = 700
Zoom = 2.0
F
Fig. S16. A typical SEM image (A), 2D (B), and 3D (C) LSCM images of a calcined, ran
domly oriented SL membrane after treatment with fluorescin and a typical SEM image
(D), 2D (E), and 3D (F) LSCM images of a calcined, uniformly b-oriented SL membrane
after treatment with fluorescin. The 3D images were built after caring out slice scans in t
he selected areas indicated in (B) and (D). Note that the laser power, zoom value, and ma
ster gain are much higher in the case of uniformly b-oriented SL membrane. The bright s
pots in (E) and (F) arise from unreacted solid gel particles residing on the membrane surf
ace owing to incomplete washing.
34
A
5
10
B
15
20
25 30 35
2 (degree)
C
2 µm
5 µm
2 µm
25 µm
25 µm
25 µm
40
45
505
10
15
20
25 30 35
2 (degree)
40
45
505
10
15
20
25 30 35
2 (degree)
40
45
50
fig. S17. (A) SEM images of a b-SLm/g plate before calcination (top) and after secondary
growth in Gel-2 (middle) and the corresponding X-ray diffraction pattern of the film (bott
om). (B) SEM images of a b-SLm/g plate after calcination (top) and after secondary growt
h in Gel-2 (middle) and the corresponding X-ray diffraction pattern of the film (bottom).
(C) SEM images of a b-SLm/g plate after calcination followed by washing with the aqueo
us 0.2 M NH4F solution (top) and after secondary growth in Gel-2 (middle) and the corre
sponding X-ray diffraction pattern of the film (bottom).
35
A
B
10 nm
D
1 m
10 nm
Si
Spot 1
10 nm
C
E
F
G
Spot 2
10 nm
H
I
J
K
Spot 3
10 nm
L
W
Fig. S18. High resolution TEM (HRTEM) images. Typical cross-sectional images of a
SL crystal in the monolayer assembled by rubbing before (a) and after (b) washing the
monolayer with a 0.2 M NH4F solution for 6 h. A piece of a uniformly b-oriented SL film
grown on a Si wafer prepared by a focused ion beam (FIB) cutter and mounted to a
copper TEM grid with W as the glue (c). A 1.2-m thick W layer was deposited onto the
SL film layer prior to cutting the SL film. The selected area electron diffraction (SAED)
patterns were taken from three spots. Lattice image of the SL film in Spot 1 (d) and its
expanded image (e) and the SAED pattern of Spot 1 (f). Lattice image of the SL film in
Spot 2 (g) and its expanded image (h) and the SAED pattern of Spot 2 (i). Lattice image
of the SL film in Spot 3 (j) and expanded image (k) and the SAED pattern of Spot 3 (l).
Uniformity of b-orientation was further supported by the analyses of the cross
sections of the films (fig. S18C) with a high resolution transmission electron microscopy.
Thus, the lattice fringes and selected area electron diffraction patterns of the cross
sections of the films are identical regardless of the spots within the film (fig. 4, D to L),
confirming once again that the SL films grow in perfect b-orientation (from the Si layer
to W layer in fig. 18C) regardless of the depth of the film. When the direction from the Si
layer to W layer is b-axis direction, the direction normal to the Cu TEM grid could be
either a-axis or c-axis direction. The observed lattice fringes and selected area electron
diffraction patterns (fig. 18, F, I, and L) coincide with the case where the direction normal
to the Cu TEM grid is the c-axis direction, indicating that the particular small portion of
the SL film cut by a focused ion beam happens to be oriented with the c-axis normal to
Cu TEM grid.
36
Fig. S19. (A) Illustration of the structure of 1-bromododecane prepared by Chem Draw®
showing its total length of 18.051 Å . (B) Illustration of 1-bromododecane molecules inclu
ded in SL channels. (C) Illustration of a side view of a perfect b-oriented 3-μm thick SL f
ilm supported on glass and the areas in the film where Br-to-Si atomic ratios were measur
ed using an EDX equipped in the SEM.
37
1.0
after 7 d
Bromine (atomic %)
0.9
0.8
5d
0.7
0.6
3d
0.5
0.4
0.3
0.5
1.0
1.5
2.0
2.5
3.0
Depth (m)
Fig. S20. Plot of Br content (atomic %) versus film depth for the three cases with
different incorporation time (as indicated).
38
180
I2/I2[qz], (%)
160
HC-15
HC-18
HC-22
HC-12
HC-09
140
120
100
80
60
40
20
0
0
10
20
30 40 50 60
0
Incident angle ( )
70
80
Fig. S21. Maker fringes of the 3-µm thick b-SLf/g/b-SLf plates incorporated with HC-n
(n = 9, 12, 15, 18, and 22) dyes. The Maker fringe appears because the uniformly b-orient
ed SL films are coated on both sides of the 1-mm thick glass plate.
39
A
B
C
D
Fig. S22. Digital camera images of a clean bare glass plate with the thickness of 1 mm
(A) and the same glass plates coated with three different types of SL films which were pr
epared by secondary growth of b-SLm/g plates in three different gels, Gel-2 (B), t-TPA ge
l (C), and TPA gel (D).
40
Transmittance (%)
100
90
80
70
60
50
40
30
20
10
0
200
Air
Bare glass
Gel-2
t-TPA gel
TPA gel
400
600
800
Wavelength (nm)
1000
Fig. S23. Transmittance spectra of air, a clean bare glass plate, and three glass plates
coated with three different types of SL films which were prepared by secondary growth
of b-SLm/g plates in three different gels, Gel-2, t-TPA gel, and TPA gel (as indicated).
41
A
B
1 m
1 m
5 µm
C
2 µm
D
1 m
5 m
1 m
5 m
Fig. S24. Top SEM views of a typical porous substrate (3 mm) composed of a 1:1
mixture of 350 nm and 600 nm silica beads prepared by pressing (150 kgf cm-2) for 30 s
and calcining at 1,020 C for 2 h (A) after surface polishing and (B) after additional
rubbing its surface with 70-nm silica beads followed by calcination at 550 °C for 8 h. (C)
SEM images of b-oriented SL monolayer assembled on porous silica supports and (D) boriented continuous SL film supported on porous silica supports prepared by the
secondary growth of the monolayer in Gel-2 at 165 C for 18 h. The insets show the
corresponding side views.
42
Sweep gas (He)
MFC (4)
MFC (3)
oven
He
on-line GC
o-xylene
Retentate
MFC (2)
mixer
3w vale
GC
By pass
p-xylene
6-port
vale
BF
MFC (1)
Dilute gas (He)
LN2 trap/ off-line GC
Rotary vacuum pump
(to remove moisture)
Moisture trap
vent
Fig. S25. The schematic illustration of the set-up used in our work to test the p-/o- xylene
separation performance of the membranes
43
References and Notes
1. J. O'Brien-Abraham, J. Y. S. Lin, in Zeolites in Industrial Separation and Catalysis, S.
Kulprathipanja ,Ed. (Wiley VCH, Verlag, Weinheim, Germany, 2010), chap. 3, pp. 307–329.
2. J. Caro, M. Noack, in Advances in Nanoporous Materials, S. Ernst, Ed. (Elsevier, Amsterdam,
2009), vol. 1, chap. 1, pp. 1–96.
3. J. Caro, M. Noack, Zeolite membranes–Recent developments and progress. Microporous
Mesoporous Mater. 115, 215 (2008). doi:10.1016/j.micromeso.2008.03.008
4. J. Choi et al., Grain boundary defect elimination in a zeolite membrane by rapid thermal
processing. Science 325, 590 (2009). doi:10.1126/science.1176095 Medline
5. J. Hedlund, F. Jareman, A. J. Bons, M. Anthonis, A masking technique for high quality MFI
membranes. J. Membr. Sci. 222, 163 (2003). doi:10.1016/S0376-7388(03)00285-0
6. Z. P. Lai, M. Tsapatsis, J. R. Nicolich, Siliceous ZSM-5 membranes by secondary growth of
b-oriented seed layers. Adv. Funct. Mater. 14, 716 (2004). doi:10.1002/adfm.200400040
7. C. J. Gump, V. A. Tuan, R. D. Noble, J. L. Falconer, Aromatic permeation through crystalline
molecular sieve membranes. Ind. Eng. Chem. Res. 40, 565 (2001). doi:10.1021/ie000553i
8. Z. P. Lai et al., Microstructural optimization of a zeolite membrane for organic vapor
separation. Science 300, 456 (2003). Medline
9. M. A. Snyder, M. Tsapatsis, Hierarchical nanomanufacturing: From shaped zeolite
nanoparticles to high-performance separation membranes. Angew. Chem. Int. Ed. 46,
7560 (2007). doi:10.1002/anie.200604910
10. H. H. Funke, A. M. Argo, J. L. Falconer, R. D. Noble, Separations of cyclic, branched, and
linear hydrocarbon mixtures through silicalite membranes. Ind. Eng. Chem. Res. 36, 137
(1997). doi:10.1021/ie960472f
11. J. O'Brien-Abraham, M. Kanezashi, Y. S. Lin, Effects of adsorption-induced microstructural
changes on separation of xylene isomers through MFI-type zeolite membranes. J. Membr.
Sci. 320, 505 (2008). doi:10.1016/j.memsci.2008.04.023
12. M. O. Daramola et al., Xylene vapor mixture separation in nanocomposite MFI-alumina
tubular membranes: Influence of operating variables. Sep. Sci. Technol. 45, 21 (2009).
doi:10.1080/01496390903402141
13. H. Guo et al., Hierarchical growth of large-scale ordered zeolite silicalite-1 membranes with
high permeability and selectivity for recycling CO2. Angew. Chem. Int. Ed. 45, 7053
(2006). doi:10.1002/anie.200602308
14. H. S. Kim et al., Aligned inclusion of hemicyanine dyes into silica zeolite films for second
harmonic generation. J. Am. Chem. Soc. 126, 673 (2004). doi:10.1021/ja037772q
Medline
15. H. S. Kim et al., Aligned inclusion of n-propionic acid tethering hemicyanine into silica
zeolite film for second harmonic generation. Adv. Mater. 19, 260 (2007).
doi:10.1002/adma.200602101
16. H. S. Kim, T. T. Pham, K. B. Yoon, J. Am. Chem. Soc. 130, 2131 (2008).
1
17. C. M. Lew, R. Cai, Y. Yan, Zeolite thin films: From computer chips to space stations. Acc.
Chem. Res. 43, 210 (2010). doi:10.1021/ar900146w Medline
18. Z. Li et al., Mechanical and dielectric properties of pure-silica-zeolite low-k materials.
Angew. Chem. Int. Ed. 45, 6329 (2006). doi:10.1002/anie.200602036
19. M. E. Davis, Ordered porous materials for emerging applications. Nature 417, 813 (2002).
doi:10.1038/nature00785 Medline
20. T. Bein, Synthesis and applications of molecular sieve layers and membranes. Chem. Mater.
8, 1636 (1996). doi:10.1021/cm960148a
21. Y. Liu, Y. S. Li, W. S. Yang, Fabrication of highly b-oriented MFI film with molecular
sieving properties by controlled in-plane secondary growth. J. Am. Chem. Soc. 132, 1768
(2010). doi:10.1021/ja909888v Medline
22. X. Li, Y. Peng, Z. Wang, Y. Yan, Synthesis of highly b-oriented zeolite MFI films by
suppressing twin crystal growth during the secondary growth. CrystEngComm 13, 3657
(2011). doi:10.1039/c1ce05094j
23. MFI and BEA are two different framework-type codes assigned by the Structural
Commission of the International Zeolite Association. Silicalite-1 and ZSM-5 belong to
MFI. ZSM-5 has aluminosilicate frameworks with varying Si/Al ratios, whereas
silicalite-1 has a pure silica framework. BEA is also often called beta zeolite. BEA has
aluminosilicate frameworks with varying Si/Al ratios, whereas pure silica BEA (Si-BEA)
has a pure silica framework.
24. C. Baerlocher, W. M. Meier, D. H. Olson, Atlas of Zeolite Framework Types (Elsevier,
Amsterdam, ed. 5, 2001), pp. 184–185.
25. O. Larlus, V. Valtchev, Synthesis of all-silica BEA-type material in basic media.
Microporous Mesoporous Mater. 93, 55 (2006). doi:10.1016/j.micromeso.2006.02.003
26. O. Larlus, V. Valtchev, Control of the morphology of all-silica BEA-type zeolite synthesized
in basic media. Chem. Mater. 17, 881 (2005). doi:10.1021/cm048799r
27. A. Mitra et al., Synthesis and evaluation of pure-silica-zeolite BEA as low dielectric constant
material for microprocessors. Ind. Eng. Chem. Res. 43, 2946 (2004).
doi:10.1021/ie034062k
28. A. Mitra et al., Synthesis and corrosion resistance of high-silica zeolite MTW, BEA, and
MFI coatings on steel and aluminum. J. Electrochem. Soc. 149, B472 (2002).
doi:10.1149/1.1507784
29. S. Mintova, M. Reinelt, T. H. Metzger, J. Senker, T. Bein, Pure silica BETA colloidal zeolite
assembled in thin films. Chem. Commun. 2003, 326 (2003). doi:10.1039/b210767h
Medline
30. K. B. Yoon, Organization of zeolite microcrystals for production of functional materials.
Acc. Chem. Res. 40, 29 (2007). doi:10.1021/ar000119c Medline
31. R. Singh, P. K. Dutta, in Handbook of Zeolite Science and Technology, S. Auerbach, K.
Carrado, P. Dutta, Eds. (Marcel Dekker, New York, 2003), chap. 2.
2
32. A. Marinkovic et al., Negative thermal expansion in hydrated HZSM-5 orthorhombic zeolite.
Microporous Mesoporous Mater. 71, 117 (2004). doi:10.1016/j.micromeso.2004.03.023
33. D. Bhange, V. Ramaswamy, Negative thermal expansion in silicalite-1 and zirconium
silicalite-1 having MFI structure. Mater. Res. Bull. 41, 1392 (2006).
doi:10.1016/j.materresbull.2005.12.002
34. J. S. Lee, J. H. Kim, Y. J. Lee, N. C. Jeong, K. B. Yoon, Manual assembly of microcrystal
monolayers on substrates. Angew. Chem. Int. Ed. 46, 3087 (2007).
doi:10.1002/anie.200604367
35. The transfer of nutrients from the dry gel to the seed layer seems to be carried out by a small
amount of moisture, which vaporizes into steam at the reaction temperature.
36. As the reaction temperature increases, the growth rate increases. However, as a possible
means to incorporate more second-order NLO dyes into the channels, we chose the
relatively low reaction temperatures (140° to 180°C) with the hypothesis that the number
of defect sites in the channels would decrease as the growth rate decreased.
37. The thickness of the seed crystal does not affect the growth rate. However, if one wants to
reach the desired thickness faster, it will be beneficial to start with thicker seed crystals.
38. H. Karsli, A. Çulfaz, H. Yücel, Sorption properties of silicalite-1 of pure silica form: The
influence of sorption history on sorption kinetics of critically sized molecules. Zeolites
12, 728 (1992). doi:10.1016/0144-2449(92)90124-8
39. D. P. Serrano, R. Van Grieken, P. Sánchez, R. Sanz, L. Rodrı́guez, Crystallization
mechanism of all-silica zeolite beta in fluoride medium. Microporous Mesoporous Mater.
46, 35 (2001). doi:10.1016/S1387-1811(01)00272-4
40. W. Stöber et al., Controlled growth of monodisperse silica spheres in the micron size range.
J. Colloid Interface Sci. 26, 62 (1968). doi:10.1016/0021-9797(68)90272-5
41. G. Xomeritakis, Z. P. Lai, M. Tsapatsis, Separation of xylene isomer vapors with oriented
MFI membranes made by seeded growth. Ind. Eng. Chem. Res. 40, 544 (2001).
doi:10.1021/ie000613k
42. M. Brinkmann et al., in Springer Handbook of Lasers and Optics, F. Träger, Ed. (Springer,
New York, 2007), chap. 5, pp. 309–331.
3