One-pot synthesis of MWW zeolite nanosheets using a rationally designed organic structure-directing agent

A new material MIT-1 comprised of delaminated MWW zeolite nanosheets is synthesized in one-pot using a rationally designed organic structure-directing agent.


Synthesis of MIT-1
MIT-1 was synthesized by combining the aqueous solution of Ada-4-16 in hydroxide form with sodium hydroxide (99.99% trace metals basis, Sigma Aldrich), and water in a Teflon jar. The mixture was stirred for 15 min, and then aluminum hydroxide (80.3% Al(OH) 3 , SPI Pharma 0250 aluminum hydroxide powder) was added. The mixture was stirred for 15 min, and colloidal silica (LUDOX® LS 30) was added. The mixture was allowed to age under stirring for 4 h at room temperature. For some samples, water evaporation was necessary and was facilitated by a stream of air and measured gravimetrically. The final gel composition was 1 SiO 2 /0.1 OSDA/0.05 Al(OH) 3 /0.2 NaOH/45 H 2 O. The gel was transferred to a Teflon-lined Parr reactor, sealed, and held at 433 K at 60 rpm rotation. Aliquots of the synthesis mixture were taken periodically and monitored by powder X-ray diffraction (PXRD). After crystallization, the solids were recovered by filtration, extensively washed with deionized (DI) H 2 O, and dried at 373 K overnight. The solid was calcined by heating under flowing N 2 at a rate of 2 K/min with a 1 h hold at 423 K, a 2 h hold at 573 K, and a 3 h hold at 813 K. The flowing gas was then switched to dry air and the temperature was held at 813 K for another 6 h.

Synthesis of MCM-22 and MCM-56
MCM-22 is synthesized following the methods of Corma et al. 1 2 Using the same precursors, with the exception of using the colloidal silica solution Ludox LS30 instead of fumed silica, the final gel composition was 1 SiO 2 /0.35 HMI/0.083 Al/0.21 Na/19.9 H 2 O. The gel was transferred to a Teflon-lined Parr reactor, sealed, and held at 418 K at 60 rpm rotation with a Teflon stir-bar added to improve mixing for 33 h. After crystallization, the solids were recovered by filtration, extensively washed with deionized (DI) H 2 O, and dried at 373 K overnight. The solids were calcined by heating under flowing dry air at a rate of 2 K/min with a 1 h hold at 423 K, a 2 h hold at 573 K, and a 6 h hold at 823 K.

Characterization
PXRD patterns were collected using a Bruker D8 diffractometer using Cu Kα radiation (40 kV, 40mA). Nitrogen isotherms were measured on a Quantachrome Autosorb iQ at liquid nitrogen temperature (77.35 K). The external surface areas were calculated by the t-plot method, and micropore and total pore volumes were determined at P/P 0 =0.01 and 0.95, respectively. Pore volumes were calculated using the Non-Local Density Function Theory (NLDFT) method for N 2 on silica on the adsorption branch at 77 K with a cylindrical pore model (ASiQwin, Quantachrome Instruments). Scanning electron microscopy (SEM) images were acquired on a JEOL 6700F at an accelerating voltage of 1 kV. Transmission electron microscopy (TEM) images and selected-area diffraction patterns were acquired on a JEOL 2010F at an operating voltage of 200 kV. Elemental analysis was performed with a CCD-based inductively coupled plasma (ICP) atomic emission spectrometer (Activa-S, HORIBA Scientific). Samples were dissolved in 48% HF and diluted into 3% HNO3 before analysis. A 5-point calibration curve was built using the following ICP standards: 1000 ppm Al in 3% HNO 3 , 1000 ppm P in H 2 O (all TraceCERT®) on the 308.215 nm Al line and the 177.440 nm P line. Simulated PXRD patterns were generated using powder pattern theorem implemented with UDSKIP. 3 Atomic coordinates for zeolite MWW structure were obtained from Camblor et al. 4 The simulated crystal had dimensions of 15 unit cells in the a-and b-directions, and 1 unit cell in the c-direction.
Magic-angle spinning (MAS) NMR was acquired using home-built 700 and 360 MHz spectrometers (courtesy of Dr. D. Ruben, FBML-MIT) equipped with 3.2 and 4 mm Varian-Chemagnetics triple resonance H/X/Y MAS probes. 27 Al MAS NMR (700 MHz, 1 H) spectra were obtained from fully hydrated zeolite samples using a short quantitative 5 single (Bloch) pulse (9º flip-angle) at a spinning frequency of 14 KHz, recycle delay of 2 s and between 4-8 k scans. An aqueous 1.0 M solution of Al(NO 3 ) 3 was used as an external reference (0 ppm). Trimethylphosphine oxide (TMPO, 99% Alfa Aesar) and tributylphosphine oxide (TBPO, 97% Alfa Aesar) were added to samples through vapor phase deposition. First samples were degassed overnight under dynamic vacuum at 423 K in a glass tube with a stopcock. Solid TMPO or TBPO was added to the sample under inert N 2 atmosphere in a glovebox. The glass tube was then evacuated at room temperature, isolated from vacuum, and allowed to equilibrate at 373 K for 3 h. The samples were cooled and packed under inert atmosphere into MAS rotors with gastight caps. 31 P MAS NMR (360 MHz, 1 H) spectra were collected using a Bloch experiment with a pulse width of 3.5 µs (γB 1 /2π = 71 kHz), an optimized recycle delay of 5 s, and a spinning frequency of 10 kHz. 31 P chemical shifts were referenced to an aqueous solution of 85% H 3 PO 4 (0 ppm). A Lorentzian deconvolution method was used to analyze the 31 P MAS NMR spectra. Quantitative analysis of acid site concentrations were calculated using a method described by Zhao et al. 6 Total acid site concentration was calculated by: (mol P from ICP)*(TMPO chemisorbed / TMPO total from NMR) / g catalyst. Total external acid site concentration was calculated by: (mol P from ICP)*(TBPO chemisorbed / TBPO total from N MR) / g catalyst. Specific peak areas are available in Table S5. 13 C MAS NMR (700 MHz, 1 H) were acquired using a crosspolarization 7 experiment with a 1.5 ms contact time and a ramp on 13 C ( 1 H γB 1 /2π = 50 kHz and 13 C γB 1 /2π = 65 kHz). All data were acquired using a spinning frequency of 12.5 kHz, variable recycle delays (i.e., 1.3xT 1 , optimized from the 1 H MAS NMR Bloch spectra, 256 scans, 5μs pulse ( 1 H γB 1 /2π = 50 kHz) and between 4,096 and 16,384 co-added transients depending on the signal-to-noise. Two-dimensional 13 C-{ 1 H} HETCOR spectra were acquired under identical conditions above, 1,024 co-added transients and 100 increments (40 μs dwell) in the indirect dimension. All 13 C data were acquired with high power twopulse phase modulation (TPPM) 8 proton decoupling ( 1 H γB 1 /2π = 83 kHz). 13 C spectra were referenced to adamantane (40.49 ppm) with respect to DDS (0 ppm).

Friedel-Crafts alkylation reactions
Prior to reaction, all materials were ion exchanged with a 0.2 M ammonium acetate solution (NH 4 C 2 H 3 O 2 /Al=10) for 16 h and filtered with DI H 2 O. The exchange is done three separate times. The solids are then dried and calcined using the methods described in section 1.2 and 1.3. The alkylation of benzene with benzyl alcohol reactions were performed in septum-sealed, thick-walled glass reactors with magnetic stirring. Typically, 5 g of 6.5 wt% benzyl alcohol (BA) in benzene and ~30 mg catalyst (specific amount of catalyst adjusted such that mol BA/mol Al = 200) were sealed into the vial and placed in a silicon oil bath at 358 K. Samples were taken at 1.5 h and 3 h by cooling the vial to room temperature, taking an aliquot, and filtering the solid. A known amount of 1,3,5-tri-tert-butylbenzene was added as an external standard. The solution was analyzed on a GC-FID (Agilent 6890N) using an HP-1 capillary column (J&W Scientific/Agilent Technologies; 30m long, 0.25mm i.d., 0.25 µm thick). The conversion was calculated based on the mole of BA reacted compared to the mole of BA in the feed; the yield is calculated as the mole of diphenylmethane (DP) or dibenzyl ether (DE) normalized by the mole of BA in the feed. Al-MCM-41 and Al-MFI were purchased from ACS Materials with product codes: Al-MCM-41, and ZSM-5 (P-38).

Molecular Modeling
The location and van der Waals (vdW) interaction energy of the Ada-4-16 OSDA with the MWW zeolite nanosheet were studied by molecular mechanics simulations using the Materials Studio 6.1 software 9 . The CVFF forcefield 10 was selected for the calculation as implemented by the Discover Module. A cutoff of 12.5 Å was employed for vdW term of the atom-atom interactions. Ewald summation was employed to compute the electrostatic energies of the system. The simulation cell consisted of 9 unit cells of MWW (a x b x c = 3 x 3 x 1). A vacuum of 30 Å was added perpendicular to the surface in the c-direction. The oxygen atoms at the zeolite surface were protonated. The positive charge of the system due to the quaternary ammoniums in the OSDA was compensated with a uniform charge background. The most stable locations for the OSDA molecule were obtained by simulated annealing. The Ada-4-16 was placed on the surface of the zeolite in a variety of configurations with the C16 tail oriented away from the surface. Molecular dynamics (MD) simulations were performed at 343 K for 10 ps using the NVE ensemble. The geometry of the equilibrated system was optimized to find the lowest energy conformations. Catalyst Si/Al a Micropore volume b (cm 3 g -1 )

Tables
Total pore volume c (cm 3 g -1 ) S ext d (m 2 g -1 ) Total acid sites e (x 10 -5 mol g -1 ) External acid sites f (x 10 -5 mol g -1 ) MCM a. Si/Al determined by ICP-AES elemental analysis, b. micropore volume calculated from N 2 adsorption isotherm at P/P 0 =0.01, c. total pore volume calculated from N 2 adsorption isotherm at P/P 0 =0.95, d. external surface area S ext calculated using the t-plot method, e. total acid sites calculated from TMPO titration and 31 P MAS NMR, f. external acid sites calculated from TBPO titration and 31 P MAS NMR.   Figure S12.