Capturing the Details of N2 Adsorption in Zeolite X Using

Benad, Jürries, Vetter, Klemmed, Hübner, Leyens, and Eychmüller. 2018 30 (1), pp 145–152. Abstract: In this study we report on mechanical propert...
1 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/cm

Cite This: Chem. Mater. 2018, 30, 296−302

Capturing the Details of N2 Adsorption in Zeolite X Using Stroboscopic Isotope Contrasted Neutron Total Scattering Daniel Olds,†,∥ Keith V. Lawler,‡,§,∥ Arnold A. Paecklar,⊥,∥ Jue Liu,† Katharine Page,*,† Peter F. Peterson,† Paul M. Forster,‡,§ and James R. Neilson⊥ †

Neutron Scattering Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States Department of Chemistry and Biochemistry, University of Nevada, Las Vegas, Las Vegas, Nevada 89154, United States § High Pressure Science and Engineering Center, University of Nevada, Las Vegas, Las Vegas, Nevada 89154, United States ⊥ Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523, United States ‡

S Supporting Information *

ABSTRACT: Porous materials have widespread industrial applications in adsorption and catalysis, but experimental studies providing atomistic data regarding gas−sorbent interactions under applicationrelevant conditions are limited. Current analytical methods give information about either the crystal structure or the macroscopic kinetics involved in these processes, but not their interplay. The development of a new combination of stroboscopic, isotopecontrasted neutron total scattering and steady-state isotopic transient kinetic analysis provides insight into both areas. An advanced data reduction procedure was developed to isolate the differential isotope signal from the stroboscopic neutron diffraction. These data, combined with measurement of adsorption isotherms and theoretical considerations from Grand Canonical Monte Carlo simulation, enabled the location of a heterogeneous distribution of nitrogen adsorption sites in calcium exchanged zeolite X under operational conditions (1 atm, 300 K). The nitrogen is found to adsorb primarily at Ca site II, drawing the associated Ca atoms further out into the cage structure. This approach has great potential for the investigation of gas sorption, separation, and catalysis processes in process-relevant conditions.

■

INTRODUCTION Porous materials, such as zeolites and metal organic frameworks (MOFs), have enabled significant industrial processes and continue to be developed for new transformative technologies.1 For example, zeolites are widely used industrially for petroleum refining2−4 and large-scale gas separations.5−7 Despite their widespread adoption, the precise nature of adsorbent/sorbate interactions in the framework structure and, thus, clear design principles for specific gas−solid interactions are not widely available in the literature. A lacking atomistic understanding regarding the binding of substrate gas molecules in porous materials under operational conditions, owing primarily to the lability and heterogeneous dispersion of the gas molecules, is a particular challenge. Conventional approaches, often employing neutron scattering or NMR to gain sensitivity to light elements, locate adsorbed molecules in porous materials either under high pressure (CD4 (9 bar) and C2H6 (5 bar)8) or at low temperatures (Xe (210 K),9 CD4 (77 K),10 CHCl3 (20 K),11 N2 (10 K),12 CO2 (4 K),12 p-xylene (1.5 K),13 and many others in MOFs14) in order to freeze the guest molecules at high occupancy to their most stable binding sites. However, it is important to understand the heterogeneity and lability of gas molecules in these materials © 2017 American Chemical Society

under operational conditions to facilitate real-world applications and development. There are many techniques that can study the functional behavior of porous materials under operational conditions. For example, steady-state isotopic transient kinetic analysis (SSITKA)15,16 provides insight into the steady-state kinetics of gas−solid interactions, which can be readily combined with DRIFT (diffuse reflectance infrared transform) spectroscopy17 to yield information about the chemical change of reactants interacting at surface interfaces. While these approaches are deeply insightful, they are not amenable to describing the nature of the range of sites occupied by guests or the framework’s structural response to guest uptake. Several X-ray scattering and spectroscopic methods have been developed to study porous framework materials. Single crystal X-ray diffraction was used to study nitrogen in metal− organic zeolites.18 More advanced techniques such as extended X-ray absorption fine structure (EXAFS)19,20 and quickscanning-EXAFS ((Q)EXAFS)21 can provide element-specific Received: November 2, 2017 Revised: December 7, 2017 Published: December 8, 2017 296

DOI: 10.1021/acs.chemmater.7b04594 Chem. Mater. 2018, 30, 296−302

Article

Chemistry of Materials

Figure 1. (a) Zeolite cage structure, viewed down the (130) axis, with the subunit ensemble overlaid in blue. (b) Subunit ensemble as determined from crystallographic refinement of pristine (top) and dosed with nitrogen (bottom) conditions, viewed normal to the (111) axis, where nitrogen positions are in green and calcium positions are in pink (site I), purple (site I′), and cyan (site II). (c) An alternate representation of the results in (b), where positions are shown within individual cage subunits. (d) View down the (130) axis, where the 6-member ring of the site II position, with associated crystallographic fit nitrogen position, has been colored. The (Si/Al)O4 tetrahedral units forming the ring are shaded blue, with oxygen atoms depicted in red.

ally, a computation approach, commonly used to understand gas adsorption studies based on diffraction,29 was used. Grand Canonical Monte Carlo (GCMC) simulations and isothermal gas adsorption experiments performed on the same N2/zeolite system are found to be consistent with the results of isotopecontrasted total scattering studies. The simulations support our crystallographic determination of Ca positions and reveal a distribution of N2 molecules clustered near one of these sites. Calculation of the isotope-contrasted neutron total scattering from the theory-derived simulation reproduces the experimental observation after complex data reduction procedures. These methods provide a unique opportunity to gain insight into chemically dynamic and heterogeneous systems in operando.

local structure information about atomic sites that may change upon gas absorption or desorption. Total scattering methods, including pair distribution function analysis, have found utility in this area owing to the fact that they can yield useful structural information even when the guest molecules do not pack into the host pores with regular periodicity. Total scattering has successfully identified the presence of guest molecules and host−guest interactions such as hydrogen and nitrogen in the Prussian blue system Mn3[Co(CN)6]2,22,23 ammonia borane in mesoporous silica,24 and the formation of AgI in a zeolite.25 To gain sensitivity to light elements while also providing chemical selectivity, isotope substitution (NatNi/62Ni) experiments with neutron total scattering have yielded insight into the interaction between acetylene and NiNa−zeolite Y.26 However, these approaches do not readily permit study of the material under steady-state, operational conditions. Here, we report the main results of steady-state isotopecontrasted neutron total scattering experiments that study the nature of N2 absorption in Ca-substituted zeolite X. Nitrogen adsorption in zeolite X was studied previously using inelastic neutron scattering.27 We previously described the gas-handling instrumentation commissioned on the “nanoscale ordered materials diffractometer” (NOMAD) at the Spallation Neutron Source at Oak Ridge National Laboratory.28 This instrumentation permits the delivery of multiple gas streams to a solid sample immobilized in the primary neutron beam; a highspeed switching valve rapidly alternates between two different gas streams (e.g., 14N2 and 15N2) with negligible change in flow rate and pressure. This precision gas handling environment permits cross-correlation of the isotope composition with the event-based time-of-flight neutron scattering data. Addition-

■

EXPERIMENTAL AND THEORETICAL METHODS

Sample Preparation. The material preparation and characterization has been described previously30 and is summarized here. The zeolite used for gas adsorption was produced through ion exchange of sodium zeolite X (NaX) powder with aqueous calcium chloride. X-ray powder diffraction showed retained crystallinity, and scanning electron microscopy showed retained morphology after the exchange step. Energy dispersive X-ray spectroscopy (EDS) and optical emission spectroscopy with inductively coupled plasma (ICP-OES) were used to determine the cation exchange level (EDS (98%) and ICP-OES (97%)) suggesting a final sample composition of Na86−2xCaxAl86Si106O384 with x ≈ 42. In preparation for the neutron scattering experiments, the powders were pressed into a pellet at 4 t, crushed into large chunks, sieved to sizes between 100 and 500 μm, and dried in vacuum at 673 K. Neutron Total Scattering Measurements. Initial neutron total scattering measurements were performed in static gas loading configurations, consisting of a quartz glass tube sealed with a plug valve, at room temperature on NOMAD. Samples were first measured 297

DOI: 10.1021/acs.chemmater.7b04594 Chem. Mater. 2018, 30, 296−302

Article

Chemistry of Materials

Figure 2. (a) Resultant Fourier difference map from fits to the statically dosed zeolite sample with a model lacking N2, (b) compared to the nitrogen positions found through GCMC simulations. To avoid any confusion by the projection of the structure, we have shown isolated views of each cage set below the figures with corresponding residual (from refinements) or simulated nitrogen density (from simulation). under vacuum for 2 h at 300 K followed by a pressure dosed condition with 15N2. The samples were then sealed at 1 psig and measured for 2 h at 300 K. Data reduction and PDFs were generated using standard data reduction protocols on NOMAD but found to have insufficient statistics for quantitative real-space analysis. Structural refinements from the diffraction data were obtained with the program TOPAS, and Fourier difference maps were created with the program JANA.31 Details of the refinements and tabulated results are provided in the Supporting Information of this manuscript. Stroboscopic neutron total scattering data were collected using the high-precision gas flow cell sample environment designed for SSITKA measurements.30 In situ SSITKA measurements with simultaneous neutron total scattering were performed by flowing alternately 15N2 and 14N2, where the 14N2 was mixed with 3% Ar to function as an inert trace, with helium as an inert carrier gas through the sample. The switching time between the two gases was set to 6 min with an overall number of 12 switching cycles between the two isotopes, resulting in a total measurement time of 150 min. An advanced data reduction procedure was developed to stroboscopically reduce the data using the Mantid data analysis framework.32 An unexpected time-dependent uptake of water by the sample during the stroboscopic measurement was observed, which was attributed to the presence of some humidity in the lines and gases. The details of the data-informed reduction procedure can be found in the Supporting Information of this manuscript. The resultant stroboscopically reduced scattering data is presented and described in the Results and Discussion. Gas Adsorption Measurements. The Ca-exchanged zeolite X sample was activated under dynamic vacuum at 673 K for 24 h followed by a N2 surface area measurement at 77 K. Adsorption isotherms were measured using a Micromeritics ASAP 2020 adsorption analyzer. Desorption measurements were taken at the end of each isotherm to monitor that no hysteresis occurred. Prior to each measurement, the sample was reactivated at 473 K under dynamic vacuum for an hour followed by an hour of equilibration in the cryostat at the target temperature. N2 isotherms were measured at 77 K and then every 10 K from 140 to 300 K. The BET surface area was evaluated over the nanoporous regime, 0.01 ≤ P/P0 ≤ 0.10, of the 77 K N2 adsorption isotherm following the recommendation of Walton and Snurr.33 Theoretical Evaluation of Adsorption Behavior. Gas adsorption was simulated with Monte Carlo in the grand canonical ensemble (GCMC) with a modified version of the MuSiC package (full details in the Supporting Information).34 A rigid 2 × 2 × 2 super cell of the experimentally determined crystal structure was used to

represent the zeolite framework with the ideal Ca-86X stoichiometry: Ca43Si106Al86O384. Ca and N2 were allowed to fluctuate. The volumes in the zeolite framework inaccessible to N2 were blocked using our energy based pore mapping program.35 The guest (cation and adsorbate)−framework Coulomb interactions were computed using electrostatic and dispersion−repulsion interactions. The T-atom framework and cation models of Boutin et al. were used for the zeolite.36,37 N2 was represented by the TraPPe-small force field.34

■

RESULTS AND DISCUSSION Ca Positions and Occupancy. Ca is known to occupy three distinct sites in the zeolite X structure,38 where each position is associated with a different component cage subunit. Here, we refer to these positions as site I (in the hexagonal prism cage), site I′ (in the beta or sodalite cage), and site II (in the alpha or super cage). These cages, as well as their relative connectivity, are shown in Figure 1a−c. Neutron total scattering is a well-suited method to determine the positions and occupancies of Ca at these three sites. The refined structure for the static pristine compared to 15N2-dosed cases can be seen in Figure 1b,c, with details of the refinement found in Supporting Information. The Ca at site II sits 0.213 Å above the plane of the 6-member ring (toward the subunit cavity). This Ca is seen to be drawn further into the subunit cage upon gas loading to a position 0.257 Å above the plane of the 6member ring. Note that occupancy and atomic displacement parameters can be correlated, so the atomic displacement parameter of Ca was fixed to 0.58 Å2 for these refinements. Initially refined populations of Ca in the pristine material were found to be 11.63(94) atoms at site I (out of a possible site multiplicity of 16), 8.00(74) atoms at site I′ (out of a site multiplicity of 32), and 23.36(58) (out of a site multiplicity of 32) atoms at site II which are in agreement with previous investigations.39,40 We find that, upon exposure to nitrogen, there is a migration of the Ca to the site II (now 23.42(0.90)) and site I′ (now 8.54(70)), presumably from the site I (now 11.04(58)). Migration behavior of extra framework cations was also observed in zeolite Y (faujasite structure) upon CO2 adsorption41 as well as benzene adsorption.42 The migration of the cations can be attributed to the strong interaction of the 298

DOI: 10.1021/acs.chemmater.7b04594 Chem. Mater. 2018, 30, 296−302

Article

Chemistry of Materials

Figure 3. (a) Time-dependent top-down view of the measured scattering, S(Q), binned into 10-s slices which demonstrates little change in the measured Bragg peaks but a steady increase in the diffuse scattering attributed to water. The average S(Q) has been overlaid in black for comparison. (b) Time-dependent weighting of the model contributions to the total pattern. (c) Isolated components which are weighted by the time-dependent model to recover the data. (d) Stroboscopically binned 14N2 data, compared to the pattern from the GCMC simulated structure. (e) Recovered residual gas signal in the data compared to the simulated GCMC nitrogen signature.

The clear preference for nitrogen to adsorb at site II can be seen as an irregular pocket of residual just outside the refined Ca position. The density map, attained from GCMC simulations for nitrogen at 300 K and 1 atm (Figure 2b), also suggests that the Ca ions at site II are the primary binding sites for nitrogen under ambient conditions. There is no N2 loading at site I and site I′ as those cations are within the sodalite cage and the hexagonal prism which are well-known to be inaccessible to adsorbed guest molecules in faujasite zeolites and were blocked by the pore mapping program in the GCMC simulations.4,15−21,45,46 The clear preference in the Fourier difference results supports the assumption that nitrogen molecules cannot easily access the hexagonal prism site I′ nor the sodalite site I. Subsequent Rietveld refinements with a nitrogen atom, representing the center of mass for the molecule in these locations, provided the occupancy for the N2 molecules. Only about 8 N2 molecules could be found based on diffraction data. However, under ambient conditions, GCMC predicts a loading of 17.9 N2 molecules/unit cell which agrees well with the adsorption isotherm measurements indicating 15 N2 molecules per unit cell. A possible explanation for the discrepancy between the number of nitrogen molecules found from neutron scattering data analysis compared to the GCMC calculations and adsorption isotherm measurements could be the diffuse nature of the nitrogen density, as the crystallographic chemical occupancies reflect only those N2 molecules that appear with some long-range order (on average). However, we cannot rule out the occlusion of some binding sites from the presence of water.

cation’s positive charge and the high electric quadrupole moment of N2 and CO2 (4.7 × 10−40 C/m2 and 13.4 × 10−40 C/m2)43 as well as the interaction with the π-electron system of the benzene molecule. This migration likely occurs via Ca passing through the shared faces of the cages. Grand Canonical Monte Carlo simulations that permitted Ca ion site hopping found 16 Ca ions/cell at site I and 27 Ca ions/cell at site II. A full occupation of site I should be viewed as the most idealized cation arrangement of a low silica evacuated zeolite X.39,40 The presence of water or residual monocations will tend to shift Ca from the most energetically favorable site I into site I′, and this will likely be coupled with a migration of ions from site II to the more favorable site I′ since the strong electrostatic repulsion at site I′ from site I is no longer present.40 Thus, deviations between the Ca occupancies in the idealized simulation and the experimental findings may be explained by water uptake of the sample. In fact, it has been hypothesized that all water cannot be removed or prevented from coadsorbing during experimental uptake measurements, and the presence of water was shown to in general reduce the adsorptive capacity for N2 and Ar into Ca containing zeolites (although at very low pressures they did observe an increase in Ar adsorption with water present).44 N Positions and Occupancy. The location and occupancy of adsorbed nitrogen were also experimentally determined using neutron diffraction. The structure was refined without the presence of nitrogen and then compared to the data to generate a Fourier difference map, shown in Figure 2a. This Fourier difference map indicated likely N2 positions and occupancy. 299

DOI: 10.1021/acs.chemmater.7b04594 Chem. Mater. 2018, 30, 296−302

Article

Chemistry of Materials

nitrogen through fit occupancy values, which vary by a ratio of 1.45 between the sample exposed to 14-N2 compared to 15-N2, which compares remarkably well with the ratio of scattering power between the two isotopes (b14‑N/b15‑N = 1.46). Details of this refinement are found in the Supporting Information. The static nitrogen-loaded data, as well as the differential signal from the data reduction process, are illustrated in Figure 3d. The isotope difference signal, represented as the total scattering function, is very weak and is therefore not analyzed as its Fourier transform, the pair distribution function. Instead, the isotope contrasted total scattering is consistent with the equivalent scattering function calculated from the GCMC simulation (Figure 3e). GCMC also gives information about the orientation of the diatomic nitrogen molecule in the structure. The simulations demonstrated that nitrogen prefers a colinear binding with the calcium ions along the N2 molecular axis. To summarize, it becomes interesting to compare the stroboscopic and isotope-contrasted results to more traditional gas-adsorption experiments performed under cryogenic conditions. Single-crystal X-ray diffraction is not able to achieve sufficient residual electron density to study the adsorption of nitrogen at room temperature in a metal organic framework.18 In another study, single crystal neutron diffraction also lacked the ability to detect adsorbed hydrogen at RT, likely due to the molecular dynamics.49 Here, we show a distribution of nitrogen molecule positions at room temperature, which provides a structural basis for such types of molecular motion. Adsorption studies in NaY could locate CO241 at 4 K as well as RT with a Uiso for CO2 of 0.145 Å2 and 0.504 Å2 respectively. Due to the large displacement coefficients of NaY and CO2 at RT, not all sodium atoms could be located compared to the measurement at 4 K, likely due to atomistic dynamics. Similarly, the high temperature factors describing the adsorption of benzene in NaY at RT is, according to the authors, not very meaningful either.44 Adsorption studies of nitrogen in the zeolite SSZ-13 located N2 at 10 K with Uiso values of 0.033 Å2 and 0.04 Å2 for N1 and N2, respectively.12 In the study of nitrogen adsorption here, at room temperature, a Uiso value of 0.067(14) Å2 is found, in good agreement with increased thermal motion expected at higher temperatures. However, it should be noted that the occupancy and atomic displacement parameters for N are found to be correlated in refinement, and the value of Uiso (fixed in final analyses) was initially determined while holding the occupancy fixed to the value derived from SSITKA analysis (further details are provided in Supporting Information). Total scattering can also provide crucial information to address questions that cannot be answered with traditional crystallographic refinements alone. For example, the binding angles in CO2 adsorbed in NaY were determined to be 149.3° with Rietveld refinement, which leaves the question if the result is physically relevant or is the result of occupational or dynamical disorder. Total scattering with pair distribution function analysis could help determine whether the distortion of the CO2 molecule is based on disorder.18 This isotope-contrasted stroboscopic neutron scattering method shows promise for the study of these and other gas/adsorbent interactions.

Nitrogen Adsorption in Operando. The absorption of N2 molecules modifies the structural framework of the zeolite. As previously reported,30 neutron diffraction data indicate that the zeolite structure responds to the presence of nitrogen by shrinking the unit cell volume upon gas loading. Since these contractions are reversible upon desorption, they are often referred to as “lattice breathing”.14 Shrinking of lattice parameters has been observed for zeolite Y upon CO2 adsorption41 whereas nitrogen adsorption led to an expansion of the unit cell in the Co(BDP) MOF47 as well as in the Prussian blue system Mn3[Co(CN)6]2.22 GCMC simulations confirm an experimentally observed shift of the Ca ions at site II further into the subunit cage upon gas loading based on the interaction with nitrogen. To isolate the signal of adsorbing nitrogen separate from any lattice breathing behavior, it is necessary to measure the system under steady-state conditions. Under steady state conditions, the lattice parameters will not change upon adsorption and desorption of different nitrogen isotopes. In conventional techniques, where a gas filled system is compared to an evacuated condition, the lattice parameters tend to change, convolving analysis of gas adsorption with changing lattice behaviors. We have avoided this complication by performing steady-state isotope varied stroboscopic neutron diffraction. The data was collected under constant flow and pressure but with symmetrically varying isotope composition of 14N2 and 15 N2, measured over 12 cycles of switching feed gas. While the aforementioned neutron density maps describe only the crystallographically ordered nitrogen site probability densities, the neutron total scattering function, S(Q) (the inverse Fourier transform of the pair distribution function), describes the scattering from all nitrogen molecules (e.g., those that are ordered and disordered). However, the diffuse scattering signature requires additional counting time to build up the needed signal-to-noise ratio. Therefore, the attained neutron total scattering data from the different isotope adsorption and desorption steps (in the steady state) was rebinned stroboscopically over all cycles to have better signal statistics. To obtain the neutron isotope-difference total scattering function, a data-informed stroboscopic reduction procedure was developed. The reduction was complicated by trace amounts of water, which was found to have infiltrated the sample during data collection, as can be seen in Figure 3a. Neutron scattering is inherently very sensitive to the presence of hydrogen due to its large incoherent cross section. Previous work has shown sensitivity of neutron diffraction to changes in hydration levels as little as 0.02%48 via changes in the incoherent scattering signal. This is well below the levels that would contribute structural features in the analysis of the refined Bragg reflections in the diffraction data. Models were developed for the dry zeolite structure and the slowly varying incoherent contributions of water (Figure 3b,c), which were then removed from the data. The integrated intensity of the remaining signal, attributed to the adsorbed nitrogen, was found to oscillate with a frequency matching the prescribed gas-dosing rates. Notably, the total intensity of this nitrogen signal is ∼0.5% of the total integrated intensity. Further details of this reduction procedure can be found in the Supporting Information. The stroboscopically reduced neutron total scattering data is consistent with a heterogeneous distribution of N2 sites localized near site II predicted by GCMC. Rietveld refinements of this data reveals sensitivity to the isotope of adsorbing

■

CONCLUSION Through a combination of isotope-contrasted stroboscopic neutron diffraction, steady-state isotope transient kinetic analysis, gas isotherm measurements, and Grand Canonical 300

DOI: 10.1021/acs.chemmater.7b04594 Chem. Mater. 2018, 30, 296−302

Article

Chemistry of Materials

KC04062, under Contract No. DE-AC05-00OR22725. The data were measured on the Nanoscale-Ordered Materials Diffractometer (NOMAD) instrument at the Spallation Neutron Source at ORNL. Some samples were prepared and additional characterization was performed at the Center for Nanophase Materials Sciences, which is a DOE Office of Science User Facility. K.V.L. and P.M.F. would like to acknowledge funding by the National Nuclear Security Administration under the Stewardship Science Academic Alliances program through DOE Cooperative Agreement No. DE-NA0001982.

Monte Carlo simulation techniques, the heterogeneous distribution of nitrogen molecules in the faujasite structure of a zeolite X was determined under ambient operational conditions (300 K, 1 atm). The approach taken here shows an increased ability to locate the absorption sites (or distributions thereof) of gas molecules in porous structures, in contrast to traditional crystallography or cryogenic experiments. These results highlight the predictive power of the relatively simple force fields employed in the GCMC simulations, which are able to reproduce both macroscopic and microscopic observables. However, the microscopic, crystal chemical observables required isotope-contrasted stroboscopic neutron scattering with advanced data treatment to provide a unique experimental structural probe for gas−solid phase interactions. Unlike in the static case, where the lattice responds to adsorbed nitrogen relative to the neat and vacuum condition, the stroboscopic method can look at the structural correlations involved in adsorption alone under steady-state conditions. The powerful combination of neutron total scattering, gas isotherm measurements, and theoretical calculations each independently informs and supports the other and builds a consistent, predictive picture. This demonstrates a fertile framework for future investigation of industrially relevant gas sorption, separation, and catalysis processes also in operando conditions.

■

■

(1) Park, K. S.; Ni, Z.; Cote, A. P.; Choi, J. Y.; Huang, R. D.; UribeRomo, F. J.; Chae, H. K.; O’Keeffe, M.; Yaghi, O. M. Exceptional chemical and thermal stability of zeolitic imidazolate frameworks. Proc. Natl. Acad. Sci. U. S. A. 2006, 103 (27), 10186−10191. (2) Kulprathipanja, S. Zeolites in industrial separation and catalysis; John Wiley & Sons: 2010. (b) Cejka, J.; Corma, A.; Zones, S., Zeolites and catalysis: synthesis, reactions and applications; John Wiley & Sons: 2010. (3) Cejka, J.; Corma, A.; Zones, S. Zeolites and catalysis: synthesis, reactions and applications; John Wiley & Sons: 2010. (4) Yilmaz, B.; Muller, U. Catalytic Applications of Zeolites in Chemical Industry. Top. Catal. 2009, 52 (6−7), 888−895. (5) Yang, R. T. Adsorbents fundamentals and applications; WileyInterscience: Hoboken, NJ, 2003. (6) Ivanova, S.; Lewis, R. Producing Nitrogen via Pressure Swing Adsorption. Chem. Eng. Prog. 2012, 108 (6), 38−42. (7) Choudary, V. N.; Jasra, R. V.; Bhat, T. S. G. Adsorption of nitrogen-oxygen mixture in NaCaA zeolites by elution chromatography. Ind. Eng. Chem. Res. 1993, 32 (3), 548−552. (8) Miller, S. R.; Wright, P. A.; Devic, T.; Serre, C.; Ferey, G.; Llewellyn, P. L.; Denoyel, R.; Gaberova, L.; Filinchuk, Y. Single Crystal X-ray Diffraction Studies of Carbon Dioxide and Fuel-Related Gases Adsorbed on the Small Pore Scandium Terephthalate Metal Organic Framework, Sc-2(O2CC6H4CO2)(3). Langmuir 2009, 25 (6), 3618−3626. (9) Wright, P. A.; Thomas, J. M.; Ramdas, S.; Cheetham, A. K. Locating the sites of adsorbed species in heterogeneous catalysts: a rietveld neutron powder profile study of xenon in zeolite-rho. J. Chem. Soc., Chem. Commun. 1984, 20, 1338−1339. (10) Getzschmann, J.; Senkovska, I.; Wallacher, D.; Tovar, M.; Fairen-Jimenez, D.; Duren, T.; van Baten, J. M.; Krishna, R.; Kaskel, S. Methane storage mechanism in the metal-organic framework Cu3(btc)(2): An in situ neutron diffraction study. Microporous Mesoporous Mater. 2010, 136 (1−3), 50−58. (11) Eckert, J.; Draznieks, C. M.; Cheetham, A. K. Direct observation of host-guest hydrogen bonding in the zeolite NaY/ chloroform system by neutron scattering. J. Am. Chem. Soc. 2002, 124 (2), 170−171. (12) Hudson, M. R.; Queen, W. L.; Mason, J. A.; Fickel, D. W.; Lobo, R. F.; Brown, C. M. Unconventional, Highly Selective CO2 Adsorption in Zeolite SSZ-13. J. Am. Chem. Soc. 2012, 134 (4), 1970−1973. (13) Pichon, C.; Palancher, H.; Hodeau, J. L.; Berar, J. F. Towards operando characterisation by powder diffraction techniques of molecular sieves. Oil Gas Sci. Technol. 2005, 60 (5), 831−848. (14) Carrington, E. J.; Vitorica-Yrezabal, I. J.; Brammer, L. Crystallographic studies of gas sorption in metal-organic frameworks. Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater. 2014, 70, 404− 422. (15) Ledesma, C.; Yang, J.; Chen, D.; Holmen, A. Recent Approaches in Mechanistic and Kinetic Studies of Catalytic Reactions Using SSITKA Technique. ACS Catal. 2014, 4 (12), 4527−4547. (16) Shannon, S. L.; Goodwin, J. G. Characterization of Catalytic Surfaces by Isotopic-Transient Kinetics during Steady-State Reaction. Chem. Rev. 1995, 95 (3), 677−695.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.chemmater.7b04594. Experimental details for the gas adsorption measurements, details on the GCMC simulations, crystallographic data refinement, Fourier difference maps, and data reduction of stroboscopically collected data (PDF)

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*(K.P.) E-mail: [email protected]. ORCID

Daniel Olds: 0000-0002-4611-4113 Keith V. Lawler: 0000-0003-1087-5815 Arnold A. Paecklar: 0000-0003-4180-1923 Jue Liu: 0000-0002-4453-910X Katharine Page: 0000-0002-9071-3383 Paul M. Forster: 0000-0003-3319-4238 James R. Neilson: 0000-0001-9282-5752 Author Contributions

∥ (D.O., K.V.L., and A.A.P.) These authors contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was primarily supported under the Department of Energy’s Office of Basic Energy Sciences and the Laboratory Directed Research and Development (LDRD) Program at Oak Ridge National Laboratory (LDRD Seed No. 7735). The presented analysis of neutron powder diffraction data was funded by the BES Early Career Award, Exploiting Small Signatures: Quantifying Nanoscale Structure and Behavior 301

DOI: 10.1021/acs.chemmater.7b04594 Chem. Mater. 2018, 30, 296−302

Article

Chemistry of Materials (17) Goguet, A.; Meunier, F. C.; Tibiletti, D.; Breen, J. P.; Burch, R. Spectrokinetic investigation of reverse water-gas-shift reaction intermediates over a Pt/CeO2 catalyst. J. Phys. Chem. B 2004, 108 (52), 20240−20246. (18) Zhang, J. P.; Zhu, A. X.; Chen, X. M. Single-crystal X-ray diffraction and Raman spectroscopy studies of isobaric N-2 adsorption in SOD-type metal-organic zeolites. Chem. Commun. 2012, 48 (93), 11395−11397. (19) Hornes, A.; Hungria, A. B.; Bera, P.; Camara, A. L.; FernandezGarcia, M.; Martinez-Arias, A.; Barrio, L.; Estrella, M.; Zhou, G.; Fonseca, J. J.; Hanson, J. C.; Rodriguez, J. A. Inverse CeO2/CuO Catalyst As an Alternative to Classical Direct Configurations for Preferential Oxidation of CO in Hydrogen-Rich Stream. J. Am. Chem. Soc. 2010, 132 (1), 34−35. (20) Hanson, J. C.; Si, R.; Xu, W.; Senanayake, S. D.; Mudiyanselage, K.; Stacchiola, D.; Rodriguez, J. A.; Zhao, H.; Beyer, K. A.; Jennings, G.; Chapman, K. W.; Chupas, P. J.; MartinezArias, A. Pulsed-reactant in situ studies of ceria/CuO catalysts using simultaneous XRD, PDF and DRIFTS measurements. Catal. Today 2014, 229, 64−71. (21) Clausen, B. S. Combined (Q)EXAFS/XRD: Technique and applications. Catal. Today 1998, 39 (4), 293−300. (22) Chapman, K. W.; Chupas, P. J.; Kepert, C. J. Selective recovery of dynamic guest structure, in a nanoporous Prussian blue through in situ X-ray diffraction: A differential pair distribution function analysis. J. Am. Chem. Soc. 2005, 127 (32), 11232−11233. (23) Chupas, P. J.; Chapman, K. W.; Lee, P. L. Applications of an amorphous silicon-based area detector for high-resolution, highsensitivity and fast time-resolved pair distribution function measurements. J. Appl. Crystallogr. 2007, 40, 463−470. (24) Kim, H.; Karkamkar, A.; Autrey, T.; Chupas, P.; Proffen, T. Determination of Structure and Phase Transition of Light Element Nanocomposites in Mesoporous Silica: Case study of NH3BH3 MCM-41. J. Am. Chem. Soc. 2009, 131 (38), 13749−13755. (25) Chapman, K. W.; Chupas, P. J.; Nenoff, T. M. Radioactive Iodine Capture in Silver-Containing Mordenites through Nanoscale Silver Iodide Formation. J. Am. Chem. Soc. 2010, 132 (26), 8897− 8899. (26) Turner, J. F. C.; Benmore, C. J.; Barker, C. M.; Kaltsoyannis, N.; Thomas, J. M.; David, W. I. F.; Catlow, C. R. A. Probing the Nature of Acetylene Bound to the Active Site of a NiNa−Zeolite Y Catalyst by in situ Neutron Scattering. J. Phys. Chem. B 2000, 104 (32), 7570−7573. (27) Jobic, H.; Schober, H.; Pullumbi, P. Adsorption and diffusion of N-2 and O-2 in LiLSX studied by neutron scattering techniques. Adsorption 2005, 11, 449−454. (28) Neuefeind, J.; Feygenson, M.; Carruth, J.; Hoffmann, R.; Chipley, K. K. The Nanoscale Ordered MAterials Diffractometer NOMAD at the Spallation Neutron Source SNS. Nucl. Instrum. Methods Phys. Res., Sect. B 2012, 287, 68−75. (29) Wu, H.; Simmons, J. M.; Liu, Y.; Brown, C. M.; Wang, X. S.; Ma, S.; Peterson, V. K.; Southon, P. D.; Kepert, C. J.; Zhou, H. C.; Yildirim, T.; Zhou, W. Metal-Organic Frameworks with Exceptionally High Methane Uptake: Where and How is Methane Stored? Chem. Eur. J. 2010, 16 (17), 5205−5214. (30) Olds, D.; Page, K.; Paecklar, A.; Peterson, P. F.; Liu, J.; Rucker, G.; Ruiz-Rodriguez, M.; Olsen, M.; Pawel, M.; Overbury, S. H.; Neilson, J. R. A high precision gas flow cell for performing in situ neutron studies of local atomic structure in catalytic materials. Rev. Sci. Instrum. 2017, 88 (3), 034101. (31) Petříček, V.; Dušek, M.; Palatinus, L. Crystallographic Computing System JANA2006: General features. Z. Kristallogr. Cryst. Mater. 2014, 229 (5), 345−352. (32) Arnold, O.; Bilheux, J. C.; Borreguero, J. M.; Buts, A.; Campbell, S. I.; Chapon, L.; Doucet, M.; Draper, N.; Leal, R. F.; Gigg, M. A.; Lynch, V. E.; Markvardsen, A.; Mikkelson, D. J.; Mikkelson, R. L.; Miller, R.; Palmen, K.; Parker, P.; Passos, G.; Perring, T. G.; Peterson, P. F.; Ren, S.; Reuter, M. A.; Savici, A. T.; Taylor, J. W.; Taylor, R. J.; Tolchenov, R.; Zhou, W.; Zikovsky, J. Mantid-Data

analysis and visualization package for neutron scattering and mu SR experiments. Nucl. Instrum. Methods Phys. Res., Sect. A 2014, 764, 156−166. (33) Walton, K. S.; Snurr, R. Q. Applicability of the BET method for determining surface areas of microporous metal-organic frameworks. J. Am. Chem. Soc. 2007, 129 (27), 8552−8556. (34) Potoff, J. J.; Siepmann, J. I. Vapor-liquid equilibria of mixtures containing alkanes, carbon dioxide, and nitrogen. AIChE J. 2001, 47 (7), 1676−1682. (35) Lawler, K. V.; Sharma, A.; Alagappan, B.; Forster, P. M. Assessing zeolite frameworks for noble gas separations through a joint experimental and computational approach. Microporous Mesoporous Mater. 2016, 222, 104−112. (36) Buttefey, S.; Boutin, A.; Mellot-Draznieks, C.; Fuchs, A. H. A simple model for predicting the Na+ distribution in anhydrous NaY and NaX zeolites. J. Phys. Chem. B 2001, 105 (39), 9569−9575. (37) Jeffroy, M.; Nieto-Draghi, C.; Boutin, A. Molecular simulation of zeolite flexibility. Mol. Simul. 2014, 40 (1−3), 6−15. (38) Pillai, R. S.; Sethia, G.; Jasra, R. V. Sorption of CO, CH4, and N-2 in Alkali Metal Ion Exchanged Zeolite-X: Grand Canonical Monte Carlo Simulation and Volumetric Measurements. Ind. Eng. Chem. Res. 2010, 49 (12), 5816−5825. (39) Pluth, J. J.; Smith, J. V. Positions of Cations and Molecules in Zeolites with Faujasite-Type Framework 0.7. Dehydrated CaExchanged X. Mater. Res. Bull. 1972, 7 (11), 1311−1321. (40) Frising, T.; Leflaive, P. Extraframework cation distributions in X and Y faujasite zeolites: A review. Microporous Mesoporous Mater. 2008, 114 (1−3), 27−63. (41) Wong-Ng, W.; Kaduk, J. A.; Huang, Q.; Espinal, L.; Li, L.; Burress, J. W. Investigation of NaY Zeolite with adsorbed CO2 by neutron powder diffraction. Microporous Mesoporous Mater. 2013, 172, 95−104. (42) Fitch, A. N.; Jobic, H.; Renouprez, A. Localization of benzene in sodium-Y-zeolite by powder neutron diffraction. J. Phys. Chem. 1986, 90 (7), 1311−1318. (43) Bae, Y. S.; Lee, C. H. Sorption kinetics of eight gases on a carbon molecular sieve at elevated pressure. Carbon 2005, 43 (1), 95−107. (44) Bai, P.; Ghosh, P.; Sung, J. C.; Kohen, D.; Siepmann, J. I.; Snurr, R. Q. A computational study of the adsorption of nperfluorohexane in zeolite BCR-704. Fluid Phase Equilib. 2014, 366, 146−151. (45) Goursot, A.; Vasilyev, V.; Arbuznikov, A. Modeling of adsorption properties of zeolites: Correlation with the structure. J. Phys. Chem. B 1997, 101 (33), 6420−6428. (46) Qin, Z.; Cychosz, K. A.; Melinte, G.; El Siblani, H.; Gilson, J.P.; Thommes, M.; Fernandez, C.; Mintova, S.; Ersen, O.; Valtchev, V. Opening the Cages of Faujasite-Type Zeolite. J. Am. Chem. Soc. 2017, 139 (48), 17273−17276. (47) Salles, F.; Maurin, G.; Serre, C.; Llewellyn, P. L.; Knofel, C.; Choi, H. J.; Filinchuk, Y.; Oliviero, L.; Vimont, A.; Long, J. R.; Ferey, G. Multistep N-2 Breathing in the Metal-Organic Framework Co(1,4benzenedipyrazolate). J. Am. Chem. Soc. 2010, 132 (39), 13782− 13788. (48) Zeidler, A.; Salmon, P. S.; Fischer, H. E.; Neuefeind, J. C.; Simonson, J. M.; Markland, T. E. Isotope effects in water as investigated by neutron diffraction and path integral molecular dynamics. J. Phys.: Condens. Matter 2012, 24 (28), 284126. (49) Spencer, E. C.; Howard, J. A. K.; McIntyre, G. J.; Rowsell, J. L. C.; Yaghi, O. M. Determination of the hydrogen absorption sites in Zn4O(1,4-benzenedicarboxylate) by single crystal neutron diffraction. Chem. Commun. 2006, 3, 278−280.

302

DOI: 10.1021/acs.chemmater.7b04594 Chem. Mater. 2018, 30, 296−302