High capacity ammonia adsorption in a robust metal–organic framework mediated by reversible host–guest interactions

To understand the exceptional adsorption of ammonia (NH3) in MFM-300(Sc) (19.5 mmol g−1 at 273 K and 1 bar without hysteresis), we report a systematic investigation of the mechanism of adsorption by a combination of in situ neutron powder diffraction, inelastic neutron scattering, synchrotron infrared microspectroscopy, and solid-state 45Sc NMR spectroscopy. These complementary techniques reveal the formation of reversible host–guest supramolecular interactions, which explains directly the observed excellent reversibility of this material over 90 adsorption–desorption cycles.


Synthesis of MFM-300(Sc)
MFM-300(Sc) was synthesised by a solvothermal method according to the literature. 1 Scandium triflate (900 mg, 1.83 mmol) and biphenyl-3,3',5,5'-tetracarboxylic acid (H 4 L, 300 mg, 0.91 mmol) were mixed in dimethylformamide (DMF, 105 mL), H 2 O (15 mL) and HCl (36.5%, 3 mL). The mixture was stirred until complete dissolution occurred. The solution was then placed in a pressure tube and heated in an oil bath to 80 o C for 72 h. The tube was then cooled to room temperature, and the colourless crystalline product was separated by filtration, washed with DMF three times and stored in acetone. Yield: 70% (based on ligand).

Characterisations
Powder X-ray diffraction (PXRD) patterns was performed for the as-synthesised, post-isotherm and post-cycling experiments samples of MFM-300(Sc) on a Philips X'pert X-ray diffractometer (40 kV and 30 mA) using Cu-Kα radiation (λ = 1.5406 Å). The data were collected at room temperature in a 2θ range of 5-50 with a scan speed of 4° min −1 . TGA was conducted on a TA Instrument Q600 under N 2 flow of 50 mL min -1 .
10 mg sample was added into an alumina pan and heated from room temperature with a ramp rate of 5 °C min -1 up to 800 °C.
BET surface areas were obtained from N 2 isotherms recorded on a 3-flex instrument at 77 K. The predried acetone-exchanged materials (100-150 mg) were loaded into a sample cell and subjected to a dynamic vacuum (1×10 −7 mbar) at 443 K for 10 h. Measurements of static adsorption isotherms (0-1.0 bar) for NH 3 were undertaken on an IGA gravimetric sorption analyser (Hiden Isochema, Warrington, UK). NH 3 gravimetric sorption isotherms were recorded at 273, 283, 293, 298, 303, and 313 K under ultra-high vacuum produced by a turbo pumping system with the temperature controlled using a programmed water bath and furnace bath.
Research-grade NH 3 was purchased from BOC and used as received. In a typical gas adsorption experiment, 40 mg of acetone-exchanged MFM-300(Sc) was loaded into the IGA system and outgassed dynamic vacuum (1 × 10 −8 mbar) at 453 K for 12 h. For the cycling experiments, the pressure of NH 3 was increased from vacuum (1 × 10 −8 mbar) to 200 mbar and the uptake recorded. The pressure was then reduced to regenerate the sample without heating. This cycling process was repeated 90 times.
Breakthrough experiments were conducted on a Hiden Isochema IGA-003 with ABR attachments and a Hiden Analytical mass spectrometer using a fixed-bed tube packed with 410 mg of MFM-300(Sc). The sample was activated by heating under a flow of He at 423 K for 12 h. The fixed-bed was then cooled to 298 K. A breakthrough curve was collected with a flow of 1000 ppm NH 3 diluted in He. The flow rate of the entering gas was maintained at 25 mL min −1 , and the concentration of NH 3 in the exhaust gas was determined by mass spectrometry and compared with the inlet concentration C 0 , where C/C 0 = 1 indicates complete breakthrough.
For tests of chemical stability, 20 mg of MFM-300(Sc) was placed in a small vial and immersed under solutions of various pH values (pH = 7-12) and different organic solvents. The vial was sealed and retained at room temperature for 12 h.
The structural determination of the binding positions of ND 3 in MFM-300(Sc) was conducted using WISH, a long wavelength powder and single crystal neutron diffractometer at the ISIS neutron and muon facility at Rutherford Appleton Laboratory (UK). The instrument has a solid methane moderator providing a high flux of cold neutrons with a large bandwidth, transported to the sample via an elliptical guide. The WISH system of divergence jaws allows tuning of the resolution according to the need of the experiment; in this case, it was setup in high resolution mode. The WISH detectors are 1m long, 8mm diameter pixelated 3He tubes positioned at 2.2m from the sample and arranged on a cylindrical locus covering a 2θ scattering angle of 10-170°. To reduce the background from the sample environment, WISH was equipped with an oscillating radial collimator that defines a cylinder of radius of approximately 22 mm diameter at 90° scattering. The sample of desolvated MFM-300(Sc) was loaded into a cylindrical vanadium sample container with an indium vacuum seal connected to a gas handling system. The sample was degassed at 1x10 -7 mbar and at 100 °C for 4 days with He flushing to remove any remaining trace of guest water. The sample was dosed with ND 3 using the volumetric method after being warned to room temperature to ensure that the gas is well dispersed throughout the crystalline structure of MFM-300(Sc). Data collection for desolvated MFM-300(Sc) and two subsequent loadings of ND 3 (0.625 and 1.3 ND 3 molecules per OH functionality) were performed controlled using a He cryostat (10 ± 0.2 K).
In situ synchrotron infrared micro-spectroscopy experiments were carried out at multimode infrared imaging and micro spectroscopy (MIRIAM) beamline at the Diamond Light Source, UK. The instrument is comprised of a Bruker Hyperion 3000 microscope in transmission mode, with 15x objective and condenser lenses and a small element (50 µm) liquid N 2 cooled MCT detector, coupled to a Bruker Vertex 80 V Fourier Transform IR interferometer using radiation generated from a bending magnet source. Spectra were collected (512 scans) in the range 400-4000 cm -1 at 4 cm -1 resolution and an infrared spot size at the sample of approximately 15 × 15 µm. A microcrystalline powder of MFM-300(Sc) was scattered onto a 0.5 mm thick ZnSe infrared window and placed within a Linkam FTIR 600 gas-tight sample cell equipped with 0.5 mm thick ZnSe windows, a heating stage and gas inlet and outlet. Ultrapure N 2 and anhydrous NH 3 gases were used as supplied from the cylinder. The gases were flowed through the gas delivery system prior to connection to the Linkam cell to remove air and moisture. The gases were dosed volumetrically to the sample cell using mass flow controllers, and the total flow rate being maintained at 100 mL min -1 for all experiments. The exhaust from the cell was directly vented to an extraction system and the total pressure in the cell was therefore 1 bar for all experiments. The sample was desolvated under a flow of dry N 2 at 100 mL min -1 and 443 K for 5 h, and was then cooled to 298 K under a continuous flow of N 2 . Dry NH 3 was then dosed as a function of partial pressure, maintaining a total flow of 100 mL min -1 . The MOF sample was regenerated with a flow of dry N 2 .
Magic angle spinning (MAS) NMR spectra were recorded using a Bruker 9.4 T (400 MHz 1H Larmor frequency) AVANCE III spectrometer equipped with a 4 mm HFX MAS probe. Samples were treated and packed into 4 mm o.d. zirconia rotors under inert conditions and sealed with a Kel-F rotor cap. Experiments were acquired at ambient temperature using a MAS frequency of 12 kHz. 1 H-pulses of 100 kHz were used for excitation and SPINAL-64 2 heteronuclear decoupling, 45 Sc-pulses of 0.5 s duration (small flip angle with radio frequency field amplitude of ~70 kHz) were employed for 45 Sc direct excitation (DE)MAS experiments, and 13 C-pulses and spin-locking at 50 kHz were used for { 1 H-} 13 C CPMAS experiments with corresponding ramped (70-100%) 1 H spin-locking at ~73 kHz (100%) for 2 ms; s Hahn-echo  r -- r sequence of 2 rotor periods total duration was applied to 13  INS spectra were collected on the TOSCA beamline at ISIS Neutron and Muon Source (UK). The sample of desolvated MFM-300(Sc) was loaded into a cylindrical vanadium sample container with an indium vacuum seal and this was connected to a gas handling system. The sample was degassed at 393 K and 10 −7 mbar for 24 h to remove any residual trace of guest water. The temperature during data collection was controlled using a closed cycle refrigerator cryostat (10 ± 0.1 K). The loading of NH 3 was performed volumetrically at room temperature, and subsequently the temperature was reduced to 10 K in order to minimize achievable thermal motion of the framework and adsorbed NH 3 molecules in the scattering measurements. Background spectra of MFM-300(Sc) were subtracted to obtain the difference spectra.
Modelling by Density Functional Theory (DFT) of the bare and NH 3 -loaded MFM-300(Sc) was performed using the Vienna Ab initio Simulation Package (VASP). 3 The calculation used Projector Augmented Wave (PAW) method 4,5 to describe the effects of core electrons, and Perdew-Burke-Ernzerhof (PBE) 6 implementation of the Generalized Gradient Approximation (GGA) for the exchange-correlation functional.
Energy cutoff was 800 eV for the plane-wave basis of the valence electrons. The lattice parameters and atomic coordinates determined by neutron powder diffraction in this work were used as the initial structure. The electronic structure was calculated on the Γ-point for the unit cell (144 atoms for the blank MOF). The total energy tolerance for electronic energy minimization was 10 −8 eV, and for structure optimization it was 10 −7 eV.
The maximum interatomic force after relaxation was below 0.001 eV/Å, and the optB86b-vdW functional for dispersion corrections was applied. 7 The vibrational eigen-frequencies and modes were then calculated by solving the force constants and dynamical matrix using Phonopy. 8 The OClimax software was used to convert the DFT-calculated phonon results to the simulated INS spectra. 9

Characterisation of Porosity
Prior to the measurement, MFM-300(Sc) was activated under dynamic vacuum at 180 °C for 12 h. The adsorption-desorption isotherms for N 2 were carried out at 77 K. Fig. S2 shows a type-I profile with a surface area of 1390 m 2 g -1 .    (b) Adsorption isotherms for N 2 at 77 K in MFM-300(Sc) before and after two cycles NH 3 adsorption.

Calculation of Isosteric Heats of Adsorption
To calculate the differential enthalpies (ΔH n ) and entropies of adsorption (ΔS n ) for NH 3 uptake as a function of gas adsorption (n), isotherms were measured over a range of temperatures and fitted to the van't Hoff isochore

Neutron Powder Diffraction
Rietveld refinements of the NPD patterns of the bare MOF and samples at various ND 3 loadings were performed using the TOPAS software package. In this treatment the guest molecules are treated as rigid bodies; we first refined the centres of mass, orientations, and occupancies of the adsorbate, followed by full profile Rietveld      MHz,  Q = 1, Gaussian broadening = 1.14 kHz for NH 3 -loaded MFM-300(Sc).

Inelastic neutron scattering
INS was applied to study the binding interaction and structure dynamics in this case because of its several unique advantages: (a) INS spectroscopy is ultra-sensitive to the vibrations of hydrogen atoms, which is ten times more visible than other elements due to its high neutron cross-section.