Solid state organic amine detection in a photochromic porous metal organic framework

A new Mg(ii) based porous metal–organic framework (MOF) has been synthesized from naphthalenediimide (NDI) chromophoric unit containing linker. This MOF (Mg–NDI) shows instant and reversible photochromism as well as solvatochromic behavior. Due to the presence of electron deficient NDI moiety, this MOF exhibits selective organic amine (electron rich) sensing in solid state.


S1: Materials and methods
All reagents were commercially available and used as received. Powder X-ray diffraction (PXRD) patterns were recorded on a Rigaku Smartlab diffractometer for Cu K  radiation ( = 1.5406 Å), with a scan speed of 2° min -1 and a step size of 0.02° in 2. Fourier transform infrared (FT-IR) spectra were taken on a Bruker Optics ALPHA-E spectrometer with a universal Zn-Se ATR (attenuated total reflection) accessory in the 600-4000 cm -1 region or using a Diamond ATR (Golden Gate). Thermo-gravimetric' analyses (TGA) were carried out on a SDT Q600 TG-DTA analyzer under N 2 atmosphere at a heating rate of 5 ºC min -1 within a temperature range of 40-900 °C. Orange color crystals were separated after the reaction and preserved in dry DMF/DEF for further applications/characterizations.

S3: Single crystal XRD and crystal structures of Mg-NDI and H 4 BINDI
As synthesized crystals of Mg-NDI and H 4 BINDI were coated with paratone-N and placed on top of a nylon cryoloop (Hampton research) and then mounted in the diffractometer. The data collection was done at 150 and 293 K, respectively. The crystals were mounted on a Super Nova Dual source X-ray Diffractometer system (Agilent Technologies) equipped with a CCD area detector and operated at 250 W power (50 kV, 0.8 mA) to generate Mo Kα radiation (λ = 0.71073 Å) and Cu Kα radiation (λ = 1.54178 Å) at 298(2) K. Initial scans of each specimen were performed to obtain preliminary unit cell parameters and to assess the mosaicity (breadth of spots between frames) of the crystal to select the required frame width for data collection. CrysAlis Pro program software was used suite to carry out overlapping φ and ω scans at detector (2θ) settings (2θ = 28). Following data collection, reflections were sampled from all regions of the Ewald sphere to redetermine unit cell parameters for data integration.
In no data collection was evidence for crystal decay encountered. Following exhaustive review of collected frames the resolution of the data set was judged. Data were integrated using CrysAlis Pro software with a narrow frame algorithm. Data were subsequently corrected for absorption by the program SCALE3 ABSPACK 2 scaling algorithm.
These structures were solved by direct method and refined using the SHELXTL 97 3 software suite. Atoms were located from iterative examination of difference F-maps following least squares refinements of the earlier models. Final model was refined anisotropically (if the number of data permitted) until full convergence was achieved. Hydrogen atoms were placed in calculated positions (C-H = 0.93 Å) and included as riding atoms with isotropic displacement parameters 1.2-1.5 times Ueq of the attached C atoms. In some cases modeling of electron density within the voids of the frameworks did not lead to identification of recognizable solvent molecules in these structures, probably due to the highly disordered contents of the 7 large pores in the frameworks. Highly porous crystals that contain solvent-filled pores often yield raw data where observed strong (high intensity) scattering becomes limited to ~1.0 Å at best, with higher resolution data present at low intensity. A common strategy for improving X-ray data, increasing the exposure time of the crystal to X-rays, did not improve the quality of the high angle data in this case, as the intensity from low angle data saturated the detector and minimal improvement in the high angle data was achieved. Additionally, diffused scattering from the highly disordered solvent within the void spaces of the framework and from the capillary to mount the crystal contributes to the background and the 'washing out' of the weaker data. Electron density within void spaces has not been assigned to any guest entity but has been modeled as isolated oxygen and/or carbon atoms. The foremost errors in all the models are thought to lie in the assignment of guest electron density. The structure was examined using the ADSYM subroutine of PLATON 4 to assure that no additional symmetry could be applied to the models. The ellipsoids in ORTEP diagrams are displayed at the 50% probability level unless noted otherwise.
Generally             shapes are representing the adsorption and desorption respectively.

S11: Electrochemical measurements
The reduction potentials of the Mg-NDI were measured using a three-electrode cell at room temperature. A platinum electrode was used as the working electrode, platinum wire as the counterelectrode, and a platinum wire as the reference. Electrochemical measurements of the analytes were carried out using 0.01 mmol solutions of tetrabutylammonium hexafluoro phosphate solution in acetonitrile. For the MOFs, the powdered materials were coated on 3 mm wide platinum electrode. The reduction potentials of the compounds were obtained from the cyclic voltammograms and corrected with respect to the Fc/Fc + internal standard.

S12: DFT calculation for Mg-NDI
For each solvent@MOF combination, 100 initial geometries were generated using the Kick 3 stochastic structure generator 5 . Positions of the solvent molecules were optimised using DFTB, 6,7 keeping the framework fixed at the experimental geometry. The optimised structures were ranked in terms of energy and the energy levels of the HOMO and LUMO were extracted for the lowest energy structure of each solvent@MOF system.
To generate cube files for the frontier orbitals, single point calculations of the DFTB optimized geometry in periodic boundary conditions were performed at DFT level. Density functional theory (DFT) was applied within the generalized gradient approximation, using the PBE XC functional. 8 The Gaussian and Plane-Wave method, as it is implemented in CP2K package, 9,10 with DZVP-MOLOPT-GTH basis set and Goedecker-Teter-Hutter pseudopotentials 11,12 was applied for all atoms. The charge density cutoff of the finest grid level is equal to 400 Ry. The number of used multigrids is 5.