A mesoporous Zr-based metal–organic framework driven by the assembly of an octatopic linker

Metal–organic frameworks (MOFs) based on high-connected nets are generally very attractive due to their combined robustness and porosity. Here, we describe the synthesis of BCN-348, a new high-connected Zr-MOF built from an 8-connected (8-c) cubic Zr-oxocluster and an 8-c organic linker. BCN-348 contains a minimal edge-transitive 3,4,8-c eps net, and combines mesoporosity with thermal and hydrolytic stability. Encouraging results from preliminary studies on its use as a catalyst for hydrolysis of a nerve-agent simulant suggest its potential as an agent for detoxification of chemical weapons and other pernicious compounds.


Synthesis of BCN-348
A solution of ZrOCl2·8H2O (26 mg, 0.08 mmol) and H8TBCPB (58 mg, 0.04 mmol) in DMF (2 mL) and trifluoroacetic acid (0.6 mL) was prepared in a 23 mL scintillation vial. Then, the sealed vial was placed into a preheated oven at 120 ºC for 5 days. After this period, colourless cubic crystals suitable for single-crystal X-ray diffraction (SCXRD) were collected by filtration and washed three times by incubating them with 20 mL of fresh DMF for 12 h. Afterwards, solvent exchange with acetone was performed by incubating the crystals three times with 20 mL of acetone for 12 h, and the resulting BCN-348 crystals were dried at room temperature.

S1.3. Characterization
Single-Crystal X-Ray Diffraction (SCXRD) data for BCN-348 were collected at 100 K at XALOC beamline at ALBA synchrotron (0.82656 Å). 2 Data were indexed, integrated and scaled using the XDS program. 3 Absorption correction was not applied. The structures were solved by direct methods and subsequently refined by correction of F2 against all reflections, using SHELXT2018 within Olex2 package. 4-6 All non-hydrogen atoms were refined with anisotropic thermal parameters by full-matrix least-squares calculations on F2 using the program SHELXL2018. 5 We treated the presence of solvent molecules in the cavities of all structures running solvent mask using Olex2 solvent mask. 6 The structure was solved using two space groups: Fm-3m and Fm-3. Although the correct symmetry corresponding to the structure is Fm-3m, we could not model the disorder of the central benzene ring of the linker (which could not be clearly defined using the maps) using this space group. For this reason, we decided to solve the crystal structure using F-3m group, which allowed us to model this benzene ring applying some restraints, such as DFIX, DELU and SIMU. The hydrogen atoms were calculated in their expected positions with the HFIX instruction of SHELXL2018, and refined as riding atoms with Uiso(H) = 1.5 Ueq(C). Powder X-Ray Diffraction (PXRD) diagrams were collected on a Panalytical X'pert mpd diffractometer with monochromatic Cu-Ka radiation (lCu = 1.5406 Å). Thermogravimetric analyses (TGA) were performed in a Pyris TGA8000, heating the sample from 25 ºC to 800 ºC at 1 ºC/min under N2 atmosphere.

Proton Nuclear Magnetic
N2 sorption isotherms were collected at 77K (N2) using an ASAP 2460 HD (Micromeritics). Temperature was controlled by using a liquid nitrogen bath. Prior to the measurements, the samples were activated by solvent exchange with acetone, 3 times per day during 3 days. The specific surface area was obtained using the Brunauer-Emmett-Teller (BET) adsorption method in the range of 0.001 -0.02 P/P0 (BETSI software). 7 Total pore volume (Vt) was calculated at P/P0 = 0.95. Pore size distribution was estimated using a density functional theory (DFT) model (HS-2D-NLDFT Carbon Cylindrical Mesopores @ N2 77K) implemented in the Microactive 4.06 software with a regularization factor of 0.03160.

Gas Chromatography analyses (GC)
were carried out in an Agilent 8860 GC chromatograph with a FID dectector and a 16 port autosampler. A HP-5 column of 50 m length, 0.320 mm diameter and 1.05 µm thickness was used, which allowed working from -60 ºC to 325 ºC.      The degradation of DIFP nerve gas simulant into diisopropylphosphate (DIP) by BCN-348 was followed by 1 H NMR and 31 P NMR spectroscopy. Specifically, 12 mg of BCN-348 were suspended in a mixture containing 0.6 mL of D2O, 0.7 μL of dimethylphosphate (DMP, used as internal reference) and 1.5 μL of DIFP (DIFP/MOF ratio of 1.5:1). The suspension was stirred during 24 hours at room temperature and the concentration of DIPF and DIP in the supernatant solution and in the adsorbate phase were quantified by NMR.

S2. Characterization of BCN-348
After 24 hours, the solid was separated from the supernatant by centrifugation (10000 rpm, 5 min). Afterwards, the solid was suspended in CDCl3 for 5 hours at room temperature to extract DIFP molecules adsorbed inside the cavities of BCN-348 and the concentration of DIFP and its degradation product DIP were determined by 1 H-NMR. Table S2. Percentage distribution of unreacted and hydrolyzed DIFP in the supernatant and adsorbate phase after 24 hours of incubation with BCN-348 as determined from 1 H NMR in D2O (0.6 mL) and after extraction with CDCl3 (0.6 mL), respectively.