Reversible coordinative binding and separation of sulphur dioxide in a robust metal-organic framework with open copper sites

have on the and human health, but SO 2 is also an important industry feedstock if it can be recovered, stored and transported efficiently. Here we report the exceptional adsorption and separation of SO 2 in a new porous material, [Cu 2 (L)] (H 4 L = 4',4'''-(pyridine-3,5-diyl)bis([1,1'-biphenyl]-3,5-dicarboxylic acid), MFM-170. MFM-170 exhibits fully reversible SO 2 uptake of 17.5 mmol g -1 at 298 K, 1.0 bar, and the SO 2 binding domains for trapped molecules within MFM-170 have been determined. Significantly, we report the first example of reversible co-ordination of SO 2 to open Cu(II) sites in a porous material, contributing to excellent adsorption thermodynamics and selectivities for SO 2 binding, as well as facile regeneration of MFM-170 post adsorption. MFM-170 is stable to water, acid and base and shows significant promise for the dynamic separation of SO 2 from simulated flue gas mixtures, as confirmed by breakthrough experiments. synthesis and characterisation of MOF samples, measurements of adsorption isotherms. GLS, XH: measurements and analysis of the breakthrough data. GLS, XZ, SPA, LJM, SJTL, SY: collection and analysis of synchrotron single crystal X-ray diffraction data. GLS, HGWG, YC, SR, AJRC: collection and analysis of neutron scattering data. GLS, SJD and CCT: collection and analysis of long duration synchrotron X-ray diffraction data. GLS, JL, NMJ, MDF, GC, TLE: collection and analysis of synchrotron IR data. SY and MS: overall design and direction of project. GLS, SY and MS: preparation of the manuscript with help from all authors.

The former is subject to an irreversible phase change on SO2 uptake, whilst the steep uptake of the latter may render it unfeasible for practical PSA applications. 26,27 Indeed, a trade-off often exists between the uptake at low-pressure and the energy cost of system regeneration. Another consideration is that sorbents for use in FGD processes must be located upstream of CO2 scrubbing units, and therefore require high selectivity for SO2 over CO2 at low partial pressures of SO2 (~2000 ppm). Although open metal sites in MOFs can improve gas binding selectivity, the resultant MOFs are often subject to severe framework degradation upon contact with water, precluding their practical applications. [28][29][30][31] Similarly, coordination of strongly complexing SO2 molecules to open metal centres can disrupt the linker-metal coordination and cause structural collapse.
Herein, we report the first example of reversible coordinative binding of SO2 to open Cu(II) sites in a remarkably robust material, MFM-170, leading to optimal adsorption and selectivity for SO2.  (3,36)-connected net with txt topology (Fig. 1b). [32][33][34] The metal cluster consists of a Cu2(O2CR)4 paddlewheel with four isophthalate units occupying the equatorial sites and one pyridyl N-donor from the ligand coordinating to the axial site of one Cu atom. The axial position of the other Cu atom of the Cu2(O2CR)4 unit is occupied by a water molecule. The framework is constructed from Cu24(RC6H3(CO2)2)24 cuboctahedron, which acts as a 36-connected node, joined in a cubic array to six adjacent cuboctahedra by four ligands each (Fig. 1c). The overall framework can be visualised as this smaller cubic net which is connected to a secondary identical net via the 12 corners of the cuboctahedra via Cu-N bonds (Fig. 1d).
Thus, each ligand is a 3-connected node, with two isophthalate moieties that each connect an edge of a cuboctahedron, and one pyridyl N atom which joins a corner of a cuboctahedron.
The interconnected void spaces in MFM-170 can be considered as three distinct cages, denoted as A, B and C (Fig. 1e) Fig. S10. To assess the long-term stability of MFM-170 to humid SO2 and water, synchrotron PXRD data were collected for wet SO2-loaded MFM-170 samples every week for 10 weeks ( Fig. S7; see SI for further details). No loss of crystallinity or change in the structure of this material was observed (Table S3), confirming the excellent chemical resilience of the framework. The remarkable stability of MFM-170 is attributed to the unusual framework connectivity where the axially-coordinated pyridyl N-donors interlock the two cubic nets and block one of the two axial Cu(II) sites.

Analysis of gas adsorption isotherms of MFM-170 and MFM-170·H2O. Desolvated MFM-170
possesses a BET surface area of 2408 m 2 g -1 (consistent with the calculated surface area of 2456 m 2 g -1 based upon the crystal structure) and a pore volume of 0.88 cm 3 g -1 (calculated from the N2 isotherm at 77 K, Fig. S9), consistent with that (0.87 cm 3 g -1 ; solvent-accessible void space of 61%) derived from the single crystal structure. Significantly, MFM-170 shows an unprecedented SO2 uptake of 19.4 mmol g -1 (or 1.24 g g -1 ) at 273 K and 1.0 bar (Fig. 3). To the best of our knowledge, this represents the highest known SO2 uptake capacity in porous materials, followed by MFM-601 (16.9 mmol g -1 ) 15 , MFM-202a (13.0 mmol g -1 ) 18 and mesoporous silicate MCM-41 (11.6 mmol g -1 ) 35 under the same conditions. The performance of state-of-the-art porous materials under ambient conditions is summarised in Table 1  [Zn2(L1)2(bipy)] (10.9 mmol g -1 , at 293 K), 17 and Ni(bdc)(ted)0.5 (10.0 mmol g -1 ). 19 At 298 K and 1.0 bar, the volumetric storage density of SO2 in MFM-170 is 307 times that of gaseous SO2 under the same conditions, or 75 times of that of compressed SO2 (P0 = 3.9 bar) in a pressure vessel (packing efficiency and system volume are not taken into consideration). Furthermore, MFM-170 shows high SO2 adsorption at elevated temperatures (11.6 mmol g -1 at 333 K and 1 bar). Uptake of SO2 in MFM-170 shows a reversible type I isotherm with high uptakes at low pressure (Fig. 3); at 273 K the uptake at 0.03 bar is 6.5 mmol g -1 .
Despite the high uptake at low pressure, the excellent reversibility of the SO2 isotherms at 273-333 K indicates that MFM-170 can be fully regenerated under pressure-swing conditions. More significantly, no loss of adsorption capacity of SO2 was detected in MFM-170 after 50 adsorption-desorption cycles at 298 K, and PXRD analysis of MFM-170 after these 50 cycles confirms the full retention of crystal structure, reflecting the exceptional chemical and thermal stability of this material (Fig. 4). Clearly, the site Cu(II)-bound SO2(2) was absent in MFM-170•H2O•3.27SO2, and as SO2 (2) is a primary site of interaction for SO2(5), the latter was not located either. However, overall the structural analysis shows that saturation of the copper sites in MFM-170 with H2O does not greatly reduce the SO2 binding capacity, consistent with the retention of high uptake capacity in MFM-170•H2O.

In situ spectroscopic analysis of host-guest binding dynamics.
In situ FTIR spectroscopic studies were conducted for MFM-170 as a function of SO2 loading (Fig. 6). The growth of a new peak at 1143 cm -1 was assigned to the ν1 symmetric stretch of adsorbed SO2, which increases as a function of SO2 partial pressure (pp). This symmetric band is red-shifted from 1152 cm -1 (Δ = -9 cm -1 ) for free SO2, confirming its interaction with the framework. A second new band, assigned to the ν3 asymmetric stretch of adsorbed SO2, grows and red-shifts from 1340 cm -1 at 0.01 ppSO2 to 1320 cm -1 at 0.10 ppSO2. These bands show larger shifts compared to gas phase SO2 (Δ= -41 cm -1 at 0.10 ppSO2), but are consistent with physisorption of SO2. 37,38 Significant vibrational changes of the framework were also observed on SO2 adsorption. The SO2 from CO2 in the presence of a large quantity of water was confirmed by repeating the breakthrough experiments with a water-saturated fixed-bed. The column was exposed to a stream of 3% H2O in He until breakthrough and saturation of water was observed. The subsequent breakthrough experiment demonstrated excellent SO2/CO2 separation under these conditions (Fig. 3d). Interestingly, whilst the breakthrough times were slightly decreased for both components than in the above experiments, CO2 is affected more severely with a much steeper breakthrough. Unlike the dry sample, a significant roll-up effect is observed for CO2 under humid conditions, indicating a large displacement of weakly bound CO2 by SO2, likely due to the formation of H2SO3 complexes in the pore. This suggests that the SO2/CO2 separation in MFM-170 could be enhanced under humid conditions. It has been suggested that 313-333 K represents a temperature range that is suitable for purifying flue gas streams from coal-fired powerplants. 2,3 Therefore, breakthrough experiments were also attempted for an activated packed bed of MFM-170 at elevated temperatures of 323 and 348 K (Fig. S25). Importantly, a very clear separation between CO2 and SO2 is evident at both temperatures, though, as expected, with reduced retention time.

Conclusions
The development of efficient strategies to fully mitigate emissions of SO2 from combustion and to achieve efficient SO2 storage and safe transport remains a fundamental challenge for many industries, power-plants and marine transport sectors. Although emerging MOF materials show great promise as sorbents for a wide range of inert gases, relatively little success has been achieved on the adsorptive removal of SO2, primarily due to the generally limited reversibility and/or stability of MOFs upon contact with highly corrosive SO2. The present work describes a high SO2 uptake of 17. A more detailed description of single crystal X-ray diffraction data can be found in the supplementary information.
Gas adsorption isotherms and breakthrough experiment: Measurements of SO2 adsorption isotherms (0-1 bar) were performed using a Xemis gravimetric adsorption apparatus (Hiden Isochema, Warrington, UK) equipped with a clean ultrahigh vacuum system. The pressure in the system is accurately regulated by mass flow controllers. Research grade SO2 and He were purchased from AIRLIQUIDE or BOC and used as received. In a typical gas adsorption experiment, 70-100 mg of MFM-170•H2O·solv was loaded into the Xemis, and degassed at 423 K and high dynamic vacuum (10 -10 bar) for 1 day to give desolvated MFM-170.
Breakthrough experiments were carried out in a 7 mm diameter fixed-bed tube of 120 mm length packed with 1.5 g of MFM-170 powder (particle size < 5 microns). The total volume of the bed was ca. 5 cm 3 . The sample was heated at 423 K under a flow of He for 2 days for complete activation. The fixed-bed was then cooled to room temperature (298 K) using a temperature programmed water bath and the breakthrough experiment was performed with streams of SO2 (0.5% diluted in He) and CO2 at atmospheric pressure and room temperature. The flow rate of the entering gas mixture was maintained at 47 mL min -1 , and the gas concentration, C, of SO2 and CO2 at the outlet determined by mass spectrometry and compared with the corresponding inlet concentration C0, where C/C0 = 1 indicates complete breakthrough. A more detailed description is given in SI.
Supplementary Information is available in the online version of the paper.