Linker-Functionalized Phosphinate Metal–Organic Frameworks: Adsorbents for the Removal of Emerging Pollutants

Metal–organic frameworks (MOFs) are attracting increasing attention as adsorbents of contaminants of emerging concern that are difficult to remove by conventional processes. This paper examines how functional groups covering the pore walls of phosphinate-based MOFs affect the adsorption of specific pharmaceutical pollutants (diclofenac, cephalexin, and sulfamethoxazole) and their hydrolytic stability. New structures, isoreticular to the phosphinate MOF ICR-7, are presented. The phenyl ring facing the pore wall of the presented MOFs is modified with dimethylamino groups (ICR-8) and ethyl carboxylate groups (ICR-14). These functionalized MOFs were obtained from two newly synthesized phosphinate linkers containing the respective functional groups. The presence of additional functional groups resulted in higher affinity toward the tested pollutants compared to ICR-7 or activated carbon. However, this modification also comes with a reduced adsorption capacity. Importantly, the introduction of the functional groups enhanced the hydrolytic stability of the MOFs.


Elemental analysis
Infrared spectra Thermogravimetric analysis

Preparation of methyl 4-(N,N-dimethylamino)phenylphosphinate (3).
A three-necked flask was three times evacuated and flushed with Ar and charged with 5 mL of freshly distilled dimethylaniline (39 mmol).After cooling with an ice/potassium oxalate bath, 10 mL of phosphorus trichloride (114 mmol) and 9 mL of pyridine (112 mmol) were added.The reaction mixture was heated at 120 °C for 4 h.The excess of phosphorus trichloride was distilled off and the residue was dissolved in 10 mL of CH2Cl2.The solution was cooled by an ice bath and 22 mL of MeOH (0.54 mol) was slowly added.The reaction mixture was stirred for 1 h at RT.The solvent was removed by rotary evaporation and the solid residue was dissolved in CH2Cl2 and washed with water 3 times.The organic fraction was dried over anhydrous MgSO4, filtered and evaporated to dryness.The crude product was purified by column chromatography on SiO2 using 95:5 dichloromethane/methanol mixture as an eluent.

Preparation of dimethyl biphenyl-4,4ˈ-bis(4-(N,N-dimethylamino)phenylphosphinate) (4).
A Schlenk tube was charged with 1.86 g of 4,4ˈ-dibromobiphenyl (6.0 mmol), 3.57 g of methyl 4-(N,N-dimethylamino)phenylphosphinate (18 mmol) and 1.1 g of Pd(PPh3)4 (0.95 mmol), three times evacuated and flushed with Ar.Then 60 mL of dry 1,4-dioxane was added followed by addition of 2.0 mL of dry triethylamine (14 mmol).The reaction mixture was stirred at 60 °C for 96 hours.After cooling to room temperature, the formed precipitate was removed by filtration and the filtrate was evaporated to dryness.The solid residue was dissolved in dichloromethane and washed with water 3 times.The organic fraction was dried over anhydrous MgSO4, filtered and evaporated to dryness.The product was isolated by column chromatography on SiO2 using 95:5 dichloromethane/methanol mixture as an eluent.

Preparation of dimethyl biphenyl-4,4ˈ-diphosphinate (1).
A two-necked round-bottom flask was charged with 3.31 g of dibromobiphenyl (10.6 mmol), three times evacuated and flushed with Ar, and dissolved in 80 mL of THF.The solution was cooled by a CO2(s)/EtOH cooling bath and 22.5 mL of 1.9M solution of t-BuLi (42.8 mmol) in pentane was slowly added.The solution was stirred for 30 min upon cooling by CO2(s)/EtOH bath and then it was replaced for an ice bath and stirred for additional 90 min.After that, a solution of 8.95 g of bis(diethylamino)chlorophosphine (42.5 mmol) in 20 mL of THF was added.After 1 h of stirring upon cooling, the cooling bath was removed and the reaction mixture was stirred for additional 16 h at RT.The reaction mixture was cooled by a CO2(s)/EtOH cooling bath and 85 mL of 1M solution of HCl (85 mmol) in diethyl ether was added.The solution was stirred 1 h upon) cooling and then 20 h at RT.The formed slurry was separated by canula filtration and the filtrate was evaporated in vacuo.The evaporated solid was dissolved in benzene.In a Schlenk tube, 8.4 mL of MeOH (208 mmol) was mixed with 11.9 mL of dry triethylamine (85 mmol) under Ar atmosphere.The mixture was cooled in an ice bath and the benzene solution from the flask was added slowly.The formed precipitate was filtered off and the filtrate was evaporated.The oily product was mixed with water and stirred for 1 h at RT.After that, the product was extracted by dichloromethane, dried over Na2SO4, filtered and rotary evaporated.Yield: 2.40 g (73 %). 1 H NMR (CDCl3): δ 7.89 (dd, 3 JPH = 13.4Hz, 3 JHH = 8.2 Hz, 4H); 7.76 (dd, 3 JHH = 8.2 Hz, 4 JPH = 3.1 Hz, 4H); 7.62 (d, 1 JPH= 570 Hz, 2H); 3.83 (d, 3 JPH= 12.0 Hz, 6H). 31 P{ 1 H} NMR (CDCl3): δ 26.9.

Preparation of dimethyl biphenyl-4,4ˈ-bis(4-methoxycabonylphenylphosphinate) (2).
A Schlenk tube was charged with 5.00 g of methyl 4-bromobenzoate (23 mmol) and 2.69 g of Pd(PPh3)4 (2.3 mmol), three times evacuated and flushed with Ar.Then 12 mL of dry 1,4dioxane was added followed by addition of 3.9 mL of dry triethylamine (28 mmol) and 2.4 g of dimethyl biphenyl-4,4ˈ-diphosphinate (7.7 mmol) dissolved in 20 mL of dry dioxane.The reaction mixture was stirred at 60 °C for 96 hours.After cooling to room temperature, the formed precipitate was removed by filtration and the filtrate was evaporated to dryness.The product was separated by column chromatography on SiO2 using 95:5 dichloromethane/methanol mixture as an eluent and purified by dissolving in dichloromethane and precipitation with diethyl ether.

Preparation of biphenyl-4,4ˈ-bis(4-carboxyphenylphosphinic acid) (H2BBP(Ph-COOH)
A round-bottom flask was charged with 2.00 g of dimethyl biphenyl-4,4ˈ-bis(4-(N,Ndimethylamino)phenylphosphinate) (3.6 mmol), evacuated and flushed with argon three times, and then 200 mL of CH2Cl2 was added followed by a dropwise addition of 1.1 mL of trimethylsilyl bromide (8.3 mmol).The reaction mixture was stirred at 40 °C for 16 h.After that, the solvent was removed by rotary evaporation.The solid residue was suspended in water, filtered off and washed with a small amount of acetone and diethyl ether.

Instrumental methods
1 H, 31 P and 13 C NMR spectra were measured on a JEOL 600 MHz NMR spectrometer.The chemical shifts were referenced to the residual 1 H and 13 C signal of the deuterated solvents.Powder X-ray diffraction (XRD) was measured using a PANalytical X'Pert PRO diffractometer in the reflexion setup equipped with a conventional Co X-ray tube (40 kV, 30 mA).Qualitative analysis was performed with the HighScorePlus software package (PANalytical, Almelo, The Netherlands, version 3.0).Thermal analyses (DTA/TGA) were carried out on a Setaram SETSYS Evolution-16-MS (Setaram, Caluire, France) instrument coupled with a mass spectrometer.The measurements were performed in synthetic air (30 mL min −1 ) from 20 to 750 °C with a heating rate of 10 °C min −1 .Fourier transform infrared (FTIR) spectra were collected with a Nicolet NEXUS 670-FT spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) with an ATR accessory.The content of C, H and N was determined by a standard combustion technique (Thermo Scientific FlashSmartTM 2000 Elemental analyzer).The content of P and Fe was measured by ICP-MS (Agilent 7900 equipped with an Ar burner, ORS 4 collision cell and orthogonal hyperbolic quadrupole mass analyser), 20 ppb indium solution was used as an internal standard.Prior the measurement, the samples were dissolved in the mixture of acids (12 mL of HCl, 4 mL of HNO3, and 4 mL HF for 10 mg of sample) under microwave irradiation.
The removal of pollutants and linker release were quantified using a high-performance liquid chromatography (HPLC) Agilent 1260 Infinity II instrument with a diode array detector (DAD) and ASTRA C18-HE (partial size 3 µm, 50 x 3 mm) chromatographic column.In each case 10 µL of the sample was injected using an autosampler.The analyses were carried out at a constant temperature of 30 °C.A mixture of 0.02M phosphate buffer (PB) solution with pH 2.5 and acetonitrile (MeCN) was used as a mobile phase.The flow rate was 0.5 mL min -1 under isocratic elution for all analytes.The composition of the mobile phase for particular analytes, wavelengths used for the detection, time of analysis, limits of detection (LOD) and limits of quantification (LOQ) are specified in Table S1.The table also shows the parameters for H2BBP(Ph) and H2BBP(Ph-NMe2), which were detected to see how much of the linker was released from the materials.The amount of H2BBP(Ph-COOEt) released from ICR-14 was determined by ICP-MS to avoid the effect of potential linker hydrolysis providing a product with a different retention on the HPLC column.The LOQ of ICP-MS was 0.005 mg L -1 of phosphorus.

Mathematical models for adsorption experiments
Collected data from adsorption experiments were evaluated using the follows equations: Adsorbed amount of the pharmaceutics, qe (mg g -1 ), was calculated using ( 1): where C0 (mg dm -3 ) is initial concentration of the pollutant, Ce (mg dm -3 ) represents equilibrium concentration of the pollutant obtained by HPLC, and m (g) and V (dm 3 ) stand for weight of the adsorbent and total volume of the pollutant solution, respectively.
The kinetic data of the adsorption were fitted with a pseudo-second order kinetic model with (2): (2) where qe (mg g -1 ) is the concentration of the pollutant adsorbed in the defined time, Qmax (mg g -1 ) is the concentration of the pollutant adsorbed in the equilibrium point.Pseudosecond order rate constant and time are in (2) marked as k2 (g mg -1 min -1 ) and t (min), respectively.[S1] Adsorption isotherms were described using Langmuir (3) and Freundlich (4) models: where qe is the adsorbed amount of the pollutant at defined concentration calculated by (1), Qmax is the Langmuir maximum sorption capacity (mg g -1 ) representing the saturation plateau; Ce represents the equilibrium concentration measured by HPLC (mg dm -3 ); and KL stands for the Langmuir constant (dm 3 mg -1 ) quantifying the affinity of the adsorbate to the adsorbent.[S2] where KF ([(mg g -1 ) (mg dm -3 ) -n ] represents the Freundlich constant and n is the parameter of surface heterogeneity, which is related to the sorption capacity.[S2] Since the final isotherm is S-shaped in the case of sulfamethoxazole adsorption by ICR-7, a modified Sips isotherm, which is a combination of Langmuir and Freundlich isotherm, was used for data fitting.Its mathematical expression is given by ( 5): where KS and n stand for Sips constant parameters and their meaning is derived from Langmuir and Freundlich model.[S3]               Table S3: Specific surface areas of the as-prepared ICR MOFs and the materials after adsorption of pollutants; BET specific surface areas in m 2 g -1 are given.

Figure S2 :
Figure S2: The side view of the 1D chains composed of octahedrally coordinated iron atoms (blue polyhedron) and tetrahedrons of phosphorus atoms surrounded by two oxygen and two carbon atoms in the structure of ICR-8 (a) and ICR-14 (b).Iron octahedra are light blue, phosphinate tetrahedra are magenta; elements are colour-coded as follows: Fe (light blue), P (magenta), O (red), C (black), N (blue) and H (white).

Figure S3 :
Figure S3: Rietveld fit of ICR-8.Measured powder XRD pattern (black), theoretical pattern calculated from the manually created crystal structure (red), Bragg's positions (green) difference between measured and calculated profile (blue) are depicted.

Figure S4 :
Figure S4: Rietveld fit of ICR-14.Measured powder XRD pattern (black), theoretical pattern calculated from the manually created crystal structure (red), Bragg's positions (green) difference between measured and calculated profile (blue) are depicted.

Figure S13 :
FigureS13: Comparison of the removal efficiency of the tested adsorbents for particular pharmaceutical pollutants; the percentage of the initial amount of the pollutant which was captured after 24 h from a 100 mg L -1 solution by 5 mg of an adsorbent is depicted.

Table S1 :
Parameters of the HPLC determination of particular analytes.

Table S2 :
Elemental composition of the ICR MOFs determined by CHN analysis and ICP-MS.The values in brackets show the theoretical composition calculated based on the empirical formula C72H54Fe2O12P6 for ICR-7, C84H84Fe2O12N6P6 for ICR-8 and C78H78Fe2O24P6 for ICR-14, respectively.