Harnessing a Ti-based MOF for selective adsorption and visible-light-driven water remediation

In pursuit of universal access to clean water, photocatalytic water remediation using metal–organic frameworks (MOFs) emerges as a strong alternative to the current wastewater treatment methods. In this study, we explore a unique Ti-based MOF comprised of 2D secondary-building units (SBUs) connected via biphenyl dicarboxylic acid (H2bpdc) ligands – denoted as COK-47 – as a visible-light-driven photocatalyst for organic dye degradation. Synthesized via a recently developed microwave-assisted method, COK-47 exhibits high hydrolytic stability, demonstrates a strong dye uptake, and shows noteworthy dye-degradation performance under UV, visible, and solar light, outperforming benchmark TiO2 and MIL-125-Ti photocatalysts. Due to its nanocrystalline structure and surface termination with organic linkers, COK-47 exhibits selective degradation of cationic pollutants while remaining inert towards anionic dyes, thus highlighting its potential for selective oxidation reactions. Mechanistic studies reveal the involvement of superoxide radicals in the degradation process and emphasize the need to minimize the recombination of photogenerated electron–hole pairs to achieve optimal performance. Post-catalytic studies further confirm the high stability and reusability of COK-47, making it a promising photocatalyst for water purification, organic transformation, and water splitting reactions under visible light.


Supplementary note 1 | TGA Analysis
First, we calculate the molecular weight of COK-47 assuming the formula Ti2O3(C14H8O4), taken from literature. 1 This calculation gives a value of 383.94 g mol -1 .Since we know the final product of COK-47 oxidative decomposition to be TiO2 (evident from Figure S2), we can calculate the theoretical weight loss, assuming a perfect COK-47 structure.Based on Equation 1, we estimate the total weight loss of the reaction to be ~58 wt% after the full conversion into TiO2 occurs.TGA curves obtained for pristine COK-47 (Figure 1c of the main text) show an overall weight loss of ~60% (in air)a value that is higher than 58 wt% expected for the stoichiometric structure.This implies the presence of excess ligands, which we believe could be present at the surface of COK-47 nanocrystals as surface passivation (as illustrated in Figure 1b of the main text), thus effectively accounting for the higher than expected ligand:Ti ratio.
Surface Analysis

Supplementary note 2 | Adsorption studies
Based on relevant literature, we assume the (interaction/adsorption) surface area (i.e.footprint) of a single methylene blue molecule to be ~ 17 x 7.6 Å 2 . 2 Based on the experimentally reported surface area of COK-47 (~285 m 2 g -1 ) and assuming a monolayer adsorption model, one can estimate a monolayer MB coverage to correspond to ~2.2 x 10 20 dye molecules per gram of MOF.This number of MB molecules can be translated to ~366 µmol or 117 mg of MB per gram of COK-47, corresponding to ~12 wt% of its weight.
Experimental adsorption tests were carried out using 10 ml of 25 ppm of MB solution and 2 mg of COK-47.The powder was homogeneously suspended in the MB solution via ultrasonication for 15 min followed by stirring the suspension for 2 h under dark.The suspension was then centrifuged at 5000 rpm for 15 min, followed by measuring the UV-Vis spectrum of the supernatant solution to quantify the remaining dye in the solution.We observed a strong decrease from 25 ppm to ~5.3 ppm, corresponding to an uptake of ~0.2 mg MB, thereby revealing an adsorption capacity of 100 mg g -1 or ~ 10 wt% of the COK-47 mass, which is in good agreement with the theoretical maximum estimated above.

Figure S1 |
Figure S1 | (a) Structural representations of (a) methylene blue (b) crystal violet (c) toluidine blue and (d) methyl orange (b) FTIR spectra of pristine COK-47 showing the characteristic COO -peaks indicated with the 3 dotted lines (c) tauc-function plot (assuming a direct band-gap) of pristine COK-47 powder.

Figure
Figure S2 | (a) XRD pattern of COK-47 after TGA measurement (heated until 700°C) in synthetic air atmosphere with blue lines corresponding to the expected TiO2 pattern (04-022-3337) (b) methylene blue adsorption studies using two adsorbent (COK-47) dosages (c) Dye adsorption curves when using COK-47 in a mixture of methylene blue and methyl orange solutions (6 ppm) after equilibration (120 min) (d) PZC curves indicating the isoelectric point for pristine COK-47.

Figure S3 |
Figure S3 | (a) UV-Vis spectrum of methyl orange on COK-47 under visible light illumination over a period of 2 h (b) Concentration profiles of methylene blue over a period of 2 h under UV and visible light illumination without the catalyst, and with (Anatase) TiO2 under visible light illumination (c) transient photocurrent response of COK-47, COK-47-bpy, and MIL-125-Ti under visible light and (d) UV illumination; (e) overall MB removal efficiency of COK-47, P25 (commercial TiO2) and TiO2 (Anatase) under UV (280-400 nm) illumination after 2 h and (f) concentration profiles of methylene blue with and without COK-47 in the presence of natural sunlight, and artificially-derived visible light (400-700 nm, Lumatec) over the period of 60 min.

Table S1 |
Overview of the adsorption capacity (Qe) values for a variety of MOFs reported previously towards dye adsorption in comparison to current work.