Uncovering the Structural Diversity of Y(III) Naphthalene-2,6-Dicarboxylate MOFs Through Coordination Modulation

Metal-organic frameworks (MOFs)—network structures built from metal ions or clusters and connecting organic ligands—are typically synthesized by solvothermal self-assembly. For transition metal based MOFs, structural predictability is facilitated by control over coordination geometries and linker connectivity under the principles of isoreticular synthesis. For rare earth (RE) MOFs, coordination behavior is dominated by steric and electronic factors, leading to unpredictable structures, and poor control over self-assembly. Herein we show that coordination modulation—the addition of competing ligands into MOF syntheses—offers programmable access to six different Y(III) MOFs all connected by the same naphthalene-2,6-dicarboxylate ligand, despite controlled synthesis of multiple phases from the same metal-ligand combination often being challenging for rare earth MOFs. Four of the materials are isolable in bulk phase purity, three are amenable to rapid microwave synthesis, and the fluorescence sensing ability of one example toward metal cations is reported. The results show that a huge variety of structurally versatile MOFs can potentially be prepared from simple systems, and that coordination modulation is a powerful tool for systematic control of phase behavior in rare earth MOFs.


[Y 2 (NDC) 3 (C 3 H 7 NO) 2 ] n (1)
Yttrium chloride hexahydrate (0.225 mmol, 0.0439 g) and naphthalene-2,6-dicarboxylic acid (0.225 mmol, 0.0486 g) were added to a 25 mL jar followed by DMF (6 mL). The solution was sonicated until material fully dispersed before placing in the oven at 120 °C for 24 hr. After cooling, the resultant crystals were rinsed several times with fresh DMF and acetone, followed by vacuum drying. Replacing yttrium chloride hexahydrate with yttrium nitrate hexahydrate resulted in the same product, suggesting that counterions do not play a significant structure directing role under these conditions.
The same material could also be synthesised rapidly via microwave assisted heating, using the above quantities and heating for 1 hr at 120 °C in a 35 mL glass vessel. The resultant material underwent the same washing process as for the solvothermal synthesis ( Figure S1). Figure S1. Comparison of the calculated and experimental PXRD patterns of 1, confirming phase purity via both solvothermal and microwave synthesis, and the ability to produce 1 using yttrium nitrate hexahydrate.
The material was could also be rapidly synthesised via microwave assisted heating, increasing the quantities by a factor of four, and heating to 120 °C for 2 hr in a 35 mL glass vessel. The resultant material underwent the same washing process as for the solvothermal synthesis. These quantities were also applied for bulk scale solvothermal synthesis ( Figure S4). (2) °, V = 893.59 (5) Å 3 , T = 100 K, space group P-1, Z = 2, 16175 measured reflections, 3231 unique (R int = 0.044), which were used in all calculations. The final R 1 = 0.054 for 3097 observed data R[F 2 > 2σ(F 2 )] and wR(F 2 ) = 0.178. Crystal structure data are available from the CCDC, deposition number 1878171.
The material could also be synthesised rapidly via microwave assisted heating, increasing the quantities by a factor of four and heating to 120 °C for 2 hr in a 35 mL glass vessel. The resultant material underwent the same washing process as for the solvothermal synthesis. These quantities were also applied for bulk scale solvothermal synthesis ( Figure S5). Approximately 20% of the cell volume is not occupied by the framework and contains diffuse and disordered solvent molecules. This electron density was accounted for using SQUEEZE within PLATON (Spek, 2015) which calculated two solvent accessible voids of 188 Å 3 and 225 Å 3 containing in total 116 electrons (the equivalent of ~1.25 molecules of DMF) per unit cell. Crystal structure data are available from the CCDC, deposition number 1878172.
To prepare the sample for 1 H NMR spectroscopy ( Figure S6), a small quantity of 5 was added to DMSO-d 6 with five drops of D 2 SO 4 , followed by heating until all material was solubilised. The integral ratios of characteristic resonances for the aromatic protons of the NDC ligand were compared to those of the in situ generated formate and dimethylammonium species. The integral ratio for NDC:dimethylammonium:formate was found to be 3:1:2.6, which correlates closely to the crystallographically derived ratio of 3:1:3. Figure S6. 1 H NMR spectrum of acid digested (D 2 SO 4 / DMSO-d 6 ) 5 with peak assignments.

TGA analysis
Thermogravimetric analysis (TGA) was performed on all MOF samples to determine thermal stability. The profiles were collected by heating from room temperature to 800 °C at a rate of 10 °C/min under an air atmosphere ( Figures S8 and S9). Figure S8. Comparison of TGA profiles of 1, 3, 4 and 5, with assigned weight loss. Figure S9. Comparison of the TGA profiles of 2 and 6 to 1, showing similarities due to the bulk phase being predominantly 1. Fluorescence Analysis The MOF sample of 5 used in the fluorescence measurements was prepared via the microwave synthesis method, followed by extensive grinding of the crystalline material to produce a fine powder. The finely ground material (2 mg) was suspended in 5 ml of 5 mM DMF/metal nitrate solutions and sonicated for 30 minutes to ensure suspension before running fluorescence measurements (λ ex = 287 nm, Figure S10). The same quantity of 5 was used in the study of Cu 2+ quenching, suspended in 5 ml solutions of the appropriate concentration. Three days following the measurements, PXRD of the samples was carried out to assess the crystallinity of the material ( Figure S11). Figure S11. Stacked PXRD patterns of 5 following cation exchange, with all samples showing retention of crystallinity, apart from exposure to Fe 3+ , which causes the decomposition of 5.