English

• 0.   Introduction

  Antibiotics remain a cornerstone in treating infectious diseases. Among them, nitrofurans, such as nitrofurazone (NZF) and nitrofurantoin (NFT), are widely used to control microbial infections in living organisms^[1-3]. However, the extensive use and subsequent discharge of these compounds have led to widespread environmental accumulation. The presence of residual antibiotics in aquatic ecosystems disrupts ecological balance and accelerates the spread of antimicrobial resistance genes, posing long-term risks to human health^[4]. Therefore, developing accurate, rapid, and sensitive analytical platforms for the trace detection of these pharmaceutical pollutants is essential for both environmental remediation and public health.

  Luminescent metal-organic frameworks (LMOFs) are highly promising platforms for environmental monitoring^[5-9]. Benefiting from structural tunability, high porosity, and distinct optical properties, LMOFs can translate specific host-guest recognition events into measurable photoluminescent (PL) signals. This sensing paradigm offers high sensitivity, rapid response, and operational simplicity^[10-11]. To construct high-performance LMOF sensors, the rational design of organic linkers is critical. One effective strategy is to incorporate fluorogenic backbones with uncoordinated heteroatoms (e.g., O, N, S), which provide intrinsic active sites for host-guest interactions^[12-14].

  Conventional bipyridine linkers with simple arene cores, such as benzene, are frequently used in MOF construction. However, these linkers often suffer from high torsional freedom between aromatic rings. This structural flexibility promotes non-radiative decay through molecular twisting and thermal dissipation, limiting the intrinsic luminescence quantum yield and packing efficiency. To address this issue, fused-ring heteroaromatics can be employed to enforce planar rigidity. For instance, the thiazolo[5,4-d]thiazole (TTZ) building block, comprising two tightly fused thiazole rings, offers a rigid, π-coplanar, and electron-rich architecture^[15]. Incorporating the TTZ core into bipyridine ligands restricts intramolecular motion and promotes ordered, TTZ core-to-core π…π stacking interactions within the framework^[16]. As systematically summarized in our recent review of TTZ-based functional MOFs^[15], this core-to-core locking mechanism not only rigidifies the network but also significantly modulates the optoelectronic properties, providing a solid structural basis for developing advanced luminescent materials.

  Our previous work demonstrated the utility of TTZ-based MOFs by developing a zinc-based LMOF, [Zn₂(OBA)₂(4bpyttz)], for the detection of Co²⁺ and nitrofuran antibiotics^[17], alongside an extended analogue for sensing Al³⁺ and oxoanions^[18]. More recently, we constructed a dual Zn(Ⅱ)/Cd(Ⅱ) MOF system utilizing an isomeric extended TTZ linker for the multi-analyte fluorescence sensing of antibiotics and Cr(Ⅵ)^[19]. While these studies confirmed the sensing potential of TTZ-based materials, a systematic, side-by-side comparison elucidating how core substitution (e.g., benzene versus TTZ) dictates framework rigidity, electronic coupling, and subsequent fluorogenic sensing performance has not yet been thoroughly explored.

  To verify the impact of ligand rigidification on the framework′s photophysics, we synthesized two 2D Cd(Ⅱ)-based metal-organic frameworks, {[Cd(4bpyttz)(CDA)]·2DMF}_n (1) and [Cd(4bbpy)(CDA)]_n (2) (4bpyttz=2,5-bis(pyridin-4-yl)thiazolo[5,4-d]thiazole, 4bbpy=1,4-bis(pyridin-4-yl)benzene, H₂CDA=4,4′-carbonyldibenzoic acid), to systematically investigate the structure-property relationships governed by distinct bipyridine-based linkers. Single-crystal X-ray diffraction reveals that both MOFs adopt a 2D network with an sql topology (Scheme 1). Notably, compared to the benzene-centered analog (MOF 2), the TTZ moiety in MOF 1 provides significant planar rigidity. This rigidification suppresses non-radiative decay, while the facilitated π…π stacking and hydrogen-bonding networks extend the π-conjugation, leading to a substantial enhancement in photoluminescence (PL) intensity. Moreover, the uncoordinated nitrogen and sulfur heteroatoms in the TTZ core of MOF 1 serve as Lewis basic sites for host-guest recognition, enabling hydrogen bonding and electrostatic interactions with the nitro and amide groups of antibiotics (Scheme 1). Consequently, MOF 1 operates as a robust fluorogenic sensor, showing a sensitive and selective turn-off response toward nitro-containing antibiotics (NZF and NFT). Mechanistic studies indicate that this quenching is primarily driven by the inner filter effect (IFE), supplemented by photoinduced electron transfer (PET) attributed to the electron-withdrawing nature of the nitro groups. Overall, this work not only highlights the crucial role of ligand coplanarity and heteroatom engineering in modulating MOF photophysics but also establishes a rational structural paradigm for designing targeted luminescent sensors.

  Scheme 1

  

  Scheme 1.  Modification of Cd(Ⅱ)-MOFs with a rigid thiazolothiazole core: enhancement of π…π interactions and significant boost in PL emission

  下载: 全尺寸图片 幻灯片

  1.   Experimental

  1.1   Materials and instrumentation

  All reagents and solvents were obtained from commercial sources and used without further purification. The organic linkers, 4bpyttz and 4bbpy, were synthesized according to the modified literature procedures^{[17, 20]}. Detailed characterization and instrumental specifications are provided in the Supporting information.

  1.2   Synthesis of MOF 1

  A mixture of H₂CDA (0.008 1 g, 0.03 mmol), 4bpyttz (0.008 9 g, 0.03 mmol), and Cd(NO₃)₂·4H₂O (0.009 3 g, 0.03 mmol) was dissolved in 5 mL of DMF. The reagents were sequentially added into a Teflon-lined stainless-steel autoclave and ultrasonicated for 5 min to ensure homogeneity. The autoclave was sealed and heated in an isothermal oven to 140 ℃ at a ramp rate of 5 ℃·min^-1, where it was maintained for 48 h. Subsequently, the system was cooled to 80 ℃ at a rate of 5-10 ℃·h^-1 for 12 h, then cooled to 30 ℃ at the same rate for 5 h, and finally allowed to cool naturally to ambient temperature. The resulting crystals were collected, washed with DMF 3-5 times to remove surface impurities, and dried under ambient conditions. Yield: 0.020 1 g (81% based on Cd). Elemental analysis calculated for C₃₅H₃₀CdN₆O₇S₂(%): C 51.07, H 3.67, N 10.21; Found(%): C 51.14, H 3.82, N 10.76. IR (KBr pellets, cm^-1): 3 412(m), 3 081(m), 3 035(m), 2 812(w), 2 727(w), 2 437(w), 2 165(w), 1 940(w), 1 916(w), 1 631(m), 1 591(s), 1 549(m), 1 482(s), 1 421(s), 1 398(m), 1 351(m), 1 292(m), 1 230(m), 1 140(w), 1 117(m), 1 088(w), 1 074(m), 1 043(m), 1 008(w), 993(m), 862(w), 832(m), 799(s), 730(m), 704(s), 663(w), 609(w), 575(w), 567(w), 543(w), 504(w), 485(w), 463(w), 417(w).

  1.3   Synthesis of MOF 2

  The synthesis of MOF 2 followed a similar procedure to that of MOF 1, except that 4bbpy (0.007 3 g, 0.03 mmol) was used instead of 4bpyttz. Specifically, CDA (0.008 1 g, 0.03 mmol), 4bbpy (0.007 3 g, 0.03 mmol), and Cd(NO₃)₂·4H₂O (0.009 3 g, 0.03 mmol) were mixed in 5 mL of DMF. After 5 min of ultrasonication, the mixture underwent the identical solvothermal program (heating, isothermal, and programmed cooling stages) and post-synthetic treatments as described for MOF 1. Yield: 0.014 4 g (68% based on Cd). Elemental analysis calculated for C₃₁H₂₀CdN₂O₅·5H₂O(%): C 52.97, H 4.30, N 3.99; Found(%): C 52.88, H 4.55, N 4.12. IR (KBr pellets, cm^-1): 3 435(m), 3 040(w), 2 812(w), 2 721(w), 2 155(w), 1 955(w), 1 631(m), 1 595(s), 1 504(w), 1 445(m), 1 404(m), 1 351(m), 1 330(m), 1 319(m), 1 237(w), 1 209(m), 1 122(w), 1 090(w), 1 065(w), 1 018(w), 990(w), 889(w), 829(m), 767(w), 694(m), 659(w), 635(w), 618(m), 507(w), 474(w).

  2.   Results and discussion

  The two 2D Cd(Ⅱ)-based frameworks, {[Cd(4bpyttz)(CDA)]·2DMF}_n (MOF 1) and [Cd(4bbpy)(CDA)]_n (MOF 2), were synthesized via a reliable solvothermal strategy. The reactions of Cd(NO₃)₂ with the V-shaped dicarboxylic acid H₂CDA and the respective bipyridine-based linkers (4bpyttz for MOF 1; 4bbpy for MOF 2) in DMF at 140 ℃ yielded phase-pure single crystals. However, acquiring high-quality single crystals of MOF 2 proved highly challenging. Despite exhaustive optimization of the synthetic parameters—including systematic variations in solvent composition (e.g., DMF/H₂O/MeCN ratios), concentrations, and ultra-slow programmed cooling rates—the harvested crystals of MOF 2 consistently exhibited microcrystalline dimensions and weak diffraction capabilities.

  2.1   Structural analysis of MOFs 1 and 2

  Single-crystal X-ray diffraction analysis reveals that both MOFs 1 and 2 crystallize in the triclinic crystal system with the P1 space group (Table 1). In MOF 1, the asymmetric unit consists of one Cd²⁺ ion, one deprotonated H₂CDA ligand (CDA^2-), one 4bpyttz ligand, and two uncoordinated lattice DMF molecules (Fig.S1). By contrast, the asymmetric unit of MOF 2 comprises one Cd²⁺ ion, one deprotonated H₂CDA ligand, and one 4bbpy ligand (Fig.S2). Due to severe disorder of the lattice solvent, the contribution of five lattice water molecules per formula unit was estimated using the PLATON/SQUEEZE procedure (Supporting information).

  Table 1

  

  Table 1.  Crystallographic data and structure refinement parameters for MOFs 1 and 2

  下载: 导出CSV

  Parameter	1	2
  Formula	C35H30CdN6O7S2	C31H20CdN2O5
  Formula weight	823.17	612.90
  Crystal system	Triclinic	Triclinic
  Space group	P1	P1
  a / nm	0.890 99(2)	0.934 55(18)
  b / nm	1.446 74(3)	1.363 1(3)
  c / nm	1.529 91(4)	1.542 7(2)
  α / (°)	111.274(2)	68.042(18)
  β / (°)	94.790(2)	84.214(16)
  γ / (°)	98.479(2)	85.800(19)
  V / nm3	1.797 43(8)	1.812 0(6)
  Z	2	2
  T / K	293(2)	293(2)
  Dc / (g·cm-3)	1.521	1.123
  Crystal size / mm	0.3×0.26×0.2	0.28×0.25×0.2
  μ(Mo Kα) / mm-1	0.780	0.653
  F(000)	836	616
  2θ range / (°)	4.674-59.27	4.384-49.996
  Reflection collected, unique	22 218, 8 049	15 281, 6 088
  Rint	0.026 5	0.148 2
  Tmax, Tmin	1.000, 0.762	1.000, 0.648
  Data, Nres, Npara	8 049, 54, 464	6 088, 21, 352
  R1, wR2 [I > 2σ(I)]b	0.036 7, 0.099 0	0.117 4, 0.291 6c
  R1, wR2 (all data)	0.044 3, 0.104 8	0.168 2, 0.341 3c
  GOF on F 2	1.042	1.078
  a Nres=number of restraints, Npar=number of parameters; b R1=∑||Fo|-|Fc||/∑|Fo|; wR2=[∑w(Fo2-Fc2)2/ ∑w(Fo2)2]1/2; c The relatively high R-indices and residual electron density for MOF 2 originate from the inherent framework lability and rapid partial loss of lattice water molecules, rather than poor data collection.

  The secondary building units (SBUs) of both frameworks are based on binuclear Cd(Ⅱ) clusters, though their local coordination environments differ significantly. In MOF 1 (Fig. 1a), the central Cd(Ⅱ) ion is seven-coordinated, bonding to two nitrogen atoms (N1 and N2^ⅲ) from two 4bpyttz ligands and five oxygen atoms (O1, O2, O4^ⅰ, O4^ⅱ, and O5^ⅰ) from three symmetry-related CDA^2- ligands, which adopt a μ₂-κ¹∶κ¹ bridging mode. To quantitatively assess this specific geometry, continuous shape measure (CShM) calculations were performed using Crystal Shape, a program developed in our laboratory and provided in the Supporting information^[21]. This software offers a streamlined and user-friendly platform by directly processing raw CIF files, significantly enhancing the efficiency of geometric analysis. The calculation yields a CShM value of 2.810 3, confirming a distorted pentagonal bipyramidal geometry for the Cd(Ⅱ) center (Table S1 and Fig. 1b). The Cd—O bond lengths range from 0.228 6(2) to 0.262 2(2) nm, and the Cd—N distances are 0.231 5(2) and 0.234 7(2) nm (Table S2).

  Figure 1

  

  Figure 1.  (a, d) SBUs of 1 and 2; (b, e) Coordination geometries of the Cd(Ⅱ) centers in 1 and 2; (c, f) 2D frameworks of 1 and 2, with the insets highlighting the π…π stacking interactions

  Symmetry codes: ^ⅰ x, y, z-1; ^ⅱ-x+2, -y+2, -z+2; ^ⅲ x+1, y+1, z+1 for 1; ^ⅳ x+1, y-1, z; ^ⅴ x, y, z+1; ^ⅵ -x+1, -y, -z+2 for 2.

  下载: 全尺寸图片 幻灯片

  MOF 2 instead features a six-coordinated Cd(Ⅱ) center with a distorted octahedral geometry (CShM value: 5.022 7, Table S3 and Fig. 1e), binding two nitrogen atoms (N1 and N2^ⅳ) from two 4bbpy ligands and four oxygen atoms (O1, O2^ⅵ, O4^ⅴ, and O5^ⅴ) from three symmetry-related CDA^2- ligands (Fig. 1d). This reduction in the coordination number stems directly from the ligand substitution. Unlike the tightly coordinated carboxylate oxygen in MOF 1, the analogous carboxylate oxygen atom (O2) in MOF 2 is displaced to a Cd…O separation of 0.313 6(10) nm, a distance too long to form a formal coordination bond. As a result, the effective Cd—O bonds in MOF 2 are comparatively shorter [0.223 7(8)-0.250 1(11) nm], while the Cd—N bonds are slightly elongated [0.234 4(9) and 0.236 8(10) nm; Table S4]. Both structures exhibit highly distorted polyhedral configurations, with O—Cd—O angles varying widely [51.92(7)°-169.75(7)° for MOF 1; 55.8(3)°-146.8(3)° for MOF 2], whereas the N—Cd—N alignments remain nearly linear [172.39(8)° and 175.2(3)°, respectively].

  Intermolecular non-covalent interactions further stabilize these frameworks. MOF 1 exhibits dense π…π stacking between the TTZ cores of adjacent 4bpyttz ligands [0.385 8(1) nm] and between adjacent pyridine rings [0.378 1(1) nm] (Fig. 1c). MOF 2 displays similar interactions between the 4bbpy ligands, with centroid-to-centroid distances of 0.412 0(1) nm for the central benzene rings and 0.402 7(1) nm for the pyridine rings (Fig. 1f). These π-electron-rich arrays reinforce structural rigidity and facilitate exciton migration, providing a crucial structural basis for fluorescence sensing.

  Topologically, the extended architectures of MOFs 1 and 2 are nearly identical despite their differing primary coordination spheres (Fig.S3 and S4). In both systems, each binuclear Cd₂ cluster acts as a 4-connected node. These nodes are linked by the bipyridine-based ligands, yielding adjacent inter-node distances of 1.529 9 and 1.736 0 nm for MOF 1, and 1.542 7 and 1.595 3 nm for MOF 2. Treating the organic ligands as linear linkers, both 2D supramolecular assemblies simplify to standard 4-connected uninodal networks mapping onto a classic sql topology with the Schläfli point symbol {4⁴·6²}^[22].

  2.2   Thermal stability, infrared spectroscopy and crystallinity

  Thermogravimetric analysis (TGA) was performed under an N₂ atmosphere to evaluate the thermal stability of MOFs 1 and 2. For MOF 1, a continuous weight loss of 16.0% was observed between 100 and 340 ℃, corresponding to the release of two uncoordinated DMF molecules (Calcd. 17.76%), with the slight deviation attributed to partial solvent volatilization during sample handling. The desolvated framework remained intact up to 350 ℃, followed by a rapid structural decomposition peaking at 423 ℃ (Fig.S5). For MOF 2, a weight loss of 12.9% occurred between 130 and 250 ℃, consistent with the theoretical removal of five lattice water molecules (Calcd. 12.81%). Its desolvated architecture was maintained up to 320 ℃, beyond which continuous thermal degradation occurs with a decomposition peak at 368 ℃ (Fig.S6). Notably, the higher decomposition temperature of 1 compared to 2 reinforces the role of the rigid TTZ core in enhancing the overall framework stability. These results confirm that both Cd(Ⅱ) complexes possess sufficient thermal stability for subsequent sensing applications in diverse environments.

  FTIR spectroscopy further validates the structural composition of the assembled frameworks. In both spectra, the absence of strong absorption peaks around 1 700 cm^-1 indicates the complete deprotonation of the carboxylic acid groups. The asymmetric (ν_as) and symmetric (ν_s) stretching vibrations of the carboxylate groups appeared at 1 591 and 1 421 cm^-1 for MOF 1, and at 1 595 and 1 404 cm^-1 for MOF 2. A broad band around 3 400 cm^-1 was observed in both MOFs and is attributed to the O—H stretching vibrations of adsorbed water molecules. In MOF 2, lattice water also contributes to this absorption. A distinguishing feature in the spectrum of MOF 1 is the sharp peak at 1 482 cm^-1, assigned to the characteristic C=N stretching of the thiazolo[5,4-d]thiazole ring in the 4bpyttz ligand, along with aromatic C—H (3 081 cm^-1) and C—S (799 cm^-1) absorptions that unambiguously confirm the integrity of the rigid heteroaromatic core. In contrast, MOF 2 lacked these signatures and instead exhibited distinct bending vibrations in the fingerprint region below 1 300 cm^-1, in line with the simpler benzene-based pillar (Fig.S7 and S8). Finally, powder X-ray diffraction (PXRD) confirms the bulk phase purity of both synthesized samples, as the experimental patterns match the simulations derived from single-crystal data (Fig.S9 and S10).

  2.3   Solid-state PL and mechanism analysis

  The solid-state PL properties of the free ligands and the assembled Cd(Ⅱ) frameworks (MOFs 1 and 2) were evaluated at room temperature. Upon excitation, the free ligands 4bpyttz, 4bbpy, and H₂CDA displayed maximum emission peaks at 435 nm (λ_ex=407 nm), 401 nm (λ_ex=356 nm), and 467 nm (λ_ex=310 nm), respectively (Fig. 2a). Following framework assembly, MOF 1 displayed a prominent emission band centered at 465 nm (λ_ex=397 nm). In contrast, the emission spectrum of MOF 2 was characterized by a primary peak at 458 nm alongside a weaker secondary peak at 486 nm (λ_ex=398 nm). Given the stable closed-shell d¹⁰ electronic configuration of the Cd(Ⅱ) ion, metal-centered d-d transitions are inherently precluded. Therefore, the fluorescence observed in these frameworks originates predominantly from ligand-based charge transfer processes^{[18, 20]}.

  Figure 2

  

  Figure 2.  (a) Solid-state PL spectra of the free ligands (H₂CDA, 4bpyttz, 4bbpy) and the corresponding Cd(Ⅱ) frameworks (MOFs 1 and 2) recorded at room temperature; (b) Calculated frontier molecular orbital energy levels for the primary ligand components in MOFs 1 and 2, illustrating the distinct charge transfer pathways and the suppression of non-radiative decay in MOF 1

  下载: 全尺寸图片 幻灯片

  To elucidate the origin of the significantly enhanced PL intensity in MOF 1 over MOF 2, density functional theory (DFT) calculations were performed to map the frontier molecular orbital energies of the ligands (Fig. 2b). In MOF 2, the HOMO and LUMO energy levels of 4bbpy and H₂CDA exhibit a staggered energy level alignment. The HOMO of 4bbpy (-6.508 eV) sits higher than that of H₂CDA (-7.068 eV), while its LUMO (-1.716 eV) is positioned above the LUMO of H₂CDA (-2.521 eV). This specific alignment drives an energetically favorable ligand-to-ligand charge transfer (LLCT) from the electron-donating 4bbpy to the electron-accepting H₂CDA. Consequently, this active LLCT pathway facilitates continuous non-radiative energy dissipation, which accounts for the relatively weak emission and the retention of ligand-like spectral features in MOF 2.

  In contrast, the incorporation of the electron-withdrawing TTZ core alters the electronic structure in MOF 1. The HOMO-LUMO gap of the 4bpyttz ligand narrows to 3.667 eV, with a HOMO of -6.335 eV and a stabilized LUMO of -2.668 eV. Consequently, the frontier orbitals of 4bpyttz encompass those of H₂CDA, forming a straddling energy level alignment. This configuration restricts the LLCT dissipation channel observed in MOF 2. Instead, the lowest-energy transition is localized within the 4bpyttz ligand, proceeding via an intraligand charge transfer (ILCT). Furthermore, the structural rigidity imparted by the coordination network and extensive π…π interactions in MOF 1 suppresses non-radiative decay associated with intramolecular rotations and vibrations. The combination of the blocked LLCT pathway and the restriction of intramolecular motion contributes to the strong fluorescence of MOF 1, providing a solid structural basis for its application as a fluorogenic sensor.

  2.4   Fluorescence sensing of antibiotics

  Strong intrinsic emission is a prerequisite for reliable luminescence sensing. As shown in Fig. 3a, MOF 1 exhibited a robust absolute quantum yield (QY) of 18.44%, whereas MOF 2 displayed negligible PL with a QY of only 1.33%. This dramatic contrast was also apparent in the emission spectra of the two frameworks dispersed in aqueous media (Fig. 3c) and in the corresponding intensity comparison (Fig. 3d). Under a 395 nm UV lamp (inset of Fig. 3d), MOF 1 exhibited intense blue-green fluorescence, while MOF 2 remained nearly dark, confirming that the high emission of MOF 1 is well-maintained in the practical sensing environment.

  Figure 3

  

  Figure 3.  (a, b) Absolute PL quantum yield (PLQY) spectra of solid-state crystals of MOFs 1 and 2 measured using an integrating sphere at room temperature; (c) PL emission spectra of MOFs 1 and 2 in water; (d) Comparison of the emission intensities of MOFs 1 and 2

  The inset shows digital photographs of MOFs 1 and 2 dispersed in aqueous suspensions under 395 nm UV light irradiation, illustrating the significant difference in emission intensity between the two frameworks.

  下载: 全尺寸图片 幻灯片

  Preliminary experiments confirmed that the extremely low baseline intensity of MOF 2 renders it unsuitable for reliable analytical applications. During antibiotic detection and titration assays, any analyte-induced spectral fluctuations were easily masked by instrumental background noise, resulting in prohibitive signal-to-noise ratios and poor reproducibility. Consequently, all subsequent sensing evaluations were conducted exclusively using the highly emissive MOF 1. Unless otherwise specified, all sensing experiments were performed at room temperature and atmospheric pressure using an aqueous suspension of MOF 1 (1 mg·mL^-1, ultrasonically dispersed in ultrapure water with an electrical conductivity of 18.2 MΩ·cm). A 2 mL aliquot of each analyte solution (10 mmol·L^-1 stock) was added to 2 mL of the MOF 1 suspension to achieve a final analyte concentration of 5 mmol·L^-1 for screening experiments. Fluorescence spectra were recorded at an excitation wavelength of 396 nm with emission monitored at 465 nm.

  Given the well-documented environmental hazards associated with antibiotic residues, we developed a MOF 1-based fluorometric platform for their detection. Ten representative antibiotics were selected as target analytes: chloramphenicol (CAP), metronidazole (MDZ), NZF, sulfadiazine (SDZ), tinidazole (TDZ), sulfamethoxazole (SMZ), NFT, ornidazole (ODZ), ofloxacin (OFLX), and penicillin (PCL). To assess the effect of each analyte on the fluorescence of MOF 1, individual antibiotics were added to MOF 1 suspensions to give a final concentration of 5 mmol·L^-1. As shown in Fig. 4, the nitrofuran-class antibiotics NZF and NFT induced pronounced fluorescence quenching. When 2 mL of 10 mmol·L^-1 NZF or NFT solution was added to the MOF 1 suspension, quenching efficiencies reached 99% and 80%, respectively, providing a clear spectral response suitable for mechanistic analysis. The apparent fluorescence shift observed with OFLX is attributed to its strong intrinsic fluorescence, which causes excessive absorption of excitation light and reabsorption of emitted photons, thereby distorting the recorded signal. These results demonstrate that MOF 1 exhibits high selectivity and sensitivity toward nitrofuran antibiotics, underscoring its promise as a fluorescent sensor for these contaminants^[23-25].

  Figure 4

  

  Figure 4.  (a) PL spectra of MOF 1 in the presence of various antibiotics (λ_ex=396 nm); (b) Relative PL intensities of MOF 1 dispersed in different antibiotic solutions, highlighting the highly selective and distinct fluorescence quenching triggered by the nitrofuran antibiotics NZF and NFT

  下载: 全尺寸图片 幻灯片

  2.5   Fluorescence titration of antibiotics

  To further evaluate the quantitative sensing performance of MOF 1 toward nitrofuran antibiotics, systematic fluorescence titration experiments were conducted by incrementally increasing the concentrations of NZF and NFT. As illustrated in the 3D PL spectra (Fig. 5a and 5c), the emission intensity of MOF 1 exhibited a progressive and massive decline upon the gradual addition of the analytes (titration volume up to 2 000 μL). The corresponding digital photographs (insets) under 395 nm UV irradiation provided a direct visual confirmation of the fluorescence quenching, with the suspension transitioning from a bright blue emission to near-extinction. The quenching efficiency was quantitatively analyzed using the Stern-Volmer (S-V) equation: I₀/I=1+K_svc_Q, where I₀ and I represent the fluorescence intensities of MOF 1 before and after the addition of the quencher Q, respectively, and K_sv is the S-V quenching constant.

  Figure 5

  

  Figure 5.  (a) Concentration-dependent 3D PL spectra of the MOF 1 aqueous suspension upon the incremental addition of NZF (titration volume: 0-2 000 μL) (inset: digital photographs under 395 nm UV light illustrating the visual fluorescence quenching); (b) Corresponding S-V plot for NZF (inset: linear fitting of the S-V plot in the low concentration region); (c) Concentration-dependent 3D PL spectra of the MOF 1 aqueous suspension upon the incremental addition of NFT (0-2 000 μL) (inset: digital photographs under 395 nm UV light); (d) Corresponding S-V plot for NFT (inset: linear fitting at the low concentration region)

  下载: 全尺寸图片 幻灯片

  As shown in Fig. 5b and 5d, the S-V plots for both NZF and NFT exhibited a strong linear relationship at lower concentration ranges (0-0.2 mmol·L^-1). Based on the linear fitting (insets), the K_sv values were calculated to be 1.3×10³ L·mol^-1 (R²=0.991) for NZF and 8.5×10² L·mol^-1 (R²=0.984) for NFT. At higher concentrations, the S-V curves significantly deviated from linearity and exhibited a distinct upward curvature. This exponential-like trajectory is typically indicative of a synergistic quenching mechanism involving the simultaneous occurrence of static and dynamic quenching, which is further reinforced by the dominant influence of the IFE as discussed in the absorption analysis^[26].

  The limit of detection (LOD) was determined based on the 3σ/k formula, where σ is the standard deviation of 20 blank measurements, and k is the slope of the linear fit at low concentrations. The calculated LOD values for NZF and NFT were 1.59 and 2.77 μmol·L^-1, respectively. Notably, a quenching efficiency of nearly 99% was achieved when the analyte concentration reached approximately 0.8 mmol·L^-1. These quantitative results demonstrate that MOF 1 exhibits high sensitivity and a low detection threshold toward nitrofuran antibiotics, providing a solid experimental basis for their rapid detection. Antibiotics lack a single regulatory limit, and their predicted no-effect concentrations (PNECs) vary widely by compound; for example, a risk assessment of 36 antibiotics in Chinese fresh surface waters reported PNECs spanning 0.001 75-2 351 μg·L^-1, with median measured concentrations of most antibiotics below 0.1 μg·L^-1[27]. The LODs of MOF 1 (1.59 μmol·L^-1, ca. 315 μg·L^-1 for NZF; 2.77 μmol·L^-1, ca. 660 μg·L^-1 for NFT) approach the upper end of the reported PNEC range, although they remain above the sub-μg·L^-1 levels typical of trace environmental monitoring. MOF 1 is therefore not intended to replace instrumental methods such as LC-MS/MS; rather, its operational simplicity, rapid response, and low cost make it a promising complementary tool for the on-site screening of relatively high-concentration contamination sources.

  2.6   Mechanism investigation of the fluorescence quenching

  To elucidate the underlying mechanism of the sensing of nitrofuran antibiotics (NZF and NFT) by MOF 1, a comprehensive study involving structural stability and spectral overlap analysis was performed. Typically, the fluorescence quenching process in luminescent MOFs can be attributed to framework collapse, PET, fluorescence resonance energy transfer (FRET), or IFE^{[19, 28-30]}.

  The structural integrity of MOF 1 after exposure to the analytes is a fundamental prerequisite for reliable sensing. PXRD and FTIR spectroscopy were employed to examine MOF 1 before and after immersion in NZF and NFT solutions for 6 h. The PXRD patterns of the recovered MOF 1 samples remained perfectly consistent with the pristine framework, displaying no significant peak shifts or the appearance of new crystalline phases (Fig.S11 and S12). Similarly, the FTIR spectra (Fig.S13 and S14) confirmed that the characteristic vibration bands of the organic ligands remained completely intact. These results unequivocally verify that the framework of MOF 1 possesses excellent chemical stability in the presence of nitrofuran antibiotics. Therefore, the observed fluorescence quenching is not caused by the collapse of the crystal structure, but rather by specific photophysical or photochemical interactions between the MOF and the analytes.

  Beyond framework stability, the high selectivity of MOF 1 for NZF and NFT arises from specific host-guest recognition. The TTZ core of the linker contains uncoordinated N and S atoms that serve as Lewis basic sites, enabling hydrogen bonding (N…H—N, S…H—N/O) and electrostatic interactions with the nitro and amide groups of the antibiotics. Furthermore, the confined pore environment of MOF 1 facilitates the enrichment of analyte molecules and brings them into proximity to these recognition sites, thereby enhancing binding affinity through pore confinement effects.

  To explore the photophysical basis of the quenching effect, the UV-Vis absorption spectra of various antibiotics were measured and compared with the excitation and emission profiles of MOF 1. As depicted in Fig. 6, the nitrofuran antibiotics, specifically NZF and NFT, exhibited broad and intense absorption bands extending from 250 to 450 nm. Notably, these continuous absorption profiles show a massive spectral overlap with the excitation spectrum of MOF 1 (centered at approximately 396 nm). This severe overlap clearly indicates that the NZF and NFT molecules can effectively compete with the MOF 1 framework for incident excitation photons (competitive absorption), which is a classic manifestation of the primary IFE^[26]. Furthermore, because the absorption spectra of NZF and NFT also partially overlap with the emission spectrum of MOF 1 (centered at 465 nm), the possibility of FRET cannot be entirely excluded. However, the dominant role of competitive absorption was further validated by subsequent UV-Vis titration experiments.

  Figure 6

  

  Figure 6.  Spectral overlap between the UV-Vis absorption spectra of various antibiotics (solid lines) and the excitation (pink shaded region) and emission (blue shaded region) spectra of MOF 1

  下载: 全尺寸图片 幻灯片

  The dynamic evolution of the absorption profiles during the sensing process was systematically tracked via UV-Vis titration. As illustrated in Fig. 7, upon the incremental addition of the antibiotic solutions (from 0 to 90 μL) to the MOF 1 aqueous suspension, the absorbance of the mixtures increased continuously at their respective characteristic wavelengths. Specifically for NZF (Fig. 7a), two distinct absorption bands centered at 260 and 375 nm progressively intensified across the entire titration process. Similarly, the continuous addition of NFT (Fig. 7b) led to a steady and massive enhancement of two primary absorption peaks located at 265 and 365 nm. The progressive enhancement of these broad absorption bands—especially the prominent peaks at 375 nm (for NZF) and 365 nm (for NFT), which exhibited tails extending to 450 nm and fully encompassed the excitation wavelength of MOF 1 (396 nm)—confirms that, as the nitrofuran concentration increases, an increasingly larger fraction of the incident excitation light is harvested by the antibiotic molecules rather than by the MOF framework. This severe competitive absorption drastically reduces the effective excitation energy reaching the MOF 1 luminophores, directly triggering the observed turn-off fluorescence response^[28]. Consequently, the fluorescence quenching of MOF 1 toward NZF and NFT results from the synergistic combination of selective host-guest recognition (via Lewis basic N/S sites and pore confinement) and the dominant IFE, with a contribution from PET due to the electron-withdrawing nitro groups. The selective binding and enrichment of analytes within the porous framework significantly enhances the efficiency of these photophysical processes.

  Figure 7

  

  Figure 7.  UV-Vis absorption spectra of the MOF 1 aqueous suspension upon the incremental addition (0-90 μL) of (a) NZF and (b) NFT

  下载: 全尺寸图片 幻灯片

  3.   Conclusions

  In summary, we have demonstrated a thiazolothiazole (TTZ) fluorophore functionalization strategy in 2D Cd(Ⅱ) MOFs that markedly enhances luminescence for aqueous antibiotic sensing. By comparing the functionalized {[Cd(4bpyttz)(CDA)]·2DMF}_n (MOF 1) with its benzene-centered analog [Cd(4bbpy)(CDA)]_n (MOF 2), combined experimental and theoretical studies establish that the rigid, fused TTZ moiety restricts intramolecular torsional freedom and suppresses non-radiative dissipation pathways (e.g., LLCT), thereby substantially improving PL. As a result, MOF 1 serves as a robust and sensitive platform for the selective detection of nitrofuran antibiotics in water, with limits of detection of 1.59 μmol·L^-1 (NZF) and 2.77 μmol·L^-1 (NFT). Mechanistic studies reveal that this selective turn-off response arises from competitive absorption (inner filter effect, IFE) coupled with photoinduced electron transfer (PET). Overall, this work provides an effective sensor for nitrofuran detection and, more broadly, illustrates a rational design strategy in which planar fluorophore functionalization rigidifies the framework to develop highly emissive MOF-based optical sensors.

  
  Acknowledgments: This research was supported by the National Natural Science Foundation of China (Grant No.21601153), the Qing-Lan Project of Jiangsu Province (HUANG Xingcai), and the State Key Laboratory of Coordination Chemistry, Nanjing University (Grant No.SKLCC2105). Supporting information is available at http://www.wjhxxb.cn
  
• 1. [1]

     ZHENG D S, YIN G Y, LIU M, CHEN C, JIANG Y H, HOU L J, ZHENG Y L. A systematic review of antibiotics and antibiotic resistance genes in estuarine and coastal environments[J]. Sci. Total Environ., 2021, 777: 146009 doi: 10.1016/j.scitotenv.2021.146009

  2. [2]

     YANG W Q, LI J, YAO Z L, LI M. A review on the alternatives to antibiotics and the treatment of antibiotic pollution: Current development and future prospects[J]. Sci. Total Environ., 2024, 926: 171757 doi: 10.1016/j.scitotenv.2024.171757

  3. [3]

     LIU C B, LI B L, LIU M, MAO S. Demand, status, and prospect of antibiotics detection in the environment[J]. Sens. Actuators B, 2022, 369: 132383 doi: 10.1016/j.snb.2022.132383

  4. [4]

     DANNER M C, ROBERTSON A, BEHRENDS V, REISS J. Antibiotic pollution in surface fresh waters: Occurrence and effects[J]. Sci. Total Environ., 2019, 664: 793-804 doi: 10.1016/j.scitotenv.2019.01.406

  5. [5]

     CUI Y J, YUE Y F, QIAN G D, CHEN B L. Luminescent functional metal-organic frameworks[J]. Chem. Rev., 2012, 112(2): 1126-1162 doi: 10.1021/cr200101d

  6. [6]

     KRENO L E, LEONG K, FARHA O K, ALLENDORF M, VAN DUYNE R P, HUPP J T. Metal-organic framework materials as chemical sensors[J]. Chem. Rev., 2012, 112(2): 1105-1125 doi: 10.1021/cr200324t

  7. [7]

     LUSTIG W P, MUKHERJEE S, RUDD N D, DESAI A V, LI J, GHOSH S K. Metal-organic frameworks: Functional luminescent and photonic materials for sensing applications[J]. Chem. Soc. Rev., 2017, 46(11): 3242-3285 doi: 10.1039/C6CS00930A

  8. [8]

     ZHANG Y M, YUAN S, DAY G, WANG X, YANG X Y, ZHOU H C. Luminescent sensors based on metal-organic frameworks[J]. Coord. Chem. Rev., 2018, 354: 28-45 doi: 10.1016/j.ccr.2017.06.007

  9. [9]

     SAMANTA P, LET S, MANDAL W, DUTTA S, GHOSH S K. Luminescent metal-organic frameworks (LMOFs) as potential probes for the recognition of cationic water pollutants[J]. Inorg. Chem. Front., 2020, 7(9): 1801-1821 doi: 10.1039/D0QI00167H

  10. [10]

      RAJA LAKSHMI P, NANJAN P, KANNAN S, SHANMUGARAJU S. Recent advances in luminescent metal-organic frameworks (LMOFs) based fluorescent sensors for antibiotics[J]. Coord. Chem. Rev., 2021, 435: 213793 doi: 10.1016/j.ccr.2021.213793

  11. [11]

      LI H Y, ZHAO S N, ZANG S Q, LI J. Functional metal-organic frameworks as effective sensors of gases and volatile compounds[J]. Chem. Soc. Rev., 2020, 49(17): 6364-6401 doi: 10.1039/C9CS00778D

  12. [12]

      CHEN B L, WANG L B, XIAO Y Q, FRONCZEK F R, XUE M, CUI Y J, QIAN G D. A luminescent metal-organic framework with Lewis basic pyridyl sites for the sensing of metal ions[J]. Angew. Chem.‒Int. Edit., 2009, 48(3): 500-503 doi: 10.1002/anie.200805101

  13. [13]

      CHEN B L, YANG Y, ZAPATA F, LIN G, QIAN G D, LOBKOVSKY E B. Luminescent open metal sites within a metal-organic framework for sensing small molecules[J]. Adv. Mater., 2007, 19(13): 1693-1696 doi: 10.1002/adma.200601838

  14. [14]

      TIAN X M, YAO S L, QIU C Q, ZHENG T F, CHEN Y Q, HUANG H P, CHEN J L, LIU S J, WEN H R. Turn-on luminescent sensor toward Fe³⁺, Cr³⁺, and Al³⁺ based on a Co(Ⅱ) metal-organic framework with open functional sites[J]. Inorg. Chem., 2020, 59: 2803-2810 doi: 10.1021/acs.inorgchem.9b03152

  15. [15]

      HUANG X C, KONG J J. Thiazolothiazole based functional metal-organic frameworks[J]. CrystEngComm, 2025, 27(17): 2611-2622 doi: 10.1039/D5CE00169B

  16. [16]

      KHATUN A, PANDA D K, SAYRESMITH N, WALTER M G, SAHA S. Thiazolothiazole-based luminescent metal-organic frameworks with ligand-to-ligand energy transfer and Hg²⁺-sensing capabilities[J]. Inorg. Chem., 2019, 58: 12707-12715 doi: 10.1021/acs.inorgchem.9b01595

  17. [17]

      YONG W, HUANG Q C, MU H Y, SHI W X, DAI B L, KONG J J, CHEN X R, HUANG X C. A luminescent Zn(Ⅱ) metal-organic framework assembled with a thiazolothiazole chromophore for sensing mainly cobalt(Ⅱ) and nitrofuran antibiotics in aqueous solutions[J]. J. Mol. Struct., 2024, 1301: 137424 doi: 10.1016/j.molstruc.2023.137424

  18. [18]

      HUANG X C, JIANG Y X, HAN Z J, YONG W, HUANG Q X, SHI W X, CHEN X R, KONG J J, WU H L. A thiazolothiazole based Zn(Ⅱ) metal-organic framework with multi-responsive fluorescent sensing toward Al³⁺ and oxoacid anions in an aqueous solution[J]. J. Mol. Struct., 2023, 1289: 135897 doi: 10.1016/j.molstruc.2023.135897

  19. [19]

      HUANG Q C, YONG W, YANG B Y, REN J, KONG J J, HUANG X C. Dual thiazolo[5,4-d]thiazole-based Zn(Ⅱ)/Cd(Ⅱ) metal-organic frameworks for the fluorescence sensing of antibiotics and Cr􀃱 ion in aqueous solutions[J]. J. Mol. Struct., 2025, 1344: 143032 doi: 10.1016/j.molstruc.2025.143032

  20. [20]

      OLESIŃSKA M, WU G L, GÓMEZ-COCA S, ANTÓN-GARCÍA D, SZABÓ I, ROSTA E, SCHERMAN O A. Modular supramolecular dimerization of optically tunable extended aryl viologens[J]. Chem. Sci., 2019, 10(38): 8806-8811 doi: 10.1039/C9SC03057C

  21. [21]

      HUANG X C. Crystal Shape v1.0[CP]. Yancheng: Yancheng Teachers University, 2026.

  22. [22]

      BLATOV V A, SHEVCHENKO A P, PROSERPIO D M. Applied topological analysis of crystal structures with the program package ToposPro[J]. Cryst. Growth Des., 2014, 14(7): 3576-3586 doi: 10.1021/cg500498k

  23. [23]

      DAS C, PATEL V D, GUPTA D, MAHATA P. Isolation of a Cd-based coordination polymer containing mixed ligands: Time- and temperature-dependent synthesis, sulfonamide antibiotics detection, and Schottky diode fabrication[J]. Inorg. Chem., 2024, 63: 3656-3666 doi: 10.1021/acs.inorgchem.3c03086

  24. [24]

      DING S, YANG G S, WANG S Q, HUI Y T, WANG Y J, TANG Z, JIANG C J, SU Z M. Carbazolyl-decorated metal-organic framework with a high fluorescent quantum yield for detection and photocatalytic degradation of organic contaminants[J]. Inorg. Chem., 2024, 63: 22572-22582 doi: 10.1021/acs.inorgchem.4c04044

  25. [25]

      RANI P, HUSAIN A, BHASIN K K, KUMAR G. Zinc(Ⅱ)-MOF: A versatile luminescent sensor for selective molecular recognition of flame retardants and antibiotics[J]. Inorg. Chem., 2024, 63: 3486-3498 doi: 10.1021/acs.inorgchem.3c04214

  26. [26]

      DATTA S, GHORAI P, CHATTOPADHYAY M K, JANA N C, BANERJEE P, MIR M H. Strategic advancement in inner filter effect controllable detection of tetracyclines and nitroaromatics in real-world matrices by two-dimensional coordination polymer[J]. Cryst. Growth Des., 2024, 24: 8645-8654

  27. [27]

      ZHANG J W, GE H, SHI J H, TAO H Y, LI B, YU X Y, ZHANG M T, XU Z L, XIAO R J, LI X Y. A tiered probabilistic approach to assess antibiotic ecological and resistance development risks in the fresh surface waters of China[J]. Ecotox. Environ. Safe., 2022, 243: 114018 doi: 10.1016/j.ecoenv.2022.114018

  28. [28]

      CAI Y Y, DONG T, BIAN Z H, LIU H, LIU X, LIU A H. Metal-organic frameworks based fluorescent sensing: Mechanisms and detection applications[J]. Coord. Chem. Rev., 2025, 529: 216470 doi: 10.1016/j.ccr.2025.216470

  29. [29]

      XU Y H, SHI Q, DENG W B, CHEN M H, LI X Y, XIONG J, ZHANG B X, WEI T, FENG Y Q, ZHANG B. Design and construction of a porphyrin-based samarium􀃮-metal-organic framework for antibiotic detection[J]. Inorg. Chem., 2025, 64(2): 1193-1202 doi: 10.1021/acs.inorgchem.4c05045

  30. [30]

      LUO F L, WANG H B, WU K, LUO D, ZHAO Y F, HUANG Y L, LU W G. Pyrazine-functionalized bio-MOFs for selective luminescence sensing of colchicine and nitrofurazone in aqueous solutions[J]. Inorg. Chem., 2025, 64: 11650-11657 doi: 10.1021/acs.inorgchem.5c01125

• 

  Scheme 1  Modification of Cd(Ⅱ)-MOFs with a rigid thiazolothiazole core: enhancement of π…π interactions and significant boost in PL emission

  下载: 全尺寸图片 幻灯片
  

  Figure 1  (a, d) SBUs of 1 and 2; (b, e) Coordination geometries of the Cd(Ⅱ) centers in 1 and 2; (c, f) 2D frameworks of 1 and 2, with the insets highlighting the π…π stacking interactions

  Symmetry codes: ^ⅰ x, y, z-1; ^ⅱ-x+2, -y+2, -z+2; ^ⅲ x+1, y+1, z+1 for 1; ^ⅳ x+1, y-1, z; ^ⅴ x, y, z+1; ^ⅵ -x+1, -y, -z+2 for 2.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Solid-state PL spectra of the free ligands (H₂CDA, 4bpyttz, 4bbpy) and the corresponding Cd(Ⅱ) frameworks (MOFs 1 and 2) recorded at room temperature; (b) Calculated frontier molecular orbital energy levels for the primary ligand components in MOFs 1 and 2, illustrating the distinct charge transfer pathways and the suppression of non-radiative decay in MOF 1

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a, b) Absolute PL quantum yield (PLQY) spectra of solid-state crystals of MOFs 1 and 2 measured using an integrating sphere at room temperature; (c) PL emission spectra of MOFs 1 and 2 in water; (d) Comparison of the emission intensities of MOFs 1 and 2

  The inset shows digital photographs of MOFs 1 and 2 dispersed in aqueous suspensions under 395 nm UV light irradiation, illustrating the significant difference in emission intensity between the two frameworks.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) PL spectra of MOF 1 in the presence of various antibiotics (λ_ex=396 nm); (b) Relative PL intensities of MOF 1 dispersed in different antibiotic solutions, highlighting the highly selective and distinct fluorescence quenching triggered by the nitrofuran antibiotics NZF and NFT

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) Concentration-dependent 3D PL spectra of the MOF 1 aqueous suspension upon the incremental addition of NZF (titration volume: 0-2 000 μL) (inset: digital photographs under 395 nm UV light illustrating the visual fluorescence quenching); (b) Corresponding S-V plot for NZF (inset: linear fitting of the S-V plot in the low concentration region); (c) Concentration-dependent 3D PL spectra of the MOF 1 aqueous suspension upon the incremental addition of NFT (0-2 000 μL) (inset: digital photographs under 395 nm UV light); (d) Corresponding S-V plot for NFT (inset: linear fitting at the low concentration region)

  下载: 全尺寸图片 幻灯片
  

  Figure 6  Spectral overlap between the UV-Vis absorption spectra of various antibiotics (solid lines) and the excitation (pink shaded region) and emission (blue shaded region) spectra of MOF 1

  下载: 全尺寸图片 幻灯片
  

  Figure 7  UV-Vis absorption spectra of the MOF 1 aqueous suspension upon the incremental addition (0-90 μL) of (a) NZF and (b) NFT

  下载: 全尺寸图片 幻灯片

  Table 1.  Crystallographic data and structure refinement parameters for MOFs 1 and 2

  Parameter	1	2
  Formula	C35H30CdN6O7S2	C31H20CdN2O5
  Formula weight	823.17	612.90
  Crystal system	Triclinic	Triclinic
  Space group	P1	P1
  a / nm	0.890 99(2)	0.934 55(18)
  b / nm	1.446 74(3)	1.363 1(3)
  c / nm	1.529 91(4)	1.542 7(2)
  α / (°)	111.274(2)	68.042(18)
  β / (°)	94.790(2)	84.214(16)
  γ / (°)	98.479(2)	85.800(19)
  V / nm3	1.797 43(8)	1.812 0(6)
  Z	2	2
  T / K	293(2)	293(2)
  Dc / (g·cm-3)	1.521	1.123
  Crystal size / mm	0.3×0.26×0.2	0.28×0.25×0.2
  μ(Mo Kα) / mm-1	0.780	0.653
  F(000)	836	616
  2θ range / (°)	4.674-59.27	4.384-49.996
  Reflection collected, unique	22 218, 8 049	15 281, 6 088
  Rint	0.026 5	0.148 2
  Tmax, Tmin	1.000, 0.762	1.000, 0.648
  Data, Nres, Npara	8 049, 54, 464	6 088, 21, 352
  R1, wR2 [I > 2σ(I)]b	0.036 7, 0.099 0	0.117 4, 0.291 6c
  R1, wR2 (all data)	0.044 3, 0.104 8	0.168 2, 0.341 3c
  GOF on F 2	1.042	1.078
  a Nres=number of restraints, Npar=number of parameters; b R1=∑||Fo|-|Fc||/∑|Fo|; wR2=[∑w(Fo2-Fc2)2/ ∑w(Fo2)2]1/2; c The relatively high R-indices and residual electron density for MOF 2 originate from the inherent framework lability and rapid partial loss of lattice water molecules, rather than poor data collection.

  下载: 导出CSV
•