Construction of a Luminescent Cadmium-Based Metal–Organic Framework for Highly Selective Discrimination of Ferric Ions

Abstract

Fluorescent metal–organic frameworks (MOFs) are ideal materials for sensors because of their adjustable pore size and functional groups, which provide them with favorable metal ion selective recognition. In this paper, a new cadmium-based MOF was synthesized using Cd(NO₃)₂·4H₂O and 3,3′,5,5′-biphenyltetracarboxylic acid by solvothermal method. CdBPTC owned three types of channels with dimensions of approximately 8.4 × 8.3 Å, 6.0 × 5.2 Å, 9.7 × 8.4 Å along a, b, and c axis, respectively. This MOF has high selectivity to ferric ions and shows excellent anti-inference ability toward many other cations. The results indicate that the fluorescence quenching efficiency of CdBPTC is close to 100% when the concentration of Fe³⁺ reaches 1.0 × 10⁻³ mol·L⁻¹. Moreover, the luminescent intensity at 427 nm presents a linear relationship at a concentration range of 2.0 × 10⁻⁴~7.0 × 10⁻⁴ mol·L⁻¹, which can be quantitatively expressed by the linear Stern–Volmer equation I₀/I = 8489 [Fe³⁺] − 0.1400, which is comparable to the previously reported better-performing materials. Competitive energy absorption and ion exchange may be responsible for the variation in fluorescence intensity of CdBPTC in different Fe³⁺ concentrations.

1. Introduction

Iron ions play an irreplaceable role in living systems, and the lack or excess of ferrum in the human body causes serious dysfunction. Iron ions are an important component of hemoglobin and myoglobin but are also related to transferrin. In addition, the level of iron ions content has a profound impact on industrial and agricultural production, as well as daily life [1,2,3,4,5]. Therefore, the detection of iron ions is of great significance for environmental detection and life science. Currently, the methods for the determination of iron ions include oxidation–reduction titration, electrochemistry, inductively coupled plasma spectroscopy (ICP), inductively coupled plasma mass spectrometry (ICP–MS), flame atomic absorption spectrometry (FAAS), etc. Each approach has its advantages and disadvantages. For example, the methods of polarography and potassium dichromate titration analysis are simple to operate but with low sensitivity. Although FAAS and ICP–MS have good sensitivity, their high price, complicated pretreatment, and operation restrict their widespread use. In this context, the detection of iron ions based on luminescence has attracted much attention due to its simplicity, rapidity, and high sensitivity [6,7,8,9,10].

Metal–organic frameworks (MOFs) are new types of porous crystal materials [11,12] mainly composed of inorganic metal ions or metal clusters and organic ligands connected by coordination bonds to form regular two- or three-dimensional network structures, which have been widely applied in separation and storage of gases [13], chemical sensing [14], catalysis [15,16], food detection [17], biomedicine [18,19] and other fields [20,21]. Fluorescent MOFs and organic porous materials [22,23,24] are ideal candidates for sensors because of their adjustable pore size and functional groups, which provide them with favorable metal ion selective recognition. It has been reported in the literature that iron quenched the fluorescence of MOFs owing to the paramagnetic feature and intensely visible absorbance of Fe³⁺, thus forming a “turn-off” sensor. For instance, Lv. R. et al. [5] synthesized a new luminescent Cd-MOF, which can detect Al³⁺ and Fe³⁺ by means of fluorescence quenching. The detection limit reached 0.56 μM for Al³⁺ and 0.3 μM for Fe³⁺, respectively. Nevertheless, some examples selectively detected Fe³⁺ via the “turn-on” effect, such as Co-based JXUST-2 [25], and Tb-based Tb-TCPP [26,27], in which turn-on MOFs were used simultaneously as fluorescent sensors for Fe³⁺, Cr³⁺, and Al, and displayed a 10-fold fluorescence enhancement. However, few studies have used MOFs for the specificity recognition of Fe³⁺ through the turn-on or turn-off effect under different concentrations.

In this paper, a new cadmium-based MOF (named as CdBPTC) was synthesized using Cd(NO₃)₂·4H₂O and 3,3′,5,5′-biphenyltetracarboxylic acid (H₄BPTC) by solvothermal method and characterized by single-crystal X-ray diffraction (SXRD), powder X-ray diffraction (PXRD), thermogravimetric analysis (TGA), Fourier transform infrared spectroscopy (FTIR), and fluorescence spectroscopy. Fe³⁺ significantly affected the fluorescence properties of CdBPTC. Interestingly, the fluorescence intensity was higher than that of CdBPTC when the concentration of Fe³⁺ was in the range of 1 × 10⁻⁹~1 × 10⁻⁵ mol·L⁻¹. Above 10⁻⁴ mol·L⁻¹, the fluorescence intensity dropped sharply with turn-off effect, showing a linear relationship in the range of 2 × 10⁻⁴~7 × 10⁻⁴ mol·L⁻¹.

2. Results and Discussion

2.1. Crystal Structure Description of {[Me₂NH₂]₂[CdBPTC]·6H₂O}n (CdBPTC)

SXRD analyses revealed that CdBPTC crystallized in a tetragonal noncentrosymmetric I4₁cd space group (No. 110). As shown in Figure 1a, the asymmetric unit of CdBPTC consists of a crystallographically independent Cd²⁺, a fully deprotonated BPTC⁴⁻ ligand. The coordination environment of the eight-coordinated Cd²⁺ center is composed of eight oxygen atoms from four chelating bidentate carboxylate groups, forming a simple mononuclear Cd(COO)₄ SBU (Figure 2a). The BPTC⁴⁻ ligand serves as a μ₄-bridge to link four Cd centers by a μ₄-η²:η²:η²:η² coordination mode through eight carboxylate oxygen atoms and extended into an infinite three-dimensional (3D) porous network (Figure 1b,c, Figure 2a, and Figure S1). CdBPTC owns three types of one-dimensional (1D) channel with dimensions of approximately 8.4 × 8.3 Å (pore A), 6.0 × 5.2 Å (pore B), and 9.7 × 8.4 Å (pore C), respectively, along the crystallographic a, b, and c axis, giving rise to the high porosity of 68.1% per unit cell volume, which is calculated by using the squeeze routine of PLATON [28]. Due to the presence of a large number of Me₂NH₂⁺ in the channel, hydrogen bonding forces may be generated with MOFs, resulting in the collapse of the frameworks after the removal of these ions. Unfortunately, we could not obtain specific surface area through conventional experiments. Crystal data and structure refinement parameters are summarized in

Table 1. Crystal data and structure refinement results for CdBPTC.

Compound	CdBPTC
Empirical formula	C16H6CdO8
Formula weight	438.61
Temperature (K)	100.15
Crystal system	Tetragonal
Space group	I41cd
a/Å	20.07520 (10)
c/Å	36.7826 (3)
Volume (Å3)	14,823.89 (19)
Z	16
ρcalc (g·cm−3)	0.786
μ (mm−1)	4.897
F (000)	3424.0
Independent reflections	5869 [Rint = 0.0332]
Data/restraints/parameters	5869/25/215
Goodness of fit on F2	1.087
R1 [I ≥ 2σ(I)] a	0.0399
wR2 [all data] b	0.1365
Flack parameter	0.478 (16)
CCDC number	2,112,156

^a R₁ = Σ(||F₀| − |F_c||)/Σ|F₀|; ^b wR₂ = [Σw(F₀² − F_c²)²/Σw(F₀²)²]^1/2.

and Table S1.

The topological network pattern of CdBPTC is shown in Figure 2. In order to reflect the shape of the MOF channel, the BPTC^4- ligand and mononuclear Cd²⁺ unit were considered to be three- and four-connected nodes instead of four- and four-connected nodes, respectively. Topological analysis indicated that the structure could be simplified as a (4, 3, 3)-connected net with stoichiometry (3-c)₂(4-c) and point symbol of {6².8⁴}{6².8}₂. The phase purity of the as-synthesized CdBPTC was characterized by PXRD. As a result, the experimental peaks were consistent with the simulated one obtained by SXRD, indicating the samples were of high purity (Figure S2). However, there were slight differences in the low 2theta degrees due to the disorder of crystal structure after grinding (Figure S12).

2.2. Framework Stability

The TGA curve of CdBPTC revealed that the framework could be thermally stable up to ca. 360 °C (Figure S4). When the temperature rose to 140 °C, the weight loss was about 17.0%, which should be the escape of six water molecules. Then, the MOF slowly lost the two dimethylamine cations and began to collapse when the temperature exceeded 360 °C. The CdBPTC powders were immersed in different organic solvents such as DMF, DMSO, EtOH, etc. for 3 days, respectively. Then, the solid samples were filtered, followed by PXRD measurements. Comparing the PXRD patterns of the experimental and simulated samples, no obvious differences were found (Figure S5). These results demonstrated the excellent thermal and chemical stability of CdBPTC.

2.3. Luminescent and Sensing Properties

As depicted in Figure 3, the luminescence characters of CdBPTC were performed in a solid state at room temperature. The sample showed blue fluorescence with an emission peak at 427 nm (excited at 354 nm), which could be mainly due to the intraligand charge transfer (ILCT) because the ligand H₄BPTC also exhibited a similar emission (Figure S6).

2.3.1. Fluorescence Detection of Metal Ions

A total of 15 mg ground CdBPTC sample was added to 15 mL DMF, then sonicated in an ultrasonic washer for 20 min; thus, suspended solutions were obtained. Na⁺, Ca²⁺, Mg²⁺, Ba²⁺, Cd²⁺, Zn²⁺, Mn²⁺, Co²⁺, Ni²⁺, Pb²⁺, Cu²⁺, La³⁺, Eu³⁺, Tb³⁺, Al³⁺, Cr³⁺, and Fe³⁺ aqueous solutions (30 μL 0.1 mol·L⁻¹) were separately added to CdBPTC suspensions (3 mL), stirred, and tested the emission spectra immediately. Obviously, the fluorescence of MOF was almost completely quenched after the addition of Fe³⁺. In comparison, the addition of other cations did not cause an obvious fluorescence response, and only partial decreased luminescence intensity was observed for Al³⁺ and Cr³⁺ (about 30% lower than the original intensity; Figure 4a). After the addition of Eu³⁺ and Tb³⁺, the unique characteristic peaks of rare-earth ions were observed, but the emission peaks of MOF at 427 nm did not decrease significantly (Figures S14 and S15). CdBPTC exhibited suitable selectivity and responsiveness to Fe³⁺ with a turn-off effect.

Considering the influence of other ions on Fe³⁺ selectivity, 1.0 × 10⁻³ mol·L⁻¹ Na⁺/Fe³⁺, Ca²⁺/Fe³⁺, Mg²⁺/Fe³⁺, Ba²⁺/Fe³⁺, Cd²⁺/Fe³⁺, Zn²⁺/Fe³⁺, Mn²⁺/Fe³⁺, Co²⁺/Fe³⁺, Ni²⁺/Fe³⁺, Pb²⁺/Fe³⁺, Cu²⁺/Fe³⁺, La³⁺/Fe³⁺, Cu²⁺/Fe³⁺, Al³⁺/Fe³⁺ competition experiments were also carried out. As shown in Figure 4b, the fluorescence intensity decreased significantly with the addition of the corresponding Fe³⁺. Additionally, the fluorescence quenching was completed immediately; that is to say, the turn-off effect still existed in the presence of other cations. The above results manifested that CdBPTC had high selectivity for Fe³⁺.

To further study the sensitivity of CdBPTC for sensing Fe³⁺, the quantitative detection was investigated through the fluorescence titration spectra analyses. The fluorescence intensity of MOF gradually increased with the increase in Fe³⁺ concentrations in the range of 1.0 × 10⁻⁹~1.0 × 10⁻⁶ mol·L⁻¹ (Figure 5a, the fluorescence measurements should be tested after 10 min with steady signals). In order to fairly evaluate the fluorescence enhancement effect of MOF in the presence of Fe³⁺ at very low concentrations, we conducted the same experiment on Cr³⁺, Tb³⁺, Eu³⁺, La³⁺, Zn²⁺, Mn²⁺, Fe²⁺ at the same concentration, respectively. As shown in Figures S13–S19, the fluorescence intensity also increased slightly in the presence of other metal ions. To eliminate the influence of a small amount of water in the solution, Fe³⁺ in DMF solution was selected for an experiment under the same conditions, and the same phenomenon was obtained, as shown in Figure S11. We speculate that this can be caused by the two Me₂NH₂⁺ contained in the MOF pore exchange ions with these metal cations. Interestingly, as the Fe³⁺ concentration reached 1.0 × 10⁻⁴ mol·L⁻¹, the fluorescence intensity dropped sharply (Figure 5b). It is worth noting that attenuation of the fluorescence signals could be observed immediately. The fluorescence-quenching efficiency was calculated by using the formula: (1 − I/I₀) × 100%, where I₀ and I are the values of the luminescence intensity of pristine CdBPTC and Fe³⁺ loaded CdBPTC, respectively, indicating that CdBPTC showed a high sensitivity in sensing fluorescence quenching. The quenching efficiency was close to 100% when the concentration of Fe³⁺ reached 1.0 × 10⁻³ mol·L⁻¹ (Figure 5c). Moreover, the luminescence-quenching coefficient (Ksv) can be quantitatively expressed by the linear Stern–Volmer equation [29,30] I₀/I = 8489 [Fe³⁺] − 0.1400. The luminescent intensity at 427 nm presented a linear relationship at a concentration range of 2.0 × 10⁻⁴~7.0 × 10⁻⁴ mol·L⁻¹ and R² > 0.99 (Figure 5d). Therefore, the Ksv value could reach 8489 L·mol⁻¹, which is comparable to the previously reported [31,32,33] better-performing materials. By the calculation of the 3σ/k formula (σ is the standard error; k is the slope), the detection limit for Fe³⁺ was found to be 9.04 ppm. In addition, we also tested the fluorescence spectra of CdBPTC in the presence of Fe²⁺ at different concentrations. As shown in Figure S19, the fluorescence intensity also showed obvious enhancement at 10⁻⁹~10⁻⁶ mol·L⁻¹. When the concentration increased to 10⁻³ mol·L⁻¹, the fluorescence quenched immediately, but the quenching efficiency (85%) was lower than that of Fe³⁺.

2.3.2. Fluorescence-Sensing Mechanism of the CdBPTC Probe

According to previous reports, the luminescence effect of MOF caused by metal ions mainly results from the destruction of MOF structure, ion exchange, host–guest interaction, and iron quenched the fluorescence of MOFs, owning to the paramagnetic feature and significant visible absorbance of Fe³⁺. In order to investigate the possible mechanism, 15 mg CdBPTC solid sample was immersed into 1 mL 1.0 × 10⁻² mol·L⁻¹ DMF solutions of different metal ions (replace the metal salt solution after 24 h), respectively, and kept in dark for three days. Then, the solid fluorescence and PXRD were measured after filtration and drying (Figures S8–S10). The structure of CdBPTC before and after the addition of metal ions was studied. As shown in Figures S8 and S12, after adding metal ions, or after grinding MOF and then sensing Fe³⁺, or after soaking in water for 6 h, the framework was maintained, as confirmed by basically unchanged PXRD diffraction peaks, indicating that the luminescence quenching was not caused by the damage of the MOF structures.

Furthermore, given the anionic framework of CdBPTC, the increased fluorescence of MOF in the presence of metal cations at low concentrations may be the result of cation exchange. With the initial addition of Fe³⁺ (10⁻⁹~10⁻⁶ mol·L⁻¹), the charge balance of the frameworks was gradually broken. After 10 min, the stable fluorescence signal indicated that the ion exchange between host and guest was complete, and the fluorescence intensity of MOF was enhanced. However, when the concentration of Fe³⁺ increased to 1.0 × 10⁻⁴ mol·L⁻¹, the fluorescence quenching of MOF was obvious without waiting. To explore the possible mechanism, the UV–Vis spectrum of MOF in the solid state was tested (Figure S20). The absorption peaks of CdBPTC appear in the range of 200–350 nm. Moreover, the UV–Vis absorption spectra of Fe³⁺ with different concentrations were measured (Figure S21). No obvious absorption signals were observed at 10⁻⁵ and 10⁻⁶ mol·L⁻¹. When the concentration reached 10⁻⁴ mol·L⁻¹, significant absorption peaks began to appear in the range of 300~400 nm, which overlapped with the absorption wavelength of MOF. Meanwhile, the UV–Vis absorption spectra of Na⁺, Mg²⁺, Co²⁺, Ni²⁺, Pb²⁺, Cu²⁺, Mn²⁺, Zn²⁺, Fe²⁺, La³⁺, Al³⁺, Cr³⁺, Tb³⁺, and Eu³⁺ were measured at the same concentration, respectively, and only Fe³⁺ had an obvious absorption peak (Figure S21). In addition, the maximum emission wavelength of CdBPTC solid sample was located at 427 nm within the Fe³⁺ concentration in the range of 1.0 × 10⁻⁹~1.0 × 10⁻⁶ mol·L⁻¹, while above 10⁻³ mol·L⁻¹, a weak shoulder peak appeared at 470 nm in the solid-state fluorescence spectrum, whereas others showed little change (Figure S10). Therefore, we speculate that ion exchange is the main reason affecting the fluorescence of MOF at low concentration, while at high concentration, the competitive absorption of excitation source energy between metal ions and MOF leads to the fluorescence quenching of MOF [33,34,35,36]. Interestingly, when CdBPTC was exposed to Fe³⁺ in the DMF solution, the color of the solution changed from yellow to brown. By contrast, the naked eye recognition and detection for Fe³⁺ could be realized.

3. Materials and Methods

3.1. Chemicals and Reagents

All reagents and solvents including the H₄BPTC ligand were purchased from commercial sources and used without further purification.

3.2. Apparatus

TGA was carried out in a nitrogen stream using the DTG-60 thermal analysis equipment with a heating rate of 10 °C·min⁻¹ (Shimadzu, Kyoto, Japan). PXRD patterns of the bulk samples were measured on a Rigaku MiniFlex600 X-ray diffractometer (Rigaku, Kyoto, Japan). FTIR was obtained by KBr disks on a Shimadzu IRAffinity-1 FTIR spectrometer in the range of 4000–400 cm⁻¹ (Shimadzu, Kyoto, Japan, abbreviations for the IR bands: w = weak, m = medium, b = broad, and vs. = very strong). Elemental analysis was carried out with Vario EL III CHNS equipment (Elementar, Langenselbold, Germany). The steady-state photoluminescence spectra (PL) for all samples were recorded on an RF-5301PC spectrofluorometer (Shimadzu, Kyoto, Japan).

3.3. Synthesis of {[Me₂NH₂]₂[CdBPTC]·6H₂O}n

A mixture of Cd(NO₃)₂·4H₂O (0.05 mmol, 16 mg), 3,3′,5,5′-biphenyltetracarboxylic acid (H₄BPTC, 0.0375 mmol, 13 mg), and mixed solvent N,N-dimethylformamide (DMF)/H₂O (4.5 mL, 4:0.5, v/v), with an additive of drops of HNO₃ (68%, 0.08 mL), was sealed in a Pyrex glass tube and heated in an oven at 90 °C for 72 h and cooled to room temperature at a rate of 5 °C·h⁻¹. Colorless prism crystals were filtered and washed with DMF. Yield: ca. 90% based on H₄BPTC. FTIR spectrum (KBr pellets, cm⁻¹): 3442(br, m), 3078(br, m), 2789(m), 2472(w), 1616(s), 1562(s), 1435(s), 1413(s), 1363(s), 1315(m), 1018(w), 1014(w), 773(m), 731(m), 657(m), 651(m), 516(w), 420(w). Elemental analysis (CHN), (C₂₀H₃₄N₂O₁₄Cd)_n, that is, {[Me₂NH₂]₂ [Cd(BPTC)]∙6H₂O)}_n, calculated (%): C 37.60, H 5.36, N 4.38; found (%): C 37.72, H 5.29, N 4.67.

3.4. Determination of the Crystal Structures

A suitable size and high-quality crystal of CdBPTC was mounted at the end of glass fiber for single-crystal X-ray diffraction analysis. Data collection was performed by a Rigaku XtaLAB PRO MM007-DW diffractometer system, which was equipped with an RA-Micro7HF-MR-DW(Cu/Mo) X-ray generator and a Pilatus3R-200K-A detector (Cu Kα, λ = 1.54178 Å). The structure was solved by direct methods and refined by full-matrix least-squares refinements based on F². Anisotropic thermal parameters were applied to all non-hydrogen atoms. The hydrogen atoms were generated geometrically. The crystallographic calculations were performed using the SHELXL programs [37,38]. The treatment for the guest molecules in the cavities of CdBPTC involves the use of the SQUEEZE program of PLATON. The existences of [Me₂NH₂]⁺ and H₂O were further verified by the FTIR spectra, elemental analysis, and TGA data, obtaining the final chemical formulas of {[Me₂NH₂]₂[CdBPTC]·6H₂O}n.

3.5. Sensing Experiment

In metal ions sensing, 3 mg of CdBPTC was ground to a fine powder and ultrasonically dispersed into 3 mL DMF to prepare the 1 mg·mL⁻¹ uniform suspensions. Various M(NO₃)x aqueous solutions were added to the above suspensions. Then, the suspensions were loaded in the cuvettes for collecting the photoluminescence spectra. The fluorescence intensity of the suspension in DMF was measured as the blank sample.

4. Conclusions

In summary, we successfully synthesized an anionic microporous MOF, CdBPTC, with good thermal and chemical stability. CdBPTC was proved to be highly selective for the discrimination of ferric ions and showed excellent anti-inference ability toward many other cations. The luminescent intensity presented a linear relationship at a concentration range of 2.0 × 10⁻⁴~7.0 × 10⁻⁴ mol·L⁻¹. The Ksv was about 8489 L·mol⁻¹, which was obtained from the linear Stern–Volmer equation. This MOF exhibited a fluorescence “turn-off” effect for Fe²⁺ and Fe³⁺, which may be attributed to the competitive absorption of excitation source energy between ferric ions and CdBPTC. This type of MOFs can be developed as a potential candidate for the detection of Fe³⁺ in complex environments and can also be used for the determination of iron ion concentration in environmental hygiene, food safety, medical health, soil, and other media.