Fluorescent probe for Fe 3+ detection based on a guest molecular luminescent metal–organic framework

Abstract

The zinc(II) bis-(8-hydroxyquinoline) (Znq₂) has excellent photoluminescence properties, and its fluorescence emission can be significantly quenched by Fe³⁺ in water. To accelerate the detection response of Znq₂ to Fe³⁺, a luminescent metal–organic framework Znq₂@ZIF-8 based on guest molecular luminescence was constructed by growing zeolite imidazolate framework-8 (ZIF-8) on the outer surface of Znq₂. The results show that the prepared Znq₂@ZIF-8 has an octahedral core–shell structure, a particle size of approximately 1–3 μm, an enhanced specific surface area of 1105.41 m² g⁻¹, and with a stable green luminescence at 495 nm. A fluorescence analytical method was developed for the detection of Fe³⁺ in water, the correlation coefficients were significant in the Fe³⁺ concentration range of 0–600 μmol L⁻¹, and the limit of detection was as low as 3.89 μmol L⁻¹. The spiked recoveries of tap water samples demonstrated that the method could be applied to practical applications. The mechanism of fluorescence detection is that Fe³⁺ participates in the competitive coordination of Znq₂@ZIF-8 metal centers, leading to the collapse of the crystal structure, meanwhile, Fe³⁺ produces a certain degree of competitive absorption of the excitation light of Znq₂@ZIF-8. This method was applied for the detection of Fe³⁺ in water with good selectivity, anti-interference ability, and has the potential to be used as a rapid detection method.

Introduction

Iron (Fe), an essential trace element, is involved in the synthesis of hemoglobin and oxidative enzymes in the human body, and if the iron content in the body deviates from the appropriate range, it may cause mental depression, metabolic disorders, growth retardation, and increased risk of cancer.^1–3 Most of the existing assays use atomic fluorescence spectrometry, atomic absorption spectrometry and X-ray fluorescence spectroscopy to determine the metal content, but most of these methods are time-consuming and expensive to perform and require skilled operators, which are shortcomings that limit their application in routine analysis. ⁴ Fluorescence detection, a method used to indicate the content of a substance through changes in the photoluminescence intensity of a fluorescent material, can be used to detect a particular substance through rational design and has the advantages of less interference, simple operation, high sensitivity and low cost.^5–7 Metal–organic frameworks (MOFs) are self-assembled from metal ions and organic ligands with high porosity, specific surface area, and good structural stability. ⁸ One class of MOFs with optical properties is known as luminescent metal–organic framework (LMOFs), where fluorescence can come from metal centers, organic ligands, and guest molecules and has been used to identify and detect environmental pollutants such as metal cations, anions, biomolecules, and pesticides in the environment. ⁹ The synthesis of LOMFs based on guest molecule luminescence is simple and can utilize the excellent optical properties of guest molecules, on one hand, and the high specific surface area and porosity of the MOF matrix, on the other hand, which can effectively enrich pollutants and improve the detection response speed.

Traditional guest luminescent molecules mainly include dyes, quantum dots, luminescent metal complexes, etc. In the synthesis of guest molecule–based luminescent LMOFs, since most MOFs need to be synthesized under hydrothermal conditions, avoiding the influence of specific synthesis conditions on the structure and fluorescence properties of guest molecules is an important prerequisite. For example, Han et al. ¹⁰ used a one-pot method to rapidly encapsulate highly luminescent AlQ₃ (tris(8-hydroxyquinoline) aluminum) as a guest molecule within the pore structure of ZIF-8 host crystals, which did not result in morphological and structural differences compared to a single ZIF-8 structure, failing to account for the morphological and structural compositions of the AlQ₃ in the composite. Many studies have utilized the structural characteristics of the MOF matrix to encapsulate dye-luminescent molecules inside its pores. However, considering that dye molecules are instability, they are difficult to apply to the detection of pollutants in aqueous environments and may generate secondary pollution.^11–13

In this study, a guest molecular luminescent–based LMOF was constructed as shown in Figure 1. Zinc(II) bis-(8-hydroxyquinoline) (Znq₂) with an octahedral morphology was synthesized by a hydrothermal method, and ZIF-8 was synthesized by in situ coating on the outer surface of Znq₂ to form a core–shell structure of Znq₂@ZIF-8, and the composite has the morphology and structure of each component. Znq₂@ZIF-8 retains the luminescent characteristics of Znq₂, while the outer coating of ZIF-8 can accelerate the rate of mass transfer recognition of targets and improve structural stability. A turn-off fluorescent probe was constructed for the trace detection of Fe³⁺ using fluorescence analysis, and the spiked recovery test demonstrated the potential of the fluorescence analysis method for practical application.

Figure 1.

Schematic diagram of the synthesis pathway and Fe³⁺ detection of Znq₂@ZIF-8.

Results and discussion

Characterization of morphology and structure

Scanning electron microscopy (SEM) images of Znq₂, ZIF-8, and Znq₂@ZIF-8 are shown in Figure 2(a)–(c), respectively. Znq₂ has a smooth octahedral structure with a particle size of 1–3 μm, whereas the particle size of ZIF-8, which is used as a shell layer material, is approximately 50 nm. The core–shell material Znq₂@ZIF-8 with Znq₂ as the core and ZIF-8 as the shell layer has a distinct octahedral profile. Compared with the smooth Znq₂ octahedra, many small ZIF-8 particles were obviously present on the outer surface of the formed Znq₂@ZIF-8, and the size of the material increased slightly with the formation of shell layer. As shown in the transmission electron microscopy (TEM) image in Figure 2(d), Znq₂@ZIF-8 has an obvious core–shell structure. Figure 2(e) shows the SEM-energy dispersive X-ray spectroscopy (EDS) mapping analysis, and C, O, N and Zn are uniformly distributed in Znq₂@ZIF-8.

Figure 2.

SEM images of (a) Znq₂, (b) ZIF-8, and (c) Znq₂@ZIF-8. (d) TEM image of Znq₂@ZIF-8. (e) SEM-EDS elemental mapping profiles of Znq₂@ZIF-8.

Fourier-transform infrared spectroscopy (FTIR), as shown in Figure 3(a), revealed that ZIF-8 is located at 1580, 1500–1350, 1350–600, and 420 cm⁻¹, which correspond to the stretching vibration of C=N, the planar stretching vibration of the imidazole ring, the planar bending vibration of the imidazole ring, and the stretching vibration peaks of the Zn–N bond, respectively, ¹⁴ these peaks are also found in Znq₂@ZIF-8. In addition, there are also C=N stretching vibration peaks and Zn–N bond stretching vibration peaks at 1580 and 420 cm⁻¹ for Znq₂, and the characteristic vibrational absorption of the benzene ring skeleton at 1500 and 1467 cm⁻¹ by quinoline structure, the stretching vibration absorption of the C–O bond at 1279 and 1114 cm⁻¹, and the absorption belonging to Zn–O and Zn–N in the range of 400–600 cm⁻¹, these peaks are also present in Znq₂@ZIF-8. ¹⁵ As shown in X-ray diffraction analysis (XRD) diffraction pattern (Figure 3(b)), the characteristic diffraction peaks of Znq₂ located at 8.9°, 10.9°, 13.7°, and 22.6° correspond to the four crystal planes of (023), (110), (101), and (311), respectively, ¹⁶ which appear in Znq₂@ZIF-8. Similarly, the diffraction peaks located at 7.4°, 12.8°, 14.7°, 16.5°, and 18.1° in ZIF-8 are attributed to the five crystal faces (011), (112), (022), (013), and (222) of the ZIF-8 crystal, respectively, ¹⁷ and the corresponding diffraction peaks can be found in the XRD pattern of Znq₂@ZIF-8. N₂ adsorption and desorption isotherm tests were performed to determine the BET specific surface area, pore size distribution, and pore volume of the material (Figure 3(c)). The results show that the specific surface area of Znq₂ is small, and its adsorption isotherm is of type II isotherm, which is a typical physical adsorption result of nonporous or macroporous adsorbents. In contrast, ZIF-8, as a porous adsorbent, has a high specific surface area, and the enhancement of the specific surface area of Znq₂@ZIF-8 through the composite is obvious, which is mainly attributed to ZIF-8, and the high specific surface area can lead to more adsorption sites. The adsorption isotherms of ZIF-8 and Znq₂@ZIF-8, on the contrary, exhibit typical type I isotherms, reflecting the phenomenon of microporous filling belonging to microporous adsorbents. ¹⁸ As shown in Figure 3(d), the pore structure of Znq₂ is less significant, which corresponds to the nonporous physical adsorption of the type II isotherm, while ZIF-8 and Znq₂@ZIF-8 are distributed with many microporous structures corresponding to the type I isotherm. The specific surface area, pore volume, and average pore size of the materials are detailed in Table 1.

Figure 3.

(a) FTIR. (b) XRD. (c) N₂ adsorption and desorption isotherms. (d) Pore size distributions.

Table 1.

The BET surface area, pore volume and mean pore size of Znq₂, ZIF-8, and Znq₂@ZIF-8.

Samples	BET surface area (m2 g−1)	Pore volume (cm3 g−1)	Average pore size (nm)
Znq2	1.39	0.01	21.27
ZIF-8	1328.08	0.95	2.43
Znq2@ZIF-8	1105.41	0.67	2.85

Characterization of fluorescence properties

To evaluate the fluorescence properties of Znq₂@ZIF-8, UV-Vis spectra and fluorescence spectra of several components were obtained to determine their fluorescence sources. As shown in Figure 4(a), both Znq₂ and Znq₂@ZIF-8 produced UV absorption near 257 nm, and a broad absorption peak at 360 nm, both of which can be attributed to the UV absorption properties of 8-HQ (8-hydroxyquinoline) and produce a certain degree of absorption redshift with respect to 8-HQ, which is related to the formation of the Zn–O/N coordination structure in Znq₂. The strong absorption band of Znq₂@ZIF-8 at 257 nm and the broad absorption band near 360 nm are attributed to π–π* electron-leap absorption in the quinoline ring and charge transfer from the electron donor O to the acceptor N, respectively. ¹⁵ 8-HQ–based metal coordination compounds have high fluorescence, whereas a single 8-HQ has low fluorescence due to excited-state intramolecular proton transfer (ESIPT). The formation of a coordination structure with metal ions, blocks the ESIPT channel, thus restoring strong fluorescence emission. ¹⁹ Therefore, 8-HQ and Znq₂ have similar fluorescence emission peaks (near 500 nm), Znq₂@ZIF-8 and Znq₂ have stronger fluorescence emission (Figure 4(b)), whereas ZIF-8 does not show fluorescence emission under the same excitation conditions, which suggests that the fluorescence source of Znq₂@ZIF-8 is 8-HQ, which is an LMOF based on guest molecular luminescence. Figure 4(c) shows the fluorescence emission spectra of Znq₂@ZIF-8 at 257, 360 nm, and nearby photoexcitation wavelengths, and the strongest emission corresponding to 495 nm was produced under excitation at 360 nm (λ_ex = 360 nm, λ_em = 495 nm). By comparison, the excitation spectrum of Znq₂@ZIF-8 under fixed emission at 495 nm (E_x@495 nm) and the emission spectrum under fixed excitation at 360 nm (E_m@360 nm) have almost the same coincidence (Figure 4(d)), which is symmetric and produces a large Stokes shift of 135 nm, which can effectively prevent interference caused by the excitation light source during the fluorescence detection process. Fluorescence stability is a prerequisite for the application of fluorescent probes in practice. Therefore, the stability of Znq₂@ZIF-8 was investigated in the range of 0–7 days (Figure 4(e)), and it can be clearly observed that there was no trend of fluorescence intensity decay in this range of time. The green fluorescence characteristics of Znq₂@ZIF-8 were determined by the CIE chromaticity coordinates (Figure 4(f)).

Figure 4.

UV-Vis spectra (a) and fluorescence emission spectra (b) of ZIF-8, 8-HQ, Znq₂, and Znq₂@ZIF-8. Excitation wavelength optimization (c), fluorescence spectra (d), fluorescence stability over 0–7 d (e), and CIE chromaticity diagram of Znq₂@ZIF-8 (f).

Optimization of detection conditions

An alkaline pH may cause serious interference in Fe³⁺ detection, and thus, only the change in the fluorescence intensity of Znq₂@ZIF-8 under acidic conditions was tested. As shown in Figure 5(a), the acidic environment causes a slight decrease in the fluorescence intensity of Znq₂@ZIF-8. The fluorescence of Znq₂@ZIF-8 was quenched to a certain extent after the addition of Fe³⁺, and the maximum quenching degree occurred at pH = 6, so this was used as the optimal detection condition. The fluorescence intensity of Znq₂@ZIF-8 was tested in several common solvents, as shown in Figure 5(b), with the strongest fluorescence intensity occurring in water and a certain degree of blueshift of fluorescence emission occurring in solvents with lower polarity. The reason may be that in the environment of organic solvents with lower polarity, the dipole moment of the material in the ground state is larger than that in the excited state, and the phenomenon of blueshift emission occurs. ²⁰ For optimal emission conditions, the reduction of secondary pollution, and economic considerations, it is more appropriate to use water as the dispersing solvent for the material. Under irradiation with a 365-nm UV lamp, Znq₂@ZIF-8 emits bright green light corresponding to the CIE chromaticity diagram (Figure 5(c)). The fluorescence of Znq₂@ZIF-8 was significantly quenched by the addition of Fe³⁺ and was basically unchanged after 1 min of addition, indicating a rapid response.

Figure 5.

(a) Changes at different pH values. (b) Changes in different solvents. (c) Changes over time.

Fluorescence detection of Fe³⁺

Figure 6(a) shows that the fluorescence intensity of Znq₂@ZIF-8 clearly decreased with increasing added Fe³⁺ concentration under the optimal detection conditions. The data were fitted to the Stern–Volmer equation according to equation (1), the linear fitted equation (Figure 6(b)) was y = 0.00143x + 1.0327, R² = 0.991, which showed a good linear relationship in the range of Fe³⁺ concentrations from 0 to 600 μmol L⁻¹. The lowest limit of detection (LOD) was calculated to be 3.89 μmol L⁻¹ according to the LOD formula in equation (2). The results of fluorescence titration showed that the prepared Znq₂@ZIF-8 had excellent fluorescence detection performance for Fe³⁺ with a high correlation and low detection limit over a wide linear range.

Figure 6.

(a) Fluorescence emission spectra of Znq₂@ZIF-8 with different concentrations of Fe³⁺. (b) Linear fitting of the Stern–Volmer equation. (c) The degree of fluorescence quenching of Znq₂@ZIF-8 by different metal ions. (d) The degree of fluorescence quenching of Znq₂@ZIF-8 by Fe³⁺ in the presence of different metal ions.

Excellent detection selectivity and anti-interference ability are prerequisites for the practical application of fluorescence probes, so a series of metal cation solutions that may cause detection interference were formulated to analyze the selectivity and anti-interference performance of Znq₂@ZIF-8 for Fe³⁺ detection. As shown in Figure 6(c), Fe³⁺ still showed an obvious fluorescence quenching effect at concentrations of various metal ions five times that of Fe³⁺. Although the presence of some interfering ions, such as Cu²⁺ and Ba²⁺, also had a certain quenching effect on Znq₂@ZIF-8, the effect was small compared to that of Fe³⁺. In addition, the anti-interference ability was explored (Figure 6(d)), and the interfering ions had no significant effect on the fluorescence quenching effect of Fe³⁺. The above results indicated that Znq₂@ZIF-8 had strong selectivity and anti-interference ability for the detection of Fe³⁺. As shown in Table 2, compared with some existing fluorescent probes, this method has better detection performance.

Table 2.

Comparison of Znq₂@ZIF-8 with other fluorescent probes.

Fluorescent probes	Detection limit (μmol L−1)	Linear range (μmol L−1)	Reference
CMC/SA/CDs	23.66	50–500	Junior et al. 21
[(CH3)2NH2]·[TbC16O8H6]	18.01	0–100	Zhao et al. 22
Al-MOF	6.62	10–70	Ma et al. 23
MAPbBr3@ZIF-8Co5%	39.00	0–1000	Chen et al. 24
Znq2@ZIF-8	3.89	0–600	This work

Study on the mechanism of fluorescence detection

The fluorescence quenching mechanism of Znq₂@ZIF-8 for Fe³⁺ was analyzed by XRD, FTIR, UV-Vis, and X-ray photoelectron spectroscopy (XPS). As shown in Figure 7(a), the structure of Znq₂@ZIF-8 after the detection of Fe³⁺ can be seen by XRD, which does not have the characteristic diffraction peaks of the two monomer materials, and the crystal structure collapses. According to the FTIR spectrum shown in Figure 7(b), the strength of the Zn–N bond in the material at 420 cm⁻¹ decreased, and competitive coordination of Fe³⁺ may have occurred. In addition, the planar bending vibrational bands at 1147, 997, and 760 cm⁻¹, all attributed to the imidazole ring, exhibited different degrees of weakening of the absorption intensity as well as displacement of the peaks, suggesting that effectively adsorbed Fe³⁺ and interacted with the above-mentioned groups. ²⁵ As shown in Figure 7(c), one of the UV absorption peaks of Fe³⁺ is at 295 nm, which is close to the fluorescence excitation peak of Znq₂@ZIF-8, resulting in a partial energy competition absorption leading to quenching. To further illustrate the quenching mechanism of Fe³⁺ on Znq₂@ZIF-8, the XPS fine spectra of N 1s and Zn 2p before and after the detection of Fe³⁺ were comparatively analyzed. As shown in Figure 7(d) and (e), both N 1s and Zn 2p decreased in intensity after Fe³⁺was detected, indicating that Fe³⁺ strongly interacted with the Zn–N bond and that competitive coordination occurred, which corresponds to the FTIR. ²⁶ The above results suggest that the fluorescence detection mechanism of Znq₂@ZIF-8 for Fe³⁺ should be the collapse of the structure due to competitive coordination, simultaneous, a certain degree of competitive absorption of excitation light energy, which has been extensively reported for metal ion-detecting fluorescent probes.^27,28

Figure 7.

XRD (a) and FTIR (b) of Znq₂@ZIF-8 before and after the detection of Fe³⁺. (c) UV-Vis spectra of Fe³⁺ and excitation spectra of Znq₂@ZIF-8. XPS Fine Spectrum Fitting: N 1s (d) and Zn 2p (e) of Znq₂@ZIF-8 before and after the detection of Fe³⁺.

Detection of Fe³⁺ in practical samples

The spiked recovery method was used to evaluate the feasibility of the present assay in practical applications. The results are shown in Table 3. The method exhibited good recovery and stability, with recoveries ranging from 92.24% to 103.62% and relative standard deviations (RSDs) ⩽8.13% for spiked amounts of Fe³⁺ ranging from 0 to 100 μmol L⁻¹. The above results indicate that the present assay has potential for practical application.

Table 3.

Detection of Fe³⁺ in tap water samples.

Samples	Added (μmol L−1)	Found (μmol L−1)	Recovery (%)	RSD (%)
Tap water	0	0.41	–	–
	5	5.02	92.24	5.91
	10	10.26	98.52	2.02
	20	21.13	103.62	5.33
	50	49.33	97.84	4.24
	100	101.94	101.54	8.13

Conclusion

In this work, Znq₂@ZIF-8 with an octahedral core–shell structure was synthesized by growing ZIF-8 on the surface of Znq₂. Guest molecularly luminescent LMOFs were formed based on the excellent fluorescence emission ability of Znq₂, and Znq₂@ZIF-8 has a tunable luminescent compared with that of ZIF-8 alone. In addition, the prepared composite material of Znq₂@ZIF-8 has a significantly larger specific surface area and richer pore structure relative to Znq₂ alone, in the study of fluorescent probes, the above characteristics are believed to be helpful to enhance the adsorption capacity of the target, which can improve the response speed and reduce the response concentration. Through a series of tests, it was verified that Znq₂@ZIF-8 has good selective recognition and strong anti-interference ability for Fe³⁺ in aqueous environments, as well as a low detection limit and wide linear range. In view of the fluorescence detection effect of Znq₂@ZIF-8, a fluorescence analytical method with simple process, fast response and high detection sensitivity can be established for the detection of low concentrations of Fe³⁺ in water.

Experimental section

Reagents and instruments

8-HQ and 2-methylimidazole were AR grade, and were purchased from Aladdin Biochemical Technology Co. Acetone, methanol, ethanol, ethyl acetate, n-hexane, and ethylene glycol, and various inorganic salts were AR grade and were purchased from Sinopharm Chemical Reagent Co.

Model SU8010 SEM (Hitachi); EDS (Bruker); TENSOR 27 FTIR (Bruker); XRD-6100 XRD (Shimadzu); ASAP 2460 surface area and porosity analyzer (BET, Micromeritics); UV-2550 spectrophotometer (UV-Vis, Shimadzu); Fluorescence spectrometer (FL, Agilent); X-ray photoelectron spectrometer (XPS, Thermo Fisher Scientific).

Synthesis of Znq₂

8-HQ and zinc acetate dihydrate (1 mmol each) were dissolved in 20 mL of ethylene glycol, respectively, stirred at room temperature for 10 min, and allowed to dissolve completely. The zinc acetate solution was mixed with 8-HQ at an injection rate of 240 mL h⁻¹ under room temperature stirring. The mixed solution was then transferred to a polytetrafluoroethylene-lined reactor and reacted at 180 °C for 24 h. After that, the solution was alternately washed by centrifugation with water and ethanol three times to remove unreacted raw materials and then finally dried in an oven at 60 °C for 24 h. ¹⁶

Synthesis of ZIF-8

Zinc nitrate hexahydrate (0.5 mmol) and 2-methylimidazole (5 mmol) were weighed at a molar ratio of 1:10, dissolved in 20 mL of methanol, continuously stirred at 50 °C for 120 min and then washed with methanol three times before being dried in an oven at 60 °C for 24 h. ²⁹

Preparation of Znq₂@ZIF-8

50 mg of the prepared Znq₂ were dispersed in 20 mL of methanol and sonicated for 30 min to make the dispersion uniform. Then, 0.5 mmol of zinc nitrate hexahydrate and 5 mmol of 2-methylimidazole were dissolved in the above dispersion, washed with methanol three times after continuous stirring at 50 °C for 120 min, and then dried in an oven at 60 °C for 24 h.

Fluorescence detection methods

The conditions were optimized by UV-Vis and fluorescence spectrometry for optimal excitation and emission light, fluorescence source, fluorescence stability, optimal solvent use conditions, optimal pH for detection, and detection response time for Fe³⁺ of the probe. The above tests were repeated three times using ultrapure water as the solvent, and the excitation and emission slits were 5 nm.

Different concentrations of Fe³⁺ (0–1000 μmol L⁻¹) were prepared, and the same concentration of Znq₂@ZIF-8 (0.25 g L⁻¹) was prepared. The substance to be detected (Fe³⁺) was mixed with the probe solution (Znq₂@ZIF-8) and then sonicated for 10 min. After the reaction was complete, the fluorescence intensity of the mixture was measured using an FL under the previously optimized conditions (λ_ex = 360 nm, λ_em = 495 nm). The linear range of detection was calculated using the Stern–Volmer equation (1), and later, the lowest LOD of the probe for Fe³⁺ was calculated by the LOD equation (2).

The Stern–Volmer equation:

$$\frac{I_0}{I}=1+K_{\mathrm{SV}}\left[M\right]$$

(1)

LOD calculation equation:

$$\mathrm{LOD}=\frac{3\mathrm{SD}}{K}$$

(2)

where I₀ is the fluorescence intensity of the probe before Fe³⁺ addition, and I is the fluorescence intensity of the probe after Fe³⁺ addition. M denotes the concentration of Fe³⁺ (μmol L⁻¹). K_sv is the constant of the Stern–Volmer equation, LOD is the lowest LOD, K is the slope of the linear fit by equation (1), and SD is the standard deviation of the 10 blank tests.

The selectivity of the probe for detecting Fe³⁺ was analyzed by preparing a solution of interfering ions at five times the concentration of the Fe³⁺ solution, sonicating for 10 min after the addition of the interfering ions, and testing the degree of fluorescence quenching for the probe solution. In addition, a solution of interfering ions at five times the concentration relative to Fe³⁺ was added during the detection of Fe³⁺, and the degree of fluorescence quenching of the probe in the mixed system was measured to test the immunity to interference for Fe³⁺ detection. The degree of fluorescence quenching was expressed as I/I₀, with I/I₀ = 1 indicating the constant fluorescence intensity, I/I₀>1 indicating the fluorescence enhancement, and I/I₀<1 indicating the fluorescence quenching.

According to the established fluorescence analysis method, tap water samples were taken and spiked with different amounts of Fe³⁺ for the recovery test, and the recovery was calculated.