A turn-on fluorescent probe for Cd²⁺ detection in aqueous environments based on an imine functionalized nanoscale metal–organic framework

Abstract

A novel fluorescent probe was synthesized and explored to detect cadmium ions (Cd²⁺) based on an imine functionalized metal–organic framework (MOF). The nanoscale MOF probe (∼100 nm) displayed a turn-on fluorescence in aqueous environments with the addition of Cd²⁺ which could be ascribed to the interaction between the C=N group and specific metal ions, presenting a low-toxicity, highly selective and sensitive detection method for Cd²⁺ with a detection limit of 0.336 μM (37.8 ppb) and a broad linear range of 0–500 μM, which is promising for application in intracellular sensing and imaging of Cd²⁺. The probing mechanism was studied and discussed in detail.

---

Introduction

As one of the extremely toxic and carcinogenic heavy metal ions, the cadmium ion (Cd²⁺) is widely used in many industrial fields such as the production of metal alloys, batteries, electroplating films and the control rods of atomic/nuclear reactors.¹ However, Cd²⁺ can be easily absorbed and accumulated in the human body via food chains with a long biological half-life (about 30 years).² Meanwhile, the intake of Cd²⁺ can damage immune systems and increase the incidence of various diseases such as bone pain, pathological fractures, and even kidney cancer and lung cancer.³ Therefore, the detection of trace Cd²⁺ is critically important.

Up to now, various techniques have been developed for the determination of Cd²⁺ such as atomic absorption spectrometry (AAS),⁴ atom emission spectrometry (AES),⁵ inductively coupled plasma mass spectrometry (ICP-MS),⁶ and synchrotron radiation X-ray spectrometry (XRS).⁷ However, normal restrictions of these analytical methods such as sophisticated instruments, complicated sample preparation and time-consuming processes made cheap, remote and on-site detection far from available. It is of vital importance to develop a simple, rapid and sensitive technique for Cd²⁺ detection. Recently, fluorescent sensors have been considered as a promising and compelling candidate, which can sensitively determine Cd²⁺ ions in different environments through the fluorescence intensity changes or wavelength shifts and avoid the complicated time-consuming identification problem.⁸ A lot of fluorescent Cd²⁺ probes based on small molecules and polymers have been reported.⁹

Among various fluorescent chemosensors, metal–organic frameworks (MOFs) have aroused tremendous attention during the past decades. By taking advantages of their fascinating luminescent properties, size-adjustable pores, and modifiable recognized sites through postsynthetic modification (PSM),¹⁰ a variety of MOFs have been explored for detecting metal ions including Ag⁺,¹¹ Al³⁺,¹² Zn²⁺,¹³ Cu²⁺,¹⁴ Pb²⁺,¹⁵ Hg²⁺ (ref. 16) etc. However, only a few examples of MOF probes capable of detecting Cd²⁺ ions, which were mainly based on luminescent lanthanide MOFs, have been reported up to date.¹⁷ There still remains two challenges in the design of MOF probes for highly selective and sensitive sensing of Cd²⁺ in water samples. One is high accuracy with low deviation of the probe, thus the nanoscale size and stable luminescence are expected. The other is satisfying selectivity and anti-interference ability toward congener elements, for which the sensing mechanism of the Cd²⁺ probe should be based on different binding affinities with various metal ions, instead of selective coordination alone.⁸ Thus, a nanoscale MOF whose recognition site has a better affinity with Cd²⁺ may be a possible solution to above problems. Given the good affinity between cadmium ions with nitrogen atoms, we can utilize the imine group as the recognition site.¹⁸ For this purpose, as the corresponding imine linkers are not commercially available and not thermal stable, the PSM strategy is needed to realize the chemical transformation from amine group.¹⁹

Enlightened by the above thoughts, we herein developed a novel nanoscale fluorescent probe based on imine functionalized MOFs via a condensation reaction of UiO-66-NH₂. As a promising method, PSM of MOFs has become a new tool for exploiting and expanding the unique properties of MOFs. By transforming amine (–NH₂) group of UiO-66-NH₂ to imine (–N=C) group, we synthesized a 3D porous MOF UiO-66-N=CH₂ as a novel Cd²⁺ fluorescent probe (Scheme 1). Due to the strong affinity and the coordination between Cd²⁺ with nitrogen atoms in imine group, the MOF probe exhibited fast response, high selectivity, and excellent sensitivity for Cd²⁺ detection in aqueous solution.

Scheme 1 Schematic illustration of design of MOF-based fluorescent turn-on probe for Cd²⁺. Postsynthetic modification of UiO-66-NH₂ to UiO-66-N=CH₂ via condensation reaction.

Experimental

Chemical reagents

All chemicals and reagents were purchased at least of analytical grade and used without further purification. The standard solutions of metal ions (Cd²⁺, Na⁺, Ca²⁺, Mg²⁺, Cu²⁺, Zn²⁺, Pb²⁺, Al³⁺ and Fe³⁺) were prepared from solving their nitrates in deionized water with the concentration of 0.01 M.

Physical measurements

Powder X-ray diffraction (PXRD) patterns were measured within the 2θ range of 3–50° on X'Pert Pro X-ray diffractometer with Cu-K_α radiation (λ = 1.542 Å) at room temperature. Thermal gravimetric analysis (TGA) were carried out on a Netzsch STA449C system at a heating rate of 10 K min⁻¹ from 30 °C to 800 °C under nitrogen atmosphere in the Al₂O₃ crucibles. Fourier transform infrared (FTIR) spectra were recorded within 4000–400 cm⁻¹ range on Thermo Fisher Nicolet iS10 spectrometer using KBr pellets. The pellets were prepared by mixing the MOF (about 1 mg) with KBr (about 150 mg). The mixture was then carefully grinded and compressed under the pressure of 10 kPa to form a transparent pellet. ¹H NMR spectra were recorded on a Bruker Advance DMX 500 spectrometer using tetramethylsilane (TMS) as an internal standard. The emission and excitation spectra for the samples were recorded by a Hitachi F-4600 fluorescence spectrometer. The morphologies of MOFs were recorded by a S-4800 field-emission scanning electron microscope (FE-SEM). X-ray photoelectron spectroscopy (XPS) was recorded with a Thermo Fisher ESCALAB 250Xi X-ray photoelectron spectrometer with Al Kα X-ray radiation for excitation. The outer luminescent quantum efficiency was determined using an integrating sphere (BaSO₄ coating) from Edinburgh FLSP920 fluorescence spectrometer. The quantum yield can be defined as the integrated intensity of the luminescence signal divided by the integrated intensity of the absorption signal.

Synthesis of UiO-66-NH₂

UiO-66-NH₂ was synthesized using a solvothermal method with a modification of previous reports.²⁰ Anhydrous ZrCl₄ (116.6 mg, 0.5 mmol, 1 equiv.), 2-aminoterephthalic acid (108.8 mg, 0.6 mmol, 1.2 equivs), benzoic acid (1.8320 g, 15 mmol, 30 equivs) and deionized water (36.0 μL, 4 equivs) were dissolved in 10 mL N,N-dimethylformamide (DMF) with ultrasonication for 15 min. Then the homogeneous solution was transferred into a teflon lined autoclave vessel and placed in a 120 °C oven for 24 h. After cooling overnight, the as-synthesized product was rinsed three times with DMF and ethanol to remove unreacted reagents. Then the faint yellow powders were isolated by centrifugation.

Synthesis of UiO-66-N=CH₂

In a typical synthesis, the activated and dried UiO-66-NH₂ (100 mg, 0.34 mmol of –NH₂) was dispersed in one flask (50 mL capacity) with 15 mL of acetone. Six equivalents (per NH₂ functionality, 163 μL) of 40% HCHO (formaldehyde) was added by microlitre pipette. After ultrasonication treated for several minutes, the flask with the mixed solution was heated under reflux and stirring in 60 °C oil bath for 4 h to produce the imine functionalized compound. After repeated centrifugation and rinsing using ethanol for three times, the solid powders was then dried under vacuum at 60 °C overnight to yield the final product, UiO-66-N=CH₂.

Luminescent experiments

The photoluminescence properties of UiO-66-N=CH₂ in different metal cation solutions were studied at 25 °C. In a typical experiment, 2.0 mg powder samples of UiO-66-N=CH₂ were weighed and introduced into 2.0 mL aqueous solution of M(NO₃)_n (Mⁿ⁺ = Cd²⁺, Na⁺, Ca²⁺, Mg²⁺, Cu²⁺, Zn²⁺, Pb²⁺, Al³⁺ and Fe³⁺) and dispersed ultrasonically to form suspensions for luminescent measurements. Then Cd²⁺ solutions at various concentrations are prepared through dilution of standard solution (0.01 M in deionized water). All the luminescent properties were observed upon excitation at 342 nm.

Digestion and analysis by ¹H NMR

¹H NMR spectra of MOFs were recorded on a Bruker Advance DMX 500 spectrometer. Approximately 3 mg of UiO-66-NH₂ and UiO-66-N=CH₂ were digested with sonication in 600 μL of DMSO-d₆ and 30 μL of DCl.

Cell culture and in vitro cytotoxicity assay

Rat pheochromocytoma (PC12) cells were incubated in a humidified incubator (37 °C, 5% CO₂) for three days employing Dulbecco's modified eagle medium (DMEM, Neuronbc) with 10% fetal bovine serum (FBS), and 1% penicillin/streptomycin (P/S, Boster). The toxicity of UiO-66-N=CH₂ was assessed by MTT assays. The crystals were cultured with PC12 cells for 24 h at five concentrations (5, 20, 50, 80, and 100 μg mL⁻¹) in 96-well assay plates (200 μL total volume per well). The MTT solution (50 μL) was added to each well tested and the plates were cultured for 4 h. The medium was removed and the blue dye appeared after the addition of DMSO (150 mL). Absorbance values of samples were determined with a microplate reader at λ = 490 nm. The cell viability was calculated as the ratio of the absorbance of the sample well to that of the control. The results were repeated in triplicate and averaged.

Results and discussion

Crystal synthesis and characterization

As the precursor of the MOF probe in this work, UiO-66-NH₂ was prepared via the solvothermal reaction of the H₂BDC-NH₂ ligand with ZrCl₄ in the solvent DMF (as shown in Scheme 1). UiO-66-NH₂ has attracted intense research interests recently by virtue of the outstanding chemical and thermal stabilities.²¹ On account of the unparticipating amine groups of the ligand in coordination to the Zr₆-octahedra [Zr₆O₄(OH)₄] clusters, these free functional groups are thus potentially available for undergoing various chemical modifications. Therefore, the fluorescent probe UiO-66-N=CH₂ was prepared through postsynthetic modification of UiO-66-NH₂ via aldehyde-amine condensation reaction. In a typical synthesis (Scheme 1), freshly activated UiO-66-NH₂ was treated by formaldehyde under reflux in acetone for 4 h to produce the corresponding imine compound UiO-66-N=CH₂.

The phase purity of UiO-66-NH₂ and UiO-66-N=CH₂ were confirmed by overlapping powder X-ray diffraction (PXRD) patterns of simulated and as-synthesized samples. As shown in Fig. 1a, the PXRD patterns of as-synthesized UiO-66-NH₂ and UiO-66-N=CH₂ matched well with simulated PXRD pattern of UiO-66, which can be featured by two peaks at 7.4° and 8.5°, representing the crystal plane (111) and (200) respectively.²²

Fig. 1 Synthesis and characterization of UiO-66-NH₂ and UiO-66-N=CH₂. (a) PXRD patterns of the simulated UiO-66-NH₂ (black), as-synthesized UiO-66-NH₂ (red) and UiO-66-N=CH₂ (blue). (b) Structure depiction of UiO-66-N=CH₂.

Furthermore, to confirm the thermal stability and structural integrity of two MOFs at elevated temperatures, samples were examined by TGA (Fig. S1†). The TGA curves revealed that UiO-66-NH₂ and the ethanol-treated UiO-66-N=CH₂ will lose entrapped solvent molecules at 100 °C and 80 °C, respectively. Similarly, both two MOFs showed high thermal stability with decomposition temperatures up to 400 °C. The permanent porosity of the prepared UiO-66-NH₂ was determined by N₂ adsorption–desorption isotherm after guest removal, exhibiting a BET surface area of 934.4 m² g⁻¹. The reactive nature of the free –NH₂ groups and the permanent porosity of UiO-66-NH₂ are potentially available for conducting various modifications. And after postsynthetic modification, the synthesized UiO-66-N=CH₂ displayed analogous N₂ adsorption–desorption behaviour (Fig. S2†), which showed a decreased BET surface area of 689.3 m² g⁻¹ as expected, indicating that UiO-66-N=CH₂ still maintains its permanent pores.

Field-emission scanning electron microscopy (FE-SEM) images (Fig. 2) were taken to reveal the morphology of the synthesized UiO-66-NH₂. In this work, 30 equivalents benzonic acid were employed as the modulator, and the as-synthesized faint yellow powders displayed perfect octahedral morphology with particle sizes about 100 nm. It is also worth mentioning here that a smaller particle size is not only conducive to guest molecules transfer within UiO-66-NH₂ crystals, but also critical to be developed as a fluorescent probe. Additionally, these materials were characterized by FTIR spectroscopy (Fig. S3†). The FTIR spectra of UiO-66-NH₂ show two peaks at 3345 and 3456 cm⁻¹ due to the existence of the symmetric and asymmetric stretching of primary amines, which demonstrates that the amino groups are free for interaction in these structures. But FTIR spectroscopy is less revealing with the framework –NH₂ stretching bands dominating, thus obscuring any imine (C=N) and –OH stretches. In addition, unambiguous characterization was provided by liquid ¹H NMR analysis (Fig. S4†). Samples of UiO-66-NH₂ and UiO-66-N=CH₂ were digested and dissolved in dilute DCl/D₂O/DMSO-d₆ solution and then analyzed by ¹H NMR spectroscopy, which confirmed the formation of the imine compound by the appearance of new signals (7.84–7.85 ppm, d, (H)C=N) (Fig. S4d†). It is obvious that the aromatic signals of the amine (7 ppm, d, 1H; 7.34 ppm, s, 1H; 7.77 ppm, d, 1H) (Fig. S4c and d†) does not completely disappear, indicating not full conversion to the imine form.

Fig. 2 FE-SEM images of UiO-66-NH₂ and UiO-66-N=CH₂.

Photoluminescence properties

The photoluminescent (PL) properties of the MOF probe obtained after successful PSM was first investigated at room temperature. The excitation and emission spectra of UiO-66-N=CH₂ are shown in Fig. 3a. It is clearly observed that the excitation spectra collected in the wavelength range of 200–400 nm are dominated by a broad absorption band in ultraviolet region centered at about 342 nm. The excitation spectra suggest that the probe can absorb the energy of ultraviolet light efficiently. Upon 342 nm radiation in water, crystals of UiO-66-N=CH₂ exhibits a broad peak centered at 468 nm.

Fig. 3 (a) Excitation (red line) and emission (blue line) spectra of UiO-66-N=CH₂; (b) the intensities at 468 nm for UiO-66-N=CH₂ dispersed in various metal ion aqueous solutions (500 μM) upon excitation at 342 nm.

To examine the potential of UiO-66-N=CH₂ as a fluorescent probe for metal ions in aqueous environment, the luminescent stability of the sample in aqueous solution has been investigated. During storage in water for 2 weeks, the luminescence intensity of the UiO-66-N=CH₂ suspension shows no significant change (Fig. S5a†), implying the excellent stability of fluorescence in water. Moreover, UiO-66-N=CH₂ also exhibits good pH-independent fluorescence in the pH range of 4–10 (Fig. S5b†), which indicates that the influence of slight fluctuation of pH value on the fluorescent intensity and sensing properties may be neglected.

Sensing of Cd²⁺

High selectivity over competing species is of vital importance for a fluorescent probe. For verifying the sensing selectivity of UiO-66-N=CH₂ of metal ions, the suspension-state luminescent experiments are performed. Different kinds of metal ions (Cd²⁺, Na⁺, Ca²⁺, Mg²⁺, Cu²⁺, Zn²⁺, Pb²⁺, Al³⁺ and Fe³⁺) have been introduced into the system of UiO-66-N=CH₂, forming 500 μM aqueous suspensions after ultrasonication treatment for 1 minute. The luminescent spectra showed that the PL intensity of UiO-66-N=CH₂ suspension is largely dependent on the interacted metal ion. As can be seen in Fig. 3b and S6,† only Fe³⁺ causes the quenching effect on the luminescence intensity of UiO-66-N=CH₂. The addition of Cd²⁺ ion enhanced the luminescence intensity most remarkably by about 3-fold as much as that of the original one, which is the most drastic enhancement effect among all the tested ions. Meanwhile, the luminescence intensity of UiO-66-N=CH₂ showed small increase by Zn²⁺, which is an important disruptor. The rest of other metal ions had a negligible effect on the luminescence of UiO-66-N=CH₂. The above results prove that UiO-66-N=CH₂ has a high selectivity for the specific recognition of Cd²⁺ in aqueous environment.

In Fig. 4, we measured the luminescence responses of UiO-66-N=CH₂ (1 mg mL⁻¹) immersed in Cd²⁺ aqueous solutions with different concentrations. It is obvious that the luminescence intensity of UiO-66-N=CH₂ incorporated with Cd²⁺ displays a strong dependence on the concentration of Cd²⁺ ion. As can be seen in Fig. 4a, the emission intensity of UiO-66-N=CH₂ increases gradually with the increase of Cd²⁺ concentration from 0 to 500 μM. Quantitatively, the enhancing effect can be analyzed by the Stern–Volmer equation:

I/I₀ = 1 + K_SV[Cd²⁺] (1)

where I₀ and I are the luminescence intensity of the suspension at 468 nm in the absence and presence of Cd²⁺, respectively, K_SV represents the Stern–Volmer quenching constant, and [Cd²⁺] is the concentration of Cd²⁺. There is a good linear relationship (correlation coefficient R² = 0.999) between I/I₀ and [Cd²⁺] in a wide concentration range (Fig. 4b), indicating that the enhancing effect of Cd²⁺ on the luminescence of UiO-66-N=CH₂ agrees the Stern–Volmer mode well. The K_SV value is calculated to be 3.81 × 10³ M⁻¹, which reveals a strong enhancing effect on the UiO-66-N=CH₂ luminescence. From the slope of the fitting line (slope) and the standard deviation (SD) for eleven replicating fluorescence measurements of blank solutions (Fig. S7†), the detection limit (3SD/slope) of UiO-66-N=CH₂ towards Cd²⁺ was calculated to be 0.336 μM (37.8 ppb).

Fig. 4 (a) The luminescent emission spectra (excited at 342 nm) of UiO-66-N=CH₂ (1 mg mL⁻¹) in Cd²⁺ solutions with different concentrations. (b) Stern–Volmer plots of I/I₀ versus Cd²⁺ concentration and the fitting line in water (R² = 0.999; K_SV = 3.81 × 10³ M⁻¹).

To evaluate the cytotoxicity of UiO-66-N=CH₂, rat pheochromocytoma (PC12) cells were selected for in vitro studies using the MTT assay, considering the wide use of PC12 cells in assessing the cytotoxicity of nanomaterials.²³ The MTT assay was carried out by incubating PC12 cells with UiO-66-N=CH₂ for 24 h at various concentrations ranging from 5 to 100 μg mL⁻¹. As revealed in Fig. S8,† the cell viability decreased slightly with the increase of the UiO-66-N=CH₂ concentration, but the cell viability was still higher than 80% even at 100 μg mL⁻¹, meaning that UiO-66-N=CH₂ was a biocompatible probe with negligible cytotoxicity.

The enhancing effect on luminescence of UiO-66-N=CH₂ by Cd²⁺ ions can be ascribed to the interaction between Cd²⁺ ions and the C=N group within UiO-66-N=CH₂ which facilitates the efficiency of deactivation from excited state to ground state. To explore the possible mechanism for such luminescence enhancement by Cd²⁺, N 1s and Cd 3d X-ray photoelectron spectroscopy (XPS) studies were carried out on probe and Cd²⁺-incorporated probe. The N 1s peak from free imine nitrogen atoms at 399.5 eV in H₂O is shifted to 400.2 eV on the addition of Cd²⁺ (Fig. 5), indicating the weak binding between Cd²⁺ and imine nitrogen atoms in Cd²⁺-incorporated probe. Such coordination binding of Cd²⁺ with the imine group of the ligand can reduce non-radiative transition and decrease energy loss, thus resulting in a ligand-centered charge transfer which may lead to the enhancing effect. The peak at 413.2 eV in the N 1s XPS spectra for Cd²⁺-incorporated probe is attributed to Cd 3d, as for the other peak at 406.6 eV, it should be ascribed to the signal of NO₃⁻ counterions.²⁴ As a control experiment, we also investigated the fluorescent response of UiO-66-NH₂ toward Cd²⁺. After immersing into the Cd²⁺ solution (500 μM), no obvious change was found from the luminescence intensity of UiO-66-NH₂, indicating the important role of the C=N group in Cd²⁺ sensing.

Fig. 5 XPS spectra of the original probe (blue) and Cd²⁺-incorporated probe (red) activated in 10 mM aqueous solution of Cd(NO₃)₂.

As a luminescent probe, UiO-66-N=CH₂ is not only highly selective and sensitive for Cd²⁺ detection, but also water-stable and nano-scale, which enable the probe to be applicable for Cd²⁺ sensing in aqueous and intracellular environments.

Conclusions

In summary, a novel turn-on fluorescent probe based on imine functionalized nano MOF for Cd²⁺ detection has been synthesized through a post-synthetic approach. The MOF probe displayed high sensitivity (detection limit, 0.336 μM), broad linear range (0–500 μM) and excellent selectivity toward Cd²⁺ over other relevant metal ions. Moreover, it is water-stable, nano-scale and low-toxic. All these indicate the potential of UiO-66-N=CH₂ for intracellular Cd²⁺ turn-on sensing and imaging.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors gratefully acknowledge the financial support for this work from the National Natural Science Foundation of China (Nos. 51372221, 51472217, 51432001, 51632008, and U1609219), and Zhejiang Provincial Natural Science Foundation of China (No. LZ15E020001).

Notes and references

1. (a) A. M. S. Mendes, G. P. Duda, C. W. A. Nascimento and M. O. Silva, Sci. Agric., 2006, 63, 328; (b) M. Lenzen, Energy Convers. Manage., 2008, 49, 2178.
2. (a) L. T. Lu, I. C. Chang, T. Y. Hsiao, Y. H. Yu and H. W. Ma, Environ. Sci. Pollut. Res., 2007, 14, 49; (b) M. Zhang, Y. Liu and B. Ye, Analyst, 2011, 137, 601.
3. (a) Y. Itokawa, K. Nishino, M. Takashima, T. Nakata, H. Katio, E. Okamoto, K. Daijo and J. Kawamura, Environ. Res., 1978, 15, 206; (b) S. Satarug, J. R. Baker, S. Urbenjapol, M. Haswell-Elkins, P. E. Reilly, D. J. Williams and M. R. Moore, Toxicol. Lett., 2003, 137, 65.
4. A. N. Anthemidis, Microchim. Acta, 2008, 160, 455.
5. A. C. Davis, C. P. Calloway Jr and B. T. Jones, Talanta, 2007, 71, 1144.
6. E. A. Hutton, J. T. van Elteren, B. Ogorevc and M. R. Smyth, Talanta, 2004, 63, 849.
7. E. Marguí, I. Queralt and M. Hidalgo, J. Anal. At. Spectrom., 2013, 28, 266.
8. (a) T. Xia, T. Song, G. Zhang, Y. Cui, Y. Yang, Z. Wang and G. Qian, Chem.–Eur. J., 2016, 22, 18429; (b) J. Yin, T. Wu, J. Song, Q. Zhang, S. Liu, R. Xu and H. Duan, Chem. Mater., 2011, 23, 4756.
9. (a) T. Cheng, T. Wang, W. Zhu, X. Chen, Y. Yang, Y. Xu and X. Qian, Org. Lett., 2011, 13, 3656; (b) Z. Liu, C. Zhang, W. He, Z. Yang, X. Gao and Z. Guo, Chem. Commun., 2010, 46, 6138; (c) L. Xu, M. He, H. Yang and X. Qian, Dalton Trans., 2013, 42, 8218; (d) H. Un, C. Huang, J. Huang, C. Huang, T. Jia and L. Xu, Chem.–Asian J., 2014, 9, 3397; (e) Y. Ren, N. Wu, J. Huang, Z. Xu, D. Sun, C. Wang and L. Xu, Chem. Commun., 2015, 51, 15153; (f) Z. Xu, G. Li, Y. Ren, H. Huang, X. Wen, Q. Xu, X. Fan, Z. Huang, J. Huang and L. Xu, Dalton Trans., 2016, 45, 12087; (g) C. Huang, L. Xu, J. Zhu, Y. Wang, B. Sun, X. Li and H. Yang, J. Am. Chem. Soc., 2017, 139, 9459.
10. (a) Y. Cui, Y. Yue, G. Qian and B. Chen, Chem. Rev., 2012, 112, 1126; (b) Y. Cui, B. Li, H. He, W. Zhou, B. Chen and G. Qian, Acc. Chem. Res., 2016, 392, 86.
11. J. Hao and B. Yan, J. Mater. Chem. A, 2014, 2, 18018.
12. X. Shi, H. Wang, T. Han, X. Feng, B. Tong, J. Shi, J. Zhi and Y. Dong, J. Mater. Chem., 2012, 22, 19296.
13. Q. Tang, S. Liu, Y. Liu, J. Miao, S. Li, L. Zhang, Z. Shi and Z. Zheng, Inorg. Chem., 2013, 52, 2799.
14. B. Chen, L. Wang, Y. Xiao, F. R. Fronczek, M. Xue, Y. Cui and G. Qian, Angew. Chem., Int. Ed., 2009, 121, 508.
15. L. Li, Q. Chen, Z. Niu, X. Zhou, T. Yang and W. Huang, J. Mater. Chem. C, 2016, 4, 1900.
16. X. Xu and B. Yan, J. Mater. Chem. C, 2016, 4, 1543.
17. (a) J. Hao and B. Yan, Chem. Commun., 2015, 51, 7737; (b) C. Liu and B. Yan, J. Colloid Interface Sci., 2015, 459, 206; (c) X. Xu and B. Yan, Sens. Actuators, B, 2016, 222, 347.
18. (a) H. Zhu, J. Fan, B. Wang and X. Peng, Chem. Soc. Rev., 2015, 44, 4337; (b) H. J. Jung, N. Singh and D. O. Jang, Tetrahedron Lett., 2008, 49, 2960.
19. S. M. Cohen, Chem. Rev., 2012, 112, 970.
20. (a) M. Kandiah, M. H. Nilsen, S. Usseglio, S. Jakobsen, U. Olsbye, M. Tilset, C. Larabi, E. A. Quadrelli, F. Bonino and K. P. Lillerud, Chem. Mater., 2010, 22, 6632; (b) A. Schaate, P. Roy, A. Godt, J. Lippke, F. Waltz, M. Wiebcke and P. Behrens, Chem.–Eur. J., 2011, 17, 6643.
21. (a) G. W. Peterson, J. J. Mahle, J. B. Decoste, W. O. Gordon and J. A. Rossin, Angew. Chem., Int. Ed., 2016, 128, 6343; (b) L. Shen, W. Wu, R. Liang, R. Lin and L. Wu, Nanoscale, 2013, 5, 9374; (c) S. S. Nagarkar, T. Saha, A. V. Desai, P. Talukdar and S. K. Ghosh, Sci. Rep., 2014, 4, 7053.
22. (a) X. Zhang, Q. Hu, T. Xia, J. Zhang, Y. Yang, Y. Cui, B. Chen and G. Qian, ACS Appl. Mater. Interfaces, 2016, 8, 32259; (b) J. H. Cavka, S. Jakobsen, U. Olsbye, N. Guillou, C. Lamberti, S. Bordiga and K. P. Lillerud, J. Am. Chem. Soc., 2008, 130, 13850.
23. K. Jiang, L. Zhang, Q. Hu, D. Zhao, T. Xia, W. Lin, Y. Yang, Y. Cui, Y. Yang and G. Qian, J. Mater. Chem. B, 2016, 4, 6398.
24. S. Liu, Z. Xiang, Z. Hu, X. Zheng and D. Cao, J. Mater. Chem., 2011, 21, 6649.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7ra11162b

---