Poly(dopamine) assisted in situ fabrication of silver nanoparticles/metal–organic framework hybrids as SERS substrates for folic acid detection

Abstract

In this contribution, a new in situ strategy for the preparation of AgNPs/MIL-101(Cr) hybrids as SERS substrates for folic acid detection was reported. In Tris–HCl buffer solution (pH = 8.5), poly(dopamine) (PDA) could be formed by self-polymerization of dopamine (DA), and was then effectively attached onto the surface of MIL-101(Cr) for further reduction of Ag⁺. By this strategy, AgNPs could be efficiently anchored on MIL-101(Cr) to form AgNPs/MIL-101(Cr) hybrids, which exhibited strong enrichment of folic acid (FA) due to electrostatic interaction, and the anchored AgNPs greatly enhanced the Raman signals of FA. Thus, the as-prepared hybrids have been used as excellent surface enhanced Raman scattering (SERS) substrates for the sensitive and selective detection of FA. Under the optimal experimental conditions, the SERS method shows good linearity in the range from 0.5 μM to 25 μM for FA with a correlation coefficient of 0.990, the detection limit was calculated to be 0.3 μM ± 0.02 μM. Moreover, it was successfully applied to detect FA in real samples with satisfactory results. These results demonstrated that the proposed SERS system has great practicability for chemical and biological assay applications, and the AgNPs/MIL-101(Cr) hybrids also offers potential for targeting cancer cells.

---

1. Introduction

As an ultrasensitive, nondestructive and important analytical technique, surface-enhanced Raman scattering (SERS) has been wildly used in biomedical detection.^1,2 The Raman intensity of the target molecules could increase by orders of magnitude when they on or near the surface of specific SERS substrates, wherein the fabrication of SERS active substrates are crucial. Among the various reported SERS active substrates, silver with surface-plasmon-resonance effect exhibits excellent performance in SERS activity.^3–5 For example, Zhan and co-workers developed cysteamine-modified silver nanoparticle aggregates, and used them for pentachlorophenol sensing by SERS.⁶ And to obtain SERS substrates with larger enhancement factors as well as good reproducibility, considerable attention has been paid to assembling silver with other nanomaterials. For instance, it has been reported that the silver nanoparticle could combine with graphene oxide through an in situ method or self-assembly method. Since graphene oxide with a large surface area and unique electronic structure could concentrate target molecules through π–π stacking or electrostatic adsorption, the resulting AgNPs/GO hybrids exhibit improved sensitivity for bioanalysis compared with the SERS substrates of AgNPs.^7,8 To further facilitate and broaden the SERS applications in biomedical systems, fabrication of new AgNPs-based multifunctional substrate composites is essential.

As a new type of porous materials, metal–organic frameworks (MOFs) also posses high surface area, and compared with graphene oxide, MOFs with intriguing properties had greater practicability in analytical application.^9–12 For instance, our group has found that iron-containing MOFs such as Fe–MIL-88, Fe–MIL-88NH₂ possess intrinsic peroxidase-like activity and could be used for sensing various bimolecular.^13–17 Recently, we developed an efficient SERS substrate by in situ synthesis of silver nanoparticles (AgNPs) on the surface of MIL-101(Fe), and based on the peroxidase-like activity of MIL-101(Fe), the hybrid material was used for ultra sensitive SERS detection of dopamine.¹⁸ However, in our experiment, we found that the Fe–MOFs such as Fe–MIL-88 or MIL-101(Fe) could not resistant to violent acid (pH < 2) or alkali (pH > 10) conditions, they collapse slowly when stored in PBS buffer, and as reported in the previous work,¹⁹ the nanosized MIL-101(Fe) particles readily decompose under basic conditions to presumably form amorphous iron hydroxide (oxide) phases. Unlike with Fe–MOFs, another well-known MOFs, MIL-101(Cr), with ultrahigh surface area, high positive charge, and much higher stability in violent acid or alkali conditions as well as in water condition has been wildly applied in adsorption and removal of anionic dye from aqueous solution.^20–22 Moreover, we found that MIL-101(Cr) could adsorb DNA with negtive charge and can be used as a low background platform for label-free DNA detection with the interacting-dye SYBR Green I as the fluorescence indicator.²³

In order to fabricate a novel SERS substrate with much higher stability which combines the excellent adsorption performance of MIL-101(Cr) and the strong SERS activity of the AgNPs, we thus designed a hybrid material AgNPs/MIL-101(Cr) by in situ synthesis of AgNPs on the outer surface of MIL-101(Cr). In our strategy (Scheme 1), the biopolymer-poly(dopamine) (PDA) was formed on MIL-101(Cr) through the self-polymerization of dopamine (DA) in aqueous solution, making Ag⁺ be directly reduced by the PDA on the surface of MIL-101(Cr) to engineer AgNPs/MIL-101(Cr) complex structure.^24,25 The obtained AgNPs/MIL-101(Cr) was further used for SERS detection of folic acid (FA) since it could capture FA through electrostatic interaction and get them sufficiently close to the surface of AgNPs. As a result, the strong inherent vibrations of FA could be observed in the SERS spectra.

Scheme 1 Schematic illustration of the preparation process for the AgNPs/MIL-101(Cr) and its application as SERS substrate for the detection of folic acid.

2. Experimental

2.1 Instrumentation

Scanning electron microscopy (SEM) images were captured using an S-4800 scanning electron microscope (Hitachi, Japan). Transmission electron microscopic (TEM) characterization was performed on an FEI Tecnai G2 F20 TEM instrument (FEI, America). X-ray photoelectron spectrometry (XPS) analyses were carried out on a Thermo Escalab 250 Xi X-ray photoelectron spectrometer using an Al Kα source (hν = 1486.6 eV). Powder X-ray diffraction (PXRD) patterns were collected on an XD-3 X-ray diffractometer with Cu Kα radiation (λ = 1.5406 Å) in the range of 5–30θ at a scan rate of 2.0° min⁻¹ (Purkinje, China). Fourier transform infrared (FTIR) spectra were recorded on a FITI-8400 (Shi-madzu, Japan) in the range of 4000–500 cm⁻¹ using the KBr disk method. Zeta-potential measurements were performed using a Zetasizer Nano-ZS90 instrument (Malvern Inc). Raman spectra were recorded using a LabRam HR 800 spectrometer (HORIBA Jobin Yvon, France).

2.2 Materials and reagents

The chromium(III) nitrate nonahydrate (Cr(NO₃)₃·9H₂O), terephthalic acid (H₂BDC) and hydrofluoric acid (HF, 48%) (Aladdin, Shanghai, China) were used to prepare MIL-101(Cr). Folic acid was purchased from Beijing Dingguo Changsheng Biotech Co., Ltd. Dopamine hydrochloride (DA) was purchased from Sigma-Aldrich (St. Louis, MO). Urine samples were obtained from three healthy volunteers of the Southwest University, and the use of human urine was approved by the University of Southwest's Institutional Review Board. A signed individual written informed consent agreement was obtained from the participants before beginning the work on this study, and the research didn't involve outside of our country of residence. All analytical reagents were used without further purification, and all solutions were prepared using ultra-pure water (18.2 MΩ).

2.3 Synthesis of AgNPs/MIL-101(Cr) hybrid

MIL-101(Cr) (200–400 nm) was synthesized according to our previous report.²² the anchored AgNPs were prepared via in situ synthesis strategy by utilizing PDA as reductant. Typically, 10 mg of MIL-101(Cr) was suspended in 9 mL of Tris–HCl buffer solution (pH 8.5), and then 1 mL of DA solution (20 mg mL⁻¹) was added dropwise under a stirring speed of 700 rpm at room temperature. After stirring for 40 min, the color of the solution changed to dark blue, indicating the formation of PDA. Then the products (PDA/MIL-101(Cr)) were isolated and washed 3 times with H₂O by centrifugation and re-dispersed in H₂O (9 mL). Afterwards, 0.5 M K₂CO₃ was used to tune the pH of the solution to 6.7. Finally, 0.25 mL of AgNO₃ (0.4 M) was added dropwise. After further stirring for 3 h, the brown product (AgNPs/MIL-101(Cr)) was isolated by centrifugation and washed with H₂O three times, and the obtained AgNPs/MIL-101 was dried in a vacuum freezing drying oven for further use.

2.4 Enhancement factor (EF) calculation of Rh 6G on the surface of hybrid AgNPs/MIL-101(Cr)

For the SERS enhancement factor measurement, the sample was prepared by drop-casting 2 μL of AgNPs/MIL-101(Cr) hybrids aqueous solution on the silicon substrate and dried in atmosphere. Then, 10 μL of Rh 6G (1 × 10⁻⁵ M) aqueous solutions was dropped on the substrate and dried. Measurements at different positions were carried out for each sample to test the reproducibility.

2.5 General procedure for FA detection

For the determination of FA, 50 μL of 0.2 M BR buffer (pH 1.9), 150 μL 1 mg mL⁻¹ AgNPs/MIL-101(Cr) aqueous solution were added to a 1.5 mL EP vial. Afterwards, the appropriate amount of FA or urine was added and the final mixture was diluted to 500 μL with ultra-pure water (18.2 MΩ). After mixing thoroughly, the solution was incubated under room temperature for 80 min and then used for FA measurement by monitoring the Raman shifts at 1595 cm⁻¹. Parameter settings were as follows: laser wavelength, 532 nm; power, 28 mW; lens, 50× objectives; acquisition time, 20 s.

3. Results and discussions

3.1 Characterization of the AgNPs/MIL-101(Cr) hybrids

In the FTIR spectra of MIL-101(Cr) (Fig. 1), the characteristic peaks located at 1624 and 1404 cm⁻¹ could be assigned to benzene skeleton vibration, and the broad band at 3444 cm⁻¹ was associated with the stretching vibrations of the O–H from the surface adsorbed water. As observed, the FTIR spectrum of the PDA/MIL-101(Cr) and AgNPs/MIL-101(Cr) show the stretching vibrations of O–H at about 3429 cm⁻¹, which is red-shifted by 15 cm⁻¹ with respect to that of the initial MIL-101(Cr) (3444 cm⁻¹), indicating that hydrogen bonding interaction between MIL-101(Cr) and PDA was occurred. Moreover, the bending vibrations of N–H at 1654 cm⁻¹ were both observed in the infrared spectra of PDA/MIL-101(Cr) and AgNPs/MIL-101(Cr), the above observations suggested that PDA successfully adsorbed on the surface of MIL-101(Cr). The presence of the peak at 1589 cm⁻¹ in the spectra of AgNPs/MIL-101(Cr) was due to oxidation of residual hydroxyl groups in PDA into quinone, at the same time, the Ag ions could be reduced to form AgNPs via accepting electrons.

Fig. 1 The FTIR spectra of MIL-101(Cr), PDA/MIL-101(Cr) and AgNPs/MIL-101(Cr).

SEM and TEM characterization clearly showed that the AgNPs are well-distributed on the surface of MIL-101(Cr) (Fig. 2). Besides, powder X-ray diffraction (XRD) pattern of AgNPs/MIL-101(Cr) showed the main diffraction peaks of MIL-101(Cr) at low angles, which indicated that the hybrid preserved the crystalline characters of the parent MIL-101(Cr) (Fig. 3), and the diffraction peaks centered at around 38.2°, 44.2°, 64.4°, and 77.4° also present in the typical XRD pattern of the as-prepared AgNPs/MIL-101(Cr), which correspond to the Ag(111), Ag(200), Ag(220), and Ag(311) lattice planes, respectively, indicating that AgNPs on the surface of MIL-101(Cr) were highly crystalline.⁵ As shown, the AgNPs diffraction peaks in the XRD patterns of AgNPs/MIL-101(Cr) are weak, probably because the content of AgNPs on the surface of MIL-101(Cr) was relatively low. The chemical composition and electronic structures of the as-prepared AgNPs/MIL-101(Cr) were analyzed by XPS. As shown in Fig. 4, it can be clearly observed that the peaks of Cr, C, O and Ag exist in the sample of AgNPs/MIL-101(Cr), and the emergence of intense Ag 3d peaks proved the deposition of AgNPs on the surface of MIL-101(Cr). The high-resolution XPS spectrum of Ag 3d orbital region shows the binding energies of Ag 3d_5/2 and Ag 3d_3/2 peaks at 368.4 and 374.4 eV, which corresponds to the Ag⁰ state.⁵ These results indicate that AgNPs were successfully anchored on the surface of MIL-101(Cr).

Fig. 2 SEM images (A) and TEM images (B) of AgNPs/MIL-101(Cr).

Fig. 3 PXRD pattern of MIL-101(Cr), AgNPs/MIL-101(Cr). Insert shows the magnification of the XRD pattern of AgNPs/MIL-101(Cr) from 30–80θ.

Fig. 4 (A) XPS survey spectra of AgNPs/MIL-101(Cr); (B) the enlarged Ag 3d region.

3.2 SERS property of the as-prepared AgNPs/MIL-101(Cr)

Since AgNPs can enhance Raman signal through the electromagnetic enhancement mechanism, the as-prepared AgNPs/MIL-101(Cr) hybrids can be potentially used as SERS substrates for sensitive analysis. To estimate the SERS activity of the AgNPs/MIL-101(Cr), Rh 6G was chosen as a probe molecule because it has been well characterized by SERS and most of the prominent Raman bands have been assigned.^26,27 Fig. 5a and b show the normal Raman spectra of the solid Rh 6G and the SERS spectra of the Rh 6G (1 × 10⁻⁵ M) adsorbed on the surfaces of AgNPs/MIL-101(Cr). As shown, when adsorbed on the surfaces of AgNPs/MIL-101(Cr), the characteristic bands of Rh 6G shift 10–30 cm⁻¹, we propose that this shift might be attributed to hydrogen-bonds or π–π stacking interaction between the AgNPs/MIL-101(Cr) and Rh 6G. To estimate the enhancement force of AgNPs/MIL-101(Cr) hybrids for Rh 6G, band at 1641 cm⁻¹ were selected as the representation for estimating the SERS enhancement factor (EF) values of Rh 6G.

Fig. 5 (a) Normal Raman spectrum of solid Rh 6G, and (b) SERS spectrum of Rh 6G (1 × 10⁻⁵ mol L⁻¹) on the AgNPs/MIL-101(Cr) substrate. Laser wavelength, 532 nm; power, 0.3 mW; lens, 50× objective; acquisition time, 20 s.

The SERS EF value was calculated using the following equation²⁶

EF = (I_SERS/N_SERS)/(I_NR/N_NR)

herein, I_SERS stands for the intensities of the vibrational mode in the SERS spectra of Rh 6G and I_NR stands for the normal Raman spectra of solid Rh 6G. Values of I_SERS and I_NR can be obtained from the SERS spectra and Raman spectra. N_SERS and N_NR are the number of Rh 6G molecules illuminated by the laser focus spot under SERS and normal Raman conditions, respectively. Supposing the molecules were uniformly dispersed on the substrates, the density of Rh 6G on the film was assumed to be 1 × 10⁻⁵ mol L⁻¹ × 2 μL × N_A/4.9 mm² (4.9 mm² stands for the surface area of the substrate), namely 2.4 × 10¹² molecules per mm², the laser spot has a diameter of 1 μm and the surface area is about 7.9 × 10⁻⁷ mm², so the N_SERS value is 1.94 × 10⁶. The density of solid Rh 6G is 1.26 g cm⁻³, the laser spot diameter is 1 μm, and the penetration depth is about 2 μm, thus the N_NR had a value of 2.6 × 10⁹ in the detected solid sample area. All the spectra were normalized for laser power and acquisition time. EF at the band at 1650 cm⁻¹ of Rh 6G was estimated to be 5.2 × 10³ for the AgNPs/MIL-101(Cr) hybrids.

3.3 AgNPs/MIL-101(Cr) as SERS substrate for FA detection

Folic acid (FA), known as vitamin B9, is widely distribute in fresh fruits, vegetables and meat. As an important water-soluble vitamin, FA plays an essential role in the synthesis of proteins, nucleic acids and participates in amino acids metabolism. Lack of FA may cause megaloblastic anemia, leucopenia. For pregnant women, adequate FA may help to prevent some birth defects such as infant cleft palate, low-body weight and congenital malformation.²⁸ With respect to that, FA detection is necessary and significant in preventing diseases related to FA deficiency.²⁹

A previous work has reported that graphene oxide/Ag nanoparticle hybrids could be formed according to a self-assembly procedure, and based on the electrostatic interaction between FA and the GO/PDDA/AgNPs hybrids, a sensitive SERS method for FA detection was established.⁸ Comparing with the self-assembled method, our in situ synthesis method proposed in this paper is more simple, and through this method, AgNPs could firmly anchored onto MIL-101(Cr) to form AgNPs/MIL-101(Cr) hybrid with stable SERS activity, thus when applied to SERS detection, the measurement result is more accurate with better reproducibility. And also the as-prepared AgNPs/MIL-101(Cr) combined the numerous Raman hot spots between the high-density AgNPs and the excellent adsorption performance of MIL-101(Cr), when applied in SERS detection, high sensitivity could be obtained. As shown in Table 1, the obtained AgNPs/MIL-101(Cr) was positively charged, and the significant decrease of the zeta-potential (from +27.3 mV to −11.4 mV) also proved the strong electrostatic interaction between AgNPs/MIL-101(Cr) and FA, based on these, we thus constructed a label-free SERS method for determination of FA by using the AgNPs/MIL-101(Cr) as substrates. As shown in Fig. 6, after loading 2.5 × 10⁻⁵ M folic acid molecules on the AgNPs/MIL-101(Cr) hybrids, intense Raman signals were detected, main vibrations of folic acid shown in the SERS spectrum were confirmed according to the reported work.³⁰ These characteristic peaks providing the information of the molecules could be used to identify folic acid in the samples. As for free FA solution, no Raman signals were observed even though the concentration is 80-folds higher than that of AgNPs/MIL-101(Cr). Since the SERS intensity of FA in the presence of AgNPs/MIL-101(Cr) was much stronger than that of the free FA solution, we propose that the conspicuous differences may result from the high local concentration of FA adjacent to the surface of AgNPs/MIL-101(Cr) which has high adsorption performance and abundant AgNPs.

Table 1 Zeta potential of MIL-101(Cr), AgNPs/MIL-101(Cr), FA and AgNPs/MIL-101(Cr) + FA

Sample name	Zeta-potential
MIL-101(Cr)	+40.5
AgNPs/MIL-101(Cr)	+27.3
FA	−15.6
AgNPs/MIL-101(Cr) + FA	−11.4

---

Fig. 6 Raman spectrum of FA (black) and AgNPs/MIL-101(Cr) (red); SERS spectra of FA on the AgNPs/MIL-101(Cr) substrate (blue). Laser wavelength, 532 nm; power, 28 mW; lens, 10× objective; acquisition time, 20 s.

To obtain the optimal condition for the determination of FA, the strongest peak located at 1595 cm⁻¹ was chosen as the signature to investigate the effect of pH and reaction time on the FA detection. As shown in Fig. 7A and B, when AgNPs/MIL-101(Cr) was used as SERS substrate, pH 1.9, 80 min were set as the optimal acidity, time respectively. Besides, the ν (C–C) ring breathing of PDA on the surface of AgNPs/MIL-101(Cr) at about 1600 cm⁻¹ may overlap the signature band of FA and disturb the detection of FA in low concentration. Therefore, the concentration of the AgNPs/MIL-101(Cr) was optimized to get stronger FA SERS spectra with a lower intensity of the predominant PDA band at 1600 cm⁻¹. Fig. 7C show SERS spectra of 5 μM FA in a series of concentration of AgNPs/MIL-101(Cr), as shown, SERS intensity of AgNPs/MIL-101(Cr) and AgNPs/MIL-101(Cr) + FA significantly enhanced with the increasing concentrations of AgNPs/MIL-101(Cr), to reduce the background interference from PDA and further improve sensitivity, 0.3 mg mL⁻¹ was chosen as the optimum concentration of AgNPs/MIL-101(Cr).

Fig. 7 Optimization of the reaction conditions pH (A), reaction time (B) and AgNPs/MIL-101(Cr) concentration (C) for 5 μM FA detection. Laser wavelength, 532 nm; power, 28 mW; lens, 10× objective; acquisition time, 20 s.

3.4 The effect of coexisting substances in urine on determination of FA

To investigate the selectivity of this SERS system for FA detection, the effect of 16 other kinds of common biomolecules as well as some probable ions present in human urine were investigated. As shown in Fig. 8, although the concentration of coexisting substances was 10-fold higher than that of FA, the obtained signals for FA had not changed much, indicading that the common coexisting substances in urine did not interfere with the determination of FA, we propose that the reasons are as follows: the Raman peak used for FA determination located at 1595 cm⁻¹, which is derived from phenyl ring deformations of FA. Since most of the coexistent substances in urine such as AA, UA, urea etc. have no phenyl in their structure, they could hardly affect the signature of FA at 1595 cm⁻¹ even though they may adsorb on the SERS substrate through electrostatic interaction. However, if those negatively changed complex matrixes exist in urine also exhibit Raman signature at 1595 cm⁻¹, we speculate that they may interfere the FA determination.

Fig. 8 SERS intensity of FA (1595 cm⁻¹) in the absence and presence of other analytes. The concentration of other biomolecules and ions that probably exit in urine are 50 μM, which are 10-folds higher than that of FA (5 μM).

3.5 Detection of FA in human urine

Under the optimal assay conditions, the SERS intensity increased with the increase of FA concentration (Fig. 9). The SERS peak at 1595 cm⁻¹ was used for the quantification, which revealed a good linear SERS response ranging from 0.5 μM to 25 μM of FA (R² = 0.990), and the limit of detection (LOD) was calculated to be 0.3 μM ± 0.02 μM. To further confirm the precision and reliability of the proposed method in real sample analysis, standard additions method was applied to detect the level of FA in human urine and the results were shown in Table 2. The good recoveries (97–101%) of known amount FA definitely demonstrated the accuracy and reliability of the present method for detecting FA in human urine.

Fig. 9 (A) SERS spectra of different concentrations of FA in water [the concentrations from bottom to top (a–j) are 0, 0.5 μM, 1 μM, 3 μM, 5 μM, 8 μM, 10 μM, 15 μM, 20 μM, 25 μM]. (B) Linear calibration curve corresponding to the Raman signal at 1595 cm⁻¹ for varying concentrations of FA (from 0.5 μM to 25 μM).

Table 2 Recovery test of FA in diluted urine sample

Sample	Amount found (μM)	Added (μM)	Found (μM)	Recovery (%)	R.S.D. (%)
Urine 1	0	1	1.01	101	7.7
Urine 2	0	3	2.95	98.0	8.3
Urine 3	0	5	4.85	97.0	3.1

---

4. Conclusion

In the present paper, AgNPs/MIL-101(Cr) hybrids were successfully prepared as SERS substrates according to an in situ method. The as-prepared AgNPs/MIL-101(Cr) with positive charge could provides large surface areas for effective concentrate of FA through the electrostatic interaction and get them sufficiently close to the surface of Ag NPs anchored on the MIL-101(Cr), leading to a dramatic enhancement of Raman intensity of FA. Thus, based on the obtained AgNPs/MIL-101(Cr), a simple, sensitive and label-free SERS strategy was developed for the detection of FA according to the special Raman scattering of FA and the SERS method was successfully used for the detection of FA in human urine samples, which suggested the present approach, had great practicability for diagnostic purposes. Moreover, the size of MIL-101(Cr) can be decreased, and we expect that the smaller size MIL-101(Cr) combined with FA will find better use in biological applications which require MOFs with small size, such as cancer cell targeting and drug delivery.

Acknowledgements

The authors are grateful for the financial support by the National Natural Science Foundation of China (NSFC, No. 21175109).

Notes and references

1. J. Kneipp, H. Kneipp and K. Kneipp, Chem. Soc. Rev., 2008, 37, 1052–1060.
2. M. D. Porter, R. J. Lipert, L. M. Siperko, G. Wang and R. Narayanan, Chem. Soc. Rev., 2008, 37, 1001–1011.
3. A. Liang, Q. Liu, G. Wen and Z. Jiang, Treatise Anal. Chem., 2012, 37, 32–47.
4. L. Chen and Y. Liu, J. Raman Spectrosc., 2012, 43, 986–991.
5. T. Yang, H. Yang, S. J. Zhen and C. Z. Huang, ACS Appl. Mater. Interfaces, 2015, 7, 1586–1594.
6. X. Jiang, M. Yang, Y. Meng, W. Jiang and J. Zhan, ACS Appl. Mater. Interfaces, 2013, 5, 6902–6908.
7. M. Liu and W. Chen, Biosens. Bioelectron., 2013, 46, 68–73.
8. W. Ren, Y. Fang and E. Wang, ACS Nano, 2011, 5, 6425–6433.
9. Z. Zhang, S. Xiang, X. Rao, Q. Zheng, F. R. Fronczek, G. Qian and B. Chen, Chem. Commun., 2010, 46, 7205–7207.
10. Y. Hu, J. Liao, D. Wang and G. Li, Anal. Chem., 2014, 86, 3955–3963.
11. C. X. Yang, H. B. Ren and X. P. Yan, Anal. Chem., 2013, 85, 7441–7446.
12. N. Yang, H. J. Song, X. Y. Wan, X. Q. Fan, Y. Y. Su and Y. Lv, Analyst, 2015, 140, 2656–2663.
13. Y. L. Liu, X. J. Zhao, X. X. Yang and Y. F. Li, Analyst, 2013, 138, 4526–4531.
14. Z. W. Jiang, Y. L. Liu, X. L. Hu and Y. F. Li, Anal. Methods, 2014, 6, 5647–5651.
15. Y. L. Liu, W. L. Fu, C. M. Li, C. Z. Huang and Y. F. Li, Anal. Chim. Acta, 2015, 861, 55–61.
16. C. Zhao, Y. L. Liu and Y. F. Li, Anal. Sci., 2015, 31, 1035–1039.
17. Z. J. Sun, J. Z. Jiang and Y. F. Li, Analyst, 2015, 140, 8201–8208.
18. Z. W. Jiang, P. F. Gao, L. Yang, C. Z. Huang and Y. F. Li, Anal. Chem., 2015, 87, 12177–12182.
19. K. M. L. T. Pashow, J. D. Rocca, Z. Xie, S. Tran and W. Lin, J. Am. Chem. Soc., 2009, 131, 14261–14263.
20. G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surblé and I. Margiolaki, Science, 2005, 309, 2040–2042.
21. F. Leng, W. Wang, X. J. Zhao, X. L. Hu and Y. F. Li, Colloids Surf., A, 2014, 441, 164–169.
22. Z. W. Jiang and Y. F. Li, J. Taiwan Inst. Chem. Eng., 2015, 1–7.
23. J. M. Fang, F. Leng, X. J. Zhao, X. L. Hu and Y. F. Li, Analyst, 2014, 139, 801–806.
24. W. Hu, G. He, H. Zhang, X. Wu, J. Li, Z. Zhao, Y. Qiao, Z. Lu, Y. Liu and C. M. Li, Anal. Chem., 2014, 86, 4488–4493.
25. B. Fei, B. Qian, Z. Yang, R. Wang, W. C. Liu, C. L. Mak and J. H. Xin, Carbon, 2008, 46, 1792–1828.
26. X. Bai, X. Li and L. Zheng, Langmuir, 2010, 26, 12209.
27. W. L. Fu, S. J. Zhen and C. Z. Huang, Analyst, 2013, 138, 3075–3081.
28. A. M. Branum, R. Bailey and B. J. Singer, J. Nutr., 2013, 143, 486–492.
29. X. Wu, X. Chen, F. Gao, W. Ma, L. Xu, H. Kuang, A. Li and C. Xu, Biosens. Bioelectron., 2016, 75, 55–58.
30. R. J. Stokes, E. McBride, C. G. Wilson, J. M. Girkin, W. E. Smith and D. Graham, Appl. Spectrosc., 2008, 62, 371–376.

---