Zeolitic imidazolate framework nanocrystals for enrichment and direct detection of environmental pollutants by negative ion surface-assisted laser desorption/ionization time-of-flight mass spectrometry

Abstract

Zeolite imidazole framework (ZIF) nanocrystals serve as sorbents and matrices for negative ion surface assisted laser desorption/ionization mass spectrometry. The unique properties of ZIFs make them suitable for analyzing a series of environmental pollutants with high signal intensity and a clean background.

---

Matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) is a soft ionization technique that was developed in the 1980s by Hillenkamp and Tanaka.^1–3 Due to its simplicity, rapidity, and easy operation, MALDI-MS has been used extensively in the analysis of biomolecules.³ Despite the successful applications of MALDI-MS in several fields, it still suffers from problems in the analysis of low molecular weight (m/z < 500 Da) compounds, such as matrix interferences and poor reproducibility, because the conventional organic matrices (e.g. α-cyano-4-hydroxycinnamic acid (CHCA) and 3-aminoquinoline (3-QA)) normally produce abundant interfering matrix peaks^4,5 which make it difficult to distinguish the sample peaks, and the existing “sweet spot” make it exhibiting poor reproducibility, especially in the low-mass range.

To solve these problems, surface-assisted laser desorption/ionization mass spectrometry (SALDI-MS) has been developed by Sunner et al.⁶ by using inorganic nanoparticles and nanostructured substrates instead of the conventional organic matrices to analyze low-mass molecules. With the development of nanomaterials, various nanomaterials have been applied for SALDI-MS as an ionization matrix,^7,8 such as gold (Au) NPs,⁹ platinum (Pt) NPs¹⁰ and silicon (Si) NPs.¹¹ In addition, existing matrix-based nanostructured surfaces are widely used, including silicon-based substrates,^12,13 carbon-based materials,^14,15 metal oxide-based substrates^16,17 and multilayer-coated hybrid substrates, such as metal coated porous alumina,¹⁸ layer-by-layer (LBL) self-assembled films.¹⁹

Metal–organic frameworks (MOFs) are a class of hybrid inorganic–organic microporous crystalline materials that are constructed from metal ions linked to organic ligands by coordination bonds.²⁰ Due to their unique properties, such as high surface areas, permanent nanoscale porosity and tunable cavities^21,22 they have great potential to be used as a matrix for SALDI-MS. MIL-100(Fe) was firstly used as a novel matrix in SALDI-MS by Huang’s group to determine polycyclic aromatic hydrocarbons (PAHs).²³ Among these MOFs, zeolite imidazole frameworks (ZIFs), including ZIF-7, ZIF-8 and ZIF-90, belong to a subclass of MOFs with zeolite-type topologies, constructed from tetrahedral metal ions bridged by imidazolate.²⁴ ZIFs not only exhibit high porosity and large surface areas like other MOFs, but also show exceptional thermal and chemical stability in water and organic solvents, making them promising sorbents and matrices for SALDI-MS. More recently, ZIF-8 coated magnetic nanocomposites were first introduced as a sorbent and new matrix to enrich and detect small biomolecules using SALDI-MS.²⁵ However, compared with the extensive studies of SALDI-MS in the field of biological analysis, there are few reports on the use of SALDI-MS for the environmental analysis of small molecules.

Bisphenols are a group of environmental pollutants, including bisphenol A (BPA), bisphenol B (BPB), bisphenol S (BPS), bisphenol F (BPF) and bisphenol AF (BPAF). BPA and its analogues are a kind of endocrine-disrupting chemical, possessing hormone-like properties and toxicity that have adverse effects on reproduction and development, neural networks, and cardiovascular, metabolic, and immune systems in humans.^26,27 Until now, several approaches have been presented for the determination of bisphenols, such as high performance liquid chromatography (HPLC)²⁸ and gas chromatography (GC).²⁹ Although high sensitivity and low detection limits are achieved, the processes of these methods are time consuming. Consequently, the development of a novel method for the fast and sensitive determination of bisphenols is urgently needed.

Herein, we firstly explore the application of a series of ZIFs (ZIF-7, ZIF-8 and ZIF-90) as both the sorbent and matrix for SALDI-MS for the enrichment and analysis of bisphenols (BPA, BPB, BPS, BPF and BPAF) via the procedure described in the graphical abstract image. In order to demonstrate the feasibility of ZIFs as new matrices, other significant environmental pollutants were also analyzed using SALDI-MS, including 1-nitropyrene (1-NP), 3-nitrofluoranthene (3-NF), 7-nitrobenz[a]anthracene (7-NBaA), 2,6-dichlorophenol (2,6-DCP), 3,5-dibromophenol (3,5-DBP) and heptadecafluorooctanesulfonic acid potassium salt (PFOS). Comparison with traditional matrices shows that ZIFs are versatile matrices with enhanced signal intensity, low background interference and good salt tolerance.

ZIF-7, ZIF-8 and ZIF-90 were synthesized and then characterized using XRD, SEM and TEM. The XRD patterns of the synthesized ZIF-7, ZIF-8 and ZIF-90 were in good agreement with the simulated ones, which shows the successful preparation of the ZIFs (Fig. S1, ESI†). The morphology and size of the ZIFs were investigated using SEM and TEM, which revealed that the particle size of ZIFs was homogenous with an average particle size of ∼50 nm for ZIF-7, ∼50 nm for ZIF-8 and ∼2 μm for ZIF-90 (Fig. 1A, B and S2, ESI†). The matrix dispersing solutions and optical images of the ZIFs show their good dispersibility in solution compared with traditional matrices of CHCA and 3-QA for MALDI-MS (Fig. 1C and S3, ESI†). Due to the small sizes of ZIF-7 and ZIF-8, and good dispersibility in solution, ZIF-7 and ZIF-8 could be more suitable as a matrix for SALDI-MS than ZIF-90 and traditional matrices, avoiding the “sweet spot” in the same sample surface and increasing the spot-to-spot reproducibility, because normally the traditional matrices of CHCA and 3-QA need to find the “sweet spot” to determine the analytes.

Fig. 1 SEM (A) and TEM (B) images of prepared ZIF-8 nanocrystals; (C) digital photographs of ZIF-7, ZIF-8 and ZIF-90 dispersed in 50% C₂H₅OH aqueous solution (1 mg mL⁻¹); (D) mass spectra of 1 mM BPA with ZIF-8 as the matrix using three different sample preparation methods, including matrix-first method (a) sample-first method (b) and on probe remix (c) in negative ion mode.

To evaluate the effect of sample preparation on the MS signal response, BPA was firstly chosen as a representative and three samples with different preparation procedures including the matrix-first method, the sample-first method and on probe remix using ZIF-8 as the matrix were investigated. As shown in Fig. 1D, the signal intensity of BPA obtained from the matrix-first method was considerably higher than that of others, probably because BPA could be ionized very effectively with the matrix-first preparation method, which provides high sensitivity for the measurement of bisphenols. Based on the results, the matrix-first method was chosen as the sample preparation method for further studies.

Five types of bisphenol environmental pollutants, including BPA, BPB, BPS, BPF and BPAF (chemical structures in Fig. S4, ESI†) were analyzed using the ZIFs and traditional matrices of 3-QA and CHCA in negative-ion mode. The characteristic peaks of all five bisphenols were detected with the use of traditional matrices of 3-QA and CHCA, but fairly low signal intensity and high background interferences were observed, seriously suppressing the target signals (Fig. S5, ESI†). This is the main factor that limits the application of MALDI-MS for quantitative measurements. The ZIFs were expected, as the novel matrix, to overcome the above problems, while maintaining excellent energy absorption and transfer capacity. As shown in Fig. 2, all five bisphenols were detected clearly with high signal intensity when using the ZIFs matrices. ZIF-8 showed the best performance among the ZIFs with high signal intensity and low background interference. Both the structures and sizes of ZIF-7 and ZIF-8 are similar, however, as reported by Jiang et al.,³⁰ the surface area of ZIF-8 (1628 m² g⁻¹) is bigger than that of ZIF-7 (247 m² g⁻¹), which probably leads to higher laser energy absorption and transfer efficiency. Due to the relatively larger size of ZIF-90 and strong background interference, ZIF-90 is not the best candidate among the ZIFs as a matrix in the low mass range. Overall, ZIF-8 is the best matrix material among the three ZIFs for SALDI-MS.

Fig. 2 Mass spectra of BPA (A–C, m/z 227.1, [M − H]⁻; m/z 211.1, [M − CH₄ − H]⁻), BPB (D–F, m/z 241.2, [M − H]⁻; 211.1; m/z [M − C₂H₆ − H]⁻), BPS (G–I, m/z, 249.1, [M − H]⁻), BPF (J–L, m/z 199.2, [M − H]⁻; m/z 197.2, [M − 3H]⁻) and BPAF (M–O, m/z 335.3, [M − H]⁻; m/z 265.2, [M − CF₃ − 2H]⁻) with different matrices (ZIF-7, ZIF-8 and ZIF-90) compared with traditional matrices (3-QA and CHCA, details in Fig. S6†) at a concentration of 1 mM in negative-ion mode.

The influence of salt concentration on ZIF-8 was investigated with BPS. The results showed that when the concentration of NaCl was up to 100 mM, the signal intensity of BPS declined slightly, less than 25%. And when the concentration of NaCl was up to 500 mM, the peak intensity was strongly influenced with an obvious decrease, indicating that the high salt concentration suppressed the signal intensity of the target. Although ZIF-8 was strongly influenced by the extremely high salt concentration, it still showed a good sample-to-sample reproducibility at the concentration of 500 mM NaCl (RSD = 10.7%, n = 15) (Fig. S6, ESI†). Moreover, the use of ZIF-8 as the matrix offers significant advantages of good spot-to-spot and sample-to-sample reproducibility when using BPA as an example (Fig. S7, ESI†). The RSDs were 5.3% (n = 10) for the spot-to-spot and 19.1% (n = 30) for the sample-to-sample assays, indicating that ZIF-8 is a homogeneously crystallized matrix without a “sweet spot”.

The feasibility of ZIF-8 as a matrix for quantitative analysis was further investigated. The results showed that bisphenols had good linearity with R² = 0.9925 for BPA, R² = 0.9918 for BPB, R² = 0.9953 for BPS, R² = 0.9973 for BPF and R² = 0.9947 for BPAF in the concentration range from 50 to 500 μM (Fig. S8, ESI†). Moreover, high sensitivity has been achieved by using the ZIF-8 matrix. The limit of detection (LOD) (S/N = 3) of BPA, BPB, BPS, BPF and BPAF was 1.60 ng mL⁻¹, 2.50 ng mL⁻¹, 1.17 ng mL⁻¹, 2.92 ng mL⁻¹ and 2.98 ng mL⁻¹, respectively. And the limit of quantitation (LOQ) (S/N = 10) is 7.0 pmol for BPA, 10.3 pmol for BPB, 4.7 pmol for BPS, 14.6 pmol for BPF and 8.9 pmol for BPAF (Table S1, ESI†). A comparison of the quantitative determination of the bisphenols with the traditional HPLC and GC methods is shown in Table S2, ESI.† Based on the above results of good linearity, and low LOD and LOQ, ZIF-8 was the best matrix to measure bisphenols in sewage. The real samples were tested and the recoveries are from 76.3 to 95.5% with RSDs of 7.9 to 5.2% (Fig. S9 and Table S3, ESI†), implying that the ZIF-8 matrix could be applied for quantitative analysis in real samples with good resistance to the complex substrate.

The application of ZIF-8 in SALDI-MS for the analysis of other environmentally significant compounds (1-NP, 3-NF, 7-NBaA, 2,6-DCP, 3,5-DBP and PFOS) were evaluated (Fig. 3, chemical structures in Fig. S4, ESI†) by comparing with ZIF-7 (Fig. S10, ESI†), 3-QA (Fig. S11, ESI†) and CHCA (Fig. S12, ESI†). As can be seen, the characteristic peaks of all six analytes were detected with high signal intensity and low background interference when using the ZIF-8 matrix. However, for ZIF-7, the characteristic peak of 2,6-DCP was not detected and the peak intensities of 1-NP, 3-NF, 7-NBaA, 3,5-DBP and PFOS were much lower than those obtained using ZIF-8. For 3-QA, only characteristic peaks of 7-NBaA and PFOS were detected, which was probably due to their easier ionization and desorption from the matrix than the other compounds. For CHCA, characteristic peaks of 1-NP, 3-NF, 7-NBaA and PFOS were observed, but the other two compounds of 2,6-DCP and 3,5-DBP were still not detected. As a result, ZIF-7 and ZIF-8 have obvious advantages over the conventional matrices of 3-QA and CHCA for the analysis of environmental pollutants, and ZIF-8 is a very promising matrix for SALDI-MS.

Fig. 3 Mass spectra of 1-NP (A, 0.4044 mM, m/z 216.8, [M − NO]⁻; m/z 261.8, [M + O − H]⁻), 3-NF (B, 0.4044 mM, m/z 217.4, [M − NO]⁻; m/z 262.4, [M + O − H]⁻), 7-NBaA (C, 0.3659 mM, m/z 243.2, [M − NO]⁻; 288.4, [M + O − H]⁻), 2,6-DCP (D, 1 mM, m/z 160.8, [M − H]⁻), 3,5-DBP (E, 1 mM, m/z 250.6, [M − H]⁻) and PFOS (F, 1 mM, m/z 498.8, [M − K]⁻) with ZIF-8 in negative-ion mode.

The special structures and specific features of ZIF-7 and ZIF-8 means they have great potential to be used as sorbents for environmental pollutants. Therefore, the adsorption kinetics and capabilities of ZIF-7 and ZIF-8 for bisphenols were investigated by binding experiments with BPS as the representative. The results showed that the adsorption equilibrium of BPS with ZIF-7 and ZIF-8 was reached within 60 min at the concentration of 40 μg mL⁻¹ (Fig. S13A, ESI†). Moreover, the maximum adsorption capacity of ZIF-7 and ZIF-8 was 25 mg g⁻¹ and 30 mg g⁻¹, respectively (Fig. S13B, ESI†), making it possible to enrich trace BPS in solution. Therefore, a series of enrichment experiments were performed. As Fig. S14 (ESI†) showed that when 10 μM BPS was added to the sewage, the characteristic peaks of BPS were not detectable without enrichment in both the ZIF-7 and ZIF-8 matrices. However, signal peaks of 10 μM BPS were detected clearly with ZIF-7 (S/N = 4.0) and ZIF-8 (S/N = 32.3) after enrichment, and the signal intensity with ZIF-8 is much higher than that of ZIF-7, probably because the larger surface area of ZIF-8 favours the adsorption of bisphenol compounds and ZIF-8 with porous structure. The detection limit (S/N = 3) of BPS was from 4.7 pmol down to 0.93 pmol after enrichment with ZIF-8. Overall, the unusual properties of ZIF-8 make it a better sorbent and matrix than ZIF-7.

In summary, ZIFs were synthesized and utilized as novel sorbents and matrices for the MALDI-TOF MS analysis of low-molecular-weight environmental pollutants. The results demonstrated that ZIFs have several advantages over traditional matrices (e.g. 3-QA and CHCA), such as high signal intensity, low background interference, good salt tolerance and excellent stability and reproducibility. Moreover, ZIF-7 and ZIF-8 can be utilized as novel sorbents with high adsorption capacities to enrich BPS in real samples. Thus the ZIFs might be applicable for the analysis of environmental pollutants, and as novel matrices to solve more analytical challenges.

Acknowledgements

This study was supported by Grant 21375018 and 21275020 from the National Natural Science Foundation of China.

Notes and references

1. K. Tanaka, H. Waki, Y. Ido, S. Akita, Y. Yoshida, T. Yoshida and T. Matsuo, Rapid Commun. Mass Spectrom., 1988, 2, 151–153.
2. M. Karas, D. Bachmann and F. Hillenkamp, Anal. Chem., 1985, 57, 2935–2939.
3. S. Zhao, K. V. Somayajula, A. G. Sharkey, D. M. Hercules, F. Hillenkamp, M. Karas and A. Ingendoh, Anal. Chem., 1991, 63, 450–453.
4. J. J. A. van Kampen, P. C. Burgers, R. de Groot, R. A. Gruters and T. M. Luider, Mass Spectrom. Rev., 2011, 30, 101–120.
5. L. H. Cohen and A. I. Gusev, Anal. Bioanal. Chem., 2002, 373, 571–586.
6. J. Sunner, E. Dratz and Y.-C. Chen, Anal. Chem., 1995, 67, 4335–4342.
7. R. Arakawa and H. Kawasaki, Anal. Sci., 2010, 26, 1229–1240.
8. Y. E. Silina and D. A. Volmer, Analyst, 2013, 138, 7053–7065.
9. J. A. McLean, K. A. Stumpo and D. H. Russell, J. Am. Chem. Soc., 2005, 127, 5304–5305.
10. R. Nayak and D. R. Knapp, Anal. Chem., 2010, 82, 7772–7778.
11. X. Wen, S. Dagan and V. H. Wysocki, Anal. Chem., 2007, 79, 434–444.
12. J. Wei, J. M. Buriak and G. Siuzdak, Nature, 1999, 399, 243–246.
13. H. Kawasaki, Y. Shimomae, T. Watanabe and R. Arakawa, Colloids Surf., A, 2009, 347, 220–224.
14. Q. Liu, M. Cheng and G. Jiang, Chem.–Eur. J., 2013, 19, 5561–5565.
15. R. N. Ma, M. H. Lu, L. Ding, H. X. Ju and Z. W. Cai, Chem.–Eur J, 2013, 19, 102–108.
16. C.-T. Chen and Y.-C. Chen, Rapid Commun. Mass Spectrom., 2004, 18, 1956–1964.
17. M.-T. Wang, M.-H. Liu, C. R. C. Wang and S. Y. Chang, J. Am. Soc. Mass Spectrom., 2009, 20, 1925–1932.
18. S. Okuno, R. Arakawa, K. Okamoto, Y. Matsui, S. Seki, T. Kozawa, S. Tagawa and Y. Wada, Anal. Chem., 2005, 77, 5364–5369.
19. J. Duan, M. J. Linman and Q. Cheng, Anal. Chem., 2010, 82, 5088–5094.
20. O. M. Yaghi, M. O’Keeffe, N. W. Ockwig, H. K. Chae, M. Eddaoudi and J. Kim, Nature, 2003, 423, 705–714.
21. Z. Y. Gu, C. X. Yang, N. Chang and X. P. Yan, Acc. Chem. Res., 2012, 45, 734–745.
22. N. Chang, Z.-Y. Gu and X.-P. Yan, J. Am. Chem. Soc., 2010, 132, 13645–13647.
23. Y.-H. Shih, C.-H. Chien, B. Singco, C.-L. Hsu, C.-H. Lin and H.-Y. Huang, Chem. Commun., 2013, 49, 4929–4931.
24. A. Phan, C. J. Doonan, F. J. Uribe-Romo, C. B. Knobler, M. O’Keeffe and O. M. Yaghi, Acc. Chem. Res., 2010, 43, 58–67.
25. Z. Lin, W. Bian, J. Zheng and Z. Cai, Chem. Commun., 2015, 51, 8785–8788.
26. C. Liao, F. Liu, Y. Guo, H.-B. Moon, H. Nakata, Q. Wu and K. Kannan, Environ. Sci. Technol., 2012, 46, 9138–9145.
27. J. Yang, Y. Li, J. Wang, X. Sun, R. Cao, H. Sun, C. Huang and J. Chen, Anal. Chim. Acta, 2015, 872, 35–45.
28. L. Wang, Z. Zhang, X. Xu, D. Zhang, F. Wang and L. Zhang, Talanta, 2015, 142, 97–103.
29. L. Q. Yu and X. P. Yan, Chem. Commun., 2013, 49, 2142–2144.
30. J.-Q. Jiang, C.-X. Yang and X.-P. Yan, ACS Appl. Mater. Interfaces, 2013, 5, 9837–9842.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra00877a

---