A small-molecule chemosensor for the selective detection of 2,4,6-trinitrophenol (TNP)

Abstract

A pyridine-based small-molecule receptor (L) for the specific recognition of TNP was synthesised and characterised by ¹H NMR, ¹³C NMR and FT-IR spectra, and the optical properties were studied by UV-Vis absorption and PL spectra. The results showed that the fluorescence of L was strongly quenched by picric acid accompanied by an obvious colour change from transparent to yellow, which indicated that the receptor L can serve as a small-molecule sensor for TNP. Crystal structure determination and theoretical calculation of the corresponding host–guest complex L·TNP were also performed.

---

The rational design and synthesis of host molecules for particular guest molecules or ions are drawing more and more active attention due to their vitally important applications in the fields of biological and environmental sensing.^1,2 Molecules possessing pyridine units in their structures could be fascinating “acceptors” for a variety of metal ions due to their strong binding abilities.^3,4 However, to our best knowledge, reports focused on neutral molecules of nitrated explosives such as trinitrophenol (TNP) are still rare.^5,6 Nitro-aromatic compounds (NACs) such as 2,4-dinitrotoluene (DNT), trinitrotoluene (TNT) and TNP, which can potentially cause respiratory damage, are widely used in manufacturing and blasting industries because of their superior explosiveness compared to TNT.^7–9 In addition to its highly explosive nature, TNP is also recognised as a threat to human health and national security as it can cause severe irritation, allergic reactions of the skin, dizziness, nausea, and damage to the liver and kidney.^10,11 The contamination of water with TNP has the potential to cause severe epidemics.^12,13 Thus, a large number of molecular and modulated chemosensors have been developed.^14–19 Some of them have shown good selectivity and sensitivity to TNP, but their structures are complicated, and they usually require a multi-step and inefficient synthetic procedure. Therefore, the development of simple and effective molecules for the specific recognition of TNP is urgent and instructive.

In the present paper, the novel, simple and pyridine-based compound (E)-4-(2-(pyridin-4-yl)vinyl)aniline (L) was designed, synthesised and characterised. It was expected to form a “host–guest” complex with TNP, resulting in entirely different UV-Vis absorption and fluorescence responses. The absorption of compound L with TNP at 350 nm increased obviously together with the appearance of a new strong absorption band at 424 nm. Meanwhile, an apparent colour change from transparent to yellow was observed accompanied by fluorescence quenching at 440 nm and a new emission band at 550 nm.

Experimental section

Chemical and instruments

Commercially available chemicals were used without further purification. All of the solvents used were purified by conventional methods before use. The synthetic route is shown in Scheme 1, and the characterisations are described. ¹H NMR and ¹³C NMR spectra were recorded on a Bruker Avance 400 MHz NMR spectrometer using DMSO-d₆ as the solvent. The instrument was calibrated using tetramethylsilane (TMS) as an internal reference. Data were reported as follows: chemical shifts (δ) in ppm; multiplicity (s = singlet, d = doublet, t = triplet, m = multiplet); coupling constants J (Hz); integration; and interpretation. The UV-Vis absorption spectra were recorded on a UV-265 spectrophotometer with a quartz cuvette (path length, 1 cm). The one-photon excited fluorescence (OPEF) spectra measurements were performed using a Hitachi F-7000 fluorescence spectrophotometer. High-resolution mass spectra (HR-MS) were recorded on a Thermo Fisher LTQ-Orbitrap XL mass spectrometer using an electrospray ion source (ESI). FT-IR spectra were obtained using KBr discs on a Nicolet 380 FT-IR spectrometer in the 4000–400 cm⁻¹ region. Optical images were obtained using a digital camera (Nikon D7000). Silica gel 60 (60–120 mesh) was used for column chromatography.

Scheme 1 Synthetic route of receptor L.

Caution! TNP, TNT, DNT, and other nitroaromatics (a–h, Fig. 1c) should be used with extreme care using the best safety precautions due to their highly explosive characters. They should be handled only in small quantities.

Fig. 1 (a) UV-Vis absorption spectra of L (10 μM) in THF upon the addition of 5 equiv. of selected NACs. Inset: the degree of change in absorbance at 350 nm. A₀ is the absorbance of the receptor L, and A is the absorbance after adding a–h. The photo shows that L exhibited an obvious selectivity for TNP. (b) Responses of L (10 μM) toward different NACs (5 equiv.) in THF. Inset: the degree of change in intensity at 440 nm. I₀ is the intensity of the receptor L, and I is the intensity of L after adding a–h. (c) Structures of the NACs used in the present study (a–h).

Preparation of the target compound L

Compound L was prepared according to the method reported in the literature.²⁰ FT-IR (KBr, cm⁻¹) selected bands: 3390 (m), 3317 (m), 3032 (w), 2923 (w), 2849 (w), 1584 (s), 1515 (s), 1415 (s), 832 (s). ¹H NMR (DMSO-d₆, 400 MHz, ppm) δ: 5.45 (s, 2H), 6.55 (d, J = 5.6 Hz, 2H), 6.83 (d, J = 5.4 Hz, 1H), 7.31 (m, 3H), 7.41 (d, J = 4.0 Hz, 2H), 8.42 (d, J = 4.0 Hz, 2H). ¹³C NMR (DMSO-d₆, 100 MHz, ppm) δ: 116.87, 122.91, 123.27, 126.17, 131.59, 132.87, 136.85, 148.35, 152.88. HR-MS (ESI-MS): m/z = 197.1107, calcd for [C₁₃H₁₃N₂]⁺ = 197.1000 ([M + H]⁺).

Results and discussion

UV-Vis absorption and fluorescence spectra

Initially, the molecular recognition behaviours of compound L toward the selected NACs were investigated by UV-Vis and fluorescence spectroscopy. As shown in Fig. 1a, the free L in THF had a maximum absorption wavelength of 350 nm. The influences of the selected NACs on its absorption were studied with a previously used method.²¹ The addition of 5 equiv. TNP into compound L affected the absorption more compared to other NACs. The absorption of compound L with TNP at 350 nm increased accompanied by the appearance of a new strong absorption band at 424 nm. This indicated that a complex may form between compound L and TNP with totally different electronic properties from those of compound L. As a result, an obvious colour change from transparent to yellow was observed after the addition of TNP to compound L, allowing the colourimetric detection of TNP by the naked eye (Fig. 1, insets).

Furthermore, the fluorescence spectra of compound L with different NACs are shown in Fig. 1b. The free L displayed an intensive emission at 440 nm in THF solution. The effects of the NACs on the fluorescence of L were also investigated in THF solution. The emission intensity of L at 440 nm decreased dramatically (about 87% quenching) with a new emission wavelength at 550 nm when 5 equiv. TNP was added, however, negligible changes in the fluorescence spectra were observed upon the addition of other NACs. The abovementioned UV-Vis absorption and fluorescent studies demonstrated that compound L could bind effectively with TNP and serve as specific recogniser of TNP.

In order to further investigate the photo-physical response of compound L to TNP, corresponding UV-Vis titration experiments (Fig. S1, ESI†) and fluorescent titration experiments were conducted. As shown in Fig. 2, upon addition of incremental amounts of 10 μM TNP to a solution of L in THF, quenching of the fluorescence emission at 440 nm was observed at concentrations as low as 20 equiv. Meanwhile, a new emission at 550 nm appeared. We then measured the fluorescence lifetime of L for different concentrations of TNP (0, 0.5 and 1.0 equiv.). The fluorescence lifetime of compound L was found to be invariant (<0.1 ns) at different concentrations of TNP (Fig. S2, ESI†), indicating that the quenching at 440 nm is static, and that a ground state complex is formed between compound L and TNP. These results implied that there are strong interactions between L and TNP. We proposed that the three NO₂ groups in TNP may act as strong electron-withdrawing groups that protonate the pyridine ring and further affect the intramolecular electron density distribution when they form hydrogen bonds with each other. The redistribution of the electron density influenced the intramolecular charge transfer (ICT), subsequently inducing an absorption and fluorescence response to TNP. Job's plot analysis was conducted to corroborate the 1 : 1 ratio between L and TNP (Fig. S3, ESI†).²² A linear Stern–Volmer plot was obtained from fluorescence quenching titration (Fig. S4, ESI†) with K_SV = 4.106 × 10⁵ M⁻¹.⁹ The detection limit was found to be 400 ppb for TNP (Fig. S5, ESI†).²³ Interference experiments to determine the effects of other nitro aromatics on the interaction of TNP with L in THF were conducted (Fig. S6, ESI†); the fluorescence emission intensity of L was almost unaffected. Considering environmental applications, an optimised aqueous solution (THF–H₂O = 9 : 1, v/v) was then selected as a testing system to investigate the spectral response of the probe L to TNP at room temperature. As shown in Fig. S7 (ESI†), both the absorption and fluorescence spectra in aqueous solution exhibited a similar response to those in pure THF, proving that L can detect TNP in aqueous solution. The ideal experimental results obtained herein demonstrated the practical applicability of the probe toward TNP sensing. Furthermore, we performed fluorescence titration experiments of L with TNP (0–20 equiv.) at various excitation wavelengths. No obvious change in quenching efficiency of TNP was observed (Fig. S8, ESI†).

Fig. 2 Change in the fluorescence of compound L (10 μM) in THF upon the addition of TNP (0–20 equiv.), λ_ex = 350 nm. Inset: the fluorescent quenching percentage as a function of TNP concentration, concentration of L = 10 μM.

Single crystal analysis

Next, we grew single crystal L·TNP and investigated the interactions between L and TNP to explain the change in emission band. By adding 5 mL of TNP (2.6 × 10⁻² M) solution to 5 mL of L (2 × 10⁻² M) solution in THF–ethanol (2 : 1), vermilion block crystals were obtained by slow evaporation of the solvent at room temperature after a few days (Fig. 3a). Single-crystal X-ray diffraction analysis revealed that the 1 : 1 host–guest complex L·TNP was formed (Fig. 3a) by L and TNP in solution. The relevant crystal data and structural parameters are listed in Table 1. All calculations and image generation were done using WinGX²⁴ and PLATON²⁵ programs.

Fig. 3 (a) Photos of L and L·TNP and the X-ray structure of the host–guest complex L·TNP; (b) hydrogen bond interactions around TNP in the complex; and (c) hydrogen bond interactions and π–π interactions around L in the complex (the hydrogen bond interactions and π–π interactions are indicated with dashed turquoise and bright green lines, respectively, and some hydrogen atoms are omitted for clarity).

Table 1 Summary of crystallographic data and structure refinement details for L·TNP

Identification code	L·TNP
Empirical formula	C19H15N5O7
Formula mass	425.36
Crystal system, space group	Monoclinic, P2(1)/c
Unit cell dimensions	a = 7.917(5) Å
	α = 90.000(5)°
	b = 12.258(5) Å
	β = 96.658(5)°
	c = 20.049(5) Å
	γ = 90.000(5)°
Volume	1932.6(15) Å^3
Z, Calculated density	4, 1.462 Mg m^−3
Temperature	298(2) K
Crystal size	0.30 × 0.20 × 0.20 mm
Absorption coefficient	0.115 mm^−1
F(000)	880
Theta range for data collection	1.95° to 25.00°
Limiting indices	−9 ≤ h ≤ 7
	−14 ≤ k ≤ 14
	−23 ≤ l ≤ 23
Reflections collected	9806/3417
Data/restraints/parameters	3417/1/280
Goodness-of-fit on F^2	1.032
Final R indices [I > 2sigma(I)]	R1 = 0.0802, wR2 = 0.1862
R Indices (all data)	R1 = 0.1211, wR2 = 0.2164

---

The crystal structure showed that the nitrogen of the pyridine group of L is protonated to form a cation, while picric acid is deprotonated to form an anion; the pyridine group was changed from a hydrogen bond acceptor to a hydrogen bond donor (Fig. 3a).²⁶ The “L cations” and “TNP anions” self-assembled to form a supramolecular structure via multiple hydrogen bond and π–π interactions. As shown in Fig. 3b and c, each picrate acted as a hydrogen bond acceptor to connect six ligands and one picrate via eight hydrogen-bonds, and each L acted as a hydrogen bond donor and acceptor to connect seven picrates and one other ligand via eight hydrogen bonds in which the D–A distances vary from 2.763 to 3.591 Å, and the corresponding D–H⋯A angles ranged from 118.53–171.29° (Table S1, ESI†). The intermolecular π–π stacking interaction between the picrate ring and phenylamine ring of L presented a π–π distance of 3.859 Å and a dihedral angle of 15.8°. Notably, the two hydrogen bonds between the N4 of the pyridine group bonded to the O7 of the hydroxyl group and the O5 of the nitro group in the picrate were very strong owing to their short distances.²⁷ These results implied that the pyridine group of L was the key site for the specific interaction with TNP.

¹H NMR titration

The interaction between the probe L and TNP in solution was further studied by ¹H NMR titration experiments (Fig. 4). The signals of the hydrogen atoms located on the aromatic rings were generally downfield shifted upon addition of 2.0 equiv. TNP. Due to the decrease in electron density of the pyridine ring moiety, Hb, Hc and Hf were shifted obviously, and the chemical shifts (Δδ) of L were 0.690, 0.577 and 0.532. These results suggested that the nitrogen atom of the pyridine ring moiety of L was the only reasonable site for protonation by TNP. To further gain insight into the recognition mechanism, the interaction between the probe L and another analogous strong organic acid, trifluoroacetic acid (TFA), in solution was also studied by ¹H NMR titration (Fig. S9, ESI†). The signals of the hydrogen atoms were shifted downfield, and Hb and Hc especially were shifted obviously. These results further confirmed that the ratiometric spectral response of L to TNP was attributed to the protonation of the pyridine ring moiety by the acidic behaviour of TNP, not to the intermolecular electron or energy transfer. In comparison with TFA, which is acidic in character, TNP also has an electron-deficient property, which can protonate the pyridine ring group more easily.

Fig. 4 The entire ¹H NMR titration spectra (400 MHz) of probe L with TNP (0, 0.5, 1.0 and 2.0 equiv.) in DMSO-d₆.

Theoretical calculation

As shown in Fig. 5, the results revealed that the energies of the highest occupied molecular orbital (HOMO) and lowest occupied molecular orbital (LUMO) of the receptor L were −5.03 eV and −1.15 eV, respectively, while those of the newly formed compound L·TNP were −6.09 eV and −2.99 eV, respectively, indicating the charge transfer from picrate to the sensor L. Meanwhile, the energies of the newly formed compound L·TNP were minimal, which indicated that the compound L·TNP was stable.

Fig. 5 HOMO and LUMO orbital of receptor L and L·TNP calculated by B3LYP method with the 6-31G* basis set.

Effect of pH

In order to understand the selectivity of receptor L toward TNP and/or other NACs, the effect of pH was investigated in water. In this case, we selected DNP (pK_a = 3.96), whose structure is closer to TNP (pK_a = 0.71), as a representative NAC.²⁸ As the acidity of TNP is stronger than that of DNP, TNP would result in the easier protonation of receptor L. This initial protonation caused an electrostatic association between the picrate anions and the protonated pyridine ends. To confirm this effect, the influence of pH on the UV-Vis response of receptor L was examined (Fig. 6a).^29,30 When pH was in the range of 6 to 14, the absorption band centred at 336 nm was unchanged. A decrease in pH from 2 to 1 resulted in a blue shift in the maximum absorption wavelength from 336 nm to 327 nm; this 9 nm shift was due to the protonation of the pyridine ring and the amine group to form an A–π–A structure. However, when pH increased from 3 to 5, the main absorption peak exhibited a red shift to 396 nm due to the protonation of the pyridine ring only, which enhanced the electron-withdrawing effect of the pyridine ring. The effect of pH on the fluorescence response of receptor L was also examined (Fig. 6b). The receptor L was fluorescent on account of the ICT that occurs from the amine group to the pyridine ring group. However, when pH is in the range of 1 to 6, the fluorescent intensity is significantly reduced due to the arrest of the ICT process as a result of the protonation of either the pyridine ring or the amine group. These results further confirmed that the spectral response of L to TNP was attributed to the protonation of the pyridine ring group by the acidic behaviour of TNP. They also demonstrated that TNP had an electron-deficient property, which was the main reason that TNP could quench the fluorescence.^31,32

Fig. 6 Effect of pH on (a) the UV-Vis response of L in water and (b) the fluorescence response of L in water. (c) A scheme of the molecular level interactions involved.

Conclusions

In conclusion, we have reported a simple and effective pyridine-based sensor (L) for the specific recognition of TNP. The crystal structure of its TNP complex (L·TNP) was successfully established by X-ray crystallography. In the complex (L·TNP), picric acid exists as picrate ion upon deprotonation, which leads to quenching of the fluorescence of L.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21101001, 50873001 and 61107014) and the Educational Commission of Anhui Province of China (no. KJ20142D02).

References

1. B. W. Xu, X. F. Wu and H. B. Li, Macromolecules, 2011, 44, 5089.
2. F. Zhang, L. Luo and Y. Sun, Tetrahedron, 2013, 69, 9886.
3. E. B. Coropceanu, L. Croitor and A. V. Siminel, Polyhedron, 2014, 75, 73.
4. B. K. Dey, J. Dutta and M. G. B. Drew, J. Organomet. Chem., 2014, 750, 176.
5. M. Kumar, S. I. Reja and V. Bhalla, Org. Lett., 2012, 14, 6084.
6. (a) V. Bhalla, A. Gupta and M. Kumar, ACS Appl. Mater. Interfaces, 2013, 5, 672; (b) Y. Salinas, R. Martínez-Máñez and M. D. Marcos, Chem. Soc. Rev., 2012, 41, 1261.
7. (a) Y. H. Lee, H. G. Liu and J. Y. Lee, Chem.–Eur. J., 2010, 16, 5895; (b) W. S. Zou, F. H. Zou and Q. Shao, J. Photochem. Photobiol., A, 2014, 278, 82.
8. K. Zhang, H. B. Zhou and Q. S. Mei, J. Am. Chem. Soc., 2011, 133, 8424.
9. G. He, N. Yan and J. Y. Yang, Macromolecules, 2011, 44, 4759.
10. V. Pimienta, R. Etchenique and T. Buhse, J. Phys. Chem. A, 2001, 105, 10037.
11. J. Y. Shen, J. F. Zhang and Y. Zuo, J. Hazard. Mater., 2009, 163, 1199.
12. J. F. Wyman, M. P. Serve and D. W. Hobson, J. Toxicol. Environ. Health, Part A, 1992, 37, 313.
13. G. Anderson, J. D. Lamar and P. T. Charles, Environ. Sci. Technol., 2007, 41, 2888.
14. X. G. Hou, Y. Wu and H. T. Cao, Chem. Commun., 2014, 50, 6031.
15. J. Li, J. Z. Liu and B. Z. Tang, RSC Adv., 2013, 3, 8193.
16. V. Bhalla, S. Kaur and V. Vij, Inorg. Chem., 2013, 52, 4860.
17. N. Venkatramaiah, S. Kumar and S. Patil, Chem. Commun., 2012, 48, 5007.
18. G. Sivaraman, B. Vidya and D. Chellappa, RSC Adv., 2014, 4, 30828.
19. L. Hong, Q. S. Mei and L. Yang, Anal. Chim. Acta, 2013, 802, 89.
20. M. Jahan, Q. L. Bao and K. P. Loh, J. Am. Chem. Soc., 2012, 134, 6707.
21. V. Sathish, A. Ramdass and Z. Z. Lu, J. Phys. Chem. B, 2013, 117, 14358.
22. S. Kumar, P. Singh and A. Mahajan, Org. Lett., 2013, 15, 3400.
23. V. Bhalla, A. Gupta and M. Kumar, Org. Lett., 2012, 14, 3112.
24. L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
25. A. L. Spek, PLATON, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1990, 46, C34.
26. Y. Q. Zhang, Y. Guo and Y. H. Joo, Chem.–Eur. J., 2010, 16, 10778.
27. H. L. Liu, X. L. Hou and L. Pu, Angew. Chem., Int. Ed., 2009, 48, 382.
28. N. Dey, S. K. Samanta and S. Bhattacharya, ACS Appl. Mater. Interfaces, 2013, 5, 8394.
29. H. P. Zhou, J. Q. Wang and Y. X. Chen, Dyes Pigm., 2013, 98, 1.
30. S. P. Wu, K. J. Du and Y. M. Sung, Dalton Trans., 2010, 39, 4363.
31. K. K. Kartha, S. S. Babu, S. Srinivasan and A. Ajayaghosh, J. Am. Chem. Soc., 2012, 134, 4834.
32. V. Bhalla, S. Pramanik and M. Kumar, Chem. Commun., 2013, 49, 895.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 987872. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c4ra07957d

---