Borate ester endcapped fluorescent hyperbranched conjugated polymer for trace peroxide explosive vapor detection

Abstract

The vapor of peroxide explosives (PEs) is difficult to detect using fluorescent probes because PEs are not typical quenching agents, not having nitro groups or aromatic units that can easily interact with electron-rich probes. Three borate ester endcapped pyrenyl–fluorene copolymers were reported for the detection of PEs, including a hyperbranched polymer (S1) and two linear polymers with borate esters on fluorenyl (S2) or pyrenyl (S3) units. It was found that the hyperbranched polymer S1 has greater steric hindrance, more external borate ester groups, higher HOMO level and higher fluorescence quantum yield, which give it higher sensitivity to H₂O₂ vapor than S2 and S3. To further amplify the sensing performance toward H₂O₂ vapor, a polymer/ZnO nanorod array composite was used, exploiting the catalytic ability and high area to volume ratio of the ZnO nanorod array. The fluorescence of the S1 film is quenched by ∼60% and ∼30% under saturated vapor of H₂O₂ and TATP, respectively, for 300 s at room temperature, and the detection limit for H₂O₂ is estimated to be 1.6 ppb. These results reveal that the S1/ZnO nanorod array composite is very promising for the preparation of a highly sensitive fluorescence device for detecting the vapor of peroxide explosives.

---

1. Introduction

During the past decade, social instability, terrorist activities and other criminal activities have caused enormous harm to human health and social stability.^1,2 A major concern in the field of public safety today is to continue to develop highly sensitive sensors for the fast and accurate identification of trace explosives.^3–7 The technology of fluorescence sensing based on fluorescent organic molecules has been intensively investigated recently because it has the advantages of high sensitivity, good specificity, rapid response, portability and simple operation. Many studies have been carried out with this technology, which is widely used in the trace detection of ions, biological molecules, explosives and environmental pollutants.^8–13

Peroxide-based explosives (PEs) such as triacetone triperoxide (TATP) are increasingly used in making improvised explosive devices in some criminal and terrorism activities^14,15 because of their simple preparation from readily available materials¹⁶ (TATP is made from just three ingredients: hydrogen peroxide (H₂O₂), an acid and acetone), poor chemical stability and easy decomposition leading to large explosive power. Compared to nitroaromatic explosives, e.g. 2,4,6-trinitrotoluene (TNT), peroxide explosive vapor is difficult to detect with fluorescence probes because PEs are not typical quenching agents, not having nitro groups or aromatic units that can easily interact with electron-rich chromophores by intermolecular π–π stacking interactions. In addition, sensitivity to mechanical stress, low stability, limited solubility and so on also make the detection of PEs more challenging.^17,18 For example, Germain et al. introduced a fluorescence detection method targeting H₂O₂ by using a chelator formed via reaction with H₂O₂, which was limited to the liquid phase.¹⁹ Prof. Deqing Zhang et al. and Hong Qun Luo et al. have also developed interesting strategies for the detection of hydrogen peroxide and glucose with two different types of fluorescent probe in solution,^20,21 which is especially suitable for biological detection. However, for peroxide explosive detection, the response rate of liquid phase detection is slow (greater than 30 min) possibly due to a slow solute diffusion process, and the additives in the probe synthesis may make the detection complicated. Todd et al. reported a method using cavity ringdown spectroscopy (CRDS) for the detection of TATP in the vapor phase but overlapping absorption bands may occur for air samples containing several analytes, leading to limited selectivity.²² Moreover, there are some other methods for detection of PEs such as gas chromatography/mass spectrometry (GC/MS), selected-ion flow tube mass spectrometry (SIFT-MS) and ion mobility spectrometry (IMS), but they require either complex operation or heavy equipment, which greatly increase the detection cost and reduce the detection efficiency.^23–25

Ling Zang et al. reported a fluorescence turn-on molecule and a fluorescence ratiometric sensor molecule that detected trace vapor of H₂O₂ by a deboration reaction.^26,27 They are both based on small molecules with a fluorescence enhancement or change of emission wavelength and only H₂O₂ was checked. In this contribution, we report three kinds of borate ester endcapped pyrenyl–fluorene copolymers based on a fluorescence “turn-off” response for detection of peroxide vapor, including H₂O₂ and TATP. The polymers were synthesized via Suzuki coupling reactions, as shown in Fig. 1, and were comprised of a hyperbranched polymer (S1) and two linear polymers with a borate ester on the fluorenyl (S2) or pyrenyl (S3) units. For efficient detection, both the molecular structure and morphology of the film are critical. From the molecular structure aspect, both pyrene and fluorene derivatives are important building blocks in fluorescent materials due to their relatively simple molecular structure, easy modification and high fluorescence quantum efficiency. The borate ester units, as the chain end groups, were introduced to selectively detect peroxides. In addition, an amplified fluorescence signal could improve the sensing properties via the “molecular wires” or the hyperbranched conjugated structure of the polymers. In particular, the properties associated with hyperbranched polymers such as modifiable surface functionality, available internal cavities, uniform film and good solubility make them attractive in biological, chemical and medical fields.²⁸ From the morphology aspect, the microstructure of films, which determines the quenching efficiency, response time, etc. has an important effect on the sensing performance. We find that using a ZnO nanorod array as the substrate of the sensing film²⁹ can effectively increase the signal strength, reaction rate and sensitivity due to a high area to volume ratio for efficient analyte permeability and catalytic light oxidation ability.³⁰ Herein, we systematically discuss the effects of the configuration of the polymer, chain end location and number of borate ester units, as well as forming a polymer/ZnO nanorod array composite, on the sensing performance for peroxide vapor detection.

Fig. 1 The polymerization of S1, S2 and S3.

2. Experimental section

2.1 Instruments

The ¹H-NMR spectra were recorded on a Bruker DRX500 instrument, and tetramethylsilane (TMS) was used as an internal standard. UV-Vis absorption and fluorescence analysis were carried out using a Jasco V-670 spectrophotometer and a Jasco FP 6500 spectrometer, respectively. High vacuum infrared spectra were obtained using a Bruker VERTEX 70v via the surface reflection–absorption model. Cyclic voltammetry (CV) experiments were performed with a CH instruments electrochemical analyzer. The electrochemical behaviors of S1, S2 and S3 were investigated in a standard three electrode electrochemical cell (a glassy carbon working electrode, a platinum counter electrode and a silver chloride electrode as a reference electrode) with a 0.1 M tetra-n-butylammonium hexafluorophosphate (Bu₄NPF₆) solution in acetonitrile, with a scanning rate of 100 mV s⁻¹ under nitrogen atmosphere at room temperature.

2.2 Synthesis

All the chemicals and solvents were obtained from commercial sources and used as received. The synthetic procedures are illustrated in Fig. 1. All the polymers were synthesized via Suzuki–Miyaura cross-coupling reactions with yields of 34%, 41% and 48%, respectively. The raw materials, 1,3,6,8-tetrabromopyrene, 9,9-dioctyl-2,7-bis(boronic acid pinacol ester)fluorene, 1,6-dibromopyrene, 2,7-dibromo-9,9-didodecyl-fluorene and 2,7-bis(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)pyrene were obtained from commercial sources.

2.2.1 Polymerization of S1. A mixture of 1,3,6,8-tetrabromopyrene (0.156 g, 0.301 mmol), 9,9-dioctyl-2,7-bis(boronic acid pinacol ester)fluorene (0.951 g, 1.480 mmol), Pd(PPh₃)₄ (98.6 mg, 0.085 mmol), K₂CO₃ (aq, 2 M, 10 mL), dioxane (20 mL), ethanol (10 mL) and aliquat 336 (two drops) was charged under nitrogen and stirred for 24 h at 100 °C. After cooling to room temperature, the solution was washed with water–CH₂Cl₂ and dried with MgSO₄. The organic phase solution was evaporated under reduced pressure to obtain the crude solid product. The crude polymer was then dissolved with the minimum amount of toluene and precipitated in methanol three times. The polymer was then further purified over a short silica column with THF as the eluent and then precipitated in MeOH again. The final product was dried to afford the polymer as a light green solid (234 mg, 34%). 1H-NMR (500 MHz, CDCl₃, 25 °C, TMS): δ = 8.20 (s, 4H), 8.07 (s, 2H), 7.85 (d, 4H), 7.83 (d, 4H), 7.78 (t, 8H), 7.67 (d, 8H), 2.04 (s, 20H), 1.26 (m, 30H), 1.12 (m, 90H), 0.83 (m, 50H). GPC (THF vs. PS): Mn = 2875, Mw = 3514, PDI = 1.22.

2.2.2 Polymerization of S2. The same polymerization procedure as for S1 was followed, with 1,6-dibromopyrene (277.2 mg, 0.77 mmol), 9,9-dioctyl-2,7-bis(boronic acid pinacol ester)fluorene (0.73 g, 1.136 mmol), Pd(PPh₃)₄ (89 mg, 0.077 mmol), K₂CO₃ (aq., 2 M/10 mL), tetrahydrofuran (20 mL) and aliquat 336 (four drops). The final product was dried to afford S2 as a greyish-green solid (250 mg, 41%). ¹H-NMR (500 MHz, CDCl₃, 25 °C, TMS): δ = 8.34 (m, 3H), 8.15 (m, 3H), 8.09 (m, 1H), 7.83 (m, 2H), 7.73 (m, 2H), 7.65 (m, 2H), 2.16 (s, 2H), 2.08 (s, 2H), 1.43 (s, 2H), 1.25 (m, 18H), 0.97 (m, 14H), 0.84 (m, 6H). GPC (THF vs. PS): Mn = 4819, Mw = 6981, PDI = 1.45.

2.2.3 Polymerization of S3. The same polymerization procedure as for S1 was followed, with 2,7-dibromo-9,9-didodecyl-fluorene (330 mg, 0.5 mmol), 2,7-bis(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-1,6-dihydropyrene (340 mg, 0.75 mmol), Pd(PPh₃)₄ (60 mg, 0.05 mmol), K₂CO₃ (aq., 2 M/15 mL), toluene (15 mL) and aliquat 336 (four drops). The final product was dried to afford the polymer as a light brown solid (170 mg, 48%). ¹H-NMR (500 MHz, CDCl₃, 25 °C, TMS): δ = 8.53 (m, 3H), 8.25 (m, 4H), 7.95 (m, 8H), 2.18 (m, 4H), 1.25 (s, 8H), 1.22 (s, 48H), 0.88 (m, 18H). GPC (THF vs. PS): Mn = 8278, Mw = 27 905, PDI = 3.371.

2.3 Vapor sensing experiment

The sensing films were prepared by dip-coating toluene solutions (10⁻⁴ M) of S1, S2 and S3 onto 10 × 20 mm quartz plates, which were vacuum-dried for half an hour before use. The fluorescence responses of the films to peroxide were determined by inserting the films into hermetically-sealed vials (3.8 mL) containing cotton and analyte at room temperature, which prevented direct contact between the probe and analyte as well as helping to maintain a constant vapor pressure. The fluorescent time-course responses were recorded as soon as the film was exposed to analyte vapor and ended at 300 s.

3. Result and discussion

3.1 Optical and electrochemical properties

Fig. 2 shows the absorption and emission spectra of S1, S2 and S3 in films and in THF solution. Their photophysical characteristics are summarized in Table 1. On the one hand, it can be found that the maximum absorption peaks of the linear polymers S2 and S3 in THF are at 383 and 346 nm, respectively. Compared with that of S2, the maximum absorption peak of S3 is blue-shifted by 37 nm and a similar result is also seen in the film state. This proves that a different connection position of the borate ester unit has a different effect on the spectral characteristics. On the other hand, the emission peak of the hyperbranched polymer S1 in THF is 465 nm, which is red-shifted by 28 nm compared to that of S2. The same result was also observed in the film state (red-shifted by 20 nm). The number of borate esters in S1 in one molecular unit is more than that of S2, and there are different configurations in S1 (hyperbranched polymer) and S2 (linear polymer), which leads to more efficient conjugation in S1. It is easy to determine that the number of borate esters on the chain end and the configuration of the polymer have different impacts on the spectral properties.

Fig. 2 UV-Vis absorption and emission spectra of S1, S2 and S3 in films (a) and in THF solution (concentration, 10⁻⁶ M) (b).

Table 1 Optical and electrochemical properties of S1, S2 and S3

	Abs, λmax (nm)		PL, λmax (nm)		HOMO (eV)	LUMO (eV)	ΔE (eV)	Φ ^a
	Solution	Film	Solution	Film
a The fluorescence quantum yields of S1, S2 and S3 in dilute THF solution were measured using a dilute solution of 9,10-diphenylanthracene (Φ ∼ 1) solution in THF as the standard.
S1	388	400	465	466	−5.49	−2.69	2.80	0.89
S2	383	396	437	446	−5.99	−2.78	3.21	1
S3	346	361	427	437	−6.04	−2.79	3.25	0.39

---

The emission peaks of the S2 and S3 films are red-shifted by 9 and 10 nm from solution to film state, respectively, while that of the S1 film remains almost the same (a red shift of only 1 nm). These results illustrate that the hyperbranched structure of S1 has larger steric hindrance, which is beneficial for preventing π–π stacking caused by the self-aggregation effect in the solid state and hence improving the sensing performance. In addition, both S1 and S2 have very high fluorescence quantum yields (Φ) of nearly 1 in their THF solutions. This is also an important factor for sensing materials. Moreover, the Stokes shifts of S1 and S3 are larger than that of S2 in both the solution state and solid state, which can prevent self-absorption by the material and reduce the interference of the background to improve the sensitivity. Last but not least, there are plenty of internal cavities³¹ and external boron ester groups in the hyperbranched configuration of S1 which make important contributions to increasing the permeability of analytes and the response rate of the sensing process.

The energy level is related to the probe’s reactivity to oxidation reagents. The electrochemical results (Table 1) reveal that the HOMOs of S1, S2 and S3 are −5.49, −5.99 and −6.04 eV, respectively, suggesting a stepwise decrease in the HOMO from S1 to S3. Their band gaps are 2.80, 3.21 and 3.25 eV, with that of S1 being the least among them. These features mean that S1 will likely be the most reactive polymer toward oxidation. The hyperbranched molecular structure and increased number of borate ester groups on the periphery of structure of S1 make conjugation more efficient and make it more reactive than the linear polymers S2 and S3. At the same time, the narrow band gap means that S1 needs excitation with light of a longer wavelength. Thus, it can not only reduce the cost of the detector but also improve the light-stability of the material. All the optical and electrochemical characteristics above suggest that the hyperbranched polymer S1 is likely a promising candidate for fluorescent sensing applications.

3.2 Detection of peroxide in the vapor phase

The sensing performances of the three polymers to peroxide vapor were monitored by fluorescence spectroscopy. The films were fabricated by dip-coating their toluene solutions, at a concentration of 10⁻⁴ g mL⁻¹, onto quartz plates. The normalized peak emission intensity change in air or explosive vapour with time is shown in Fig. 3. It shows that the quenching efficiencies of the three probes towards the saturated vapor of 30% aqueous H₂O₂ are 40% (S1), 24% (S2) and 23% (S3). As can be seen, the hyperbranched polymer S1 shows a much better sensing performance than those of the linear polymers S2 and S3.

Fig. 3 (a–c) Stability and sensing properties of S1–S3 films on different substrates exposed to air or saturated H₂O₂ vapor for 300 s (a: S1; b: S2; c: S3); (d) quenching efficiency of S1–S3 films on different substrates in saturated H₂O₂ vapor within 300 s.

Since the sensing performance is also related to their film morphology, SEM was used for the morphology characterization. Fig. 4 shows that a more uniform film is formed for the hyperbranched polymer S1, while S2 and S3 suffered from heavy aggregation. The SEM results suggest that the linear polymer structure with polar borate ester end groups will lead to serious aggregation, which not only prevents the penetration of H₂O₂ vapor but also reduces the specific surface area of the films, resulting in a lower sensitivity.

Fig. 4 SEM images of S1–S3 films prepared with toluene solvent (all concentrations, 10⁻³ g mL⁻¹) and ZnO nanorod array as film substrate (a: S1; b: S2; c: S3; d: ZnO nanorod array).

The morphology of the substrate may also greatly influence the sensitivity, since a high specific surface area and surface activity of the films will contribute to the sensing performance. Our group has successfully developed ZnO nanorod arrays with different surface morphologies, which were proven to improve the sensing efficiency via catalytic functionality to accelerate the reaction rate and an evanescent effect to increase the emission intensity.²⁹ Therefore, we coated the fluorescent materials on the surface of ZnO nanorod arrays (Fig. 4(d)) vertically arranged on a quartz plate to make a composite. Fig. 3(a) shows that the S1/ZnO nanorod array composite produces enhanced fluorescence quenching with a response rate of about 30% within 50 s and finally 60% within 300 s upon exposure to vapor of 30% aqueous H₂O₂, corresponding to a 20% increase relative to that on quartz plate. Under the same conditions, the quenching efficiency of the hyperbranched polymer S1 with H₂O₂ vapor is also much better than that of the linear polymers S2 and S3 (43% and 30% quenching efficiency within 300 s, respectively). This demonstrates that the hyperbranched polymer S1, with more external boron ester end groups and internal cavities, is conducive to peroxide vapor penetration and contributes to more effective detection.

In order to interpret the sensing mechanism, both fluorescence and high vacuum reflection–absorption infrared spectra (IRAS) were used to monitor the changes in the S1/ZnO nanorod array composite film before and after H₂O₂ vapor exposure. Fig. 5(a) shows that emission changes occur for S1 by dissolving the composite film into THF after exposing it to H₂O₂ for a period of 20 minutes. The emission peak of S1 declines gradually with a slight red shift from 461 to 465 nm. Fig. 5(b) show the IRAS changes: after the exposure, the band at 1343 cm⁻¹ (characteristic peak of B–O) disappears, along with the appearance of a new band at 3253 cm⁻¹ (characteristic peak of O–H). This proves that a deboronation reaction occurred between the external boron ester end groups of S1 and peroxide vapor leading to the fluorescence turn-off, as shown in Fig. 5(c). In addition, the fluorescent conjugated structure of the hyperbranched polymer S1 can amplify the fluorescence signal due to its effective conjugation and multiple borate esters structure. On the other hand, the ZnO nanorod arrays not only change the surface morphology of the film but also increase the reactivity to H₂O₂ vapor, which will accelerate the deboronation reaction of S1 under UV light.³⁰

Fig. 5 (a) Changes in the emission spectral pattern of S1 film on ZnO nanorod array during reaction with saturated vapor of H₂O₂, followed by dissolving the film in THF solvent within the reaction time of 20 minutes. (b) High vacuum infrared spectra of S1/ZnO nanorod array composite dissolved in THF before (red) and after (black) exposure to hydrogen peroxide vapor for 1 h. (c) Deboronation reaction of boron ester group by H₂O₂.

The light-stability will determine the lifetime of the sensing devices. The photo bleaching of S1 on a quartz plate is measured to be ∼7% after 300 s under air. However, the light bleaching of the S1/ZnO nanorod array composite can be well controlled within 2%, suggesting that the ZnO nanorod arrays are also favorable for enhancing the stability of the material. Nevertheless, compared with the good stability of S1 and S2, the photo bleaching rate of S3 film either on a quartz plate or ZnO nanorod array is not ideal (∼9%). The results illustrate that the connection position of the borate ester groups has an obvious influence on the light-stability and the sensing performance of the sensing films. A directly connected borate ester on fluorene causes better light stability and sensing performance than that of pyrene. This is likely related to the increased intermolecular interaction among pyrene molecules due to the polar borate ester units. The S3 film is very difficult to used as an efficient sensor due to the decreased photo stability, which leads to a short lifetime and false alarm signals of the detection device.

The selectivity is another important issue for the sensing performance. Fig. 6 shows the fluorescence responses of the S1/ZnO composite to saturated vapor of different analytes. The S1 film presents an obvious fluorescence quenching response to H₂O₂ (∼60%) and TATP (∼30%). It is mentionable that compared with H₂O₂, TATP is hard to detect with high sensitivity owing to the lower saturation vapor pressure (78 ppm). Hence, S1 will be a promising fluorescent probe for TATP vapor detection. The other commonly used solvent vapors, such as methylene chloride, acetone and ethyl alcohol, show very low quenching efficiencies, which will not interfere with the sensing process. It is important to note that some solvents such as toluene can lead to enhanced fluorescence intensity due to a swelling effect. However, this will not produce interference in the peroxide detection based on the fluorescence quenching process. In addition, the quenching efficiencies of S1 with some saturated vapor mixtures of different analytes after an exposure of 300 s are all less than 15%, so this will not interfere with the sensing process (as shown as Fig. S4†).

Fig. 6 Fluorescence responses of S1 film on ZnO nanorod array to saturated vapor of different compounds after an exposure of 300 s.

In order to determine the detection limit of the S1/ZnO nanorod array composite film, different concentrations of H₂O₂ solution were prepared by diluting the 30 wt% H₂O₂ solution with deionized water to produce a corresponding equilibrium vapor pressure (1, 19, 38 and 225 ppm, corresponding to 90, 10, 5, and 1 times dilution of the commercial 30 wt% H₂O₂ solution) at room temperature (25 °C).³² The quenching efficiencies towards different concentrations of H₂O₂ vapor are shown in Fig. 7(a). Then, the data of the different quenching rates and corresponding equilibrium vapor pressures were given a linear fit, which is well-fitted to the Langmuir equation. As shown in Fig. 7(b), the detection limit of H₂O₂ could be as low as 1.6 ppb if the triple multiple signal to noise ratio of the fluorescent detection device was considered as 0.01.

Fig. 7 (a) Quenching efficiency of the S1 film on ZnO nanorod array with different vapor pressures of H₂O₂ after an exposure of 300 s. (b) Quenching efficiency of the S1 film on ZnO nanorod array as a function of the vapor pressure of H₂O₂, fitted with the Langmuir equation.

4. Conclusions

In summary, we reported a relatively simple, sensitive and selective borate ester endcapped pyrene–fluorene hyperbranched copolymer for fluorescent detection of peroxide explosive vapor. The sensing performance is based on a deboronation reaction under peroxide and a signal amplification effect of the polymer resulting in fluorescence quenching, which is related to the molecular structure, substitution position on the aromatic ring and the number of borate units. A composite of the sensory polymer/ZnO nanorod array can improve the stability and sensitivity of the material because of the catalytic light oxidation ability and surface morphology, with a high area to volume ratio for efficient analyte permeability. The results show that compared with the two linear analogues S2 and S3, the hyperbranched polymer S1 demonstrates a more sensitive response toward peroxide (∼60% quenching and 30% for TATP within 300 s) due to the larger steric hindrance, higher HOMO level, greater number of internal cavities and external boron ester groups of the hyperbranched structure. The detection limit for H₂O₂ vapor is estimated to be 1.6 ppb. This method provides a new method for trace and on-site chemical detection of peroxide-based explosives (PEs) for human health and public safety.

Acknowledgements

This work is supported by the National Natural Sciences Foundation of China (nos 61325001, 21273267, 61321492, 51473182), the Shanghai Science and Technology Committee (no. 11JC1414700), and Shanghai Municipal Commission of Economy and Informatization. We would also like to express our thanks to Mr Pengcheng Xu for IR measurements and helpful discussions.

Notes and references

1. S. Meyer, Secure J., 2012, 25, 309–325.
2. J. Roach, P. Ekblom and R. Flynn, Secure J., 2005, 18, 7–25.
3. S. W. Thomas, G. D. Joly and T. M. Swager, Chem. Rev., 2007, 107, 1339–1386.
4. M. E. Germain and M. J. Knapp, Chem. Soc. Rev., 2009, 38, 2543–2555.
5. T. Naddo, Y. Che, W. Zhang, K. Balakrishnan, X. Yang, M. Yen, J. Zhao, J. S. Moore and L. Zang, J. Am. Chem. Soc., 2007, 129, 6978–6979.
6. C. Zhang, Y. Che, X. Yang, B. R. Bunes and L. Zang, Chem. Commun., 2010, 46, 5560–5562.
7. J. W. Grate, R. G. Ewing and D. A. Atkinson, Trends Anal. Chem., 2012, 41, 1–14.
8. C. Chen, R. Y. Wang, L. Q. Guo, N. Y. Fu, H. J. Dong and Y. F. Yuan, Org. Lett., 2011, 13, 1162–1165.
9. Y. Liu, C. Deng, L. Tang, A. Qin, R. Hu, J. Z. Sun and B. Z. Tang, J. Am. Chem. Soc., 2011, 133, 660–663.
10. S. Shanmugaraju, S. A. Joshia and P. S. Mukherjee, J. Mater. Chem., 2011, 21, 9130–9138.
11. H. N. Kim, W. X. Ren, J. S. Kim and J. Yoon, Chem. Soc. Rev., 2012, 41, 3210–3244.
12. Z. Huang, J. Du, J. Zhang, X. Q. Yu and L. Pu, Chem. Commun., 2012, 48, 3412–3414.
13. C. Rena, J. S. Leea and T. E. Glassa, Supramol. Chem., 2014, 26, 607–611.
14. R. Schulte-Ladbeck, M. Vogel and U. Karst, Anal. Bioanal. Chem., 2006, 386, 559–565.
15. G. M. Whit, J. Forensic Sci., 1992, 37, 652–656.
16. G. J. McKay, Kayaku Gakkaishi, 2002, 63, 323–329.
17. P. Wilson, B. Prince and M. McEwan, Anal. Chem., 2006, 78, 575–579.
18. F. Dubnikova, R. Kosloff, J. Almog, Y. Zeiri, R. Boese, H. Itzhaky, A. Alt and E. Keinan, J. Am. Chem. Soc., 2005, 127, 1146–1159.
19. M. E. Germain and M. J. Knapp, Inorg. Chem., 2008, 47, 9748–9750.
20. F. Hu, Y. Y. Huang, G. X. Zhang, R. Zhao and D. Q. Zhang, Tetrahedron Lett., 2014, 55, 1471–1474.
21. Y. Ling, N. Zhang, F. Qu, T. Wen, Z. F. Gao, N. B. Li and H. Q. Luo, Spectrochim. Acta, Part A, 2014, 118, 315–320.
22. M. W. Todd, R. A. Provencal, T. G. Owano, B. A. Paldus, A. Kachanov, K. L. Vodopyanov, M. Hunter, S. L. Coy, J. I. Steinfeld and J. T. Arnold, Appl. Phys. B, 2002, 75, 367–376.
23. D. Muller, A. Levy, R. Shelef, S. Abramovich-Bar, D. Sonenfeld and J. Tamiri, Forensic Sci., 2004, 49, 935–938.
24. P. F. Wilson, B. J. Prince and M. J. McEwan, Anal. Chem., 2006, 78, 575–579.
25. G. A. Buttigieg, A. K. Knight, S. Denson, C. Pommier and M. B. Denton, Forensic Sci. Int., 2003, 135, 53–59.
26. M. Xu, J. M. Han, Y. Q. Zhang, X. M. Yang and L. Zang, Chem. Commun., 2013, 49, 11779–11781.
27. M. Xu, J. M. Han, C. Wang, X. M. Yang, J. Pei and L. Zang, ACS Appl. Mater. Interfaces, 2014, 6, 8708–8714.
28. A. K. Patri, I. J. Majoros and J. R. Baker, Curr. Opin. Chem. Biol., 2002, 6, 466–471.
29. D. F. Zhu, Q. G. He, Q. Chen, Y. Y. Fu, C. He, L. Q. Shi, X. Meng, C. M. Deng, H. M. Cao and J. G. Cheng, ACS Nano, 2011, 5, 4293–4299.
30. C. He, D. F. Zhu, Q. G. He, L. Q. Shi, Y. Y. Fu, D. Wen, H. M. Cao and J. G. Cheng, Chem. Commun., 2012, 48, 5739–5741.
31. Q. G. He, H. M. Huang, F. L. Bai and Y. Cao, Macromol. Rapid Commun., 2006, 27, 302–305.
32. S. L. Manatt and M. R. R. Manatt, Chem.–Eur. J., 2004, 10, 6540–6557.

---

Footnote

† Electronic supplementary information (ESI) available: Fourier transform infrared spectra of S3, S2, S1 and the contrast, the GPC and CV curves of S1–S3 polymer, the quenching efficiency (1 − I/I₀) of the S1 film on ZnO nanorod array for different vapor pressures of H₂O₂ after an exposure time of 300 s. See DOI: 10.1039/c5ra02472b

---