Polyaniline nanofibers and their self-assembly into a film to be used as ammonia sensor

Abstract

In this research, polyaniline (PANI) nanofibers were chemically synthesized by the hydrochloric acid (HCl) assisted interfacial polymerization method, where the morphologies and electrical properties of the nanofibers were controlled by the concentration of the HCl solution. The as-synthesized PANI nanofibers were assembled into a sensitive film with controlled thickness and nice uniformity using a liquid–gas interfacial self-assembly method. Field emission scanning electron microscopy, Fourier transform-infrared spectroscopy, and X-ray diffraction were used to characterize the PANI nanofibers. The sensing experiments were performed at room temperature with the injection of ammonia at different concentrations. All the sensors showed a good linear response to the target gas, especially for the sample of PANI doped with [HCl] = 1 mol L⁻¹, for which the reproducibility, continuity and selectivity were further tested. The results obtained in this research will be useful for potential applications in related fields of ammonia detection.

---

1. Introduction

Ammonia (NH₃) exists widely in our daily life both in nature and industrial areas and examples include animals and plants rotting, industrial nitrogen fixation and concrete fillers. NH₃ is harmful to the human body; when people are exposed to low concentrations of NH₃ it can harm the nasal cavity and skin whereas at high concentrations it can even lead to death.^1,2 However, NH₃ can react with ˙OH and ˙NO₃ radicals and ozone which exists widely in atmosphere in a complex way to generate some harmful and toxic products, such as formaldehyde, formic acid and hydrogen cyanide^3,4 all of which affect human health and the environment. So it is quite important to be able to have rapid, trace detection of NH₃. The semiconductor gas sensor offers a good choice for detection of NH₃ with several advantages in terms of high sensitivity, fast response, low cost, flexibility, and simplicity in fabrication.⁵ In the past years, nanostructured materials such as tungsten trioxide,⁶ tin dioxide,⁷ and titanium dioxide⁸ and some other conducting polymers^9,10 have been widely fabricated and used as gas sensors. Among those materials, conducting polymers such as polypyrrole,¹¹ polythiophene,¹² and polyaniline (PANI)¹³ have shown to be superior in some aspects such as low working temperature in contrast to the traditional metal oxide semiconductor materials^5,14 which often need a high working temperature because of the sensing mechanism.¹⁵ In all the conjugated conducting polymers, PANI has attracted a lot attention because of its easy synthesis, environmental stability, and simple acid/base reaction property.^16,17 Furthermore, it can selectively react with NH₃ at room temperature which leads to a change of conductivity, so it is of potential application in creating a gas sensor which will operate at room temperature for specific detection and monitoring. Recently, various PANI nanostructures such as nanorods,¹⁸ nanospheres,¹⁹ nanowires,²⁰ and nanofibers²¹ have been synthesized. Compared to special structure directing materials added to the polymerization reactor, which serve as hard templates, such as porous membranes,²² PANI nanofibers were quite easy to obtain under the easily controlled reaction condition. Li et al.¹⁶ have used a simple interfacial polymerization method to obtain PANI nanofibers with high uniformity in size. In the method of interfacial polymerization, no hard templates and organic acids are needed. Common mineral acids such hydrochloric acid (HCl), sulfuric acid or nitric acid can be used as dopants to obtain high quality PANI nanofibers.²³ The easily fabricated PANI nanofibers supply the fundamental sensing materials. To build the gas sensors, another important problem is how to fabricate nanomaterial-based thin films on the substrates of the devices. The common methods used to deposit nanomaterial thin films on the gas sensing devices are drop-casting²⁴ or screen-printing.²⁵ Although these methods have broad applicability for many materials, they suffer from inhomogeneity in film thickness and poor reproducibility. To solve this problem a simple water–gas interface self-assembly method was devised. PANI nanofibers were self-assembled, arranged on the water surface to form a uniform thin film and were then easily transferred onto the substrate of the sensor devices, in which the thickness and the reproducibility were well controlled. Fig. 1 shows the schematic illustration of the fabrication of PANI film on the wafer substrate, and the same route is used to fabricate PANI film on the sensor substrate. The sensing results indicate that the self-assembly PANI nanofiber thin film shows good sensing performances to NH₃ sensing. The fabrication processes and the sensing properties are discussed in detail in the following sections.

Fig. 1 Schematic illustration of the fabrication of PANI film on a wafer substrate: (a) dropping the as-prepared PANI/water/ethanol mixture onto the water surface; (b) self-assembly of PANI on the water surface; (c) the self-assembled PANI film is picked up by a wafer substrate; (d) the wafer substrate coated with PANI film.

2. Experimental

2.1. Materials

Aniline (C₆H₇N), ammonium persulfate [(NH₄)₂S₂O₈, APS], carbon tetrachloride (CCl₄), potassium bromide (KBr) and HCl were all purchased from the Sinopharm Chemical Reagent Company. All the chemical reagents were analytical grade or better and were used directly without any further purification. Deionized water (DI water, 18.2 MΩ cm) was obtained using an ultrafilter system (Milli-Q, Millipore, Marlborough, MA).

2.2. Synthesis of PANI nanofibers

The PANI nanofibers were synthesized using the interfacial polymerization method.²⁶ In a typical synthesis strategy, 10 ml of CCl₄ was used as organic phase with 3.2 mmol aniline dissolved in it by ultrasonic processing for 60 min, 10 ml of HCl solution of a specific concentration with 0.8 mmol of APS dissolved in it was used as the water phase. Then the APS/HCl solution was carefully transferred onto the surface of an aniline/CCl₄ solution, and an interface was soon formed. The reaction was carried out at room temperature for 24 h. The upper phase was then collected and followed by washing alternately with DI water and ethanol, and it was then centrifuged and dried under vacuum at 45 °C for 24 h. The dried samples were stored for further use.

2.3. Water–gas interface self-assembly and construction of sensors

A portion (10 mg) of the dried powder was dispersed in 5 ml of DI water and ethanol mixture (50% : 50% v/v), which was then treated ultrasonically for 60 min to obtain a green coloured mixture with a concentration of 2 mg ml⁻¹. Then 50 uL of the well dispersed mixture was carefully dropped from one side on the DI water surface in a clean beaker with a diameter of 5 cm. It was observed that the green mixture spread out on the water surface and a thin film was then formed.

To fabricate the sensors, the self-assembled thin film was transferred onto the flat electrode. The distance between the two electrodes of the flat electrode was 200 μm. The fabrication route is shown in Fig. 1. A piece of flat electrode or wafer was dipped into the water and then used to slowly scoop up the thin film which was then dried in air at room temperature.

2.4. Characterization

The morphologies of the as-prepared samples were observed using field-emission scanning electron microscopy (FESEM, FEI Sirion 200), and some samples were tilted on a worktable for a cross view. X-ray diffraction (XRD) was measured with an X-ray diffractometer (Philips X'Pert) with a Cu-Kα line (0.15419 nm). Fourier transform-infrared (FTIR) spectra of the samples were obtained using a Nicolet MX-1E FT spectrometer with a KBr tablet. The optical photographs of the self-assembly processing and the fabrication procedures were obtained using a digital camera (Sony RX100).

2.5. Measurement of gas sensing performances

The gas sensing experiments were carried out in a static system (a cuboid chamber with a volume of 20 L with a multimeter (Agilent U3606 A) and a DC power supply (Agilent U8002A)) by measuring the electrical voltage changes of the sensing devices under different conditions. The testing gas atmosphere was made by injecting the specific gas. A certain amount of NH₃ was injected into the chamber with devices in the system and the NH₃ concentration in the chamber was calculated, based on the injected amount. All the measurements were carried out at 25 °C in ambient air.

3. Results and discussion

3.1. Fabrication and characterization of PANI

Fig. 2 shows the XRD patterns of the samples synthesized with different concentrations of HCl. The results for all the samples are in good agreement with those found in the literature^27. In general, all the samples exhibited an amorphous phase, indicating that the degree of crystallinity of the samples was not high. Two broad peaks centered at about 2θ = 19° and 26° were observed, which were in good agreement with the (200) diffraction plane of the ES-1 structure of HCl doped PANI.²⁸ The crystallinity of the PANI could be ascribed to the repetition of benzenoid and quinoid rings in the polymer chains, which indicated the doping level of HCl.²⁹ It could be seen that the intensity of the peak at 19° and 26° were highest for the PANI doped with [HCl] = 1 mol L⁻¹, which indicates that the sample has a higher molecular alignment compared to the other samples.³⁰ The higher molecular alignment means that the electrons can transport along the polymer chain more easily than the others.³¹

Fig. 2 XRD patterns of PANI prepared using different concentrations of HCl.

As is already known, the band structure change between atoms and groups of atoms in the molecule can lead to the change of the characteristic vibrational frequency. The structure of the synthesized PANI was further demonstrated in this research using FTIR spectroscopy in the range of 400–4000 cm⁻¹ as shown in Fig. 3. As can be seen, the broad band at wave numbers >3200 cm⁻¹ is from the N–H stretching,³² which represents the electronic extended conjugation of the PANI salt.³³ In this test, the quantity of every sample is almost the same, so it can rule out the influence of the quantity of the samples. It was found that the peak intensity increased with the increase of the HCl concentration ranging from 0.1 mol L⁻¹ to 1 mol L⁻¹. Whereas by further increasing the HCl concentration to 2 mol L⁻¹, the peak intensity decreased. The peak intensity also represented the electronic extended conjugation of the PANI salt, which reflects the polymer growth and order of the polymer chains. The higher the intensity, the longer and more ordered the polymer chain becomes. It can also be seen that the absorption shifted to the lowest wave number when the HCl concentration was 1 mol L⁻¹. This proves that the HCl concentration of 1 mol L⁻¹ produced the optimum level of the conjugation length. The other main characteristic peaks are assigned as follows: the bonds at 1565 and 1488 cm⁻¹ are attributed to C=N and C=C stretching mode of vibration for the quinonoid and benzenoid units of PANI, respectively, and the bonds at 1291 cm⁻¹ and 1133 cm⁻¹ are the stretching peak of C–N and C=N, respectively. The band at 797 cm⁻¹ is assigned to the C–H bending vibration out of the plane of the para-disubstituted benzene rings. In addition, the intensity ratio of the peak at 1584 and 1495 cm⁻¹ (quinoid band/benzenoid band) is a little higher when compared with the other samples which may indicate that PANI chains doped with a [HCl] of 1 mol L⁻¹ is much richer in quinoid units than the others. It can also be assumed that PANI doped with [HCl] of 1 mol L⁻¹ has a longer conjugation length as a result of the higher concentration of HCl on its polymer chain.³⁰

Fig. 3 FTIR spectra of PANI prepared using different concentrations of HCl: (a) 0.1 mol L⁻¹, (b) 0.5 mol L⁻¹, (c) 1 mol L⁻¹, (d) 2 mol L⁻¹.

Fig. 4 shows the typical FESEM images of the synthesized PANI samples, and as can be seen, all the PANI samples are nanofibers under the described conditions. According to the relevant literature,³⁴ interfacial polymerization tends to produce PANI nanofibers whereas the traditional oxidative chemical polymerization tends to yield granular PANI. In the previous work by Zhang et al.,³⁵ the type of doping acid does not affect the formation of the PANI nanofibers, whereas the concentration of the doping acid does affect it. It can be seen that the morphology of the PANI shows large changes at different concentrations of HCl. When [HCl] = 0.1 mol L⁻¹, the nanofibers seem to be growing on top of each other and show a vague edge, and, the nanofibers are short and coarse. When the [HCl] = 0.5 mol L⁻¹ and 1 mol L⁻¹, clear and angular nanofibers are obtained with a diameter of 50 nm, and with the concentration of 0.5 mol L⁻¹ are much longer. Interestingly, when further increasing the concentration of HCl to 2 mol L⁻¹, the obtained products seem to agglomerate with each other. The reasons why the concentration of HCl affects the morphology may contribute to the nucleation–diffusion mechanism.³⁶ It is known that there are two types of nucleation for the formation of PANI, namely, homogeneous nucleation and heterogeneous nucleation. If heterogeneous nucleation plays a dominant role, the new molecules will grow on the formed nanofibers, which results in the agglomerated particles. In contrast, homogeneous nucleation tends to produce well dispersed nanofibers. In the interfacial polymerization process, PANI forms only at the interface and then moves into the water phase even though its intensity is much higher than water while at the same time the reaction terminates. The diffusion ratio of the synthesized nanofibers in the HCl solution of different concentrations is typically different. With the increase of the HCl concentration, the number of ions in the solution increases; as a result, the mutual repulsion between the ion atmosphere, and thus the viscosity of the HCl solution, increases. For the positively charged PANI polymers, it is hard for diffusion to occur in the HCl solution with a higher concentration. When the water phase has a high concentration, the synthesized PANI nanofibers cannot diffuse immediately and supply the sites for the secondary growth.³⁶ As a result, the concentration of HCl determines the morphology of the PANI nanofibers.

Fig. 4 FESEM images of PANI prepared using different concentrations of HCl: (a) 0.1 mol L⁻¹, (b) 0.5 mol L⁻¹, (c) 1 mol L⁻¹, (d) 2 mol L⁻¹.

3.2. PANI sensors by self-assembly and their sensing performance to NH₃

As described previously, the PANI nanofiber film created using self-assembly was transferred on to a wafer for FESEM characterization, the results of which are shown in Fig. 5. From the top view, it can be seen that the nanofibers ranged together to form a rather uniform film. The film showed porous nature. From the cross view it can be seen that the thickness of the fabricated thin films was about 1 μm which can be well controlled.

Fig. 5 FESEM images of the PANI self-assembled films prepared using different concentrations of HCl: (a) 0.1 mol L⁻¹, (b) 0.5 mol L⁻¹, (c) 1 mol L⁻¹, (d) 2 mol L⁻¹; the insets show the corresponding cross-sectional view.

To fabricate sensing devices, PANI self-assembly film was transferred on to a flat electrode which acted as the substrate of the device, which was used as sensor for next measurement. The NH₃-sensing properties of the PANI self-assembly film sensors were measured in ambient air at room temperature. In this research, the gas sensing response is defined as R_g/R_o, where R_o and R_g are the resistances of the sensing film in the device before and after exposure to NH₃, respectively. Fig. 6 shows the gas sensing response of the PANI self-assembly film sensors exposed to NH₃ of various concentrations at room temperature. It can be seen from the image that the gas sensing response of the sensors increases dramatically when the sensors were exposed to NH₃ and decreases immediately as the NH₃ was pumped out. The response of the [HCl] = 1 mol L⁻¹ doped PANI nanofibers sensor is much higher than that of the others. In this research, the response time was defined as the time for reaching 90% of the full response change of the sensor after the testing gas is added and the recovery time is defined as the time for falling to 10% of its maximum response after the testing gas is removed. As can be calculated, the response time is about 10 s, and the recovery time is about 300 s. The gas response versus concentration of NH₃ is shown in Fig. 7. From the image it can be seen that the responses of the sensors to the increasing concentration of NH₃ are nearly linear, the [HCl] = 1 mol L⁻¹ doped PANI sensor shows superiority in sensitivity.

Fig. 6 The responses (R_g/R_o) of PANI self-assembly thin film sensors to various concentrations of NH₃ at room temperature, with the concentration ranging from 20 ppm to 100 ppm. The sensors are indexed in the patterns with different color and noted as (a)–(d): (a) 0.1 mol L⁻¹, (b) 0.5 mol L⁻¹, (c) 1 mol L⁻¹, (d) 2 mol L⁻¹.

Fig. 7 The gas response (R_a/R_GO) comparison chart of the sensors to NH₃ at various concentrations.

In addition, it is also observed that the response decreases after the peak appeared when the sensor was exposed to NH₃, and the decrease becomes sharper with the increase of NH₃. As is known, the NH₃ catches the protons on the polymer chains and leads to the decrease of conductivity. Whereas the NH₃ molecule is much smaller than the HCl molecule, and the distance between the polymer chains is shortened, which benefits the increase of conductivity.³⁷

Fig. 8(a) shows the chemical reaction equation of the formation of PANI. The HCl molecule transports the nitrogen atom of the imine group, which is known as the doping (protonation) process.²⁴ The PANI molecule gains protons from the HCl molecule to form N⁺–H chemical bonds, and as a result the positively charged local centers located at the nitrogen atoms are formed. Valance electrons can hop from one center to another, along the polymer chain or between the adjacent molecules, leading to conduction. So it is obvious that the conjugation length and the regularity of the polymers influence the conduction of PANI.³ During the polymerization process, the concentration of HCl affects the doping level and the crystallinity of PANI, and thus influences the conductivity.

Fig. 8 (a) Formation reaction of HCl doped PANI, and (b) the potential interaction between NH₃ and PANI.

Fig. 8(b) shows the interaction between NH₃ and PANI. When PANI interacts with NH₃, the unshared pair electrons from the N atoms of the polymer backbones are released. However, the NH₃ molecule is polar in nature, and the electronegativity value of N is larger than that of H, and as a result, partial polar bonds between the N and H atoms are formed, which finally destroy the conjugation along the polymer chain and increase the inner chain distance in PANI. The destruction of the conjugation along the polymer chain and the increase of the inner chain distance influences the electron transport along the chain and hopping between the adjacent polymer chains; as a result, the resistance of the PANI film increases and the emeraldine salt state is converted to the emeraldine base state. When the NH₃ is removed from the test chamber, the N atoms of the PANI chains regain the unshared pair electrons and the emeraldine base state is converted to the emeraldine salt state, because of which the conductivity of PANI is restored.³⁸ This reversible process can be simplified into the following eqn (1):

PANI–H⁺ + NH₃ ↔ PANI + NH₄⁺ (1)

When in the presence of NH₃, this reaction equilibrium goes predominantly towards the right hand side, but when in ambient air, the reaction equilibrium is restored.³⁹

So the dopant on the polymer chains also affects the sensing performance. The longer conjugation length and the higher crystallinity of the polymer result in a higher sensitivity to NH₃.⁴⁰ When the concentration of HCl is increased from 0.1 mol L⁻¹ to 1 mol L⁻¹, the doping amount increased and leads to the increase of the conjugation length. By further increasing the concentration of HCl, more Cl⁻ combines with the PANI chains and thus enhances the steric effects between the polymer chains. So further increasing the concentration of HCl reduced the crystallinity of the polymers. These were confirmed from the analysis of the FTIR spectra and XRD patterns. Depending on the previous discussions, the PANI doped with an [HCl] of 1 mol L⁻¹ shows superiority in NH₃ sensing when compared with PANI doped with other concentrations.

Sensing reproducibility of the sensor prepared using [HCl] = 1 mol L⁻¹ was further tested, by exposing the sensor to 100 ppm of NH₃ with five cycles at room temperature, the results of which are shown in Fig. 9. From the image it can be seen that the gas sensing responses to 100 ppm NH₃ are all about 7, which indicates that the sensor exhibits good reproducibility and reversibility.

Fig. 9 The sensing reproducibility test of the PANI self-assembly thin film sensor prepared at [HCl] = 1 mol L⁻¹. The test was carried out by exposing the sensor to 100 ppm NH₃ over five cycles at room temperature.

The sensing performance of the sensor to ammonia was further tested using lower concentrations such as 1.0, 0.8, 0.6, 0.4 and 0.2 ppm (results shown in Fig. 10), and the results showed that the prepared sensor with PANI nanofibers doped with [HCl] of 1 mol L⁻¹ can detect ammonia concentrations as low as 0.2 ppm with a sensitivity of 1.03.

Fig. 10 A continuous test using 0.2–1 ppm NH₃ concentration at room temperature (RT) for the PANI self-assembly thin film sensor prepared at [HCl] = 1 mol L⁻¹.

Generally, it is necessary to test the selectivity of the chemical sensor. Nitrogen dioxide (NO₂), acetone (C₃H₆O), hydrogen sulfide (H₂S), methane (CH₄), ethanol (C₂H₆O), HCl, sulfur dioxide (SO₂) and hydrazine (NH₂NH₂) were tested as interfering gases. The real time response and sensitivity contrast histogram are shown in Fig. 11. According to the special sensing mechanism, H₂S enhances the protonation effect and thus exhibits the opposite response compared to the other gases. The CH₄ or C₂H₆O concentration of 100 ppm does almost not change the resistance of the sensor. The gas response to NO₂ and C₃H₆O concentrations of 100 ppm are 3 and 2, respectively, which are obviously lower than the gas response to the NH₃ concentration of 100 ppm. At the same time, the sensor shows little response to HCl, SO₂ and NH₂NH₂ gas concentrations of 100 ppm. The sensing performance to NO₂ and C₃H₆O is quite different to that of NH₃, which can be seen in Fig. 11(a), the response time of the sensor to NO₂ and C₃H₆O is quite long (about hundreds of seconds). However, the response time of the sensor to NH₃ is only about tens of seconds. So it can be concluded that the [HCl] = 1 mol L⁻¹ doped PANI nanofiber self-assembly sensor shows good selectivity to NH₃.

Fig. 11 The responses of the PANI nanofiber film sensor prepared by using 1 M HCl to various gases at a concentration of 100 ppm. (a) The real time patterns; (b) gas sensitivity response histogram.

4. Conclusion

In summary, the PANI nanofibers were synthesized in the presence of HCl at different concentrations using interfacial polymerization and the sensors were fabricated by transferring the self-assembly film onto the plate electrodes. Their sensing properties to NH₃ were studied in detail. It was found that when [HCl] = 1 mol L⁻¹, the corresponding sensor showed optimized sensing performances, which may be because of the high doping level. This work may provide a new construction route for PANI-based NH₃ sensors at room temperature.

Acknowledgements

The authors acknowledge the financial supports from National Key Research and Development Plan (2016YFC0201103), the financial supports from Natural Science Foundation of China (Grant No. 11674320 and 51471161), Anhui Provincial Natural Science Foundation for Distinguished Young Scholar (1408085J10), Youth Innovation Promotion Association CAS, and Key Research Projects of the Frontier Science CAS (QYZDB-SSW-JSC017).

References

1. B. Timmer, W. Olthuis and A. van den Berg, Sens. Actuators, B, 2005, 107, 666–677.
2. U. V. Patil, et al., Appl. Surf. Sci., 2015, 339, 69–74.
3. J.-C. Chiang and A. G. Macdiarmid, Synth. Met., 1986, 13, 193–205.
4. C.-J. Nielsen, H. Herrmann and C. Weller, Chem. Soc. Rev., 2012, 41, 6684–6704.
5. Z. Xu, et al., Chemistry, 2014, 20, 6040–6046.
6. G. Wang, et al., J. Phys. Chem. B, 2006, 110, 23777–23782.
7. J. Zhang, et al., J. Phys. Chem. C, 2009, 113, 1662–1665.
8. B. Karunagaran, et al., Mater. Charact., 2007, 58, 680–684.
9. H. Bai and G.-Q. Shi, Sensors, 2007, 7, 267–307.
10. J. Janata and M. Josowicz, Nat. Mater., 2003, 2, 19–24.
11. S. Bibi, et al., J. Phys. Chem. C, 2015, 119, 15994–16003.
12. S.-T. Navale, et al., Synth. Met., 2014, 195, 228–233.
13. N.-E. Agbor, M.-C. Petty and A.-P. Monkman, Sens. Actuators, B, 1995, 28, 173–179.
14. Z. Dai, et al., Sci. Rep., 2013, 3, 1669.
15. A.-L. Kukla, Y.-M. Shirshov and S. A. Piletsky, Sens. Actuators, B, 1996, 37, 135–140.
16. D. Li, J. Huang and R.-B. Kaner, Acc. Chem. Res., 2009, 42, 135–145.
17. M. Matsuguchi, et al., Synth. Met., 2002, 128, 15–19.
18. H.-B. Xia, et al., J. Phys. Chem. B, 2005, 109, 12677–12684.
19. X. Guo, et al., J. Mater. Chem., 2011, 21, 8618–8625.
20. C.-R. Martin, Chem. Mater., 1996, 8, 1739–1746.
21. J.-X. Huang, Pure Appl. Chem., 2006, 78, 15–27.
22. M. Wan, Macromol. Rapid Commun., 2009, 30, 963–975.
23. J.-X. Huang and R.-B. Kaner, J. Am. Chem. Soc., 2004, 126, 851–855.
24. S.-Z. Wu, et al., Eur. Polym. J., 2000, 36, 679–683.
25. N. Promphet, et al., Sens. Actuators, B, 2015, 207, 526–534.
26. J.-C. Michaelson and A.-J. McEvoy, J. Chem. Soc., Chem. Commun., 1994, 1, 79–80.
27. J.-P. Pouget, et al., Macromolecules, 1991, 24, 779–789.
28. H.-C. Pant, et al., Bull. Mater. Sci., 2006, 29, 379–384.
29. B.-K. Sharma, et al., J. Alloys Compd., 2009, 477, 370–373.
30. K.-P. Sambasevam, S. Mohamad and S.-W. Phang, Mater. Sci. Semicond. Process., 2015, 33, 24–31.
31. Y. Wei, R. Hariharan and S.-A. Patel, Macromolecules, 1990, 23, 758–764.
32. G.-M. Morales, et al., Polymer, 1997, 38, 5247–5250.
33. P.-P. Sengupta and B. Adhikari, Mater. Sci. Eng., A, 2007, 459, 278–285.
34. X.-B. Yan, et al., Sens. Actuators, B, 2007, 123, 107–113.
35. X.-Y. Zhang, et al., Synth. Met., 2004, 145, 23–29.
36. D. Li and R.-B. Kaner, J. Am. Chem. Soc., 2006, 128, 968–975.
37. Q.-Q. Lin, Y. Li and M.-J. Yang, Synth. Met., 2012, 162, 2242–2249.
38. M. Das and D. Sarkar, J. Mater. Sci.: Mater. Electron., 2016, 27, 4109–4119.
39. Z. Du, et al., J. Mater. Sci.: Mater. Electron., 2011, 22, 418–421.
40. H.-D. Zhang, et al., Sens. Actuators, A, 2014, 219, 123–127.

---