Humidity sensor based on electrospun MEH-PPV:PVP microstructured composite

Abstract

The present study demonstrates a solution-processed humidity sensor based on poly[2-methoxy-5-(2-ethylhexyloxy)-1,4-phenylenevinylene]:polyvinylpyrrolidone (MEH-PPV:PVP) organic microstructured composite thin film. The capacitive type humidity sensor has been fabricated in the surface type geometry of Al/MEH-PPV:PVP/Al, wherein organic composite (MEH-PPV:PVP) has been deposited onto the aluminium electrodes by electrospinning technique. The structural and morphological properties of organic thin film have been characterized by X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM) and atomic force microscopy (AFM). The humidity sensing properties of the sensor have been investigated at ∼1 volt AC operational bias, by measuring the capacitance as a function of a broad range (20–90%) of relative humidity (RH). The temperature and operational frequency dependency of the capacitance of the sensor has also been analyzed in detail. The proposed sensor exhibits high sensitivity (114 fF/% RH @ 100 Hz), small hysteresis (∼2% RH) and fast response (18 s and 8 s for adsorption and desorption processes, respectively). Compared with traditional spin-coated and drop-casted organic humidity sensors, the microstructure composite based sensor has demonstrated significantly improved sensing parameters, highlighting the unique advantages of the electrospinning process for humidity sensor fabrication. The possible humidity sensing mechanism of the proposed capacitive sensor has also been elaborated.

---

1. Introduction

The precise control and reliable estimation of the relative humidity in the atmosphere is a vital requirement for electronics industry, meteorology and pharmaceuticals manufacturing.^1,2 Recently, organic semiconductors have gained considerable attention for humidity sensing application because of their intrinsic hygroscopic property, low dielectric permittivity and interesting moisture dependent electrical characteristics.^3,4 In pursuit to develop promising humidity sensors, various transduction techniques such as capacitive, resistive, optical, field effect transistor (FET) and surface acoustic wave (SAW) have been exploited.^5–8 Each of these aforementioned techniques possesses own performances and specific conditions of application. Capacitive type humidity transduction however remains a dominant theme by virtue of the cost-effective and facile device design, linearity in response and low power dissipation.⁹ By virtue of these benefits, more than 75% of the miniaturized humidity sensors available in the commercial market use a capacitive technique.⁷

The development of a humidity sensor which exhibits high sensitivity, broad operating range, a good stability over utilization periods, negligible hysteresis and robust response and recovery time is highly desirable. Apart from these stringent requirements, the sensing layer of such ideal sensor should also not be susceptible to swell, shrink or peel off from the substrate at higher RH levels.¹⁰ Optimization of sensing parameters of humidity sensors therefore requires optimization of the sensing film in the first place. M. Matsuguchi et al.,¹¹ suggest that the morphology of the sensing materials should be taken into consideration to improve the water adsorption ability of the sensing film. Selection of the sensing material with improved water sorption capable morphology is of prime importance. H. Farahani et al.,¹² reported that the porosity of the humidity sensitive polymer thin film can be manipulated to improve the water sorption capability. So far, our previous research strives pertaining to polymer based humidity sensors have been based on spin-cast or drop-cast organic sensing layers.^3,13,14 Surface topographies obtained from these traditional spin-cast and drop-cast techniques are usually compact with negligible intergranular voids.¹⁵ In contrast to these traditional techniques, electrospinning is a facile and cost effective technique for the synthesis of nanofibers and microstructures. It is believed that such structures are promising candidates for humidity sensing application due to their unique characteristics viz., small size, developed porosity and large surface to volume ratio.

Present work is motivated by our recent report regarding influence of MEH-PPV concentration on the morphological and optical properties of electrospun MEH-PPV:PVP composite.¹⁶ In this aforementioned study, electrospun composite of 3 wt% MEH-PPV and 5 wt% PVP chloroform solution resulted in the formation of hexagon like structures with significant density of inter microstructure voids. We believe that the microstructures can provide an effective base for the clump of water molecules from the surroundings to subside and consequently change the electrical properties of the sensing film. Further, when the organic composite is exposed to ambient humidity, the voids may facilitate the humidity diffusion through the bulk. The present investigation has therefore been undertaken to characterize the water sorption behavior of MEH-PPV:PVP microstructured composite thin films for a capacitive-type humidity sensor. Capacitive type humidity sensor (Al/MEH-PPV:PVP/Al) in the surface type configuration have been fabricated. The capacitance of the sensor has been monitored as a function of varied relative humidity levels (% RH) at varied operational frequencies of AC input bias.

2. Experimental

Poly[2-methoxy-5-(2-ethylhexyloxy)-1,4-phenylenevinylene] (MEH-PPV) and polyvinylpyrrolidone (PVP) were purchased from Sigma Aldrich and were used without any further purification. Fig. 1(a) and (b) show molecular structures of MEH-PPV and PVP.

Fig. 1 Molecular structure of (a) poly[2-methoxy-5-(2-ethylhexyloxy)-1,4-phenylenevinylene] (MEH-PPV) and (b) polyvinylpyrrolidone (PVP).

For the device fabrication, commercially available glass substrates (dimension 25 × 25 mm) were primarily cleaned by sonication in soap water, deionized (DI) water, acetone, ethanol and DI water, respectively (15 min each). The substrates were later dried by nitrogen blow in a dust free environment. Thin film of aluminium (thickness ∼100 nm) was deposited on glass substrates by Edward auto 306 thermal evaporator through a shadow mask. A pair of rectangular aluminium electrodes separated by 40 μm was obtained on the glass substrate. During thermal deposition of aluminium thin film, the pressure inside the chamber was kept at 10⁻⁵ mbar, whereas the aluminium deposition rate was maintained at 0.1 nm s⁻¹.

3 wt% concentrated solution of MEH-PPV and 5 wt% concentrated solution of PVP were prepared separately in chloroform. Both solutions were mixed in 1 : 1 volumetric ratio and the resulting composite solution was stirred for 2 h to form a homogeneous solution. The composite solution was then loaded into a plastic syringe with a needle and electrospinning process was performed using a controller interfaced syringe pump (flow rate adjusted at 2.0 ml h⁻¹). Composite (MEH-PPV/PVP) microstructures were deposited to cover the separation (40 μm) between the aluminium electrodes. During the electrospinning process, applied bias of 20 kV was provided and the distance between the syringe needle tip and collector was set at 15 cm. The active sensing layer of MEH-PPV:PVP composite with an average thickness of ∼2 μm and the surface area ∼8 mm × 25 mm, was obtained by the electrospinning technique. The cross sectional view of the fabricated Al/MEH-PPV:PVP/Al surface type sensor is depicted in Fig. 2. Aluminium thin film was opted as the electrodes by virtue of its significantly higher adhesion to glass substrate¹⁷ which is usually not affected by repeated use of sensor.

Fig. 2 Schematic cross section view of Al/MEH-PPV:PVP/Al humidity sensor.

In the present study, crystallinity and structure of composite was characterized by X-ray diffraction (XRD) pattern obtained from the Panalytical Xpert pro. The surface morphology and microstructures distribution of the organic composite was studied by Hitachi SU-8000 field emission scanning electron microscope (FESEM) and Agilent Technologies 5500 atomic force microscope (AFM). The electrical characterization of the humidity sensor was performed in a laboratory assembled humidity control chamber (dimensions ∼40 cm × 60 cm). The inlet and outlet valves for the flow of humid air and dry nitrogen gas were provided in the humidity chamber. To increase the humidity level, the chamber was exposed to the humid air sourced by commercially available Rossmax NB80 humidifier. To decrease the RH level in the humidity chamber, dry nitrogen gas was allowed to enter through the inlet valve, whereas humid air was ejected out through an outlet valve. The capacitance of the fabricated sensor was monitored by GW Instek LCR-829 meter at varied humidity levels. To monitor the temperature and humidity levels inside the humidity chamber, Mastech MS6503 humidity meter (humidity sensitivity ±2.5% RH, temperature sensitivity ±0.7 °C) was utilized.

The repeatability of the electrical parameters of the sensor was monitored 7 times and was found within ±3%. All measurements were taken at ambient temperature (T ∼ 25 °C), however, to study the effect of temperature on the capacitance of the sensor, a hot plate was placed to vary the temperature of the sensor. Fig. 3, portrays the characterization setup utilized for the electrical characterization of humidity sensor at various humidity levels.

Fig. 3 Characterization setup, utilized for electrical characterization of the capacitive type Al/MEH-PPV:PVP/Al humidity sensor.

3. Results and discussion

The XRD patterns of pristine MEH-PPV thin film and the MEH-PPV:PVP microstructured composite in the range of 5° < 2θ < 80° are shown in Fig. 4. The broad hump at 2θ ∼25° in the MEH-PPV diffractogram suggests the amorphous nature of the polymer, which is in accordance to the previous finding.¹⁸ However, the XRD pattern of the MEH-PPV:PVP microstructured composite shows diffraction peaks at 30°, 35°, 50°, and 60° which verify the generation of crystalline structure in electrospun composite film.

Fig. 4 XRD patterns of pristine MEH-PPV polymer and MEH-PPV:PVP composite thin films.

The hygroscopic nature and surface morphology of the sensing film primarily determines the performance of the humidity sensors.¹⁹ Fig. 5 shows the FESEM images of the MEH-PPV:PVP composite thin film. At low magnification, 6k (shown in Fig. 5(a)), it can be observed that hexagonal shaped microstructures microstructures are distributed with indiscriminate orientation and micro-voids are contained in the composite thin film. These voids are believed to play critical role in adsorbing water vapors.²⁰ At higher magnification, 12k (see Fig. 5(b)), the average size of the microstructures can be easily estimated. The average size of the MEH-PPV:PVP microstructures is estimated to be ∼6 μm. In general, the FESEM micrographs of the sensing film, suggest surface irregularities and higher level of voids serving as admittance openings for humidity.²¹ The microstructured morphology obtained from electrospinning has led to an increased effective surface area of the sensing film. Such microstructured morphology is expected to allow greater ambient humidity exposure which ultimately would benefit to the humidity sensing property of the sensor.^22,23 The existence of microvoids distribution also suggests that swelling of polymer thin film is not expected at higher RH levels.²⁴

Fig. 5 FESEM micrographs of MEH-PPV:PVP thin film at (a) 6k and (b) 12k magnification, respectively.

Fig. 6(a)–(c) show the 2D, 3D AFM micrographs and cross section analysis of the MEH-PPV:PVP composite humidity sensing film, respectively. In general, the high resolution (3 × 3 μm) AFM images are commendably accorded with the aforementioned FESEM images and depict uneven surface morphology of the microstructures. The root mean square roughness value of the microstructures has been estimated to be ∼6.50 nm. The rough and irregular surface morphology (as portrayed by both FESEM and AFM micrographs) indicates that MEH-PPV:PVP microstructured composite is a potential matrix for humidity sensing application.

Fig. 6 (a) 2D micrograph (b) 3D AFM micrograph and (c) cross section analysis of electrospun MEH-PPV:PVP composite thin film.

Fig. 7, plots the capacitance–relative humidity relationship of Al/MEH-PPV:PVP/Al humidity sensor. At low RH levels, the value of capacitance is small and C–% RH curves are flat, indicating that capacitance is independent of humidity at low % RH levels. Below 20% RH, the coverage of humidity absorbed on the sensing surface is usually not continuous and restricted to immobile monolayer. The water adhesion on the sensing layer at this stage is characterized by the physical hydrogen bonds through the weak van der Waals interaction of water vapors with the polymer molecules.¹¹ When the adsorption of humidity on sensing film is progressively increased, physisorbed immobile monolayer transforms into mobile multilayers.²⁵ At this stage, the capacitance of the sensor starts increasing significantly and monotonically which indicates that electrical capacitance variation is closely related to the amount of adsorbed water molecules. Generally, the capacitive humidity sensors follow a nonlinear response as a function of relative humidity.^14,26 For the proposed sensor, with increasing RH from 20% to 90%, the capacitance of the device has increased by about 1.61 times in magnitude (f ∼ 100 Hz). The changes in capacitance as a function of relative humidity, at higher operational frequencies of input bias, i.e., 1 kHz and 10 kHz have been found to be 1.2 times, 1.03 times, respectively. The sensitivities of the proposed sensor (within 20 and 90% RH humidity bandwidth) at varied operational frequencies have been estimated to be 114 fF/% RH (100 Hz), 20 fF/% RH (1 kHz) and 5 fF/% RH (10 kHz). After 3 months aging of humidity sensor at ambient condition, a 2.4% decrease in capacitance was observed at 100 Hz operational frequencies, whereas 2% decrease was observed for higher orders of frequencies.

Fig. 7 Capacitance–relative humidity plots of Al/MEH-PPV–PVP/Al humidity sensor at 1 kHz, 10 kHz, 100 Hz (inset) operational frequencies of AC input bias.

The capacitance of the proposed humidity sensor (expressed as eqn (1)) depends upon the area of aluminium electrodes (A), separation between the aluminium electrodes (d), vacuum permittivity (ε_o) and relative dielectric permeability of the of the organic composite humidity sensing thin film (ε_d).

As observed from Fig. 7, the capacitance of surface type humidity sensor based on MEH-PPV–PVP thin film is highly sensitive to RH variations. When sensor is exposed to humidity, water vapors diffuse into the humidity sensing layer by virtue of the porous morphology of MEH-PPV:PVP composite. Since, the dielectric permittivity of water is ∼80, which is immensely higher than that of organic semiconductors (usually ∼5),²⁷ therefore, the water uptake in the sensing film is beneficial to increase the net dielectric permeability of the sensing film. Improved permeability of sensing film with rising RH levels, leads to higher electrical capacitance of the sensor. By virtue of the strong polarity of water molecules, capacitance measurement is therefore usually considered an effective means to characterize the water sorption behavior. The relation between dielectric constant and capacitance can be described by eqn (2).²⁸

where ε_d, and ε_w are the dielectric permittivity constants of the sensing film at dry and wet conditions, respectively. “n” is the factor related to morphology of the dielectric. As, explained earlier that water molecule is polar whose dipole moment changes under the influence of external electric field. The polarization of the water molecules can be expressed as eqn (3).

P = αE (3)

where, P represents polarization, α represents polarizability and E is the external electric field sourced by applied bias. The dielectric constant of the sensing element (ε_d) is dictated by the polarizability of the sensing material.^29,30 This polarizability may be ionic (α_i), dipolar (α_dip) or electronic (α_e) in nature.^31–33 The relation between the dielectric constant and polarizability (α_d) is expressed by the Clausius–Mosotti equation (eqn (4)).³⁴

From Fig. 7 (inset), it can be observed that, the change in capacitance with humidity is remarkable (114 fF/% RH) at 100 Hz operational frequency of the AC input bias (V_rms ∼ 1 V). In general for the humidity sensors, pronounced sensitivity is observed in low operational frequency range.^3,26 At higher order of operational frequencies such as 1 kHz and 10 kHz, the capacitance becomes small and changes to a relatively less extent with increasing RH levels. Admittedly, according to dielectric physical theory, the capacitance is independent of the operational frequency.³⁵ The operational frequency dependency of capacitance is however a common phenomenon in humidity sensors. At low RH levels only small amount of water is adsorbed by the sensing layer, and capacitance seems independent of operational frequency. However at high RH levels, when mobile multilayers of water contents are physisorbed on sensing layer, leak conductance appears.³⁶ At this stage, the capacitance–operational frequency interdependence of the sensor takes the complex form and is represented by using the following expression:^7,36

where, ε is complex dielectric parameter, ε′ is real part and ε′′ is the dielectric loss factor, respectively, σ is the leak conductivity and ω is the operational frequency. Since, C and ω are in inverse relationship, therefore at low operational frequencies, the magnitude of C is greatly affected. Hence, for the estimation of the sensing parameters of the humidity sensor we have opted for 1 kHz operational frequency.

The temperature versus capacitance relationship of the proposed humidity sensor has also been studied. The magnitude of capacitance shows significantly higher stability with temperature variation until 55 °C, showing negligible capacitance change (∼2%). However with further increase in temperature from (55–100 °C), the capacitance of the sensor seems to exhibit a direct relationship with enhancement in temperature. Capacitance variation was estimated to be ∼1.6 times the initial value when temperature was steadily increased from 55 to 100 °C. Albeit, the proposed sensor shows significant thermal drift at elevated temperatures, however it is worth noting that under normal operating conditions (below 55 °C) thermal drift is negligible.

Response and reset time estimation is quite crucial while optimizing the sensing performance of the humidity sensors. Response time is referred as, time duration to achieve 90% of the stable plateau value when sensor is suddenly exposed to humid environment.³⁷ While reset time is the duration required to a sensor to recover back to its initial value, after prompt removal of humidity. The dynamic characteristics of the proposed humidity sensor have been studied for three repeated cycles of humidification and dehumidification at 1 kHz oscillation frequency of input bias. The dynamic response–recovery curves of the humidity sensor have been portrayed in Fig. 8. In the present study, the response–recovery measurement has been carried out by alternately placing the proposed sensor into two different chambers with the two distinctive RH levels i.e., 20% and 95% RH. From Fig. 8, the consistency of the dynamic behavior of the sensor can be well observed for all 3 repeated cycle curves. The response and recovery times for the proposed sensor have been estimated to be 18 and 8 seconds, respectively.

Fig. 8 Dynamic response–recovery cycles (20–95% RH level) of the Al/MEH-PPV:PVP/Al humidity sensors.

The hysteresis curve of the proposed sensor has also been studied by evaluating the calibration curves during humidity absorption and desorption processes. The graph shown in Fig. 9, depicts the formation of small and consistent hysteresis gap in response to the increasing and deceasing of surrounding RH levels. The hysteresis at a certain humidity level is calculated as the percentage difference in measured capacitance during a de-humidification cycle in contrast to that measured during humidification cycle. Typically, the hysteresis of the humidity sensor is expressed as the average value of hysteresis on the total span of humidity scale. The average hysteresis value in the present study has been estimated to be ∼2% as the sensor is exposed to moisten ambient conditions within the range of 20 to 95% RH. The obtained hysteresis value suggests competency for this sensing device since it is below the threshold of 3%, a maximum value set for a practically competent sensor.³⁸ Furthermore, other research works have shown the correlation of small hysteresis value to heavy distribution of finely porous structures permeating surface of the sensing film.^39–41

Fig. 9 Hysteresis characteristic of Al/MEH-PPV:PVP/Al surface type humidity sensor.

Table 1, shows the comparison of the humidity sensing parameters of the proposed sensor with those previously reported in literature. It can be well observed that the humidity sensing performance of the present sensor is improved.

Table 1 Comparison of key humidity sensing parameters of MEH-PPV:PPV composite based sensor with previously reported sensors

Material	Sensitivity (@1 kHz)	Bandwidth	Response recovery time	Hysteresis
DMBHPET^13	7 fF/% RH	30–80% RH	10 and 15 s	—
Polyimide^42	5 fF/% RH	20–85% RH	∼15 and 10 s	∼3%
MEH-PPV:PVP	20 fF/% RH	20–90% RH	∼18 and 8 s	∼2%
	114 fF/% RH (100 Hz)

---

4. Conclusion

Capacitive-type humidity sensing device based on electrospun MEH-PPV:PVP microstructured composite has been fabricated and evaluated. Upon exposure to analyte (humidity), an increase in electrical capacitance has been predominantly observed. The capacitance variation has been explained to be associated with the difference in dielectric permittivity of humid and desiccated organic sensing film. The experimental results indicate that the sensor exhibits improved sensitivity as compared to previously reported spun-cast and drop-cast thin films. The reason of improved sensing parameters of the proposed humidity sensor is believed to be associated with the porous and coarse surface morphology of the organic microstructured composite. Present study, therefore, highlights that along with the choice of suitable sensing material, mode of sensing layer deposition is equally important. The fabricated sensor has shown an acceptable value of hysteresis (∼2%) in its response during the humidity adsorption and desorption cycles.

Acknowledgements

Authors are thankful to the Ministry of Education for the financial support under High Impact Research (HIR) grant UM.S/625/3/HIR/MOE/26 with account number UM.0000080/HIR.C3 and University Malaya Research Grant (UMRG) under grant number RP007A-13AFR. This project was also partially funded by the University Malaya postgraduate grant PG089-2012B.

References

1. Y. Zhang, K. Yu, D. Jiang, Z. Zhu, H. Geng and L. Luo, Appl. Surf. Sci., 2005, 242, 212–217.
2. N. M. Kiasari, S. Soltanian, B. Gholamkhass and P. Servati, Sens. Actuators, A, 2012, 182, 101–105.
3. K. S. Karimov, M. Saleem, Z. Karieva, A. Mateen, M. T. S. Chani and Q. Zafar, J. Semicond., 2012, 33, 073001.
4. I. Murtaza, K. S. Karimov, Z. Ahmad, I. Qazi, M. Mahroof-Tahir, T. Khan and T. Amin, J. Semicond., 2010, 31, 054001.
5. Z. Chen and C. Lu, Sens. Lett., 2005, 3, 274–295.
6. C. Y. Lee and G. B. Lee, Sens. Lett., 2005, 1, 1–15.
7. Z. Rittersma, Sens. Actuators, A, 2002, 96, 196–210.
8. Y. Yao, X. Chen, H. Guo and Z. Wu, Appl. Surf. Sci., 2011, 257, 7778–7782.
9. M. Saleem, M. Sayyad, K. S. Karimov, Z. Ahmad, M. Shah, M. Yaseen, I. Khokhar and M. Ali, J. Optoelectron. Adv. Mater., 2008, 10, 1468.
10. C.-W. Lee, H.-S. Park, J.-G. Kima, B.-K. Choi, S.-W. Joo and M.-S. Gong, Sens. Actuators, B, 2005, 109, 315–322.
11. M. Matsuguchi, S. Umeda, Y. Sadaoka and Y. Sakai, Sens. Actuators, B, 1998, 49, 179–185.
12. H. Farahani, R. Wagiran and M. N. Hamidon, Sensors, 2014, 14, 7881–7939.
13. M. I. Azmer, Z. Ahmad, K. Sulaiman and A. G. Al-Sehemi, Measurement, 2015, 61, 180–184.
14. Z. Ahmad, Q. Zafar, K. Sulaiman, R. Akram and K. S. Karimov, Sensors, 2013, 13, 3615–3624.
15. P.-G. Su and L.-N. Huang, Sens. Actuators, B, 2007, 123, 501–507.
16. M. I. Azmer, Z. Ahmad and K. Sulaiman, Mater. Lett., 2015, 150, 73–75.
17. K. Mittal, Act. Passive Electron. Compon., 1976, 3, 21–42.
18. C. Ran, M. Wang, W. Gao, Z. Yang, J. Deng, J. Ding and X. Song, Phys. Chem. Chem. Phys., 2014, 16, 4561–4568.
19. M. Tahir, M. Hassan Sayyad, F. Wahab, F. Ahmad Khalid, F. Aziz, S. Naeem and M. Naeem Khalid, Int. J. Mod. Phys. B, 2014, 28, 1450032.
20. F. ur Rehman, M. Tahir, S. Hameed, F. Wahab, F. Aziz, F. Khalid, M. N. Khalid and W. Ali, Mater. Sci. Semicond. Process., 2015, 39, 355–361.
21. M. Joulazadeh, A. H. Navarchian and M. Niroomand, Adv. Polym. Technol., 2014, 33, 21461.
22. J. G. Webster and H. Eren, Measurement, Instrumentation, and Sensors Handbook: Spatial, Mechanical, Thermal, and Radiation Measurement, CRC press, 2014.
23. Y. L. Yang, L. H. Lo, I. Y. Huang, H. J. Chen and W. S. Huang, IEEE Sens. J., 2002, 1, 511–514.
24. M. Matsuguchi, M. Shinmoto, Y. Sadaoka, T. Kuroiwa and Y. Sakai, Sens. Actuators, B, 1996, 34, 349–355.
25. F. Aziz, M. H. Sayyad, K. Sulaiman, B. Majlis, K. S. Karimov, Z. Ahmad and G. Sugandi, Meas. Sci. Technol., 2012, 23, 014001.
26. M. Björkqvist, J. Salonen, J. Paski and E. Laine, Sens. Actuators, A, 2004, 112, 244–247.
27. H. Bässler and A. Köhler, in Unimolecular and Supramolecular Electronics I, Springer, 2012, pp. 1–65.
28. J. G. Korvink, L. Chandran, T. Boltshauser and H. Baltes, Sens. Mater., 1993, 4, 323–335.
29. M. A. Omar, Elementary solid state physics: principles and applications, Addison-Wesley, 1993.
30. F. Gutman and L. Lyons, Organic Semiconductors – Part A, Robert E. Publishing company Malabar, FL, USA, 1981.
31. M. Saleem, M. H. Sayyad, K. S. Karimov, M. Yaseen and M. Ali, Sens. Actuators, B, 2009, 137, 442–446.
32. M. Tahir, M. H. Sayyad, J. Clark, F. Wahab, F. Aziz, M. Shahid, M. A. Munawar and J. A. Chaudry, Sens. Actuators, B, 2014, 192, 565–571.
33. R. Akram, M. Saleem, Q. Zafar and K. S. Karimov, Sens. Lett., 2013, 11, 494–499.
34. M. Shah, Z. Ahmad, K. Sulaiman, K. S. Karimov and M. Sayyad, Solid-State Electron., 2012, 69, 18–21.
35. T. Zhang, Y. He, R. Wang, W. Geng, L. Wang, L. Niu and X. Li, Sens. Actuators, B, 2008, 131, 687–691.
36. P. Li, X. Zheng, Y. Zhang, M. Yuan, B. Jiang and S. Deng, Ceram. Int., 2015, 41, 14251–14257.
37. A. Tripathy, S. Pramanik, J. Cho, J. Santhosh and N. A. Abu Osman, Sensors, 2014, 14, 16343–16422.
38. M. Ueda, K. Nakamura, K. Tanaka, H. Kita and K.-I. Okamoto, Sens. Actuators, B, 2007, 127, 463–470.
39. Z. Imran, S. Batool, H. Jamil, M. Usman, M. Israr-Qadir, S. Shah, S. Jamil-Rana, M. Rafiq, M. Hasan and M. Willander, Ceram. Int., 2013, 39, 457–462.
40. A. Sun, L. Huang and Y. Li, Sens. Actuators, B, 2009, 139, 543–547.
41. M. Matsuguch, T. Kuroiwa, T. Miyagishi, S. Suzuki, T. Ogura and Y. Sakai, Sens. Actuators, B, 1998, 52, 53–57.
42. L. Gu, Q.-A. Huang and M. Qin, Sens. Actuators, B, 2004, 99, 491–498.

---