Open Access
Add to BibSonomyAbstract
A high sensitivity and simple ethanol sensor based on an un-cladded multimode plastic optical fiber (UCPOF) coated with carbon nanotubes (CNTs) for the detection of different concentrations of ethanol in de-ionized water is developed and demonstrated. The UCPOF probe is fabricated by chemically removing the fiber cladding and integrated with CNT as a sensing layer. The effect of surface morphology on the sensor performance is investigated by characterizing another UCPOF coated with GO nanomaterial. The developed fibers are coated with CNTs and GO using drop casting technique. Energy dispersive X-ray spectroscopy (EDX), atomic-force microscopy (AFM) and scanning electron microscope (SEM) are used to investigate the element and morphology of the synthesized nanomaterials. The experimental results indicated that the absorbance spectrum of the CNT-based UCPOF sensor increases linearly with a higher sensitivity of 0.68/vol% and magnitude change of 95.4% as compared to 0.19/vol% and 56.3%, respectively, for the GO-based sensor. The UCPOF coated with CNT exhibits faster response and recovery than that of GO. The sensor shows high selectivity to ethanol amongst a range of diluted organic VOCs. The superior sensing performance of the developed fiber sensor indicates its high efficiency for ethanol detection in various industrial applications.
© 2017 Optical Society of America
1. Introduction
Ethanol is a volatile organic compound (VOC) that is commonly used in the food, beverage, fuel and pharmaceutical industries [1,2]. Exposure to high concentrations of ethanol may cause irritation to the skin and inflammation of the nasal mucous membrane. Therefore, ethanol detection in a liquid medium is crucial for the quality control of food, beverages and medicines [3,4]. Several approaches have been considered for ethanol detection, such as liquid chromatography, mass spectrometry [5], conductometric sensing [6], and electrochemical sensing [7]. Although these sensors attain relatively good sensitivity and selectivity, they suffer from drawbacks that include high loss and high working temperature. Optical fiber sensors have been gaining much attention in the last few decades because they offer enormous advantages over traditional electronic devices. They are immune to electromagnetic interference, micro-sized, and nondestructive [8]. Most research efforts focus on typical silica fiber based ethanol sensors with nanostructured material coatings [9,10]. They offer good performance in chemical sensing applications. However, fiber optical sensors with small waist diameter (< 100μm) are more fragile and cannot be used as practical sensors in harsh environment. Moreover, research on reducing the cost and intricacy of these sensors is required.
Plastic optical fibers (POFs) offers numerous advantages, including low cost, ease of manipulation, high degree of flexibility and large-diameter cores. POF-based sensors have been widely applied in low-cost applications such as chemical [11], environmental [12], and medicinal research [13]. The large cores and numerical apertures and ease of handling of the POFs contribute improve the sensing performance in terms of enhanced sensitivity compared with silica optical fibers [14]. In order to enhance the optical fiber sensor performance, it is important to have strong interaction between the sensing layer and light propagates in the fiber core. For this purpose, several configurations have been developed such as tapered optical fiber [15], and side-polished optical fiber [16]. These modifications tend make the optical fibers fragile and cause handling difficulties. Uncladded-POF (UCPOF) has attracted considerable attention in chemical sensing applications due to its simplicity, low cost, and flexibility even after the cladding removal as well as it preserves the geometrical circular structure of the optical fiber.
Recent works have identified that the use of nanostructured materials can improve the chemical sensing performance with regards to sensitivity, selectivity and response time [17–19]. Graphene and graphene oxide (GO) have been in the limelight recently as nanomaterials that can enhance the performance of sensors due to its distinctive optical, chemical properties and strong hydrophilic nature, as well as large surface area [20, 21]. Carbon nanotubes (CNTs) have attracted tremendous interest in chemical sensing applications because of their prominent optical and physical characteristics in addition to their capability of detecting various types of chemicals [22,23].
Graphene-coated POF for glucose detection demonstrated a stable output towards different glucose concentrations due to the good adsorption and atomically thick nature of graphene [24]. Furthermore, an ethanol sensor based on tapered POF coated with monolayer graphene was reported and demonstrated [25]. The results revealed that the tapered POF coated with graphene shows better sensing performance compared to the uncoated fiber. This can be ascribed to the large change of refractive index (RI) in evanescent waves resulted from the high interaction between the graphene surface and ethanol molecules. CNT-coated clad modified sensor was also proposed and demonstrated to monitor ethanol concentration of the range 0–500 ppm; it showed that thick CNT layer indicates better sensing performance toward ethanol and methanol due to the high interaction between the nanotubes and target molecules [26]. Although much research has been conducted on POF in the area of chemical sensing [28–31], there are great opportunities in examining the sensing characteristics of CNT and GO coatings on UCPOF for ethanol sensing applications.
Here, we report the excellent response of ethanol sensors based on UCPOFs coated with CNT and GO. The achieved results are carried out at room temperature, which indicates the high efficiency of the POF sensors as compared to the conventional ethanol sensors.
2. Experiment
2.1. Fabrication of the UCPOF
A multimode POF with a polymethyl-methacrylate (PMMA) core diameter of 980μm and a fluorinated polymer cladding thickness of 20μm was used to fabricate the proposed UCPOF ethanol sensor. The cladding of POF was removed according to the chemical etching procedure reported by Merchant et al. [32] using acetone, sand paper (1000 grit size) and de-ionized water. Firstly, 2 cm of the fiber protective jacket was removed using a POF stripper. To soften the POF cladding, the stripped region was immersed in diluted acetone until a whitish layer appears on the POF surface. Later, the fiber was polished using a fine abrasive to remove the POF top layer and then cleaned using de-ionized water to eliminate any residuals. Both ends of the fiber were stripped and polished using sand paper to improve the coupling of light in the fiber. The achieved sensing region was approximately 2 cm. Figure 1 exhibits the microscopic images of the POF (taken with Motic Microscope-BA310E, South Korea) before and after removal of the cladding. The figure indicates that fiber with a 1-mm diameter is etched to approximately 975μm without affecting the cylindrical geometry of the optical fiber.
2.2. Synthesis of the CNT and GO nanomaterials
To produce functionalized CNTs that can dispersed in ethanol, CNTs were synthesized by adding raw CNTs (from Hangzhou Company, China) to a mixture of sulfuric acid as a chemical treatment method to obtain CNTs containing different COOH contents [33]. Chemically treated CNTs (2 mg) were added to 8 ml of ethanol in a covered vial, followed by ultrasonication in a water bath for one hour at room temperature. The CNT layer was deposited on the UCPOF via drop-casting technique. Before the CNT deposition process, the fiber was heated at 70 °C in an oven for 15 min to improve adhesion of the CNT material on the fiber. After the solution was drop-casted onto the UCPOF, the coated POF sample was heated to 70 °C in the oven for 2 hour and left to dry at room temperature for 24 hour.
To attain a complete assessment of the CNT based sensor performance, GO nanomaterial was synthesized, using the simplified Hummer’s method [34], and utilized as a sensing layer. GO was dispersed in deionized water at a concentration of 0.25 mg/mL and then sonicated for one hour. Prior to the coating process, the UCPOF was heated to 70 °C for 15 min. The GO was drop-casted onto the UCPOF for micro-characterization and ethanol sensing. Subsequently, the sample was heated at 70 °C for 2 hour, followed by an additional 24 hour period of drying at room temperature. Following the same deposition procedure, silica wafer (SiO2) substrates were used to deposit both nanomaterials (CNT and GO) for EDX characterization purposes.
2.3. Ethanol sensing experimental setup
The UCPOF ethanol sensor was fixed in a customized liquid chamber and then connected on one end to a tungsten-halogen light source (Ocean OpticsTM HL 2000, wavelength range of 360 nm to 2400 nm). The other POF end was attached to a spectrophotometer (Ocean OpticsTM USB-4000) with a spectral wavelength range between 200 nm to 1100 nm which is used for the absorbance measurement. A computer was attached to the spectrophotometer via USB port, as shown in Fig. 2.
The data collected from the spectrophotometer was processed and analyzed using the Spectra-Suite software (version 6.2) installed on the computer. The absorbance, Aλ, was calculated by the software, as represented by the following mathematical equation:
where Sλ is the light intensity detected at wavelength λ upon exposure to ethanol. Dλ represents the intensity of the dark reference signal and Rλ is the reference intensity for the blank fiber at wavelength λ. For sensing purposes, different concentrations of diluted ethanol (20% – 100%) were carried out using different ratios of de-ionized water mixed with pure ethanol. All the experiments were performed at room temperature (26 °C).3. Results and discussion
3.1. Characterization of the nanomaterials
The surface morphology of the CNT and GO coatings are examined using a scanning electron microscope (SEM-Hitachi S-3400N), as exhibited in Figs. 3(a)–3(d). Figures 3(a) and 3(b) show typical SEM images of the prepared CNT and GO materials that are coated onto the UCPOF; the images enable observation of their homogeneity on the fiber. This figure further confirms the successful deposition of the nanomaterials onto the surface of the fabricated fiber. Figure 3(c) shows that the multi-walled CNT (MWCNT) coating is composed of randomly assembled, tortuous, long as well as tangled tubular structures formed into bundles interspersed by a large number of pores. Figure 3(d) shows a thin and smooth wrinkled surface topography of the GO nanomaterial, with no obvious defects or cracks. Energy Dispersive X-Ray (EDX) analysis is used to examine the chemical compositions (shown in Figs. 3(e) and 3(f)), which assured the presence of Carbon (C) and Oxygen (O) in the CNT and GO sensing layers. The existence of the high Si peak in both samples is due to the silica (SiO2) substrate used to deposit the nanomaterials.
Fig. 3 SEM images of UCPOF coated with ((a) and (c)) CNT and ((b) and (d)) GO. EDX spectra of (e) CNT and (f) GO coated glass (SiO2) substrate.
To measure the thicknesses of the CNT and GO coatings, a portion of the fiber is covered with aluminum tape during the coating process to obtain the difference between the coated and uncoated fiber. Figures 4(a) and 4(b) depict the atomic force microscope (NT-MDT Solver NEXT AFM) 3D image of the CNT and GO layers coated on the uncladded probe, respectively. From Fig. 4(a), the average surface roughness is found to be 34.272 nm, and the average thickness of the CNT coating is estimated to be 2 ±0.2μm. Hence, this level of roughness demonstrates that the effect of light scattering on the sensing performance is insignificant. From the 3D image (shown in Fig. 4(b)), the average surface roughness and thickness for the GO nanomaterial are estimated to be approximately 7.330 nm and 2 ± 0.17μm, respectively.
Fig. 4 3D topography AFM image of the boundary area between uncoated and coated UCPOF for (a) CNT and (b) GO sensing layers.
3.2. Sensor performance
The optical responses of the UCPOF sensor based on CNT and GO materials toward different concentrations of ethanol are investigated to assess the sensing performance of the proposed sensor. In optical fiber sensor, an interaction occurs between the active sensing layer and the light traveling through the fiber, causing obvious changes in the optical sensing performance of the developed fiber sensor.
The normalized absorbance spectra of CNT and GO coated UCPOF sensors toward different ethanol concentrations are illustrated in Figs. 5(a) and 5(b). Note that the absorbance responses of CNT and GO sensors are inversely dependent on the ethanol concentrations in the wavelength range of 500 nm to 750 nm and 660 nm to 820 nm, respectively. In other words, the absorbance levels are observed to decrease when ethanol concentration increases. Both the transition of the overtone and combination tone of O-H stretched absorption are suggested to be the reason for the decrease in the absorbance levels. An obvious dip in the ranges of 700 nm to 720 nm and 700 nm to 730 nm appeared in the absorbance spectra of the CNT and GO coated UCPOF, respectively, due to the changes in the optical properties of the nanomaterials; more specifically, the RI [24]. Figure 5(a) shows that the depth broadens as the ethanol concentration increases. This behavior is due to the changes in the real and imaginary part of the RI. A change in the real part of the RI causes the wavelength of the dip to change. When the imaginary part of the RI changes, the dip will broaden. Furthermore, a wavelength shift of approximately 5 nm is observed in Fig. 5(b) that is attributed to the increase in RI when the ethanol concentration increased.
Fig. 5 Normalized absorbance spectra obtained for (a) CNT and (b) GO nanomaterial-coated UCPOF towards different ethanol concentrations in de-ionized water.
For sensing performance comparison, Figs. 6(a) and 6(b) demonstrate the dynamic response of CNT and GO thin films toward various ethanol concentrations over wavelength ranging from 600 nm to 800 nm at room temperature. Figure 6(a) shows that the UCPOF-based CNT sensor exhibits relative absorbance, response, and recovery times of 55%, 45 s, and 48 s, respectively, for 20% ethanol and 95.4%, 6.8 s, and 13 s, respectively, for 100% ethanol.
Fig. 6 Normalized absorbance based dynamic response obtained for (a) CNT based POF sensor towards ethanol and (b) GO nanomaterial based POF sensor towards ethanol. (c) Normalized absorbance change as a function of ethanol concentration for CNT and GO sensing layer.
The response of the CNT based sensor is better than the response of the tapered silica optical fiber sensor reported by Shabaneh et al. [27]. It is found that the response and recovery times decrease with increasing ethanol concentration due to the interaction between the ethanol molecules and the adsorbed oxygen ions [14]. The response magnitude for the CNT thin film-based sensor increased proportionally to the ethanol concentrations, in agreement with the results in [10]. The dynamic performance of the GO thin film (depicted in Fig. 6(b)) shows an acceptable response by having different absorbance magnitudes for different ethanol concentrations below 60%. The GO thin film-based sensor demonstrates poor response, reversibility and recovery times compared with the UCPOF-based CNT sensor. This observation indicates saturation of the sensor at ethanol concentrations above 60%. To investigate the reliability of the developed sensors, a repeatability investigation has been performed by exposing the sensors to another cycle (C2) of a 20% ethanol concentration, as shown in Figs. 6(a) and 6(b). Both sensors show almost the same response as in the case of the first cycle (C1) for 20% ethanol by producing absorbance magnitude changes of 53% and 40% for the CNT and GO sensors, respectively. This exhibits the excellent repeatability of the developed sensors which is very important for accurate detection of ethanol.
The normalized absorbance change as a function of ethanol concentration for the proposed thin films over the wavelength range of 600 nm to 800 nm is depicted in Fig. 6(c). This figure further confirms the excellent dynamic performance of the CNT coated sensor in Fig. 6(a). The absorbance change increased proportionally with the increase of the ethanol concentrations. Furthermore, the CNT-based sensor shows higher sensitivity towards ethanol concentrations compared to the slight change observed in the GO-based sensor. The sensitivity of the CNT- and GO-based UCPOF sensors is 0.68/vol% and 0.19/vol%, respectively, as shown in Fig. 6(c). The CNT-based sensor shows higher sensitivity, that is, approximately 4 times than that of the GO-based sensor. Meanwhile, the CNT- and GO-based UCPOF sensors exhibit slope linearity of 99% and 86%, respectively. It is believed that the morphology and roughness of the material surface play major roles in the sensor response to the measurand, i.e., ethanol in this case. As indicated in Fig. 4(a), the CNT thin film has a high surface area with many sites in its mesh structure available to adsorb ethanol molecules; such sites may help in the sensitivity enhancement. Thus, the porous surface of the CNT thin film allows the ethanol molecules to deeply penetrate the sensing layer and interact with the evanescent wave, as shown in Fig. 7(a), unlike the sheet structure of the GO shown in Fig. 7(b). The difference is because chemical molecules easily diffuse within materials with porous structures [35]. Meanwhile, the average surface roughness of the CNT coating is approximately five times higher than that of the GO coating, which may interestingly contribute to the high sensitivity of the CNT-based un-cladded sensor compared with that of the GO-based sensor. Resolution is the smallest incremental change in the tested analyte that will cause an obvious alteration in the output signal [36]. Figures 8(a) and 8(b) show the resolutions of the developed UCPOF sensors coated with CNT and GO at low ethanol concentrations ranging from 0.2–0.6%. As shown in Fig. 8(a), the resolution limit of the ethanol sensor based on CNT sensing layer is found to be as low as 0.2% which is lower than the detection limit reported in [9,17]. In addition, ethanol concentration of 0.2% lead to an absorbance reduction of 34.4% with respect to the baseline in the CNT-based sensor with high response and recovery times of 25 s and 38 s, respectively. From Fig. 8(b), it can be noticed that the UCPOF sensor coated with GO shows detection limit of 0.6% with small absorbance difference of 1.77% and 1.785% from 0.2% and 0.4% ethanol concentrations, respectively. Thus, the CNT based sensor attains lower resolution limit than the GO based sensor. Furthermore, the sensitivity of the developed UCPOF coated with CNT towards low ethanol concentrations (0.2% – 0.6%) is found to be approximately 3 times higher than the one coated with GO.
Fig. 7 Schematic representation of the proposed interaction mechanism between ethanol molecules and (a) CNT and (b) GO thin films.
From the results, in addition to its high repeatability, response and recovery times, the sensitivity of the CNT-based sensor exhibits no obvious saturation at high ethanol concentrations. In contrast, the GO-based sensor shows a very poor response compared to the CNT-based sensor. Therefore, it is possible to regard the CNT-based sensor as reliable and ideal for practical use. The ethanol sensing mechanism of the CNT thin film can be explained on the basis of the model reported by Sin et al. [37]. Pristine CNT treated with different oxidizing agents (acids) might introduce the covalent attachment of -COOH on the CNT surface, this group consists of a carbon atom double bonded to an oxygen as well as an -OH group [38]. The OH groups of ethanol molecules come into contact with the carboxylic group attached to the CNT nanostructured surface through hydrogen bonds in the customized liquid chamber, leading to improved sensitivity toward ethanol molecules due to the hydrogen bonding of the dipole-dipole interactions between the polar groups on the CNT surface as well as that of the ethanol molecule [37].
Table 1 summarizes the sensitivity and resolution of CNT coated on UCPOF as compared to existing works on CNT coated on tapered MMF tip and coiled optical fiber sensor towards ethanol. The UCPOF sensor shows the highest sensitivity and resolution than our previous sensor based on tapered MMF tip as well as coiled optical fiber sensor. The superior performance of the UCPOF sensor than the tapered MMF tip is because of the absorbance sensing mechanism deployed in the former instead of reflectance based as in the latter. Absorbance mechanism involves measurement of the light directly propagates in the UCPOF. In contrast, the reflectance mechanism uses weak reflected light from the sensing tip interaction with ethanol and thus, less power measured at the receiver as compared to the absorbance measurement. The low sensitivity with unknown resolution towards ethanol using the coiled optical fiber indicates the significant enhancement on the ethanol sensing response with the presence of CNT sensing layer coated on the optical fiber sensors.
Table 1. Sensitivity and resolution comparison with existing ethanol optical sensor
One of the essential characteristics for a chemical sensor is its selectivity to other types of chemicals compounds. Thus, the CNT sensor’s selectivity with respect to acetone, methanol, xylene, and chloroform is examined. Figure 9 depicts the experimental results of the UCPOF coated with CNT, and exposed, at room temperature, to five different organic VOCs. As can be clearly noticed, the CNT based UCPOF sensor indicate high selectivity to ethanol than those of the other tested analytes.
Fig. 9 Normalized absorbance change of CNT based UCPOF as a function of different concentrations of ethanol, acetone, methanol, xylene, and chloroform in water.
According to the illustrated results, ethanol similar alcoholic compounds like methanol and acetone give high response on the developed sensor, compared with the non-similar ones such as chloroform and xylene. From Fig. 9, the sensitivity is found to be about 0.678/vol%, 0.254/vol%, 0.0197/vol%, 0.2627/vol%, and 0.098/vol% for ethanol, acetone, xylene, methanol, and chloroform, respectively, which indicates that our sensor reveals significant selectivity to ethanol and it is much recommended for ethanol detection. This can assigned to the polar nature of both the CNT coating and the tested analyte molecule. Alcohol molecules can form hydrogen bonds together with the polar groups on the CNT surface. Thus, methanol and acetone showed better response than other examined VOCs (xylene and chloroform) [40].
In addition, the selectivity coefficient (S) is calculated which can be presented as S =At/Ai [41], where At and Ai are the absorbance response of the sensor to 100% concentration of ethanol and other tested analytes, respectively. The bigger value of S indicates that the sensor has a better ability to distinguish ethanol from other tested organic compounds, as summarized in Table 2. These specific values can be used to discriminate between ethanol and other organic analytes under test.
Table 2. S values of the UCPOF sensor coated with CNT
4. Conclusion
In conclusion, we fabricated a simple ethanol sensor based on UCPOF coated with CNT thin film at room temperature. A simple and low cost chemical process was employed to prepare the UCPOF sensing area. In order to investigate the alteration of the coating’s surface morphology on the sensor performance, another UCPOF coated with GO nanomaterial has been examined. CNT and GO were prepared and drop-casted on the UCPOF surface as sensing layers for detecting different ethanol concentrations in the range of 20% to 100% in water. Experiments demonstrate that the CNT-based UCPOF sensor shows significant sensitivity to aqueous ethanol of approximately 4 times higher than the sensor coated with GO. This difference is mainly attributed to the porous structure and rough surface of the CNT coating compared to the GO coating. The resolution of the ethanol sensor based on CNT sensing layer is found to be as low as 0.2% with a high sensitivity of 15.1/vol%. Furthermore, the sensor shows superior selectivity to ethanol amongst a range of diluted organic VOCs. The significant sensitivity, high repeatability, excellent selectivity and remarkable stability of the UCPOF sensor coated with CNT towards ethanol indicates its high efficiency to be used in practical applications for ensuring safety necessities in the industrial fields.
Funding
Universiti Putra Malaysia (UPM) (501100004530).
Acknowledgments
The authors would like to acknowledge Universiti Putra Malaysia for the project funds No. FRGS/1/2014/TK03-/UPM/03/1 and FRGS/2/2014/TK03/UPM/01/1.
References and links
1. P. Hu, G. Du, W. Zhou, J. Cui, J. Lin, H. Liu, D. Liu, J. Wang, and S. Chen, “Enhancement of ethanol vapor sensing of TiO2 nanobelts by surface engineering,” ACS Appl. Mater. Interfaces 2 (11), 3263–3269 (2010). [CrossRef] [PubMed]
2. C. Elosua, I. R. Matias, C. Bariain, and F. J. Arregui, “Volatile organic compound optical fiber sensors: a review,” Sensors 6 (11), 440–1465 (2006). [CrossRef]
3. S. B. Khan, M. M. Rahman, K. Akhtar, A. M. Asiri, J. Seo, H. Han, and K. Alamry, “Novel and sensitive ethanol chemi-sensor based on nanohybrid materials, ” Int. J. Electrochem. Sci. 7, 4030–4038 (2012).
4. Ş. Alpat and A. Telefoncu, “Development of an alcohol dehydrogenase biosensor for ethanol determination with toluidine blue O covalently attached to a cellulose acetate modified electrode,” Sensors 10 (1), 748–764 (2010). [CrossRef] [PubMed]
5. T. Buttler, L. Gorton, H. Jarskog, G. Marko-Varga, B. Hahn-HäGerdal, N. Meinander, and L. Olsson, “Monitoring of ethanol during fermentation of a lignocellulose hydrolysate by on-line microdialysis sampling, column liquid chromatography, and an alcohol biosensor,” Biotechnol. Bioeng. 44 (3), 322–328 (1994). [CrossRef] [PubMed]
6. G. Neri, A. Bonavita, G. Micali, N. Donato, F. Deorsola, P. Mossino, I. Amato, and B. De Benedetti, “Ethanol sensors based on Pt-doped tin oxide nanopowders synthesised by gel-combustion,” Sens. Actuat. B-Chem. 117 (1), 196–204 (2006). [CrossRef]
7. B. Tao, J. Zhang, S. Hui, and L. Wan, “An amperometric ethanol sensor based on a Pd–Ni/SiNWs electrode,” Sens. Actuat. B-Chem. 142 (1), 298–303 (2009). [CrossRef]
8. C. R. Zamarreno, I. R. Matias, and F. J. Arregui, “Nanofabrication techniques applied to the development of novel optical fiber sensors based on nanostructured coatings,” IEEE Sens. J. 12 (8), 2699–2710 (2012). [CrossRef]
9. S. Girei, A. Shabaneh, H. Lim, N. Huang, M. Mahdi, and M. Yaacob, “Absorbance response of graphene oxide coated on tapered multimode optical fiber towards liquid ethanol,” J. Eur. Opt. Soc.-Rapid Publ. 10, 15019 (2015). [CrossRef]
10. A. Shabaneh, S. Girei, P. Arasu, W. Rahman, A. A. A. Bakar, A. Sadek, H. Lim, N. Huang, and M. Yaacob, “Reflectance response of tapered optical fiber coated with graphene oxide nanostructured thin film for aqueous ethanol sensing,” Opt. Commun. 331, 320–324 (2014). [CrossRef]
11. D. R. Raj, S. Prasanth, T. Vineeshkumar, and C. Sudarsanakumar, “Ammonia sensing properties of tapered plastic optical fiber coated with silver nanoparticles/PVP/PVA hybrid,” Opt. Commun. 340, 86–92 (2015). [CrossRef]
12. Z. Harith, N. Irawati, H. A. Rafaie, M. Batumalay, S. W. Harun, R. M. Nor, and H. Ahmad, “Tapered plastic optical fiber coated with Al-doped ZnO nanostructures for detecting relative humidity,” IEEE Sens. J. 15 (2), 845–849 (2015). [CrossRef]
13. L. Bilro, N. Alberto, J. L. Pinto, and R. Nogueira, “Optical sensors based on plastic fibers, ” Sensors 12 (9),12184–12207 (2012). [CrossRef] [PubMed]
14. N. Zhong, Q. Liao, X. Zhu, M. Zhao, Y. Huang, and R. Chen, “Temperature-independent polymer optical fiber evanescent wave sensor, ” Sci. Rep. 5, 11508 (2015). [CrossRef]
15. M. Batumalay, Z. Harith, H. A. Rafaie, F. Ahmed, M. Khasanah, S. W. Harun, R. M. Nor, and H. Ahmad, “Tapered plastic optical fiber coated with ZnO nanostructures for the measurement of uric acid concentrations and changes in relative humidity,” Sens. Actuat. A-Phys 210, 190–196 (2014). [CrossRef]
16. A. L. Khalaf, F. S. Mohamad, P. T. Arasu, A. A. Shabaneh, N. A. Rahman, H. N. Lim, S. Paiman, N. A. Yusof, M. A. Mahdi, and M. H. Yaacob, “Modified plastic optical fiber coated graphene/polyaniline nanocomposite for ammonia sensing,” in Proceedings of IEEE 6th International Conference on Photonics (IEEE, 2016), pp. 1–3.
17. A. Aziz, H. Lim, S. Girei, M. Yaacob, M. Mahdi, N. Huang, and A. Pandikumar, “Silver/graphene nanocomposite-modified optical fiber sensor platform for ethanol detection in water medium,” Sens. Actuat. B-Chem. 206, 119–125 (2015). [CrossRef]
18. Y. Zhao, X. Li, X. Zhou, and Y. Zhang, “Review on the graphene based optical fiber chemical and biological sensors,” Sens. Actuator B-Chem. 231, 324–340 (2016). [CrossRef]
19. F. R. Baptista, S. A. Belhout, S. Giordani, and S. J. Quinn, “Recent developments in carbon nanomaterial sensors,” Chem. Soc. Rev. 44, 4433–4453 (2015). [CrossRef]
20. K. Toda, R. Furue, and S. Hayami, “Recent progress in applications of graphene oxide for gas sensing: a review,” Anal. Chim. Acta 878, 43–53 (2015). [CrossRef] [PubMed]
21. S. Basu and P. Bhattacharyya, “Recent developments on graphene and graphene oxide based solid state gas sensors,” Sens. Actuat. B-Chem. 173, 1–21 (2012). [CrossRef]
22. R. H. Baughman, A. A. Zakhidov, and W. A. de Heer, “Carbon nanotubes–the route toward applications,” Science 297 (5582), 787–792 (2002). [CrossRef] [PubMed]
23. C. Piloto, F. Mirri, E. A. Bengio, M. Notarianni, B. Gupta, M. Shafiei, M. Pasquali, and N. Motta, “Room temperature gas sensing properties of ultrathin carbon nanotube films by surfactant-free dip coating,” Sens. Actuat. B-Chem. 227, 128–134 (2016). [CrossRef]
24. H. Qiu, S. Xu, S. Jiang, Z. Li, P. Chen, S. Gao, C. Zhang, and D. Feng, “A novel graphene-based tapered optical fiber sensor for glucose detection,” Appl. Surf. Sci. 329, 390–395 (2015). [CrossRef]
25. S. Jiang, H. Qiu, S. Gao, P. Chen, Z. Li, K. Yu, W. Yue, C. Yang, Y. Huo, and S. Wang, “Evanescent wave absorption sensor based tapered plastic optical fiber coated with monolayer graphene for ethanol molecules detection,” Chin. J. Chem. 34, 1039–1047 (2016). [CrossRef]
26. S. Manivannan, A. M. Saranya, B. Renganathan, D. Sastikumar, G. Gobi, and K. C. Park, “Single-walled carbon nanotubes wrapped poly-methyl methacrylate fiber optic sensor for ammonia, ethanol and methanol vapors at room temperature,” Sens. Actuat. B-Chem. 171, 634–638 (2012). [CrossRef]
27. A. Shabaneh, S. Girei, P. Arasu, M. Mahdi, S. Rashid, S. Paiman, and M. Yaacob, “Dynamic response of tapered optical multimode fiber coated with carbon nanotubes for ethanol sensing application,” Sensors 15 (5), 10452–10464 (2015). [CrossRef] [PubMed]
28. N. Cennamo, A. Donà, P. Pallavicini, G. D’Agostino, G. Dacarro, L. Zeni, and M. Pesavento, “Sensitive detection of 2, 4, 6-trinitrotoluene by tridimensional monitoring of molecularly imprinted polymer with optical fiber and five-branched gold nanostars,” Sens. Actuat. B-Chem. 208, 291–298 (2015). [CrossRef]
29. N. Cennamo, G. D’Agostino, M. Pesavento, and L. Zeni, “High selectivity and sensitivity sensor based on MIP and SPR in tapered plastic optical fibers for the detection of L-nicotine,” Sens. Actuat. B-Chem. 191, 529–536 (2014). [CrossRef]
30. V. Semwal, A. M. Shrivastav, R. Verma, and B. D. Gupta, “Surface plasmon resonance based fiber optic ethanol sensor using layers of silver/silicon/hydrogel entrapped with ADH/NAD,” Sens. Actuat. B-Chem. 230, 485–492 (2016). [CrossRef]
31. Y. Jin and A. M. Granville, “Polymer Fiber Optic Sensors–A Mini Review of their Synthesis and Applications,” Biosens. Bioelectron. 7, 2 (2016).
32. D. Merchant, P. Scully, and N. Schmitt, “Chemical tapering of polymer optical fibre,” Sens. Actuat. A-Phys 76 (1), 365–371 (1999). [CrossRef]
33. Y. T. Shieh, G. L. Liu, H. H. Wu, and C. C. Lee, “Effects of polarity and pH on the solubility of acid-treated carbon nanotubes in different media,” Carbon 45 (9), 1880–1890 (2007). [CrossRef]
34. C. Ching, S. Lee, P. Ooi, S. Ng, Z. Hassan, H. A. Hassan, and M. Abdullah, “Optical and structural properties of porous zinc oxide fabricated via electrochemical etching method,” Mater. Sci. Eng. B-Adv. Funct. Solid-State Mater. 178 (15), 956–959 (2013). [CrossRef]
35. H. Tai, Y. Jiang, G. Xie, J. Yu, and X. Chen, “Fabrication and gas sensitivity of polyaniline–titanium dioxide nanocomposite thin film,” Sens. Actuat. B-Chem. 125 (2), 644–650 (2007). [CrossRef]
36. K. Kalantar-zadeh and B. Fry, Nanotechnology-enabled sensors (Academic, 2007).
37. M. L. Y. Sin, G. C. T. Chow, G. M. K. Wong, W. J. Li, P. H. W. Leong, and K. W. Wong, “Ultralow-power alcohol vapor sensors using chemically functionalized multiwalled carbon nanotubes,” IEEE Trans. Nanotechnol. 6 (5), 571–577 (2007). [CrossRef]
38. W. Xue and P. Li, “Dielectrophoretic deposition and alignment of carbon nanotubes,” in Handbook of Carbon Nanotubes - Synthesis, Characterization, Applications, S. Yellampalli, ed. (Academic, 2011), pp. 172–190.
39. F. B. Xiong and D. Sisler, “Determination of low-level water content in ethanol by fiber-optic evanescent absorption sensor,” Opt. Commun. 283 (7), 1326–1330 (2010). [CrossRef]
40. B. Philip, J. K. Abraham, A. Chandrasekhar, and V. K. Varadan, “Carbon nanotube/PMMA composite thin films for gas-sensing applications,” Smart Mater. Struct. 12, 935 (2003). [CrossRef]
41. M. Zhao, X. Wang, J. Cheng, L. Zhang, J. Jia, and X. Li, “Synthesis and ethanol sensing properties of Al-doped ZnO nanofibers,” Curr. Appl. Phys. 13, 403–407 (2013). [CrossRef]
