Hydrogen gas sensing properties of Pt-WO3 films on spiral microstructured fiber Bragg grating (FBG) has been demonstrated. Pt-WO3 film was prepared by hydrothermal method. The spiral microsturctured FBG was fabricated using femtosecond laser. Spiral microstructure FBG hydrogen sensor can detect hydrogen concentration from 0.02% H2 to 4% H2 at room temperature, and the response time is shortened from a few minutes to 10~30 s. Double spiral microstructure at pitch 60 μm and sputtered with 2 μm Pt-WO3 film recorded hydrogen sensitivity of 522 pm/%(v/v) H2 responding to hydrogen gas in air. This translated to approximately 2~4 times higher than the unprocessed standard FBG. The humidity has little effect on the sensing property. The sensor has fast response time, good stability, large detection range and has the good prospect of practical application for hydrogen leak detection.
© 2017 Optical Society of America
Hydrogen gas as a clean energy has tremendous applications in aerospace, petrochemical plant, metallurgical industry, fuel cell and so on [1, 2]. During hydrogen gas transmission, storage and exploitation, it is easy to leak due to its minimum atomic diameter. If the concentration of hydrogen gas is over 4% in air, the hydrogen explosion may occur . Therefore, it is necessary to monitor hydrogen gas leakage avoiding a serious accident.
FBG hydrogen sensors have been widely researched due to its distributed measuring and intrinsic safety . To make FBG hydrogen sensors, different types of hydrogen-sensitive thin films are developed and used as coating on the surface of FBG fiber. Pd or Pd alloy has been widely used due to its high selection towards hydrogen gas . Pd film swells when it absorbs hydrogen gas, thus introducing change of stress and refractive index on the fiber, which results in a central wavelength shift of FBG. However, fiber hydrogen sensors based on Pd films have some defects. The detected limit of hydrogen concentration is not high enough for the weak swelling fiber. Although some researchers improved the sensitivity of FBG fiber sensor various methods such as taper , corrosion  and femtosecond laser ablation , the sensitivity of FBG fiber is still not sufficiently high. Another drawback is that Pd undergoes irreversible change in crystal structure after several cycles of absorption and de-absorption of H2. Although some methods can relieve the phenomenon of hydrogen embrittlement, such as addition of metals (Ag, Ni) into the Pd film, however, the hydrogen embrittlement can’t be fully eliminated. Thus, sensors based on Pd or Pd alloy are not very suitable for practical applications. Noble metal (Pt, Pd, etc.) loaded with WO3, present another kind of sensing material, which have been widely used to make FBG hydrogen sensor based on exothermic reaction at a hydrogen sensor film. Different WO3 grain morphology greatly affects the sensing performance, such as sensitivity and response time. There have been several WO3 structures with different morphologies such as, nanospheres , nanowires [10, 11], nanorods  and nanolamellae . It was reported that hydrogen sensors prepared with nanolamellae showed the highest response while the nanowires presented the shortest response time . Caucheteur  reported a highly sensitive FBG hydrogen sensor. However, it couldn’t detect hydrogen concentration below 0.6% H2. J. X. Dai  developed Pt-WO3 FBG hydrogen sensor with sensitivity below 1% H2 through Pt-WO3 optimization process combined with temperature sensitivity enhancement. This sensor showed 13 pm wavelength shift towards 0.04% H2. In summary, the response time of Pt-WO3 sensor is usually higher than 1 minute [15–19], thus, improving the response time of Pt-WO3 sensor would be an advantage for its practical application.
In this paper, microstructured hydrogen sensor based on Pt-WO3 film which detects the hydrogen concentration down to 0.02% H2 and up to 4% H2 at room temperature is presented. The sensor structure is composed of a microstructured FBG which is fs laser ablated combined with rotating jig. The microstructured FBG is more sensitive to thermal expansion, ultimately, resulting to greater wavelength shift and fast response time. The demonstrated sensor poses a sterling performance in sensitivity, stability and response time.
2.1 Preparation of spiral microstructured FBG
The 10 mm long FBG was written into the core of standard single mode fiber by first removing polyimide layer, then irradiating the fiber using a KrF excimer (248 nm) laser by phase mask technique. The spiral microstructure, 10 mm long was fabricated on the cladding by fs laser (IFRIT, Cyber Laser, Tokyo, Japan) which has a wavelength of 780 nm, pulse duration of 180 fs and repetition rate of 1 kHz. The fiber was mounted on a precision three-dimensional stage, while a specially designed jig held and rotated the fiber during micro-machining process. The laser beam was focused on the fiber surface using an objective lens (Sigama, Koki, Japan) with a focal length of 60 mm. The entire process was monitored on real-time by CCD. Simultaneously, the fiber was rotated around its axis at a constant rotational speed ω while still advancing at a uniform speed v along the fiber axis. Therefore, the spiral microstructure fabricated has a constant pitch that can be expressed as:Table 1. The sketch of double and single spiral microstructured FBG is shown in Fig. 1. By hydrothermal method, Pt-WO3 particles were coated on the surface of microstructured FBGs which have enhanced surface area and better stretchability as supported by a recent work .
2.2 Preparation of Pt-WO3
Firstly, the WO3•H2O nanolamellae was prepared by the hydrothermal method using Na2WO3•2H2O, citric acid and HCl as starting material , acetyl acetone platinum (Pt(acac)2) was later added to WO3•2H2O powder for catalyst preparing Pt. Subsequently, the mixture was fine ground. After 2 hours of thermal annealing process at 315 °C, Pt loaded WO3 powder was prepared. Finally, Pt-WO3 powder mixed with appropriate volume of deionized water at mole ratio of 1:5 was deposited on the surface of microstructured FBG.
2.3 Hydrogen gas testing
The characterization was performed at room temperature (20 °C) and at the relative humidity of about 27%. Hydrogen concentration was calibrated using commercial electrochemical hydrogen meter (RBT-6000 -ZLG/A; Ruian Company, China), fitted with alarm to monitor H2 concentration beyond 4% in the chamber. A hydrogen concentration meter based on chemical electrics principles was connected to the chamber for calibration. The varying hydrogen concentration is provided by changing flowing rate of pure H2 injected into a gas chamber. The sensors in the chamber were connected to a FBG demodulator based on CCD demodulation principle, which has high precise to detect 1 pm wavelength shift. The output port of the demodulator was connected to a computer to collect and analyze the measured data.
In comparison with the standard FBG sensor, the sensitivity of microstructured FBG hydrogen sensor can be enhanced. The reasons are as follows: Firstly, more heat will be produced because microstructured FBG have more surface area to accommodate more nanostructure films. Secondly, the expansion of fiber also changes the grating pitch although the expansion coefficient of fiber is small (0.55 × 10−6/K), generally, we can ignore fiber expansion effect. Since, the Pt-WO3 expansion coefficient is higher than that of SiO2, the microstructured FBG sensor has a higher flexibility and ultimately experiences greater sensitivity enhancement. Thirdly, we can also ignore the thermal stress because of the uniform distribution of the thermal stress along the fiber. However, the internal stress distribution of fiber maybe inhomogeneous after fs ablation, thus, the wavelength shift may increase due to non-uniform distribution of the thermal stress. Combination of these factors will improve the sensitivity of microstructured FBG probe greatly. The response time of microstructured FBG probe also can be improved compared with a standard FBG probe. The greater the microgroove depth, the greater is the amount of nano-powder closer to the fiber core. As a result, the greater heat is generated by exothermic reaction and faster transferred to the fiber core. Therefore, the microstructured FBG hydrogen sensor has faster response time than that of standard FBG sensor.
4. Results and discussions
4.1 Characterization of Pt-WO3
Figure 2(a) demonstrates the surface morphology of double spiral microstructured FBG with pitch of 60 μm coated with Pt-WO3 using digital microscope (Keyence, VHX-100, Japan). It can be seen that the film is basically uniformly coated on the surface of FBG. The crystal morphology of Pt-WO3 was analyzed by Field Emission Scanning Electron Microscopy (FE-SEM, Zeiss Ultra plus, German) and the element composition was analyzed using Energy Dispersive X-ray Spectrometer (EDS) attached to the FE-SEM. Figure 2(b) shows the cross section picture of film coated on the surface of fiber. The thickness of the film is about 2 μm. The particles of Pt-WO3 present in nanolamellaes with different sizes are shown in Fig. 2(c). It evident from Figs. 2(b) and 2(c) that the morphology is much porous which permit hydrogen gas permeate into WO3 film quickly. Figure 2(d) shows the atomic ratio of Pt:W = 1:5, which have higher sensitivity than other proportion of Pt-WO3 .
4.2 Hydrogen gas sensing performance
Figure 3 presents hydrogen sensing performance of four samples. All the samples responded quickly as soon as the hydrogen gas flowed into the gas chamber. Figure 3(a) demonstrates the wavelength shift of the double spiral microstructure sensor (sample-1). The amount of central wavelength shifts are 11, 25, 55, 71, 173, 195, 271, 364 and 530 pm under the concentration of 0.02, 0.04, 0.08, 0.12, 0.36, 0.4, 0.52, 0.68 and 1% H2, respectively. It also can be seen that two cycles of hydrogen response under 0.36% H2. It shows good repeatability during the cycle test. The FBG hydrogen sensors present a linear response from 0.02% H2 to 1% H2 as shown in Fig. 3(f). The sensitivity of sample-1 is about 522 pm/%(v/v) H2 derived from the Fig. 3(d).
For single spiral microstructure sensor (sample-2) shown in Fig. 3(b), the wavelength shifts are 10 and 293 pm at the hydrogen concentration of 0.04% and 1% H2, respectively. Obviously, the amount of wavelength shift for single spiral sensor is lower than that of double spiral sensor under the same processing parameters. The sensitivity of sample-1 is about 1.8 times that of sample-2. The reason is that the double spiral microstructure FBG sensor has double micro-trenches and thus accommodates more films, thus becoming more flexible in comparison to the single spiral FBG sensor.
For sample-3 presented in Fig. 3(c), the processing parameters is same as sample-1 except for laser energy. The wavelength shifts are 9 pm and 353 pm at 0.04% H2 and 1.04% H2, respectively. Thus, it is evident that the wavelength shift of sample-3 is lower than that of sample-1. It demonstrates that higher laser energy can improve the sensitivity of the hydrogen sensors. The main reason is that the higher laser energy makes deeper microgroove which accommodates more Pt-WO3 nanoparticles.
Figure 3(d) shows the performance of standard FBG sensor. The wavelength shifts are 6, 15, 28, 64, 122 and 270 pm at 0.08, 0.24, 0.32, 0.48, 0.68 and 1% H2, respectively. Due to the nonlinear relationship between the hydrogen concentration and wavelength shift, the sensitivity of standard FBG sample was 75, 62.5, 87.5, 133.3, 179.4 and 270 pm/%(v/v) H2 at 0.08, 0.24, 0.32, 0.48, 0.68 and 1% H2, respectively. We simply took the average of the sensitivity below 1% H2, it appropriately considered that the maximum sensitivity of standard FBG sensor is 270 pm/%(v/v) H2, the average sensitivity is about 130 pm/%(v/v) H2 below 1% H2. As shown in Fig. 3(e), hydrogen sensitivity of sample-1 is the highest, followed by sample-2, sample-3 and standard FBG sensor. Compared with other FBG sensors based on Pd thin film, the sensitivity of microstructured FBG sensor based on Pt/WO3 is greatly improved. For instance, ref . reported the tapered FBG sensor with a maximum sensitivity of 81.8 pm/%(v/v) H2. Although the sensitivity of tapered FBG sensor was enhanced, its performance is inferior compared to the laser-assisted-microstructured FBG based on Pt/WO3. The sensitivity of sample-1 is about 2 times than that of standard FBG sensor under higher concentrations (>1% H2), furthermore, the sensitivity of sample-1 is 2-4 times compared to standard FBG sensor under low hydrogen concentration (<1% H2). One of the reasons may be the activation of the reaction by the light. Reference  reported that temperature can be increased by a light coupling to the sensitive layer. It can be inferred that the higher sensitivity is in partly contributed by the effect of light. At low hydrogen concentrations, since the majority of light in standard FBG is coupled into the fiber core, the light energy can’t activate the sensitive layer, the sensitivity and response speed is uniform no matter what the hydrogen concentration it is. However, for spiral microstructured FBG, the partial light energy may have leak from the fiber core due to grooves’ depth. Thus, the light energy could have enhanced the temperature in case of a slight light leaks. As grooves’ depth and approaches the core region, the possibility of light leakage is greater. Such interaction with light, may increase activation of the sensitive layer as shown in Fig. 3(e). The slope seems the same even at high hydrogen concentration. The reason behind this effect may be due to the autocatalytic reaction that rises temperature increases and reaction rate at high concentration of H2. Another reason could be that at low hydrogen concentrations, the proportion of increased the heat amount is greater than that of high hydrogen concentration, thus, the double spiral microstructure sensor has higher sensitivity under lower hydrogen concentration. It also can be seen in Fig. 3(e), the slopes seem to become the same at high hydrogen concentration, the reason is contributed to the effect of autocatalytic reaction that the temperature rises increases the reaction rate at high concentration of H2.
As shown in Figs. 4(a)-4(c), two typical cycle responses of sensors under 1% H2 were chosen from Figs. 3(a), 3(b) and 3(d). Response time is defined as 90% wavelength shift of FBG from the hydrogen gas flow into the gas chamber, while recovery time is defined as decreasing 90% wavelength shift of FBG from samples exposed to air. The response time of 90% wavelength shift is about 15 s, 28 s and 70 s shown in Figs. 4(a)-4(c), respectively. When the hydrogen gas was injected into the gas chamber, the wavelength sharply increased and reached 90% in a short time for sample-1. However, the response speed underwent relatively slow growth for standard FBG. After hydrogen gas was switched off, the wavelength quickly recovered for sample-1 (about 16 s) and sample-2 (about 37 s), while the recovery time is about 33 s for standard FBG sample. One reason is that the three dimensions microstructures can conduct heat faster from the surface of fiber to the fiber core due to its enhanced surface area. Another reason is also the effect of light as above discussed. Compared with sample-1, the depth of grooves of smaple-2 is smaller. Hence, the response time is lower. In a word, the response time and recovery time of sample-1 is greatly improved.
The response and recovery time changes under different hydrogen concentrations. The response time is about 10~20 s when the concentration of hydrogen is below 1% H2, while the recovery time is about 8~20 s. The response time is about 20~30 s as hydrogen concentration is higher than 1% H2, while the recovery time is about 10~100 s. The recovery time is gradually increased with the growing of hydrogen concentrations due to increasing exothermic temperature. It can be seen that the response time and recovery time is basically same under the lower concentrations due to a reversible reaction and limited exothermic temperatures. Therefore, the central wavelength is quickly resumed. Okazaki et al. reported a hydrogen sensor based on Pt/WO3 with response times above 250 s for 1% H2 in N2 and Sumida et al. presented a Pt/WO3 based multimode fiber sensor which displayed a response time of 5 min for 1% H2 in N2 [22, 24]. Notably, Pt/WO3 films coatings in both case were prepared by sol-gel method. However, the reported response time is longer than that of sensor presented here. The enhanced performance is achieved through optimization of hydrogen composite film at (Pt:W = 1:5) atomic ratios. Additionally, we integrated hydrothermal coating method followed by medium-high temperature annealing process (temperature 315°C). Ultimately, a porous Pt/WO3 with nanolamellaes morphology was developed which is more suitable for hydrogen sensing.
In recent years, researches on Pt-WO3 hydrogen gas sensors have increased dramatically. The response time reported by ref [11–14] is over 1 minute, thus the microstructured FBG sensor presented in this paper has greatly improved response time parameter to less than a minute. Interferometer hydrogen sensors have very high sensitivity [17, 25], however, temperature cross sensitivity of photonic crystal fiber and panda fiber was also high and can’t to be easily compensated. Compared with interferometer hydrogen sensors, FBG hydrogen sensor is suitable for distributing measure. The performance of microstructured FBG hydrogen sensor towards low hydrogen concentrations is highly improved than that reported in [14, 15]. Most importantly, spiral microstructured FBG hydrogen sensor enjoys a wide range of detection of hydrogen concentration from 0.02% to 4% H2 without compromising on its sensitivity.
There are some different reports about the influence of humidity on the hydrogen gas sensing on WO3. It was reported that the hydrogen response of WO3 nano-dot-based micro sensors was insensitive to relative humidity above 50% . Y. Yamaguchi reported the declined performance of Pt-WO3 film with an increase of humidity . In order to investigate the influence of humidity on the hydrogen response of Pt-WO3 FBG probe, we placed supersaturated sodium chloride solution and sodium bromide solution into the gas chamber, as a result, relative humidity of 85% and 52% was generated in the chamber. Figure 5 presents the hydrogen response of sample-1 under different relative humidity at 0.2% H2. The wavelength shifts are 96, 88 and 87 pm under 27% RH, 52% RH and 85% RH, respectively. Compared with the response under 27% RH, the wavelength shift was reduced slightly at 85% and 54% RH. It was observed that the response speed was slightly decreased with the growing of humidity as shown in Fig. 5. When the probe was placed in the high humidity environment, tremendous water molecules were absorbed on the surface of film, which hindered the penetration of hydrogen atoms. However, hydrogen atoms remain fast penetrated into the film due to porous structure of Pt-WO3 film, the response time was just delayed for 1-2 seconds at high relative humidity environment. It was considered that the humidity has little influence on the performance of Pt-WO3 probe. Similar results were also reported by ref [14, 15].
Figure 6 shows the hydrogen response for a microstructured FBG sensor after two months exposure in ambient air. The wavelength shifts are 20, 45, 77, 136, 272, 362, 540, 1000 and 2210 pm under the hydrogen concentration of 0.04, 0.08, 0.12, 0.28, 0.52, 0.68, 1, 1.8, 4% H2, respectively. A few pico-meters wavelength fluctuations were observed, which demonstrated good stability of the probe. Compared with other hydrogen sensors based on Pd films which are easily oxidized after long time in air exposure, the Pt-WO3 FBG sensor is very suitable for long time monitoring of hydrogen leakage in air.
To investigate the hydrogen selectivity of the sensor, the samples were exposed to CH4 and CO gases in air at 4% volume. Since the central wavelength did not change, we concluded that the Pt/WO3 film was not responsive to CH4 and CO gases and thus it has good selectivity toward hydrogen gas at room temperature.
Spiral microstructured FBG hydrogen sensor based on Pt-WO3 composite film has been proposed and demonstrated. The response time was greatly improved compared to that of standard FBG hydrogen sensor. The sensitivity of hydrogen sensor was also improved 2~4 times. The mole ratio of Pt:W was 1:5 which has higher sensitivity. At room temperature of 20 °C and relative humidity of 27%, the double spiral microstructured FBG hydrogen sensor has a 11 pm, 25 pm, 530 pm and 2.21 nm wavelength shift at 0.02%, 0.04%, 1% and 4% H2, respectively. The experiment demonstrated that double spiral probe has the strongest response to hydrogen gas, followed by single spiral type sensor, standard FBG type, and all the samples have good stability during the hydrogen response. The humidity has little effect on the sensing performances. As a result, integrating microstructured FBG and Pt-WO3 composite film, demonstrate good prospects for hydrogen gas monitoring in future.
This work is finically supported by the Project of National Natural Science Foundation of China, NSFC (Number: 61475121, 61290311).
References and links
1. I. P. Jain, “Hydrogen the fuel for 21st century,” Int. J. Hydrogen Energy 34(17), 7368–7378 (2009). [CrossRef]
2. Y. R. Kim, H. J. Lee, S. Kim, and I. S. Jeung, “A flow visualization study on self-ignition of high pressure hydrogen gas released into a tube,” Proc. Combust. Inst. 34(2), 2057–2064 (2013). [CrossRef]
3. D. A. Crowl and Y. D. Jo, “The hazards and risks of hydrogen,” J. Loss Prev. Process Ind. 20(2), 158–164 (2007). [CrossRef]
4. T. Hübert, L. Boon-Brett, G. Black, and U. Banach, “Hydrogen sensors-A review,” Sens. Actuators B Chem. 157(2), 329–352 (2011). [CrossRef]
5. N. Javahiraly, “Review on hydrogen leak detection: comparison between fiber optic sensors based on different designs with palladium,” Opt. Eng. 54(3), 030901 (2015). [CrossRef]
6. S. Silva, L. Coelho, J. M. Almeida, O. Frazao, J. L. Santos, F. X. Malcata, M. Becker, M. Rothhardt, and H. Bartelt, “H2 sensing based on a Pd-coated tapered-FBG fabricated by DUV femtosecond laser technique,” IEEE Photonics Technol. Lett. 25(4), 401–403 (2013). [CrossRef]
7. J. Dai, M. Yang, Z. Yang, Z. Li, Y. Wang, G. Wang, Y. Zhang, and Z. Zhuang, “Enhanced sensitivity of fiber Bragg grating hydrogen sensor using flexible substrate,” Sens. Actuators B Chem. 196(196), 604–609 (2014). [CrossRef]
8. X. Zhou, Y. Dai, M. Zou, J. M. Karanja, and M. Yang, “FBG hydrogen sensor based on spiral microstructure ablated by femtosecond laser,” Sens. Actuators B Chem. 236, 392–398 (2016). [CrossRef]
9. J. Li, Q. Zhao, G. Zhang, J. Chen, L. Zhong, L. Li, J. Huang, and Z. Ma, “Synthesis of monoclinic WO3 nanosphere hydrogen gasochromic film via a sol-gel approach using PS-b-PAA diblock copolymer as template,” Solid State Sci. 12(8), 1393–1398 (2010). [CrossRef]
10. L. F. Zhu, J. C. She, J. Y. Luo, S. Z. Deng, J. Chen, and N. S. Xu, “Study of physical and chemical processes of H2 sensing of Pt-coated WO3 nanowire films,” J. Phys. Chem. C 114(36), 15504–15509 (2010). [CrossRef]
11. J. Kukkola, M. Mohl, A. Leino, J. Mäklin, N. Halonen, A. Shchukarev, Z. Konya, H. Jantunen, and K. Kordas, “Room temperature hydrogen sensors based on metal decorated WO3 nanowires,” Sens. Actuators B Chem. 186(18), 90–95 (2013). [CrossRef]
12. M. Horprathum, T. Srichaiyaperk, B. Samransuksamer, A. Wisitsoraat, P. Eiamchai, S. Limwichean, C. Chananonnawathorn, K. Aiempanakit, N. Nuntawong, V. Patthanasettakul, C. Oros, S. Porntheeraphat, P. Songsiriritthigul, H. Nakajima, A. Tuantranont, and P. Chindaudom, “Ultrasensitive hydrogen sensor based on Pt-decorated WO3 nanorods prepared by glancing-angle dc magnetron sputtering,” ACS Appl. Mater. Interfaces 6(24), 22051–22060 (2014). [CrossRef] [PubMed]
13. A. Boudiba, C. Zhang, P. Umek, C. Bittencourt, R. Snyders, M. Olivier, and M. Debliquy, “Sensitive and rapid hydrogen sensors based on Pd–WO3 thick films with different morphologies,” Int. J. Hydrogen Energy 38(5), 2565–2577 (2013). [CrossRef]
14. C. Caucheteur, M. Debliquy, D. Lahem, and P. Megret, “Hybrid fiber gratings coated with a catalytic sensitive layer for hydrogen sensing in air,” Opt. Express 16(21), 16854–16859 (2008). [CrossRef] [PubMed]
15. J. Dai, M. Yang, Z. Yang, Z. Li, Y. Wang, G. Wang, Y. Zhang, and Z. Zhuang, “Performance of fiber Bragg grating hydrogen sensor coated with Pt-loaded WO3 coating,” Sens. Actuators B Chem. 190(1), 657–663 (2014). [CrossRef]
16. Z. Li, M. Yang, J. Dai, G. Wang, C. Huang, J. Tang, W. Hu, H. Song, and P. Huang, “Optical fiber hydrogen sensor based on evaporated Pt/WO3 film,” Sens. Actuators B Chem. 206, 564–569 (2015). [CrossRef]
17. B. Xu, C. L. Zhao, F. Yang, H. Gong, D. N. Wang, J. Dai, and M. Yang, “Sagnac interferometer hydrogen sensor based on panda fiber with Pt-loaded WO3/SiO2 coating,” Opt. Lett. 41(7), 1594–1597 (2016). [CrossRef] [PubMed]
18. Y. Wang, D. N. Wang, F. Yang, Z. Li, and M. Yang, “Sensitive hydrogen sensor based on selectively infiltrated photonic crystal fiber with Pt-loaded WO3 coating,” Opt. Lett. 39(13), 3872–3875 (2014). [CrossRef] [PubMed]
19. S. Silva, L. Coelho, O. Frazao, J. Santos, and F. Malcata, “A review of palladium-based fiber-optic sensors for molecular hydrogen detection,” IEEE Sens. J. 12(1), 93–102 (2012). [CrossRef]
20. W. Zeng, Y. Li, B. Miao, and K. Pan, “Hydrothermal synthesis and gas sensing properties of WO3•H2O with different morphologies,” Physica E 56, 183–188 (2014). [CrossRef]
21. C. Caucheteur, M. Debliquy, D. Lahem, and P. Megret, “Catalytic fiber Bragg grating sensor for hydrogen leak detection in air,” IEEE Photonics Technol. Lett. 20(2), 96–98 (2008). [CrossRef]
22. S. Okazaki, H. Nakagawa, S. Asakura, Y. Tomiuchi, N. Tsuji, H. Murayama, and M. Washiya, “Sensing characteristics of an optical fiber sensor for hydrogen leak,” Sens. Actuators B Chem. 93(1), 142–147 (2003). [CrossRef]
23. S. Magne, S. Rougeault, M. Vilela, and P. Ferdinand, “State-of-strain evaluation with fiber Bragg grating rosettes: application to discrimination between strain and temperature effects in fiber sensors,” Appl. Opt. 36(36), 9437–9447 (1997). [CrossRef] [PubMed]
24. S. Sumida, S. Okazaki, S. Asakura, H. Nakagawa, H. Murayama, and T. Hasegawa, “Distributed hydrogen determination with fiber-optic sensor,” Sens. Actuators B Chem. 108(1-2), 508–514 (2005). [CrossRef]
25. Y. Wang, M. Yang, G. Zhang, J. Dai, Y. Zhang, Z. Zhuang, and W. Hu, “Fiber optic hydrogen sensor based on Fabry–Perot interferometer coated with Sol-Gel Pt/WO3 coating,” J. Lightwave Technol. 33(12), 2530–2534 (2015). [CrossRef]
26. R. Calavia, A. Mozalev, R. Vazquez, I. Gracia, C. Cané, R. Ionescu, and E. Llobet, “Fabrication of WO3 nanodot-based microsensors highly sensitive to hydrogen,” Sens. Actuators B Chem. 149(2), 352–361 (2010). [CrossRef]
27. Y. Yamaguchi, S. Imamura, K. Nishio, and K. Fujimoto, “Influence of temperature and humidity on the electrical sensing of Pt/WO3 thin film hydrogen gas sensor,” J. Ceram. Soc. Jpn. 124(6), 629–633 (2016). [CrossRef]