Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1225-0562(Print)
ISSN : 2287-7258(Online)
Korean Journal of Materials Research Vol.29 No.5 pp.297-303
DOI : https://doi.org/10.3740/MRSK.2019.29.5.297

Flexible Hydrogen Sensor Using Ni-Zr Alloy Thin Film

Deok-Whan Yun,Sung Bum Park,Yong-il Park
Department of Advanced Materials Engineering, Kumoh National Institute of Technology, 61 Daehak-ro, Gumi, Gyeongbuk 39917, Repubilc of Koea
Corresponding author E-Mail : yipark@kumoh.ac.kr (Y. I. Park, Kumoh Inst.)
April 24, 2019 April 24, 2019 April 25, 2019

Abstract


A triple-layered PMMA/Ni64Zr36/PDMS hydrogen gas sensor using hydrogen permeable alloy and flexible polymer layers is fabricated through spin coating and DC-magnetron sputtering. PDMS(polydimethylsiloxane) is used as a flexible substrate and PMMA(polymethylmethacrylate) thin film is deposited onto the Ni64Zr36 alloy layer to give a high hydrogenselectivity to the sensor. The measured hydrogen sensing ability and response time of the fabricated sensor at high hydrogen concentration of 99.9 % show a 20 % change in electrical resistance, which is superior to conventional Pd-based hydrogen sensors, which are difficult to use in high hydrogen concentration environments. At a hydrogen concentration of 5 %, the resistance of electricity is about 1.4 %, which is an electrical resistance similar to that of the Pd77Ag23 sensor. Despite using low cost Ni64Zr36 alloy as the main sensing element, performance similar to that of existing Pd sensors is obtained in a highly concentrated hydrogen atmosphere. By improving the sensitivity of the hydrogen detection through optimization including of the thickness of each layer and the composition of Ni-Zr alloy thin film, the proposed Ni-Zr-based hydrogen sensor can replace Pd-based hydrogen sensors.



초록


    © Materials Research Society of Korea. All rights reserved.

    This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

    1. Introduction

    Hydrogen is widely regarded as the future clean fuel of major industries. However, since hydrogen sensor is still not perfect in order to prevent hydrogen leakage accident, DOE or NASA of USA have been continuing research and development on hydrogen sensor. The reason why hydrogen sensors cannot detect hydrogen completely is because of the special properties of hydrogen gas, which requires more precise and complete detection and function than similar combustible gases(CH4, C3H8, etc.) detection sensors. In addition, since hydrogen has a very high flame speed, even if a small amount of hydrogen leaks into the air, it easily explodes with a little ignition heat. Therefore, it is essential to secure the stability of hydrogen gas. Research on the detection of leaking hydrogen gas is being carried out in parallel with the development of various hydrogen technologies such as production, transportation, and storage of hydrogen energy.1-6)

    Recent studies have been carried out with Pd, which has excellent selectivity for hydrogen gas. As Pd generates a hydrogen compound due to its interaction with hydrogen, a change in mass, volume, and electrical resistance occurs, which can be used as a signal to read the leakage of hydrogen gas. However, despite the advantages of Pdbased hydrogen sensor, the structure of the thin film changes due to the volume expansion by hydrogen absorption, and the resistance hysteresis during hydrogen absorption-desorption occurs. Furthermore, there is also a cost problem because Pd is very expensive as a noble metal.7) As a solution to these problems, improvement studies have been carried out by alloying Pd with other metals such as Ag, Ni, and Mg. The Pd-Ni sensor reported by Hughes8) can detect hydrogen from 0.1 to 100 % at 300 K and Pd-Ni alloy thin films formed on the Al2O3 substrate reported by Cheng9) and Huang10) were also able to detect hydrogen at 5% at room temperature and 0.7-6 % at 100 ºC, respectively. However, the sensing ability of hydrogen decreases with increasing temperature. M. Wang,11) who applied Pd-Ag alloy with high hydrogen selectivity and permeability to hydrogen sensor, fabricated a thin film sensor with a composition ratio of Pd : Ag = 3 : 2 and 2~3 % of hydrogen could be detected. In this study, the fracture problems due to hydrogen expansion which have been reported in the Pd film was improved through alloying, but the hydrogen sensing ability decreased and the reaction time increased as the content of non-Pd metal increased. Thereafter, a hydrogen sensor with a dual layer of Pd layer on top and Sm layer on bottom12) and a hydrogen sensor with a composition of Mg90Pd10 with high hydrogen affinity13,14) were developed. However, these sensors also could not avoid the limitation of using Pd.

    In this study, amorphous Ni64Zr36 thin film fabrication process was developed and a flexible sensor using the Ni64Zr36 alloy thin film was fabricated. The hydrogen sensor has a triple sandwich structure(PMMA/Ni64Zr36/ PDMS) centered on the amorphous Ni64Zr36 alloy thin film (Fig. 1). The PMMA having excellent hydrogen selectivity and preventing the oxidation of the metal film is positioned on the upper side and the PDMS having excellent diffusion rate of hydrogen and flexibility is positioned in the lower portion in contact with the hydrogen gas.

    2. Experimental Procedure

    A solution of PDMS(polydimethylsiloxane, Sylgard 186) and a curing agent(Sylgard 184 B) in a weight ratio of 10 : 1 was spin-coated on a Si wafer(1.5 × 2.0 cm2) for 30 seconds at 3,000 rpm. To obtain a dense film, a PDMS layer having a thickness of about 200 μm was prepared by drying in a vacuum oven at 60 ºC for 3 hours. Then, a Ni64Zr36 metal film was deposited on the top of the PDMS using DC magnetron sputtering. An alloy target with a composition of Ni64Zr36 was used. The vacuum of the inner chamber before deposition was reduced to 2 × 10−6 Torr, the applied power was 10 W, and the internal argon pressure was maintained at 100 mTorr. The Ni64Zr36 film was deposited on a Si substrate of 1.5 × 2.0 cm2 coated with PDMS and the distance from the target was 15 cm. After sputtering, a PMMA solution(Microchem, 950PMMA C2) solution was spincoated on top of the fabricated Ni64Zr36/PDMS membrane at 5,000 rpm for 30 seconds and then dried on a hot plate at 60 ºC to form a 200 nm thick PMMA layer. After drying, the prepared PMMA/Ni64Zr36/PDMS triple layer film was separated from the Si wafer to complete the hydrogen sensor. The thickness of the deposited Ni64Zr36 film was confirmed by XPS depth profiling. The surface state of the polymer films was confirmed by SEM images.

    A specially designed hydrogen sensing device was used for the measurement of hydrogen sensing characteristics. A gas mixture of argon and hydrogen was controlled through a mass flow controller(MFC) to measure the degree of resistance change depending on gas composition. To verify the hydrogen permeability, a customized hydrogen separation membrane device(Nara cell-tech) was used. The device measured hydrogen permeability through a mass flow controller(MFC) and a gas chromatograph( GC)(Dong-il Shimadzu Corp.). The concentration of hydrogen in the GC system was determined by using a thermal conductivity detector(TCD) to measure the difference in thermal conductivity between argon as the carrier gas and the measured components. Fig. 2 shows a schematic diagram of the above devices.

    3. Results and Discussion

    Fig. 3(a) shows the FE-SEM images of the surface of the hydrogen sensor showing that the surface condition is not smooth but wrinkled. It is considered that the wrinkles were formed by Ar plasma in the process of depositing Ni64Zr36 film on the PDMS. In general, PDMS membranes are hydrophobic, but wrinkles are formed on the surface of the membrane by O2, Ar, N2 and NH3 plasma arbitrarily,15) and -OH groups are formed on the surface while varying the surface properties from hydrophobic to hydrophilic.16) The thickness of Ni64Zr36/PMMA film was estimated to be about 180-200 nm, and the thickness of Ni64Zr36 was estimated to be about 15-20 nm. The top PMMA layer was observed to be about 200 nm thick. There was no peeling problem from the bottom PDMS substrate. This may be positive in terms of increased surface area that affects sensing ability, but may result in irregular wrinkle structures and cracking or fracture of the upper membrane when exposed to strong plasma for a long time.17) Fig. 4 shows that many cracks occurred on the N64Zr36 surface after strong O2 plasma treatment with power of 50 W. However, no fracture or crack was found on the surface of the fabricated sensor which was fabricated using Ar plasma with power of 10 W [Fig. 3(a)].

    Fig. 5 shows the XRD patterns of the Ni64Zr36 alloy thin film deposited on the PDMS substrate. Due to the nature of PDMS, it shows a sharp peak near 12.8o and a broad peak near 22.6º.18) The XRD patterns of the PDMS film and the Ni64Zr36 alloy film deposited PDMS film are almost the same because the actual thickness of Ni64Zr36 film is very thin and its phase is amorphous.

    Fig. 6 shows the result of AFM(Atomic Force Microscopy) analysis using Ni64Zr36/PDMS/Si structure with Ni64Zr36 layer deposited on PDMS. The AFM line mapping were used to check the height and condition of the wrinkled surface. The area used for the AFM analysis was 40 × 40 μm2, but the extended lengths of the lines measured for wrinkles for the three line-mapping sections were 97.1 μm, 99.4 μm and 90.6 μm, respectively, with an average of 95.7 μm. This is almost 240 % longer than 40 μm. The calculated area obtained by using the increased length showed an increase of 5.7 times when compared with the initial plane of 40×40 μm2. The increase of the surface area is thought to be beneficial for the hydrogen sensing in terms of the improvement of the reactivity by hydrogen. Mesoporous thin films19) or nanotube arrays20,21) show that the hydrogen sensing capability is increased by having a larger surface area than conventional flat hydrogen sensors. However, as mentioned above, when the intensity of the sputtering plasma used for deposition the Ni64Zr36 layer on the PDMS film and the exposure time are increased, a large internal stress is generated. Therefore, a reduction in durability and fracture of the thin film due to severe wrinkling may occur(especially when a thick film is used).

    Fig. 7 shows the hydrogen permeability of the fabricated hydrogen sensor and the PDMS, PMMA, and Ni64Zr36 thin film, respectively. PDMS is a polymer containing Si and is widely known as a polymer that has a very good permeability to gas molecules due to a wide gap between the inner rings.28) The fabricated hydrogen sensor showed a hydrogen permeability of about 1/50~1/100 of those of the PDMS and the Ni64Zr36 film. This is because the hydrogen permeability of PMMA is as low as about 10 to 15 mol/m·s·Pa and is about 1/500 to 1/1,000 of the hydrogen permeability of PDMS.22) Therefore, the hydrogen sensing ability possibly be improved by simply removing the top PMMA layer from the triple-layered sensor. Nonetheless, PMMA is useful because of its excellent protection for Ni64Zr36 thin film from an oxidation, not to mention of the high selectivity to hydrogen, which prevents the passage of large gas molecules(CO, CO2, O2, etc.).23) Furthermore, the PMMA layer can be used as a barrier to prevent contamination of the Ni64Zr36 layer by other external materials.

    Fig. 8(a) shows the electrical characteristics of Ni64Zr36 hydrogen sensor reacting at high concentration of hydrogen. It can be seen that as the hydrogen is applied after the Ar purge at the initial room temperature, the resistance value increases. According to Balla et al.,24) it is reported that the resistance of Ni64Zr36-based glassy alloy changes from a minimum of 40 % to a maximum of 60 % when hydrogen is applied at a pressure of 3 bar to 10 bar at 100 ºC. This is because the added hydrogen reacts with the internal Zr to form Zr-H, which changes the electron density in the vicinity of the Fermi energy. The formation of Zr-H causes the electron conduction of the Zr atom to decrease in the valence band, and this leads to a decrease in the electrical conductivity. It also explained that this property depends on the composition of the alloy. It can be seen that the degree of change is about 20 %(10-30 %) at the initial stage, and when Ar purge is performed again, it is recovered close to the initial resistance value. These results show that Ni-Zr-based hydrogen sensors can replace Pd-based hydrogen sensors that are difficult to use in high-concentration hydrogen environments.

    Fig. 8(b) shows the electrical characteristics of the hydrogen sensor at 5 % hydrogen concentration. As the hydrogen is applied at the initial room temperature, the resistance value changes by 2.8 %. However, the recovery rate was about 1/2 of that of the previous high concentration hydrogen atmosphere, and it was found that only about 1.4 % of the resistance value was changed when the test was performed with 5 % hydrogen. The sensing ability at low hydrogen concentration is actually reduced, and the Zr-H state is partially maintained in the adsorptiondesorption process after the formation of Zr-H, thereby decreasing the rate of the resistance change.

    Fig. 8(c) shows the test result using a Pd77Ag23 sensor fabricated with the same triple layer structure as the Ni64Zr36 hydrogen sensor. The change of resistance was about 1.5% at room temperature, 5% hydrogen atmosphere, which was about half of the resistance of the Ni64Zr36 sensor. However, unlike the Ni64Zr36 sensor, the saturation time(reaching time to maximum resistance) was relatively reduced from 450 s to 130 s and the reaction was instantaneous when hydrogen was turned on and off. It can be explained that the catalytic performance of Pd-Ag is superior to that of Ni-Zr alloy. However, in the case of Pd-Ag alloy sensor. There was also a change between the initial resistance and the second resistance, possibly due to the time-delayed partial generation of the internal residual hydride.

    4. Conclusion

    The PMMA/Ni64Zr36/PDMS triple-layered hydrogen sensor with high hydrogen permeability, hydrogen selectivity and anti-oxidation function was designed and fabricated. The hydrogen sensor has good reactivity with hydrogen and has excellent flexibility by using PDMS as a substrate without using a conventional Si wafer or alumina substrate. At a high hydrogen concentration of 99.9 %, the resistance change of the Ni64Zr36 hydrogen sensor changes to about 20 % at the initial stage, which is better than the conventional Pd-based hydrogen sensor, which is difficult to use in high hydrogen concentration environments. At low hydrogen concentration of 5 %, the resistance change of the Ni64Zr36 sensor is about 2.8 %, which is not significantly different from 1.5 % of the Pd77Ag23 sensor. Despite using a low-cost Ni64Zr36 alloy as the main sensing element, it can achieve similar performance to conventional Pd sensors in a highly concentrated hydrogen atmosphere. Therefore, the proposed Ni-Zrbased hydrogen sensor has the potential to replace the Pd-based hydrogen sensor by improving the sensitivity of hydrogen detection.

    Acknowledgement

    This paper was supported by Research Fund, Kumoh National Institute of Technology

    Figure

    MRSK-29-5-297_F1.gif

    The structure of Ni64Zr36 alloy thin film hydrogen gas sensor.

    MRSK-29-5-297_F2.gif

    The schematic of Ni64Zr36 alloy thin film hydrogen gas sensor test system.

    MRSK-29-5-297_F3.gif

    The FE-SEM image of the fabricated PMMA/Ni64Zr36/ PDMS hydrogen sensor; (a) surface and (b) cross-section.

    MRSK-29-5-297_F4.gif

    FE-SEM image of the surface of Ni64Zr36/PDMS after strong O2 plasma treatment.

    MRSK-29-5-297_F5.gif

    The X-ray diffraction of (a) Ni64Zr36/PDMS and (b) bare PDMS substrate.

    MRSK-29-5-297_F6.gif

    The AFM surface image and line scans of the Ni64Zr36/PDMS membrane.

    MRSK-29-5-297_F7.gif

    The hydrogen permeability of PMMA/Ni64Zr36/PDMS, PDMS, PMMA, Ni64Zr36 thin film.

    MRSK-29-5-297_F8.gif

    The electric resistance change of the hydrogen sensor using; (a) Ni64Zr36 alloy thin film at 99 % hydrogen concentration, (b) Ni64Zr36 alloy thin film at 5 % hydrogen concentration, (c) Pd77Ag23 alloy thin film at 5 % hydrogen concentration.

    Table

    Reference

    1. S.-I. Yamaura, Y. Shimpo, H. Okouchi; M. Nishida and O. Kajita, Mater. Trans., 44, 1885 (2003).
    2. L. A. Ruth, Mater. High Temp., 20, 7 (2003).
    3. D. S. Newsome, Catal. Rev., 21, 275 (1980).
    4. S. E. Nam, S. H. Lee; K. H. Lee, J. Membr. Sci., 153, 163 (1999).
    5. J. Li, R. Fan, H. Hu and C. Yao, Mater. Lett., 212, 211 (2018).
    6. W. L. Watkins and Y. Borensztein, Sens Actuators B Chem,, 273, 527 (2018).
    7. B. A. McCool and Y. S. Lin, J. Mater. Sci., 36, 3221 (2001).
    8. R. C. Hughes and W. K. Schubert, J. Appl. Phys., 71, 542 (1992).
    9. C. C. Brown and R. E. Buxbaum, Metall. Trans. A, 19, 1425 (1988).
    10. S.-M. Kim, D. Chandra, W.-M. Chien, N. K. Pal, M. D. Dolan, A. Talekar, J. Lamb, S. N. Paglieri and T. B. Flanagan, Int. J. Hydrogen Energy, 37, 3904 (2012).
    11. S. Hara, N. Hatakeyama, N. Itoh, H. M. Kimura and A. Inoue, J. Membr. Sci., 211, 149 (2003).
    12. S. Hara, K. Sakaki, N. Itoh, H.-M. Kimura, K. Asami and A. Inoue, J. Membr. Sci., 164, 289 (2000).
    13. S. Hara, N. Hatakeyama, N. Itoh, H.-M. Kimura and A. Inoue, Desalination, 144, 115 (2002).
    14. S.-I. Yamaura, Y. Shimpo, H. Okouchi, M. Nishida, O. Kajita and H. Kimura, A. Inoue, Mater. Trans., 44, 1885 (2003).
    15. G. L. Holleck, J. Phys. Chem., 74, 503 (1970).
    16. H. Katsuta, R. J. Farraro and R. B. McLellan, Acta Metall., 27, 1111 (1979).
    17. M. D. Dolan, S. Hara, N. C. Dave, K. Haraya, M. Ishitsuka, A. Y. Ilyushechkin, K. Kita, K. G. McLennan, L. D. Morpeth and M. Mukaida, Sep. Purif. Technol., 65, 298 (2009).
    18. S.-I. Yamaura, M. Sakurai, M. Hasegawa, K. Wakoh, Y. Simpo, M. Nishida, H. Kimura, E. Matsubara and A. Inoue, Acta Mater., 53, 3703 (2005).
    19. C. Lu and Z. Chen, Sens. Actuators, B, 140, 109 (2009).
    20. J. Lee, D. H. Kim, S.-H. Hong and J. Y. Jho, Sens. Actuators, B, 160, 1494 (2011).
    21. E. Sennik, Z. Colak, N. Kılınc, Z. Ziya Öztürk, Int. J. Hydrogen Energy, 35, 4420 (2010).
    22. H. Katsuta, R. J. Farraro and R. B. McLellan, Acta Metall., 27, 1111 (1979).
    23. M. D. Dolan, N. Dave, L. Morpeth, R. Donelson, D. Liang, M. Kellam, S. Song, J. Membr. Sci., 326, 549 (2009).
    24. S. Balla, B. Vehovszky, A. Bárdos and M. Kovalaková, J. Phys.: Condens. Matter, 144, 012012 (2009).