JangDongJun1
ChoU Jin1
JeonYouhyeong1
LeeTaeYong1
KimRyangHa1
KimYounglae1
KwonMin-Woo1*
-
(Department of Electric Engineering, Gangneung-Wonju National University, Gangneung,
Gangwondo 25457, Korea)
Copyright © The Institute of Electronics and Information Engineers(IEIE)
Index Terms
Hydrogen, gas sensor, palladium nanoparticles, molybdenum disulfide
I. INTRODUCTION
Importantly, the depletion of fossil fuels and emissions of hazardous gases such as
NO$_{\mathrm{x}}$ and CO$_{\mathrm{x}}$ have highlighted alternative energy fuels.
Various fuels are being developed, but the development of high-performance fuels is
insufficient. Recently, H$_{2}$ fuel energy is rapidly gaining attention as a clean
energy source. H$_{2}$ has colorless and odorless properties when the concentration
in the atmosphere is 4% or higher, making it difficult for human senses to detect
explosiveness [1-3]. H$_{2}$ gas has wide flammability and metal corrosiveness. Therefore, real-time
sensing and monitoring of H$_{2}$ leaks in various H$_{2}$ infrastructures and fields
are important. In addition, the sensitivity and response/recovery time in detecting
H$_{2}$ under various environment & temperature conditions are critical issues that
need to be addressed.
H$_{2}$ sensing technology has been extensively studied based on catalytic combustion,
electrochemical, and semiconductor metal oxide (SMO) sensors. Table 1. describes the
characteristics and mechanisms of each sensor. To improve the sensing parameters of
H$_{2}$ sensors, various strategies have been developed for nanomaterial-based H$_{2}$
gas sensors, which have gained attention for their high sensing performance and simple
structure [12,13]. Firstly, Pd-NP [14], palladium nanowires (Pd-NW), and palladium nanotubes (Pd-NT) have been developed
as a Pd network not only reduce the activation energy of H$_{2}$ adsorption but also
exhibit high catalytic properties for H$_{2}$ sensing [15]. Secondly, SMO (TiO$_{2}$ [10], Nb$_{2}$O$_{5}$ [16], ZnO)-based H$_{2}$ gas sensors are known to have high reactivity and catalytic activation
to detect H$_{2}$. Lastly, 1D single-walled carbon nanotubes (SWCNTs) and 2D transition
metal dichalcogenides (TMDs) nanosheet-based H$_{2}$ gas sensors also show excellent
H$_{2}$ sensing performance at low operating temperatures and promising detection
platforms in practical H$_{2}$ applications [17-19].
In this study, Fig. 1 presents nanocomposites H$_{2}$ gas sensor decorated with Pd on MoS$_{2}$ layer as
a channel. We analyze the H$_{2}$ sensing mechanism according to MoS$_{2}$ synthesis
and Pd coating techniques.
Fig. 1. Proposed nanocomposites H$_{2}$ gas sensor structure.
Table 1. Comparison of various H$_{2}$ gas sensors
II. NANO MATERIALS
Nanomaterial technology is being actively researched in various industries and applications.
Among the various infrastructures, MoS$_{2}$ of 2D materials is also attracting attention
in the field of gas sensors. MoS$_{2}$ nanosheets allow gas molecules to penetrate
and diffuse between layers through van der Waals forces. MoS$_{2}$ is also highly
reactive to toxic gases due to its high surface area-to-volume. However, MoS$_{2}$
is insensitive to nonpolar molecular sensing such as H$_{2}$ gas [20]. To effectively detect H$_{2}$, Pd with high catalytic properties for H$_{2}$ was
decorated. Fig. 2 illustrates the operation of Pd-H$_{2}$ detection. When H$_{2}$ is injected into
a gas sensor, the absorbed H$_{2}$ reacts with Pd to produce a Pd hybrid (PdH$_{\mathrm{x}}$),
which combines with O$_{2}$ in the air to form H$_{2}$O. The technology based on Pd
nanoparticles operates on the principle that H$_{2}$ is dissociated within the lattice
of Pd to form PdH$_{\mathrm{x}}$ and induce a change in resistance. Also, the Pd-H$_{2}$
mechanism enables a fast response time for H$_{2}$ detection through electrochemical
modulation. As the high sensing specific surface area of MoS$_{2}$ increases, it exhibits
high sensing sensitivity and excellent selectivity for H$_{2}$ detection. Moreover,
the high electron mobility of MoS$_{2}$ activates the reaction of H$_{2}$ as it induces
rapid oxygen ionization within the MoS$_{2}$ channel. Fig. 1(a) proposed an H$_{2}$ sensor by applying a nanocomposite of Pd-MoS$_{2}$. The optimization
of Pd decoration and MoS$_{2}$ synthesis is critical. If the Pd deposit thickness
is as 1 nm, Pd molecules on MoS$_{2}$ interfaces may be unclear and Pd nanoparticles
may have low coverage density [22]. When Pd nanoparticles are deposited more than 8 nm, the distance between particles
disappears and the H$_{2}$ reaction volume immersion decreases. The formation of Pd
nanosheets not only reduces the H$_{2}$ detection area of MoS$_{2}$ but also weakens
the H$_{2}$ sensing characteristics. The bandgap changes according to the number of
MoS$_{2}$ deposited layers. Monolayer MoS$_{2}$ is a semiconductor with a direct bandgap
of 1.8 eV, whereas multilayer MoS$_{2}$ has an indirect bandgap that decreases to
1.2 eV as the number of layers increases. Additionally, the number of layers in MoS$_{2}$
controls the interaction properties of electrons and the electrical conductivity characteristics.
In the case of multilayer MoS$_{2}$, there is increased electron mobility, leading
to a decrease in the bandgap and improved electrical properties. Through the optimization
of the fabrication process, various Pd-MoS$_{2}$ gas sensor samples were fabricated,
and I-V characteristics and sensing I-T curve were analyzed.
Fig. 2. Pd-H$_{2}$ of reaction mechanism.
III. FABRICATION METHOD
Fig. 2(a) shows an overview of the fabrication process of the Pd-MoS$_{2}$ based H$_{2}$ sensor.
Fig. 3(b) shows the channel and metal electrode deposition size of the fabricated nanocomposites
H$_{2}$ gas sensor. After P$^{+}$ (Boron) implantation a Si wafer substrate, SiO$_{2}$
layer thickness of 300 nm was accumulated on the Si body by wet oxidation. To deposit
the channel, MoS$_{2}$ is synthesized through chemical vapor deposition (CVD). Monolayer
MoS$_{2}$ was deposited approximately 4 layers (0.7 nm) and multilayer MoS$_{2}$ was
deposited about 8 layers (1.4 nm). Fig. 4(b) measured a sample of H$_{2}$ gas sensor by Raman spectroscopy of MoS$_{2}$. After
the pattern is formed through a lithography process, Pd is decorated on MoS$_{2}$
using DC sputter as shown in Fig. 4(c). The sputtering process is conducted under two conditions: 4 seconds and 10 seconds
at 100 W and Ar flow rate of 30 CC. If the Pd thickness exceeds 10 nm, the Pd nanoparticles
are formed into Pd nanosheets. As shown in Fig. 4(d), Pd nanoparticles were decorated with 3 nm and 8 nm. As the last step, a two-terminal
Ti/Au electrode is deposited with thicknesses of 10 nm/40 nm.
Fig. 3. (a) The fabrication process flow of Pd-MoS$_{2}$ network gas sensor; (b) The schematic drawing of proposed H$_{2}$ gas sensor structure.
Fig. 4. (a) The schematic diagram of MoS$_{2}$ synthetic method; (b) Raman spectroscopy of the MoS$_{2}$ on SiO2/Si, showing a distribution from 350-450 cm$^{-1}$; (c) The schematic diagram of DC sputter fabrication; (d) Pd thickness conditions as deposition time for DC sputtering.
IV. GAS SENSOR MECHANISM
A detailed H$_{2}$ sensing mechanism is presented in Fig. 5. Prior to the injection of H$_{2}$ gas, oxygen molecules are adsorbed onto the surface
of MoS$_{2}$ and converted into oxygen ions. When the gas sensor is exposed to H$_{2}$
gas, the Pd coating on the surface adsorbs H$_{2}$. As the concentration of H$_{2}$
gas increases, H$_{2}$ diffuses from Pd to the MoS$_{2}$ surface through the spillover
effect. Subsequently, the oxygen ions and H$_{2}$ within the MoS$_{2}$ channel react,
resulting in the production of H$_{2}$O and electrons. These chemical reactions induce
changes in resistance in the MoS$_{2}$ channel. When observing the MoS$_{2}$-Pd network
gas sensor, the Schottky barrier is lowered, allowing current to flow in response
by H$_{2}$ injection.
Fig. 5. H$_{2}$ sensing mechasim of Pd decorated MoS$_{2}$ gas sensor: (a) When H$_{2}$ gas is not exposed to the sensor; (b) When H$_{2}$ gas is injected to the sensor.
V. MEASUREMENT RESULT
Fig. 6 shows the I-V characteristics of the fabricated Pd-decorated MoS$_{2}$ gas sensor.
The I-V curves of the typical sensor in air exhibit obvious nonlinear operation at
room temperature by vacuum probe station. A voltage DC sweeps of -5 V to +5 V was
applied to the two-terminal electrodes. A significant difference of 1,000 times [3
orders] was observed when comparing the two devices coated with Pd 3 nm, with multilayer
MoS$_{2}$ showing a larger difference compared to monolayer MoS$_{2}$. The number
of MoS$_{2}$ layers not only affects the bandgap but also leads to different resistance
values. Moreover, Pd 8nm decorated multilayer MoS$_{2}$ demonstrates characteristics
similar to a selector device in resistive random-access memory.
Fig. 7(a) and 5(b) are the results of extracting H$_{2}$ detection responses of monolayer MoS$_{2}$
and multilayer MoS$_{2}$ deposited with Pd 3 nm. To perform I-T curve measurements,
a nanocomposite H$_{2}$ gas sensor was placed inside the gas probe station chamber.
The N$_{2}$ gas was injected for 1 minute at a flow rate of 100 standard cubic centimeters
per minute (SCCM) to minimize the reaction with other gases. When injecting H$_{2}$
gas at a 20 SCCM, all gases were controlled by a mass flow controller (MFC). During
the measurement, a voltage of 3~V was applied to the Au/Ti electrode at room temperature.
The gas concentrations were supplied at 10000 ppm, 20000 ppm, 30000 ppm, and 40000
ppm to detect the LOD. The sensitivity was calculated as R$_{0}$/R$_{\mathrm{g}}$,
where R$_{\mathrm{g}}$ denotes the channel resistance value of the sensor after H$_{2}$
gas injection, and R$_{0}$ presents the resistance of the gas sensor in the presence
of an N$_{2}$ gas without H$_{2}$. To extract the response time, N$_{2}$ is injected
to stabilize the resistance. To extract the response time, N$_{2}$ was injected to
stabilize the resistance. The response time was measured as the time exceeding 90%
of the initial resistance change from the injection of H$_{2}$ gas.
With the supply of H$_{2}$ gas, Fig. 7(a) shows that the sensitivity of the Pd 3 nm - monolayer MoS$_{2}$ gas sensor is about
22.19 and the response time is about 77.4 sec. In Fig. 7(b), the sensitivity of the Pd 3 nm - multilayer MoS$_{2}$ gas sensor was 6.513, and
the response time was extracted as 220.8 sec. Therefore, it may be seen that MoS$_{2}$
as multilayer structure has excellent reaction characteristics of H$_{2}$ gas. On
the other hand, Fig. 7(c) shows that there is no reactivity to H$_{2}$ gas when the Pd thickness is 8 nm. If
deposited as Pd nanosheets rather than Pd nanoparticles, the reaction area of the
sensing gas is reduced, and the catalytic properties of H$_{2}$ are degraded [21].
Fig. 6. I-V characteristics of the Pd decorated MoS$_{2}$ network gas sensors measured by a typical DC double sweep.
Fig. 7. Dynamic I-T curves of (a) Pd 3 nm – monolayer MoS$_{2}$ gas sensor; (b) Pd 3 nm – multilayer MoS$_{2}$ gas sensor; (c) Pd 8 nm – multilayer MoS$_{2}$ gas sensor as 1% to 4% H$_{2}$ concentration sensing.
V. CONCLUSIONS
In this study, we proposed a resistive-type H$_{2}$ gas sensor with Pd-NP decorated
on MoS$_{2}$ as channel network. Then, the chemiresitive H$_{2}$ detection mechanism
in the sensing layer according to MoS$_{2}$ CVD synthesis and Pd nanoparticles coating
technique is introduced. The sensitivity and response time of the fabricated H$_{2}$
gas sensors are measured through the input voltage signal. We demonstrated high H$_{2}$
reaction volume density properties through deposition of Pd nanoparticles. In addition,
we compared the sensing parameters of H$_{2}$ gas following the interaction of electrons
according to different energy band gaps with MoS$_{2}$. Finally, the H$_{2}$ sensor
was optimized through Pd nanoparticle application technology, and the electrical H$_{2}$
gas sensing behavior according to the number of MoS$_{2}$ layers was confirmed through
the input voltage.
ACKNOWLEDGMENTS
This research was supported by the National R&D Program through the National Research
Foundation of Korea (NRF) funded by the Ministry of Science and ICT (NRF-2022M3I7A1078936)
and this research was supported by "Regional Innovation Strategy (RIS)" through the
National Research Foundation of Korea (NRF) funded by the Ministry of Education (MOE)(2022RIS-005)
and also, supported by the National Research Foundation of Korea (NRF) grant funded
by the Korea government (MSIT) (2021R1G1A1093786)
References
Sundarrajan, Subramanian, Suleyman I. Allakhverdiev, and Seeram Ramakrishna. "Progress
and perspectives in micro direct methanol fuel cell." International Journal of Hydrogen
Energy 37.10 (2012): 8765-8786.
Xie, Huaqing, et al. "Thermodynamic study for hydrogen production from bio-oil via
sorption-enhanced steam reforming: Comparison with conventional steam reforming."
international journal of hydrogen energy 42.48 (2017): 28718-28731.
Koo, Won-Tae, et al. "Chemiresistive hydrogen sensors: fundamentals, recent advances,
and challenges." ACS nano 14.11 (2020): 14284-14322.
Sze, Simon M., Yiming Li, and Kwok K. Ng. Physics of semiconductor devices. John wiley
& sons, 2021.
Grimes, Craig A., Elizabeth C. Dickey, and Michael V. Pishko, eds. Encyclopedia of
sensors. Vol. 10. Stevenson Ranch: American Scientific Publishers, 2006.
RaviPrakash, J., et al. "Hydrogen sensors: Role of palladium thin film morphology."
Sensors and Actuators B: Chemical 120.2 (2007): 439-446.
Cho, Minkyu, et al. "Fast Flexible Bottom‐Gated Hydrogen Sensor Based on Silicon Nano-membrane."
Advanced Materials Technologies 6.5 (2021): 2000847.
Honarpazhouh, Yasaman, et al. "Electrochemical hydrogen storage in Pd-coated porous
silicon/graphene oxide." International Journal of Hydrogen Energy 41.28 (2016): 12175-12182.
Sarkar, Shreya, and Sebastian C. Peter. "An overview on Pd-based electrocatalysts
for the hydrogen evolution reaction." Inorganic Chemistry Frontiers 5.9 (2018): 2060-2080.
Li, Zhong, et al. "Resistive-type hydrogen gas sensor based on TiO2: A review." International
Journal of Hydrogen Energy 43.45 (2018): 21114-21132.
Cheng, I-Kai, Chun-Yu Lin, and Fu-Ming Pan. "Gas sensing behavior of ZnO toward H$_{2}$
at temperatures below 300° C and its dependence on humidity and Pt-decoration." Applied
Surface Science 541 (2021): 148551.
Deng, Suzi, et al. "Reduced graphene oxide conjugated Cu2O nanowire mesocrystals for
high-performance NO2 gas sensor." Journal of the American Chemical Society 134.10
(2012): 4905-4917.
Paska, Yair, and Hossam Haick. "Interactive effect of hysteresis and surface chemistry
on gated silicon nanowire gas sensors." ACS applied materials & interfaces 4.5 (2012):
2604-2617.
Xiao, Mengmeng, et al. "Batch fabrication of ultrasensitive carbon nanotube hydrogen
sensors with sub-ppm detection limit." ACS sensors 3.4 (2018): 749-756.
Luo, Yifan, and Chao Zhang. "Pt-activated TiO2-MoS$_{2}$ nanocomposites for H$_{2}$
detection at low temperature." Journal of Alloys and Compounds 747 (2018): 550-557
R.A. Rani, A.S. Zoolfakar, J.Z. Ou, M.R. Field, M. Austin, K. Kalantar0zadeh, Sens.
Actuators B: Chem. 176 (2013) 149.
Schroeder, Vera, et al. "Carbon nanotube chemical sensors." Chemical reviews 119.1
(2018): 599-663.
Meng, Zheng, et al. "Electrically-transduced chemical sensors based on two-dimensional
nanomaterials." Chemical reviews 119.1 (2019): 478-598.
Li, Hua-Min, et al. "Metal-semiconductor barrier modulation for high photoresponse
in transition metal dichalcogenide field effect transistors." Scientific reports 4.1
(2014): 4041.
Baek, Dae-Hyun, and Jongbaeg Kim. "MoS$_{2}$ gas sensor functionalized by Pd for the
detection of hydrogen." Sensors and Actuators B: Chemical 250 (2017): 686-691.
Hao, Lanzhong, et al. "Highly enhanced H 2 sensing performance of few-layer MoS 2/SiO
2/Si heterojunctions by surface decoration of Pd nanoparticles." Nanoscale research
letters 12 (2017): 1-10.
Luo, Yifan, et al. "Hydrogen sensors based on noble metal doped metal-oxide semiconductor:
A review." International Journal of Hydrogen Energy 42.31 (2017): 20386-20397.
DongJun Jang received a B.S. degree in electronic engineering from Gangneung-Wonju
University (GWNU, Korea) in 2022. Since 2022, he has now been working on M.S. course
at GWNU. Currently, he is conducting resistive switching devices (RRAM) and chemiresistive
hydrogen gas sensors research at the Intelligent Semiconductor Device & Circuit Design
Laboratory (ISDL) according to Professor Min-Woo Kwon.
U Jin Cho has been studying in the Department of Electronic Engi-neering at Gangneung-Wonju
National University (GWNU, Korea) in 2019. Currently, he is conducting Nanocomposite
hydrogen sensor and Machine Learning research with Professor Min-Woo Kwon in the Intelligent
Semiconductor Device & Circuit Design Laboratory (ISDL). He is currently attending
school.
Youhyeong Jeon has been studying in the Department of Electronic Engineering at
Gangneung-Wonju National university (GWNU, Korea) in 2019. Since 2022, Hydrogen Sensor
research were conducted with Min-Woo Kwon. He is currently attending school.
TaeYong Lee has been studying in the Department of Electronic Engineering at Gangneung-Wonju
National University (GWNU, Korea) since 2018. He is conducting research on chemiresistive
hydrogen sensors and resistive switching devices (RRAM) with Professor Min-woo Kwon
in the Intelligent Semiconductor Device Circuit Laboratory (ISDL) from 2022 to 2023.
He is currently attending school.
RyangHa Kim received a B.S degree from electronic engineering at Gangneung-Wonju
National Uni-versity (GWNU, Korea) in 2023. She has been working on M.S. course at
GWNU. Presently, Hydrogen sulfide gas sensor fabrication and measurement were researched
with Young-Lae Kim professor.
Younglae Kim received his Ph.D. from Department of Electrical and Computer Engineering
at North-eastern University (Boston, MA, USA) in 2013. After Ph. D., he worked at
Intel Corporation (Hillsboro, OR, USA) as a PTD Engineer from 2013 to 2018. In 2018,
he joined Gangneung-Wonju National University (GWNU, Korea) as a Professor, working
in the Department of Electronic Engineering.
Min-Woo Kwon was born in Inchon, south Korea in 1987. He received B.S. and Ph.
D. degrees in department of Electrical and Computer Engineering from Seoul National
University (SNU) in 2012 and 2019, respectively. From 2019 to 2021, he worked at the
Samsung semiconductor Laboratories, where he contributed to the development of 1x
nm DRAM cell transistor and its characterization. In 2021, he joined Gangneung-Wonju
National University (GWNU, Korea) as an assistant professor in the Department of Electric
Engineering, where he is currently a professor.