Abstract
Although hydrogen has been recognized as a promising material for global energy transportation, its inadvertent leakage or generation into enclosed spaces may pose a risk of an explosion hazard in engineering facilities, including nuclear power plants. To manage the associated risks, it is crucial to identify the characteristics of hydrogen dispersion in air and its stratification behavior. In this study, we conducted an experimental investigation on the distribution of hydrogen in an enclosed vessel under steam-condensing conditions by using helium as a substitute for hydrogen. A series of steady-state and transient tests were carried out in a cylindrical test vessel, in which the interacting effect with steam condensation was simulated by employing a vertical condenser tube to promote the mixing of the steam-air-helium mixture. During transient tests, the impact of the jet momentum created by helium injection into the enclosure and subsequent buoyancy-driven redistribution of helium in the postinjection phase was observed.
1. Introduction
Hydrogen has gained recognition as a promising material for global energy transportation due to its high energy density per unit mass compared to other fuels [1–8]. However, the potential safety hazards associated with highly combustible and explosive nature of hydrogen must be addressed for its safe utilization as an energy carrier [9–15]. Understanding hydrogen behavior is indispensable to control the associated risks and prevent dreadful accidents that had taken place in the past. It is also essential to take proactive measures to ensure the safe handling and utilization of hydrogen in energy systems [16–20].
Especially, the behavior of hydrogen is one of primary safety concerns in nuclear power plants where hydrogen could be generated from oxidation reactions between fuel cladding and coolant during accidents [20, 21]. The situation becomes more complex in a large-scale enclosed space, such as a containment building, as hydrogen mixes with steam and air and forms nonuniform spatial distributions due to its lightweight nature. Furthermore, the steam is expected to condense on the walls of internal structures within the containment, or on heat-exchange pipes just in case that a cooler system is utilized to suppress the pressure increase.
Several experimental studies have investigated characteristics of stratification and mixing of a gas mixture, including hydrogen, in a large enclosure. The TOSQAN facility employed a 7 m3 vessel to simulate steam injection in air or air/helium atmospheres with wall condensation and measured gas velocity and concentration fields [22]. The THAI test focused on gas distribution and behavior of hydrogen and aerosols in a 60 m3 cylindrical vessel [23]. The HM tests of the THAI research program investigated the interaction of a buoyant jet with a stratified light gas layer in the upper space. The PANDA facility, with six cylindrical vessels totaling 460 m3, was used to address the impact of jet/plume on obstruction geometries, thermal radiation, and activation of spray or cooler systems in the HYMERES-2 project [24]. Besides, the SPOT, MISTRA, and HYMIX facilities were involved in the ERCOSAM-SAMARA project to investigate containment thermal hydraulics [25].
To evaluate the risk of combustion resulting from high local hydrogen concentration, it is crucial to predict hydrogen distribution in the presence of interacting processes. Especially, steam condensation on the wall changes local gas density in the vicinity of the surface and drives natural convection, enhancing the mixing of the gas mixture. However, the combined effects of gas entrainment by jet or plume, steam diffusion by wall condensation, and buoyancy forces from density differences make the analysis of hydrogen distribution in an enclosure very complex [25]. Thus, an experimental study simulating highly mixing conditions induced by steam condensation, on either steady-state or transient conditions, and measuring local gas concentrations in an enclosed vessel is needed to gain further insights into stratification and mixing characteristics of gas mixtures including hydrogen.
In this study, we conducted an experimental investigation to simulate the coupled effect of the transport behavior of hydrogen coexisting with air and steam in an enclosed space and the steam condensation that drives the circulation of the gas mixture. We used helium as a substitute for hydrogen and performed a series of steady-state and transient tests at the JERICHO (JNU Experimental Rig for Investigation of Condensation Heat transfer On tube) facility, which includes a test vessel equipped with a condenser tube. The axial distribution of gas temperature, the local mole fraction of each gas species, and the mean heat transfer coefficient over the condenser tube were measured. The characteristics of time-dependent distribution of the light gas were studied under different helium injection rates and durations in the transient tests.
2. Experiments
2.1. Test Facility
Experiments were conducted in the condensation test facility at the Jeju National University, named JERICHO [26, 27], which consists of a primary system to simulate the containment building, a secondary system for cooling the primary system, and a helium measurement system as shown in Figure 1. The primary system comprises a cylindrical stainless steel tank with a diameter of 609 mm and a height of 1950 mm, including upper and lower heads of 267 mm in height. Steam was supplied into the test tank from a steam generator with a maximum power of 120 kW. In cases where a constant system pressure was required, the steam flow rate was regulated using a proportional-integral-derivative (PID) controller for the submerged heaters in the steam generator. The test vessel was equipped with 14 thermocouples positioned at the center between the condenser tube and the wall of the vessel. At the same elevation, two thermocouples were placed at 90-degree intervals in the azimuthal angle. The temperatures from the thermocouples were averaged to determine the bulk temperature of the gas. To account for nonuniform conditions, the average gas temperature was calculated by excluding the temperature measurement from .
Air and helium were injected into the top of the test tank and the location between the inlet of the condenser tube and the wall of the test tank, respectively, as depicted in Figure 2(a). Helium gas (99.995% of purity) was supplied from a cylinder tank. The pressure of the helium was adjusted to 5 bar (absolute) using a pressure regulator. The flow rate of the injected helium was controlled and monitored using a mass flow controller. To achieve a gas temperature that was close to the bulk gas temperature required for transient test conditions, heaters were used to heat the helium supply line that delivered the helium. The injection point for the helium was positioned between two thermocouples, which were located at and spaced at 45-degree intervals. The gas was injected in an upward direction.
(a) Test vessel and helium measurement system
(b) Condenser tube
A condenser tube, 1 m in length and 21.5 mm in diameter, was placed in the center of the stainless tank. The measurement points for the condenser tube are illustrated in Figure 2(b). Local heat fluxes at six elevations were calculated by measuring the temperature difference obtained from two thermocouples located at different depths at each elevation.
The secondary system was designed to supply water into a condenser tube with a controlled temperature and flow rate. The inlet temperature of the cooling water to the condenser tube was controlled using a PID-controlled preheater located downstream of the water storage tank to maintain a designated temperature. Further details can be found in [26, 27].
The helium measurement system of the experimental facility enabled us to determine the composition of the gas mixture. The system comprises three trains of sampling lines and filters to remove moisture, condensation tanks, and gas analyzers. Each train is connected to two sampling lines made of 1/4 stainless steel tubes. We could take simultaneous gas samples from three measurement points. Figure 2(a) shows the locations of the gas sampling lines (SLs) and temperature measurement points near them. Three sampling lines (SL1-SL3) located near the top of the test tank were aligned along the center of the test tank. The other three lines (SL4-SL6) were positioned between the condenser tube and the wall of the test tank.
We calculated the partial pressure of steam in the sampled gas based on the measured gas temperature near the gas sampling lines, assuming that the steam was in a saturated state. We could determine the partial pressure of helium and air by substituting the partial pressure of steam from the total pressure and taking into account the helium concentration among the noncondensable gas measured from the gas analyzers (FTC300, Messkonzept GmbH). The JERICHO facility is equipped with three gas analyzers, which determine the concentration of helium among the noncondensable gas without steam based on the thermal conductivity of the gas mixture in each train. To protect the gas analyzers from steam, which could damage the system, we installed a moisture filter, a 5-meter-long helical coil, and a condensation tank between the sampling line and the gas analyzer at each train. The sampling rate is adjustable with the rotameter embedded in the gas analyzer. Further details about the helium measurement system are provided in a previous study [26].
2.2. Data Reduction
The concentrations of each gas were derived from a data reduction process. The first step involves calculating the partial pressure of steam from the gas temperature by assuming the saturated condition. If the steam is not saturated, this assumption will cause errors in the measurement of gas concentration; in cases where the steam is superheated, the steam partial pressure will be overestimated, thereby reducing the estimated fraction of noncondensable gases in the steam-air-helium mixture. In preliminary tests, we installed two relative humidity probes (HX86A, OMEGA) in the test vessel of the JERICHO facility and confirmed the state of the steam mixed with air and helium under steam-condensing conditions. It was revealed that the relative humidity of the steam in the mixture of air and helium was maintained very close to 100%, indicating that the steam was in the saturated state.
Subsequently, we can obtain the partial pressure of noncondensable gases, including air and helium, as follows: where represents the bulk temperature of the gas mixture, which is defined as the temperature of the main volume of the steam-air-helium mixture away from the condensing surface. In this test, was obtained through measurements at Tg1 to Tg7, 144 mm away from the wall of the condenser tube, as shown in Figure 2(a). As a next step, the partial pressure of helium was calculated using the measured concentration of helium among the noncondensable gas (), measured by the gas analyzer, as
The partial pressure for the air was obtained by subtracting the partial pressure of helium from the partial pressure of noncondensable gas. The mass fraction of noncondensable gas () that has been considered a control variable in previous studies can be calculated in terms of mass fraction using the ideal gas law as where M denotes the molecular weight of each gas species.
The global heat transfer coefficient over the condenser tube was obtained based on the energy balance of the coolant. That is, the condensation heat transfer coefficient was calculated by balancing the heat flow across the tube from the bulk to the wall with the increase in the coolant energy from the inlet to the outlet of the tube, given by where and denote the average temperature of the gas mixture in the bulk and the exterior surface of the condenser tube, respectively. These average temperatures were determined using local temperature data from equidistant measurement points. Specifically, the average bulk temperature was calculated as the arithmetic mean of the temperatures recorded at Tg1, Tg3, Tg4, Tg5, Tg6, and Tg7, as shown in Figure 2(a). The variable A in Eq. (4) represents the surface area of the cylinder tube where the heat transfer through condensation occurs. For the transient case, temperatures of the bulk and the wall were not time-averaged for calculating the heat transfer coefficient.
The measurement errors of the main experimental variables are presented in Table 1. For steady-state tests, the average uncertainty of the condensation heat transfer coefficient was estimated to be 10.6%.
2.3. Experimental Conditions
The experimental conditions of this study are summarized in Table 2. The experimental program consisted of steady-state tests with a total of 39 runs and two series of transient tests. In steady-state tests, the pressure of the test tank and the composition of the steam-air-helium mixture were stabilized by balancing the steam supply rate with the steam condensation rate within the test tank. In transient tests, on the other hand, helium was introduced into the test tank where a mixture of steam and air preexisted through the injection nozzle, creating a vertical buoyant jet. The two series of transient tests can be further divided into transient case A, where PID-controlled heaters in the steam generator were used to maintain the system pressure, and transient case B, where the pressure of the test vessel varied due to constant steam injection. Descriptions of each test condition are illustrated in Figure 3.
(a) Steady-state case
(b) Transient case A (fixed pressure)
(c) Transient case B (fixed steam injection rate)
For each case, steam was first produced in the boiler. Then, we slightly pressurized the test tank and discharged preexisting noncondensable gases by injecting steam and venting the internal gas for several minutes. Next, we maintained the system pressure by controlling the steam generator heaters. We adjusted the initial concentration of each gas at the desired value by injecting air and helium just in case. The wall subcooling degree was also regulated by adjusting the flow rate and inlet temperature of the water flowing in the condenser tube. For the initial conditions of each case, the temperature difference between the inlet and outlet was set to 5 K throughout the cases.
During the steady-state tests, the flow rate of injected steam was controlled by a PID system to maintain the system pressure. The heat transfer coefficient was obtained by measuring the temperatures of fluids and tube walls for 240 s at a frequency of 1 Hz after a steady-state condition was reached. The gas mixture was sampled from each elevation following the temperature measurement. To reduce flow disturbances due to sampling, a minimum of 30 L/h of the gas mixture was sampled from the test tank for each line. Then, we could obtain profiles of partial pressure for each gas component under steady-state conditions.
Despite the fact that the test section of the JERICHO facility was completely covered with insulation material, it is anticipated that some heat loss occurred from the test vessel, resulting in a slightly lower wall temperature compared to the gas temperature. In steady-state tests, the heat loss out of the test vessel can be estimated by calculating the difference between the rate of energy inflow carried by the injected steam and the heat removal rate by the condenser tube. The heat transfer rate of steam condensation is easily calculated by multiplying the heat transfer coefficient with the wall subcooling degree (40 K), and the heat transfer area of the condenser tube. Consequently, the average rate of heat loss from the test section was estimated to be 2.5 kW.
In transient case A, the steam generator operated to sustain the system pressure at 3 bar. That is, the partial pressure of steam was reduced upon the introduction of helium into the test tank. We varied the flow rate to inject the same amount of helium in three separate runs. The experiment was terminated as soon as the helium concentration among noncondensable gases reached 0.5. This case allowed us to analyze the impact of flow rate and helium injection rate on the stratification behavior of helium.
In contrast, in transient case B, the steam injection rate into the test tank remained constant for the entire duration of the test, at the rate needed to establish an initial equilibrium state of steam and air before initiating the transient run. Hence, the pressure of the test tank kept increasing when helium is introduced. Unlike in transient case A, gas sampling was conducted continuously during the experiment, with a flow rate of 12 L/h for each line to mitigate the effect of sampling on gas behavior in the test tank. This methodology allowed for the evaluation of partial pressure transitions amid helium injection.
3. Results and Discussions
3.1. Steady-State Tests
In the steady-state tests, the pressure of the test vessel and the composition of the gas mixture were unchanged by regulating the steam injection rate and thus sustaining heat balance of the test section. The total mass fraction of noncondensable gas ranged from 0.1 to 0.7, and the molar fraction of helium in noncondensable gas varied from 0.0 (i.e., air only) to 0.6 while the subcooling degree of the condenser tube was kept at 5 K. A total of 39 experimental runs were conducted for steady-state tests.
The axial profiles of the gas temperature inside the test vessel are presented in Figure 4. The helium fraction in noncondensable gas was identified as the governing parameter for helium stratification in a steam condensation environment. When the mean helium fraction in noncondensable gas was 0.3 or less, the vertical temperature distribution of the gas mixture was uniform and exhibited little dependence on the total mass fraction of noncondensable gas. However, as increased to 0.4, a temperature gradient of the gas mixture was observed in a limited space near the top of the test tank when the mass fraction of noncondensable gas was low (). That is, the gas temperature decreased in the upper space of the test tank, indicating a higher concentration of helium. A further increase in caused the field with variable gas temperature to expand as a thicker stratified layer of helium developed in the upper space with an increase of the helium concentration.
(a)
(b)
(c)
(d)
The molar fractions of each gas species were estimated based on the temperature distribution of the gas mixture from Figure 4 and the local obtained through gas analyzer measurements. The axial profiles of the molar fraction of steam, air, and helium under gas stratification conditions are plotted separately in Figure 5. It is worth noting that a stratified light gas layer floating on the top of the gas mixture caused steam to be pushed downwards, resulting in a higher steam concentration around the condenser tube.
(a)
(b)
(c)
(d)
(e)
(f)
The reason why the stratification behavior of the air-steam-helium mixture relies predominantly on the helium fraction in noncondensable gas can be explained by the changes in the density of the gas mixture near the condensing surface. Free convection around the condenser tube, which promotes the mixing of the gas mixture, is driven by the density difference between the wall and the bulk. When steam carried from the bulk is condensed on the cold wall, the concentration of the noncondensable gas mixture builds up in the close vicinity of the wall. If the gas mixture with accumulated noncondensable gases near the wall is heavier than the gas mixture in the bulk, then a naturally driven flow is sustained within an enclosure by the density difference between the wall and the bulk. However, a higher helium fraction in the noncondensable gas reduces the gas density near the wall, thereby decreasing the density difference and driving force of mixing. If the helium fraction in noncondensable gas exceeds a threshold value, the gas mixture density at the wall becomes lower than that in the bulk, leading to the stratification of a light gas with limited mixing of the gas mixture.
Figure 6 shows the measured condensation heat transfer coefficient according to the composition of the gas mixture, and Figure 7 depicts a map to represent the generation of a light gas stratification in terms of and , along with the contour plot of heat transfer coefficients. An increase in either or resulted in a reduction of the condensation heat transfer coefficient. The degradation of steam condensation with an increase of the helium concentration among the noncondensable gas was reported and explained in our previous studies [26, 27]. An unexpected escalation of the mean heat transfer coefficient over the condenser tube was observed in certain stratified conditions, as compared to when the steam-air-helium mixture was uniformly mixed. Specifically, at Wnc =0.2, the heat transfer coefficient measured at XHe/Xnc =0.5 was higher than the values obtained at lower helium concentrations, i.e. 0.2 ≤ XHe/Xnc ≤0.4. As shown in Figure 6, steam was transported downwards in the test tank when a stratified layer of helium was formed in the upper space. Given that the condenser tube is positioned between 511 mm and 1511 mm from the top of the test tank, the steam-rich section around the condenser tube could have resulted from the stratification of the light gas.
Inspection of Figure 7 revealed that the magnitude of the condensation heat transfer coefficient or heat flux could not be the sole parameter to distinguish whether the stratified light gas layer would form or not. In addition, the lower limit of , which induces a stratified atmosphere in the steam-air-helium mixture, tended to slightly decrease as the steam concentration in the gas mixture increased.
3.2. Transient Case A: Fixed Pressure
Transient tests were conducted to investigate the transport behavior of helium injected vertically upward through a nozzle into a test vessel filled with a mixture of steam and air. In transient case A, the pressure of the test tank was kept constant at 3 bar through PID control of the steam supply rate into the test vessel. The helium injection rate and duration are shown in Figure 8(a), and the recorded steam injection rate to maintain the pressure of the test tank is presented in Figure 8(b). Three jet velocities were selected to systematically investigate the impact of jet characteristics on the stratification of the light gas inside the enclosed space while keeping the total injection amount constant. Consequently, the injection time varied according to the helium injection rate. The gas temperatures were continuously monitored even after the helium injection was terminated so that the temporal distribution of helium in the presence of the driving force for the convection, exerted only by steam condensation, could be observed.
(a)
(b)
The temporal variation of gas temperatures measured at 10 locations is plotted in Figure 9. During the helium injection phase, the gas temperatures kept decreasing since the partial pressure of steam was reduced due to the inflow of helium at a fixed pressure. Note that there was little difference between the measured gas temperatures at different elevations in this phase. This may be attributed that the jet momentum, combined with the influence of steam condensation, was high enough to induce vigorous circulation of the gas mixture inside the test vessel. The only exception was the temperature measured by the bottommost thermocouple, of Figure 2(a), which varied little during the helium injection phase. Considering that was positioned at the same elevation as the jet exit, this implies that the distribution of the gas mixture due to the helium injection was mainly affected above the jet exit due to the gas entrainment by the jet, whereas helium was almost absent below the level of the jet exit.
(a)
(b)
(c)
As soon as the helium injection was terminated, the gas temperature rose sequentially from the bottom as shown in Figure 9. In the absence of jet momentum, helium migrated upward by buoyancy, while steam was accumulated in the lower section. On the contrary, the gas temperatures measured at the upper head of the test vessel (–) kept decreasing over the entire period of the transient. This indicates that a stratified helium layer was formed at the uppermost space of the enclosure. As time elapsed, a noticeable temperature gradient of the gas mixture was observed especially above the condensing section, and the redistribution of the steam-air-helium mixture inside the test vessel occurred gradually due to the diffusive transport of helium.
Figure 10 presents the axial profile of the gas temperature and the mole fraction of helium at the end the transient tests. With an increase in the jet velocity, the local concentration of helium at the upper head of the test vessel also increased. However, the gas compositions around the condensing section showed little dependence on the helium injection rate and become quite uniform by the end of the postinjection phase. The temporal evolution of the mean heat transfer coefficient over the condenser tube is plotted in Figure 11. The heat transfer coefficient continued to drop during the helium injection phase since the light gas lowered the density difference of the steam-air-helium mixture between the wall and the bulk as reported in [26, 27]. During the postinjection phase, however, the gradual formation of a stratified helium layer increased the steam content around the condenser tube, causing the heat transfer coefficient to rise again and maintain a steady value for the rest of the transient.
(a)
(b)
3.3. Transient Case B: Fixed Steam Injection Rate
In transient case B, steam was supplied at a constant rate which balanced out with the steam condensation rate in the pretest stage when the test vessel was filled with the steam-air mixture. Three runs of transient tests with different helium injection rates and corresponding durations were carried out. Unlike transient case A, the gas mixture inside the test vessel was continuously sampled to measure the time-varying helium concentration among the noncondensable gas and helium mole fraction. Figure 12(a) shows the helium injection rate and duration for the three runs, while Figure 12(b) presents the change in the pressure of the test vessel. In transient case B, the pressure of the test vessel kept increasing owing to not only the introduction of helium to the enclosure but also the degraded rate of steam condensation by the light gas. The tests were terminated as soon as the pressure of the test vessel reached 4 bar as depicted in Figure 12(b).
(a)
(b)
(c)
The temporal variation of gas temperatures and mole fractions of each gas species is shown in Figures 13–15, respectively. Unlike transient case A, the gas temperatures kept increasing gradually at all measurement points during the helium injection phase as the partial pressure of steam rose. This was mainly attributed to the degradation of the steam condensation rate on the condenser tube by the introduction of helium (see Figure 16). Once the helium injection ended, the gas temperature increased sequentially from the bottom of the enclosure. Figures 14 and 15 reveal that the helium mole fraction continuously increased at the uppermost sampling point (SL1 in Figure 2(a), 15 mm from the ceiling of the test vessel) during the postinjection phase in all runs, indicating the formation of a stratified light gas layer at the top. On the other hand, at the lowest sampling point SL3, the steam fraction increased rapidly after the helium injection was terminated and then reached a plateau. Alike transient case A, the buoyancy-driven redistribution of helium after the jet momentum vanished caused steam to descend, and thus, the steam content rose gradually from the bottom.
(a)
(b)
(c)
(a)
(b)
(c)
(a)
(b)
(c)
The change of the condensation heat transfer coefficient averaged over the condenser tube is plotted in Figure 16. The inflow of helium in the injection phase resulted in a continuous reduction in the steam condensation rate. Once the helium injection was completed, however, the heat transfer coefficient increased again rapidly and even exceeded the initial value measured in the absence of helium. This was due to the rise in steam concentration around the condensing section in the postinjection phase. One could observe from the measurement data from SL3 in Figure 14 that the steam mole fraction became larger than the initial value at the late stage. Additionally, the increase in the total pressure of the gas mixture also contributed to an increase of the condensation heat transfer coefficient, as reported in [28].
It was also observed that the gas temperature at the bottommost measurement point () began to rise prior to the termination of helium injection. Since steam was introduced into the lower section of the enclosure and a vigorous circulation of the steam-air-helium mixture was established above the jet exit, steam built up to high concentrations at the lower space of the test vessel, while little helium was present. In Figure 15, the stratification process of helium during the transient tests was clearly observed. It was noted that the peak mole fraction of helium near the ceiling of the enclosure increased with an increase in the helium injection rate at the end time of the transient.
4. Conclusions
In this study, we performed an experimental investigation on the stratification behavior of a light gas coexisting with air and steam in an enclosed cylindrical vessel where the transport of the gas mixture was influenced by steam condensation on a vertical tube. In steady-state tests, the axial distribution of the gas mixture for predetermined compositions was measured. The concentration of helium in the noncondensable () was found to be the most significant factor for helium stratification as it determines the density difference between the condensing wall and the bulk.
In transient tests, we intended to figure out the transport behavior of helium into a test vessel filled with a mixture of steam and air. While the gas compositions showed little variation based on the positions above the gas exit due to the jet momentum exerted during helium injection, the formation of the stratified light gas layer by the buoyancy-driven redistribution of helium and the corresponding descent of steam were observed in the postinjection phase. Even though the helium injection rate influenced the peak helium concentration at the end of the transient, the helium transport behavior turned out to be very similar.
The experimental data obtained from this study can be used to validate a predictive model for light gas stratification in an unventilated space. Furthermore, they are expected to be useful in evaluating the efficacy of a CFD analysis tool in simulating the hydrogen distribution under a steam condensation environment in an enclosed volume, such as the containment building of nuclear power plants.
Nomenclature
| : | Area (m2) |
| : | Specific heat capacity (J kg−1 K−1) |
| : | Heat transfer coefficient (W m-2 K-1) |
| : | Molecular weight (g mol-1) |
| : | Mass flow rate (kg s−1) |
| : | Pressure (Pa) |
| SL: | Sampling line |
| : | Temperature (K) |
| : | Mass fraction |
| : | Mole fraction. |
| : | Air |
| : | Bulk |
| : | Coolant |
| : | Gas |
| : | Helium |
| : | Inlet |
| : | Noncondensable gas |
| : | Outlet |
| : | Steam |
| : | Saturated |
| : | Top |
| : | Wall. |
Data Availability
Data is available on request.
Additional Points
Highlights. (i) Experiments on hydrogen distribution in steady-state and transient conditions were conducted. (ii) The characteristics of gas stratification under steam condensation were studied. (iii) In transient tests, the impact of the buoyant helium jet was investigated.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (no. RS-2022-00144494) and the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Ministry of Trade, Industry & Energy (MOTIE) (no. 20224B10300050).