Novel Gas-Sensitive Material for Monitoring the Status of SF₆ Gas-Insulated Switches: Gese Monolayer

Abstract

Detecting the decomposition components of SF₆ insulating gas is recognized as an effective means to monitor the operating status of the SF₆ insulating switch. In this paper, the adsorption characteristics of a new two-dimensional material GeSe for five SF₆ decomposition gases (SO₂, SOF₂, SO₂F₂, H₂S and HF) are reported by first-principles simulation. Through the analysis of the change of energy band structure, density of states distribution, and gas desorption time, it is found that GeSe has the potential as a gas-sensitive material for the selective detection of SO₂F₂, and the computational work in this paper provides theoretical guidance for the development of new gas-sensitive sensors applied in monitoring SF₆ insulated switches.

1. Introduction

Gas Insulated Switchgear (GIS) has been widely operated around the world, with many advantages such as small ground occupation, high reliability, and long maintenance cycle [1,2,3,4,5]. GIS equipment also has inherent problems [6]; when the external moisture or conductive impurities enter into the sealed cavity or lead to the possible deterioration of the internal insulation of the GIS, there is the occurrence of flashover failure, due to the GIS using a unique fully sealed structure, so the fault location and maintenance is very difficult [7].

For sulfur hexafluoride (SF₆) gas-insulated fully enclosed power distribution devices, it is possible to monitor the operating status of the equipment by detecting the type and concentration of the decomposition products generated by SF₆ gas in a fault, predicting the potential insulation deterioration defects of the equipment and determining the type of fault of the equipment [8,9,10,11]. The key to online monitoring technology based on fault characteristic gas is the accurate detection of gas components, and the current detection method based on semiconductor gas sensors is one of the most effective means. The advantages of semiconductor sensors are a simple structure, low price and fast response, and they are widely used in industrial production, in medical and other industries [12,13].

Conventional semiconductor gas sensors require high operating temperatures, a poor selectivity for multi-component gases, and a poor reliability on the complex and changing environment of the actual industrial production [14,15]. Therefore, new selective gas sensors with a low operating temperature have been the target of scholars’ exploration. Among them, gas-sensitive sensors based on two-dimensional nanomaterials have received much attention in recent years due to their unique properties, such as when Wu et al. reported a novel CsxWO₃/MoS₂ nanocomposite based on solvothermal synthesis and fabricated a sensor based on interdigital electrodes [16]. The results show that the sensor has an excellent selective sensing capability for hydrogen, with a high response at room temperature. Zhang et al. found that cobalt-doped In₂O₃ nanoparticles and MoS₂ nanoflower composites could be used as high-performance carbon monoxide gas sensors with a high sensitivity and fast response detection at room temperature [17]. Two-dimensional materials have great potential in the field of gas sensing, and the purpose of this paper is to explore new two-dimensional materials suitable for the online monitoring of GIS equipment [18,19].

In this paper, the adsorption properties of SO₂, SOF₂, SO₂F₂, H₂S and HF on the GeSe surface are investigated based on the density functional theory. The energy band structure, density of states, electron density distribution and desorption time of the adsorption systems are analyzed to investigate the gas-sensitive performance of the GeSe monolayer on SF₆ decomposition components, so as to provide theoretical guidance for the possible application of the GeSe monolayer in the industry.

2. Materials and Methods

All spin-unrestricted calculations presented in this paper were performed by Dmol3 package, using formal spin as initial. The generalized gradient approximation (GGA) based on Perdew–Burke–Ernzerhof (PBE) is used as the basic generalization to handle the exchange-correlation between electrons. The double numerical polarization (DNP) basis set of functions deals with the linear combination of atomic orbitals (LCAO). The DFT seminuclear pseudopotential (DSSP) is used to solve the problem of transition metals in the system [20,21]. The convergence criterion for the geometry optimization calculation in this paper is set as follows: the geometry optimization is considered as reaching convergence when the energy difference between two geometry optimizations is extremely small (less than 1.0 × 10⁻⁵ Ha, 1 Ha = 27.2114 eV), or the force on each atom is extremely small (less than 0.002 Ha/Å), or the maximum displacement distance of each atom is extremely small (less than 0.005 Å). In addition, the Grimme method was used for dispersion correction in the optimization of each model. To ensure the accuracy of the calculations, exactly 8 × 8 × 1 k points are used in the geometry optimization and electron distribution calculations [22]. For the frontline molecular orbitals in Dmol³, only 1 × 1 × 1 k points are available [1].

As shown in Equation (1), the adsorption energy (E_ads) is defined in order to visually characterize the adsorption properties of the system [23,24].

$$E_{\mathrm{ads}}{=E}_{\mathrm{GeSe}+\mathrm{gas}}{-E}_{\mathrm{gas}}{-E}_{\mathrm{GeSe}}$$

(1)

where E_gas and E_GeSe represent the energy of the gas molecule and GeSe monolayer, and E_GeSe+gas is the total energy of the GeSe adsorption system. Besides, to characterize the adsorption behavior in detail, we define Qt as the charge transfer in the adsorption system, and when Qt is positive, it represents the gain of electrons by the gas molecules during the adsorption process. The distance (d) is used to describe the adsorption distance of the gas molecules.

3. Results

3.1. Geometric Structure and Adsorption Parameter Analysis

First, we established the molecular models of five possible decomposition products of SF₆. During the operation of the GIS equipment, the insulation defects inside the equipment will make the SF₆ molecules decompose into low-fluorine sulfide under the action of high temperature and arc, and characteristic products such as SO₂, SOF₂, SO₂F₂, H₂S and HF will be generated after the action from impurities such as water air inside the equipment. The five SF₆ decomposition products are shown in Figure 1a–e, and their structural parameters are consistent with the results of previous studies [25].

Then, we established the monolayer model of GeSe, as shown in Figure 2a. The 4 × 4 × 1 supercell of the GeSe monolayer contains 18 Ge atoms and 18 Se atoms, and the lattice parameters of the optimized GeSe monolayer are a = 4.26 Å and b = 3.99 Å, which are consistent with the previous studies [26,27,28]. For this model, a vacuum layer of 20 Å was built to shield the surrounding structures from interference [29], and the adsorption models of SO₂, SOF₂, SO₂F₂, H₂S and HF on GeSe are shown in Figure 2b–f.

As shown in Figure 2, the adsorption distances of several gas molecules on the GeSe surface are relatively large except for HF gas, which has an adsorption distance of 2.343 Å. The relevant parameters for all adsorption models are shown in

Table 1. Adsorption parameters of five SF₆ decomposition components on GeSe monolayer.

Adsorption System	Eads (eV)	Qt (e)	d (Å)
SO2	−1.04	−0.200	3.015
SOF2	−0.78	−0.020	3.578
SO2F2	−0.26	−0.019	3.566
H2S	−0.32	0.051	3.510
HF	−0.38	−0.032	2.343

. Excluding the special case of the adsorption distance of HF, it is easy to find that SO₂ seems to have the best adsorption parameters, as the E_ads of SO₂ on the GeSe monolayer surface reaches −1.04 eV, which is the only adsorption system with an absolute value over 0.8 eV, while its charge transfer is −0.2 e, which is much larger than the adsorption systems of other gas molecules. According to the rules of gas adsorption studies, the adsorption behavior can be recognized as chemisorption when the adsorption energy is greater than 0.8 eV. In other words, the GeSe monolayer pair only exhibits chemisorption for SO₂, as judged by the adsorption parameters, and the forces for the other gases are van der Waals-dominated physical adsorption.

Furthermore, it is noteworthy that SO₂, SOF₂, SO₂F₂ and HF all gain electrons during the adsorption process and exhibit oxidant properties, while H₂S acts as a reducing agent transporting a small amount of electrons to the GeSe monolayer, which is consistent with the redox nature of the gas itself. Based on the analysis of the adsorption parameters, GeSe appears to be a potential sensitive agent for the selective detection of SO₂ in SF₆ decomposition products, and we will continue with a further analysis based on the effect of the electron density analysis on the adsorption behavior produced by each system.

3.2. Electron Density Analysis for Adsorption System

The distribution of the deformation charge density (DCD) of the adsorption system of five gas molecules is shown in Figure 3, in which the red area represents the electron depletion region and the blue area represents the electron-rich regions, and the charge transfer in the adsorption process can be qualitatively analyzed by the charge distribution calculation. In Figure 3a–c,e, as mentioned above, the gas molecules in these systems act as oxidizing agents to draw electrons from the GeSe monolayer, which is reflected in the diagrams in that they approach the electron-aggregating region of the GeSe monolayer through atoms in the electron depletion region. In contrast, H₂S gas, which acts as a reducing agent, is connected through the red region, where the S atoms are located in the electron-aggregating region where GeSe is located. In addition, the large area of the red region where the S atom is located, shown in Figure 3a, and the close distance to the blue region in the GeSe monolayer verify that the SO₂ molecule has the largest charge transfer in each system.

To further describe the interaction between gas molecules and GeSe monolayers, we explored the electronic structure of each adsorption system using DOS analysis. Figure 4a,c,e,g,i show the TDOS distributions for the SO₂, SOF₂, SO₂F₂, H₂S and HF adsorption systems, respectively, while PDOS is shown in Figure 4b,d,f,h,j.

Focusing first on the TDOS, for the SO₂F₂ and H₂S systems, it is relatively similar, so that their DOS curves before and after adsorption both differ only slightly in a few small ranges; for example, the density of states of GeSe increases at −4 eV after HF adsorption, and H₂S increases at −6 eV, −4 eV and −2 eV. These new peaks originate from the contribution of gas molecules; however, the density of states of the whole system does not change in any other way than the introduction of gas molecules, and as we can see, the vast majority of the TDOS curves before and after the adsorption of the above two gases shown in Figure 4g,i overlap. For SOF₂ and SO₂F₂ systems, in addition to the increase of gas molecular states, there are some other changes in the state density distribution of the whole system. For example, in the TDOS curve of SOF₂ gas, the peak near 2 eV is slightly shifted to the right, while in the SO₂F₂ adsorption system, the peak near −4.5 eV is shifted. This phenomenon indicates that the adsorption of the two gases on the GeSe monolayers has a weak effect on the electronic structure of the intrinsic material. However, it can be observed that the TDOS curves of the two adsorption systems do not change significantly near the Fermi level, so it can be inferred that the adsorption behavior of the gas does not cause significant changes in the band gap of the material. For the adsorption of SO₂, in addition to the contribution of the state of gas molecules, in contrast to other adsorption systems, the adsorption of SO₂ makes the DOS curve of the whole system move slightly toward the direction of low energy, so that the adsorption of SO₂ molecules on the GeSe monolayer can generate electrical signals. According to the above analysis, we can reach a preliminary conclusion that the GeSe monolayer only has chemical adsorption characteristics for SO₂ and that the adsorption behavior of other characteristic gases is only physical adsorption. Therefore, the GeSe monolayer could selectively detect SO₂ within the environment of a variety of gas components, which also verifies the conclusion of the geometric structure and adsorption parameter analysis.

The PDOS curve can analyze the interaction force between gas molecules and adsorption materials through the hybridization of atomic orbitals; as shown in Figure 4b, the 2p orbitals of S atoms (S-2p) of gas molecules and 4p orbitals of Se atoms of base materials in the SO₂ adsorption system have orbital hybridization in the continuous intervals of −7, 1 eV and −5 to 0 eV. This indicates that the O-Se chemical bond is formed between the gas molecule and GeSe monolayer and that the SO₂ molecule has been firmly captured by the adsorption material. Such a conclusion is also applicable to the adsorption systems of SOF₂ and SO₂F₂. Taking the adsorption system of SO₂F₂ as an example, the curve representing the 2p orbital of the F atom overlaps with the curve of Se-4p in the range of −8 to −7 eV, and F-2p, O-2p, Se-4p and Ge-4p hybridize with each other near 1 eV. This indicates that GeSe exerts a certain chemical force on SO₂F₂, but judging from the overlapping interval and area of the PDOS curve, this effect is not as good as the chemical adsorption of GeSe to SO₂. As for the adsorption of H₂S and HF, it is obvious that there is no significant hybridization between the atomic orbital of the gas molecule and that of the base material in the PDOS curve, which also verifies that the interaction force between GeSe and these two gases is very weak.

3.3. Band Structure Analysis

It is generally believed that there is a relationship between the band gap and conductivity, as shown in Equation (2) [30].

$$\mathsf{\sigma }\propto \frac{\exp \left({-E}_{\mathrm{g}}\right)}{{\mathrm{k}}_{\mathrm{B}}T}$$

(2)

where ${\mathrm{k}}_{\mathrm{B}}$ is the Boltzmann’s constant and $T$ represents the temperature. $E_{\mathrm{g}}$ is the band gap of the semiconductor material. According to this formula, we can find that at a certain temperature, the conductivity of the material is inversely proportional to the energy band; that is to say, the conductivity of the material can be estimated by calculating the band gap. Therefore, we calculated the band gap of the GeSe monolayer and the band gap of five adsorption systems, and further explored the possibility of the material as a gas-sensitive material for detecting SF₆ decomposition components according to the change of the band gap before and after the material adsorption.

The band structure of each system is shown in Figure 5, the band gap values of the five adsorption models are shown in the broken line diagram in Figure 6, and the change in the band gap generated by the corresponding adsorption behavior is shown in the bar diagram in Figure 6. First of all, let us pay attention to the band structure. It is not difficult to find that, with the exception of the band structure of the SO₂ adsorption system shown in Figure 5b, the difference between the band structure of other gases after adsorption and that of the intrinsic material is not great. In particular, the band structure of the SOF₂ gas adsorption system is almost identical to that of the GeSe monolayer. This indicates that the adsorption of SOF₂ molecules does not cause significant changes in the band gap and band structure of the adsorbed material, so one can consider that the conductivity of the GeSe monolayer will not change significantly due to contact with SOF₂ gas molecules. For the absorption of SO₂, it can be seen that in the band structure of the adsorbed system, a new energy level appears near the Fermi level, which acts as the conduction band of the band structure and greatly reduces the band gap of the structure. In other words, after SO₂ is absorbed by the GeSe monolayer, the conductivity of the monolayer will increase significantly, which should be represented as the reduction of resistance of the GeSe sensor from the perspective of macroscopic devices.

The unique selective adsorption properties of the GeSe monolayer on SO₂ can be more clearly demonstrated in Figure 6. The band gap of the intrinsic GeSe monolayer is 1.504 eV, while the band gap decreases sharply to 1.025 eV after SO₂ adsorption. The band gaps of the GeSe monolayer are 1.499, 1.495 and 1.4, respectively, after the adsorption of other SF6 characteristic decomposition components. This shows that the GeSe monolayer is only sensitive to SO₂ molecular selectivity, while the adsorption of other SF₆ decomposition components will not cause the response of the GeSe monolayer.

Besides, based on the frontier molecular orbital theory, the energy band of the system can be calculated according to the energy difference between the lowest unoccupied orbital and the highest occupied molecular orbital. In order to further verify the change in the GeSe monolayer energy band caused by the adsorption behavior of the system, we calculated the energy band in another way by calculating the HOMO and LUMO of each model. However, DMOL³ only supports 1 × 1 × 1, so the calculation results of the frontier molecular orbital are only suitable for a qualitative analysis. As shown in Figure 7, consistent with the conclusion obtained from the energy band structure analysis, we calculate that according to the molecular orbital, the SO₂ adsorption system has the smallest band gap, which is much smaller than for other adsorption systems shown in Figure 7b–e, while the band gap of other adsorption systems is almost unchanged compared with the intrinsic material (1.67 eV).

3.4. Desorption Time Analysis

For gas sensors, desorption time is a factor that must be considered in testing and practical applications. The development of a gas sensor based on new two-dimensional materials can realize desorption at a lower temperature, which is also a major advantage of this kind of gas sensor. Herein, we calculated the desorption time of five gas decomposition components from the GeSe monolayer at 298, 348 and 398 k based on Equation (3) [9]. The calculation results are shown in Figure 8.

$${\mathsf{\tau }=\mathrm{A}}^{-1}{\mathrm{e}}^{-\frac{E_{\mathrm{a}}}{{\mathrm{K}}_{\mathrm{B}}T}}$$

(3)

where A is the attempt frequency (1012 s⁻¹), T is the temperature, K_B is the Boltzmann constant (8.318 × 10⁻³ kJ/(mol·K)), and E_a is assumed to have the same value as Ead, representing the energy consumed for gas desorption. For SOF₂, SO₂F₂, H₂S and HF adsorption systems, because the adsorption energies are too small, according to the equation, these gas molecules will only desorb on the GeSe monolayer in extreme time, which also verifies the weak adsorption performance of the GeSe monolayer to these gases. As for SO₂, the first thing that needs to be emphasized again is that it can be firmly adsorbed on the surface of the GeSe monolayer. Secondly, SO₂ can be desorbed in an acceptable time at a lower temperature. As shown in the figure, the desorption time of SO₂ at 348 k is 1139 s, while SO₂ can be separated from the GeSe monolayer in only 14.63 s at 398 k, which can fully meet the requirements of a reusable gas sensor.

4. Discussion and Conclusions

In this paper, the monolayer model of GeSe and the adsorption model of SO₂, SOF₂, SO₂F₂, H₂S and HF on the monolayer are constructed. Through the analysis of adsorption parameters and electron density, the following conclusions are drawn:

(1)

The GeSe monolayer has a good adsorption effect on SO₂ and is not sensitive to other decomposition components of SF₆, so it can selectively adsorb SO₂ molecules.

(2)

The adsorption behavior of SOF₂, SO₂F₂, H₂S and HF will not significantly change the electronic structure of the GeSe monolayer, while the adsorption of SO₂ will shift the overall density of states of the GeSe monolayer, resulting in a significant reduction in the band gap of the adsorption system and an increase in the conductivity of the material.

(3)

SO₂ can be desorbed from the GeSe monolayer in only 14.63 s at 398 k. The GeSe monolayer is a potential gas sensing material for the selective detection of SO₂ with a rapid recovery speed.