Effect of the Arc Extinguishing and Insulation Properties of C₄F₇N/CO₂ Mixtures with Oxygen Addition: Experimental Investigations and Comparative Analysis

Abstract

The C₄F₇N/CO₂ mixture is considered one of the most promising alternatives to sulfur SF₆. Recent studies have shown that the addition of O₂ to the C₄F₇N/CO₂ mixture can suppress carbon precipitation following electric arc discharges. This paper conducts arc-burning experimental research on SF₆, 10%C₄F₇N/90%CO₂, and 10%C₄F₇N/85%CO₂/5%O₂ mixtures. Measurements were taken of the arc voltage and arc current under a 10 kA breaking current for these three gases. Additionally, the pressure at the nozzle throat during arc and cold flow conditions, as well as the pressure in the storage chamber, were measured. The post-arc current and Rate of Rise of Recovery Voltage (RRRV) for the three gases were calculated. The study also compared the solid precipitates in the gas medium after multiple arc-burning experiments. The results indicate that adding O₂ to the C₄F₇N/CO₂ mixture can increase the pressure at the nozzle throat during the arc phase, and the inclusion of O₂ has minimal impact on the peak arc-extinguishing voltage and critical RRRV. It is hypothesized that a small amount of oxygen has a negligible effect on the thermal recovery properties of the C₄F₇N/CO₂ mixture, while also inhibiting the precipitation of carbon following electric arc discharges. This research could provide a reference for developing and optimizing eco-friendly high-voltage circuit breakers.

1. Introduction

SF₆ gas is commonly used as the insulating medium in high-voltage equipment such as Gas-Insulated Switchgear (GIS) and gas circuit breakers [1,2,3,4]. The Kyoto Protocol, enacted in 1997, designated SF₆ as one of the six gases targeted for global greenhouse gas emissions reduction, making searching for eco-friendly alternative gas-insulating mediums a focal point and hotspot of research among scholars worldwide [5,6,7,8,9]. In recent years, C₄F₇N, newly discovered by 3M, has demonstrated electrical performance comparable to SF₆ while causing minimal atmospheric pollution, and is considered the most promising alternative to SF₆, garnering continuous attention from researchers globally since its inception [10,11,12]. With a boiling point of only −4.7 °C at atmospheric pressure, C₄F₇N typically uses CO₂ or N₂ as a buffer gas for mixing to meet usage requirements [13,14].

Following the introduction of C₄F₇N, many scholars have conducted research on its insulating properties. The results of experiments conducted by the French company Alstom using 145 kV GIS equipment, as reported in [15], indicate that when the volume fraction of C₄F₇N is between 18% and 20%, the breakdown voltage of the C₄F₇N/CO₂ mixtures is comparable to that of SF₆. B.Y. Zhang et al., in [16], investigated the partial discharge characteristics of the C₄F₇N/CO₂ mixtures and SF₆, obtaining and comparatively analyzing the inception voltages of partial discharges for both gases. The results show that at pressures of 80 kPa, 100 kPa, and 120 kPa, the inception voltages of the C₄F₇N/CO₂ mixtures with a 15% volume fraction of C₄F₇N are almost identical to those of SF₆. X. Lin et al., in [17], discovered through dynamic breakdown experiments that compared to pure SF₆, the C₄F₇N mixtures exhibits a significant “low-pressure breakdown” phenomenon, with greater variability in dynamic breakdown voltages. However, considering that C₄F₇N decomposes under the action of high-energy electric arcs, and the decomposed gas components can deposit on electrodes forming a layer of ash, thereby affecting the gas’s performance and posing potential threats to equipment, O₂ is introduced as a second buffer gas. The purpose is to react with the precipitated carbon to regenerate CO₂, thus reducing carbon formation under arc conditions [18]. While there are reports on the insulating properties of the C₄F₇N/CO₂/O₂ mixtures [19,20,21], studies on their arc-extinguishing characteristics are still relatively scarce.

For the reasons outlined above, this paper conducts arc combustion experiments on SF₆, 10%C₄F₇N/90%CO₂, and 10%C₄F₇N/85%CO₂/5%O₂ mixtures. Arc voltages and currents were measured under a breaking current of 10 kA for these gases. Additionally, the pressures in the nozzle throats during both arcing and cold flow states, as well as the pressures in the gas storage chambers, were measured. The study also calculated the post-arc currents and the critical Rate of Rise of Recovery Voltage (RRRV) for these gases (RRRV is the rate at which the voltage across the terminals of a circuit breaker increases after the interruption of the fault current and the extinction of the arc. It is a critical parameter in assessing the performance of a circuit breaker, as it reflects the circuit breaker’s ability to withstand the transient recovery voltage (TRV) without re-ignition of the arc). A comparison of the solid precipitates within the gaseous media after multiple arc combustion experiments was conducted. During these experiments, the storage tank pressure was maintained at 8 bar, while the chamber was pressurized to 4 bar. The findings from this research could guide the design and optimization of eco-friendly gas circuit breakers.

2. Experimental Methods

To measure the parameters of electric arcs in different gaseous media, this study constructed an arc generation experimental platform consisting of three main components: a power supply circuit, an arcing device, and a measurement system. The overall schematic of the platform is shown in Figure 1.

The platform utilizes an LC oscillation circuit as its power source, which includes a charging switch MS, a capacitor Cs, an inductor Ls, a discharging switch Sx, a closing switch MB, and a discharging resistor Rx. Before the start of the experiment, the capacitor is pre-charged to store energy, which is then discharged through the inductor at the onset of the experiment to produce a sinusoidal current. By presetting the parameters of the capacitor and inductor, the charging voltage of the capacitor can be adjusted to control the frequency and peak of the current. The maximum charging voltage of the capacitor in the device is 2 kV, with an oscillation output frequency and current of 50 Hz, 0~100 kA (15 Hz, 0~25 kA).

The arc generation device primarily consists of an experimental tank and a storage tank, with the experimental tank internally equipped with a storage chamber and a nozzle. The test tank has a height of 300 mm and an internal diameter of 248 mm. The gas storage chamber has a height of 200 mm and an internal diameter of 87 mm. The capacity of the gas storage tank is 30 L. The main material is 304 stainless steel. The lower electrode is installed in a fixed position inside the storage chamber. Initially, the upper electrode passes through the nozzle to contact the lower electrode. An arc is ignited by separating the movable upper electrode, driven by a pneumatic actuator, from the fixed lower electrode. Figure 2a and Figure 2b, respectively, display the physical and schematic diagrams of the arc generation tank. The storage chamber is connected to the storage tank via a high-pressure rubber hose, controlled by an electromagnetic valve to regulate the flow. The inflation pressure of the storage tank is approximately twice that of the experimental tank. When the electromagnetic valve is opened, gas from the storage tank rapidly enters the storage chamber. The blockage of the nozzle by the upper electrode quickly increases the pressure within the storage chamber, and after the upper electrode moves away from the nozzle, a blow-out arc is formed due to the rapid pressure change. This experiment employs two methods for measuring dynamic pressure at the storage chamber and nozzle throat: direct measurement and measurement via a pressure transducer tube. Figure 2c and Figure 2d illustrate the connections of pressure sensors at the two measurement locations, respectively.

This study used an upper and lower electrode, a cylindrical pin electrode and a hollow electrode, respectively, with the front end made of Cu20%/W80% copper–tungsten material. The diameter of the upper electrode is 18 mm. The nozzle material is polytetrafluoroethylene (PTFE) doped with 5% boron nitride, with a height of 80 mm, an inlet diameter of 61 mm, an outlet diameter of 56 mm, a throat diameter of 20 mm, and a nozzle throat length of 8 mm. The proportions of the C₄F₇N mixtures are 10%C₄F₇N/90%CO₂ and 10%C₄F₇N/85%CO₂/5%O₂, respectively. The gas storage tanks were pressurized to an absolute pressure of 8 bar, with a tank inflation pressure of 4 bar, and the current used during the arc ignition experiments was 10 kA.

The equipment used for conducting experimental measurements includes the following components:

(1)

Rogowski coil: CWT150 (precision 0.1%, sensitivity 0.2 mV/A, conversion rate 40 kA/µs), utilized for measuring the arc current;

(2)

Differential probe: HT8100 (precision 2%, rise time 3.5 ns) used for measuring arc voltage;

(3)

Displacement sensor: Hermit (precision 0.1 mm) employed for tracking the trajectory of the upper electrode;

(4)

Pressure probe: 8530b–500 (precision 0.5%) for measuring dynamic pressure in the nozzle throat and gas storage chamber.

The specific experimental process is divided into a preparation phase and an experimental phase. During the preparation phase for the mixture arc ignition experiments, the internal gas storage container of the gas recovery system is used to mix the gases. Initially, a smaller amount of C₄F₇N gas is loaded and the gas flow rate is strictly controlled to prevent condensation. The gases are left to stand for 24 h to ensure even mixing, and a gas mixing ratio detector is used to ensure the error is less than 0.5%. During the experimental phase, the experimental tank, gas storage tank, and pipelines are first evacuated to ensure a vacuum level below 50 Pa. Then, the gas from the storage container is transferred into the experimental and storage tanks. After each experiment, some of the gas inside the experimental tank is recovered and the gas inside the storage tank is replenished to maintain a constant pressure difference.

3. Results and Discussion

3.1. Study on Pressure Characteristics

This paper simulates gas blowing by pre-charging the gas storage chamber from the gas storage tank and shortening the distance from the nozzle outlet to the nozzle throat for ease of installation. Consequently, the paper analyzes the pressure variations under cold flow conditions in comparison with the position of the upper electrode to explain the changes observed in the pressure measurements. Figure 3a displays the dynamic pressures measured in the gas storage chamber and nozzle throat under cold flow conditions. Figure 3b illustrates the dynamic stroke of the upper electrode, which helps elucidate the variations in air pressure. Position 1 is the initial position of the upper electrode. At approximately 30 ms, the valve between the gas storage tank and the gas storage chamber opens, allowing gas to flow from the storage tank into the storage chamber, causing a rapid increase in chamber pressure. However, because the upper electrode blocks the nozzle throat while also covering the pressure measurement port of the nozzle throat, there is no significant rise in the pressure at the nozzle throat until about 133 ms. At this point, the upper electrode moves to Position 2, where the nozzle throat’s pressure port is no longer obstructed, leading to a rapid increase in the pressure of the nozzle throat. By 137 ms, or Position 3, the valve between the storage tank and storage chamber closes, and the upper electrode also moves away from the nozzle throat. This removal of the obstruction allows the gas to flow rapidly out of the storage chamber, causing a swift decrease in the pressure in both the storage chamber and the nozzle throat. It is also observed that the pressure in the nozzle throat decreases faster than in the storage chamber and remains below the initial tank pressure for a period, confirming the presence of a strong airflow. Position 4 is the stopping position of the upper electrode.

In high-voltage switchgear, gas (air) purging methods are commonly used to ensure that the arc is effectively extinguished after it has been interrupted. By comparing the flow characteristics of the extinguishing media in the nozzle, this study analyzes the extinguishing capabilities of different gases. During arc extinguishing experiments with SF₆, C₄F₇N/CO₂, and C₄F₇N/CO₂/O₂ mixtures, dynamic pressure changes in the nozzle throat were monitored and the pressure differential between the nozzle throat and the storage chamber after the electrode left the nozzle throat was calculated; the calculation results are shown in Figure 4b.

As shown in Figure 4a, the pressure in the nozzle throat during the arcing stage increased by approximately 0.5 bar compared to the cold flow stage in the C₄F₇N/CO₂ mixtures. In contrast, in the C₄F₇N/CO₂/O₂ mixtures, the pressure in the nozzle throat during the arcing stage increased by approximately 1.1 bar compared to the cold flow stage, which is significantly higher than that in the C₄F₇N/CO₂ mixtures. It is hypothesized that the addition of O₂ enhances the radial heat conduction of the arc, resulting in a larger arc diameter in the C₄F₇N/CO₂/O₂ mixtures. In practical circuit breakers, higher throat pressures in the nozzle imply that more gas can be compressed within the expansion chamber prior to the major nozzle opening, and consequently, a stronger blast intensity can assist in the medium’s recovery after the nozzle opens. Hence, it is beneficial to add O₂ to the C₄F₇N/CO₂ mixtures to enhance the extinguishing performance of the blend. Figure 4b illustrates the changes in the pressure differential between the gas storage chamber and the nozzle throat, with time zero corresponding to the moment when the upper electrode departs from the nozzle throat. At the moment the upper electrode leaves the nozzle throat, the velocity of the airflow is at its fastest, leading to a rapid decline in the pressure within the nozzle throat. Consequently, there is a brief rise in the pressure differential between the gas storage chamber and the nozzle throat, peaking at approximately 9 ms. After a portion of the gas inside the storage chamber is depleted, the pressure differential between the storage chamber and the nozzle throat begins to decrease, eventually stabilizing. By examining the highest pressure differentials and the stabilization time in the three gases, it is evident that the SF₆ gas exhibits a significantly higher peak pressure differential and a slower rate of pressure decline. This suggests that in practical circuit breakers, those utilizing SF₆ gas as the filling medium can maintain a more stable blasting strength during the interrupting process, aiding in arc extinguishing.

3.2. Study on Arc Voltage and Current Characteristics

This paper further compares the arc voltage and arc conductivity of three different gases. It is important to note that during the first half-cycle of the arc when the arc current is high and the arc voltage is low, the arc conductivity often exhibits a very high value. Our primary focus is on the pattern of changes in arc conductivity during the half-cycle before extinguishing. Additionally, during the first half-cycle of the arc, the electrode has not yet left the nozzle throat, lacking the cooling effect of the blast on the arc. Therefore, this study only conducts a calculation and analysis of the arc conductivity during the second half-cycle of the arc. The starting moment of the arc conductivity in Figure 5b corresponds to the zero-crossing point of the first half-cycle of the arc.

As shown in Figure 5a, arcing begins at approximately 2.5 ms in all three gases, with a corresponding current of 7.9 kA. The near-cathode voltage drop at the moment of arcing is approximately 30 V in SF₆ gas and about 20 V in C₄F₇N mixtures. The appearance of the near-cathode voltage drop is due to the small electrode gap at the moment of arcing, while electrons rapidly move towards the anode within the arc gap, forming a positive space charge region near the cathode. As the gap widens, the arc resistance increases, and consequently, the voltage across the arc rises. At the zero-crossing instant of the current, there is no significant difference in the arc voltage among the three gases. After the current zero-crossing, the polarity of the arc voltage reverses, resulting in a typical “saddle-shaped” voltage waveform during the second half-wave of arcing.

During the second half-wave of the arc, the arc voltage in SF₆ gas is significantly lower than that in the mixtures. At 19 ms, as the current crosses zero, the arc extinguishes, and a noticeable extinguishing voltage spike occurs. In SF₆ gas, this extinguishing voltage spike reaches 1.125 kV, while in the C₄F₇N/CO₂/O₂ mixtures and the C₄F₇N/CO₂ mixtures, the spikes are both 0.9 kV. This indicates that the addition of O₂ has a minimal effect on the extinguishing voltage spike, suggesting that a small amount of oxygen has little impact on the thermal recovery characteristics of the C₄F₇N/CO₂ mixtures.

$$g={\displaystyle \frac{i}{u}}$$

(1)

$g$ is the arc conductivity, $i$ is the arc current, and $u$ is the arc voltage.

Figure 5b presents the arc conductivity during the second half-wave of the current for three gases. As shown in Equation (1), the arc conductivity in this paper is calculated by directly dividing the arc voltage by the arc current. From the figure, it is evident that the arc conductivity in SF₆ gas is higher than in the C₄F₇N/CO₂/O₂ mixtures and the C₄F₇N/CO₂ mixtures. Specifically, the peak arc conductivity in SF₆ gas is approximately 60 S, while the maximum arc conductivities in both mixtures are the same, around 42 S. The overall trend of arc conductivity initially increases and then decreases. During the rising phase, there are no significant differences between the two mixtures in terms of arc conductivity. However, during the declining phase, the arc conductivity in the C₄F₇N/CO₂/O₂ mixtures is slightly lower than that in the C₄F₇N/CO₂ mixtures.

3.3. Thermal Interrupting Capability Assessment

The post-arc current is considered a crucial indicator of interrupting capability. Compared to the Cassie arc model, which is applied to the high-current phase, the Mayr arc model can more accurately describe the arc behavior during the current zero-crossing phase. Therefore, this paper uses the Mayr arc model to calculate the post-arc current and the critical TRV for three different gases. As shown in Equation (2), the calculation of post-arc current using the Mayr arc model requires the determination of the arc time constant and the dissipation coefficient. These parameters are calculated using the arc voltage and arc conductivity data from the 50 microseconds preceding current zero-crossing, as shown in Figure 6. The results indicate that the time constant and dissipation coefficient for the SF₆ arc are 3.44 μs and 193.15 kW, respectively. For the C₄F₇N/CO₂ mixtures, these values are 3.54 μs and 143.08 kW, and for the C₄F₇N/CO₂/O₂ mixtures, they are 3.60 μs and 142.75 kW.

$${\displaystyle \frac{d\left(g\right)}{dt}}={\displaystyle \frac{1}{\tau }}\left({\displaystyle \frac{P}{u^2}}-g\right)$$

(2)

$g$ is the arc conductivity, $u$ is the arc voltage, $\tau$ is the time constant, and $P$ is the dissipation coefficient.

Using the calculated time constants and dissipation coefficients, the post-arc currents of the three gases under different RRRV values were computed, with the results shown in Figure 7.

The critical RRRV for SF₆ gas is 1.26 kV/μs, while the critical RRRV for the C₄F₇N/CO₂ mixtures and the C₄F₇N/CO₂/O₂ mixtures are close, at 1.06 kV/μs and 1.04 kV/μs, respectively. This further indicates that the addition of a small amount of oxygen has a minimal impact on the thermal recovery characteristics of the C₄F₇N/CO₂ mixtures.

3.4. Solid Product Analysis

Analysis of the precipitates inside the experimental vessel after arc combustion in different gases showed that the precipitates in the C₄F₇N/CO₂ mixtures appeared as black powder, while those in the C₄F₇N/CO₂/O₂ mixtures were dark grey. Reference [22] utilized energy-dispersive X-ray spectroscopy to study the solid products after disconnecting the C₄F₇N mixed gas and found that the products were primarily composed of C, F, O, and Cu elements. These sediments are prone to floating within the switchgear due to gas blowing and mixing with the gas medium, potentially impairing the gas medium’s arc-extinguishing and insulation performance. Additionally, they can easily adhere to the switchgear’s internal structure, posing a threat to its bonding structure. The lighter color of the sediments in the C₄F₇N/CO₂/O₂ mixed gas compared to the black sediments in the C₄F₇N/CO₂ mixed gas indicates that the addition of oxygen can effectively inhibit the carbon precipitation phenomenon of C₄F₇N and CO₂ under the influence of the arc. The solid deposits inside the experimental tank after the arc-firing tests in the C₄F₇N mixtures are shown in Figure 8.

4. Conclusions

In this paper, the arc parameters in SF₆ gas, C₄F₇N/CO₂/O₂ gas mixtures, and C₄F₇N/CO₂ mixtures under the action of free-flame arc and gas flow and the dynamic pressure changes in the gas storage chamber and nozzle throat were measured, and the experimental results were compared and analyzed, and the following conclusions were obtained:

• The arc in the C₄F₇N/CO₂/O₂ mixtures increases the pressure at the nozzle throat more than in the C₄F₇N/CO₂ mixtures, suggesting that in practical circuit breakers, the arc in the C₄F₇N/CO₂/O₂ mixtures can more effectively block the nozzle throat, enhancing the blasting strength. Additionally, the flow rate of the C₄F₇N mixtures is higher than that of SF₆, which in practical circuit breakers may lead to a rapid drop in the pressure of the interrupter chamber during the interrupting process, potentially failing to maintain sufficient blasting strength to assist in arc extinguishing at the current zero moment;
• Comparing the arc extinguishing voltage peaks and critical RRRV between C₄F₇N/CO₂ and C₄F₇N/CO₂/O₂ mixtures reveals that there is no significant difference in the extinguishing voltage peaks, with the critical RRRV differing by only 0.02 kV/μs. This suggests that the addition of a small amount of oxygen has a minimal impact on the thermal recovery characteristics of the C₄F₇N/CO₂ mixtures;
• After conducting breaking experiments in three different gases, a comparison of the solid residues inside the vessels reveals that the color of the product in the C₄F₇N/CO₂/O₂ mixtures is significantly lighter than that in the C₄F₇N/CO₂ mixture. This indicates that oxygen effectively inhibits the deposition of carbon under the influence of an electric arc in fluorocarbon gases, thereby enhancing their insulation properties. Therefore, using the C₄F₇N/CO₂/O₂ mixtures as the filling medium for circuit breakers appears to be the better choice;
• To provide more comprehensive support for the design and optimization of new environmentally friendly circuit breaker structures and to more thoroughly evaluate the application prospects of the C₄F₇N mixtures, we plan to conduct experiments in our future work to measure the arc temperatures in different C₄F₇N mixtures and perform a more detailed analysis of the composition of solid products after the arc-firing tests.