Open access peer-reviewed chapter

The Performance of Insulation and Arc Interruption of the Environmentally Friendly Gas CF3I

Written By

Dengming Xiao

Submitted: 24 January 2018 Reviewed: 05 July 2018 Published: 05 November 2018

DOI: 10.5772/intechopen.79968

From the Edited Volume

New Trends in High Voltage Engineering

Edited by Reza Shariatinasab

Chapter metrics overview

1,212 Chapter Downloads

View Full Metrics

Abstract

Many researches of trifluoroiodomethane (CF3I) have shown that CF3I has many excellent properties that make it one of the possible alternatives of SF6. This paper reveals the effect laws of CF3I gas content, gap distance, gas pressure, polarity, and electric field nonuniform coefficient on the insulation performance of CF3I gas mixtures. In general, CF3I-N2 gas mixtures present a superior dielectric strength than CF3I-CO2 under different electric field sets. The experimental results indicate that 20 and 30% content CF3I-N2 gas mixtures can achieve nearly 50 and 55% insulation strength of pure SF6. In addition, to evaluate the arc interruption performance of environmentally friendly gas CF3I, we set up a CF3I transient nozzle arc model to study its thermodynamic and transport property. The analysis shows that CF3I gas has a good arc interruption capability, which mainly functions thermodynamic and transport properties approach that of SF6, and some are even better than SF6. The decomposition process is also aggravated by impurities including metal and water. The main by-products are greenhouse gases with GWP below that of SF6 and are lowly toxic and incombustible.

Keywords

  • gas insulation
  • environmentally friendly gases
  • CF3I

1. Comparison of CF3I and its mixtures with SF6 and its mixtures on insulation property

In order to assess the insulation strength of CF3I and its mixtures, power frequency breakdown voltages of SF6 and 20%SF6–80%N2 mixture are measured in the same experimental condition [1]. The result is shown in Tables 1 and 2.

Gas U SF 6 (kV) U 20 % SF 6 80 % N 2 (kV)
d (mm)/P (MPa) 0.3 0.2 0.15 0.1 0.3 0.2 0.15 0.1
5 126.8 82.0 60.7 34.9 82.2 53.7 39.0 25.3
10 222.3 150.2 112.9 82.9 170.5 112.6 82.5 57.7
15 315.0 220.1 174.8 120.0 246.0 166.7 123.9 90.7

Table 1.

Power frequency breakdown voltages of SF6 and 20%SF6–80%N2 gas mixtures in slightly nonuniform electric field.

Gas U SF 6 (kV) U 20 % SF 6 80 % N 2 (kV)
d (mm)/P (MPa) 0.3 0.2 0.15 0.1 0.3 0.2 0.15 0.1
5 44.0 42.6 30.0 21.5 34.4 24.7 18.6 14.8
10 56.4 74.4 64.1 47.9 69.6 53.1 41.6 30.1
15 64.1 95.2 93.4 72.4 91.3 79.0 61.5 44.5
20 72.4 111.0 120.3 97.0 110.1 102.3 82.4 59.4

Table 2.

Power frequency breakdown voltages of SF6 and 20%SF6–80%N2 gas mixtures in highly nonuniform electric field.

Seen from Table 1, when P = 0.1 MPa and d = 10 mm, breakdown voltage of SF6 in slightly nonuniform electric field is 82.9 kV, approximate to the standard value of breakdown voltage in uniform electric field, 89 kV/(mm∙MPa). Moreover, breakdown voltage of 20%SF6–80%N2 mixture is 57.7 kV, roughly 70% of the value of SF6, which is in agreement with data of other researchers.

Figure 1 compares the insulation property of CF3I-N2 gas mixtures [2] with SF6 and SF6-N2 gas mixtures in slightly nonuniform electric field. It is obvious that the breakdown voltage of CF3I-N2 mixture is lower than SF6 and SF6-N2 in the whole pressure range and distance range but higher than N2 in the same condition. Table 3 shows the insulation strength of CF3I-N2 mixture relative to pure SF6 and 20%SF6–80%N2 mixture in different distances and pressures. It can be known from Table 3 that the insulation strength of 20%CF3I–80%N2 in slightly nonuniform electric field is as high as 50% of SF6 and 70% of 20%CF3I–80%N2. If the mixing ratio of CF3I increases to 30%, the insulation strength of CF3I-N2 mixture becomes 55% of SF6 and 78% of 20%SF6–80%N2. In this paper, CF3I-CO2 mixture, whose insulation strength is 97% of CF3I-N2 in the same condition if the mixing ratio is 30%, is researched [3]. With this proportion relationship, insulation property of CF3I-CO2 mixture relative to SF6 and 20%SF6–80%N2 can be calculated [4].

Figure 1.

Comparison of the insulation property of CF3I-N2 mixture with SF6 and SF6-N2 mixture in slightly nonuniform electric field.

P (MPa) 0.3 0.2 0.15 0.1 0.3 0.2 0.15 0.1
d (mm) U 20 % CF 3 I 80 % N 2 / U SF 6 U 20 % CF 3 I 80 % N 2 / U 20 % SF 80 % N 2
5 0.53 0.49 0.49 0.50 0.82 0.75 0.77 0.67
10 0.54 0.52 0.51 0.48 0.70 0.70 0.70 0.69
15 0.57 0.52 0.48 0.49 0.73 0.69 0.68 0.64
d (mm) U 30 % CF 3 I 70 % N 2 / U SF 6 U 30 % CF 3 I 70 % N 2 / U 20 % SF 6 80 % N 2
5 0.58 0.55 0.58 0.55 0.89 0.84 0.85 0.76
10 0.57 0.57 0.57 0.54 0.75 0.76 0.77 0.77
15 0.61 0.57 0.55 0.55 0.78 0.77 0.78 0.72

Table 3.

Insulation strength of CF3I mixture relative to SF6 and 20%SF6–80%N2 in slightly nonuniform electric field.

About the situation in highly nonuniform electric field, breakdown voltage of CF3I-N2 mixture with the distances of 10 mm and 20 mm as examples is compared with SF6 and 20%SF6–80%N2, as shown in Figure 2. Because of the strong electronegativity of SF6 and CF3I, the mixtures present drastic changing breakdown property. It is useless to define the relative insulation strength in highly nonuniform electric field generally, and analysis according to specific pressure condition is necessary [5].

Figure 2.

Comparison of the insulation property of CF3I-N2 gas mixtures with SF6 and SF6-N2 mixture in highly nonuniform electric field.

Seen from Figure 2, power frequency breakdown voltage of SF6 has obvious “hump effect” as a function of pressure with the distance of 10 and 20 mm. Similarly, the increase of breakdown voltage of 20%-mixed SF6-N2 mixture with pressure also tends to saturate. It is discovered by comparison that 30%CF3I–70%N2 possesses comparative insulation property to 20%SF6–80%N2, which is around 60% of SF6. When P = 0.2 MPa, the insulation property of 30%CF3I–70%N2 mixture is as high as 80% of 20%SF6–80%N2 mixture and more than 55% of pure SF6 [6]. When the pressure increases to 0.3 MPa, breakdown voltage of SF6 declines rapidly because of the “hump effect.” In this way, the breakdown voltage of 30%CF3I–70%N2 gas mixtures even exceeds pure SF6, becoming 1.17 times higher than the latter. Because the power frequency breakdown voltage of 20%SF6–80%N2 always increases with the pressure, insulation strength of 30%CF3I–70%N2 is only around 75% of 20%SF6–80%N2 [7]. It can be indicated that the insulation strength of CF3I-N2 gas mixtures relative to SF6-N2 declines gradually with the increase of pressure, but the breakdown level is always more than 55% of SF6. It is partly because the “hump effect” of SF6 in the pressure is 0.1–0.3 MPa and also because interfusing buffer gas N2 improves the abnormal breakdown phenomenon in highly nonuniform electric field [8]. Table 4 shows the insulation strength of CF3I gas mixtures relative to SF6 and 20%SF6–80%N2 in the same condition. It can be seen that the insulation property of 20%CF3I–80%N2 in highly nonuniform electric field is more than 49% of SF6 and 69% of 20%SF6–80%N2 in the same condition.

d (mm) 10 20 10 20
P (MPa) U 30 % CF 3 I 70 % N 2 / U SF 6 U 30 % CF 3 I 70 % N 2 / U 20 % SF 6 80 % N 2
0.10 0.61 0.58 0.97 0.95
0.15 0.59 0.62 0.91 0.91
0.20 0.55 0.73 0.77 0.79
0.30 0.80 1.17 0.75 0.77
P (MPa) U 20 % CF 3 I 80 % N 2 / U SF 6 U 20 % CF 3 I 80 % N 2 / U 20 % SF 6 80 % N 2
0.10 0.56 0.51 0.90 0.84
0.15 0.51 0.54 0.79 0.78
0.20 0.49 0.64 0.68 0.69
0.30 0.79 1.15 0.69 0.75
P (MPa) U 30 % CF 3 I 70 % CO 2 / U SF 6 U 30 % CF 3 I 70 % CO 2 / U 20 % SF 6 80 % N 2
0.10 0.64 0.62 1.02 1.02
0.15 0.62 0.59 1.05 0.86
0.20 0.55 0.42 0.77 0.46
0.30 0.67 0.59 0.55 0.39

Table 4.

Insulation strength of CF3I gas mixtures relative to SF6 and 20%SF6–80%N2 gas mixtures in highly nonuniform electric field.

CF3I-CO2 gas mixtures, also affected by “hump effect,” present obvious decline in high pressure. Under the pressure 0.1 and 0.15 MPa, the breakdown voltage of 30%-mixed CF3I-CO2 gas mixtures is 60% of pure SF6 and 1.05 times of 20%SF6–80%N2 gas mixtures [9]. However when the pressure rises to 0.3 MPa, the insulation property of CF3I-CO2 gas mixtures is only around 40% of 20%SF6–80%N2 mixture [10]. Relative insulation strength of different pressures is shown in Table 4 in detail.

The insulation properties of CF3I-N2 and CF3I-CO2 gas mixtures can be concluded with the analysis above.

In uniform electric field, the insulation strength of 30%-mixed CF3I-N2 gas mixtures can approach 72% of pure SF6 theoretically and in slightly nonuniform electric field more than 55% of pure SF6 [11]. In highly nonuniform electric field, the relative insulation strength of CF3I-N2 gas mixtures is influenced by the pressure seriously, which ranges from 55 to 117% in the researched pressure range [12].

The insulation strength of 30%-mixed CF3I-CO2 gas mixtures can approach 68% of pure SF6 theoretically in uniform electric field, and in slightly nonuniform electric field, it can be more than 53% [13]. In highly nonuniform electric field, the relative insulation strength is also influenced by the pressure. The insulation strength of 30%-mixed CF3I-CO2 gas mixtures is 0.42–0.62 times of pure SF6.

Detailed comparative data and the insulation strength of 20%SF6–80%N2 gas mixtures are listed in Table 5. Seen from the table, whether in slightly nonuniform electric field or highly nonuniform electric field, the insulation strength of CF3I-N2 is better than CF3I-CO2 in the same condition. Therefore, in actual equipment application, CF3I-N2 gas mixtures with the mixing ratio of 20–30% should be considered preferentially.

Gas Electric field U CF 3 I X / U SF 6 U CF 3 I X / U SF 6 N 2
Uniform 0.72 0.88
30%CF3I–70%N2 Slightly nonuniform 0.55 0.78
Highly nonuniform 0.55–1.17 0.75–0.97
Uniform 0.68 0.83
30%CF3I–70%CO2 Slightly nonuniform 0.53 0.75
Highly nonuniform 0.42–0.67 0.39–1.02

Table 5.

Insulation strength of CF3I gas mixtures relative to SF6 and 20%SF6–80%N2 gas mixtures.

Then, principles of applying CF3I mixture in actual electrical equipment are discussed with the example of 40.5 kV cubicle-type gas-insulated switchgear (C-GIS).

C-GIS, vulgarly named “gas-filled cabinet,” is suitable for situations that have high requirement on reliability and limited space and floor area such as rail transit, high buildings, and industrial enterprises. 40.5 kV C-GIS is the most popularly employed. Known from the survey of products on the market, the most popular practice currently is employing SF6 with the pressure 0.12–0.15 MPa (abs) as the main insulation medium of primary loop. For example, the SF6 charge pressures of XGN46–40.5 model C-GIS developed by Xi’an High Voltage Apparatus Research Institute and ZX2 model C-GIS produced by ABB are, respectively, 0.15 and 0.13 MPa. Meanwhile, some companies employ higher pressure, e.g., 8DA10 model C-GIS produced by Siemens, whose SF6 charge pressure is 0.26 MPa. C-GIS is complicated on structure and different on shape, containing function units such as breakers, disconnectors, current/voltage transformers, etc., and forming both slightly nonuniform electric field and highly nonuniform electric field. In accordance to relative insulation strength shown in Table 5, replacing SF6 in C-GIS with CF3I-N2 with the mixing ratio of 20–30% as insulation medium without changing the gap distance and pressure, insulation strength is not strong enough, so it is necessary to adjust the charge pressure or structure properly.

Through analysis, we can know that increasing pressure or increasing gap distance can effectively improve the insulation level of CF3I gas mixtures. If the gap distance is kept constant, the CF3I-N2 gas mixtures of 20–30% mixing ratio at 0.25–0.3 MPa will reach the insulation level of SF6 at 0.15 MPa. In this case, there is no need to change the existing C-GIS internal structure; only a thickened metal enclosure is needed to ensure the operation pressure and the leakage level meet the requirements. At the same time, the liquefaction temperature of CF3I-N2 gas mixtures around this pressure can be maintained below −25°C, which is in accordance with the requirements of GB/T 11022-2011 on the operating temperature of the switch equipment. If the operating pressure is kept constant, the data in Figure 1 and Table 1 show that to achieve the same insulation strength as SF6, the gap distance of 30%CF3I–70%N2 gas mixtures needs to be increased by about twice as much.

Therefore, although the insulation strength of the CF3I gas mixtures is weaker than pure SF6, CF3I gas mixtures can not only achieve the same insulation level as the SF6, but also the liquefaction temperature can meet the operation requirement of the switch equipment by adjusting the operating condition or structure parameters properly. More importantly, CF3I-N2 gas mixtures will not affect global warming. It can solve the unfriendly environmental problems of SF6 and realize the green upgrading of electrical equipment. Therefore, it is suggested that CF3I-N2 gas mixtures with the mixing ratio of 20–30% can be used as alternative medium for SF6 in low- and medium-voltage electrical equipment.

Based on the results of power frequency and lightning impulse tests, it is shown that the insulation strength of CF3I-N2 gas mixtures increases with the increase of CF3I mixing ratio in slightly nonuniform electric field or in highly nonuniform electric field. For CF3I-CO2 gas mixtures, the breakdown voltage varies with the mixing ratio in slightly nonuniform electric field, similar to CF3I-N2. However, in highly nonuniform electric field, the breakdown voltage of 10%-mixed CF3I-CO2 gas mixtures is higher than mixtures with mixing ratio of 20 and 30%, due to the abnormal breakdown of the high pressure. It can be seen that higher mixing ratio of electronegative gas in the CF3I gas mixtures does not always cause better insulation performance, and it will also be influenced by many factors such as the gap distance, the pressure, and the unevenness of the electric field.

From the change of the breakdown voltage with the gap distance, the breakdown voltage of CF3I-N2 and CF3I-CO2 gas mixtures increases linearly with the increase of the gap, and the increase speed is proportional to the pressure in slightly nonuniform electric field. In highly nonuniform electric field, the breakdown voltage of CF3I-N2 gas mixtures shows a tendency to saturate with the gap distance, while the CF3I-CO2 combination has obvious nonlinear characteristics.

From the change of the breakdown voltage with the pressure, the insulation strength of the CF3I-N2 and CF3I-CO2 gas mixtures in the slightly nonuniform field increases linearly with the increase of the pressure and even shows a certain degree of negative synergy. However, under the highly nonuniform electric field environment, there is a clear “hump” in the power frequency breakdown voltage of CF3I-CO2 gas mixtures with the change of pressure, and the positive lightning impulse coefficient in the “hump” section is always less than 1.

From the effect of polarity on discharge, the breakdown voltage of CF3I gas mixtures under positive polarity is higher than that of negative polarity in slightly nonuniform electric field. But in highly nonuniform electric field, the opposite is true. In addition, the uniformity of electric field will also affect the insulation performance of CF3I gas mixtures. Therefore, in the actual product application, the uniformity of the electric field should be improved as much as possible for the purpose to ensure that the insulation strength of the CF3I gas mixtures can be maintained at a high level.

Finally, by comparing with the SF6 and 20%SF6–80%N2, it is found that the 20%CF3I–80%N2 gas mixtures can reach the insulation level of 50% of pure SF6 gas and about 65% of 20%SF6–80%N2 gas mixtures. When the mixing ratio of CF3I is increased to 30%, insulation strength of CF3I-N2 gas mixtures can reach about 55% and 75% of SF6 and 20%SF6–80%N2 gas mixtures, respectively. For the combination of CF3I-CO2, 30%-mixed CF3I-CO2 gas mixtures can reach the insulation level of more than 53% of pure SF6 in slightly nonuniform electric field, but in highly nonuniform electric field, relative insulation strength depends on the pressure and can only reach 42–67% of the latter.

Advertisement

2. The radial temperature distribution characteristics and leading energy transport process of CF3I nozzle arc

The transient temperature distribution of arc is decided by the mutual equilibrium among the energy input of arc, different energy transport processes, and changing rate of energy storage, which determines the characteristics of arc in turn. We analyze the arc extinction characteristics of CF3I through the thermodynamic characteristics of CF3I arc.

The temperature of arc plasma changes with the changes of time and position, and that’s decided by the interactions of the produce of the joule heat of arc and different energy transport processes in itself. The following are the calculation and analysis of the energy balance of the core area (in the radius with 83.3% highest temperature) and conducting regions (in the radius with 4000K temperature) of arc in the conditions with 900 A large current, 50 A small current, and current-zero period.

By the radial temperature analysis of arc, during the process that the head of the moving contact moves from the nozzle upstream to the throat, because the space of the throat is small, and the moving contact and thermal plasma of arc will block the flow of the fluid to the catchment area through the nozzle, and the slow speed of the fluid causes a bigger radius of arc so that the core area is not obvious and the temperature is relatively lower. We choose the time of 900 A current to analyze in Table 6. In the time of large current, the sinusoidal current waveform changes relatively slowly with the change of time, and the changing rate of energy storage of the core area of arc can almost be negligible, and as the slow decline of arc, global energy of arc also goes down, so the change of energy storage in the radius of arc reaches more than 20%. As the interior of the thicker arc cannot be influenced by the cold fluid, the electric power produced in the core area of arc is lost more than 90% by the radiation process, and beyond the core area of arc, the interactions of arc and fluid are more drastic, and the energy loss of radiation lowers obviously. As a whole, the axial convection process of arc does positive work, and the core area does less work. Although the radial convection of arc is small, it takes the joule heat away, and that’s because the cold fluid in upstream moves from the entrance to the lower right and contacts with the thermal plasma with a bigger oblique angle to form convection, especially the cold fluid which flows along the slope of fixed contact even forms eddy in the interior of arc. Therefore, the radial convection in the arc boundary has a good cooling effect, especially in the reabsorption area of the radiation energy.

Arc boundary Electric power input (103W) Radiation loss (%) Radial thermal conductance (%) Axial thermal convection (%) Radial thermal convection (%) Energy loss of pressure to do work (%) Changing rate of energy storage of electric power (%)
Boundary of core area (R833) 4.185 −93.9 −14.0 13.6 −11.2 −0.1 2.4
Arc boundary (R4k) 4.768 −25.8 −52.6 25.8 −68.7 0.2 21.9

Table 6.

Percentage of electrical power input associated with various energy transport processes for the whole CF3I arc length at core and arc boundary at 900 A.

When the current lowers to 50 A, the moving contact moves to the catchment of nozzle, and the distance of contact reaches more than 18 mm. The throat of the nozzle opens completely, and the thermal characteristics of the nozzle arc are decided by the energy distribution and transport process in Table 7. Differing from large current, because of the temperature distribution characteristics of arc and the effect of high-speed cold fluid, the loss ratio of radiation energy decreases sharply. After the reabsorption process of radiation in the arc boundary, the loss ratio of radiation energy approaches zero. Around arc, the high-speed movement of the cold fluid in the nozzle makes arc form an obvious high-temperature core area. The fast change of radial temperature makes the radial thermal conduction become strong, and the energy transport ratio in the core area and arc area is both 70%. In comparison, the convection effect works little to the change of energy. Because the moving contact pulls to the right and arc plasma strengths in axial, the axial convection does positive works in the core area of arc, but in the arc boundary, the high-speed cold fluid around arc effectively takes the energy away, and it has an obvious cooling effect to arc. The radial convection in the core area begins to do negative works, and that’s because the temperature gradient in the core area of arc is very big, and the gas with relatively lower temperature in the radiation reabsorption region begins to enter into the core area to maintain the conservation of mass in the interior of arc; however, out of the core area of arc, the decline of the radial temperature becomes slow, and the radial convection effect still does positive works.

Arc boundary Electric power input (103W) Radiation loss (%) Radial thermal conductance (%) Axial thermal convection (%) Radial thermal convection (%) Energy loss of pressure to do work (%) Changing rate of energy storage of electric power (%)
Boundary of core area (R833) 4.179 −46.6 −69.4 17.8 −6.2 −10.5 11.7
Arc boundary (R4k) 5.666 −0.1 −79.4 −25.7 21.0 −16.2 4.7

Table 7.

Percentage of electrical power input associated with various energy transport processes for the whole CF3I arc length at core and arc boundary at 50 A.

In the current-zero area, shown in Table 8, the joule heat of arc is zero, and the temperature of plasma lowers to less than 10,000 K, and the radiation effect is almost 0. The radius of arc is very small, and the thermal conduction is relatively weaker. Convection mainly consumes arc’s energy, whether in the core area or within the arc boundary, and the radial thermal convection takes more than 30% energy away as the axial reaches about one half of the change of arc’s energy.

Arc boundary Electric power input (103W) Radiation loss (%) Radial thermal conductance (%) Axial thermal convection (%) Radial thermal convection (%) Energy loss of pressure to do work (%) Changing rate of energy storage of electric power (103W)
Boundary of core area (R833) 0 −0.1 −13.3 −35.8 −50.4 −2.4 1.796
Arc boundary (R4k) 0 −0.0 −19.0 −37.4 −48.7 −1.8 1.841

Table 8.

Percentage of electrical power input associated with various energy transport processes for the whole CF3I arc length at core and arc boundary at current zero.

Note: Because the current input in the current-zero area is zero, the change of arc’s energy storage is the changing power of energy, and it is regarded as the energy input to measure the strength of every energy transport process.

Advertisement

3. The arc’s characteristics near the arc-zero area

The thermal interruption of nozzle arc is decided by the arc’s temperature around the current-zero area. Through the discussion of the above section, when it’s close to the current-zero area (25 A and lower), the arc’s temperature and radius decrease rapidly and become a key time of arc interruption, and this process is worth being researched more. When the current is very small, a strong radial thermal conduction makes arc form a relatively bigger radial temperature gradient in the radiation reabsorption area, which takes the energy in the arc’s interior away effectively. And this promotes more entrance of cold fluid of outside into the high-temperature plasma of arc (to maintain the conversation of mass), so that the radial convection heat dissipation becomes the main form of energy transport when the current approaches zero. The superposition of two effects explains the sharp declines of the arc’s temperature and radius in dozens of microseconds before current zero-crossing in Figures 3 and 4.

Figure 3.

Arc temperature varied with time at nozzle throat near current zero.

Figure 4.

Arc radius varied with time at nozzle throat near current zero.

The changes of nozzle throat arc’s temperature and radius with time near current-zero area are shown in Figures 3 and 4. They show that in 30 s before arc-zero area, CF3I arc’s temperature shows a trend of faster decrease without obvious phenomenon that the declines of temperature about time decrease. And the arc’s radius decreases with time. From the characteristic time of arc (see Table 9), at 10 μs before current zero-crossing (arc current is about 5 A), the characteristic time about arc’s radius and temperature is 24.75 and 40.64 μs which are equivalent to SF6 arc’s values in the same time. In the last 10 μs, the characteristic time of CF3I arc in the current-zero area lowers to about 3 μs which is much smaller than air and CO2 arc and approaches SF6 arc. The comparison shows that the declines of CF3I arc’s temperature and radius near current-zero area are equivalent to SF6 and much bigger than air and CO2 arc.

About arc’s radius About arc’s temperature
10 μs before current zero-crossing 24.75 μs 40.64 μs
Current-zero area 3.90 μs 2.27 μs

Table 9.

Characteristic time of CF3I arc near current zero.

Advertisement

4. Analysis of the by-product after CF3I interrupts arc

There are three reasons to cause gas decomposition, and they are decomposition caused by electron collision, thermal decomposition, and photodecomposition. There are mainly first two types in the high-voltage electrical equipment. The main discharge forms in switching arc or equipment such as GIS are high-power arc discharge, spark discharge, and partial discharge. Among them, high-power discharge has large current and a long time of duration, and energy can accumulate in a short time, and temperature can reach more than 20,000 K, and it’s the most serious discharge form of decomposition reaction.

Although physical parameters of CF3I show good thermal interruption characteristics, the same with SF6, it will have decomposition reaction in the conditions of high temperature and high gas pressure. Though most decomposition reactions are reversible reaction, after arc interrupt, there are still some by-products to be produced. These by-products will influence subsequent quenching of arc, and their toxicity and environmental characteristics will influence the application of CF3I in switching arc equipment. This section does spectrum analysis to the arc-later gases, combining the calculating results of the thermomotive equilibrium state gas composition, and evaluates the types of arc-later by-products of CF3I and the possible influence from the impurities.

4.1. Spectrum analysis to the arc-later compositions of CF3I

The arcing experiment is done in the SF6 switch cabinet with nominal voltage of 12 kV and nominal current of 630 A, and we use the sine-input arc current of effective value of 400 and 630 A. We conduct sampling of the arc-later gas of CF3I after five times’ arcing experiment. The experimental gas is CF3I with purity of more than 99.5% which has known impurities of CO2 and CF3Br and water with volume ratio lower than 18 ppm (1.8 × 10−3%). Before the experiment, we have done the insulation test and moisture content test to the switch cabinet to ensure that it can reach experimental standard.

Gas detection uses PLOT-Q chromatogram column to separate and detect the gas, and the detection is done 48 hours after sampling of gas. We use the cleansed needle tubing for sampling and put into the detection equipment for automatic detection, gas separation, and spectrum distinction. Spectrometer has a distinguishing uniformity of more than 80% to the gas composition, and we can trust the judgment of gas composition. As needle tubing sampling will inevitably mix with air, being a mass of N2 and trace amounts of CO2, we will find N2 and CO2 in the detection. Every gas sample is about 10 μL.

Arc-later gas composition gotten by spectral testing is shown in Table 10. In the experiment, 400 A current successfully interrupts for five times, but 630 A current restrikes in the first interruption experiment and continues to burn for dozens of milliseconds. Therefore, arcing experiment under 630 A current is just done for one time, and we obtain the gas after unsuccessful interruption as testing sample. In the results, except the main composition CF3I, other impurities or gas decompositions are N2, CF4, CO2, C2F6, CHF3, H2O, and C4F8. N2 and H2O are impurities because of mixture of air. CO2, CHF3, and C4F8 exist in the original gas, so they are the impurities in the producing process of gas. Therefore, CF4 and C2F6 are the products of the high-voltage and large-current arcing experiment of CF3I. The comparisons of the response values of CF3I, CF4, C2F6, C4F8, and CHF3 in original gas and after different times’ arcing experiment are shown in Figure 5.

Original gas 400 A five times 630 A one time
N2
CF4
CO2
C2F6
CHF3
H2O
C4F8
CF3I

Table 10.

Gas composition after arc for CF3I arc obtained through mass spectrum measurement.

From the spectrum data of CF3I, CF3I decomposes little after five times’ successful interruption of 400 A arc, and CF3I is 97.0% of original gas; the decomposition can be ignored. But after the unsuccessful interruption of 630 A current, CF3I is only 3.4% of original gas. Other gases’ spectrum data shows that under conditions of successful interruption, only 3% CF3I produces CF4 and C2F6 and little C4F8 and CHF3 as the gas reaches room temperature from arc’s high temperature. According to conservation of elements, it will separate out a little iodine. When the arcing interruption fails, more than 95% CF3I becomes CF4 and separates out iodine at the same time. This shows that the ratio of by-product after successful arcing interruption of CF3I is very low and the main by-products are CF4 and C2F6. However, when arc interruption fails and arc continues to burn, a mass of CF3I will decompose and produce much CF4 so that it cannot be used as arc-quenching medium. By examining the switch cabinet after the failure of arc quenching, we can find that there are much black solid attached on the surface of insulator and contact and this solid is carbon and iodine. The production of elementary substance shows that during the drastic burning of arc-restrike gas, many organic compounds decompose into elementary substance due to the lack of oxidizer (such as oxygen). Therefore, when CF3I is applied in interrupting arc, we should reserve enough margin for interrupting current, and at the same time, we can use some absorbents or make air-blast-arc method to eliminate produced iodine.

4.2. Characteristic analysis of CF3I decomposing products

Based on environmentally friendly purpose and toxicity, we analyze the composition of arc-later gases, and the results are listed in Table 11. It shows that compared with SF6, the global warming potential (GWP) of decomposing gases of CF3I under conditions of large current decreases at different levels. Except CF4, the existence time in the atmosphere of other gases drastically reduces. As for toxicity, the arc-later decompositions of CF3I belong to perfluorocarbon and partial fluorocarbon, which both have low toxicity, so it will cause headache, nausea, or dizziness under conditions of long-time or high-density inhalation. While doing experiment, we should note ventilation and appropriate self-protection.

Gas Dielectric strength compared with SF6 Liquefaction temperature (°C) GWP Existence time in the atmosphere (year) Toxicity
C2F6 0.78–0.79 −78 9200 10,000 Low toxicity
CF4 0.39 −186.8 6500 50,000 Low toxicity
CHF3 0.18 −78.2 11,700 264 Low toxicity
C2HF5 0.59 −48.5 14,800 32.6 Low toxicity

Table 11.

Environmental and toxic analysis of CF3I gas composition after arc.

In a word, after CF3I arc discharges under conditions of high voltage and large current, main by-products are perfluorocarbons, CF4 and C2F6; in the event of mixing trace amounts of water, it will produce a little hydrofluorocarbons such as CHF3 and C2HF5. Perfluorocarbon and hydrofluorocarbon with few carbon atoms both have low toxicity; the consequences of inhalation are connected with the density. Only in closed environment, when human inhales higher-density gas, it will damage the cardiovascular system to a certain degree. During experiment and use, we should note ventilation and appropriate self-protection to ensure a safe process.

Figure 5.

Mass spectrum data of CF3I gas composition after arcs: (a) CF3I, (b) CF4, (c) C2F6, (d) C4F8, and (e) CHF3.

Advertisement

5. Conclusions

  1. After adjustment to operation conditions or structure size of the electrical equipment, CF3I gas mixtures can reach equivalent insulation level of SF6, and at the same time, the liquefaction temperature can satisfy operation conditions of switch equipment. Because CF3I-N2 gas mixtures will not influence global warming, it can resolve environmentally unfriendly problems and realize the green upgrade of the electrical equipment. Therefore, we suggest that 20–30% CF3I-N2 gas mixtures can be applied as SF6 alternatives in medium- and low-voltage electrical apparatus.

  2. CF3I has a good arc interruption characteristic, and before current zero-crossing, it approaches SF6, and some thermodynamic properties are even better than SF6. CF3I decomposes easily after large-current arcs, high-temperature decomposing products are hard to recombine after arc extinction, and they are easy to be influenced by impurities such as water and so on and produce a little hydrofluorocarbon. Few-carbon-atom perfluorocarbon and hydrofluorocarbon both have low toxicity, so we should take appropriate actions such as absorbent or gas mixtures to eliminate or restrain decomposing products.

Advertisement

Acknowledgments

This work is supported by the National Natural Science Foundation of China (Grant No.51337006).

References

  1. 1. Taki M, Maekawa D, Odaka H, Mizoguchi H, Yanabu S. Interruption capability of CF3I gas as a substitution candidate for SF6 gas. IEEE Transactions on Dielectrics and Electrical Insulation. 2007;14:341-346
  2. 2. Cressault Y, Connord V, Hingana H, Teulet P, Gleizes A. Transport properties of CF3I thermal plasmas mixed with CO2, air or N2 as an alternative to SF6 plasmas in high-voltage circuit breakers. Journal of Physics D: Applied Physics. 2011;44:495202
  3. 3. Wang W, Rong M, Wu Y, Yan JD. Fundamental properties of high-temperature SF6 mixed with CO2 as a replacement for SF6 in high-voltage circuit breakers. Journal of Physics D: Applied Physics. 2014;47:255201
  4. 4. Liu J, Zhang Q, Yan J, Zhong J, Fang M. Analysis of the characteristics of DC nozzle arcs in air and guidance for the search of SF6 replacement gas. Journal of Physics D: Applied Physics. 2016;49:435201
  5. 5. Zhang X, Dai Q, Han Y, et al. Investigation towards the influence of trace water on CF3I decomposition components under discharge. High Voltage Technology. 2016;42:172-178
  6. 6. Yokomizu Y, Ochiai R, Matsumura T. Electrical and thermal conductivities of high-temperature CO2–CF3I mixture and transient conductance of residual arc during its extinction process. Journal of Physics D: Applied Physics. 2009;42:215204
  7. 7. Yokomizu Y, Ochiai R, Matsumura T. Particle composition of CO2-CF3I mixture at temperatures of 300-30,000 K. The Transactions of the Institute of Electrical Engineers of Japan B. 2007;127:1281-1286
  8. 8. Qiu Y. GIS Equipment and Its Insulation Technologies. Beijing: China Water & Power Press; 1994
  9. 9. Xiao S, Cressault Y, Zhang X, Teulet P. The influence of Cu, Al, or Fe on the insulating capacity of CF3I. In: Physics of Plasmas (1994-present), vol. 23. 2016. p. 123505
  10. 10. Christophorou LG, Olthoff JK, Green DS. Gases for Electrical Insulation and Arc Interruption: Possible Present and Future Alternatives to Pure SF6. NIST TN-1425, vol. 82011. p. 391
  11. 11. Deng Y. Fundamental research of the application of environmentally friedly insulation gas CF3I in the electrical equipment [PhD thesis]. Shanghai: Shanghai Jiao Tong University; 2016
  12. 12. Christophorou LG. Insulating gases. Nuclear Inst & Methods in Physics Research A. 1988;268:424-433
  13. 13. Zhao X. The research of arc extinguishing and insulation characteristics of environmentally friendly insulation gas CF3I [PhD thesis]. Shanghai: Shanghai Jiao Tong University; 2018

Written By

Dengming Xiao

Submitted: 24 January 2018 Reviewed: 05 July 2018 Published: 05 November 2018