This project aims to develop underpinning knowledge on the characteristics of C4F7N gas mixtures, which may be used as alternatives to SF6, with a specific focus on the long-term gas stability to demonstrate performance in service in new equipment. It will consider credible in-service scenarios where the gas may be put under stress that could lead to changes in the chemical make-up of the gas mixture, including any changes that could result from interaction with materials used in the manufacture of the equipment and low levels of moisture and air contamination. The resulting by-products will be assessed for their potential health impacts, their reactivity with other materials and the prospect that they may be use as diagnostic markers for asset health.
Benefits
This research project is intended to deliver benefits through the facilitation of the use of SF6 alternatives and a consequential net environmental benefit and reduction in carbon (CO2 equivalent) emissions.
Learnings
Outcomes
Many of the different factors investigated (water, air, light, pressure) that could impact by-product formation in C4F7N gas mixtures suggest that no unexpected products are formed during in-service conditions compared with thermal ageing of the gas mixture. The introduction of electrical ageing and also the presence of epoxy resin suggests that different by-products might be created under these conditions. This will feed into the next stage of the project where the presence of by-products will used diagnostically.
Despite the non-toxic nature of both SF6 and C4F7N mixtures, both generate small amounts of toxic breakdown products. It is recommended that handling guidelines for in-service C4F7N gas mixtures should be based on IEC 62271-4.
Although the GWP of C4F7N by-products are high, causing some concern over the environmental impact, the quantities of the by-products are very small. Loss of the gas to the atmosphere should still be avoided through careful handling procedures.
Compatibility studies involving EPDM with the clean C4F7N gas mixture caused a marked decrease in the C4F7N peak in the GCMS. Other nitriles present in decomposed gas were also seen to decrease suggesting some degree of incompatibility between EPDM and C4F7N, confirming reports by others. This would have implications for retro-filling equipment with EPDM components.
Recommendations for further work
None at this stage
Lessons Learnt
In designing a test set up for HVAC testing of gas mixtures where the effect of pressure is being considered as one of the variables, it was identified that conventional sensors respond too slowly. An integrated-electronics piezo-electric (IEPE) sensor is being trialled in combination with another method. Thermal sensors are also slow, and an optical technique was employed.
A commercial gas sensor for C4F7N was used for by-products but has challenges because some potential by-products are not detectable. GC-MS was therefore chosen as the best option for identifying them. The commercial detector is better than GC-MS at separating carbon monoxide from other gases (oxygen and nitrogen) and enabling it to be quantified.
Repetitive breakdown testing (400 shots) under LI and HVAC conditions in small gas volumes shows evidence of breakdown of the C4F7N gas mixture. CF4 is the first product identified and then C2N2. The -CN group is lost readily and combines to produce C2N2, but there are not many other -CN containing products and it may recombine to reform C4F7N. C4F10 is more clearly identified in larger gas testing experiments.
Increasing the moisture content in the breakdown tests has a number of impacts. For example, it reduces the dielectric strength and breakdown voltage and may affect the recombination process of the gas mixture, however it does not appear to change the type of by-products present. Adding atmospheric air affected the concentration of by-products but not the compounds produced. Increasing pressure had no effect on by-products either.
2024/2025 update
It was observed in the comparison of by-products of SF6 and C4F7N mixture that SF6 generates products of a lower molecular weight (MW). While this is mostly true of C4F7N, loss of nitrogen and reaction to add fluorine (either C4F8 or C4F10) creates slightly larger molecules. C4F8 and other breakdown products have lower Lethal Concentration (LC50) values 50. LC50 describes the amount of chemical inhaled by test animals that causes death in 50% of test animals used during a toxicity test study and is an indication of the possible toxic effect on humans. This observation is similar to the known effects of SF6 byproducts.
The global warming potential (GWP) of SF6 is 24.300 and is higher than any other known substance. Inevitably, therefore, the by-products created when it decomposes are lower than SF6, albeit that some of them are still high. By-products of C4F7N, which has a GWP of 2,100, can be significantly higher e.g. C2F6 (11,100), C4F10 (9,320) and C3F6 (9,310). The atmospheric lifetimes of many of the by-products of both gases are high.
Clean, new C4F7N gas mixture and a comparable gas mixture exposed to arcing and containing decomposition products was exposed to light for 30 days to determine the effect of ultraviolet (UV) radiation on the gas mixture. There was no discernible effect on the composition of the gas, and it was concluded that, for the exposure level used, UV has no impact.
Electrothermal ageing and compatibility testing under thermal and electrothermal conditions of the C4F7N gas mixtures with epoxy resins, ethylene propylene diene terpolymer (EPDM) rubber and bromobutyl rubber (BIIR) have been completed and various decomposition products have been identified, many of which have not previously been identified in the literature. CO and C3F6 have been reported widely, CHF3 has been identified by one research group many others have also been found. Further work will be carried out to understand the presence of these by-products. The presence of CH3Cl in one of the thermal ageing tests with epoxy resin was unexpected and could indicate contamination although chlorinated compounds in some epoxy resins have been reported.
Dissemination
All three GB transmission owners are invited to join project update meetings so that the work proceeds collaboratively, and all results are received at the same time.
Four papers reporting results from this project have been submitted for presentation at the International Symposium on High Voltage in Nagano, Japan in August 2025. The titles are as follows:
“Gas Analysis of C4F7N Gas Mixture After the Exposure of Epoxy Resin and PTFE to Repetitive Surface Flashover Events”
“Decomposition of the C4F7N-CO2-O2 Gas Mixture Under Repetitive Lightning Impulse Breakdown Events”
“Thermal decomposition of C4F7N-CO2-O2 mixture based on molecular dynamics”
“Exploring the compatibility between C4F7N-gas mixture and common GIS Insulator materials using the thermal method”.
A paper entitled “Compatibility of the C4F7N/O2/CO2 Mixture with Epoxy Resin After Long-Term Thermal and Electro-Thermal Tests” has been submitted to the IEEE Conference on Electrical Insulation and Dielectric Phenomena for presentation in Manchester in September 2025.
