Much has been written about RTV silicone coated insulators and their performance. Currently, IEC is working on an important future standard (IEC 63432) to establish consensus for formal procedures and test protocols based on laboratory and field experience gained over decades. This edited contribution to INMR by experts at Sediver answers frequently asked questions on this topic and supplements a past INMR publication that focused on ageing and end-of-life of coatings.
Hydrophobicity Transfer & Recovery Properties
Hydrophobicity is a unique property of silicone materials linked to their inherent water repellency. This property is dynamic and can evolve as a function of different environmental stress conditions. When a silicone rubber surface is subjected to continuous dry band arcing or corona activity, hydrophobicity is suppressed. While this loss can sometimes be permanent, hydrophobicity usually recovers after some time. This property is termed ‘recovery’.
By contrast, once a silicone insulator is exposed to external pollution, its surface is progressively covered with dust and other pollutants whose crust is typically not hydrophobic. Fortunately, low weight molecular species in the silicone bulk material migrate through the crust of accumulated pollution such that the deposit layer itself becomes hydrophobic. This property is called ‘transfer’. It is the transfer property that makes silicone rubber such an interesting material for use in polluted service environments.
There is an important consequence of the transfer property when it comes to evaluating the pollution layer on a silicone surface. The fact that surface pollution is progressively encapsulated in the silicone bulk results in a relatively reduced amount of pollution being collected when measuring ESDD/NSDD.
For example, one study showed that, due to structural differences between the two chemistries, RTV silicone-coated insulators have a slightly faster transfer time than do commercially available HTV silicone materials used in polymeric insulators. Fig. 1 shows test data from surfaces polluted with Kaolin at ESDD=0.1mg/cm² and NSDD= 0.2mg/cm².

The upcoming IEC 63414 standard describes test procedures for artificial pollution tests on HTM insulators. It stipulates that solid layer pollution tests must be performed:
• within 24h after deposit of the layer if the goal is to determine insulator performance with limited recovery; and
• within 64h to 68h for a test with partial recovery.
There is also a provision about maximum NSDD to consider for testing. If kaolin is used, this limit is NSDD=0.25mg/cm² while with tonoko it is NSDD=0.4mg/cm² (Note: Compared to kaolin, tonoko less impacts conductivity of the pollution layer on an insulator surface). This issue is the more important given requests by some utility engineers who want artificial pollution tests performed with NSDD levels above 1mg/cm² or even 2mg/cm². It is unrealistic to conduct such tests since transfer cannot take place in a realistic time scale. Therefore, they are not representative of what happens in the field, where pollution accumulates progressively, and hydrophobicity rebuilds itself as well. Fig. 2 illustrates these dynamics at play in the pollution layer with an evolution of performance over time of approx. 25% over a period slightly more than a month.

Accuracy of ESDD/NSDD Sampling on Silicone Surfaces
Another issue related to silicone’s transfer property is to quantify level of pollution on an insulator in service. This technique is widely used to evaluate site pollution, such as in insulator test stations. Yet transfer of hydrophobicity will progressively encapsulate pollutants and alter accuracy of evaluating pollution level using the typical method of collecting dust by washing.
Fig. 3 reports on an investigation made by Sediver’s Research Center on a variety of kaolin species. Type 1 is common kaolin used for pollution testing and has properties in accordance with IEC 60507. Types 2 to 4 represent a kaolin having granulometry below the IEC recommendation and which can be compared to thin dust particles, such as in desert environments. Type 5 is Kieselguhr.

Slurries made with these different types of contaminants were applied by dipping on slabs of silicone coated tiles (as in Fig. 5). Level of pollution on these slabs was set at ‘Heavy’ as per the IEC60815-1 classification. One slab of each type was used as reference to measure ESDD/NSDD at the beginning of the test.

After 70 days, slabs were cleaned to evaluate ESDD and NSDD levels and compared with values from a reference slab that was initially polluted together with test samples. As seen in Fig. 6, a typical pattern common to all types of kaolin tested shows that the silicone surface will retain most of the NSDD while the ESDD would typically be mostly (but not completely) removed during the cleaning phase. This means that cleaning a silicone surface to measure ESDD/NSDD in the field will provide a largely underestimated NSDD whereas ESDD can be considered as only slightly reduced.

To validate findings, SEM images were taken of the slabs after they were cleaned to remove pollution from their surfaces. Fig. 7 shows the surfaces of several samples, consistent with all other samples during this test. Large quantities of the original minerals are still there, encapsulated in the silicone surface and therefore are not accounted for during measurement of NSDD and, to a lesser degree, ESDD.

Electric Field Simulation of Coated Insulators
This is frequently asked and is a legitimate question from the theoretical point of view. But it is important to keep in mind that hydrophobicity is a dynamic process. It is unpredictable and always with heterogeneity along the insulator surface, even if higher electric stress is by nature located around the pin.
Testing discussed above shows how difficult it can be to theoretically describe the dynamic evolution of silicone in service. Fig. 8 shows a model used for electric field simulation of a silicone-coated glass insulator for a short string. The calculation was processed through COMSOL. A string of 5 fog type insulators was modeled with a 350µm silicone coating over the glass and with the string energized at 80 kV.

Evaluation of voltage gradient was determined and compared to values for an uncoated string. As shown in Fig. 9, which describes the potential distribution on each of the 5 units, it appears clear that there is no significant difference if silicone is added to the surface or not.
Similarly, Fig. 10 shows electric field along the profile of insulators for the first two units, which have highest stress applied. Here again, it can be seen there is no significant difference between coated and uncoated units. Evaluation of E-field on coated insulators is therefore pointless since it yields no scientific or practical knowledge. Moreover, once hydrophobicity varies, it will become impossible to correctly model the insulator’s surface.


How Coatings React to Acids
Polymeric insulators can be damaged (even up to a point of failure) when corona and electric arcing occur and remain active along the core. This can result in erosion and cracking of the sheath, and the insulator can fail from acids generated by corona degrading the fiberglass rod. This risk is different for coated insulators since both glass and porcelain do not contain an organic fiberglass rod and will not fail due to presence of acid. This is one of the benefits of coated insulators.
At the same time, there is the matter of how coatings behave in an acidic environment. Not all are the same and their chemistries can show different reactions. Sediver’s Research Center, for example, evaluated behavior of a variety of coatings under sulfuric acid and nitric acid. Fig. 11 shows the condition of different coatings after 100h and 200h in 1N concentrations of sulfuric and nitric acid at 50°C.

Only one of the 4 different chemistries showed degradation with cracks. Coatings, in general, seem less sensitive to acid attack than do some HTV materials (as seen in Fig. 12 from 3 commercially available HTVs used in composite insulators). The most significant difference is that, even with cracks, there is hardly any impact on behavior of a cap & pin suspension insulator since there is no fiberglass rod nor organic interfaces.

Table 1 shows variations in hardness and hydrophobicity for different types of RTV and HTV silicones. Still, it is difficult to use such parameters to discriminate benefits other than through visuals, e.g. where one type of RTV (Type 3) showed cracks and greatest change in hardness. All HTVs tested showed cracks and HTV 3 showed severe degradation yet was the sample whose hardness changed least. Moreover, changes in hydrophobicity measured through variations in contact angle did not yield value since most samples registered a slight increase in hydrophobicity due to attack on the polymer.

Verifying such properties for HTV silicones used in polymeric insulators might make sense due to the stress applied on the material either around the rod or at the base of the sheds, e.g. during severe conditions. But such testing is not needed for coatings which are applied on solid substrates.
Conclusions
• Understanding the effect of silicone’s transfer properties is key to addressing pollution testing;
• Measuring ESDD/NSDD on silicone surfaces can lead to underestimation given that encapsulation is taking place with the transfer of hydrophobicity over pollutants;
• Electric field simulation of silicone coated glass or porcelain insulators provides no real benefit in predicting flashovers or pollution related performance;
• Acid tests on silicone-coated insulators offer little value, even if some properties change under acid attack (mainly by nitric acid). Yet this it is not as important as the need to evaluate ageing of HTV polymeric insulators.
References
[1]: CIGRE Brochure 837 Coating for improvement of electrical performance of outdoor insulators under polluted conditions. June 2021
[2]: IEC 63414 Draft. Artificial pollution tests on high voltage polymeric insulators to be used on AC and DC systems.
[3]: INMR 2023 Bangkok. “Coating…of course”. JM George, S. Suc, F. Virlogeux.




















