Prevention of silver corrosion issues in oil-filled power transformers - Selective approaches for elemental sulfur quantification and its removal from mineral insulating oil

Abstract

The electrical grid is one of the most complex and important systems of our era. The grid transmits and provides electricity to millions of homes, businesses, schools, hospitals, etc. in an impressive network that can reach extensions of thousands of miles. The power grid, in particular the power transformers, are constantly subjected to numerous stresses that degrade the mechanical and dielectric properties of the liquid insulating system (usually mineral oil), eventually compromising its function. The environmental and economic concerns regarding the disposal of used mineral insulating oil have driven to major efforts on extending the life time of the insulating system. A number of regeneration processes, usually referred to as reclamation processes, have been successful to reduce the level of contaminants and oil degradation products in the oil. Despite considerable effort, issues remain with the presence of elemental sulfur (S₈), typically found following some oil regeneration procedures, where procedures utilised to remove other corrosive sulfur compounds (e.g., dibenzyl disulfide) were proven ineffective for S₈. In recent decades, the occurrence of power transformer failures associated to the presence of corrosive sulfur in the insulating mineral oil has experienced a marked increase, and corrosion of silver-coated electrical contacts in switching compartments due to trace levels of elemental sulfur is a known contributor to this increase. It has been reported that S₈ levels, as low as 1 ppm, may be sufficient to induce silver corrosion within the power transformer. The standard corrosion test (DIN 51353) is time consuming, requires 18 h for completion, and is followed by a visual subjective rating process. Hence, the development of a rapid and reliable analytical method to monitor the presence of elemental sulfur is of critical importance. To improve this, two analytical techniques to detect trace levels of S8 in insulating mineral oil have been developed, the analytical methods were optimised and laboratory results that demonstrate the potential of the approaches adopted are presented in this thesis. Both methods are based on the selective reaction of elemental sulfur with triphenylphosphine (TPP). The derivatisation of the elemental sulfur required minimal sample preparation and resulted in the formation of a single compound, namely triphenylphosphine sulfide (TPPS). The developed GC-MS method allowed S₈ concentrations to be monitored down to 0.5 ppm, while the UHPSFC-MS was able to quantify S₈ as low as 5 ppb. These methods can be used as a routine test or to confirm the presence of S₈ in samples where corrosion has occurred. Herein is also presented an effective laboratory reclamation approach to selectively remove S₈ from insulating mineral oil using polymers loaded with trisubstituted phosphines. The general concept of this method is based on the capability of the trisubstituted phosphine to react selectively with elemental sulfur present in the mineral oil. The reaction leads to the in-situ formation of the corresponding phosphine sulfide, converting the elemental sulfur in a suitable form to be removed. The use of the trisubstituted phosphine-loaded polymers enabled not only the selective and near total removal of S₈ from mineral oil but also the possibility of the polymer to be regenerated and reused. The reliability and robustness of our method was demonstrated by evaluating the capability of four different polymers to selectively remove S8 from a real transformer oil sample that was found to be corrosive towards silver. The results showed that the polymers were capable of efficiently removing ≥ 99.5% of the S₈ present in the oil sample, i.e., from circa 17 ppm to below 0.01 ppm. Which demonstrates the ability of our reclamation process to extend the life-time of oil insulated power transformers, by converting a corrosive sample into non-corrosive. Considering the operational conditions of our system, i.e., equipment, operational costs, time frame, efficiency, etc., a scale up to an industry level should be conceivable.

More information

Identifiers

ORCID for Richard Brown: orcid.org/0000-0003-0156-7087

Catalogue record

Export record