Although not specified by the IEC, many of those affected by European directives include a 3% by weight limit for polycylic aromatic compounds (PACs) as measured by IP 346 so that the oils purchased do not require special hazard labels. PACs are also called polycyclic aromatics (PCAs) or polynuclear aromatics (PNAs). In the IP 346 method, solvent extraction with dimethyl sulfoxide (DMSO) is used to isolate PACs from the portion of the oil in the boiling range of 300˚C and higher. The amount of PACs is determined gravimetrically. The method is not suitable for used oils.
It is desirable to have a test that better characterizes the composition of aromatic compounds than aniline point. Changes in crudes and refining techniques can affect the aromatic content of a product, and therefore aniline point can be used to provide a measure of the consistency or continuity of product. One of the difficulties of specifying aromatic content is that generally similar, overall aromatic content in different crudes with dissimilar compositions of PACs, polar compounds, heterocyclic nitrogen, and sulfur compounds, etc., may yield oils with different performance characteristics. Gassing tendency under partial discharge conditions, impulse breakdown voltage, and oxidation stability are all affected by the amount and types of aromatics.
The impulse breakdown voltage of oils tends to decrease with increasing concentration of PACs, with the effect being more pronounced with increasing number of rings. With the pressure today for oils to have lower aromatic contents, and particularly lower PAC contents to alleviate environmental concerns, the impulse breakdown voltage of electrical insulating oils easily meets this requirement.
The gassing tendency can be defined as the rate of gas evolved or absorbed by an insulating oil when subjected to electrical stress of sufficient intensity to cause ionization. The oil sample in the test cell is saturated and blanketed with hydrogen gas (rarely, nitrogen). The characteristic is positive if gas is evolved and negative if gas is absorbed. The use of different saturating gases can affect the absolute value of the gassing tendency but typically does not affect the relative ranking of oils.
The overall rate of change of gas volume in an ASTM D 2300 test cell is the result of the rates of two processes in the oil/gas froth where the partial discharge occurs. Activated products, primarily hydrogen, are produced by the decomposition of the oil. Activated hydrogen species are formed at the same time in the hydrogen blanket. The active species formed in both processes hydrogenate aromatic hydrocarbons in the oil and are incorporated into the condensed phase at a rate limited by the concentration and type of aromatics present.
Excess activated species revert to hydrogen gas in its ground state and add to the volume of gas in the cell. Oil in which the rate of uptake of active hydrogen exceeds the rate of its production from the decomposition of the oil in the discharge "absorbs" hydrogen from the gas blanket. Oil in which active hydrogen is generated by the decomposition of hydrocarbons in the discharge at a greater rate than the rate of uptake of hydrogen by aromatic hydrocarbons "evolves" gas to the blanket. Gas evolution in this test increases as the aromatic hydrocarbon contents of oils decrease.
The wide variety of conditions under which partial discharges occur in operating transformers cannot be reproduced in a standard test performed on oil in a laboratory. However, molecular hydrogen is the principal degradation product of low energy electrical discharges found in mineral oil-filled transformers. The solubility of hydrogen is low in transformer oils, and gas bubbles readily can form in the oil in the vicinity of a discharge. Gas bubbles are weak dielectrics and their presence can intensify the partial discharge or initiate further breakdown in highly stressed regions nearby. No specific correlation has been established between the rates of gas evolution of transformer oils measured by the D 2300 test and the performance of these oils in transformers. The likelihood of damage to the insulation system in a transformer by low energy discharges is clearly higher in oil in which a high gas evolution rate is measured in test method D 2300. For some equipment, such as cables, bushings and instrument transformers, certain capacitors, or for insulation which is only partially impregnated, low gassing tendency oils may mitigate the effects of partial discharges.
The ASTM specification lists limits for two test methods, D 2300 procedure A (maximum of 15 μL/min) and procedure B (maximum of 30 μL/min). There is no requirement in the IEC specification.
When extreme values for gassing tendency is compared with impulse breakdown voltage and aromatic content values as shown in Table below, the relationship between these properties is revealed. Oils with lower gassing tendency also tend to have lower impulse breakdown voltages and higher total aromatic contents. There appear to be other factors which have an influence, as Oils M and O both have impulse breakdown voltages of 184 kV and yet have very different gassing tendencies and aromatic contents.
This may be due to differences in the relative amounts of mono- and poly-aromatic species present in the oils. The method for determining aromatic content, ASTM D 2140, does not provide an indication of the relative portions of the different types of aromatic hydrocarbons.