STP108

    Oil Ash Corrosion of Materials at Elevated Temperatures

    Published: Jan 1951


      Format Pages Price  
    PDF (2.7M) 55 $25   ADD TO CART
    Complete Source PDF (5.8M) 55 $55   ADD TO CART


    Abstract

    Ash from residual fuel oils can be highly corrosive to materials at elevated temperature, particularly when it contains substantial quantities of vanadium. Crude oils high in vanadium are especially prevalent in some foreign oil fields and are also quite common in the United States. Heavy fuel oils may contain up to 3.0 per cent ash. An oil ash from domestic fuel oils containing approximately 6 per cent vanadium, 5 per cent sodium, and 10 per cent sulfur was corrosive to S-588 heat-resisting alloy in stress-to-rupture tests at 1350 F. It was determined that the influence of stress was relatively unimportant, the attack being confined to the surface of the specimens. The heaviest corrosion occurred above the line of ash contact. Alloy 19-9DL was not so heavily attacked by this ash. On the other hand, domestic ash containing approximately 3 per cent vanadium, 1 per cent sodium and 8 per cent sulfur was not corrosive to either of these alloys, or to a wide selection of heat-resisting materials, in unstressed tests at temperatures of 1000, 1350 or 1500 F. The unstressed tests were run with only about one-half of each specimen underneath the ash. Two ashes from Venezuelan crude oils and one ash from Middle East crude oil were compared against nine representative metals at 1000, 1200, and 1350 F. A liquid phase was observed in all three ashes at 1000 F., and slight attack occurred on all the alloys at 1000 and 1200 F. At 1350 F. these ashes are liquid, and heavy attack occurs, either above or below the ash line or both. In many cases, “catastrophic oxidation” was observed. One of the Venezuelan ashes with approximately 44.5 per cent vanadium, 2 per cent sodium and 3 per cent sulfur was selected for “screening” tests on over 80 different materials or protective treatments. The tests were run for 168 hr. at 1350 F. The “superalloys” had the worst corrosion resistance, whereas the relatively weak low nickel or nickel-free alloys had the best. Ordinary 18-8 stainless steel had the best resistance in the austenitic or “superalloy” class. Several ceramics were also severely attacked. Of the protective coatings, “Ihrigizing” was the best for carbon steel, and “Chromizing” was also effective in protecting carbon steel and type 430 stainless steel. “Siliconizing” reduced the amount of attack on some of the superalloys.Metallizing with aluminum appeared to have promise in protecting 18-8 steel. However, when some of the better materials or protective coatings were tested for a longer time at 1350 or 1500 F., heavy attack occurred. In a study of the effects of various percentages of V2O5 in sodium sulfate (Na2SO4) on the corrosion of 18-8 at 1350 F., it was found that no corrosion occurred at 2 per cent or less of V2O5. This coincided with the percentage at which the mixture has no liquid components at 1350 F. The Venezuelan ash did not corrode any of the stainless or heat-resisting alloys in outdoor exposure tests for over seven months. It was determined that the attack of hot oil ash on metals is confined to the surface, with the exception of high-nickel alloys, where intergranular penetration was observed. It is not necessary for the material to be in contact with the ash in order to experience severe corrosion. It is concluded that no straightforward metallurgical solution is in sight which can be expected to take care of the more severe cases of oil ash corrosion of materials at elevated temperatures. The use of additives in the oil is suggested as a promising approach to an over-aal soution of this problem. Lime (CaO) is the most promising of the additives which have been investigated.


    Author Information:

    Evans, C. T.
    Chief Metallurgist, Elliott Co., Jeannette, Pa.


    Paper ID: STP46733S

    Committee/Subcommittee: G01.05

    DOI: 10.1520/STP46733S


    CrossRef ASTM International is a member of CrossRef.