Zero Inclusion Steels

摘要

An examination of the influence of inclusions onsteel properties indicates several general trends, although thespecifics of each case can vary considerably. The first of the trendsis that the most deleterious inclusions are the largest ones, whichare almost invariably collections of small particles rather than onesingle precipitate. The second is that 'large' is a function of theproperty affected and of the alloy concerned, but non-metallicinclusions of less than one micron dimension, separated bydistances which are greater than 10 microns, essentially have noeffect on mechanical properties. Unfortunately, 'large' in mostcases is not quite large enough to enable ultrasonic or other NDTmethods to be used as a reliable quality filter. The third is that asmelting processes become more refined, and in consequence thecontent of large inclusions in the product becomes smaller, ourpresent methods of inclusion assessment become less and lesseffective in judging quality. In the cleanest alloys presently madeby remelting techniques, the probability of metallographicdetection of the small fraction of large inclusions which wouldaffect properties is negligible and metallographic assessment istherefore not p055ible. Since this small fraction of largeinclusions also constitutes a negligible fraction of the total oxygencontent, any assessment based on chemical analysis is equally invalid.The corollary to the above situation is that as inclusion qualityassessment becomes more problematical, the reliance on processcontrol becomes greater - a fact of production which in the aero-engine industry has led to a proliferation of specifications andfrozen practices. The logical end-result of the quality improvementprocess has produced the idea of 'zero inclusion' steels. These arealloys in which the process of melting itself guarantees the absenceof inclusions which are large enough to influence mechanicalproperties. It appears from research that the only process capableof this degree of cleanness is electron-beam hearth remelting, andit is in this context that we have carried out the developmentsdescribed in this work.The alloys which have been studied using BBCHM are austeniticstainless steels (for electronic applications) and maraging 250 steel(for aerospace applications). The compositions and proceduresrequired for the zero inclusion condition were determined inlaboratory experiments, and the alloys were subsequently meltedusing these techniques on a large scale. Results are reported for thestainless steel on the scale of 1 tonne ingot, and for the maragingsteel at the scale of 5 tonne ingots.It appears from the results of this programme that the zeroinclusion condition can be achieved in appropriate alloycompositions i.e. those in which the evaporation experienced inFBCHM would not be prohibitive. (Two alloying elements whichare a particular problem in this regard are Mn and Cu.) Since theEBCHM process is very amenable to on-line process controlincluding image analysis of the liquid surface, we believe that itmeets all of the quality control requirements of at least the nearfuture in very high quality alloys. The cost attached to the productis significant, but not unreasonable in view of the quality attainedand the critical nature of the end product. The EBCHM process isan accepted industrial tool in Japan, the USA and in Russia and theequipment required for the zero inclusion alloy production on therequired scale is presently available.