The nature of the FeAl 3 liquid-(FeMn)Al 6 reaction in the Al-Fe-Mn system

  • PDF / 419,863 Bytes
  • 5 Pages / 594 x 774 pts Page_size
  • 94 Downloads / 170 Views

DOWNLOAD

REPORT


I.

INTRODUCTION

IT has previously been

established that there are two eutectic troughs in the aluminum-rich comer of the AI-Fe-Mn system, as shown by Figure 1 which is taken from Denholm et al. t These authors also report that the nature of the third univariant reaction emanating from the ternary eutectic point, as shown by point S of Figure 1, is uncertain. Thus, Phillips 2 previously concluded that the third reaction was peritectic at higher temperatures but became eutectic near the ternary eutecfic point. However, Phillips based his conclusion on the fact that his results required a very steeply sloping line to pass through the composition of his eutectic point as shown by point U in Figure 1 and this, in turn, required the tangent to this line to pass between the two points representing the compositions of the simultaneously precipitating solid FeA13 and (FeMn)A16 phases, as required by a eutectic reaction. Since Denholm et al. ~ found that the ternary eutectic point, S, lies well to the left of that reported by Phillips, thus requiring a much less steeply sloping line, they postulated that the third reaction may be peritectic throughout. The present paper provides evidence to show that the third univariant reaction is peritectic in the vicinity of the ternary eutectic point.

ing solution, a further crystallization run was made. This procedure was continued through five incremental additions of iron hardener, leading to an alloy composition of 2.00 pct Fe and 0.49 pct Mn. Three incremental additions of manganese hardener (2 to 5 g, 75 pct Mn) were then made, leading to the final alloy composition of 1.99 pct Fe and 0.63 pct Mn. The compositions of these alloys are shown by the points A to I of Figure 2 together with the points K and L representing the compositions of two alloys previously considered by Denholm et al. l The sizes of the circles and squares chosen to represent each point is considered to be a reasonable indication of the degree of precision of the sampiing and analytical methods used. The compositions of the intermetallics obtained during the crystallization runs are shown in Table I, and the nature of the primary crystals obtained from each run is shown by the symbols [] and (3 of Figure 2, representing FeA13 and (FeMn)A16, respectively.

I

2.0

F /

P'HIMAHY

1.8!-o

II.

FoA,3

CRYSTALS 1 I ~ 6 6 3 /

I

/I

I

s

I oH ~ / /

EXPERIMENTAL

The apparatus, the methods of crystallization and sampiing, and the alloys used are the same as those previously reported by Denholm et al.,~ the major impurity being 0.04 pct silicon which is not considered sufficient to alter the composition of the intermetallics which precipitate close to the ternary point under very slow cooling conditions. The present procedure consisted of first preparing 4.4 kg of a liquid aluminum alloy containing 1.75 pct Fe and 0.49 pct Mn, corresponding to point A of Figure 2, a composition which lies just to the right and above the ternary eutectic point shown by J. Samples of liquid or slurry were then taken while cooling,