Yancy W. Riddle, Ph.D. Project Leader Postdoctoral Researcher yancy_riddle@yahoo.com	Håkon Hallem Diploma student  (dr.ing. student starting 6/2002) hakoh@stud.ntnu.no	Nils Ryum, prof. dr. techn. Institute Leader nils.ryum@chembio.ntnu.no

Research Area:  Study of Recrystallization in Aluminium Alloys

Current Research Topic: Addition of Sc and Zr to Aluminium Alloys: Improving Recrystallization Resistance
 

Introduction:
    Preventing recrystallization in wrought aluminium alloys may result in the retention of strength, hardness, corrosion resistance, fracture toughness, as well as decrease weld cracking.  These properties and more are often vital for the success of an aluminum alloy in a given application.  For the same reasons the alloy may be required to survive long elevated temperature conditions in-service and retain an unrecrystallized state.  However careful control of alloy composition and  material processing conditions are needed if these conditions are to be met.  In particular the various processing temperatures, processing times, and severity of deformation all contribute to the challenge of recrystallization prevention.

    In this research we employ various dispersoid phases to pin grain and subgrain boundaries thus preventing recrystallization.  This is termed "Zener drag".  The effectiveness of dispersoids varys with composition, size, volume fraction, distribution, and degree of coherency to the matrix. To date Al(3)Zr is the most effective dispersoids in use in aluminium alloys.  Unfortunately Al(3)Zr is often difficult to precipitate resulting in strongly heterogeneous distributions having a less than optimal ability to retard recrystallization.

    Recent research has shown scandium, when precipitated as Al(3)Sc, to be a particularly effective Zener drag agent.  Ease of nucleation, high volume fraction, and homogeneous distribution contribute to its effectiveness.  Furthermore scandium heat treatment is compatible with current practices making it an attractive addition to aluminium.  However Al(3)Sc coarsens much faster than Al(3)Zr reducing  its usefulness over longer high temperature annealing times.  It has been discovered that combining Zr and Sc in the same alloy can, with proper heat treatment, result in an Al(3)(Sc, Zr) dispersoid which takes advantage of the positive properties of Al(3)Zr and Al(3)Sc.  Al(3)(Sc, Zr) nucleates rapidly and homogeneously with a high volume fraction and slow coarsening rate (compared to Al(3)Sc).  The reader is referred to (M.Sc.) and (Ph.D.) for more detailed information regarding the behavior of Al-Sc-Zr alloys.  There is currently very limited use of scandium in industry making it a very expensive alloying addition.  Further understanding of its potency, discovery of alternate sources, and increased demand for use in aluminium alloys may significantly drop scandium's price in the future.

Our goals:
    Prevention of recrystallization in some aluminum alloys is increased above the current limit by adding both Sc and Zr.  Neither element acting alone can provide this level of recrystallization resistance.  Relatively little is known about Al3(Sc,Zr).  Our goal is to continue characterizing this new Al3(Sc,Zr) dispersoid phase, understand why it is so effective, and use this information for optimizing its performance.   To accomplish this we are utilizing optical, scanning, and transmission electron microscopy, microprobe analysis, and a variety of mechanical tests.  The Al-Sc-Zr system may also provide a suitable alloy for studying very early nucleation behavior which is of interest to a more general academic metallurgical audience.  For this we are planning an experiment using the ESRF syncrotron.

Recent Results:
    Recently we have been improving the recrystallization properties of an experimental extrudable Al-Mn-Mg alloy by adding Zr and Sc to form Al3(Sc, Zr) dispersoids.  Both commercial purity and high purity alloys are under investigation.  Without Sc addition we find the precipitation of Al(3)Zr following preheating to be very poor and heterogeneous as demonstrated in figure 1a.  No alternate heat treatment was found to improve the distribution of Al(3)Zr.  With addition of Sc both the distribution and volume fraction of dipsersoids, Al3(Sc,Zr) in this case, are both dramatically improved as seen in figure 1b.  The micrographs in figure 1 are from a transmission electron microscope (TEM).
 

Figure 1a:  Distribution of Al3Zr in an Al-Mn-Zr alloy	Figure 1b:  Distribution of Al3(Sc,Zr) in an Al-Mn-Zr-Sc alloy

    After extrusion these alloys were subject to annealing at 500, 550, or 600 C for 1 hour and recrystallization behavior observed in the optical microscope using anodized samples.  The grain structures of the non-Sc and Sc alloy shows markedly different behavior as demonstrated in figure 2.  Without Sc surface recrystallization begins after 1 hour at 500C with complete recrystallization occuring after 1 hour at 550 C.  By adding Sc to the alloy recrystallization is surpressed to at least 600C for 1 hour with only a thin layer of recrystallized material forming at the surface.
 
 

Figure 2:  Optical micrographs of extruded material after 1hour at 500, 550, and 600C               showing degree of recrystallization

The details of the current research will be published at the 8th International Conference on Aluminum Alloys in Cambridge, England in July 2002.  If you are interested in this research and/or would like more information please contact us.  Check back for periodic updates.

Last update:  November 13, 2001.