[Skip to content]

.

Rotary friction welding of an Fe3Al based ODS alloy

P D Sketchley*, P L Threadgill* and I G Wright**

* TWI, Cambridge, UK
** Oak Ridge National Laboratory, USA

Presented at 5th International Conference on Structural and Functional Intermetallics, Vancouver, Canada, 17-19 July 2000 and to be published in Intermetallics Journal.

Abstract

An Fe 3Al based oxide dispersion strengthened (ODS) alloy is under consideration for possible use as tubes in advanced heat exchangers, and it is necessary to investigate methods of joining the alloy to itself, and to Haynes 230 alloy. Previous experience on iron aluminides has shown them to be weldable by several processes, but it is known that fusion processes invariably lead to a loss of the Y 2O 3 oxide dispersion which is an essential feature of ODS alloys. Therefore, solid state processes offer a potential advantage, and in this work continuous drive rotary friction welding has been investigated as a method to join the Fe 3Al ODS alloy in both the recrystallised and unrecrystallised condition. Trials were also undertaken to join both recrystallised and unrecrystallised material to Haynes 230 alloy.

All welds were made in 15mm diameter material, using a conventional continuous drive rotary friction welding machine. Welds were evaluated initially using bend tests, and detailed metallographic observations of the grain size, and the nature of the solid state interface between the materials.

It was found possible to make high quality welds containing no flaws for a variety of welding conditions. The microstructural condition of the Fe 3Al ODS alloy had no apparent influence on weldability, and no difficulty was encountered in making the dissimilar metal joints. The results obtained are discussed in terms of the microstructures obtained, and are compared with other studies on joining iron aluminides and ODS alloys.

Introduction

Intermetallic alloys have been the subject of many investigations to exploit their excellent high temperature mechanical properties and oxidation resistance. Recent studies at ORNL have developed improved alloys based on the Fe 3Al system, whereby these alloys are used as the basis for ODS alloys. These alloys are made by a powder attrition and compaction method, and contain a very fine dispersion of oxide particles (usually Y 2O 3) which improve the creep resistance by pinning dislocations, and a stable oxide layer of Al 2O 3, which provides excellent oxidation and corrosion resistance. Such materials of great interest for new generation heat exchangers and other high temperature equipment.

While development of such alloys continues, it is important to remember that techniques have also to be developed to enable them to be processed and fabricated into useful components. Part of this of course involves the need to join the alloy to itself, and possibly to other materials.

Earlier work on ODS alloys by several authors has established that the use of fusion welding techniques will seriously compromise the high temperature mechanical properties of the alloys, as the fine dispersion of oxides in the alloy will be destroyed. However, many ODS alloys can be welded without problems of cracking or porosity, even though the high temperature properties are poor. Non-fusion techniques are therefore intrinsically more attractive, as these will not destroy the critical oxide distribution.

So far, no studies are known which has investigated the solid state joining of the present alloy. However, ODS alloys such as MA956 and PM2000 have been successfully joined by continuous drive rotary friction welding and diffusion bonding [1,2,3] . Recent work has also reported the successful joining of an ODS alloy based on Fe 60Al 40 [4] . Studies on welding ODS alloys to date have concentrated on room temperature mechanical properties and on understanding the microstructures formed, and there appears to be little data in the public domain on the high temperature performance of these alloys.

In the present work, a feasibility study has been carried to investigate various aspects of joining a Fe 3Al-ODS alloy using continuous drive rotary friction welding. These were as follows:

  • Fine grained Fe 3Al-ODS to fine grained Fe 3Al-ODS.
  • Coarse grained Fe 3Al-ODS to coarse grained Fe 3Al-ODS.
  • Fine grained Fe 3Al-ODS to Haynes 230 alloy.
  • Coarse grained Fe 3Al-ODS to Haynes 230 alloy.

Experimental approach

Materials:

All materials were supplied as 15mm diameter bar by ORNL. Analyses of the alloys are given in Table 1. All Fe 3Al-ODS alloy was supplied in a fine grained (approximately 1µ average grain diameter) condition. Some material was subsequently heat treated at 1300°C to convert it to the coarse grained condition. In ODS alloys, maximum creep resistance is achieved in this condition, and the grains can be of several millimetres in length. In hot rolled rod as used here, the coarse grains would have a high aspect ratio, probably 5:1 or greater.

Table 1 - Analyses of alloys:

Cr 21.95 Al 0.29
Co 0.74 B <0.003
Fe 1.65 C 0.10
Mo 1.50 Cu 0.02
W 13.70 Mn 0.50
P 0.005 S <0.002
Si 0.38 Ti 0.01

Welding equipment:

All welds were made using continuous drive rotary friction welding. A modified Verson AI machine was used. This is capable of rotation speeds up to 4500rpm, and can apply an axial force of 100kN. The transmission power of the spindle motor is approximately 7.5kW. In the continuous driver rotary friction welding process, one component is rotated against another under an applied pressure. Frictional heat causes the materials to soften and plastically deform. After a pre-set displacement (known as burn-off) has occurred, the machine is rapidly braked, and the pressure increased to generate a high quality solid state friction weld.

During welding, the primary parameters (friction force, forge force, rotation speed and displacement) were continuously monitored and recorded.

Prior to welding, the mating faces of all specimens were machined perpendicular to the rotational axis, and the samples were degreased with petroleum ether and dried.

Metallographic examination:

Samples for metallographic examination were mounted in Bakelite and polished to a 1micron finish or better using conventional metallographic techniques, followed by etching in a HNO 3-glycerol solution.

Results

Fine grained to fine grained Fe 3Al-ODS:

No great difficulty was found in generating good quality rotary friction welds, a result in line with previous experience. A typical macrosection is shown in Figure 1. The flash formation is ideal, with the diameter of the sound extruded material exceeding the original bar diameter, and the HAZ is wider at the outer diameter of the bar, reflecting the greater heat input in this region due to the higher relative velocity in this area. Tensile properties were variable, with tensile strengths between 634MPa and 1305MPa. As can be seen in Table 2 attached, the other tensile values were 1140 and 1288MPa.

Fig. 1. Macrosection of friction weld between fine grained ODS alloy and itself
Fig. 1. Macrosection of friction weld between fine grained ODS alloy and itself

The weld microstructures are essentially divided into four regions, which can be defined as follows, working from the parent material to the bond line:

A: Parent material not visibly affected by the welding process.
B: Area in which the orientation of the elongated parent grains changes progressively from parallel to the bar rolling direction to parallel to the plane of the weld. There is no evidence of recrystallisation in this region, but the formation of equiaxed subgrains in the more severely deformed regions suggest that recovery of the high dislocation density may be occurring.
C: An area where the microstructure consists of fine, equiaxed recrystallised grains, typically <5microns in diameter.
D: An area at the bond line where the microstructure has been dynamically recrystallised, giving a typical grain diameter of 5-8microns.

A typical trace of welding parameters believed to give a near optimum weld is shown in Figure 2.

Fig. 2. Parameter trace for weld W19 between fine grained Fe 3Al-ODS and itself
Fig. 2. Parameter trace for weld W19 between fine grained Fe 3Al-ODS and itself

Typical microstructures from these regions are shown in Figures 3 to 5. It is noted that these structures are typical of those found in friction welds in other iron based ODS alloys [2,4] .

Fig. 3. Microstructure of fine grained Fe 3Al-ODS alloy
Fig. 3. Microstructure of fine grained Fe 3Al-ODS alloy
Fig. 4. Microstructure of deformed region in HAZ of friction weld in fine grained Fe 3Al-ODS alloy
Fig. 4. Microstructure of deformed region in HAZ of friction weld in fine grained Fe 3Al-ODS alloy
Fig. 5. Microstructure of bond line region of friction weld in fine grained Fe 3Al-ODS alloy
Fig. 5. Microstructure of bond line region of friction weld in fine grained Fe 3Al-ODS alloy

Coarse grain Fe 3

Al-ODS to coarse grain Fe 3Al-ODS:As for the fine grained alloy, no particular difficulty was experienced in making a high quality rotary friction weld in the coarse grained material, as can be seen from the macrosection shown in Figure 6. The same features were observed, although the extent of the deformed but unrecrystallised region was slightly reduced. At the bond line, the microstructure was fully recrystallised, and in fact this microstructure was virtually identical to that found when welding the fine grained material, as shown in Figure 7. It should be noted that a higher forge force was used (80kN as opposed to 70kN in the fine grained material), leading to a more rapid burn-off rate, and this is consistent with the smaller region of deformed microstructure. A typical trace of parameters is shown in Figure 8.

Fig. 6. Macrosection of friction weld between coarse grained ODS alloy and itself
Fig. 6. Macrosection of friction weld between coarse grained ODS alloy and itself
Fig. 7. Microsection of bond line section of friction weld between coarse grained ODS alloy and itself
Fig. 7. Microsection of bond line section of friction weld between coarse grained ODS alloy and itself
Fig. 8. Parameter trace for weld W21 between coarse grained Fe 3Al-ODS and itself
Fig. 8. Parameter trace for weld W21 between coarse grained Fe 3Al-ODS and itself

Fine grained Fe 3Al-ODS to Haynes 230 alloy:

Welds were made under a number of conditions, but no great difficulty was encountered in achieving a good bond. A typical macro-section is shown in Figure 9. It is evident that the Haynes 230 alloy has deformed much less then the Fe 3Al-ODS alloy, an indication of the greater high temperature strength of the nickel based alloy. Detailed examination of the bond line showed, as expected, no evidence for mixing, and in both alloys very fine grained regions were observed at the interface.

Fig. 9. Macrosection of friction weld between fine grained Fe 3Al-ODS alloy and Haynes 230
Fig. 9. Macrosection of friction weld between fine grained Fe 3Al-ODS alloy and Haynes 230

A typical; parameter trace is shown in Figure 10. Mechanical tests at room temperature all failed at the bond line, at UTS values between 636 and 720MPa. The highest strength was achieved with the highest forge force (100kN). This is somewhat below the tensile strength of the two alloys, as expected.

Fig. 10. Parameter trace for weld W25 between fine grained Fe 3Al-ODS and Haynes 230
Fig. 10. Parameter trace for weld W25 between fine grained Fe 3Al-ODS and Haynes 230

Coarse grained Fe 3

Al-ODS to Haynes 230 alloy: Only limited trials were undertaken with this combination due to shortage of material, and parameters which had given good results for the fine grained region were chosen. This gave a sound weld, although the mechanical strength was lower than for the fine-grained Fe 3Al-ODS to Haynes 320 weld. Metallographically, the features were similar to the fine grained weld, as shown in Figure 11. The parameter trace is shown in Figure 12.

Fig. 11. Macrosection of friction weld between coarse grained Fe 3Al-ODS alloy and Haynes 230
Fig. 11. Macrosection of friction weld between coarse grained Fe 3Al-ODS alloy and Haynes 230
Fig. 12. Parameter trace for weld W23 between coarse grained Fe 3Al-ODS and Haynes 230
Fig. 12. Parameter trace for weld W23 between coarse grained Fe 3Al-ODS and Haynes 230

Discussion

Weld quality:

In all the welds examined, the quality of the welds as determined by optical examination was very high. No cracks, areas of incomplete bonding or other flaws were observed. This is in accordance with experience on other ODS alloys, which have been found to respond very well to friction welding. It is also noteworthy that intermetallic alloys in general respond well to friction welding, and there are several instances of using the technique to join alloys based on Ti 3Al [5,6,7] , TiAl [8,9] and Ni 3Al [10,11] .

Weld microstructures:

The microstructures observed in the welds joining ODS alloys resemble closely those reported for MA956 and Fe 60Al 40 ODS alloys. The elongated fine or coarse grains were plastically deformed close to the bond line, and the material was recrystallised in an area extending approximately 0.5mm from the weld line. In the recrystallised region, the grains appeared equiaxed in the two-dimensional sections taken, and were slightly coarser at the bond line, presumably due to some limited grain growth associated with the slightly larger thermal cycle at this point.

Mechanical properties:

The failure of the tensile tests at the weld line in all of the welds comes as no surprise. Where the Fe 3Al-ODS was welded to itself, the bond line bond line strength must be considerably reduced by the recrystallisation process, which will drastically reduce the number of dislocations. In the generally similar MA956 ODS alloy, North et al [2,3] have reported agglomeration of the Y 2O 3 particles, suggesting that the fine oxides become bonded to nitride or carbide particles. Inkson and Threadgill [4] also reported agglomeration of the Y 2O 3 particles in friction welds in an Fe 60Al 40-ODS alloy, but no association with carbides or nitrides was observed. No transmission microscopy study has yet been carried out on the welds reported here, and so the extent and mechanism of possible Y 2O 3 agglomeration remains unknown. The effect of agglomeration will of course reduce the number of particles able to pin dislocations, and increase the interparticle spacing, and both of these effects will reduce the strength of the alloy.

It is believed that attainment of full parent material strength in friction welds in ODS alloys in this or other ODS alloys will not be achieved, as the microstructure at the bond line will be irreversibly damaged. The same phenomena will also adversely affect the high temperature tensile and creep behavior of the welds for the same reasons, and it is considered unrealistic to expect parent material properties in friction welds. However, sound welds can be made with relative ease, and the prospects for solid state processes look far better than for fusion processes, where the Y 2O 3 particles are believed to agglomerate and float out of the weld pool. This area has been explored in greater detail elsewhere. [12]

Although friction welding has great potential for joining ODS alloys, and is arguably the most suitable process, designs employing the process will have to allow for the reduction in mechanical properties.

Conclusions

This work, which is preliminary in nature, has demonstrated that the Fe 3Al-ODS alloy can be joined easily to itself and Haynes 230 alloy using continuous drive rotary friction welding, regardless of the metallurgical condition of the alloy.

Tensile failures at ambient temperature occur at the bond line due to substantial and irreversible changes to the microstructure during welding. It is expected that high temperature mechanical properties of the welded joints would also be compromised.

Acknowledgements

This work was supported by the United States Department of Energy under contract DE-AC05-96OR22464.

References

  1. Bucklow I A, Dunkerton S B and Threadgill P L: 'Investigation into the joining of creep resistant oxide dispersion strengthened alloys'. TWI Members Report 570/1996.

  2. Kang C Y, North T H and Perovic D D: 'Microstructural features of friction welded MA956 superalloy material'. Met. And Mat. Trans A, 27A, 12, 4019-4029, 1996.

  3. Shinozaki K, Kang C Y, Kim Y C, Aritoshi M, North T H and Nakao Y: 'The metallurgical and mechanical properties of ODS alloy MA956 friction welds'. Welding Journal Research Supplement, August 1997,pp289-299.

  4. Inkson B J and Threadgill P L: 'Friction Welding of FeAl40 Grade 3 ODS alloy'. Materials Science and Engineering, A258 (1998) 313-318.

  5. Threadgill P L and Baeslack W A; 'Metallurgical aspects of joining an α 2 titanium aluminide'. ProcJIMIS-6, Intermetallic Compounds, Sendai, Japan, pp1021-1025, June 1991.

  6. Juhas M J, Baeslack W A, Fraser H L, Threadgill P L, Phillips D and Broderick T F; 'Interface characteristics of solid phase welds between Ti-6Al-2Sn-4Zr-2Mo-0.1Si and Ti-14Al-21Nb titanium aluminide'.7th International Titanium Conference, San Diego, 29 June - 2 July 1992., vol 2 pp1453 - 1460. Ed. F H Froes and I L Caplan., TMMS.

  7. Juhas M J, Hou K H, Baeslack W A, Threadgill P L and Phillips D; 'Interface characteristics of solid state welds in titanium aluminides'. Proc. Processing, Properties and Applications of Metallic andCeramic Materials, Birmingham, UK, September 7-10 1992, Birmingham University.

  8. Godfrey S P, Strangwood M and Threadgill P L; 'Linear friction welding of a gamma titanium aluminide alloy'. SF2M/ONERA conference, Paris , October 1996.

  9. Smith L S and Threadgill P L: 'The physical and welding metallurgy of titanium aluminides: A review'. TWI Members Report 633/1998, April 1998.

  10. Threadgill P L; 'Joining of a nickel aluminide alloy'. Proc 4th International Conference on Trends in Welding Research, Gatlinburg, Tennessee, 5th-8th June 1995, pp317-322, ASM.

  11. Threadgill P L; 'Solid state joining of nickel aluminides: preliminary studies'. Proc 3rd ASM Paris Conference on Synthesis, Processing and Modelling of Advanced Materials, Paris, France, 25-27 June1997, pp153-158.

  12. Threadgill P L: 'Joining of ODS alloys - the state of the art'. Paper accepted for Pittsburgh Coal Conference, Pittsburgh PA, September 2000.

 

Presented with permission from Elsevier