Page 122

Welding Journal | January 2013

Fig. 9 — Histograms of current density from indents performed within the weld nugget and at the outer edge of each sample. High current density “outliers” are linked to the presence of Cu-rich precipitates. a decrease in FSW torque or in the microstructure. A reduction in the volume fraction of the θ phase accompanied by a coarsening of the θ′ phase cannot be explained by equilibrium kinetics, which would predict the dissolution of the smaller particles and coarsening of the larger particles within a constant temperature field. Table 4 summarizes the bulk eddy current measurements. Similar readings were obtained for the 150 and 200 rev/min specimens, whereas the 300 rev/min specimen was significantly lower. It has been reported that the hardness does not have a 1-to-1 correlation with electrical conductivity in heat-treatable alloys (Ref. 45). At sufficiently high temperatures, dissolution of particles increased the amount of solidsolution solute causing a decrease in electrical conductivity. The increased solute presence results in natural aging of the weld nugget postweld thereby increasing the hardness. This hardness reversion with decreased electrical conductivity has been reported in other 2xxx series aluminum alloys (Refs. 45, 58) similar to the findings in this study. Although the combined use of eddy current and hardness testing was not generally used for identification of 2xxx series aluminum alloys (Ref. 45), it was useful for understanding the precipitate state in the difference zones of a FSWs by correlation with complementary microscale techniques. To further probe the bulk eddy current measurements, corresponding nanocurrent density measurements were calculated from indents applied in the center SZ region and the base metal region which was assumed to be near the edge of the transverse specimen. Figure 9 presents a bar graph plot showing the relative occurrence of each current density for the SZ (lighter color) and the base metal (darker color). At all rev/min conditions, a low occurrence of current densities in the range of 13–27 A/mm2 is observed only in the SZ region. Comparing Fig. 9A and B, corresponding to the 150 and 200 rev/min specimens respectively, an increase can be observed in the occurrence of the current densities in the range of 11–21 A/mm2. This increase in higher current densities for the 200 rev/min specimen corresponds to a decrease in the occurrence of the lower current densities (< 10 A/mm2). For the 300 rev/min specimen in Fig. 9C, the major occurrence of current densities is in the range of 1–5 A/mm2 with similar behavior noted for the SZ and the base metal. Very few higher current densities in the SZ are observed in the narrower range 18–22 A/mm2. To understand this variation, individual current density measurements were made directly on the Cu-rich particles and compared with the Al matrix as shown in Fig. 10. As can be observed, a higher current density range of 11–16 A/mm2 was associated with the large Cu-rich particles. with a lower current density range of 1–7 A/mm2 was associated with the matrix. Thus, the histograms can be interpreted as the 150 and 200 rev/min FSWs having a higher concentration of larger Cu-rich particles in the SZ than in the 300 rev/min FSW. This corresponded with the decrease in eddy current measurements as the volume fraction of large Cu-rich particles decreased. This was also consistent with predominant coarsening of the θ phase at lower revs/min and greater dissolution at the higher revs/min. To investigate the details of the precipitate state, TEM images were obtained as summarized in Figs. 11–13 for FSWs at 150, 200, and 300 rev/min respectively. At 150 and 200 rev/min, a mixed precipitate state was observed, which included a range of large Cu-rich precipitates that were identified as CuAl2 or θ phase. Smaller θ′ disc-shaped strengthening precipitates, ranging from 20–50 nm, were also observed in Fig. 11, which coarsen to 50 to 150 nm in Fig. 12. In Fig. 13, for the 300 rev/min specimen, a more uniform coarsening of the smaller θ′ precipitates was observed, which was also observed in the superlattice reflections in the accompanying SAD pattern of Fig. 13C due to increased volume fraction (Ref. 20). The microstructure of the 300 rev/min specimen showed almost none of the larger overaged phase or CuAl2 precipitates as compared with Figs. 11 and 12. Instead, the microstructure was similar to that of JANUARY 2013, VOL. 92 16-s WELDING RESEARCH Fig. 10 — Piezo-automation results confirming higher current density for indents placed on Cu-rich precipitates. Table 5 — FSW Temperatures Calculated Using the Alternative Heat Index Specimen Calculated Temperature (°C) 150 523 200 532 300 542 Frequency Occurence Frequency Occurence Frequency Occurence 150 RPM 200 RPM 300 RPM


Welding Journal | January 2013
To see the actual publication please follow the link above