The progressive precipitation of the Cu present in solution at the SOL as a result of irradiation has been recognized since the 1980s. There are a large number of studies that show that a high density of small CECs are formed under irradiation, and that the number density, size, and volume fraction are strongly depen­dent on the irradiation fluence, flux, and the material composition.

For MnMoNi steels and Fe-Cu alloys, the number density and volume fraction increase rapidly with increasing fluence, and then there is an appearance of saturation, that is, a pattern of behavior that mir­rors the shape of the curve in Figure 6(a). This is illustrated in Figure 8 from the work of Odette et al. (see data presented in Eason eta/.29 where the volume fraction, fp, and radius have been derived from SANS data). It can be seen that the radius increases with fluence in this example. Auger et a/.62 had found a similar pattern of behavior in SANS and AP data from ten steels and two Fe-Cu alloys with <0.2 wt% Cu. Saturation occurred at a similar fluence to that of Odette et a/., that is, ~1 x 1019 n cm~2 (for irradiation temperatures close to 290 °C).

The rate at which the volume fraction increases with fluence is also strongly dependent on the irradi­ation flux and the composition of other elements such as Ni. Figure 9 shows the SANS measurements of Williams and Phythian63 on MnMoNi SAWs. It shows the effect of dose rate, Cu, and Ni on the development of CEC volume fractions with dose. Decreasing the Cu decreases both the volume fraction and CEC size. In the high Cu welds (seen most clearly at high dose rate), increasing the Ni clearly increases the total volume fraction of CECs formed at all doses. At the same time, the mean CEC diameter is somewhat decreased in the higher Ni weld (thus the volume fraction increase is associated with a large increase in cluster number density). It is also evident that, while the precipitated volume fraction appears to be satur­ating at the highest dose in the low Ni welds, there is no sign that saturation is close in the high Ni welds.

A number of authors have found similar results.29,64,65

Figure 9 Effect of bulk composition and dose rate on CEC size and volume fraction in MnMoNi SAWs. In (a) data are from low Ni welds, high Cu ~0.08Ni, 0.55Cu; low Ni, low Cu ~0.08Ni, 0.15Cu; and in (b) data are from high Ni welds, high Cu ~1.65Ni, 0.55Cu; high Ni, low Cu ~1.5Ni, 0.05Cu (all wt%). In both figures could be seen high flux ~5.5 x 10~9, medium flux ~6.3 x 10~10, and low flux ~9 x 10-11 (all dpas-1). Reprinted with permission from Williams, T. J.; Phythian, W. J. In Effects of Radiation on Materials, 17th International Symposium, ASTM STP 1270; Gelles, D. S., Nanstad, R. K., Kumar, A. S., Little, E. A., Eds.; American Society for Testing and Materials: Philadelphia, 1996; p 191. Copyright ASTM International.

It is to be noted that Soneda65 found that when com­paring Cu clusters in low-fluence-irradiated steels formed at a flux of 109 and 1010ncm_2s_1 E > 1 MeV, there was significant effect of dose rate on cluster size rather than number density.

The effect of Ni was established early on,66 but it has only just been established that Mn also has an important effect. Odette et al. demonstrated from SANS studies that at constant Cu and Ni, increasing Mn decreased the size of the clusters but increased their number density as illustrated for 0.4 wt% Cu, 0.8 wt% Ni, and 3.4 x 1019 E > 1 MeV in Figure 10. In higher Ni steels, Burke et a/.67 have also demon­strated that removing Mn from a steel significantly lowers the resultant embrittlement and the level of observable solute-related damage.

It was thought for many years (e. g., Jones and Bolton2) that CEC-related hardening would reach a maximum level once all the Cu had precipitated, and remain at this level as the CEC size remained constant — probably as a result of a balance between cluster nucleation and growth, and cluster destruction in cascades, that is, overaging did not occur. Jones and Bolton2 reported measurements of Cu cluster diame­ter using SANS on unirradiated and irradiated C-Mn SMA welds. It was shown that under surveillance conditions, and at temperatures below about 300 °C, Cu clusters grew to about 2 nm in diameter. Even after subsequent accelerated irradiations (of the sur­veillance samples irradiated to the lowest doses) to doses between ^200 x 10~5 and 1200 x 10~5dpa, the mean precipitate diameter was still ^2 nm.

Figure 10 SANS data on cluster radius, rp, number density, Np, and volume fraction, fp for 0.4 wt% Cu split melt model steels irradiated at high IVAR flux at 290 °C. Effects on Mn variations in alloys with ~0.8 wt% Ni. Reproduced from Eason, E. D.; Odette, G. R.; Nanstad, R. K.; Yamamoto, T.; EricksonKirk, M. T. A Physically Based Correlation of Irradiation-Induced Transition Temperature Shifts For RPV Steels; Oak Ridge Report ORNL/TM-2006/530, 2007.

More recent observations at the onset of the pla­teau in CEC formation in a number of commercial steels have shown that only around half of the avail­able Cu had precipitated (Auger etal.)62 This suggests that particle coarsening and overaging could occur in irradiated RPV steels as well as in thermally aged steels, once the precipitation level was high. Particle coarsening has been reported in MnMoNi surveillance material from the Doel-1 and Doel-2 reactors,68,69 as shown in Figure 11. This particle coarsening should result in overaging (i. e., softening) after the hardness reaches a maximum value.

Figure 12 illustrates schematically the influence of selected material and irradiation variables on the

CEC volume fraction (nominally for a Cu-containing MnMoNi steel irradiated at ^290 °C).

The Cu level in the matrix at the SOL is an important parameter as it is the matrix Cu that is available for precipitation of CECs. This has been determined by either thermodynamic modeling or by direct measurement. For example, modeling calcula­tions were performed by Buswell and Jones70 to determine matrix Cu levels in Magnox SMA welds with bulk Cu contents between 0.13 and 0.31 wt%. They found that the precipitation of Cu during the final weld stress-relief heat treatment (at 590-600 ° C), and also during the subsequent extremely slow cooling (5 °Ch-1) of the RPV before reactor operation, reduced the maximum Cu available to precipitate during irradiation to no more than 0.15-0.20 wt%.70 Precipitation during the final weld stress relief also occurs in US steels. Indeed, a consensus has emerged that there is an upper limit to SOL-dissolved Cu, which is dependent on heat treatment. McElroy and

Lowe have shown that even differences of 20-30 °C in the heat treatment temperature can markedly affect the dissolved Cu content,71 as can the slow cooling of real structures from the stress-relief temperature.