The DDRs for MnMoNi steels presented in the last section provide convincing examples of the applica­tion of fundamental insight to the prediction of changes in mechanical properties of operating RPVs due to radiation damage. Mechanistic understanding is continually developing as research continues and more data are obtained. Advances may lead to modifications in the form, or the values of, the fitting parameters. The major topics are the following:

• The effect of flux

• The role of Ni, Mn, and Si

• The possibility of new mechanisms at fluences

beyond the range for which there are data in the

current surveillance databases

There are two aspects of the effect of flux: first, the prediction of embrittlement at low fluxes and second, improvements in the general description of the effect of flux on embrittlement. It was described in the previous section that recent BWR data from the SSP capsules have greatly expanded the available BWR data, leading to an improved shift model. Carter et at}2 pointed out that, although this pro­vides a better description of BWR plate data, the model still tends to underpredict the embrittlement of BWR welds for measured AT41j greater than 60 °C. This suggests that there may be further improvements necessary in the description of embrit­tlement in the low flux range. Indeed, there may be general improvements in the description of flux. Odette considers that there is a systematic flux effect in the range of 0.8-8 x 1011 n cm-2 s-1 E > 1 MeV in the IVAR database which is not predicted by the EONY model.30 Further analysis of the IVAR data­base may lead to improvements in the description of the flux dependence of embrittlement at both low (surveillance) fluxes and high (MTR) fluxes.

The DDRs for MnMoNi steels discussed in the previous section really apply to only steels with Ni < 1.3 wt%. High Ni welds have been used in a limited number of civil PWRs, notably VVER 1000 reactors. High Ni welds were selected because vessel designers wished to take benefit from the greater hardenability and superior SOL properties (com­pared to lower Ni steels). At present the response of Cu-containing high Ni steels to irradiation doses of <60mdpa can be understood in terms of the framework of the understanding developed for Cu-containing MnMoNi steels with <1.2 wt% Ni, that is, hardening from MD and solute-enriched clusters (see, e. g., the work of Williams et at}2 ). A notable difference is that there is little evidence for a plateau in hardening from CECs as the fluence increases; rather a continuous increase with fluence is observed (for fluences up to ^60 mdpa).

It is possible, however, that an additional high fluence embrittlement mechanism may operate in Ni and Mn-containing steels. Specifically, it has been suggested that at long times and or at high doses, Mn, Ni, and Si could form a new phase in RPV steel.47’130 This late-blooming phase (LBP) would produce an additional increment of hardening at high fluences, that is, late-onset embrittlement or anomalous hardening at high doses. If this is the case, then the DDRs described in the previous sec­tion may become nonconservative. Recently, there have been an increasing number of observations of MnNiSi clusters in irradiated low Cu steels (see, e. g., Auger et a/.131 and Soneda et a/.127), and there is intensive research aimed at establishing whether NiMnSi clusters represent segregation to small microstructural features (thereby lowering interfacial or strain energies) or represent precipitates of a distinct Ni-Si-Mn-enriched phase that is thermo­dynamically favored at RPV operating temperatures and RPV steel compositions. It is the latter possibility that gives rise to the possibility of a late-onset embrittlement.