·Table of Contents
·Materials Characterization and testing
Experimental Study of Irradiation Damage Effect by Steam MeasurementsB. Acosta, L. Debarberis, M. Beers, C. McGirl and F. Sevini
Commission of the European Communities, Joint Research Centre
Institute For Advanced Materials, IAM,
Postbus 2, 1755 ZG Petten, The Netherlands
Tel: +31-224-565435, Fax: +31-224-561432
The MODEL ALLOY project, carried out by the JRC as 'enabling action' within the frame of the European Network AMES, is focussing on the objective of understanding the role and the influence of the content of impurities such Phosphorus, Copper, and Nickel on the mechanical properties of steels. The irradiation of the model alloys will allow determining the role of such impurities above mentioned in the steel embrittlement due to neutron irradiation.
The model alloys have been offered by Russian Research Center "Kurchatov Institute" (KI) and the project is related to the Letter of Intent for Scientific collaboration between KI and JRC Petten. The results arising from this project will provide a better understanding of the causes of embrittlement of reactor pressure vessel steels.
Parametric variation of Nickel, Copper and Phosphorus will enable to understand the relative influence of steels to irradiation embrittlement. The selected model alloys are covering a wide range of variation of content regarding the P, Cu and Ni elements. Five different Nickel maximum content levels are available; varying from ~0.004 to ~2 wt%. Copper levels range from ~0.005 to ~1 wt% and Phosphorus levels range from ~0.001 to ~ 0.039.
This paper summarises the results arisen from the application of a non-destructive technique, such is STEAM (Seebeck and Thomson Effects on Aged Materials), on the Model Alloys in both fresh and irradiated conditions. These results are correlated with the chemical composition in order to determine the role of Ni, Cu and P contents on the STEAM value. Furthermore, studies of the relationship between the shifts on Charpy test Transition Temperature and the changes on STEAM values due to neutron irradiation embrittlement, have been performed with the aim of ascertain the STEAM performance on irradiated steels.
The material tested with the STEAM technique consists on MODEL ALLOYS offered by Kurchatov Institute, which are covering a large spectrum of ferritic steels with parametric variation of Copper, Nickel and Phosphorus, known to play a significant role in material characteristic and degradation.
All the alloys followed the same heat treatment: quenching at 980-100 ° C, oil cooling, followed by tempering at 650-670 ° C, 10h, cooled in air.
For this study a set of 32 model alloys, covering a wide range of variation of content regarding the P, Cu and Ni elements, was selected. The Table 1 shows the Cu, Ni, and P contents as well as the grouping criteria used in the data analysis.
||Very low Nickel
||11 data sets
range: 0.004 ¸ 0.009 wt %
||3 data sets
||4 data sets
||7 data sets range: 1.14 ¸ 1.21 wt %
||Very High Nickel
||7 data sets
range: 1.97 ¸ 2.0 wt %
||14 data sets
range: 0.001 ¸ 0.002
plus 1 data at 0.004 wt %
||10 data sets
range: 0.006 ¸ 0.014 wt %
||8 data sets
range: 0.029 ¸ 0.039 wt %
||5 data sets
range: 0.005 ¸ 0.006 wt %
||14 data sets
range: 0.09 ¸ 0.12 wt %
||10 data sets
range: 0.39 ¸ 0.41 wt %
||Very High Copper
||3 data sets
range: 0.97 ¸ 0.99 wt %
Table 1. Grouping Criteria of the Model Alloys|
This material were machined as miniaturised Charpy specimens (3´4´27 mm) and fully characterised (Charpy impact test, hardness and STEAM) at the JRC-IAM . Afterwards model alloys were irradiated in the HFR, at Petten (The Netherlands), in the LYRA irradiation rig designed for the AMES European Network irradiation programme .
Details about the irradiation history are given in Table 2. The fluence and dpa have been so far estimated taking into account the available MCNP nuclear calculations at a nominal distance of 50 mm (trolley position) from the core box wall, the known fluence rate distribution for the pool side facility as a function of the distance from the core box wall, and the measured distance of the capsule during irradiation.
|Cycle||Beginning / End||Full Power Days||Average Temperature °C (*)||Accumulated Fluence(nm-2)||Accumulated DPA|
|99-06||01.07.99 / 26.07.99||25.25||270||3.31 ´ 1022||0.0049|
|Table 2: Irradiation History of LYRA 3|
The thermoelectric voltage generated in inhomogeneous circuits can be used to investigate metallic materials and alloys, especially in non-destructive testing. The JRC-IAM has developed prototype equipment called STEAM, Seebeck and Thomson Effects on Aged Material, to evaluate the stage of ageing of ferritic steels .
STEAM measurements have been performed on the Model Alloys in both conditions, fresh and irradiated with the aim of correlate the irradiation induced embrittlement (mainly described by the shift on the ductile to brittle transition temperature, D TT) and the change on the STEAM value due to irradiation. The Table 3 shows the values of the STEAM measurements, for the fresh and the irradiated model alloys, as well as the correspondent D TT.
The procedure of performing STEAM measurements consists in:
This procedure was repeated at least three times for each material, and for both conditions (fresh and irradiated). The sets of values obtained on each measurement were very stable, even for the irradiated material tested in a hot cell in different environmental conditions than in the laboratory. The standard deviation of the measurements vary in the range of 0.08 to 0.16 m V/° C, hence the maximum error of the measurement can be estimated at about 3 %.
As the Table 3 shows, there exists a substantial difference on STEAM values between the different alloys (for both sets, non-irradiated and irradiated), this can be related with the difference of impurities content for the alloys.
The neutron irradiation induces changes in the steel's microstructure (i.e. copper-rich precipitates, phosphide precipitates and stable matrix defects) which causes the hardening of the material and a shift of the transition temperature. Therefore, irradiation damage depends on the combination of metallurgical variables (Cu, Ni contents at first instance, and then P and Mn) and irradiation conditions (mainly fluence, irradiation temperature and flux).
For this study the irradiation parameters were the same for all the model alloys, as described above. Hence the shifts in the transition temperature and in the STEAM values can be related with the alloy's chemical composition.
The first step of the analysis is to determine weather the STEAM technique is capable to detect the material degradation due to irradiation. It is observed that in most of the alloys irradiation increases the STEAM value, so this difference between STEAM on fresh and irradiated can be correlated with the transition temperature shifts (D TT). On the basis of the major importance of Cu and Ni content on the irradiation embrittlement, the analysis was divided in sub-groups as a function of the Ni content, taking Cu and P as independent parameter, as described in the Table 1.
|Fig 1: DTT versus DSTEAM for Very Low Nickel Content|
|Fig 2: DTT versus DSTEAM for Medium Nickel Content|
In Figure 1, Figure 2 and Figure 3 the existing correlation between DTT and the STEAM value change (D STEAM) due to irradiation can be observed. This is done for the sub-groups of: very low, medium, and high nickel content on the selected alloys. For the remaining sub-groups not enough points were available to perform a regression analysis.
|Fig 3: DTT versus DSTEAM for High Nickel Content|
The second part of the analysis consists on finding out a "model" to predict the change on STEAM measurement (DSTEAM) as a function of the impurities content, for the given irradiation dose. For this study non-linear least square techniques minimising the sum of squared residuals about the model, together with residuals analysis were used. It has to be remarked that this "model" is merely a prospective analysis on the basis of the obtained results for this set of 32 model alloys, further studies on different steels, and various irradiation conditions need to be done in order to improve and corroborate the model.
The model for predicting the D STEAM can be taken as:
where DSTEAM is calculated in percentage, to obtain easy readable figures, Cu and Ni are expressed in weight %, and A is equal to 6.3.
As can be seen, there is not phosphorus contribution in the model, indeed the synergetic effects on DSTEAM of Cu and Ni results much stronger than the P effect. Possibly, the effect of P (mainly grain boundary segregation) cannot be easily observed since the STEAM technique is more powerful in detecting matrix damage.
Regarding to the irradiation parameters, i.e. fluence, irradiation temperature, and flux, as they were the same for all the studied alloys, it can be only said that the equation's first coefficient (A = 6.3) can be taken as an "irradiation factor" modelling the irradiation conditions.
|Fig 4: DSTEAM predicted by the Model versus DSTEAM calculated|
|Fig 5: Plot of residual values for the DSTEAM model|
The Figure 4 shows the fitting of the DSTEAM predicted with the observed values. The obtained R2 for the DSTEAM model is 0.91. The standard deviation of the residual values results 11.3 for this model, as can be seen in the Figure 5.
STEAM measurements have been performed on the Model Alloys with the aim of determine the shifts induced by irradiation, and correlate them with the changes in the mechanical properties, in particular with the shift in transition temperature. The capability of the STEAM method, as non-destructive technique, to detect the irradiation damage has been demonstrated.
In addition statistical studies have been performed to obtain a prediction model of the irradiation-induced shift in STEAM as from the chemical composition (i.e. carbon, copper, nickel and phosphorus contents). It has to be remarked that this "model" is merely a prospective analysis on the basis of the obtained results for this particular set of 32 model alloys, further studies on different steels, and various irradiation conditions need to be done in order to improve and corroborate the model.
The obtained model takes into account the synergetic effects of Cu and Ni, the phosphorus contribution could not be established. Possibly, the effect of P (mainly grain boundary segregation) cannot be easily observed since the STEAM technique is more powerful in detecting matrix damage.
The results arising out of this research will help to the development of the STEAM as a non-destructive technique to assess RPV embrittlement state. Further studies, like influence of the steel's electrical properties (e.g. resistivity) on the STEAM measurement, influence of the irradiation fluence and temperature are foreseen to achieve a complete development of the STEAM technique.
|© AIPnD , created by NDT.net|||Home| |Top||