1. Introduction
In recent years, carbon dioxide emissions have been limited to overcome the climate crisis caused by global warming worldwide. Thus, efforts have been made to secure energy sources that can replace coal-fired power generation. Nuclear fusion is a power generation system that produces enormous amounts of electricity using the energy of neutrons generated by nuclear fusion reactions between deuterium and tritium in the ultra-high-temperature plasma. Such nuclear fusion power generation involves a significantly lower risk of radioactive leakage compared to conventional nuclear power generation, and its fuel (deuterium) is not likely to be depleted as it is abundant in seawater. In this regard, various studies have been conducted to develop and construct fusion reactors in advanced countries, including South Korea, the United States, Japan, and the EU.
The blanket, a key component of a fusion reactor, is close to plasma and thus requires durability in high neutron irradiation and high temperature conditions. Reduced-activation ferritic/martensitic steel (RAFM steel) has been developed as the most promising candidate material for the blanket1-5). RAFM steel utilizes 9Cr ferritic heat-resistant steel, a material for conventional thermal power generation, as a basic component, but replaces the high-activation alloying elements contained in 9Cr ferritic heat-resistant steel (e.g., Mo and Nb) with low-activation alloying elements (e.g., W, V, and Ta). This alloy design is applied because Mo and Nb can be converted into highly radioactive isotopes by the neutrons generated during nuclear fusion reactions between deuterium and tritium. During the production of RAFM steel, normalizing and tempering heat treatments are performed after processing, including rolling. This allows RAFM steel to have a complex microstructure in which Cr-rich M23C6 and (Ta,V)-rich MX precipitates are distributed at the grain boundaries and grain interiors in the matrix of tempered martensite, respectively.
For the production of fusion reactors, welding is essentially required. Thus, studies on the welding characteristics of RAFM steel have also been actively conducted6-10). In general, the heat-affected zone (HAZ) of RAFM steel exhibits lower strength than the base metal, which is related to the coarsening of intergranular M23C6 precipitates in the over-tempered HAZ (OTHAZ) adjacent to the base metal6). It is reported that the weld metal generally has higher strength than the base metal and HAZ.
In this study, TIG welding was performed on two RAFM steel types with different W contents using wires that have the same composition as the base metal to investigate the effects of the W content on the microstructure and impact characteristics of the TIG welded joint of RAFM steel. To this end, two RAFM steel types were prepared by varying the W content to 1.2 and 2.0 wt.% based on the Fe-0.1C-0.5Mn-9Cr-0.2V-0.06Ta composition, and wires of the same compositions were fabricated. Post-weld heat treatment (PWHT) was performed after TIG welding. After PWHT, weld microstructures were analyzed using optical microscopy (OM), scanning electron microscopy (SEM), and transmission electron microscopy (TEM), and weld properties were assessed through Vickers hardness measurements and Charpy impact testing.
2. Experimental Method
The chemical components of the RAFM steels used in this study are summarized in Table 1. As can be seen from the table, two RAFM steel types with different W contents were prepared in this study. They were expressed as 1.2W alloy and 2W alloy according to their W contents. Fig. 1 shows the results of calculating the equilibrium phase diagrams of the two RAFM steel types using the Thermo-Calc. software. It can be seen that the ferrite region expanded at high temperatures as the W content increased. This appears to be due to the role of W, a ferrite stabilizing element.
Table 1
Chemical composition of RAFM steels investigated in this work (in wt.%)
| C | Si | Mn | Cr | W | V | Ta | Ti | N | Fe | |
|---|---|---|---|---|---|---|---|---|---|---|
| 1.2W alloy | 0.1 | 0.13 | 0.48 | 9.10 | 1.19 | 0.20 | 0.06 | 0.011 | 0.010 | Bal. |
| 2W alloy | 0.1 | 0.13 | 0.47 | 9.10 | 1.95 | 0.20 | 0.06 | 0.009 | 0.011 | Bal. |
Fig. 1
Equilibrium phase diagrams calculated by Thermo-Calc. software, (a) 1.2W alloy and (b) 2W alloy
TIG welding was performed on the two RAFM steel types in Table 1. Detailed welding conditions are summarized in Table 2. In this study, TIG welding was performed by fabricating filler wires that have the same composition as the base metal shown in Table 1. After welding, each sample was subjected to PWHT at 730°C for 60 minutes.
Table 2
Welding conditions performed in this work
| Current, A | Voltage, V | Welding speed, cm/min | Pass, # | Preheat temperature, ℃ | Interpass temperature, ℃ |
|---|---|---|---|---|---|
| 215-245 | 12-14 | 22 | 25-28 | 250-300 | 252-276 |
The microstructure of each alloy weld metal was analyzed using OM, SEM, and TEM. For SEM observation, the specimens were mechanically polished using SiC paper and then etched. In this instance, a mixture of ethanol (85 ml), hydrochloric acid (10 ml), and nitric acid (5 ml) was used as the etching solution. The mechanical properties of the weld metal were then evaluated through Vickers hardness and Charpy impact testing. Charpy impact testing was conducted at 20℃ intervals from 0℃ to -100℃.
3. Experiment Results and Discussion
3.1 Weld metal microstructure analysis results
Fig. 2 shows the weld metal microstructure of each alloy after PWHT. As can be seen from Figs. 2(a) and 2(b), the weld metal with a W content of 1.2 wt.% was solely composed of tempered martensite (TM). The weld metal with a W content of 2 wt.%, on the other hand, exhibited a large amount of δ-ferrite (δ) in addition to tempered martensite as can be seen from Figs. 2(c) and 2(d). The generation of δ-ferrite in the 2W alloy weld metal is the result attributed to the impact of W as a ferrite stabilizing element as mentioned earlier in Fig. 1. Lee et al.11) reported similar research results in a study on RAFM steel base metals.
Fig. 2
Weld metal microstructures after PWHT, (a-b) OM and SEM micrographs of 1.2W alloy weld metal and (c-d) OM and SEM micrographs of 2W alloy weld metal. TM: Tempered Martensite, δ: δ-ferrite
Fig. 3 shows the results of analyzing the precipitates distributed in the weld metal using TEM. It can be seen that coarse Cr-rich M23C6 carbides were precipitated along the prior austenite grain boundary (PAGB) and martensite lath boundary while fine (V,Ta)-rich MX particles precipitated in the lath interior. In particular, W was detected together with Cr according to the TEM-EDS analysis results for M23C6. According to the results of a previous study, the addition of W into Cr-Mo heat-resistant steels, which have similar compositions to RAFM steels, forms the precipitates of the [Fe4(Cr, W)19]C6 type as W together with Fe replaces some of Cr in Cr23C612). In other words, the addition of W into the RAFM steels used in this study also appears to have generated (Cr,W)-rich M23C6 carbides during PWHT for the TIG welded joint. Figs. 4(a) and 4(b) show the results of measuring the size distribution of the M23C6 carbides observed from the 1.2W and 2W alloy weld metals through an image analyzer. It can be seen that relatively large sizes and a high fraction of M23C6 are distributed in the 2W alloy weld metal with a higher W content. The results of a previous study reported that W contributes to maintaining the fine size of M23C6 by inhibiting the coarsening of M23C6 during long-term heat treatment12). During short PWHT as in this study, however, it is judged that an increase in W content under the same Cr content relatively increased the size and fraction of M23C6 measured after PWHT due to the increased nucleation of M23C6 as there was insufficient time for coarsening after the nucleation of M23C6.
3.2 Evaluation results for weld metal mechanical properties
Figs. 5(a) and 5(b) show the hardness profiles measured from the 1.2W and 2W alloy weld metals, respectively. As can be seen from Fig. 5(a), the hardness of the weld metal after PHWT decreased due to the tempering effect compared to immediately after welding. OTHAZ with the lowest hardness was observed from the HAZ adjacent to the base metal for both the 1.2W and 2W alloys. An interesting result in Fig. 5 is that the hardness of the 2W alloy weld metal significantly varied depending on the position regardless of PWHT. This is related to the generation of δ-ferrite mentioned earlier in Fig. 2. In other words, a large amount of δ‑ferrite is distributed in the 2W alloy weld metal unlike the 1.2W alloy weld metal. Since such δ‑ferrite has a much lower hardness value than tempered martensite, it appears that the hardness value significantly varied depending on the hardness measurement position for the 2W alloy weld metal as shown in Fig. 5(b). For reference, Fig. 6 shows the results of measuring the hardness of δ-ferrite and tempered martensite that are distributed in the 2W alloy weld metal. It can be seen that the hardness of δ-ferrite is more than 100 HV lower than that of tempered martensite.
Fig. 5
Hardness profile from weld metal (WM) to base metal (BM), (a) 1.2W alloy weld and (b) 2W alloy weld
Finally, Fig. 7 shows the impact test results of the two alloy weld metals. It can be seen that the 2W alloy weld metal exhibits significantly lower impact characteristics than the 1.2W alloy weld metal. This is related to the δ-ferrite generation and the increase in the size and fraction of M23C6 at grain boundaries caused by the increased W content as mentioned earlier in Figs. 2 and 4. In other words, the δ-ferrite generated in RAFM steel adversely affects impact characteristics as it serves as a major crack propagation path during impact testing11), and the increase in the size and content of M23C6 at grain boundaries makes the grain boundaries vulnerable, thereby degrading impact characteristics13).
4. Conclusions
In this study, two reduced-activation ferritic/martensitic steel (RAFM steel) types with different W contents were prepared and TIG welding was performed using filler wires that have the same composition as the base metal. The weld metal microstructure of each sample was analyzed and their mechanical properties were evaluated. The results of this study are as follows.
1) The weld metal with a W content of 1.2 wt.% was solely composed of tempered martensite while the weld metal with a W content of 2 wt.% exhibited a large amount of δ-ferrite in addition to tempered martensite. This is attributed to the ferrite stabilizing effect of W.
2) In the weld metal, coarse M23C6 carbides were precipitated along the prior austenite grain boundary (PAGB) and martensite lath boundary while fine MX particles were precipitated in the lath interior. In particular, the size and fraction of M23C6 carbides increased as the W content increased.
3) Over-tempered heat-affected zone (OTHAZ) with the lowest hardness was observed from the HAZ adjacent to the base metal for both the 1.2W and 2W alloy weld metals. In the case of the 2W alloy weld metal, the hardness significantly varied depending on the position regardless of post-weld heat treatment (PWHT). This is because δ-ferrite and tempered martensite with significantly different hardness values are distributed together in the 2W alloy weld metal.
4) The 2W alloy weld metal exhibited significantly lower impact characteristics than the 1.2W alloy weld metal. This results from the generation of δ-ferrite and the increased size and fraction of M23C6 at grain boundaries in the 2W alloy weld metal.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
