The AMC26M materials were built on a steel plate following the laser-based powder-bed fusion (LPBF) additive manufacturing (AM) method using argon atomized powders purchased from Oerlikon Metco Inc. (US). The C26M powders were sieved to − 45 µm/+15 µm, which is a typical particle size distribution range for LPBF-AM process. The target composition range and the actual chemistry of the powders are listed in Table 1.
Table 1
Composition of the C26M powders in wt% used to manufacture AM specimens for corrosion testing.
Element | Fe | Cr | Al | Mo | Others |
Lot B238-240 | Bal. | 12.25 | 5.95 | 1.99 | 0.21Si, 0.06Y, 0.01C, 0.026O, 0.003N, 0.008P, < 0.001S |
Min-Max range | Bal. | 11.75–12.75 | 5.85–6.15 | 1.75–2.25 | 0.15-0.25Si, 0.03-0.07Y, 0.01C*, 0.03O*, 0.01N*, 0.009S*, 0.009P* |
*Maximum |
The elements Al, Cr, and Mo were analyzed by ICP-OES; P, Si, and Y were analyzed by ICP-MS; and O, N, C, and S by LECO methods. The actual powder composition met the specified target range from GE. LPBF-AM machine and laser process parameters were selected based on initial trials. The initial trials also helped solve two issues; (1) the initial poor powder flowability was resolved by prebaking the powder and (2) the initially observed cracks on the AM built parts were resolved by using a stage plate of similar CTE as the AM C26M built samples. In a LPBF-AM process, a part is constructed layer-by-layer by spreading a uniform powder layer (e.g., 20 to 50 µm in depth), which is subsequently fused using laser energy to create a solid layer. The mechanics of spreading involves a powder-feeder that drops a predefined amount of powder on the build plate. Then a recoating system, typically a blade that translates across the exposed powder, pushing it across the build chamber where it is deposited onto a build plate. The powder is then selectively melted by a laser beam, to form a two-dimensional slice of a three-dimensional component. Once a layer has been completed, the build plate is lowered by a specified height, and a fresh layer of powder is spread over the build-plate for the next layer. The newly melted material fuses with previously deposited layers, and the process repeats until the three-dimensional component is realized. A build plate of ferritic SS430 was used to avoid a coefficient of thermal expansion (CTE) mismatch between the wanted block blanks of AMC26M and the build plate. After the AM building process was completed, the entire structure seen in Fig. 1 was stress relieved at 650°C for 1 h in air. After this the shapes or blocks were removed from the build plate by wire EDM. Figure 1 shows the blocks or samples of AMC26M built for testing. Slices of structures marked 4, 5, and 6 in Fig. 1 were used for 12-month immersion corrosion testing reported here. The specimens were slices perpendicular to the build direction, so the largest exposed surface of each specimen was normal to the build direction. Before testing, the specimens were polished to 600 grit finish and ultrasonically cleaned in ultra-high purity (UHP) water followed by isopropyl alcohol. The AM specimens were tested alongside powder metallurgy tube specimens made of APMT2 (FA-SMT) and PMC26M having a 0.3 mm wall thickness 25,26.
The AMC26M specimens for immersion tests were discs with a typical diameter or 12.5 mm and a thickness of 1.58 mm having a total area of 11.1 cm² and an average mass of 5.552 g ± 0.020 (for nine tested discs). The PMC26M and APMT2 tube specimens had a typical length of 12 mm, an outside diameter of 10.24 mm and a wall thickness of 0.29 mm giving a surface area of 7.6 cm² for each coupon. The average mass of the PMC26M nine specimens was 0.790 ± 0.008 g and that of APMT2 specimens was 0.828 ± 0.016g.
Out-of-pile autoclave immersion tests were conducted for four periods of three months to a total immersion time of one year. Every three months the specimens were removed from the autoclaves, cleaned, and weighed, and reintroduced into the autoclaves for the next test period. The test autoclaves and operating conditions are listed in Table 2. The water recirculated in the three autoclaves at about 200 cm³/min. The high purity water is forced into the autoclave by a high-pressure pump, the pressure in each autoclave is controlled by a back pressure regulator at 10 MPa for the BWR systems (288°C) and at higher pressure for the PWR system (330°C). On exiting the autoclave, the water is cooled down to ambient temperature and continuously filtered and its conductivity is continuously monitored and maintained constant before reinjection into the autoclave.
Table 2
Out-of-pile testing conditions
Autoclave S6 | Simulated BWR Normal Water Chemistry NWC (1 ppm O2), 288°C. |
Autoclave S5 | Simulated BWR Hydrogen Water Chemistry HWC (0.3 ppm H2, < 5 ppb O2), 288°C. |
Autoclave S2 | Simulated PWR High Purity Water, 3.75 ppm H2, < 5 ppb O2, 330°C. |
Microstructural Characterization of the AMC26M specimens
Figure 2 shows the typical microstructure of the AMC26M specimens tested in the three autoclave systems (Table 2). These microstructures were obtained by sectioning the tested coupons after the 12-month immersion tests. Columnar grains 10–20 µm wide and 50–100 µm tall are observed in the AM built direction. The stress relief heat treatment did not alter the microstructure of the as-deposited AMC26M blocks. Figure 2 also shows porosity and discontinuities in the microstructure of the specimens. Typically, the porosity in AM materials could result from lack of fusion or melt boundaries and porosity developed by entrapped gas. By their appearance in Fig. 2 it is likely that the porosity originated by the lack of fusion during the fabrication process.