Academia.eduAcademia.edu

Figure 8 – uploaded by Kenneth Göransson

See full PDFdownloadDownload figure
Fig. 9. STEM EDX maps for specimen exposed for 2 h. The morphology evolution during exposure is illustrated in the SEM images in Fig. 5. The specimen exposed for 2 h has nano-sized oxide particles distributed on the surface. They have a typical size of 40—50 nm, covering 2% of the surface area. There are also areas, such as inside machining grooves, where there are larger (up to 200 nm) particles present. After exposure for 24 h, crystals with blocky shape (0.5 pm) are present on the surface, with a coverage about 30%. There are also particles between the blocky crystals that are much smaller. When the exposure reaches 168 h, there is a drastic morphology change. The oxide coverage of the metal sur- face is now about 80% and the crystals have increased in size to about 1 ym. In areas not covered by large crystals, very small, often needle-shaped, particles can be observed. After exposure for 840 h, the surface is covered to practically 100% with blocky crystals. The size of these oxide crystals is only marginally larger than after exposure for 168 h. This suggests that coarsening is limited and that the crystals grow until they impinge with each other. Any subse- quent growth must then result in thickening of the crystal layer. From the XRD results it is clear that the large crystals must be The STEM/EDX maps in Fig. 9 show that the nano-particles present on the surface after exposure for 2 h consist of Ni, Fe, Ti and O, but very little Cr. Using EDX point analysis, it could be concluded that the particle is likely Ni(Fe, Ni, Ti)204 spinel with Fe/ Ni = 1.2. In the map, the thin oxide covering the entire metal sur- face cannot be resolved. The EDX maps after exposure for 24 h in Fig. 10 reveal the presence of a bilayered oxide with the outermost layer (with a thickness of 0.6 + 0.1 um at this location) made of non- stoichiometric (regarding the Fe/Ni-ratio = 1.2) NiFe2O,4 spinel crystals. The inner scale, 0.10 + 0.02 um thick here, is porous and

Figure 9 STEM EDX maps for specimen exposed for 2 h. The morphology evolution during exposure is illustrated in the SEM images in Fig. 5. The specimen exposed for 2 h has nano-sized oxide particles distributed on the surface. They have a typical size of 40—50 nm, covering 2% of the surface area. There are also areas, such as inside machining grooves, where there are larger (up to 200 nm) particles present. After exposure for 24 h, crystals with blocky shape (0.5 pm) are present on the surface, with a coverage about 30%. There are also particles between the blocky crystals that are much smaller. When the exposure reaches 168 h, there is a drastic morphology change. The oxide coverage of the metal sur- face is now about 80% and the crystals have increased in size to about 1 ym. In areas not covered by large crystals, very small, often needle-shaped, particles can be observed. After exposure for 840 h, the surface is covered to practically 100% with blocky crystals. The size of these oxide crystals is only marginally larger than after exposure for 168 h. This suggests that coarsening is limited and that the crystals grow until they impinge with each other. Any subse- quent growth must then result in thickening of the crystal layer. From the XRD results it is clear that the large crystals must be The STEM/EDX maps in Fig. 9 show that the nano-particles present on the surface after exposure for 2 h consist of Ni, Fe, Ti and O, but very little Cr. Using EDX point analysis, it could be concluded that the particle is likely Ni(Fe, Ni, Ti)204 spinel with Fe/ Ni = 1.2. In the map, the thin oxide covering the entire metal sur- face cannot be resolved. The EDX maps after exposure for 24 h in Fig. 10 reveal the presence of a bilayered oxide with the outermost layer (with a thickness of 0.6 + 0.1 um at this location) made of non- stoichiometric (regarding the Fe/Ni-ratio = 1.2) NiFe2O,4 spinel crystals. The inner scale, 0.10 + 0.02 um thick here, is porous and

Related Figures (18)

Chemical composition of the investigated material in wt% and at.%. Table 1
Chemical composition of the investigated material in wt% and at.%. Table 1
Fig. 1. Autoclave loop schematic. The loop system consists of two consecutive auto- claves, a main pump, a heat exchanger, a preheater, a H202-injection pump, a cooler, sensors for pressure, temperature, oxygen and conductivity and other control equip- ment including a data acquisition system. grain boundary attacks. Afterward, the specimens were exposed in an autoclave for four different times: 2, 24, 168 and 840 h. A BWR environment was simulated by continuously adding 500 ppb hydrogen peroxide to the water at 286 °C, under 80 bar. A sche- matic of the autoclave system is shown in Fig. 1. The loop system consists of two consecutive autoclaves, a main pump, a heat exchanger, a preheater, a H202-injection pump, a cooler, sensors for pressure, temperature, oxygen and conductivity and other control equipment including a data acquisition system. The autoclave and the hot part of the loop are made of titanium alloys, to prevent contamination. Degassed ultra-pure water is pumped in the auto- clave pipeline and then heated up to 286 °C, at this stage it is then mixed with the H202-solution before entering into the test sections (Autoclave 1 and Autoclave 2 in Fig. 1). In each autoclave, several test coupons are mounted in sample assemblies, which are con- nected in series. In Fig. 2 the specimen assembly is shown. During autoclave exposure, the pressurized water passes through a jet nozzle section and generates a water jet of high flow velocity (10 m/ s in this study) at the nozzle mouth. The water jet impinges on the central region of the test coupon surface. It generates a high flow at the coupon surface facing the jet. The water flows pass the pe- riphery and back side of the coupon and enters the flow channels in the sample assembly and finally arrives at the next sample as- sembly. The exposure system is the same as in former work [1], [6], [9], [10]. The samples were dried and weighed with an analytica balance having 0.1 mg precision before and after each exposure. The surface of the specimens had uniform color, indicating that spallation was limited and that the adhesion of the oxide was good during exposure (at 286 °C). After cool-down it was observed that spallation could occur, and also during FIB-preparation the oxide sometimes cracked. With longer exposure time the color got gradually darker. The thickness of the material was 0.30 mm and the area facing the water jet was 20 x 20 mm”.
Fig. 1. Autoclave loop schematic. The loop system consists of two consecutive auto- claves, a main pump, a heat exchanger, a preheater, a H202-injection pump, a cooler, sensors for pressure, temperature, oxygen and conductivity and other control equip- ment including a data acquisition system. grain boundary attacks. Afterward, the specimens were exposed in an autoclave for four different times: 2, 24, 168 and 840 h. A BWR environment was simulated by continuously adding 500 ppb hydrogen peroxide to the water at 286 °C, under 80 bar. A sche- matic of the autoclave system is shown in Fig. 1. The loop system consists of two consecutive autoclaves, a main pump, a heat exchanger, a preheater, a H202-injection pump, a cooler, sensors for pressure, temperature, oxygen and conductivity and other control equipment including a data acquisition system. The autoclave and the hot part of the loop are made of titanium alloys, to prevent contamination. Degassed ultra-pure water is pumped in the auto- clave pipeline and then heated up to 286 °C, at this stage it is then mixed with the H202-solution before entering into the test sections (Autoclave 1 and Autoclave 2 in Fig. 1). In each autoclave, several test coupons are mounted in sample assemblies, which are con- nected in series. In Fig. 2 the specimen assembly is shown. During autoclave exposure, the pressurized water passes through a jet nozzle section and generates a water jet of high flow velocity (10 m/ s in this study) at the nozzle mouth. The water jet impinges on the central region of the test coupon surface. It generates a high flow at the coupon surface facing the jet. The water flows pass the pe- riphery and back side of the coupon and enters the flow channels in the sample assembly and finally arrives at the next sample as- sembly. The exposure system is the same as in former work [1], [6], [9], [10]. The samples were dried and weighed with an analytica balance having 0.1 mg precision before and after each exposure. The surface of the specimens had uniform color, indicating that spallation was limited and that the adhesion of the oxide was good during exposure (at 286 °C). After cool-down it was observed that spallation could occur, and also during FIB-preparation the oxide sometimes cracked. With longer exposure time the color got gradually darker. The thickness of the material was 0.30 mm and the area facing the water jet was 20 x 20 mm”.
To investigate the phases present on the specimens, grazing incidence X-ray diffraction (XRD) was used. The XRD measure- ments were carried out on the sample surface at ambient tem- perature using a Siemens D5000 diffractometer in a grazing
To investigate the phases present on the specimens, grazing incidence X-ray diffraction (XRD) was used. The XRD measure- ments were carried out on the sample surface at ambient tem- perature using a Siemens D5000 diffractometer in a grazing
Fig. 4. XRD analysis of exposed samples. The only phase detected in addition to the substrate is the spinel NiFe20, phase. Fig. 3. Mass change as a function of exposure time.
Fig. 4. XRD analysis of exposed samples. The only phase detected in addition to the substrate is the spinel NiFe20, phase. Fig. 3. Mass change as a function of exposure time.
Fig. 2. The specimen assembly consists of a nozzle (titanium cylinder where a hole is drilled along the central line), tube spacers, test coupon and the test coupon holder. The water jet is re-created after it has impinged the coupons.
Fig. 2. The specimen assembly consists of a nozzle (titanium cylinder where a hole is drilled along the central line), tube spacers, test coupon and the test coupon holder. The water jet is re-created after it has impinged the coupons.
Fig. 5. SEM top-view images of samples exposed for different times. a) 2 h, b) 24 h, c) 168 h and d) 840 h. incidence set-up (7.5° theta) in detector scan mode with a Gobel mirror and an energy dispersive detector. In order to investigate the oxide morphology, top-view scanning electron microscopy (SEM) images were recorded with a Leo Ultra 55 FEG-SEM. An FEI Versa 3D FIB/SEM was used to produce cross-sections for SEM investi- gation and thin foil lift-out specimens for transmission electron microscopy (TEM). Imaging and energy dispersive X-ray (EDX) analyses were carried out with an FEI TEM Tecnai microscope T20 The measured mass change as a function of exposure time is
Fig. 5. SEM top-view images of samples exposed for different times. a) 2 h, b) 24 h, c) 168 h and d) 840 h. incidence set-up (7.5° theta) in detector scan mode with a Gobel mirror and an energy dispersive detector. In order to investigate the oxide morphology, top-view scanning electron microscopy (SEM) images were recorded with a Leo Ultra 55 FEG-SEM. An FEI Versa 3D FIB/SEM was used to produce cross-sections for SEM investi- gation and thin foil lift-out specimens for transmission electron microscopy (TEM). Imaging and energy dispersive X-ray (EDX) analyses were carried out with an FEI TEM Tecnai microscope T20 The measured mass change as a function of exposure time is
Fig. 6. SEM cross-section images of samples exposed for different times. a) 2 h, b) 24 h, c) 168 h and d) 840 h.
Fig. 6. SEM cross-section images of samples exposed for different times. a) 2 h, b) 24 h, c) 168 h and d) 840 h.
Fig. 7. TEM bright field images of sample exposed for different times. a) 2 h, b) 24 h, c) 168 h and d) 840 h. The following features are present; spinel crystals, nano-grained “NiO’ Ti-rich needle shaped oxides. In Fig. 4, XRD traces of specimens (top-surface) exposed for 2 h, 24 h and 168 h are presented. The only phase, apart from the fcc substrate, that was observed is NiFe2O, spinel that is identified presented in Fig. 3. The specimens suffer mass loss, which means that more mass has been lost because of metal ion dissolution than what has been gained by the incorporation of oxygen in the oxides being formed, as no signs of spallation were observed (before
Fig. 7. TEM bright field images of sample exposed for different times. a) 2 h, b) 24 h, c) 168 h and d) 840 h. The following features are present; spinel crystals, nano-grained “NiO’ Ti-rich needle shaped oxides. In Fig. 4, XRD traces of specimens (top-surface) exposed for 2 h, 24 h and 168 h are presented. The only phase, apart from the fcc substrate, that was observed is NiFe2O, spinel that is identified presented in Fig. 3. The specimens suffer mass loss, which means that more mass has been lost because of metal ion dissolution than what has been gained by the incorporation of oxygen in the oxides being formed, as no signs of spallation were observed (before
Fig. 8. HAADF STEM image of a nanosized (about 50 nm) spinel crystal after 2 h exposure.
Fig. 8. HAADF STEM image of a nanosized (about 50 nm) spinel crystal after 2 h exposure.
Fig. 10. STEM EDX maps for specimen exposed for 24 h. The sub-surface nodules are titanium carbonitrides formed during processing, present in the material before the exposure.
Fig. 10. STEM EDX maps for specimen exposed for 24 h. The sub-surface nodules are titanium carbonitrides formed during processing, present in the material before the exposure.
Fig. 12. STEM EDX maps for specimen exposed for 168 h, excluding oxygen in the quantification to reveal the relative changes in the metallic content. Fig. 11. STEM EDX maps for specimen exposed for 168 h. exposure the specimen shows a double spinel layer. In the outer spinel layer (pos. 1 in Fig. 14) the Fe/Ni ratio is very close to the stoichiometric value of 2, implying that the composition is close to the ideal NiFe2O.. In the inner spinel layer (pos. 2 in Fig. 14) the contains small grains. This layer is depleted in Fe and enriched (relative to the metal) in Cr (27 at.% of the metallic content) and Ti, but the main metallic element is Ni (70 at.% of the metallic content). The specimen exposed for 168 h shows three different oxide layers, see Fig. 11: the outer part, 0.8 + 0.2 um thick, is made of large spinels of Ni(Fe, Ni, Ti)zO4, (Ni 44 at.%, Fe 46 at.%, Ti 10 at.%) the middle layer, 0.6 + 0.2 jm in thickness, is very porous and contains Ni-—Ti enriched needle-like structures and the fine-grained inner scale of 0.20 + 0.05 um in thickness is enriched in Cr and Ti (but the main element is still Ni). In Fig. 12, EDX maps from the same location as in Fig. 11 are presented, but oxygen was not included in the quantification, giving an idea of the relative concentration of the metallic elements. Again it can be seen that the inner oxide layer is enriched in Cr and Ti, and depleted in Fe and Ni, compared to the overall metal composition. An EDX line-scan across the inner layer is further shown in Fig. 13. In this layer, the concentration of Ni and Fe decreases more, in relative terms, than the concentration of Cr and Ti, compared with the concentrations in the metal. Still, Ni is the dominating element. It is possible that this layer is mainly NiO, as suggested in Ref. [1], or rather (Ni, Cr, Fe, Ti)O. However, the layer is nano-grained and porous, so it is possible that the layer consists of a mixture of different fine-grained oxides (M,Oy). This would be expected considering the low solubility of Cr in NiO [12]. CTCAAICDNIY manc af tha cnariman avnncad far QAN hh ara chniarmn
Fig. 12. STEM EDX maps for specimen exposed for 168 h, excluding oxygen in the quantification to reveal the relative changes in the metallic content. Fig. 11. STEM EDX maps for specimen exposed for 168 h. exposure the specimen shows a double spinel layer. In the outer spinel layer (pos. 1 in Fig. 14) the Fe/Ni ratio is very close to the stoichiometric value of 2, implying that the composition is close to the ideal NiFe2O.. In the inner spinel layer (pos. 2 in Fig. 14) the contains small grains. This layer is depleted in Fe and enriched (relative to the metal) in Cr (27 at.% of the metallic content) and Ti, but the main metallic element is Ni (70 at.% of the metallic content). The specimen exposed for 168 h shows three different oxide layers, see Fig. 11: the outer part, 0.8 + 0.2 um thick, is made of large spinels of Ni(Fe, Ni, Ti)zO4, (Ni 44 at.%, Fe 46 at.%, Ti 10 at.%) the middle layer, 0.6 + 0.2 jm in thickness, is very porous and contains Ni-—Ti enriched needle-like structures and the fine-grained inner scale of 0.20 + 0.05 um in thickness is enriched in Cr and Ti (but the main element is still Ni). In Fig. 12, EDX maps from the same location as in Fig. 11 are presented, but oxygen was not included in the quantification, giving an idea of the relative concentration of the metallic elements. Again it can be seen that the inner oxide layer is enriched in Cr and Ti, and depleted in Fe and Ni, compared to the overall metal composition. An EDX line-scan across the inner layer is further shown in Fig. 13. In this layer, the concentration of Ni and Fe decreases more, in relative terms, than the concentration of Cr and Ti, compared with the concentrations in the metal. Still, Ni is the dominating element. It is possible that this layer is mainly NiO, as suggested in Ref. [1], or rather (Ni, Cr, Fe, Ti)O. However, the layer is nano-grained and porous, so it is possible that the layer consists of a mixture of different fine-grained oxides (M,Oy). This would be expected considering the low solubility of Cr in NiO [12]. CTCAAICDNIY manc af tha cnariman avnncad far QAN hh ara chniarmn
-STEM/EDX maps of the specimen exposed for 840 h are shown in Fig. 14 and quantitative results are presented in Table 2. After this Fig. 13. STEM EDX line scan in the specimen exposed for 168 h. The scan direction is from bottom to top (from metal to oxide).
-STEM/EDX maps of the specimen exposed for 840 h are shown in Fig. 14 and quantitative results are presented in Table 2. After this Fig. 13. STEM EDX line scan in the specimen exposed for 168 h. The scan direction is from bottom to top (from metal to oxide).
Fig. 14. STEM EDX maps for specimen exposed for 840 h. mass is decreasing although oxygen is being incorporated and because there are no signs of spallation of the samples. In order to estimate the amount of metal dissolution, equation (1) from Ref. | 1] [6] [10] was used.
Fig. 14. STEM EDX maps for specimen exposed for 840 h. mass is decreasing although oxygen is being incorporated and because there are no signs of spallation of the samples. In order to estimate the amount of metal dissolution, equation (1) from Ref. | 1] [6] [10] was used.
ratio is about 1.2 as for the shorter exposure times. The Ti content is much higher in the inner spinel layer (8.5 at.% of the metallic content) than in the outer spinel layer (2.8 at.% of the metallic content). The interface between the inner and outer spinel layer is strikingly sharp, as seen in the Fe-map of Fig. 14. Below the double spinel layer a large-grained nickel-rich oxide (presumably NiO) layer is observed (pos. 3 in Fig. 14), that was not present after exposure for 168 h, where some Ti rich-needles are embedded. The inner oxide layer (positions 4 and 5 in Fig. 14) consists of very fine grains and the layer seems to be porous, as observed also for the shorter exposure times. Its composition (metallic part) is similar to the composition of the base metal. A diffraction pattern (Fig. 15) from this area shows that nanosized grains of NiO are present. It is also worth noting that in the innermost oxide, the Cr content is very low (pos. 5 in Fig. 14). This is in contrast to what is observed for the shorter exposure times.
ratio is about 1.2 as for the shorter exposure times. The Ti content is much higher in the inner spinel layer (8.5 at.% of the metallic content) than in the outer spinel layer (2.8 at.% of the metallic content). The interface between the inner and outer spinel layer is strikingly sharp, as seen in the Fe-map of Fig. 14. Below the double spinel layer a large-grained nickel-rich oxide (presumably NiO) layer is observed (pos. 3 in Fig. 14), that was not present after exposure for 168 h, where some Ti rich-needles are embedded. The inner oxide layer (positions 4 and 5 in Fig. 14) consists of very fine grains and the layer seems to be porous, as observed also for the shorter exposure times. Its composition (metallic part) is similar to the composition of the base metal. A diffraction pattern (Fig. 15) from this area shows that nanosized grains of NiO are present. It is also worth noting that in the innermost oxide, the Cr content is very low (pos. 5 in Fig. 14). This is in contrast to what is observed for the shorter exposure times.
Fig. 15. Diffraction pattern of nano-crystalline NiO from the region 4 in Fig. 14. EDX results in at.% (metallic part) obtained from the regions marked in Fig. 14. Table 2
Fig. 15. Diffraction pattern of nano-crystalline NiO from the region 4 in Fig. 14. EDX results in at.% (metallic part) obtained from the regions marked in Fig. 14. Table 2
Fig. 17. Proposed mechanism of oxide formation. NiFe20, spinel crystals (black), inner oxide (Ni, Cr)O layer (grey line) and Ti-rich oxide needles (red), NiO layer (grey square), innermost oxide (grey circles). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 17. Proposed mechanism of oxide formation. NiFe20, spinel crystals (black), inner oxide (Ni, Cr)O layer (grey line) and Ti-rich oxide needles (red), NiO layer (grey square), innermost oxide (grey circles). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 16. a) Oxide thickness for each exposure time measured from TEM images. b) Calculated dissolved metal thickness (h dissolved) for each exposure time.
Fig. 16. a) Oxide thickness for each exposure time measured from TEM images. b) Calculated dissolved metal thickness (h dissolved) for each exposure time.
Related topics:
Materials EngineeringMaterials ScienceNuclear Materials
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved