In this work, fuel cladding tube fragments with length 50 mm and diameter 10.5 mm (cladding thickness 0.4-0.5 mm), made of 12% chromium steel EP823-Sh (12Cr–1SiMoWVNb) were used for coating deposition and further corrosion tests. The chromium coatings were obtained by electroplating using direct current with different durations of deposition process. Aqueous solution of chromium anhydride (250 g/l) and sulfuric (2.5 g/l) acid was used as an electrolyte. During the deposition process, the electrolyte was maintained in isothermal equilibrium at different temperatures. A lead wire was used as an insoluble anode, which was passed through the fuel cladding. The anode-equipped cladding was immersed in a bath with electrolyte. The thickness of the Cr layer increased linearly with increasing plating time. The temperature range for the electrolyte was chosen as 10-45 °C, and the electrodeposition rate was ~0.35 μm/min. The thickness of the obtained chromium coating on the studied samples was ~10 μm. Fig. 3 presents a sample of fuel cladding with an internal chromium coating.

Figure 3. 

Sample of fuel cladding with internal chromium coating.

The surface microstructure and cross sections of the samples were studied using an EVO 50 XVP scanning electron microscope manufactured by Carl Zeiss (Germany). SEM images were obtained in backscattered electrons during the study of microstructure. The elemental composition of the samples was analyzed by X-ray microanalysis using energy dispersive (EDS, INCA 350x-Act) and wave dispersive (WDS, INCA Wave 500) spectrometers manufactured by Oxford Instruments, combined with an EVO 50 scanning electron microscope. Measurements were performed at an accelerating voltage of 10–20 kV and a probe current of 5–50 nA.

For corrosion tests in lead and simulant fission products (CsI+Te), fuel claddings with and without chromium coating were washed in an ultrasonic bath in acetone, then in deionized water, dried and placed in an argon box. For tests in lead, C0 grade lead (Pb>99.992%) and niobium shavings as an oxygen getter were loaded into the washed samples in a ratio of 25:1, respectively. According to our estimates, the oxygen concentration in molten lead was ~10^-5 wt.%, with the initial oxygen concentration in C0 grade lead being ~10^-4 wt.%. For testing in a simulant fission products environment, an oxygen getter and a CsI+Te mixture was loaded into the washed samples. The samples were sealed, then moved to a muffle furnace and held at a temperature of 650 °C for 100 hours. After the corrosion tests were completed, the fuel claddings were cut into fragments. After cleaning from corrosive components (Pb, CsI+Te, getter), cladding fragments were submitted for metallographic analysis.

Fig. 4 presents SEM images of the cross-section microstructure of the inner surface of cladding after deposition of coatings in the temperature range 10–45 °C.

Figure 4. 

SEM images of the cross-section microstructure of the inner surface of cladding after deposition of coatings in the temperature range 10-45 °C. а. 10 °C; b. 25 °C; c. 35 °C; d. 45 °C

The coating obtained at 10 °C has poor adhesion to the substrate and flakes off (Fig. 3a). The coatings are porous and consist of submicron-sized grains. When the electrolyte temperature is increased to 25 °C (Fig. 3b), the adhesion of the coating improves significantly and no delamination occurs, the grain size on the surface increases to 1–2 µm, and pores are also observed in the coating structure. At an electrolyte temperature of 35 °C (Fig. 3c), dense chromium coatings are formed, the transverse grain size on the coating surface changes from 1 to 5 µm. A further increase in electrolyte temperature to 45 °C (Fig. 3d) results in the formation of smoother coatings, however, there are micro cracks in the coatings. The deposition rate drops by a factor of five compared to the regime at an electrolyte temperature of 35 °C. At electrolyte temperatures above 60 °C the coating on the inner surface of the sample is not deposited, the surface of the steel at the grain boundaries is etched. Thus, as the electrolyte temperature increases from 10 °C to 45 °C, the number of pores in the coatings decreases while the thickness of the chromium coating formed on the surface decreases. The structure of the coating is represented by small equiaxed grains with the size of 0.5–5 μm. The coatings obtained in the temperature range of 25–35 °C have good adhesion to the surface, without pores and cracks and they were selected for corrosion tests.

Fig. 5 shows the distribution of the main elements of 10 µm thick coating obtained in the temperature range of 25–35 °C. The distribution of chromium over the coating thickness is homogeneous. Within the sensitivity of the energy dispersive detector no impurities were detected in the chromium coating.

Figure 5. 

X-ray maps of iron and chromium distribution in the cross section of the inner surface of the chromium-coated cladding.

For ferritic-martensitic steels, the most dangerous factor in terms of corrosion interaction with fission products is Cs with the formation of localized corrosion areas of the inner surface to a depth much greater than the depth of uniform corrosion. Under normal conditions, Fe-Cr alloys are covered with a passive protective oxide film Cr₂O₃, which, according to thermodynamic data, does not react with Cs, Te or caesium tellurides. The authors of works (Sasaki 2015, 2017) propose in the case of ferritic-martensitic steels with carbide hardening the mechanism of intergranular corrosion development under the action of Cs, Te and oxygen. At the initial stage of the corrosion test at 680 °C, damage to the protective oxide film occurs because of sensitization. Corrosive compound Cs₂Te₃ would not be solely molten but would be the coexistence phase of Cs₂Te₃, Cs₂Te and corrosive agent Te as the equilibrium. Further, the corrosion agent Te reacts with the grain boundary carbides of M₂₃C₆ and with the matrix near the grain boundary to form Cr₃Te₄ according to the following equations:

4Te+2/23 Cr₂₃C₆ →Cr₃Te₄ + 18/23 C

4Te+3Cr→Cr₃Te₄

It is accepted that the corrosion enhancement and intergranular corrosion observed in oxide fuel cladding because of volatile fission products (Cs, I, Te). Ferritic-martensitic steels under normal conditions are covered by a protective oxide layer, the probability of direct dissolution of steel components along grain boundaries under the action of Te and Se should be negligible if the protective film retains its integrity. The most dangerous for this type of steel is the formation of chromates under the action of Cs, penetrating to depths much greater than the depth of uniform corrosion. Thus, one of the main tasks is to prevent the penetration of caesium into the cladding and the formation of chromates, which can be successfully achieved by creating a uniform dense protective film on the surface of the steel. The natural oxide film present on the surface of ferritic-martensitic steels is not sufficient.

Fig. 6 presented SEM images of microstructure of coated and uncoated specimens after corrosion tests in CsI+Te at 650 °C for 100 h.

Figure 6. 

SEM images of uncoated (a) and coated (b) samples after corrosion tests in CsI+Te at 650 °C for 100h.

After corrosion tests at 650 °C for 100 h, no uniform corrosion is observed on the inner surface of the uncoated fuel cladding (Fig. 6a). No signs of intergranular corrosion were found. Analysis of the microstructure of the inner surface of the initial fuel cladding after corrosion tests showed the presence of local corrosion damage with depth up to ~1–3 µm and width of ~6 µm. The presence of such corrosion damage indicates the destruction of the protective oxide film Cr₂O₃ in local areas as a result of sensitization. The deepest local corrosion damage corresponds to the areas of large (~1 μm) M₂₃C₆ carbides. If the protective oxide film Cr₂O₃ is sensitized, Te reacts both with grain-boundary carbides M₂₃C₆ and with the matrix near the grain boundary to form Cr₃Te₄, activating the development of intergranular corrosion. Thus, localized corrosion on uncoated samples, located near large carbides, are potentially dangerous locations for further development of intergranular corrosion.

After corrosion testing, no change in coating morphology was observed in the chromium-coated samples (Fig. 6b). No evidence of either uniform or intergranular corrosion was observed. Localized corrosion areas are also absent. The investigated coatings do not tend to sensitize and effectively block diffusion of fission products and oxygen into the material of the fuel cladding.

Fig. 7 presents a SEM image of the inner surface of uncoated (a) and coated (b) samples after corrosion tests in liquid lead at a temperature of 650 °C for 100 h. The oxide film has a two-layer structure: an external oxide layer with a light grey contrast with a thickness of ~0.6–1.4 µm and an internal oxide layer with a dark grey contrast with a thickness of ~150 nm (Fig. 7a). There are areas with no adhesion of the oxide film to the steel surface, the oxide film has a loose structure. EDS analysis of the film cross-section showed that the internal oxide layer with a thickness of ~150 nm is Cr₂O₃ type oxide, while the external oxide layer is a mixed iron and chromium oxide.

Figure 7. 

SEM image of uncoated (a) and coated (b) samples after corrosion tests in liquid lead at 650 °C for 100 h.

One of the key parameters affecting the oxidation rate in liquid lead is the oxygen content. Its content determines the corrosion mechanism of the material. For a temperature of 650 °C, at low oxygen concentration in the lead melt (С_[O]<10^-7 wt.%), the steel components can undergo selective dissolution, and this leads to penetration of lead deep into the material at grain boundaries. At high oxygen concentration (С_[O]=10^-5-10^-3 wt.%) there is intensive formation of oxide layer, in this case the corrosion products of EP823-Sh steel are external oxide layer of magnetite, chromium spinel of non-stoichiometric composition of Fe(Fe_1-x,Cr_x)₂O_4. The formation of a porous layer of Fe₃O₄ and plumboferrites (_nPbO*_mFe₂O₃) formed at the boundary of the outer oxide layer and lead is also observed.

Samples with internal chromium coating do not show any significant changes of microstructure after corrosion tests (Fig. 7b). The integrity of the coating is preserved over the entire cross-section of the fuel element cladding fragment. Adhesion of the coating to the base material is not disturbed, no cracks are detected. On the surface of chromium coatings, a typical picture of uniform corrosion development is observed. The formation of a dense Cr₂O₃ oxide film with a thickness of ~500–600 nm takes place. The main protective mechanism of metal coatings is the formation of protective oxide films, which do not chemically interact with lead and are resistant to dissolution. The continuous oxide film Cr₂O₃ film, impedes the penetration of reagents to each other and their growth is accompanied by self-inhibition of the process. Oxidant accumulation on the inner surface of the oxide film does not occur due to kinetic inhibition of the oxide film growth process along with inhibition at the mass transfer stage. The absence of mass transfer in the oxide layer of the chromium-coated sample leads to the fact that its thickness is ~3 times less than that of the sample without chromium coating. Fig. 8 presents the X-ray distribution maps of iron, chromium, and oxygen at the chromium coating/steel interface after corrosion tests in liquid lead.

Figure 8. 

X-ray distribution maps of iron, chromium, and oxygen in the chromium coating after corrosion tests in liquid lead at 650 °C for 100 h.

SEM-image analysis of the microstructure of the cross-section of the samples after oxidation for 100 h shows the presence of a uniform chromium layer on the steel substrate (green layer in Fig. 8). The chromium coating has a thin oxygen layer on its surface, indicating the presence of Cr₂O₃ oxide (blue layer in Fig. 8). No voids are observed between the Cr₂O₃ oxide layer and the residual chromium, which could be formed due to outward directed cationic diffusion of chromium from the coating and reverse diffusion of vacancies in the oxide-coating system. No cracking of the coating was identified. Cracks may occur due to delamination of the oxide film, as well as at the Cr/steel interface during polishing of microscopy specimens after corrosion testing. The absence of cracks indicates that the adhesion of both the coating and the oxide film on the chromium is high and this prevents deep penetration of corrosive components. However, there are pinholes in the chrome coating. They are formed in the process of electrodeposition of chromium, which can be seen in Fig. 4b. X-ray mapping shows the presence of oxygen (blue layer in Fig. 8) in these voids. It is most likely that chromium electrolyte remains in the pitting voids, which reacts with the chromium coating during the high temperature test at 650 °C to form pitting oxide inclusions. As stated above, the pitting oxides in the chrome coating during the 650 °C test for 100 h did not cause any cracking or discontinuity and therefore did not affect its corrosion resistance properties.

Corrosion tests in liquid lead at 650 °C for 100 h allow us to conclude that, under these test conditions, the chromium coating effectively blocks the diffusion of oxygen and lead into the cladding material. High corrosion resistance of fuel element cladding is necessary to exclude penetration of fuel fission products into the coolant, which may result in the spread of radioactive elements into the core and the first reactor cooling circuit. A more resistant and safer cladding, in turn, leads to longer fuel assembly life, longer fuel company duration and deeper heavy atoms burnup. The application of a 10 µm thick chromium coating does not result in a noticeable change in the geometric dimensions of the fuel element cladding. EP823-Sh steel contains ~12 wt.% of chromium, presumably chromium coating of such thickness should not significantly affect thermal-hydraulic and neutron-physical parameters of the reactor core. However, this circumstance requires a detailed study to confirm the assumption made.