Decommissioning of nuclear reactors involves dismantling of the reactor components and the relocation of spent fuel to repository sites for final disposal. In case of the absence of a geological repository site, it is commonly accepted that dry storage inside nuclear power plants or at selected concentrated sites as an interim solution to handling spent fuel has been considered as an appropriate option (Wang et al., 2020). During the period of dry storage, metal canisters, sometimes with additional concrete or metal shields, are used to accommodate and preserve spent nuclear fuel for up to 40 years, prior to reprocessing or final disposal in a geological repository. Austenitic stainless steels are widely used as the substrate materials for dry storage canisters owing to its excellent corrosion resistance, thermal conductivity, and mechanical properties. The foregoing properties make stainless steels a desirable material for applications in various industrial plants, from food storage to solar power generation (Günen et al., 2014a). It is important to note that some of the dry storage sites, such as the nuclear power plants, may be located at areas close to seashore and the metallic materials are thus more prone to salt deposition under highly humid environments in these areas. The deposited salt could induce localized stress corrosion cracking (SCC) near the welded regions of the metallic materials and could eventually lead to structural failure of the canisters. Serious failures of the canisters accompanying with through-wall cracks could lead to the release of radionuclides toward the environment. Therefore, adequate corrosion protection measures are required for the canister materials such as stainless steels (Yeom et al., 2020). Researchers also attempted to enhance the corrosion resistance of stainless steels by surface modification techniques such as boronizing and nitriding (Günen et al., 2014b; Çetin et al., 2021). However, these techniques usually lack flexibility and require abundant efforts that make them almost impossible for large scale application.

During the past decade, cathodic protections for corrosion mitigation in stainless steels by specific metal-oxide coatings such as TiO₂ (Yeh et al., 2013), ZnO₂ (Techapiesancharoenkij et al., 2017), or ZrO₂ (Yeh et al., 2008) irradiated with ultraviolet (UV) have drawn considerable interests. Among them, TiO₂ coatings are more popular due to their inherent advantages such as superior photoelectrical properties and long-term stability in corrosive environments (Bu and Ao, 2017). Under excitation in the presence of UV illumination, TiO₂ coatings can generate electron and hole pairs in the space charge layer. The generated photoelectrons are then transferred to the stainless steel surface that is exposed to a corrosive environment, making its open circuit potential more active than its original corrosion potential. The holes on the other hand would move toward the electrolyte leading to the oxidation of H₂O (Yuan et al., 1994). In principle, the TiO₂ coating actually acts as a non-sacrificial anode for photo-cathodic protection against corrosion in stainless steels. Efforts were devoted to the application of TiO₂ coating over stainless steel substrates for corrosion mitigation, and positive results were reported (Fujisawa and Tsujikawa, 1994; Imokawa et al., 1994). However, due to the wide band gap energy of TiO₂ and the recombination of electron and hole pairs in the absence of UV illumination, practical applications of TiO₂ coatings have not been widely accepted (Shen et al., 2005). Other approaches have been made on metal oxides that also bear semiconductor property and have similar band gap energies compared to TiO₂. Investigated examples included Fe₂O₃ (Liu et al., 2014), SnO₂ (Subasri and Shinohara, 2003), CeO₂ (Yang and Cheng (2017), and WO₃ (Tatsuma et al., 2001) that all acted as photo-anodes and were separately coupled with stainless steels for corrosion mitigation. This type of photo-cathodic protection would require a relatively complicated system setup and would need to deal with the degradation of organic substances in the photo-anodic half-cell, leading to increasing production costs for large-scale industrial applications. These shortcomings may simply be overcome by the use of TiO₂ coatings directly applied over stainless steels. Several techniques such as sol–gel coating, chemical vapor deposition, sputtering, spray pyrolysis, anodization, and plasma spraying for TiO₂ deposition are currently available (Xu et al., 2021) (Erol et al., 2014). Among them, the sol–gel method offers many advantages in terms of coating homogeneity, well-controlled TiO₂ phase structure, and good adherence over larger area. On the other hand, the factors that influence the functionality of these TiO₂ coatings over stainless steels, especially low carbon Type 304 stainless steel (304LSS), have not yet been investigated in detail. For example, TiO₂ coatings prepared by the sol–gel method usually require a thermal treatment at 400–500 C to achieve a desirable photocatalytic effect under UV illumination. However, such a high-temperature treatment to TiO₂ coatings would also alter the microstructure of 304LSS surfaces, which could eventually lead to a poorer photocatalytic activity of these coatings (Konishi and Tsujikawa, 1997).

The undesirable outcomes of worsened catalysis in TiO₂ coatings on the 304LSS substrate during high-temperature treatments have to be addressed properly before such coatings being adopted for large-scale applications. In this study, we intend not only to retain the catalytic efficiency of plain TiO₂ coatings but also to prolong the cathodic protection period of the coatings even after the shutoff of UV illumination, by precisely controlling the diffusion of Fe ions from the 304LSS substrate into TiO₂ coatings during annealing treatments. For a precise control over the diffusion of Fe ions, a number of parameters such as annealing temperature, annealing time, and coating thickness were considered and tested. In addition, for comparison purposes, the influences of Fe ions on the photocatalytic activity and the cathodic protection period of Fe-doped plain TiO₂ coatings were investigated on indium-doped tin oxide (ITO) glass substrates. An external dopant such as ferric nitrate nanohydrate was used as a prime source of Fe in these TiO₂ coatings.

Multi-layered TiO₂ coatings were prepared by dipping the 304LSS substrates in TiO₂-containing sol–gel solutions for a number of times, followed by appropriate annealing treatments at various temperatures for different time durations.

Electrochemical polarization analyses were conducted to evaluate the corrosion behavior of the 304LSS substrates exposed to an aqueous alkaline solution without and with the TiO₂ coatings in the absence and presence of UV illumination. Surface morphologies of the substrate samples were studied by scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDX). Crystalline characteristics of TiO₂ coatings were evaluated by an X-ray diffraction (XRD) pattern. Depth profile analysis by secondary ion mass spectroscopy (SIMS) was conducted to quantify the amount of Fe ions having diffused into the TiO₂ coatings from the substrate materials.

TiO₂ coatings and coatings doped with Fe ions play important roles on both enhanced photocatalytic activity and prolonged cathodic protection for improving the corrosion resistance of coated 304LSS substrates exposed to corrosive environments. It is anticipated that this particular coating treatment could lead to a more effective reduction in the corrosion rate of the coated substrate in the presence of UV illumination and to a prolonged low corrosion rate even after the UV illumination has been switched off. During the consecutive annealing treatments for the inner anatase-type TiO₂ coating and the outer amorphous TiO₂ coating over the 304LSS substrate, it is expected that a considerable amount of Fe element present in the substrate would diffuse into the inner-coating layer and form Ti–O–Fe bonds, with the simultaneous formation of α-Fe₂O₃ and Fe₃O₄ in the interface between the substrate and the coating layer (Zhu et al., 2001; Yeh et al., 2013). In addition to Fe, Cr, and Ni would also diffuse into the TiO₂ coating, but Fe tends to react more readily with TiO₂ owing to its inherent tendency toward oxygen (Maruska and Ghosh, 1979). Furthermore, the diffusion coefficient of Fe is greater than that of Cr, which makes α-Fe₂O₃ and Fe₃O₄ to be predominant in the interfacial oxide layer. Although major ions in the interfacial oxide actually consists of both Fe²⁺ and Fe³⁺ ions, most Fe elements are present in the oxidation state of Fe³⁺ rather than Fe²⁺ since conversion from Fe²⁺ to Fe³⁺ would usually take place during planned calcination treatments (Yu et al., 2006). It is well known that Fe₂O₃ and TiO₂ are both n-type semiconductors with two different band gap energies of 2.2 and 3.2 eV, respectively. Moreover the conduction band energy of α-Fe₂O₃ is lower than that of TiO₂ by outer 0.38 eV. Hence, the n–n type heterojunction formed during the calcination treatment allows the photo-excited electrons to migrate easily from the outer and inner TiO₂ coatings to the substrate (Huang et al., 1997).

In this study, the key to achieving a better photocatalytic activity and a prolonged cathodic protection is the preparation of multi-layered TiO₂ coatings via the sol–gel-based dipping process, followed by calcination treatments at various temperatures. Suggested multi-layered TiO₂ coatings depositing over the 304LSS substrate consist of two outer coatings of amorphous TiO₂ and one inner anatase-type TiO₂ coating underneath the amorphous coatings. It is presumed that this inner anatase-type TiO₂ layer, after accommodating the Fe²⁺ and Fe³⁺ ions diffusing from the interfacial oxide during the calcination treatment, would act as a free electron donor under UV illumination. In the meantime, this specifically engineered Fe-doped layer would also become a charge storage container to trap the free electrons produced on the site and from the outer amorphous TiO₂ layers, making the substrate’s electrochemical corrosion potential shifted to a more active value. The two outer amorphous TiO₂ layers also serve as a barrier for preventing the stored electrons from escaping from the inner TiO₂ layer once the UV illumination is switched off.

In the simultaneous presence of Fe³⁺ and Fe²⁺ ions, the Fe³⁺ ions would undergo a reduction reaction and become Fe²⁺ ions via the process of intercalated cations capturing some of the photoelectrons generated inside the TiO₂ coatings during UV illumination. The remaining excess electrons would migrate through the interfacial oxide layer to the 304LSS substrate, offering photo-cathodic protection to the alloy (Huang et al., 1998). On the other hand, when the UV illumination is shutoff, Fe²⁺ ions tends to re-oxidize into Fe³⁺ ions by releasing the electrons that have been captured during UV illumination. The released electrons would continue to protect the substrate alloy from corrosion even in the absence of UV illumination. Furthermore, the two outer layers of amorphous TiO₂ coatings would make it difficult for the preserved electrons to escape from coatings and react directly with the oxidizing species in corrosive environments. Eventually, the electrochemical corrosion potential of the substrate, instead of instantly moving back to the original value, would slowly rise up once the UV illumination is shutoff (Huang et al., 1998). This unique charge storage and slow release characteristics of the multi-layered TiO₂ coatings, as shown in Figure 1, may help maintain the effectiveness of cathodic protection over a prolonged period of time in the absence of UV illumination. The fundamental mechanism of electron capturing and releasing is as follows (Shinohara, 2001): Capturing:F e 3+ + e - → F e 2+ ( Under   UV   illumination ) (1) Releasing:F e 2+ → F e 3+ + e - ( Dark   condition ) . (2)

A continuous, uniform, and compact TiO₂ coating was prepared on Type 304LSS substrates by the sol–gel method for prolonged corrosion mitigation. The TiO₂ coating on the 304LSS/3TO sample exhibited the best photo-cathodic protection property in the corrosive environment. The 304LSS/3TO sample was cathodically protected for more than 10 h even after the cutoff of UV illumination. The corrosion current density of the 304LSS/3TO sample was still one order of magnitude less than that of the uncoated sample for 10 h without UV illumination. The novel multi-layered TiO₂ coating developed in this study is able to provide an effective and prolonged photo-cathodic protection for the stainless steel substrate.