Formation of a photocatalytic WO3 surface layer on electrodeposited Al–W alloy coatings by selective dissolution and heat treatment

In this study, we explored the feasibility of WO3 surface layer formation on electrodeposited Al–W alloy coatings by selective dissolution and heat treatment, with the aim of providing corrosion-resistant Al–W alloy coatings with photocatalytic self-cleaning properties under visible light illumination. The selective dissolution of Al and oxidation of residual W was carried out by immersing Al–W alloy films in an aqueous solution of nitric acid. A nanostructured H2WO4·H2O surface layer was formed on the alloy film by this process. The H2WO4·H2O layer was dehydrated to WO3 by heat treatment, yielding a multilayered WO3/Al–W alloy film with an approximately 300 nm thick WO3 layer. The WO3/Al–W alloy film exhibited photocatalytic self-cleaning, as demonstrated by the photodegradation of stearic acid and methylene blue. We also confirmed that selective dissolution and heat treatment did not significantly diminish the corrosion resistance of the Al–W alloy films.

Herein, we describe a new process for imparting corrosion-resistant Al-W alloy films with self-cleaning properties through the formation of a WO 3 surface layer. This process is comprised of a chemical dissolution step and heat treatment. Al is selectively removed by dissolution in an acidic solution, and a W-enriched layer is formed at the surface of the Al-W alloy film. Subsequent heat treatment in air converts the W-enriched surface layer to photocatalytic WO 3 . Through this process, a corrosion-resistant Al-W coating with self-cleaning ability under visible light illumination is obtained. Many strategies for preparing WO 3 surface layers have been reported in the literature, including sputtering [31][32][33] , chemical vapor deposition 34 , vacuum evaporation 35 , spin-coating 27 , spray pyrolysis 36,37 , sol-gel processing 38,39 , and hydrothermal growth 28,40 . The process described herein differs from these routes in that W present in the alloy film itself serves as the WO 3 precursor; hence, no additional W precursor is required. For this reason, this process is cost-effective and facilitates successive and large-scale production.
In this study, we examined the feasibility of forming photocatalytic WO 3 layers on Al-W alloy films via this process. The conditions required for the formation of photocatalytic WO 3 on Al-W alloy films were determined. Selective dissolution of Al-W alloy films. In the first attempt to form a surface oxidation layer, the electrodeposited Al-W alloy film was heated in air. However, thermal oxidation of the as-deposited film generated a complex oxide (Al 2 O 3 ·3WO 3 ) rather than photocatalytically active WO 3 ( Supplementary Fig. S1). The photocatalytic activity of the thermally oxidized film (Al 2 O 3 ·3WO 3 /Al-W alloy) was negligible ( Supplementary Fig. S2).
To obtain WO 3 instead of the complex oxide, the W content of the alloy film had to be increased prior to thermal oxidation. The Al-W alloy films were immersed in 0.56 M (3.5 wt.%) aqueous nitric acid (HNO 3 ) solution before heat treatment, with the intention of dissolving Al selectively to form a W-enriched surface layer. Figure 1b shows a typical EDX spectrum collected from the surface of an alloy film immersed in aqueous HNO 3 for >15 h. The amount of Al was significantly reduced after immersion in aq. HNO 3 , indicating that Al was selectively dissolved from the Al-W alloy films. This result is consistent with the expected behavior based on potential-pH diagrams 41 , which show that Al dissolves in acidic solutions, whereas W is passivated. The increased amount of O indicated that oxidation of residual W proceeded simultaneously during the dissolution process. The XRD patterns of the films after selective dissolution for different durations are shown in Fig. 2b-d. The pattern of the film after 9 h (Fig. 2b) contained no diffraction peaks except for the amorphous halo pattern of the Al-W alloy. In contrast, faint diffraction peaks attributed to H 2 WO 4 ·H 2 O (JCPDS: 18-1420) and an intense peak near 28° were observed in the pattern of the film treated for 12 h (Fig. 2c). The peak near 28° may be derived from WO 3 ·0.33H 2 O, but this is not a certainty owing to the absence of other diffraction peaks. The film treated for >15 h (Fig. 2d) 45 . Figure 3 shows surface scanning electron microscopy (SEM) images of the alloy films after selective dissolution for various durations. Small cracks a few hundred nanometers in length were observed at 9 h (Fig. 3a). These www.nature.com/scientificreports www.nature.com/scientificreports/ cracks were attributed to volume shrinkage caused by Al dissolution at the surface. With increasing duration, portions of the film along the cracks were exfoliated, but one side remained attached to the film surface to form petal-like structures, as indicated by arrows in Fig. 3b. Based on the XRD pattern shown in Fig. 2c, the petals in Fig. 3b were composed of WO 3 ·0.33H 2 O and H 2 WO 4 ·H 2 O. The petals covered the entire surface by 15 h (Fig. 3c), at which point they were composed of H 2 WO 4 ·H 2 O (Fig. 2d). The width and thickness of each H 2 WO 4 ·H 2 O petal were <300 nm and <50 nm, respectively. A similar petal-like morphology is commonly observed following selective dissolution of sputtered Al-W alloy films containing >45 at.% W 42 and electrodeposited Fe-W alloy films 45 , although the residue of the former is composed of β-W and that of the latter is composed of H 2 WO 4 ·H 2 O. Increasing the dissolution duration from 15 to 24 h (Fig. 3c,d) did not cause a notable change in the petal-like morphology, but the film treated for 24 h suffered from significant volume shrinkage that generated macroscopic cracks a few micrometers in width. Based on these results, a dissolution duration of 15 h was selected for the following experiments.  Fig. S3a). The XPS Al 2 s spectra confirmed that the WO 3 phase after heat treatment contained no elemental Al ( Supplementary Fig. S3b).

Dehydration of H
Surface SEM images (Fig. 4a,b) revealed that the petal-like grains were almost unchanged by heating at 300 °C, whereas the grains were sintered at 350 °C to form more compact grains. The cross-sectional image of an Al-W alloy film following heat treatment at 350 °C (Fig. 4c) shows that an ~300 nm thick WO 3 layer was formed on the film. The EDX spectrum of the Al-W alloy underlayer (Fig. 4d) indicated that it remained unoxidized, and the alloy composition was nearly identical to that of the as-deposited film.
It was thought that heat treatment at even higher temperatures might generate WO 3 with enhanced photocatalytic performance 49 . However, heat treatment at temperatures of 400 °C or higher crystallized the amorphous Al-W alloy to form intermetallic compounds such as Al 5 W and Al 12 W (Supplementary Fig. S4). The films containing these compounds were quite brittle and thus impractical for use as coatings. www.nature.com/scientificreports www.nature.com/scientificreports/ photocatalytic self-cleaning properties of Wo 3 /Al-W alloy films. Photocatalytic degradation of SA and methylene blue (MB) on the WO 3 /Al-W alloy films obtained by selective dissolution and heat treatment as described above was examined to evaluate the self-cleaning properties of the films. The photodecomposition of SA was monitored by measuring the change in the water contact angle on the film coated with SA. Absorption spectra of MB solutions in contact with the films were analyzed to monitor the photodegradation of MB.
The water contact angle on each film was measured under illumination with visible light after the film was treated with SA solution in heptane. For comparison, the contact angle on a bare Al-W alloy film was also measured following treatment with SA solution in heptane. As shown in Fig. 5, the contact angle on each film prior to illumination (t = 0 h) exceeded 50° due to the hydrophobicity of the SA adsorbed on the film surfaces. The contact angle for the bare Al-W alloy films remained almost constant regardless of illumination time, indicating the SA was present on the surface. In contrast, the contact angles on the c-WO 3  Based on the photocatalytic degradation results for SA and MB, the o-WO 3 /Al-W alloy film photodegraded organic compounds adsorbed on its surface more effectively than the c-WO 3 /Al-W film. Thus, the self-cleaning ability of the o-WO 3 /Al-W alloy film is superior to that of the c-WO 3 /Al-W film. corrosion resistance of the Wo 3 /Al-W alloy films. The resistance of the o-WO 3 /Al-W alloy film to pitting corrosion was investigated by measuring the pitting potential in 3.5 wt.% aqueous NaCl solution through potentiodynamic polarization tests. The polarization curve of the o-WO 3 /Al-W alloy film is shown in Fig. 7, with the polarization curves of bare Al-W alloy films containing 12.4 and 10.5 at.% W and an Al plate included for comparison. In each of the curves, the anodic current density exhibited a steep rise at a certain potential. This www.nature.com/scientificreports www.nature.com/scientificreports/  www.nature.com/scientificreports www.nature.com/scientificreports/ steep rise was attributed to pitting corrosion on the surface of the alloy films. The pitting potential of the o-WO 3 / Al-W alloy film was lower than that of the bare alloy film with a similar W content (12.4 at.% W); however, it was higher than those of the Al-W alloy film with 10.5 at.% W and the Al plate. conclusions A photocatalytic WO 3 surface layer was formed on electrodeposited Al-W alloy films by selective dissolution and heat treatment. The selective dissolution of Al and oxidation of W proceeded during immersion of the Al-W alloy films in an aqueous HNO 3 solution and nanostructured H 2 WO 4 ·H 2 O was formed on the alloy surface. Subsequently, H 2 WO 4 ·H 2 O was dehydrated to c-WO 3 or o-WO 3 by heat treatment at 300 or 350 °C, respectively. The orthorhombic WO 3 /Al-W alloy film exhibited superior visible-light photocatalytic activity for the photodegradation of SA and MB adsorbed on the surface. Although the resistance of the WO 3 /Al-W alloy film to pitting corrosion was slightly lower than that of the untreated Al-W alloy film, it was still better than that of the Al-W alloy film with 10.5 at.% W. These results demonstrate the feasibility of selective dissolution and heat treatment as a new process to obtain corrosion-resistant alloy films with photocatalytic self-cleaning abilities under visible light illumination.  www.nature.com/scientificreports www.nature.com/scientificreports/

Methods
Electrodeposition of Al-W alloy films. The electrodeposition of Al-W alloy films was carried out in an EMIC-AlCl 3 -W 6 Cl 12 bath placed in an argon-filled SDB-1AO glove box (Miwa Manufacturing Co., Japan). An EMIC-AlCl 3 melt was first prepared by slowly adding AlCl 3 (99%, Fluka, USA) to EMIC (97%, Tokyo Chemical Industry, Japan) in a molar ratio of 2:1. EMIC was vacuum-dried at 120 °C prior to use. The prepared melt was stored in a 25 mL glass vessel, which served as the electrolytic cell. W 6 Cl 12 , synthesized by a method described elsewhere 17,18 , was added to the melt to a final concentration of 49 mM. The bath temperature was maintained at 80 °C throughout the experiment with a TJA-550 thermostat (ASONE Corp., Japan) connected to a rubber heater wound around the cell and a thermocouple immersed in the bath.
Galvanostatic electrodeposition was performed on a 100 nm thick Cu film formed on a glass substrate by sputtering. A section of the Cu/glass substrate was covered with polytetrafluoroethylene tape, such that a defined 1 × 1 cm 2 area was exposed to the bath. An Al plate served as the counter electrode. The Cu/glass substrate and Al plate were placed vertically in the cell containing the melt. The Cu/glass substrate and Al plate were parallel to each other, and the distance between them was approximately 10 mm. During electrodeposition, the bath was stirred at 150 rpm with a stirrer bar (15 × 5 mm) and a PC-420D magnetic stirrer (Corning, USA). Electrodeposition was performed at 20 mA cm −2 for 25 min, with the current density controlled by a 660 C electrochemical analyzer (ALS Co., Japan).

Selective dissolution and heat treatment of Al-W alloy films. The electrodeposited Al-W alloy films
were mechanically polished to obtain smooth surfaces. The portion of each Cu substrate on which the alloy film was not deposited was covered with KTC-AC-828T masking resin (Kakoki Trading Co., Japan), after which the films were immersed in 0.56 M (3.5 wt.%) aqueous HNO 3 solution at room temperature.
The films were heat treated in a KBF848N1 electric furnace (Koyo Thermo Systems Co., Japan) in air. The films were heated from room temperature to the desired temperature at a rate of 2 °C min −1 , held at that temperature for 10 h, and then cooled slowly over several hours to room temperature.
The surface and cross-sectional morphologies of the alloy films were examined by field-emission scanning electron microscopy (FE-SEM), and the elemental compositions of the films were determined by EDX on a SU6600 field-emission scanning electron microscope (Hitachi, Japan) equipped with a Quantax Xflash 4010 detector (Bruker, USA). To prevent charging effects during the SEM analysis, a thin Au coating was deposited onto the samples by sputtering. The crystal structures of the films were determined by XRD analysis using an X'pert PRO-MPD diffractometer (Malvern Panalytical, UK). The XPS spectra of the film surfaces were measured by JPS-9030 (JEOL, Japan) with Mg Kα (1253.6 eV) X-ray source. The spectra in this paper were calibrated using C1s peak at 285.0 eV. evaluation of photocatalytic self-cleaning properties. The photocatalytic self-cleaning properties of the films were evaluated by monitoring the photodegradation of SA and MB 21 . The reagents used in these tests were purchased from Wako Chemical Co. (Japan) and used as received. For both tests, the films were illuminated with visible light at wavelengths from 380 to 520 nm. The light was emitted from a HAL-320 xenon solar simulator lamp (Asahi Spectra, Japan) with a light intensity of 1 sun passed through a Super Cold 750 filter and an SHPF-25C-533 short-pass filter (SIGMA KOKI, Japan) with a transmittance range of 380-520 nm. Wavelengths of >520 nm were blocked to prevent self-decomposition of MB via light absorption.
The photodegradation of SA was monitored according to the Japan Industrial Standards R_1753 method (2013). Each alloy film was illuminated with a LUV-16 UV lamp (ASONE Co., Japan) for 24 h in a dark room to ensure that the surface was devoid of interferents. Each film was then coated with a thin layer of SA by spin-coating 1 mL of 0.3 wt.% SA solution in heptane at 2000 rpm and dried at 70 °C on a hot plate for 10 min. Following SA coating, each film was illuminated under visible light from the solar simulator lamp in a dark room. The contact angle of a water droplet (<1 μL) on each film was measured with a ME2 contact angle meter (Asumi Giken Co., Japan).
For the photodegradation of MB, each film was first immersed in a large volume of aqueous 5 μM MB solution for >12 h to ensure that the adsorption/desorption equilibrium was reached. The film was then placed in a 4.5 mL acryl vial containing 1.5 mL of 5 μM MB solution and illuminated with visible light. The absorption spectrum of the MB solution was collected with a UV-2450 UV-VIS spectrometer (Shimadzu, Japan). The absorbance at 663 nm was used to determine the MB concentration according to Beer's law. evaluation of corrosion resistance. The resistance of the alloy films to pitting corrosion was evaluated by potentiodynamic polarization in a 3.5 wt.% aqueous NaCl solution. The NaCl solution was deaerated by bubbling with argon gas for 1 h prior to use. The electrochemical measurements were carried out with a three-electrode system using a 25 mL glass vessel as an electrolytic cell. A section of each alloy film was covered with a masking resin to expose a defined 0.2 mm 2 area that served as the working electrode. A platinum plate and a Ag/AgCl/ saturated KCl electrode were used as the counter and reference electrodes, respectively. The potential was scanned in the positive direction from −800 mV vs. Ag/AgCl/sat. KCl at a rate of 0.5 mV s −1 .

Data availability
The datasets generated and analyzed during the present study are available from the corresponding author on reasonable request.