**Characterization, Bioactivity and Antibacterial Properties of Copper-Based TiO2 Bioceramic Coatings Fabricated on Titanium**

#### **Salih Durdu**

Department of Industrial Engineering, Giresun University, Giresun 28200, Turkey; durdusalih@gmail.com or salih.durdu@giresun.edu.tr; Tel.: +90-4543104114

Received: 15 October 2018; Accepted: 18 December 2018; Published: 20 December 2018

**Abstract:** The bioactive and anti-bacterial Cu-based bioceramic TiO2 coatings have been fabricated on cp-Ti (Grade 2) by two-steps. These two-steps combine micro-arc oxidation (MAO) and physical vapor deposition–thermal evaporation (PVD-TE) techniques for dental implant applications. As a first step, all surfaces of cp-Ti substrate were coated by MAO technique in an alkaline electrolyte, consisting of Na3PO4 and KOH in de-ionized water. Then, as a second step, a copper (Cu) nano-layer with 5 nm thickness was deposited on the MAO by PVD-TE technique. Phase structure, morphology, elemental amounts, thickness, roughness and wettability of the MAO and Cu-based MAO coating surfaces were characterized by XRD (powder- and TF-XRD), SEM, EDS, eddy current device, surface profilometer and contact angle goniometer, respectively. The powder- and TF-XRD spectral analyses showed that Ti, TiO2, anatase-TiO2 and rutile-TiO2 existed on the MAO and Cu-based MAO coatings' surfaces. All coatings' surfaces were porous and rough, owing to the presence of micro sparks through MAO. Furthermore, the surface morphology of Cu-based MAO was not changed. Also, the Cu-based MAO coating has more hydrophilic properties than the MAO coating. In vitro bioactivity and in vitro antibacterial properties of the coatings have been investigated by immersion in simulated body fluid (SBF) at 36.5 ◦C for 28 days and bacterial adhesion for gram-positive (*S. aureus*) and gram-negative (*E. coli*) bacteria, respectively. The apatite layer was formed on the MAO and Cu-based MAO surfaces at post-immersion in SBF and therefore, the bioactivity of Cu-based MAO surface was increased to the MAO surface. Also, for *S. aureus* and *E. coli*, the antibacterial properties of Cu-based MAO coatings were significantly improved compared to one of the uncoated MAO surfaces. These results suggested that Cu-based MAO coatings on cp-Ti could be a promising candidate for biomedical dental implant applications.

**Keywords:** micro arc oxidation (MAO); Cu nano-layer; hydrophilic surface; apatite; in vitro bioactivity; antibacterial properties

#### **1. Introduction**

Commercially pure titanium (cp-Ti) materials are preferred for dental implant applications, owing to its low density, low elastic modulus (closer to that of bone), low thermal conductivity, non-magnetic properties, high specific strength, corrosion resistance, good mechanical properties, fracture resistance and fatigue resistance and biocompatibility [1–3]. It is well-known that titanium has corrosion resistant and biocompatibility properties. These are related to the native TiO2 layer spontaneously formed on its surface [4,5]. However, titanium implant materials cannot bond directly to the bone owing to their bio-inert nature (not bioactive), unlike bioactive ceramics such as bio-glass, glass ceramic, hydroxyapatite (HA), ZrO2 and TiO2 etc. [6,7]. As a result of this, the bone tissue around the implant is absorbed. This leads to slow healing and the loosening of the implant–bone interface [8]. Therefore, bioactive ceramics such as HA or TiO2 on titanium were coated to enhance bioactivity [9,10].

TiO2-based coatings have been suggested to improve corrosion resistance, bioactivity and biocompatibility of implant surfaces. They have recently received great attention for the biomedical applications owing to their more stable chemical composition [11,12]. TiO2 were fabricated on cp-Ti by various surface coating methods including sol–gel [13], anodic oxidation [14], magnetron sputtering [15], electrophoretic deposition [16], acid etching [17,18], laser surface treatment [19], plasma spraying [20,21] and micro arc oxidation [22–25] etc. However, certain problems like non-homogenous structure, micro-structural control problems, micro-cracks formation, the presence of phase impurity, poor adhesion strength were observed in these methods except for MAO technique [26–28]. Thus, implant application areas of the coatings produced by these techniques are limited.

Micro-arc oxidation (MAO) is one of the most applicable methods to deposit a porous and rough bioceramic layer on valve metals such as Ti, Al, Mg, and Zr surfaces [3,29–31]. The MAO coating promotes bioactivity, biocompatibility, wear and corrosion resistance respect to other surface coating methods [32–39]. Also, the adhesion strength between the substrate/MAO coatings is excellent, owing to its in situ growth [40]. Furthermore, it is reported that the porous and rough surfaces on the implant surfaces are beneficial for the formation and Osseo-integration of new bone tissue [41]. However, the bacterial adhesion and colonization may occur on the implant surfaces under body conditions. These lead to infections at the implant site and it results in the loss of implants [42].

In order to overcome bacterial adhesion and colonization, the surface modifications of implants were increasingly carried out by numerous anti-bacterial agents such as Cu [12], Ag [43] and Zn [44] over the last few years. The Cu is one of the basic trace elements necessary for human existence. It contributes in synthesis and release of life-sustaining proteins and enzymes within the living organisms [45]. Furthermore, it actively takes part in various enzyme-based processes, for bone metabolism stimulates new vessel formation and accelerates early skin wound healing [46,47]. Therefore, it is beneficial for bone tissue formation [12]. Moreover, copper exhibits excellent antibacterial properties versus a broad spectrum of bacteria, including gram-positive and gram-negative bacteria by interfering DNA replication and disrupting cell membranes [48–50].

In particular, there have been some promising studies on the fabrication of antibacterial Cu/CuO-nanoparticles containing Cu-incorporated TiO2 coatings on cp-Ti by using the MAO technique for the last five years [42,45,51–56]. Wu et al. investigated the formation and investigation of antibacterial resistances of Cu-incorporated TiO2 coatings by MAO and hydrothermal treatment (HT) [42]. Yao et al. investigated the antibacterial properties of Cu nanoparticles containing TiO2 coating synthesized by MAO [45]. Huang et al. fabricated the Cu-incorporated bioceramic coatings by MAO and HT and they evaluated osteoblast response [51]. Zhu et al. produced Cu-containing micro arc oxidized TiO2 coatings and evaluated anti-bacterial properties [52]. Zhang et al. examined the antibacterial properties of TiO2 coatings doped with various amounts with Cu nanoparticles deposited on titanium by MAO [53]. Huang et al. prepared Cu-containing TiO2 coatings on Ti by MAO and then, biocompatibility and antibacterial properties were evaluated [54]. Zhang et al. directly fabricated Cu-doped TiO2 coatings in alkaline electrolyte containing β-glycerophosphate disodium, calcium acetate and various amounts of copper acetate on Ti via a single step MAO process [55]. They then investigated in vitro biocompatibility and antibacterial activity for gram positive *S. aureus* [55]. However, in the above studies, Cu nanoparticles were randomly separated through the whole MAO surface. Also, for Cu-incorporated- and Cu containing-MAO coatings, Cu was not homogeneously distributed throughout the MAO surface. So, antibacterial adhesion properties on the Cu-MAO surface could not be sufficiently improved. Furthermore, Wu et al. produced Cu-doped TiO2 coatings on cp-Ti by magnetron sputtering with MAO [56]. The CuTi layers were formed on the titanium substrate by magnetron sputtering at the first stage. Then, the bioceramic coatings were produced by MAO technique at the second stage. Afterwards, their antibacterial properties were evaluated [56]. However, in that study, the Cu was observed around micro discharge channels and could not be dispersed throughout the MAO surface. Moreover, in vitro bioactivity of the Cu-MAO

coatings were not investigated, although in vitro biocompatibility investigations were carried out in the above aforementioned studies.

In this work, unlike the literature, convenient two-step MAO and PVD-TE techniques were devoted to synthesis uniform, bioactive, biocompatible and anti-bacterial novel Cu-based TiO2 bioceramic composite coatings on cp-Ti substrate. A bioactive and biocompatible anatase and rutile-based bioceramic structure on the cp-Ti substrate were coated by MAO technique in an alkaline electrolyte, consisting of Na3PO4 and KOH in de-ionized water at the first step. Then, a copper (Cu) nano-layer with 5 nm thickness was accumulated on the MAO coatings by the PVD-TE technique at the second step. The phase structures, morphologies, elemental amounts, functional groups, thicknesses, roughness and wettability of the MAO and Cu-based MAO coating surfaces were characterized by XRD (powder- and TF-XRD), SEM, EDS, FTIR, eddy current device, profilometer and CAG in detail, respectively. In vitro bioactivity of all coatings was evaluated by immersion tests in SBF at body temperature (36.5 ◦C) for 28 days. Then, for gram-positive bacteria (*Staphylococcus aureus*) and gram-negative bacteria (*Escherichia coli*), in vitro antibacterial properties of all coatings were investigated in detail.

#### **2. Materials and Methods**

#### *2.1. Sample Preparation*

The cp-Ti (Grade 2; commercially pure titanium) substrates were cut appropriate sizes (60 mm × 25 mm × 5 mm) by using a water jet. The surfaces of cp-Ti were ground by using silicon carbide (SiC) sand papers from No. 120 to No. 1200. And then, they were cleaned in an ultrasonic bath containing acetone for 60 min and dried in an oven at 50 ◦C.

#### *2.2. Micro Arc Oxidation (MAO) Process*

In order to produce bioceramic coatings on cp-Ti, an alternating current (AC) MAO device (MDO-100WS) running up to 100 kW was used. The MAO device mainly contained four pieces of equipment, consisting of an AC power supply, a double walled stainless steel tank, water cooled chiller and air flow stirrer. The cp-Ti was used as an anode, while the stainless-steel container was used as a cathode during MAO. The MAO solution was prepared by the dissociation of 10 g/L Na3PO4 and 1 g/L KOH in de-ionized water, respectively. The MAO treatment was performed on a constant current mode in the range of 0.325 A/cm<sup>2</sup> in Na3PO4 and KOH. The treating time was carried out at 5 min. The detailed MAO parameters and analysis results are given in Table 1. The temperature was maintained below 30 ◦C by a chiller in the tank during the MAO treatment. After the MAO treatment, all substrates were rinsed by de-ionized water and dried again in an oven at 50 ◦C for 24 h. Afterwards, they were preserved in desiccators. In order to ensure repeatability during MAO process, three MAO surfaces were produced on three cp-Ti specimens by the same parameters.



#### *2.3. Physical Vapor Deposition-Thermal Evaporation (PVD-TE) Process*

A Cu thin film layer with 5 nm (Copper: 99.999% purity of Alfa Aesar, Ward Hill, MA, USA) was accumulated on the MAO coatings using PVD-TE (Vaksis, Bilkent, Turkey, PVD/2T) at a deposition speed of 0.5 nm/s at room temperature. The vacuum chamber pressure was set at about 1 × <sup>10</sup>−<sup>6</sup> mbar before the PVD-TE process started. The vacuum chamber pressure was maintained under vacuum (base pressure) of 3 × <sup>10</sup>−<sup>5</sup> mbar. In order to avoid the abrupt evaporation of Cu powders, the changeable current was gradually increased up to 45 A. The evaporated materials (Cu powders) with a grain size of -100 mesh were placed at the bottom of a wolfram crucible, that was approximately 15 cm away from the MAO coating surfaces. The average thickness of Cu nano-layer on the MAO surfaces was measured as about 5 nm by XTM integrated to PVD-TE device (Vaksis, Bilkent, Turkey, PVD/2T). Cu vapor products were then deposited onto the MAO surfaces. In order to ensure repeatability during the PVD-TE process, three Cu nano-layers were deposited on three MAO coating specimens by the same PVD-TE parameters.

#### *2.4. Characterization of the Coatings*

The phase structure of the MAO surfaces was detected by powder-XRD (powder-X-ray diffractometer, Bruker D8 Advance, Billerica, MA, USA) with Cu-Kα (λ = 1.54 Å) between 2θ values of 10◦ and 90◦ with a scanning rate of 0.1◦·min−1. The phase structure of Cu-based MAO coatings was analyzed by TF-XRD (Thin film-X-ray diffractometer, PANalytical X'Pert PRO MPD, Philips, Amsterdam, The Netherlands) with Cu-Kα between 2θ values of 10◦ and 90◦ with a scanning rate of 0.001◦·min−1. The surface morphologies of all coatings were analyzed by SEM (Scanning electron microscope, Philips XL30S FEG, Amsterdam, The Netherlands). Also, the elemental amounts of all coatings were evaluated by EDS (Energy dispersive spectrometer, Philips, Amsterdam, The Netherlands). The 3-D surface topography and surface roughness were evaluated by profilometer (surface profiler, KLA Tencor P-7, Milpitas, CA, USA). The surface roughness values were achieved by the scanning of mechanical contact at 500 μm × 500 μm area in 3-D. The surface wettability (hydrophilicy/hydrophobicity) and the contact angle values of all surfaces were analyzed by CAG (Contact Angle Goniometer, Dataphysics OCA 15EC, San Jose, CA, USA) sessile drop technique. The contact angle data was recorded throughout every 10 s from 0 to 90 s after the de-ionized water droplet of volume 1 μL was contacted onto the both surfaces.

#### *2.5. In Vitro Bioactivity Tests*

To be informed of in vitro apatite-forming ability on the MAO and Cu-based MAO surfaces, all coatings were immersed in 1.0× SBF (simulated body fluid) at body temperature (36.5 ◦C) for 28 days. The SBF procures the formation of bone-like apatite layer on the implant surfaces. So, the apatite-forming ability on the implant materials represents information predicted about in vitro bioactivity. The MAO and Cu-based MAO coating samples were immersed in 1.0× SBF with ion concentrations almost equal to that in human blood plasma. The SBF was prepared by dissolving reagent grade chemicals consisting of NaCl, NaHCO3, KCl, K2HPO4·3H2O, MgCl2·6H2O, CaCl2 and Na2SO4 in distilled water at 36.5 ◦C, respectively. They were then buffered at pH 7.4 with (CH2OH)3CNH2 and 1 M HCl at 36.5 ◦C. The surface area' ratio (in mm2) of all surfaces to SBF was almost set 1 equal to 10 in the direction of the Kokubo and Takadama' recipe [57]. To maintain ion concentration of the SBF, it was renewed during every 24 h. All coating specimens were taken out from SBF at the post-immersion test and they were gently washed in de-ionized water. Eventually, they spontaneously dried under room temperature. All immersed dried coating specimens were kept in desiccators at pre-characterization. All experimental studies were carried out in triplicate.

After immersion treatment was completed in SBF, the morphologies, elemental structures, phase structures and functional groups of all coating surfaces were analyzed by SEM, EDS, TF-XRD and FTIR, respectively. The SEM images were taken with up to 20000× magnification. In addition, to reveal newly formed elements on immersed surfaces, all coatings surfaces were examined by EDS. The phase compositions of both coating surfaces were investigated by TF-XRD. The FTIR (Fourier transform infrared spectroscopy; JASCO FT/IR 6600, JASCO, Easton, MD, USA) spectra were collected over the range in the spectral range of 450–4000 cm−<sup>1</sup> at post-immersion in SBF.

#### *2.6. In Vitro Antibacterial Activity of the Coatings*

The antibacterial activities of all coating surfaces were investigated versus to *S. aureus* and *E. coli* by colony counting method. All coating surfaces were immersed in 5.0 mL of the bacterial suspension (1 × <sup>10</sup><sup>7</sup> CFU/mL). They were then incubated at 37 ◦C for 24 h. All coatings samples were washed by 150 mM NaCl at post-incubation and put into a tube including 2 mL phosphate buffer solution. Subsequently, to detach the bacteria from the surfaces to solution, they were shaken on a vortex for 2 min. Aliquots of the solution with 100 μL were plated onto muller hinton agar (MHA) plates. The active bacteria colonies on the surfaces were then incubated at 37 ◦C for 48 h, and were counted. All experimental studies were carried out in triplicate.

#### **3. Results and Discussion**

#### *3.1. Phase Structures of the Coatings*

The phase structures of the MAO and Cu-based MAO coatings on cp-Ti were investigated by powder XRD and TF-XRD as shown in Figures 1 and 2, respectively. In addition to the cp-Ti substrate diffraction peaks (JCPDS card number: 044-1294), the existences of characteristic peaks of anatase-TiO2 (JCPDS card number: 21-1272) and rutile-TiO2 (JCPDS card number: 21-1276) on the MAO surface were indicated by the powder-XRD pattern as shown in Figure 1.

**Figure 1.** XRD spectra pattern of the MAO coating.

**Figure 2.** TF-XRD spectra pattern of the Cu-based MAO coating.

After Cu was accumulated on the MAO surface, the phase structures formed on the surface were detected by TF-XRD as shown in Figure 2. The Ti (JCPDS card number: 04-004-8480), TiO2 (JCPDS card number: 01-070-2556), anatase-TiO2 (JCPDS card number: 01-083-5914) and rutile-TiO2 (JCPDS card number: 04-006-8034) were obtained on the Cu-based MAO surface as illustrated in Figure 1. The presence of crystalline Cu and/or Cu-based compounds were not verified by TF-XRD.

The phase of TiO2 has two polymorphs, which are also known as anatase and rutile. The phase of rutile is stable at high temperatures, while the anatase is a metastable phase at low temperature. The phase of anatase forms, which is a metastable structure, on cp-Ti surface at initial steps of MAO process. It is reported that the amount of anatase in the coatings decreases, as the amount of rutile increases. It is clearly stated that the rutile modifier becomes predominant after a critical period of MAO parameters such as treatment time, voltage and current. So, the anatase transforms to thermodynamically stable rutile with increasing experimental parameters, such as treatment time, voltage and current at the next stages of the MAO process [9,58].

Anatase and rutile phases formed on the cp-Ti surface through the MAO process occurred by the ionization mechanism and electrostatic interactions of oppositely charged ions (anion and cation). The alkaline electrolyte consists of Na3PO4, and KOH compounds contain Na+, PO4 <sup>3</sup>−, K<sup>+</sup> and OH<sup>−</sup> ions. The Na3PO4 is dissolved in distilled water and ionizes to Na+ and PO4 <sup>3</sup><sup>−</sup> (Equation (1)). Similarly, another compound of KOH is dissolved in de-ionized water and transforms to K+ and OH− ions (Equation (2)). Ti metals were dissolved and lost four electrons through MAO. Thereby, it transformed to positively charged cationic Ti4+ ions (Equation (3)). Synchronically, O2 gaseous is released. The O2 will either evolve as a gas, or dissolve into the solution as atoms and ionize to O2−. Also, positively charged Ti4+ and negatively charged O2<sup>−</sup> and/or OH<sup>−</sup> ions react with each other due to the electrostatic interaction through MAO (Equations (4) and (5)). Eventually, the anodic oxidation reactions occur between cationic Ti4+ and anionic O2−/OH<sup>−</sup> ions on Ti substrate through the MAO. So, TiO2 phase structure forms on cp-Ti substrate. The Gibbs energy of the anatase/rutile transition is negative because the phase of anatase is thermodynamically unstable for all temperature values. Anatase/rutile transition begins above 880 K and a completes at 1190 K [59]. The anatase to rutile transformation facilitates because the local temperature in micro discharge channels can reach 8000 K due to electron collisions through the MAO process [60]. Therefore, the amount of rutile increases with increasing MAO parameters such as treatment time, voltage and current.

Dissolution reactions in electrolyte:

$$\text{Na}\_3\text{PO}\_4 \rightarrow \text{Na}^+ + \text{PO}\_4^{3-} \tag{1}$$

$$\text{KOH} \rightarrow \text{K}^+ + \text{OH}^- \tag{2}$$

$$\text{Ti} \rightarrow \text{Ti}^{4+} + 4\text{e}^- \tag{3}$$

Anodic oxidation reactions through MAO process:

$$\text{Ti}^{4+} + 4\text{OH}^- \rightarrow \text{TiO}\_2 + 2\text{H}\_2\text{O} \tag{4}$$

$$\text{Ti}^{4+} + 2\text{O}^{2-} \rightarrow \text{TiO}\_{2} \tag{5}$$

#### *3.2. Surface Structures of the Coatings*

The surface morphologies of the MAO and Cu-based MAO coatings on cp-Ti were evaluated by SEM as shown in Figure 3. As seen in Figure 3, the surfaces of both coatings are porous and rough. The MAO coatings contain many crater-like or volcano-like micro pores and a few micro cracks due to the existence of thermal stresses during whole process. The various-sized micro pores on the MAO surface are called as micro discharge channels occur by micro spark discharge. The micro sparks form at weak regions such as sites and edges on cp-Ti substrate during oxide film due to the existence of dielectric breakdown at the initial steps of the MAO process. Thus, these resulted in the increase of the intensity of micro discharge channels throughout the MAO process. The micro discharge channels

have various sizes and occur on the substrate materials though the MAO process. So, volcano-like structures are observed on the MAO surfaces. Molten oxide structures term as oxide magma occur in micro discharge channels owing to the existence of local high temperature (up to about 10<sup>4</sup> K) and high pressure (approximately 100 MPa) [61,62]. The molten oxides in micro discharge channels is rapidly cooled and solidified because it comes into contact with the electrolyte during MAO. Eventually, it stacked instantaneously to form the MAO coatings. These porous and rough surfaces are beneficial to cell attachment and lead to increased cell adhesion [63]. Moreover, it is reported that porous TiO2 layers that promote the sinking of liquid into the pores owing to capillary forces are favorable for the seeding and spreading of cells [64,65].

**Figure 3.** Surface SEM morphologies of the coatings: (**a**) the MAO coating and (**b**) Cu-based MAO coating.

The morphologies of both coating surfaces are nearly identical as shown in Figure 3, although TE treatment is applied on the MAO surface. The Cu layer has approximately 5 nm thickness and could not change the MAO surface or fill porous structure. In our previous studies [66,67], the hydroxyapatite-based MAO surface on zirconium was coated by anti-bacterial silver and zinc elements with 20 nm thickness and the surface morphology could not be changed. However, the surface chemistry, hydrophilicity/hydrophobicity, apatite forming ability and antibacterial activity are changed although morphologies of the MAO coating surfaces are maintained after the TE process. The 3-D mapping average roughness values of the MAO and Cu-based MAO surfaces are obtained as 1.10 and 1.16 μm, respectively. Thus, the average roughness of both surfaces was not significantly changed.

#### *3.3. Elemental Chemical Analysis of the Coatings*

The elemental analysis spectra results (the elemental amounts) on the MAO and Cu-based MAO coatings were analyzed by EDS and are shown in Table 2. A trace of Cu element was detected on the Cu-based surface while the elements of Ti, O and P were observed on both surfaces. The Ti peaks originate from the substrate material and TiO2 structure in MAO coating. Furthermore, the O and P elements consist of anionic compounds such as PO4 <sup>3</sup><sup>−</sup> and OH<sup>−</sup> in Na3PO4 and KOH-based alkaline MAO electrolyte. Anionic compounds migrate from electrolyte to substrate due to the existence of opposite charged ions under electrical field through MAO. Afterwards, they react with positively charged cationic ions (Ti4+) accumulate on the MAO coating. However, the P element is not a crystalline form as seen in Figures 1 and 2 because it could not be detected by powder- or TF-XRD. The existence of the possible Ti-P-based compounds could not be proven in Figures 1 and 2. It can be expressed that they could not be transformed from amorphous to crystalline form on the surfaces whereas Ti4+ and PO4 <sup>3</sup><sup>−</sup> react with each other during MAO as supported by Figures 1 and 2. So, it could be stated that P-based compounds are an amorphous structure on the MAO-based surfaces. Moreover, the Cu was homogeneously stratified at a nanometer scale through the whole surface by TE process in that it is deposited on the MAO surface. The existence of the Cu element on the MAO surface is verified by the EDS-area whereas the crystallinity of it could not be confirmed by TF-XRD as shown in Figure 2. Thus, it is observed as a trace amount on the MAO and TE combined surface. The Cu alters the surface chemistry of the MAO without any morphological changing as shown in Figure 3. Furthermore, it can be concluded that the Cu film on the MAO surface is an amorphous structure.


**Table 2.** EDS area spectra results of the MAO and the Cu-based MAO coatings.

#### *3.4. Wettability of the Coatings*

The wettability (hydrophilicity/hydrophobicity) of both surfaces was evaluated by a CAG device as shown in Table 3. Also, the average contact angle values of the coatings dependent on contacting time were given in Table 3 after the water droplet made contact with the surface. The CAG measurement is an efficient method to stay on top of the surface wettability and the surface free energy [68]. A small contact angle value refers to good wettability as a high contact angle value indicates a poor wettability. For orthopedic and dental implant applications, a good surface hydrophilicity is necessary for adherent growth of cell and tissue and it represents a good biocompatibility [68–71]. Furthermore, this was supported by in vitro apatite-forming ability results (in Section 3.5).


**Table 3.** The average contact angle values of the MAO and Cu-based MAO surfaces at post-contacting time of droplet.

The surface wettability depends on many factors such as surface morphology, surface chemistry, roughness etc. [72,73]. The volcano-like pores on the surface absorb contacted distilled water by owing to capillary forces. So, for both surfaces, the contact angle values gradually decrease with increasing contact time as expected. Also, both surfaces exhibit hydrophilic characteristics because the contact angle values are lower than 90◦ [74]. It is clearly stated that TiO2-based MAO surfaces are rough, indicating hydrophilic properties [75].

There is no extreme difference in contact angle values of both coatings, as shown in Table 3. However, it is obvious that the Cu-based surface is more hydrophilic than the MAO surface. In this study, the wettability of the surfaces change, whereas the surface morphologies of them are nearly identical. In a previous study [66], the surface of Zn-based with 5 nm thickness hydroxyapatite-based coating on zirconium was super hydrophilic with respect to the plain MAO surface. Similarly, another study [67], the surface of Ag-based with 20 nm thickness hydroxyapatite-based coating on zirconium was observed more hydrophobic than the one of plain MAO surface. So, it could be concluded that the hydrophilic/hydrophobic nature of the surfaces are strongly related to the surface chemistry even if the surfaces have identical morphology. Moreover, it is clear that the hydrophilicity of the MAO

surface is improved after the Cu with 5 nm thickness is deposited on the MAO surface. Essentially, these two different surfaces' wettability is related to the polarity of them. The polar surfaces indicate that hydrophilicity/lower contact angles improve the wettability, while the opposite trend is observed in non-polar surfaces [76]. Eventually, it could be concluded that the accumulation of Cu onto the MAO increases the wettability/hydrophilicity. Also, this situation is beneficial for the attachment of cell and tissue for medical applications.

#### *3.5. In Vitro Bioactivity of the Coatings*

It is claimed that newly formed bone-like apatite layers on its surface under living body conditions is an essential requirement for binding bone tissue of the implant materials. This situation refers to in vivo apatite formation on the implant materials. An apatite structure on the surface can occur by immersion in SBF up to week-long periods under body temperature (36.5/37 ◦C) except for in vivo experimental conditions. The apatite formation/apatite forming ability on the surface provides predictive information about in vitro bioactivity. Thus, in vitro apatite forming ability on the surface is an important assessment to evaluate the bioactivity of the implants. However, "Apatite-forming ability is just a necessary but by no means sufficient precondition of "bioactivity". "Bioactivity" is a very complex interplay of many factors, where apatite-forming ability is just one of many" [77]. So, in order to be predicting information about bioactivity of all coating surfaces, in vitro immersion test was carried out under SBF condition at 36.5 ◦C for 28 days. After immersion tests were completed, the morphologies, elemental amount, phase structure and functional groups of all surfaces were characterized by SEM, EDS-area, TF-XRD and FTIR, respectively as seen in Figure 4, Table 4, Figures 5 and 6.

**Figure 4.** Surface morphologies of the coatings at post-immersion in SBF at 36.5 ◦C for 28 days: (**a**) 500× and (**b**) 20000× for the MAO surface and (**c**) 500× and (**d**) 20000× for the Cu-based MAO surface.



**Figure 5.** TF-XRD spectra pattern of the coatings at post-immersion in SBF at 36.5 ◦C for 28 days: (**a**) the MAO coating and (**b**) Cu-based MAO coating.

**Figure 6.** FTIR spectra of the coatings at post-immersion in SBF at 36.5 ◦C for 28 days: (black line) the MAO coating and (red line) Cu-based MAO coating.

After immersion in SBF for 28 days, the morphologies of both coating surfaces were investigated by SEM as shown in Figure 4. It could be clearly observed that the fine-dispersed particles are well dispersed throughout the whole surface at the post-immersion process. These fine-dispersed particles are nearly coated on the surfaces as seen in Figure 4. These particles are well dispersed through the whole surface. However, it is observed that the dispersed particles formed on the Cu-based surface are a little thinner than the one on the MAO surface at micron scales. It is well known that the crystalline apatite structure formed on the surfaces refers to predicting bioactivity. Therefore, as seen in TF-XRD and FTIR spectra results, the existence of crystalline apatite structure on the Cu-based MAO surface are greater than one the MAO surface whereas the apatite layer on the Cu-based MAO surface is a little thin respect to the MAO surface at macron scales scanning (106 × 106 <sup>μ</sup>m2). So, it can be stated that the Cu-nano layer on the MAO surface triggers increasing the crystallinity of apatite structure under SBF.

In order to get information, the elemental structures and the elemental amounts of both surfaces at post-immersion in SBF, the MAO and Cu-based MAO surfaces were analyzed by the EDS-area as shown in Table 4. In addition to the existence of Ti, O, P and Cu detected on the surfaces at pre-immersion in SBF as given in Table 2, an extra Ca element is obtained on both surfaces at post-immersion in SBF as given in Figure 5 and Table 4. The presence of this is connected with the diffusion Ca2+ ions of Ca-based SBF surfaces onto the surfaces during the immersion process. Positively charged Ca2+ ions in SBF migrate on the Ti-OH based surfaces due to the electrostatic interactions of oppositely charged ions. Then, PO4 <sup>3</sup><sup>−</sup> and OH<sup>−</sup> ions in SBF diffuse to both surfaces. These migrations and reactions are carried out through the immersion process in SBF. Eventually, they formed an apatite layer on the surfaces during the post-immersion process. Thus, an extra Ca element is observed on both surfaces during the post-immersion process. However, the electron penetration depth varies from 0.2 to 2 μm, depending on the accelerating voltage in EDS analysis. Thus, only presented elements in these particles formed at the post-immersion process in SBF cannot be observed by EDS. Besides the elements in these particles, the elements such as Ti, O, P and Cu existed in the MAO and PVD layer were observed by EDS analysis.

The phase structures of these newly formed dispersed structures were characterized by TF-XRD and FTIR as shown in Figures 5 and 6. The phases of Ti (JCPDS card number: 04-004-8480), anatase (JCPDS card number: 01-083-5914), rutile (JCPDS card number: 04-006-8034), TiO2 (JCPDS card number: 01-070-2556), TCP (tri-calcium phosphate: (JCPDS card number: 044-1294)) and apatite (JCPDS card number: 00-009-0432) on the MAO and Cu-based MAO surfaces at post-immersion in SBF were observed in Figure 5a,b, respectively. The existences of Ti, TiO2, anatase and rutile on both surfaces have been verified in Figures 1 and 2 at pre-immersion in SBF. The phases of TCP and apatite on both surfaces form at post-immersion in SBF under body temperature for 28 days. So, the fine-dispersed particles monitored in Figure 4 refer to TCP and apatite phases on immersed surfaces. As seen in Figure 5a,b, the intensity and amount of crystalline apatite structure on the Cu-based MAO are greater than ones on the MAO whereas the elemental amounts of Ca on the Cu-based MAO are lower than one on the MAO as reported in EDS-area spectra and amount results. The apatite-forming ability on the Cu-based MAO surface is high compared to the MAO surface according to TF-XRD spectra.

The functional groups and phases on the MAO and Cu-based MAO coatings at post-immersion in SBF were investigated by FTIR as seen in Figure 6. Also, at post-immersion in SBF, FTIR spectrum analysis results such as wave numbers, band modes, band assignments and phase were presented in Table 5. The band structures obtained on both surfaces at post-immersion in SBF are (PO4 <sup>3</sup>−), (C–C), (CO3 <sup>2</sup>−), (OH−), (P–H) and (Ti–O–Ti). The FTIR spectra curves found at 560 and 1026, 962 and 1050, 1403, 1420–1425 and 1460–1465, 1641, 1980 and 2302–2388, 3397–3448 and 3640–3742 cm−<sup>1</sup> correspond to (PO4 <sup>3</sup>−), (PO4 <sup>3</sup>−), (C–C−), (CO3 <sup>2</sup>−), (OH−), (P–H), (Ti–O–Ti) and (OH−), respectively [22,43,78–95]. The ATR-FTIR peaks of both surfaces verify the presence of (C–C) at 1403 cm−<sup>1</sup> [43,78–80], the presence of (OH−) ion at 1641 cm−<sup>1</sup> [43,78,79] and the presence of (Ti–O–Ti) in the regions of 3397–3448 cm−<sup>1</sup> [43,78–80]. In addition to XRD spectra, the presence of TiO2 structure on both surfaces are indicated by the bands of (C–C), (OH−) and (Ti–O–Ti) on FTIR once again [43,78–80]. The FTIR spectra curves on both surfaces refer to the presence of (PO4 <sup>3</sup>−) at 560 cm−<sup>1</sup> [81–83] and 1026 cm−<sup>1</sup> [83–86], the presence of (PO4 <sup>3</sup>−) at 962 cm−<sup>1</sup> [83–87] and 1050 cm−<sup>1</sup> [83–87], the presence of (P–H) in the regions of 1980 cm−<sup>1</sup> [88–93] and 2302–2388 cm−<sup>1</sup> [88–93] and the presence of (OH−) at 3640–3742 cm−<sup>1</sup> [88,91–94]. The apatite and TCP structures formed on both surfaces at post-immersion in SBF are proved by the presence of (PO4 <sup>3</sup>−), (P–H) and (OH−) bands on FTIR curves [81–94]. The FTIR spectra curves on both surfaces verify the presence of (CO3 <sup>2</sup>−) at 1420–1425 cm−<sup>1</sup> [22,95] and 1460–1465 cm−<sup>1</sup> [22,95]. The substituted carbonated-apatite formed on both surfaces at post-immersion

in SBF is proved by the presence of (CO3 <sup>2</sup>−) [22,95]. It is clear that the induced apatite is a carbonated apatite at post-immersion in SBF for 28 days. It is well known that the sharp and deep of FTIR peaks provide information about the crystallinity of phases on the surface [96]. As supported in Figure 5, the sharp and deep peaks of (PO4 <sup>3</sup>−) bands in the Cu-based MAO verify the existence of highly crystalline apatite structure respect to the MAO in Figure 6. Thus, it can be stated that the Cu on the MAO improve predicting bioactivity compared to the MAO surface.

**Table 5.** ATR-FTIR spectrum analysis results for the MAO and Cu-based MAO coatings after immersion in SBF.


SBF is a metastable calcium phosphate-based electrolyte supersaturated compared to the apatite structure [97]. However, it is stated that a chemical stimulus is required to trigger the heterogeneous nucleation of apatite from the SBF because the homogeneous nucleation threshold of apatite is very high [98]. The hydroxyl groups such as Ti–OH on the surfaces are basically essential to induce the apatite nucleation. The provision of abundant Ti–OH groups and the enrichment of calcium and phosphate trigger the nucleation of apatite on the MAO surface [97]. After apatite nuclei forms on the surface, the ions of Ca2+, PO4 <sup>3</sup><sup>−</sup> and CO3 <sup>2</sup><sup>−</sup> in SBF diffuse to TiO2-based surfaces to combine with apatite nuclei owing to the electrostatic interactions of opposite charged ions. As a result, a novel apatite layer is formed on both coating surfaces.

#### *3.6. In Vitro Antibacterial Activity of the Coatings*

In order to determine the antibacterial contribution of the Cu layer on the MAO process, the level of bacterial colony adhering to the cp-Ti, the MAO (TiO2) and Cu-based MAO surfaces was investigated. Figure 7 shows active colony ratios of gram-positive (*S. aureus*) and gram-negative (*E. coli*) bacteria adhered to all tested surfaces. Also, Figures 8 and 9 show *S. aureus* and *E. coli* colony plates formed by bacteria adhering to all surfaces after incubation, respectively. It was determined that the adhesion of gram-positive and gram-negative bacteria on the cp-Ti surfaces was lower than that of the MAO surface. The MAO surfaces containing bioactive and biocompatible TiO2 structures are porous and rough. Thus, the surface energy of the MAO is greater than one of the smooth surfaces. Hence, it can be concluded that the active colony ratios of *S. aureus* and *E. coli* adhered on the MAO surfaces increased in very small amount compared to the smooth cp-Ti surfaces. However, this difference was not statistically significant (*p* > 0.05). So, the numbers of bacteria adhering to cp-Ti and the MAO surfaces were very close to each other.

The antibacterial activity of the Cu-based MAO surfaces against *E. coli* was determined as 68.0%. Also, the antibacterial activity of the Cu-based MAO surfaces against *S. aureus* was observed as 69.6%, respectively. It has been observed that Cu-based MAO coating process provides antibacterial property to the substrate and MAO surfaces. Also, the Cu coating increases the antibacterial property by 1.11-1.18-fold against *S. aureus* and *E. coli*, respectively. This result shows that the Cu-based MAO

coating on the surface has a significant effect on antibacterial activity. The increase in antibacterial activity by increasing the coating can be explained by the more intense interaction of Cu ions on the surface and bacteria.

**Figure 7.** Active colony ratios of *S. aureus* as gram positive bacteria and *E. coli* as gram negative bacteria on cp-Ti substrate, the MAO and Cu-based MAO surface cultivation.

**Figure 8.** Culture plate photographs of *S. aureus* after re-cultivation: (**a**) cp-Ti substrate, (**b**) the MAO and (**c**) Cu-based MAO surface cultivation.

**Figure 9.** Culture plate photographs of *E. coli* after re-cultivation: (**a**) cp-Ti substrate, (**b**) the MAO and (**c**) Cu-based MAO surface cultivation.

The increase in antibacterial activity of surfaces after Cu-based coating may be explained by the toxic effect of Cu ions on the bacterial cell. In mediums containing Cu-based surfaces, bacteria are immobilized to the coated surfaces; proliferation and movement are restricted, and cell death occurs [99]. Copper causes toxic effects on the bacteria by multiple mechanisms and prevents the resistance formation in the bacteria. The first mechanism is that copper ions deform the cell wall and cause cell death. In another mechanism, copper ions inactivate membrane

proteins and enzymes, disrupt transporter molecules and lead to cell death [100,101]. Previously, some researchers have investigated the antibacterial properties of Cu-based MAO surfaces against various microorganisms [102,103]. Trapalis et al. [104] reported that copper-coated surfaces (Cu/SiO2) exhibited significant antibacterial activity against *E. coli*.

Another important observation in this study is that Cu-based surfaces show higher antibacterial activity against *S. aureus* than *E. coli*. The Cu-based surfaces, which exhibit the highest antibacterial activity in this study, showed 1.02 times higher antibacterial activity against *S. aureus* compared to *E. coli*. This result can be explained by the structural differences of gram-positive and gram-negative cells. Gram-negative bacteria have an outer membrane in addition to the cell wall, while gram-positive bacteria do not have an outer membrane. The outer membrane acts as a barrier to reduce the transport of metals to the bacteria. For this reason, copper penetrates rapidly into the cell in gram-positive cells and a serious toxic effect occurs. However, in gram-negative bacteria containing outer membrane, copper transport to the cell is restricted; the formation of toxic effect is delayed and reduced [50,105]. Similar studies in the literature support these results. Wang et al. [105] reported that selenium-coated surfaces are more effective against gram-positive bacteria and that this result is due to the cellular difference between *S. aureus* and *E. coli*.

#### **4. Conclusions**

A novel Cu-based TiO2 coating has been produced on cp-Ti surfaces by using combined two-step MAO and PVD-TE methods for dental implant applications. The TF-XRD spectra results showed that the existence of crystalline Cu on the MAO surfaces could not be observed whereas it was detected as elemental structure by EDS attached to SEM. Also, the SEM analyses indicated that the surface morphologies of the MAO and Cu-based MAO coatings were porous and rough. The morphology and topography of both coating surfaces were not changed by PVD-TE treatment. However, the hydrophilicity of Cu-based MAO surfaces was improved with respect to the MAO surfaces owing to the enhancing polarity of Cu on the MAO surface. Moreover, in vitro bioactivity of Cu-based TiO2 surfaces was increased compared to the MAO surfaces after immersion test in SBF for 28 days. These were supported by TF-XRD, ATR-FTIR and SEM as written in Section 3.5. Furthermore, it was observed that the amounts of active-bacteria colonies lived on Cu-based MAO surfaces were lower than ones of the MAO surface for gram positive (*S. aureus*) and gram negative (*E. coli*).

There have been some remarkable studies on the production of antibacterial Cu/CuO-nanoparticles containing Cu-incorporated- and Cu-containing-TiO2 coatings on Ti by a single or hybrid MAO methods [42,45,51–56]. However, Cu nanoparticles were randomly separated through the whole MAO surface in aforementioned studies. Also, Cu was not uniformly separated on Cu-incorporated- and Cu-containing-MAO surfaces. So, antibacterial adhesion properties on the Cu-MAO surface could not be sufficiently improved. Moreover, the Cu was observed around micro discharge channels and could not be dispersed during combined hybrid MAO surfaces. Moreover, in vitro bioactivity of the Cu-nanoparticles, Cu-incorporated and Cu-containing-MAO coatings were not investigated although in vitro biocompatibility investigations were carried out in aforementioned studies. In conclusion, it can be stated that the Cu-based TiO2 coatings are porous, rough and have the potential for biomedical surface coating applications due to their hopeful properties such as surface chemistry, morphology, hydrophilicity, in vitro bioactivity and in vitro antibacterial resistance with respect to other literature studies.

#### **Funding:** This research received no external funding.

**Acknowledgments:** The author would like to special thank F. Unal for running Physical Vapor Deposition Thermal Evaporation System at Giresun University GRUMLAB, A. Nazim for running SEM and EDS, A. Sen for running XRD at Gebze Technical University and B. Alcan and Y. Ozturk for running TF-XRD at TUBITAK MAM Materials Institute.

**Conflicts of Interest:** The author declares no conflict of interest.

#### **References**


© 2018 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Plasma Electrolytic Oxidation of Titanium in H2SO4–H3PO4 Mixtures**

#### **Bernd Engelkamp 1,\*, Björn Fischer <sup>2</sup> and Klaus Schierbaum <sup>1</sup>**


Received: 19 December 2019; Accepted: 28 January 2020; Published: 30 January 2020

**Abstract:** Oxide layers on titanium foils were produced by galvanostatically controlled plasma electrolytic oxidation in 12.9 M sulfuric acid with small amounts of phosphoric acid added up to a 3% mole fraction. In pure sulfuric acid, the oxide layer is distinctly modified by plasma discharges. As the time of the process increases, rough surfaces with typical circular pores evolve. The predominant crystal phase of the titanium dioxide material is rutile. With the addition of phosphoric acid, discharge effects become less pronounced, and the predominant crystal phase changes to anatase. Furthermore, the oxide layer thickness and mass gain both increase. Already small amounts of phosphoric acid induce these effects. Our findings suggest that anions of phosphoric acid preferentially adsorb to the anodic area and suppress plasma discharges, and conventional anodization is promoted. The process was systematically investigated at different stages, and voltage and oxide formation efficiency were determined. Oxide surfaces and their cross-sections were studied by scanning electron microscopy and energy-dispersive X-ray spectroscopy. The phase composition was determined by X-ray diffraction and confocal Raman microscopy.

**Keywords:** titanium dioxide; plasma electrolytic oxidation; anatase and rutile

#### **1. Introduction**

The oxide layer of metals such as titanium can be tailored for specific applications. The most common technique used to artificially grow a passive layer is anodic oxidation, in which moderate voltages promote a denser and thicker oxide layer compared with the naturally formed oxide. With the increasing scope of applications, new demands on materials have evolved. To meet these needs, researchers have developed new techniques from classical anodic oxidation. One derived technique is plasma electrolytic oxidation (PEO), in which the applied voltage exceeds a critical point and causes the initial oxide layer to reform by characteristic breakdowns, which often induce plasma conditions. The complex interplay between chemical, electrochemical, and thermodynamic reactions creates unique oxide layers and enables versatile changes in these layers by slightly changing the process parameters. Therefore, PEO-treated titanium can be used for a variety of applications, such as biomedical prostheses, automotive components, and photocatalytic devices [1].

Furthermore, PEO-treated titanium has recently been used as a gas sensor material at room temperature for various gases [2]. In this case, H2SO4 at an exceptionally high concentration of 12.9 M is used as an electrolyte and leads to a characteristic porous oxide structure with a layer thickness of approximately 5.5 μm. The morphology and thickness suggest a high surface-to-bulk ratio, which is beneficial for gas sensor technology. In general, breakdowns during PEO promote the formation of crystalline titanium dioxide phases, namely, anatase and the high-temperature phase rutile [3–5]. Both phases differ in significant properties (e.g., band-gap energy and electron–hole recombination rate) for potential application as a gas sensor material [6]. When investigating the effect of the crystal phase composition on the gas–oxide interaction, the ability to systematically control the rutile to anatase fraction is desirable. In 12.9 M H2SO4, the dominant crystal phase in the oxide is rutile, while the anatase fraction dominates in lower concentrations [7]. However, lower concentrations adversely affect the oxide layer by decreasing its thickness and porosity.

In recent studies on PEO for medical applications, electrolytes with phosphoric acid (H3PO4) have been frequently used, and oxide layers of comparable porosity and thickness can be formed [8–10]. Anatase is the dominant phase in these layers, and rutile is almost non-existent. When used as an electrolyte, phosphoric acid not only affects the phase composition but also drastically changes the outcome of the PEO process. The combination of both electrolytes provides an interesting approach to tailoring the properties of the oxide scale. Since the ratio of the two compounds in the mixture is critical for obtaining specific properties, we explored the effect of small amounts of H3PO4 in concentrated H2SO4.

Our PEO experiment is based on a galvanostatic DC operation mode. The resulting constant current offers a simple method of treating and evaluating samples for a systematic PEO study. For instance, it can be split into several contributions and classified into ionic and electronic currents [8,11]. The ionic current reflects the migration of ions in the oxide layer and is the driving force in conventional anodic oxidation. The electronic current is mainly induced by breakdowns. In the course of the PEO process, a transition from ionic to electronic current can be observed.

Starting with concentrated sulfuric acid (12.9 M) as an electrolyte, we investigated the impact of adding H3PO4 at molar fractions of 1% and 3%. Before and after PEO, the samples were analyzed for weight gain with a microbalance. Scanning electron microscopy and X-ray diffraction were used to systematically study oxide surfaces and cross sections at different stages of the process. We derived information about the phase distribution in the oxide layer from confocal Raman microscopy of cross sections, and elemental composition was investigated by energy-dispersive X-ray spectroscopy.

#### **2. Materials and Methods**

Samples (surface area of approx. 3.634 cm2) were cut from titanium foil (thermally annealed, 99.6% purity; 125 μm thickness) by means of a laser (PowerLine F30, ROFIN-SINAR Laser GmbH, Hamburg, Germany). The samples were cleaned ultrasonically for approximately 10 min in acetone and 10 min in deionized water. The electrolytic cell consisted of a glass vessel with an integrated glass shell, which enabled the temperature regulation of the electrolyte by pumping cold thermal fluid through the shell. The thermal fluid temperature was kept constant at 15 ◦C with a recirculating cooler (FL1201, JULABO GmbH, Seelbach, Germany). The reaction chamber was filled with 114 ± 5 mL of the electrolyte. The electrolyte was based on 12.9 M H2SO4 (75 wt%). It was enriched with 25 wt% H3PO4, which resulted in molar fractions of n(H3PO4)/n(H2SO4) = 0%, 1%, and 3%. Specifically, the last two fractions corresponded to c1%(H3PO4) = 0.1 M plus c1%(H2SO4) = 12.3 M and c3%(H3PO4) = 0.3 M plus c3%(H2SO4) = 11.4 M. A magnetic stirrer prevented spatial temperature differences in the electrolyte and reduced the disturbance of gas accumulations on the electrode surfaces. The electrolytic system was completed by the titanium sample as the anode and a graphite rod as the cathode (area of immersion of 7.38 ± 0.6 cm2) at a distance of 23 ± 4 mm.

A constant current density of 55 mA/cm2 was applied by using a highly stable current power supply (FUG MCP 350-350). Voltage and current were adjusted and recorded in 250 ms intervals using an in-house developed LabVIEW program. After treatment, the sample was rinsed in deionized water and dried in air. The weight of the sample was measured before and after the process with an analytical balance (ABT 120-5DM, Kern und Sohn GmbH, Germany) with a repeatability of 0.02 mg.

The microstructure of the oxide layer was investigated by field emission scanning electron microscopy (SEM; JSM-7500F, JEOL Ltd., Tokyo, Japan). Surface images of secondary electrons were captured with 5 kV excitation. Cross sections were prepared by argon ion milling (Cross Section Polisher IB-09010CP, JEOL Ltd., Tokyo, Japan). For energy-dispersive X-ray spectroscopy (EDX), images were created with 15 kV excitation energy and detected with an XFlash Detector 5030 (Bruker AXS GmbH, Karlsruhe, Germany). Quantitative results of elemental composition were obtained by averaging over at least three comparable sections to minimize local fluctuations.

The phase composition was determined by X-ray diffraction (XRD). Diffraction data were collected on a Bruker D2 Phaser diffractometer with Cu-K*α* radiation (*λ* = 1.54184 Å, 30 kV, 10 mA) in Bragg–Brentano geometry and a LYNXEYE 1D detector. XRD patterns were measured with a flat silicon, low-background rotating sample holder (5.0 min−1) with 24.5◦ < 2*θ* < 29.5◦, a scan speed of 2 s/step, and a step size of approximately 0.024◦.

Raman measurements were performed with a confocal Raman microscope alpha300 R (WITec GmbH, Ulm, Germany). A fiber-coupled single-mode DPSS laser with an excitation wavelength of 532 nm was used. The laser power applied to the sample was set to 20 mW. A Zeiss EC Epiplan-Neofluar DIC 100x/0.9 NA was selected as the microscope objective, and the samples were scanned with a step size of 200 nm. In this way, a spatial resolution of about 300 nm could be achieved. The system also featured real-time laser profilometry, so the sample surface remained in the focal plane during the entire measurement period. The spectrometer used was a WITec UHTS 300 combined with an Andor iDus Deep Depletion CCD detector, which was cooled to −60 ◦C. The Raman scattered light was spectrally dispersed by a reflection grating with 1200 mm<sup>−</sup>1. The average spectral resolution was about 2 cm<sup>−</sup>2/pixel. The software WITec FIVE version 5.2.4.81 was used to evaluate the measurement data and create Raman images, including cosmic ray removal and background subtraction by the implemented shape function.

#### **3. Results**

#### *3.1. Voltage Response and Mass Change*

The voltage response of the PEO process in 12.9 M H2SO4, as shown in Figure 1a, reveals information about the current character. The ratio between electronic and ionic currents varies during the PEO process. In the beginning, the electric field was insufficient to cause electric breakdowns, and conventional anodic oxidation occurs. The ionic current predominated and promoted the formation of a dielectric oxide layer. Consequently, the cell resistance increased. The constant current was sustained by the rapid increase in applied voltage. Above a critical voltage, electrical breakdowns become visible by electroluminescence. The charge transfer by breakdowns represents an electronic current and is energetically favored compared with ion migration. The electronic current increasingly contributes to the total current. Eventually, a linear voltage region is reached, which indicates a mainly electronic current character. This linear stage is also known as the microarc stage [9,12].

The impact of breakdowns on oxide formation was further investigated by terminating the process at different charge densities and determining the mass change, presumably due to oxide formation, with an analytical balance. Figure 1b presents the mass change with varying charge densities. The slope of the polynomial fit represents the mass change per transferred charge, i.e., the oxide formation efficiency [13]. The efficiency for 0% H3PO4 was positive until 8.4 C/cm2, after which it was negative. This turning point corresponded to the beginning of the linear microarc stage. This indicates that breakdowns during the microarc stage in concentrated sulfuric acid cause mass loss of the oxide. The initial passivation by breakdowns transforms into a destructive reforming with combined mass loss.

The progress drastically changed by adding small amounts of H3PO4 to concentrated sulfuric acid. The voltage response for n(H3PO4)/n(H2SO4) = 1% in Figure 1a already differed from the voltage response in pure H2SO4. The transition from ionic to electronic current was also observable by the subsequent decreasing voltage rate. However, it was less pronounced. The interruption of the steep increase between 1.5 and 7 C/cm2 is due to the previously reported transition from a grooved morphology to a porous morphology [13,14]. The mass loss in Figure 1b was suppressed compared with samples prepared in pure H2SO4. The efficiency was positive until 19.4 C/cm2. Shortly after the efficiency changes to negative values, the microarc region started, and no noticeable voltage gain occurred. Hence, the applied voltage was sufficient to sustain the defined current density, even though the oxide formation efficiency in this region (Figure 1b) was negative. Therefore, a steady state between repassivation of the dielectric layer and its destruction by breakdowns can be assumed.

With 3% phosphoric acid, the trend continued more drastically. Higher voltages, even above 200 V, were necessary to sustain the given current density. The transition from ionic to electronic currents was completed even later, while the mass change was always positive. At approximately 55.8 C/cm2, the efficiency changed from positive to negative values, and the microarc region started. However, above approximately 70 C/cm2, inhomogeneities were visible on the oxide surface and restricted the process in 3% H3PO4 at higher charge densities.

**Figure 1.** Process information at different charge densities for 0%, 1%, and 3% additional H3PO4 in 12.9 M H2SO4: (**a**) voltage response and (**b**) mass change with interpolation and marked highest oxidation efficiency.

#### *3.2. XRD Investigation*

The crystallographic structure of the oxide was investigated by X-ray diffraction (XRD). The weight fraction of rutile can be estimated for each sample by using the (101) reflection of anatase and the (110) reflection of rutile [7,15]. Figure 2 presents the calculated rutile fraction for different charge densities. The rutile fraction in the sample prepared with 0% H3PO4 increased continuously with the charge density until the sample was almost exclusively rutile. The titanium reflections of the titanium substrate decreased continuously as a result of the growing oxide layer [7]. The transition was similar in 1% H3PO4, but it was less pronounced. At a low charge density, mainly anatase was present, and the rutile fraction increased with increasing charge density. However, the maximum value and the slope of the fitted curve are smaller. For 3% H3PO4, our findings were remarkably different, and the typical increase in the rutile to anatase fraction was no longer identified. The fraction was below 11% for any charge density.

**Figure 2.** (**a**) Rutile to anatase fractions (with interpolation) versus the charge density for samples prepared with 0%, 1%, and 3% additional H3PO4 in 12.9 M H2SO4. The fractions are derived from XRD intensities. (**b**) Representative diffractograms of chosen samples with high and low current densities.

#### *3.3. SEM Surface Images*

The scanning electron microscope (SEM) images of surfaces in Figure 3 demonstrate the modification of surfaces resulting from variations in the transferred charge and the amount of H3PO4. In the left column, the surfaces of samples prepared with 0% H3PO4 are shown. Since the breakdown voltage was already exceeded for 9 C/cm2, circular sinkholes of former discharge channels, i.e., micropores, were clearly visible. However, the even distribution caused the surface to appear regular and flat. With increasing charge transfer, the surface became rougher. At 41 C/cm2, some minor plateaus and some cracks were noticeable. For 69 C/cm2, plateaus and cracks were clearly visible and impaired the circular shape of pores.

**Figure 3.** SEM images of sample surfaces produced in H2SO4 with the addition of 0%, 1%, or 3% H3PO4. Samples produced with similar charge densities (±1.0 C/cm2) are presented in the same row.

In the second column, the surfaces of the samples prepared in 1% H3PO4 are shown. The porous structure remained apparent. However, the pore size increased, and the pore density decreased. As observed previously, the pore network dissolved at a higher charge density, and different levels of depth evolved. In the right column, the surfaces of samples prepared in 3% H3PO4 are presented. The pore size further increased, and the density further decreased. The destructive character of the breakdowns can be observed, although it was less pronounced compared with the samples prepared in 0% or 1% H3PO4.

#### *3.4. SEM Cross Sections*

The result of our SEM cross section investigation in Figure 4 shows the depth profile of the oxides. The total thickness distinctly increased with the fraction of phosphoric acid in the electrolyte. Values were approximately 5.0 μm for 0%, 7.3 μm for 1%, and 10.9 μm for 3% (±0.5 μm). Different layers could be distinguished in the oxide layer because they abruptly changed in morphology or elemental composition. Two layers were as described for PEO in sulfuric acid: a compact layer beside the titanium substrate with a relatively small thickness and a porous layer with a major contribution to the total thickness [7,16]. Furthermore, in Figure 4c, a smooth area in the near the surface was clearly distinguishable from the porous layer below.

**Figure 4.** SEM images of oxide cross sections produced in H2SO4 with the addition of 0% (**a**), 1% (**b**), and 3% (**c**) H3PO4. The charge density in all plasma electrolyte oxidation (PEO) processes is 41.3 <sup>±</sup> 1.0 C/cm<sup>−</sup>2.

#### *3.5. EDX Images*

While the compact layer is rather difficult to resolve in the SEM cross section image, it is clearly identifiable by its elemental composition. This is shown in Figure 5 by the energy-dispersive X-ray spectroscopy (EDX) images. Remarkably, the compact layer in all samples exhibited an increased sulfur concentration of approximately 1.1 ± 0.3 at % (while it was below 0.3 ± 0.3 at % in the remaining oxide). In samples prepared with H3PO4, an increased phosphor concentration throughout the oxide could be detected. For 3% H3PO4, the value is approximately 1.9 ± 0.3 at %. In comparison, the phosphor concentration in a sample produced in 1% H3PO4 amounted to 0.6 ± 0.3 at %. Another feature was detectable in the sample produced in 3% H3PO4 and less pronounced in the sample produced in 0% H3PO4: a smooth area near the surface was observed in several samples and independent of the electrolyte composition. The EDX analysis in Figures 5a,c reveals that the region exhibited an increased titanium concentration, while the oxygen and phosphor concentration decreased.

#### *3.6. Confocal Raman Microscopy*

Since a common drawback of XRD phase analysis is the low spatial resolution, we additionally performed confocal Raman microscopy. Figure 6 shows false-color images, which are derived from Raman microscopy of cross sections prepared with 41.3 ± 1.0 C/cm−<sup>2</sup> in mixtures with 0%, 1%, and 3% additional H3PO4. The surrounding background image is the result of light microscopy. The different colors in the false-color images indicate the types of Raman spectra, which are presented below. The anatase phase is recognized by an intense Eg mode around 147 cm−<sup>1</sup> [17]. Additionally, Raman-active modes derived from anatase around 395 cm−<sup>1</sup> (B1g), 515 cm−<sup>1</sup> (B1g and A1g), and 637 cm−<sup>1</sup> (Eg) are used for classification [7]. The rutile phase can be identified by two Raman modes around 442 cm−<sup>1</sup> (Eg) and 605 cm−<sup>1</sup> (A1g) [7].

**Figure 5.** Energy-dispersive X-ray spectroscopy (EDX) images of oxide cross sections produced in 12.9 M H2SO4 with 0% (**a**), 1% (**b**), and 3% (**c**) H3PO4. The transferred charge density is around 41.3 <sup>±</sup> 1.0 C/cm−<sup>2</sup> for each sample. The color intensity in the analyzed segment is only comparable to other elemental maps for the same cross section.

**Figure 6.** False-color images derived from confocal Raman microscopy with underlying light microscopy images of three cross sections. The transferred charge density is around 41.3 <sup>±</sup> 1.0 C/cm−<sup>2</sup> for all cross sections, while the fraction of H3PO4 in the electrolyte changes, i.e., 0% (**a**), 1% (**b**), and 3% (**c**). Red represents anatase, blue represents rutile, and cyan corresponds to a signal with large background. The integration time of each single spectrum varies, with 0.2 s in 0%, 0.1 s in 1%, and 0.05 s in 3%. Averaged spectra from the corresponding color-marked area are shown below each cross section.

Without H3PO4, the porous layer mainly exhibited rutile, while a distinct anatase fraction could be identified from the spectra of the compact layer. For 1% H3PO4, the colored areas, which indicate anatase and rutile, were similar in size and homogeneously distributed. For 3%, the intensities of the rutile modes in the spectra vanished. The entire oxide mainly exhibited anatase. For 0% and 3%, small areas near the surface were marked with cyan and fitted the previously mentioned smooth area in our SEM and EDX images. The corresponding cyan spectra resembled the spectra from rutile. However, it clearly differed by a frequency shift and an increased background signal, which may have originated from near-surface groups.

#### **4. Discussion**

Our results demonstrate that adding phosphoric acid to concentrated sulfuric acid as an electrolyte has a drastic influence on oxide formation in PEO. With 1% or 3% additional phosphoric acid, higher potentials are necessary to sustain the constant current density. This indicates an enhanced dielectric layer in terms of electrical resistance, with either increased thickness or higher electrical resistivity. The SEM cross sections confirm an increasing thickness. The promoted oxidation is associated with the mass gain in Figure 1b, which is mainly negative in 12.9 M H2SO4, partly positive with 1% H3PO4, and positive for all investigated charge densities with 3% H3PO4.

While oxide formation is promoted, breakdown effects are inhibited by adding H3PO4, which results in diminished breakdowns [10,14]. One indicator is the suppressed surface destruction, as shown in Figure 3. Breakdowns may cause plasma oxidation; however, almost all electric energy is used for ionization, water vaporization, joule heating, and gas evolution, especially in the microarc region [18]. As a consequence, plateaus and cracks evolve on the surface. After adding H3PO4 to the electrolyte, these effects decrease. Closely related is the mentioned mass loss in Figure 1b. It is assumed that the mass loss is promoted by breakdowns. Since the mass change becomes positive with additional H3PO4, lower breakdown intensity is expected.

Furthermore, reduced breakdown intensity can be identified from the phase composition. Rutile is a high-temperature modification and formed from anatase in an irreversible, time-dependent transformation [6]. In PEO, the energy for transformation is brought into the system by electrical breakdowns. When breakdowns diminish in the process, transformation is reduced or even disabled [7]. Results derived from the XRD analysis in Figure 2 show that the rutile fraction decreases with an increasing amount of H3PO4. Confocal Raman microscopy data, as presented in Figure 6, confirm this observation. Our results suggest that the energy liberated by breakdowns decreases with an increasing amount of H3PO4. With even higher concentrations of H3PO4, it is assumed that the crystallization is further inhibited and that even the anatase fraction is reduced [19].

In our discussion of the impact of additional H3PO4, we must consider that the concentrations of both acids decrease if one mixes appropriate amounts of H2SO4 (75 wt%) and H3PO4 (25 wt%). For example, when adjusting a fraction of n(H3PO4)/n(H2SO4) = 3%, the resulting concentration of H2SO4 is 11.4 M, while the concentration of H3PO4 is 0.3 M. To understand the dilution influence of H2SO4, we conceived a PEO experiment with 11.4 M H2SO4. As a result, no significant change in the breakdown character was observed. In conclusion, the additional H3PO4 causes the drastic change observed in the PEO process.

To rationalize our findings, we correlated the impact of additional H3PO4 in 12.9 M H2SO4 to the corrosion of titanium in the electrolyte. In pure H2SO4 and the absence of an applied potential, the maximum corrosion rate is between 12.5 [20] and 13.7 M [21]. With an applied concentration of 12.9 M, a strong chemical etching can therefore be expected. In contrast, a simple corrosion experiment without applied potential in 12.9 M H2SO4 with 3% H3PO4 clearly demonstrates that the corrosion rate drastically decreases. A possible explanation is that solvated H3PO4 anions preferentially adsorb onto the anodic area. A similar effect is known to occur for corrosion inhibitors in H2SO4 [20]. As a consequence of the preferential adsorption, the active area is blocked for the more reactive H2SO4 anions.

The concept of the described corrosion behavior can be applied to our PEO experiment. In 12.9 M, the breakdowns are exceptionally strong, whereas the breakdowns are suppressed in a mixture with 3% H3PO4. We assume that, during PEO, the preferential adsorption of H3PO4 anions blocks the anodic surface for H2SO4 anions, which tend to favor breakdowns instead of ion migration. Hence, destructive breakdowns and the related corrosion are suppressed, which directly reduces the mass loss in the process, as confirmed by our mass investigation in Figure 1b. Leach and Sidgwick proposed that the different behaviors of SO2<sup>−</sup> <sup>4</sup> and PO3<sup>−</sup> <sup>4</sup> are the result of their different molecular charges [22], which is plausible since the electrical field in PEO is exceptionally high.

Furthermore, the high H3PO4 anion adsorption to the anodic surface leads to the preferred incorporation of H3PO4 anions. This is in accordance with our EDX investigation, which reveals an increased concentration of phosphor species compared with the low sulfur concentration in the oxide. Similarly, earlier studies have confirmed that phosphorous or phosphate ions penetrate more easily through the titanium oxide layer during anodization compared with sulfur or sulfate ions [19]. Certainly, these phosphorous species contribute to the high mass gain in Figure 1b for electrolytes with additional H3PO4.

With the favored incorporation of H3PO4 anions, enhanced ion migration can be assumed, which increases the fraction of the ionic current. Because of the constant current mode in our process, the total current density always remains constant. When ionic migration is promoted, the total current in the process comprises less electronic current, i.e., breakdowns. Therefore, the additional H3PO4 reinforces the inhibition of breakdowns.

The reduced electronic current has a major impact on the oxide layer. According to the literature, two effects limit the lifetime of a discharge. On the one hand, the expansion of gases in the channel lead to cooling, and the plasma collapses [23]. On the other hand, the gas forms a bubble on top of the channel; thus, it increases the electrical resistivity of the channel and terminates the plasma [4,18]. The formation of a new plasma is prevented while the gas is inside and above the channel. When the gas escapes, the electrolyte fills the void [3]. As a consequence, the next discharge is created in the same channel. This hypothesis predicts that the current per channel per discharge is limited by the lifetime of the plasma channel.

According to our results, more charge compensation in the form of breakdowns occurs in pure H2SO4 compared with mixtures containing 1% or 3% H3PO4. As a consequence of the limited charge transfer per discharge in a single channel, more channels are necessary for higher charge compensation. Hence, the pore density increases compared with surfaces prepared in mixtures with 1% or 3% H3PO4, as shown in the SEM images in Figure 3. On the other hand, fewer channels for charge compensation by breakdowns are necessary in mixtures with H3PO4. Hence, the pore density decreases. However, the pore size increases in mixtures with H3PO4 because of recurring breakdowns in identical channels.

It should be noted that anatase is the predominant phase in the compact layer for all investigated samples. Previous studies confirm our conclusion that the thin compact layer is composed of nanocrystalline anatase [8,24]. The dense compact layer likely evolves from a temperature gradient [9]. The thermal mass of the titanium substrate enables the rapid cooling of the plasma. Hence, the time is too short and the temperature is too low for the phase transformation to rutile for an anatase-to-rutile transformation, and therefore, only a nanocrystalline anatase structure evolves. Also remarkable is the increased sulfur concentration in the compact layer, as seen in Figure 5. This may result from the electrolyte becoming trapped in interfacial nanopores during the fast cooling [14].

Another notable feature is apparent in the cross sections of several samples, which can be identified in our study in three ways. First, SEM cross section images, especially in Figure 4c, reveal a rather smooth morphology between the porous layer and the surface. Second, the EDX cross sections in Figures 5a,c show a different elemental composition compared with the porous layer in the previously mentioned region. Third, the Raman investigations in Figures 6a,c reveal a distinct change in the phase composition near the surface. We assume that this region originates from molten titanium, which is ejected from the channels and rapidly cools down in the vicinity of the electrolyte [25]. Because of quenching, oxide formation is suppressed, and consequently, oxide stoichiometry is not achieved. For this reason, we expect a predominant amorphous structure with small contributions of titanium dioxide phases.

#### **5. Conclusions**

Oxide coatings on titanium were produced in a galvanostatically controlled PEO process with a constant current density. With a concentrated 12.9 M H2SO4 electrolyte as the starting material, small amounts of H3PO4 were added to investigate the impact on oxide formation. Pure 12.9 M H2SO4 is highly suitable for promoting breakdowns. With a higher charge density, breakdowns cause a destructive reforming and induce the phase transition from predominantly anatase to almost entirely rutile. Upon reaching 35 C/cm2, the rutile to anatase fraction is over 90%.

By adding small amounts of H3PO4, i.e., 1% or 3%, breakdown effects are drastically reduced. The rutile fraction in the process does not exceed 11%, even for charge transfers as high as 89 C/cm2. The drastic change is explained by the preferential adsorption of H3PO4 anions to the anodic area. Therefore, the H3PO4 anions block the surface for more reactive H2SO4 anions and suppress breakdowns. The enhanced concentration of H3PO4 anions at the surface reinforces their incorporation and, consequently, migration in the oxide layer. Therefore, the ionic current in the process increases, and the electronic current fraction, including breakdowns, decreases. With fewer breakdowns, the destructive reforming diminishes. Moreover, the increase in ion migration promotes oxide formation with increased thickness and mass gain of the oxide layers, which are produced in mixtures with H3PO4.

**Author Contributions:** B.F. designed, conceived, and performed the confocal Raman microscopy; B.E. designed, conceived, and performed the other experiments; B.E. validated and analyzed the results; K.S. supervised the project; B.E. wrote the paper. All authors have read and agreed to the published version of the manuscript.

**Funding:** We acknowledge support by the Heinrich Heine University Duesseldorf. This research was also supported by Bundesministerium für Wirtschaft und Energie (BMWi) under project no. ZF4185502ZG6.

**Acknowledgments:** We gratefully thank Denis Netschitailo for supplementary PEO experiments and related discussions.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **LDH Post-Treatment of Flash PEO Coatings**

#### **Rubén del Olmo 1,\*, Marta Mohedano 1, Beatriz Mingo 2, Raúl Arrabal <sup>1</sup> and Endzhe Matykina <sup>1</sup>**


Received: 25 April 2019; Accepted: 28 May 2019; Published: 30 May 2019

**Abstract:** This work investigates environmentally friendly alternatives to toxic and carcinogenic Cr (VI)-based surface treatments for aluminium alloys. It is focused on multifunctional thin or flash plasma electrolytic oxidation (PEO)-layered double hydroxides (LDH) coatings. Three PEO coatings developed under a current-controlled mode based on aluminate, silicate and phosphate were selected from 31 processes (with different combinations of electrolytes, electrical conditions and time) according to corrosive behavior and energy consumption. In situ Zn-Al LDH was optimized in terms of chemical composition and exposure time on the bulk material, then applied to the selected PEO coatings. The structure, morphology and composition of PEO coatings with and without Zn-Al-LDH were characterized using XRD, SEM and EDS. Thicker and more porous PEO coatings revealed higher amounts of LDH flakes on their surfaces. The corrosive behavior of the coatings was studied by electrochemical impedance spectroscopy (EIS). The corrosion resistance was enhanced considerably after the PEO coatings formation in comparison with bulk material. Corrosion resistance was not affected after the LDH treatment, which can be considered as a first step in achieving active protection systems by posterior incorporation of green corrosion inhibitors.

**Keywords:** PEO; LDH; active protection; corrosion; aluminium

#### **1. Introduction**

Plasma electrolytic oxidation (PEO) is a plasma-assisted electrochemical surface treatment characterized by the utilization of high voltages (100–600 V) in alkaline electrolytes to produce ceramic-like coatings on light alloys such as aluminium [1], magnesium [2] and titanium [3]. This technology may become a potential alternative to conventional toxic and highly carcinogenic chromic acid anodizing (CAA) for niche applications [4–6].

The PEO process can be conducted under direct current (DC) [7], alternate current (AC) [8], unipolar [9] or bipolar pulsed regimes [10] involving polarization of the alloy to voltages above the dielectric breakdown of the oxide. This results in the formation of multiple short-lived microdischarges on the metallic surface that trigger the formation of highly stable ceramic phases [1,11,12]. The resulting PEO coatings present high hardness, thermal stability and adherence to the substrate, which lead to enhanced corrosion and tribological properties. Moreover, the coatings' microstructures and compositions can be tailored as required by controlling electrical parameters during the coating synthesis by electrolyte selection and by the application of pre- and post-treatments [13,14].

However, the costs associated with PEO technology are relatively high, which is mainly due to high energy consumption. General current densities and voltages for the PEO process are in the range of 1.5–15 A·dm−<sup>2</sup> and 400–500 V for Mg [15], 6–20 A·dm−<sup>2</sup> and 200–500 V for Ti [16] and 10–60 A·dm−<sup>2</sup> and 200–400 V for Al [17]. Typical PEO treatment times are within a 15–60 min range.

In the particular case of Al and its alloys, the efficiency of conversion of the anodic charge into the final coating material may be as low as 20%, due to gas generation and dissolution processes [18–20]. Therefore, energy consumption of the process must be reduced in order to make this technology commercially viable for high-volume applications [10,14,20,21]. Strategies to achieve this goal involve waveform design and cell geometry [22], electrolyte design [23–27]) and a pre-anodizing approach [18]. An additional strategy consists of "flash" PEO—processes of a short duration (1–3 min) that produce thin coatings (1–5 μm) suitable and sufficient for many corrosion resistance-demanding applications.

PEO coating properties can be improved even further by incorporating active functionalities. This can be achieved by introducing smart agents into the coatings matrix, which are released in situ, triggered by an external stimulus [23–25]. Coatings based on LDH have increasing potential not only for developing smart functionalities on ceramic materials, but in many other fields, such as biomaterials, due to their low toxicity [23].

LDHs can be expressed by the general formula [M2<sup>+</sup>1−*<sup>X</sup>*M3<sup>+</sup>*X*(OH)2] *<sup>X</sup>*+(A*m*−)*X*/*m*·*n*H2O], where M<sup>2</sup><sup>+</sup> and M3<sup>+</sup> represent divalent metallic cations (e.g., Mg2<sup>+</sup>, Zn2<sup>+</sup>, etc.) and trivalent metallic cations (e.g., Al3<sup>+</sup>, Co3<sup>+</sup>, etc.), respectively, and A*m*<sup>−</sup> is the interlayer anion (e.g., NO3 −, Cl−, etc.) [26]. It consists of uniform flakes that grow roughly vertically on the substrate and act as intelligent nanocontainers that release, in a controlled way, the previously loaded corrosion inhibitors. The active corrosion protection mechanism is based on anion-exchange reactions induced by particular triggers, such as increases in the aggressiveness of the surrounding environment or initiation of the corrosion process [26,27].

The most common approach to synthesizing LDH is the in situ growth method, due to its low cost and ease of synthesizing at laboratory and industrial scales [28]. In this method, the source of Al3<sup>+</sup> ions required to grow the LDH structure is the metallic substrate, and it results in coatings with better adherence compared to LDH coatings synthesized by other routes such as hydrothermal treatment [29–31], which is characterized by the use of autoclave to get high temperatures, or the urea hydrolysis method, which is used to form well-crystallized large LDH [32–34].

The knowledge of multifunctional surfaces with active protection based on ceramic-like coatings (PEO) is at an embryonic stage [35]. Y. Zhang [36] investigated the growth behavior of LDH layers on PEO-coated aluminum alloy. It was found that LDH grains are preferentially formed on the micro-pores of PEO coatings to provide effective film repairs. The most dominant factor which determines the PEO/LDH system behavior is the interaction between the electrolyte anions and Al(OH)2 <sup>+</sup> cations from PEO coating surface. This phenomenon has a major impact on the kinetic mechanism of the formation of LDH coating [37,38]. F. Chen [39] demonstrated that flakes were preferentially formed in the PEO pores at the initial stage of LDHs growth; it was also found that long-term corrosion resistance was significantly enhanced due to the limited penetration of corrosive ions [38]. The corrosive degradation of the substrate coated with PEO is accelerated by chloride ions in the environment. From a corrosion-relevance perspective, the entrapment of chloride anions via the ion-exchange mechanism is the most critical factor for designing efficient corrosion protection strategies. In fact, partial entrapment is possible when the PEO layer is sealed with LDH [40].

Although the concept of LDH-based post-treatment for sealing the PEO layers has been recently proven, many variables remain unexplored. In particular, this work is focused on the formation of in situ Zn-Al LDH on low energy consumption PEO coatings. For that, three PEO coatings were selected from 31 processes (with different combination of electrolytes, electrical conditions and time) based on corrosive behavior and energy consumption. The selected materials were thoroughly characterized and evaluated in terms of corrosion resistance.

#### **2. Materials and Methods**

#### *2.1. Material*

The composition of the 1050-H18 aluminum alloy (Famimetal S.L., Madrid, Spain) in wt.% is: 0.07 Zn, 0.05 Mn, 0.25 Si, 0.05 Cu, 0.40 Fe, 0.05 Mg, 0.05 Ti, 0.03 V and Al balance.

#### *2.2. Specimens Preparation*

The samples were cut from sheets into specimens of 30 <sup>×</sup> 20 <sup>×</sup> 1.5 mm3 dimensions, ground to P1200 silicon carbide abrasive paper, rinsed in distilled water and methanol and dried in warm air. Prior to the PEO processing, specimens were etched in 15 wt.% sodium hydroxide solution for 20 s, rinsed in deionized water and dried in warm air. The working area was then limited to <sup>∼</sup>3 cm2 using a commercial resin (Lacquer 45, MacDermid plc., Birmingham, UK).

#### *2.3. Surface Treatment Based on PEO*

PEO coatings were obtained using different electrolytes and conditions in order to grow thin PEO coatings with low energy consumption (Table 1) under vigorous agitation. The experimental system was carried out with a DC voltage/current-controlled power supply (SM400-AR-8 Systems electronic) equipped with a thermostatic jacket (25 ± 1 ◦C) under continuous electrolyte agitation. An AISI 316 steel plate of 7.5 <sup>×</sup> 15 cm<sup>2</sup> size was used as a counter electrode. After the PEO process, all samples were rinsed in distilled water and isopropanol and dried in warm air.


PEO treatment conditions: *V* (V): 400; *j* (A·cm<sup>−</sup>2): 0.1; *t* (s): 180; PEO treatment conditions \*: *V* (V): 350; *j* (A·cm<sup>−</sup>2): 0.05; *t* (s): 200; \*\* Na2SiO3 alludes to the use of water glass with specific gravity of 1.3 g/L.

#### *2.4. Synthesis of Zn-Al-LDH Growth*

Zn-Al LDH-nitrate (LDH-NO3) was synthesized on AA1050-H18 aluminium alloy. The specimens were immersed in the solution for different times under continuous stirring in order to form LDH (Table 2), then rinsed in deionized water and dried in air at room temperature.


All treatments were developed in 100 mL of aqueous solution at 95 ◦C. pH values were adjusted to 6.5 using 1 vol.% ammonia.

#### *2.5. Characterization*

Planar and cross-sectional views of the specimens were examined using a JEOL JSM 6335F (Tokyo, Japan) field emission scanning electron microscope (FESEM) working at 20 kV and equipped with an energy dispersive (EDS) spectrometer (OXFORD X-MAX, Oxford, UK). Coating cross-sections were ground through successive grades of SiC paper and polished to a 1 μm diamond finish.

Phase composition was examined by X-ray diffraction (XRD), with a Philips X'Pert MRD (Amsterdam, The Netherlands, Cu Kα = 1.54056 Å). The XRD patterns were taken using grazing incidence with a step size of (0.01◦–1◦) and a dwell time of 6 s per step at room temperature.

The specific energy consumption was calculated by integration of the voltage-time and current-time transients acquired by the power supply (SM 400AR-8 Systems electronic) during the PEO treatment (Equation (1)). With the obtained results, the energy consumption was calculated in terms of kW·h·m−<sup>2</sup> according to Equation (1), and the specific energy consumption in kW·h·m−2·μm−<sup>1</sup> was obtained dividing *P*tot by the coating thickness.

$$P\_{\rm tot} = \int\_{t\_0}^{t} [V \cdot j] \left[ \frac{\mathcal{W} \cdot \mathbf{s}}{m^2} \right] \tag{1}$$

#### *2.6. Electrochemical Behavior*

Electrochemical impedance spectroscopy (EIS) was used to evaluate the corrosion resistance of the different coatings in an aqueous saline solution (NaCl 3.5 wt.%) at 25 ◦C. For that, a GillAC (ACM Instruments, Cumbria, UK) computer-controlled potentiostat and a three-electrode cell were used. The specimen was connected as a working electrode with an exposed area of 1 cm2. A graphite electrode and a silver–silver chloride (Ag/AgCl) electrode used as the counter and the reference electrode, respectively. The solution inside the reference electrode was KCl 3 M, which provided a potential of 0.210 V with respect to the standard hydrogen electrode. The tests applying a sinusoidal perturbation of 10 mV RMS amplitude in the frequency range of 30 kHz–0.01 Hz were carried out after 1 h of immersion. All measurements were duplicated to ensure reproducibility.

#### **3. Results and Discussion**

#### *3.1. PEO Coating Screening*

The first screening process to select one PEO coating per electrolyte type was conducted in accordance with three factors: (i) presence of microdischarges during coating formation, (ii) visually uniform coating morphology and (iii) coating thickness ≥1 μm.

Among the different alternatives based on different electrolyte compositions, PEO coatings developed in phosphate and silicate electrolytes showed more promising results (with lower breakdown voltage values, beneficial for low energy consumption) (Table 3) and therefore additional process conditions were also tried (*<sup>V</sup>* (V): 350; *<sup>j</sup>* (A·cm<sup>−</sup>2): 0.05; *<sup>t</sup>* (s): 200) (Table 1).


**Table 3.** Energy consumption values and electrolytes characteristic of selected flash PEO coatings.

Finally, only the A3.1, A3.2, P2.1, S1.3, S2.6 and S4 PEO coatings (Table 1) fulfilled the previous requirements.

The last step of the screening process consisted of a corrosion evaluation based on the value of the modulus of the impedance obtained by electrochemical impedance spectroscopy (EIS). Figure 1 shows the |Z| <sup>×</sup> 10−<sup>2</sup> Hz values of the studied materials, providing an estimation of the corrosion resistance, where higher values of |Z| indicate a lower corrosion rate [41].

**Figure 1.** Scatter diagram of impedance modulus at 10−<sup>2</sup> Hz; selected PEO coatings on AA1050 alloy, citing two measurements per condition.

With the aim of studying the influence of electrolytes on LDH growth, three PEO coatings were selected (one per electrolyte composition: aluminate A3.1, phosphate 2.1 and silicate S4). The specific energy consumption of selected coatings was calculated (Table 3) by integration of the voltage-time and current-time transients (Figure 2) recorded during the PEO process to verify that the developed coatings were energy efficient. The obtained values reveal the effect of electrolyte composition on energy consumption, coating growth rate and breakdown voltage values (Table 3).

**Figure 2.** Voltage-current time curves for (**a**) A3.1, (**b**) P2.1 and (**c**) S4 PEO coatings on AA1050 alloy.

As can be seen in Figure 2, the current drop was observed only in aluminate and silicate electrolyte cases (Figure 2a,c) when 400 and 350 V limitations (Table 1) were achieved after 40 and 75 s of treatment, respectively, and the power supply switched to a constant voltage-control mode. Typically, when the current density was below 20 mA·cm−2, the microdischarges extinguished; the anodizing, however, was carried on till set time in order to repair microdefects in the oxide material [42].

The fact that the limiting voltage was reached (hence the current drop) could be explained by high coating density and, consequently, higher resistance of the oxide to charge and mass transfer that were achieved at early stages [18]. High value of energy consumption in the case of aluminate is mainly due to the fact that aluminate species in the electrolyte gave rise to the formation of coating composed of nearly pure alumina, which has very low electron conductivity (i.e., the current flows mainly by ion and not by electron transfer) [12]. Further, the high breakdown voltage (320 V) in the aluminate electrolyte compelled the use of a higher voltage limit (400 V) in order to ensure a long enough period of sparking in order to achieve a uniform coating of a significant thickness; this yielded a higher specific energy consumption value.

As a result, the dielectric breakdown voltage was high, the limiting set voltage was achieved quickly and, as a consequence of the current drop, the sparking period was short, hence the low coating growth rate [22,43]. Similarly, high value of energy consumption in the case of phosphate electrolyte, where sparking was observed until the end of the treatment and coating growth rate was relatively high, was due to the absence of current drop, because the limiting 350 V were never reached. In this case, higher resistance of the oxide to charge and mass transfer were achieved at 60 s, giving rise to intense microdischarges and voltage variations, and therefore the treatment was stopped at 115 s in order to maintain coating uniformity.

The lowest energy consumption of 2.2 kW·h·m−2·μm−<sup>1</sup> was achieved in case of silicate electrolyte. This was mainly the result of its high electrical conductivity and, therefore, low *U*bd. The onset of microdischarges early in the treatment and the relatively long sparking period before current decay resulted in the intermediate coating growth rate value (Table 3).

The specific energy consumption values obtained under DC conditions in the present work are similar to those reported in studies carried out under AC conditions. For example, E. Matykina et al. developed a PEO coating on pure aluminium using silicate electrolyte, and obtained a growth rate of 1.3 <sup>μ</sup>m·min−<sup>1</sup> and energy consumption values of 4.77 KW·h·m−2·μm−<sup>1</sup> [18]. Y.L. Cheng et al. reported a growth rate of 11.3 <sup>μ</sup>m·min−<sup>1</sup> and energy consumption values of 5 KW·h·m−2·μm−<sup>1</sup> for Al-Cu alloy in concentrated aluminate electrolyte [44]. It is well known that DC conditions promote low growth rates in comparison with AC conditions [18,22]; however, in this study the values obtained were considerably lower compared with the available data for different PEO treatments on commercial Al alloys, which can be as high as 26.7 kW·h·m−2·μm−<sup>1</sup> [18]. The present findings demonstrate that in order to reduce specific energy consumption under DC conditions it is necessary to (i) limit the final forming voltage that ensures a current drop, and (ii) use electrolytes with conductivity, which ensures low microdischarges onset voltages and extended sparking periods, as in the case of the S4 electrolyte.

#### *3.2. PEO Coatings Characterization*

Figure 3 shows the planar view and cross-section scanning electron micrographs of AA1050 coated by selected PEO coatings. All selected treatments show a thin oxide layer of 1–2.5 μm (Table 3). This is particularly evident in aluminate electrolyte-based PEO coating (A3.1), where the Al-Fe intermetallic compounds from the substrate are still visible in the coating (Figure 3a, inset) due to its low thickness. This coating is also more heterogeneous (Figure 3b) than the rest, which is attributable to its high breakdown voltage values (Figure 2) [37,45,46] and, as mentioned before, its low coating growth rate. Phosphate electrolyte-based PEO coating (P2.1) showed a homogeneous (Figure 3c) surface appearance (Figure 3c), and the highest thickness value (Figure 3d). This was mainly due to the presence of polyphosphate species that participated in PEO coating formation and favored its high coating growth rate [47]. Silicate electrolyte-based PEO coating (S4) (Figure 3e), with the lowest breakdown voltage, showed a homogeneous surface morphology with very sparse submicrometric pores. The latter may be attributable to the formation of a thin superficial glassy layer of SiO2, which can be surmised from the EDS analysis where the presence of 1.5 at.% Si in the coatings and a greater content of oxygen than in the other two coatings was confirmed.


**Figure 3.** Planar view (**a**,**c**,**e**) and cross-section view (**b**,**d**,**f**) of secondary electron images of the PEO coatings (A3.1, P2.1, S4), respectively. EDS analysis was performed on the areas corresponding to the planar views of the coatings.

#### *3.3. LDH Screening*

The effect of reactant composition and treatment time during the growth of LDH coatings were investigated. Figure 4 depicts the XRD patterns of the different LDH treatments (Table 2) grown on the bulk material. The presence of peaks at 9.6◦ and 19.9◦ corresponding to the characteristic (003) and (006) reflections of LDHs intercalated with NO3<sup>−</sup> [32,48,49] indicates the formation of LDH under the different conditions.

**Figure 4.** XRD patterns of Zn-Al-LDH-coated AA1050 alloy at different conditions.

It was revealed that treatments containing NaNO3 in the solution led to more defined and intense peaks, probably due to the presence of sodium ions in the LDH gallery [50]. On the contrary, the presence of NH4NO3 drove the formation of broadened peaks [51]. In fact, just in the case of LDH grown in NH4NO3, a small peak was revealed at 9.9◦ that could be associated with an LDH phase intercalated with carbonate, due to the formation of LDH layers under atmospheric conditions [52].

LDHs formed under long treatment times (LDH 2 and 4) showed very strong peaks in comparison with LDHs formed under short treatment times (LDH 1 and 3), mainly because an increment in the LDH degree of crystallinity took place [51].

The correlation between XRD patterns of studied LDH coatings and planar view scanning electron micrographs were investigated. Figure 5 shows secondary electron images of the LDH treatments (Table 2) grown on the pure aluminium.

The typical flake-like LDH structure could be clearly observed for LDH carried out in NH4NO3 at short treatment times, whereas LDH developed in NaNO3 formed this structure at long exposure times. According to Figure 5a, the LDH structure carried out in the presence of NH4NO3 is in good agreement with the XRD patterns that showed broadened peaks and, consequently, a highly open LDH structure. Additionally, Figure 5b clearly shows (also consistent with the XRD pattern) a non-defined LDH structure, which is usually attributed to the incorporation of carbonate ions into the LDH gallery [48]. However, LDH developed in NaNO3 showed typical curved plate-like LDH microcrystals at long exposure times, and a non-defined structure at short exposure times (Figure 5c,d). This was mainly due to an increment in the degree of LDH crystallinity that was observed in XRD patterns (Figure 4).

In order to evaluate the correlation between the corrosion protection and the structure of studied LDHs, a screening process based on corrosion performance (EIS) was carried out. Figure 6 depicts the Bode and Nyquist plots for AA1050 alloy with studied LDH coatings.

**Figure 5.** Secondary electron images of the LDH 1 (**a**,**b**) LDH 2 (**c**,**d**), LDH 3 (**e**,**f**) and LDH 4 (**g**,**h**) coatings.

**Figure 6.** Bode plots for AA1050 alloy with studied LDH coatings.

From the point of view of coating structure, the presence of spheroidal particles in flake-like LDH 1 and LDH 4 (Figure 5) [14] was associated with the presence of secondary phases [35–37] that favored aluminium cation dissolution due to their highly cathodic behavior [48]. In addition, the porous structure of these spheroidal particles also favored Cl− anion penetration into the LDH gallery, and for this reason these coatings showed the lowest corrosion protection among the studied LDH coatings (Figure 6) [53–55]. From the point of view of corrosion protection of non-defined LDH coatings, it should be noted that LDH 2 showed similar corrosion behavior in comparison with LDH 4, which can be attributed to its intermediate non-porous LDH structure (Figure 5). On the contrary, LDH 3 provided a beneficial effect to corrosion protection in comparison with all the studied LDH coatings. This may be due to the presence of sodium ions in the LDH gallery, which facilitate the formation of non-porous corrosion-protective LDH coating (Figure 6). For this reason, LDH 3 was the selected treatment to use for the selected PEO coatings and study their anti-corrosion properties (Figure 6).

#### *3.4. PEO-LDH Coating Characterization*

Figure 7 depicts the XRD patterns of selected PEO coatings (A3.1, P2.1 and S4) with LDH 3 and discloses the effects of the PEO coating compositions on LDH growth.

**Figure 7.** XRD patterns of Zn-Al-LDH-coated A3.1, P2.1 and S4 PEO coatings.

PEO coatings based on aluminate and silicate electrolytes showed the typical peaks at 9.6◦ and 19.9◦, which corresponded to the characteristic (003) and (006) reflections of LDH intercalated with NO3 <sup>−</sup> [32,48,49]. Additionally, the presence of ZnO and Al2Si2O5(OH)4 characteristic peaks in the S4-LDH XRD pattern came from LDH chemical composition (Table 2) and silicate electrolytes, respectively (Table 1).

In the particular case of PEO coating developed in phosphate electrolyte, there were no peaks detected in that range. This could be attributed to the formation of non-crystalline phases, or to only a small amount that could not be detected at the selected scan rate (Figure 7).

Figure 8 highlights a detailed morphology of the selected PEO-LDH coatings, which reveals the importance of PEO coating composition. The characteristic flake-like LDH structure can be clearly observed for the A3.1-LDH coating, in which LDH flakes are covering the whole PEO coating (Figure 8a,b). This is also observable in the S4-LDH coating, but in this case is more heterogenous (Figure 8e,f). These results are in accordance with XRD patterns that showed the presence of these characteristic reflections (Figure 7). However, for the P2.1-LDH coating there is a drastic decrease of the density of LDH-like flakes (Figure 8c,d), which is in accordance with the XRD patterns (Figure 7).

**Figure 8.** Secondary electron images of the planar view of the PEO coatings after LDH 3 coating formation: (**a**,**b**) A3.1, (**c**,**d**) P2.1, (**e**,**f**) S4.

This fact is in strong agreement with a high dependence on the availability of Al(OH)2 <sup>+</sup> cations necessary to form Zn-Al-LDH. According to previous studies [40,56], Zn-Al LDH synthesis can be explained via the following chemical reactions:

$$\text{Al}\_2\text{O}\_3 + 3\text{H}\_2\text{O} \to 2\text{Al}(\text{OH})\_3 \tag{2}$$

$$\text{Al(OH)}\_{3} + \text{NH}\_{4}^{+} \rightarrow \text{Al(OH)}\_{2}^{+} + \text{NH}\_{3}\text{H}\_{2}\text{O} \tag{3}$$

$$\text{Zn(OH)} + \text{Al(OH)}\_{2}^{+} + 2\text{NO}\_{3} \rightarrow \text{LDH} - \text{NO}\_{3} \tag{4}$$

As mentioned before, the in situ growth method was used in the present work and, consequently, Al(OH)2 <sup>+</sup> cations were an essential requirement to form the LDH layers. Due to the porous structure of PEO coatings and their compositions, two sources can provide Al(OH)2 <sup>+</sup> cations: (i) the aluminium metal matrix (due to the electrochemical interactions with LDH solutions), and PEO coating thickness [57].

Firstly, in consideration of the PEO coating cross-section (Figure 3) and planar view micrographs after LDH treatment (Figure 8), it can be concluded that the amount of LDH flakes on the selected PEO coatings was highest for A3.1, and lowest for the P2.1 and S4 coatings. This could firstly be explained by thickness, and secondly by the chemical composition of the PEO coating surfaces. Due to the low thickness of the A3.1 PEO coating (~1 μm) (Figure 3a,b), the migration capacity of Al(OH)2 <sup>+</sup> cations from the aluminium metal matrix towards the coating surface was ensured. However, the P2.1 (Figure 3c,d) and S4 (Figure 3e,f) PEO coatings showed higher thickness values (~1.5–2 μm), and, consequently, the migration capacity of Al(OH)2 <sup>+</sup> was reduced. With respect to the chemical composition of PEO coating surfaces, the aluminium content decreased in the order A3.1 > P2.1 > S4 (Figure 3, EDS analysis table) due to the presence of aluminate species in the A3.1 electrolyte. Further, greater charge passed during the P2.1 treatment, and intense sparking from the onset of the microdischarges until the end of the treatment (Figure 2b) resulted in greater thickness and density of the P2.1 coatings. Consequently, a highly heterogeneous LDH layer was achieved

in comparison with the S4 PEO coating. This justifies the absence of peaks in XRD patterns of the P2.1-LDH (Figure 7).

#### *3.5. Corrosion Resistance of PEO* + *LDH Coatings*

In order to evaluate the effect of LDH formation on selected PEO coatings, corrosion resistance was measured by electrochemical impedance spectroscopy (EIS) for 1 h of immersion in 3.5 wt.% NaCl solution at room temperature (Figure 9).

PEO-LDH treatments

**Figure 9.** Scatter diagram of impedance modulus at 10−<sup>2</sup> Hz; selected PEO-LDH coatings on AA1050 alloy.

As can be seen in Figure 9, PEO coatings with and without LDH treatment showed high similitude of the impedance modulus.

It is important to note that, in this work, the corrosion resistance of different PEO coatings was strongly connected with PEO coating porosity, because the lack of pores restricts the Cl− penetration and prevents its detrimental effects [58]. For this reason, A3.1 showed the lowest corrosion resistance due to the combination of pores and heterogeneities in comparison with the P2.1 and S4 coatings. The highest corrosion protection exhibited by the S4 PEO coating (~5 <sup>×</sup> <sup>10</sup><sup>6</sup> <sup>Ω</sup>·cm2) may be attributed to its sparse surface porosity due to the formation of a glassy silica-rich layer.

In this work, the results showed similar behavior of PEO-LDH compared with PEO coatings without any clear improvement after the post-treatment. This could be attributed to several factors, for instance: (i) LDH flake resistance was negligible compared with that of the PEO coating, and (ii) PEO coating lost some of its barrier properties during the formation of LDH, somewhat retracting from the possible small beneficial effect of the LDH layer.

It should be noted that no studies of LDH formation on flash PEO coatings (at <5 min anodizing time) have been reported so far. However, based on this work and few works carried out with non-flash PEO of aluminium, some conclusions can be drawn regarding the effects of LDH post-treatments. For instance, when an LDH layer is not loaded with inhibitors, the corrosion resistance remains unchanged or degrades slightly, as has been shown in [37]. On the other hand, when LDH is intercalated with an inhibitor (e.g., vanadate ions), an improvement in corrosion resistance is observed with immersion time due to an active protection effect [37,38].

The present findings highlight that the development of LDH-container layers did not deteriorate the corrosion resistance of flash PEO coatings, which has a potential for added active protection functionality. Therefore, the first stage of active protection system building can be considered successful. The second stage would consist of loading the LDH scaffold with corrosion inhibitors that would ensure enhanced corrosion protection.

In conclusion, these results are highly relevant for understanding the relation between coherent and uniform LDH layer formations on flash PEO coatings, which is the first step in achieving active protection systems through the incorporation of green corrosion inhibitors into the LDH layer.

#### **4. Conclusions**

The following can be summarized from this preliminary study of LDH growth on flash PEO coatings:


**Author Contributions:** Formal Analysis, R.d.O., M.M., B.M., R.A. and E.M.; Funding Acquisition, M.M.; Investigation, R.d.O. and M.M.; Methodology, R.d.O. and M.M.; Resources, M.M. and E.M.; Supervision, R.A. and E.M.; Writing—Original Draft, R.d.O., M.M. and B.M.; Writing—Review & Editing, R.A. and E.M.; Conceptualization, R.d.O., M.M., B.M., R.A. and E.M.; Software, R.d.O., M.M., B.M., R.A. and E.M.; Validation, R.d.O, M.M., B.M., R.A. and E.M.; Data Curation, R.d.O., M.M. and B.M.; Visualization, R.d.O., M.M., B.M., R.A. and E.M.; Project Administration R.A. and E.M.

**Funding:** This work was partially supported by (MAT2015-73355-JIN), ADITIMAT-CM (S2018/NMT-4411) and RTI2018-096391-B-C33 MCIU/AEI/FEDER, UE. MM is grateful to the Ramon y Cajal Programme (MICINN, Spain, RYC-2017-21843).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Correlation between Defect Density and Corrosion Parameter of Electrochemically Oxidized Aluminum**

**Hao-Ren Lou 1, Dah-Shyang Tsai 1,\* and Chen-Chia Chou <sup>2</sup>**


Received: 5 December 2019; Accepted: 24 December 2019; Published: 27 December 2019

**Abstract:** It has been recognized that a connection may exist between defects of oxide coating and its corrosion protection. Such a link has not been substantiated. We prepare two coatings of anodized aluminum oxide (AAO) and plasma electrolytic oxidation (PEO), and analyze them with Mott-Schottky plots and potentiodynamic polarization scans. The as-grown and annealed AAO coatings exhibit both p-type and n-type semiconductor behaviors. Polarization resistance of the AAO coating increases from (1.8 <sup>±</sup> 1.7) <sup>×</sup> 10<sup>8</sup> to (4.3 <sup>±</sup> 0.5) <sup>×</sup> 10<sup>8</sup> <sup>Ω</sup>·cm2, while corrosion current decreases from (6.1 <sup>±</sup> 3.6) <sup>×</sup> 10−<sup>7</sup> to (2.3 <sup>±</sup> 0.9) <sup>×</sup> 10−<sup>7</sup> A·cm−2, as annealing temperature increases from room temperature to 400 ◦C. The parameter analysis on AAO indicates a positive correlation between corrosion current and donor density, a negative correlation between polarization resistance and donor density. The attempt on correlating corrosion potential gives rise to considerable deviation from a linear fit. The results suggest protection of AAO hinges on its donor density, not acceptor. On the PEO coatings, only the n-type behavior is observed. Intriguingly, the donor density of PEO coating is influenced by the annealing temperature of its pre-anodized layer. The most resistant PEO coating, with pre-anodized and 400 ◦C annealed AAO, exhibits polarization resistance (2.1 <sup>±</sup> 0.4) <sup>×</sup> <sup>10</sup><sup>9</sup> <sup>Ω</sup>·cm<sup>2</sup> and corrosion current (1.7 <sup>±</sup> 0.4) <sup>×</sup> <sup>10</sup>−<sup>8</sup> <sup>A</sup>·cm<sup>−</sup>2.

**Keywords:** anodized aluminum; corrosion resistance; Mott-Schottky analysis; defect; annealing; plasma electrolytic oxidation

#### **1. Introduction**

Electrochemical oxidation offers several value-added attributes to the aluminum surface such as color, hardness, corrosion, and scratch resistances, which enhance its aesthetic and functional purposes. In industrial practice, oxidation of the surface is performed through anodizing or plasma electrolytic oxidation (PEO). Anodizing is commonly carried out in acidic solutions, with an imposed voltage sufficiently low such that electric discharges did not occur. When the imposed voltage is raised to a point that electric discharges emerge and travel on the metal surface, the processing enters the phase of PEO [1–5]. However, the division between anodizing and PEO may not be as clear-cut as described. For example, the treatment of PEO is usually performed in alkaline electrolytic solutions using the constant current mode. The metal surface has to go through a voltage escalating period to reach the state of traveling microdischarges. Thus, PEO is often preceded by a brief period of anodizing.

Anodizing may produce two morphologies of anodic aluminum oxide (AAO): self-ordered porous films and non-porous compact barrier films. In the last two decades, the anodic aluminum oxide composed of regular nanometer pores has attracted tremendous attention, since researchers are in a fervent pursuit of well-defined porous templates that allow them to mold their nanomaterials. The studies on porous-type film have yielded detailed knowledge on how electric current and solution

composition can be varied to control the pore size, the interpore spacing, and even the pore diameter in vertical direction [6,7]. On the other hand, a barrier-type film of planar geometry is also desirable. The compact barrier layer with high dielectric strength finds its applications in the electronic devices of metal-insulator-metal capacitor and the microfluidic devices of electrowetting on dielectrics [8,9].

The morphological dissimilarity between porous- and barrier-type films arises from a high-to-low level of incorporating electrolyte anions (highest for regular-pore films, least for non-porous barrier films). In the acidic electrolyte, the incorporating anions could be SO4 <sup>2</sup>−, C2O4 <sup>2</sup>−, or PO4 <sup>3</sup>−. In the neutral or alkaline electrolyte, the anion is the hydroxyl group [6]. Despite the notable morphology differences [10–12], a few aspects are in common. Those oxygen-carrying anions migrate inward, driven by an imposed electric field. Outward migrating Al3<sup>+</sup> cations diffuse in the opposite direction and both contribute to oxide growth. Both the porous- and the barrier-type films are featured with a barrier layer, which is the oxide adjacent to metallic substrate.

Ion migration in the oxide coating is made possible by point defects. The barrier layer may be viewed as semiconductor because of these frozen point defects [13–16]. When soaked in an electrolytic solution, the barrier layer has two interfaces: the oxide/electrolyte and oxide/metal interfaces. In the AAO literature, researchers concur on the physical picture that one interface is p-type and the other n-type, but cannot agree upon which side is p-type or n-type. The review article of Diggle [17], along with the works of Takahashi [18] and Mibus [19], assumed that the region near oxide/electrolyte interface was anion excess in stoichiometry, therefore, the p-type region. The inner region of oxide/metal interface was excess in metallic cation and n-type behavior. An opposite view was given in the works of Vrublevsky [20–23] and also Benfedda [24], who considered that negative charges such as electrons were trapped at the oxide/electrolyte interface and acted as donors for the n-type behavior. Positive charges, such as holes, were trapped at the oxide/metal interface, acting as acceptors responsible for the p-type behavior.

Researchers have recognized the connection between defects and corrosion and formulated three mechanisms to account for corrosion current via electron and proton conduction [25–27]. Nonetheless, a straightforward correlation of experimental data, to the best of our knowledge, is not reported. Corrosion of the anodized Al alloy 6061 may serve as an excellent example, since 6061 is widely used as structural materials in the aviation and marine industries, and extensively attacked in a chloride containing environment. In this work, we prepare a dense and conformal barrier layer on the 6061 surface. The oxide coating displays the n-type and the p-type behaviors both. We vary the dopant densities with thermal annealing, and show that the corrosion resistance of oxide coating can be correlated to the defect density of donor, not acceptor. Further discussion indicates the external interface of coating is n-type, responsible to the corrosion protection. On the other hand, the inner interface is p-type and is largely related to the growth behavior.

#### **2. Materials and Methods**

Two types of coatings on 6061 aluminum alloy were prepared: the AAO coating and the PEO coating with pre-anodized oxide. Anodization of the pre-cleaned surface was performed with a pulsed current of square bipolar waveform in the aqueous solution of 5.5 g·dm−<sup>3</sup> ammonium pentaborate octahydrate (NH4B5O8·8H2O). The pH value of electrolytic solution was 8.6 and its conductivity 1.45 mS·cm−<sup>1</sup> at room temperature. The electrical parameters of potentiostatic anodization were set as follows: 200 V in positive polarization and 40 V in negative polarization, 50 Hz in frequency, and 40% in duty ratio. This set of electrical parameters were abbreviated as 200 V (+)/40 V (−), which did not give rise to electrical discharges throughout anodization. The frequency was defined as (*T*+on + *T*+off + *T*−on + *T*−off) <sup>−</sup>1, in which *T*+on and *T*−on were the duration periods of positive and negative pulses, respectively. *T*+off and *T*−off were the resting periods between the positive and negative pulses. The duty ratio was defined as *T*+on/(*T*+on + *T*+off + *T*−on + *T*−off). The anodization period of 200 V (+)/40 V (−) was 20 min. Most of anodized samples were annealed, then subject to the galvanostatic PEO treatment. A few anodized samples were etched in sulfuric acid. The parameters of subsequent

PEO were set 45.3 mA·cm−<sup>2</sup> (0.7 A) for positive polarization and 51.8 mA·cm−<sup>2</sup> (0.8 A) for negative polarization, with frequency 50 Hz and duty 40%. This set of PEO parameters was abbreviated as 0.7 A (+)/0.8 A (−). The entire PEO treatment lasted two min and the cell voltage was recorded. In electrochemical oxidations, the sample was mounted at the central position of the electrolytic solution, and the electrical current was sent by a direct current (DC) power supply (DCG-100A, ENI Emerson Electric Co., Saint Louis, MO, USA) with a pulse waveform generator (SPIK-2000A-10H, MELEC, GmbH, Shanghai, China). More details on the setup of anodization and PEO, Figure S1, along with the working procedure, Figure S2, can be found in Supplementary Materials and our previous publication [28].

Annealing of the anodized sample was executed in flowing nitrogen at 100, 150, 250, 300, and 400 ◦C for one hour using a tubular reactor. Etching of a few anodized samples was performed in 2.0 M sulfuric acid for 15, 30, 45, and 60 min. These samples were subject to Mott-Schottky analysis before etching. The analysis procedure was repeated after etching and washing. For Mott-Schottky analysis, the capacitance in aqueous solutions was recorded every 0.05 V between +2.0 and −2.0 V (vs. Ag/AgCl). The value of space charge capacitance was calculated as the inverse of the imaginary component of impedance at 1 kHz, which was recorded with an electrochemical workstation (Autolab PGSTAT302N, Herisau, Switzerland). Given that double-layer capacitance exceeds space charge capacitance sufficiently and the two are in series, the measured capacitance is dominated by the value of space charge capacitance. The three-electrode setup of impedance measurement involved a solution of 5.5 g·dm−<sup>3</sup> NH4B5O8 housed in a 500 mL beaker, with the reference electrode of Ag/AgCl (3.0 M KCl) and the counter electrode of stainless mesh. The working electrode of anodized sample was placed at the beaker center, surrounded by the stainless mesh electrode, and was 7.6 cm in diameter. The Ag/AgCl reference was located adjacent to the working electrode.

The surface morphology was examined with a field-emission scanning electron microscope (SEM, JSM-7900F, JEOL, Tokyo, Japan). The surfaces were metallized with platinum prior to SEM observations. The coating thickness was taken as the average value of six different locations of the mounted specimen. Phase analysis was performed with a wide-angle X-ray diffractometer (D2 phaser, Bruker, Billerica, MA, USA), equipped with a CuKα<sup>1</sup> radiation source and nickel filter. Diffraction results are plotted in Supplementary Materials. Corrosion resistance was evaluated using the technique of potentiodynamic polarization scan. The measurement was done in a solution of 3.5% sodium chloride at room temperature with a three-electrode setup. The setup involved a working electrode with an exposed area 1.0 cm2, a reference electrode Ag/AgCl (1.0 M KCl), and a counter electrode of platinum coated titanium mesh 20 mm × 20 mm. Potentiodynamic polarization data were taken using a 1287A electrochemical interface (Solartron Analytical, Leicester, UK). Current data were recorded between <sup>−</sup>2.0 and 0.0 V at scan rate 5 mV·s<sup>−</sup>1. The corrosion current (*J*corr) and the corrosion potential (*E*corr) were read from the intersection point of anodic and cathodic extrapolated Tafel lines. With the anodic and cathodic Tafel slopes, *b*<sup>a</sup> and *b*c, the polarization resistance (*R*p) was calculated with the Stern-Geary equation as shown in Equation (1).

$$R\_{\rm P} = \frac{b\_{\rm a} \times b\_{\rm c}}{2.303 \times f\_{\rm corr} \times (b\_{\rm a} + b\_{\rm c})} \tag{1}$$

#### **3. Results and Discussion**

#### *3.1. Defect Density and Corrosion Protection of AAO*

Figure 1a,b shows the cross-sectional and top-view images of as-grown barrier layer. As expected, the AAO layer is nonporous and compliant, with minor surface undulations due to of scratches left after polishing. There is no discharge damage found in the oxide since no spark has occurred under 200 V (+)/40 V (−) in the pH 8.6 solution. Thickness of the barrier layer is measured 286 ± 30 nm. The time profiles of positive and negative current, Figure 1c, are consistent with the literature description on self-limiting growth [6]. As time progresses, the positive current density decreases exponentially from

an initial value 16.2 mA·cm−<sup>2</sup> to a steady value 0.65 mA·cm<sup>−</sup>2. Similarly, the negative current descends from an initial value 4.53 mA·cm−<sup>2</sup> to a steady one, 0.65 mA·cm−2. The self-limiting growth occurs when the metal piece is made an anode in the electrolytic solution that furnishes oxygen-containing species, and then a film develops uniformly to oppose the ongoing growth since the dielectric film obstructs diffusion of ions, along with their associated charge-transfer reactions. Consequently, holding the voltage constant, the current diminishes with increasing oxide thickness. If the defect concentration of newly-added oxide is the same with that of grown oxide, the barrier layer resistance increases linearly with the layer thickness and the electric current drops exponentially. Hence, the final thickness of barrier layer is largely determined by the imposed voltage, irrelevant to anodization time. The ratio of layer thickness over applied voltage has been reported 1.1–1.4 nm·V−<sup>1</sup> in literature [9,29], depending on the electrolytic solution. Oxide growth of our 200 V (+)/40 V (−) anodization obeys this rule of thumb, showing a ratio of AAO thickness divided by imposing voltage, ~1.4 nm·V<sup>−</sup>1.

**Figure 1.** Morphological qualities of the AAO layer and the current density profile in anodizing. The images of (**a**) cross-sectional view and (**b**) top view for the barrier layer grown in the solution of 5.5 g·dm−<sup>3</sup> NH4B5O8·8H2O. The associated (**c**) anodizing current is performed with the potentiostatic condition 200 V (+)/40 V (−).

Crystal defects provide the essential diffusion channels during growth. After anodization, these remaining defects are vital to the properties of barrier layer. Thermal annealing is known to diminish the defect density of oxide effectively. Figure 2 shows a series of Mott-Schottky plots for the six samples with low-to-high annealing temperature. For each plot of the inverse square of space charge capacitance *C*SC−<sup>2</sup> versus electrode potential *E*, the acceptor density *N*<sup>a</sup> or the donor density *N*<sup>d</sup> can be extracted from the slope of linear segment, as shown in Equations (2) and (3).

$$\text{C}\_{\text{SC}}^{-2} = \frac{-2}{\text{e} \,\epsilon\_0 \text{e} N\_a} (E - E\_{\text{fb}} - \frac{\text{k}T}{\text{e}}), \text{ p-type} \tag{2}$$

$$C\_{\rm SC}^{-2} = \frac{2}{\epsilon \epsilon\_0 \epsilon N\_d} (E - E\_{\rm fb} - \frac{\rm kT}{\rm e}), \text{ n-type} \tag{3}$$

in which and <sup>0</sup> denote the relative dielectric constant and the vacuum permittivity, κ is the Boltzmann constant, *T* is the absolute temperature, e is the electrical charge and *E*fb is the flat-band potential. For each annealing temperature in Figure 2, one cave-in of the *C*SC<sup>−</sup>2-*E* plot can be found around the electrode potential −0.6 V. Another cave-in may be detected at the more negative potential, related to hydrogen evolution reaction. The cave-in of V-shape indicates the p-type and n-type defects both exist in the barrier layer, since two correlation lines of negative and positive slopes can be drawn. Thus the anodized barrier layer is a p-n heterojunction that may be separated by a neutral region, consistent with the literature. The two slopes generally increase, with increasing annealing temperature. In other words, the acceptor and donor densities decrease. Of the two correlation lines, the intersection potential is assumed to be the flat-band potential value. The *E*fb value shifts with annealing temperature in the positive direction from −1.4 (as-grown), −0.67 (100 ◦C), −0.66 (150 ◦C), −0.53 (250 ◦C), −0.52 (300 ◦C), −0.50 V (400 ◦C), suggesting the defects of barrier layer shift in a systematic manner.

**Figure 2.** Mott-Schottky plots of annealed AAO layers. The inverse square of space charge capacitance is plotted versus electrode potential between −1.8 and 0.6 V for the (**a**) as-grown barrier layer, the barrier layers after annealing at (**b**) 100, (**c**) 150, (**d**) 250, (**e**) 300, (**f**) 400 ◦C for 1 h. The positive and negative slopes of linear segment are marked.

Acceptor and donor densities decrease with increasing annealing temperature, as shown in Figure 3. The acceptor density *N*<sup>a</sup> is higher than the donor density *N*<sup>d</sup> in the as-grown barrier layer. Meanwhile, the acceptor density exhibits a higher temperature dependence than the donor density, that is, annealing decreases the acceptor density faster than the donor density. Annealing at 100 ◦C is sufficient to reduce the acceptor density from 1.7 <sup>×</sup> 10<sup>18</sup> (as-grown) to 5.2 <sup>×</sup> 10<sup>17</sup> cm−3. On the other hand, the donor density decreases slightly from 8.6 <sup>×</sup> 10<sup>17</sup> to 8.4 <sup>×</sup> 1017 cm−3. The significant drop in donor density occurs at the higher annealing temperature of 150 ◦C. Further decline in the defect density is less drastic with increasing temperature. The *N*<sup>a</sup> value of 300 ◦C sample 1.8 <sup>×</sup> 10<sup>16</sup> cm−<sup>3</sup> is near that of 400 ◦C, 1.6 <sup>×</sup> <sup>10</sup><sup>16</sup> cm<sup>−</sup>3. The *<sup>N</sup>*<sup>d</sup> values of 300 and 400 ◦C are similar in magnitude as well, with 5.1 <sup>×</sup> <sup>10</sup><sup>16</sup> (300 ◦C) and 4.4 <sup>×</sup> 1016 cm−<sup>3</sup> (400 ◦C).

Figure 4a presents the typical potentiodynamic polarization curves of annealed barrier layers, in contrast to those of the as-grown barrier layer and the 6061 surface with natural oxide. Comparison of these polarization curves indicates that thermal annealing improves corrosion protection of the barrier layer against 3.5% NaCl solution. Raising the annealing temperature diminishes the corrosion current *J*corr, raises the polarization resistance *R*p, and shifts the corrosion potential *E*corr in the positive direction. The improvement on anticorrosion appears to be progressive among annealed samples. The most visible enhancement is noted between the as-grown barrier layer and the natural oxide on 6061 surface, yet the overall improvement through annealing is also impressive.

**Figure 3.** Defect densities of the AAO layer plotted against the annealing temperature. The defect densities of (**a**) acceptor and (**b**) donor decrease with increasing annealing temperature. The annealing temperature of as-grown barrier layer is assigned to be 30 ◦C.

**Figure 4.** Potentiodynamic polarization curves of the AAO and correlations between corrosion parameters and donor density. (**a**) Potentiodynamic polarization curves at scan rate 5 mV·s−<sup>1</sup> show the anticorrosion trend. Linear correlation is displayed between (**b**) corrosion current, (**c**) polarization resistance of the AAO layer and its donor density. Considerable deviation is shown in the correlation between (**d**) corrosion potential and donor density.

It is particularly intriguing to attain a linear correlation between corrosion current *J*corr and donor density *N*d, as shown in Figure 4b. Such a linear correlation can also be found between polarization resistance *R*<sup>p</sup> and *N*d, Figure 4c. In contrast to *J*corr and *R*p, corrosion potential *E*corr appears less correlated to *N*d, Figure 4d. The above statement is vindicated in the percent deviation of linearity, which is 18% between *E*corr and *N*d, substantially higher than the percent deviation between *J*corr and *N*d, 8.1%, also much higher than the percent deviation between *R*<sup>p</sup> and *N*d, 5.7%. The linear correlation is a strong piece of evidence supporting the view that n-type defects are the origin of chloride anion corrosion. We also try to correlate corrosion parameters with acceptor density and obtain a much higher percent deviation of 32% between *E*corr and *N*a, 38% between *J*corr, and *N*a, 38% between *R*p and *N*a (Figure S3, Supplementary Materials). Evidently, the p-type defects are not the reason of corrosion.

It seems worthwhile to discuss the reason why corrosion potential *E*corr is less correlated to donor density *N*<sup>d</sup> in contrast to the superior correlation between *J*corr and *N*d, and also that between *R*<sup>p</sup> and *N*d. The reported *N*<sup>d</sup> value has been extracted from the linear portion of Mott-Schottky plot, denoting the donor density of the average coating surface. Considering the surface area of our samples, 15.4 cm2, certain inhomogeneity is bound to exist on the coating. A good correlation between *J*corr and *N*<sup>d</sup> suggests that the corrosion current is the sum of individual contribution by the typical donor defects, so is *R*<sup>p</sup> and *N*d. However, the corrosion potential is not a sum of individual potentials of point defects. The *E*corr value is more likely influenced by a small group of point defects that do not contribute to corrosion current.

Etching in sulfuric acid offers a different way to identify which side of the barrier layer is n-type. Mott-Schottky plots of the 400 ◦C-annealed and etched samples are plotted in Figure 5. We note that the *C*SC−<sup>2</sup> values of the etched samples are less than those of unetched samples in Figure 2f, indicating that *C*SC increases in magnitude. In terms of slope decline, the positive slope decreases more than the negative slope does. Hence, the rise in *N*<sup>d</sup> is more pronounced than that in *N*<sup>a</sup> after etching. For example, etching for 15 min, the correlated *N*<sup>a</sup> value rises from 1.6 <sup>×</sup> 1016 to 8.6 <sup>×</sup> 1017 cm<sup>−</sup>3; while the correlated *N*<sup>d</sup> value increases more significantly, from 4.5 <sup>×</sup> 1016 to 1.8 <sup>×</sup> 1019 cm−<sup>3</sup> as shown in Figure 5a. Figure 5b presents the Mott-Schottky plot after 45 min etching, the positive slope is almost zero. The above etching results indicate more donor defects after oxide removal. A plausible explanation is that when the vulnerable oxide being etched, more defects and higher surface heterogeneity are created at the same time.

**Figure 5.** Mott-Schottky plots of the etched AAO layers. The samples of barrier layer have been 400 ◦C annealed for 1 h are then etched in sulfuric acid for (**a**) 15 and (**b**) 45 min. Note the positive slope flattens as etching time increases.

In view of the preferential etching on n-type oxide, the oxide/electrolyte interface ought to be assigned as the donor interface since the preferential etched surface must directly face the sulfuric acid. Furthermore, this assignment is also supported by the correlation results of Figure 4. Only when the oxide/electrolyte (external) interface is n-type, the corrosion protection could be hinged on the donor

density. Otherwise, corrosion of this barrier layer would depend on the acceptor density. Our conclusion on the n-type oxide/electrolyte interface is consistent with the conclusions of Vrublevsky [20–23] and Benfedda [24].

#### *3.2. The PEO Coating with Pre-Grown AAO*

The morphological features of the PEO coating are shown in Figure 6. The sample has been pre-anodized under 200 V (+)/40 V (−), then micro-arc treated at 0.7 A (+)/0.8 A (−). The coating thickness is 760 ± 35 nm, Figure 6a, much thicker than its pre-anodized barrier layer, considering the PEO period is 2 min only. Electrical discharges of PEO leave scars, which may be elusive in the cross-sectional image of the coating, yet distinct in the top-view image as shown in Figure 6b. The coating surface shows many pin holes of submicron size associated with a terrain of frozen lava swellings, much different from the pre-anodized surface. Also encircled in red are a few micron-sized grains of hexagonal facet, which might be the precursor of alpha alumina. The crystalline phase of the PEO coating is gamma aluminum oxide, as indicated in its X-ray diffraction result (Figure S4, Supplementary Materials).

**Figure 6.** Morphology of the PEO coating with pre-anodized oxide. The SEM images of (**a**) cross-sectional view and (**b**) top view of the PEO coating, which is preceded by an AAO layer prepared at 200 V (+)/ 40 V (−). Large circles mark the electrical discharge damages.

Mott-Schottky plots of two PEO samples are shown in Figure 7, indicating only the n-type behavior is detected. One PEO sample has been treated with a pre-anodized layer without annealing, Figure 7a, the other with a pre-anodized layer being 250 ◦C-annealed, Figure 7b. Annealing temperature can <sup>a</sup>ffect the donor density of the PEO coating. Without annealing, the donor density is 2.92 <sup>×</sup> 1017 cm<sup>−</sup>3. The donor density of PEO coating decreases with increasing pre-annealing temperature, with 2.67 <sup>×</sup> <sup>10</sup><sup>17</sup> (150 ◦C), 2.37 <sup>×</sup> <sup>10</sup><sup>17</sup> (250 ◦C), and 1.95 <sup>×</sup> <sup>10</sup><sup>17</sup> cm−<sup>3</sup> (400 ◦C).

Since the extent of *N*<sup>d</sup> decline in PEO coating is narrow, its influences on corrosion protection are moderate. Figure 8 shows how much corrosion protection is affected by the donor density of PEO coating. Figure 8a indicates the corrosion potential *E*corr increases with increasing annealing temperature of the AAO layer, −0.77 ± 0.055 V (without annealing), −0.74 ± 0.03 V (250 ◦C annealed AAO), −0.72 ± 0.02 V (400 ◦C annealed AAO). The corrosion current *J*corr of PEO coating decreases with increasing annealing temperature; (3.3 <sup>±</sup> 0.9) <sup>×</sup> <sup>10</sup>−<sup>8</sup> (AAO without annealing), (1.7 <sup>±</sup> 0.5) <sup>×</sup> <sup>10</sup>−<sup>8</sup> (250 ◦<sup>C</sup> annealed AAO), (1.7 <sup>±</sup> 0.4) <sup>×</sup> <sup>10</sup>−<sup>8</sup> <sup>A</sup>·cm−<sup>2</sup> (400 ◦C annealed AAO). And the polarization resistance *<sup>R</sup>*<sup>p</sup> increases with increasing annealing temperature; (1.1 <sup>±</sup> 0.3) <sup>×</sup> <sup>10</sup><sup>9</sup> (AAO without annealing), (2.3 <sup>±</sup> 0.8) <sup>×</sup> <sup>10</sup><sup>9</sup> (250 ◦C annealed AAO), (2.1 <sup>±</sup> 0.4) <sup>×</sup> <sup>10</sup><sup>9</sup> <sup>Ω</sup> cm<sup>2</sup> (400 ◦C annealed AAO).

**Figure 7.** Typical Mott-Schottky plots of the PEO coatings. The plots of *C*SC−<sup>2</sup> versus electrode potential are drawn for the PEO coatings with an AAO layer (**a**) without annealing, (**b**) 250 ◦C annealing.

**Figure 8.** Influences of the AAO layer on anticorrosion of following PEO coating. Comparison of potentiodynamic polarization curves at scan rate 5 mV·s−<sup>1</sup> shows (**a**) the influences of annealing temperature on anticorrosion of its PEO coating. Correlation between donor density and (**b**) corrosion current, (**c**) polarization resistance, (**d**) corrosion potential of the PEO coating.

We ought to mention that linear correlation of the PEO coating contains substantial uncertainty inferior to that of the AAO layer. Correlation between corrosion protection and donor density of PEO coatings, the percent deviation is 27% between *R*<sup>p</sup> and *N*d, 42% between *J*corr and *N*d, 43% between *E*corr and *N*d. We attribute the weak correlation to the damaging effects of electrical discharge. Extra thickness of the PEO coating provides additional protection, but its porosity also brings more uncertainty.

It is of interest to look into why the donor density of PEO coating is affected by annealing temperature of its pre-anodized AAO. One plausible cause is defects of the pre-anodized AAO layer alter the voltage profile of PEO treatment. Figure 9a contrasts the voltage-time profiles of 400 and 250 ◦C annealed samples, with that of the sample without annealing. The voltage profiles of 400 and 250 ◦C annealing are consistently higher than that of the one without annealing. Thus, a higher positive voltage is required for the AAO layer of less defects to reach the preset current density. If the pre-anodized layer is partially etched away, the remaining oxide affects the voltage-time profile of PEO differently. Figure 9b compares the voltage profile of the pre-anodized sample without etching with those of the etched samples. As a whole, the positive voltage of the samples etched for 15, 30, 45, and 60 min begins with a lower value than the unetched sample. However, the voltage of etched samples surpasses that of the unetched sample after 5 or 9 s. We generally observe the first spark at 10 s (Video S1, Supplementary Materials), therefore, the *E*10S value is physically meaningful. The *E*10S values in Figure 9b are listed as: 507 (15 min), 475 (30 min), 461 (45 min), 437 (60 min), and 421 V (unetched). An etched sample seems to develop a coating with higher dielectric strength than the unetched sample after several PEO seconds, so that its positive voltage is higher. Therefore, defects of the pre-anodized oxide exert influences on its following PEO coating. Compared with the unetched sample, the etched AAO layer grows faster, so does the positive voltage when PEO time is between 2 and 18 s. The acceptor defects and diminished outer layer seem to be responsible for the fast growth in this brief period.

**Figure 9.** Positive voltage evolution in the initial stage of PEO treatment. Comparison of the voltage-time profiles of PEO that began with (**a**) 400 and 250 ◦C annealed barrier layer, and the barrier layer without annealing. Comparison of the voltage-time profiles of PEO that started with (**b**) the barrier layers of 15, 30, 45, and 60 min etching and the barrier layer without etching and annealing.

#### **4. Conclusions**

Point defects are known to affect corrosion protection properties in various manners, but researchers seldom connect the two directly. In this research, the point defect densities of acceptor and donor are varied through annealing and etching of the AAO layer, and the anticorrosion parameters of the AAO layer are measured with potentiodynamic polarization scans. We find that a linear correlation between corrosion current and donor density, also between polarization resistance and donor density. The correlation between corrosion potential and donor density can be fitted with considerable deviation. For PEO coatings, only the n-type behavior is observed. Similar correlations can be established between the corrosion properties and donor density for PEO coating. However, deviations from linear correlation are substantial, since the electrical discharges during PEO introduce porosity and uncertainty in anticorrosion properties.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2079-6412/10/1/20/s1, Figure S1: Schematic diagram of setup for anodizing the 6061 sample, following by PEO treatment, Figure S2: Experimental procedure of electrochemical oxidation, characterization, and corrosion analysis, Figure S3: Linear correlation between corrosion parameters and acceptor density. In contrast to Figure 4b–d, the correlations are poor with respect to acceptor density. Correlation is attempted between (a) *J*corr and *N*a, (b) *R*p and *N*a, (c) *E*corr and *N*a, Figure S4: XRD patterns of AAO and PEO coatings. (a) The result of 400 ◦C annealed barrier layer displays the diffraction lines of metallic aluminum substrate only. (b) The diffraction result of PEO coating that began with an AAO precursor shows the gamma aluminum oxide features in addition to the diffraction lines of aluminum metal, Video S1: The first appearance of microdischarges in PEO.

**Author Contributions:** Conceptualization, formal analysis and writing, D.-S.T.; investigation and validation, H.-R.L.; resources, C.-C.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** The authors would like to thank Ministry of Science and Technology of Taiwan for financial support of this work through the project MOST-106-2221-E-011-119-MY3.

**Acknowledgments:** We would like to thank Ching-Hwa Ho of NTUST for allowing the oscilloscope and other electronic measurement instruments in this work.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
