Cathodic voltage-dependent composition, microstructure and corrosion resistance of plasma electrolytic oxidation coatings formed on Zr-4 alloy

Abstract

Plasma electrolytic oxidation (PEO) coatings are fabricated on Zr-4 alloy by a pulsed bipolar power supply. When the anodic voltage remains constant, the variation of cathodic voltage exhibits a significant impact on the microstructure and corrosion resistance of the oxide coatings. Here we systematically investigate the influence of cathodic voltage on the phase composition, morphology, thickness, and elemental composition of the PEO coatings. Corrosion behaviors are evaluated by electrochemical impedance spectroscopy (EIS). The coating thickness and the electrolyte borne elements incorporated in the coatings both increase with the increase of cathodic voltage. It is interesting to note that the relative content of tetragonal ZrO₂ and monoclinic ZrO₂ also shows a strong cathodic voltage dependence. The coating formed at 50 V cathodic voltage shows the most compact microstructure with the largest amount of tetragonal ZrO₂ and correspondingly exhibits the optimized corrosion resistance. The presence of t-ZrO₂ is found to be beneficial for dense oxide coatings and better corrosion resistance. Therefore, cathodic voltage is an important parameter during PEO process to adjust the microstructure and the corrosion resistance performance of PEO coatings.

---

1. Introduction

Zr alloys are currently utilized in the majority of commercial nuclear plants as cladding material due to their excellent corrosion resistance, high temperature mechanical properties and thermal neutron absorption ability.¹ However, in case of the typical loss of coolant accidents (LOCAs), the violent reaction between steam and Zr alloy results in the generation of appreciable amounts of heat and hydrogen gas, which may lead to the hydrogen explosion.² The recent accident at Fukushima Dai-ichi once again revealed the importance and impendency of accident-tolerant fuel (ATF) system, which is designed to prevent radioactive release in the case of LOCAs and reactivity initiated accidents (RIAs). From this perspective, a wide variety of different approaches have been envisioned. They can be grouped into three categories: improved high temperature oxidation resistance and/or strength of Zr alloy cladding (e.g., highly adherent oxidation-resistant coatings), non-zirconium cladding with high strength and oxidation resistance (e.g., silicon carbide fiber-reinforced SiC ceramic composites), and alternative fuel forms with improved performance.^3–6 Compared with the latter two aspects, surface treatment of Zr alloy is more practical and cost-effective. More importantly, this approach can retain the fuel systems to a large extent based on Zr alloy cladding which has already been highly optimized over 60 years of development.

Among the variety of surface treatment technologies, plasma electrolytic oxidation (PEO) can greatly satisfies the main desired attributes of ATFs which require matched thermal expansion coefficients, high adhesion and enough integrity. PEO or microarc oxidation (MAO) is an in situ coating growth technology processed in specific electrolyte. When compared with other surface treatments such as plasma spray, chemical vapor deposition (CVD) and physical vapor deposition (PVD), it can be processed at room temperature and meanwhile achieves the higher bonding strength.^7–9 Moreover, complex shaped alloys can also be prepared with uniform coatings. Therefore, in terms of the cost-effectiveness, simplicity in processing and the coating properties, PEO is a promising surface treatment technology.

The PEO coatings on alloys such as Al, Mg and Ti have been extensively studied, and remarkable progress has been made in areas of anticorrosion, biomedical and wear protection accordingly.^10–13 However, to our best knowledge, the research on the coatings formed on Zr alloys that used as fuel structural material in the reactor core is still rare. PEO is a multifactor-controlled process. The qualities of PEO coatings are determined by many factors, such as electrolyte composition, applied power modes, electrical parameters, etc.^14–19 In respect to the PEO coatings on Zr alloys, Ying et al. studied the effects of current density on the coating properties.²⁰ They used autoclave experiments to investigate the corrosion properties of the coatings at 300 °C and 10 MPa in LiOH solution. It showed that PEO coatings prepared at low current densities had lower weight gains. Besides, many studies have focused on the electrolyte composition. For example, Cheng et al. have studied the PEO coatings formed in silicate, pyrophosphate and silicate–pyrophosphate electrolytes, respectively.¹¹ They observed that the coating was comprised with compact outer layer and porous inner layer, the former mainly contained tetragonal ZrO₂ (t-ZrO₂) while the latter was dominated by monoclinic ZrO₂ (m-ZrO₂). Yan et al. fabricated Al₂O₃/ZrO₂ coatings in NaAlO₂ electrolyte with different NaAlO₂ content.²¹ The results showed the coating with more t-ZrO₂ and α-Al₂O₃ performed better in microhardness and bond strength. Ya et al. have investigated the effect of Ce-contained additive on the microstructure and properties of PEO coatings.²² It suggested that Ce can be incorporated into the coating to improve its t-ZrO₂ content.

The t-ZrO₂ is widely considered to be superior in many physical and chemical properties, thus the stabilization of t-ZrO₂ would be beneficial for the enhancement of coating properties. Some studies on the Al, Mg and Ti alloys have found that the change of anodic or cathodic voltage can affect the phase composition of PEO coatings. The study of Xu et al. showed that the amount of γ-Al₂O₃ in the PEO coatings formed on NiTi alloy were affected by the anodic voltage.²³ Li et al. investigated the effects of cathodic voltage on the elemental composition and surface morphology of PEO coatings formed on Al alloy.²⁴ When it comes to the Zr alloy, Zou et al. found that the anodic voltage can influence the microhardness and corrosion resistance of PEO coatings on the Zr–1Nb alloy.²⁵ But the coatings in this work were mainly consist of m-ZrO₂ and besides, the variation of anodic voltage showed no significant influence on the phase composition.

In the present work, we innovatively regulated the relative content of t-ZrO₂ and m-ZrO₂ by adjusting the cathodic voltage of a pulsed bipolar power supply. The coating thickness and the electrolyte borne elements incorporated in the coatings also shown cathodic voltage dependence. Correspondingly, the morphology and corrosion resistance of the coatings both varied with the variation of cathodic voltage.

2. Experiment procedure

2.1. Preparation of the coatings

Commercial purity Zr-4 alloy was obtained as plates of 1 mm thickness. The chemical composition of the alloy was as follows: (wt%) 1.50 Sn, 0.20 Fe, 0.10 Cr, Ni < 0.007, and Zr balance. Specimens with coupon dimensions of 10 mm × 10 mm × 1 mm were cut from the plates by using the wire electrical discharge machining (WEDM). Prior to PEO treatment, the surfaces of the specimens were ground to 2000 grits with abrasive paper, and cleaned with ethanol and deionized water in an ultrasonic bath, then dried at room temperature. A 20 kW homemade pulsed bipolar power supply was employed to carry out the PEO process. The specimens were used as anode and placed in a water-cooled electrolyser which was made of stainless steel and served as cathode. The Zr-4 specimens were then treated in the electrolyte solution containing 10 g L⁻¹ Na₅P₃O₁₀, 2 g L⁻¹ NaOH, 2.17 g L⁻¹ Ce(NO₃)₃·6H₂O and 1.41 g L⁻¹ C₄O₆H₄KNa·4H₂O. C₄O₆H₄KNa·4H₂O was used as complexing agent considering that Ce(NO₃)₃ would hydrolyze in the alkaline solution. For each PEO coated alloy, the PEO process was carried out for 5 min under constant voltage mode. The constant anodic voltage of 375 V and different cathodic voltages of 0 V, 25 V, 50 V, 75 V, 100 V were applied, noted as CV0, CV25, CV50, CV75 and CV100, respectively. The frequency and duty cycle were selected to be 1000 Hz and 35%, respectively. During the PEO processes, reaction temperature was controlled to be below 30 °C by adjusting the cooling water flow and constant mechanical stirring. After the PEO process, specimens were rinsed with distilled water and air dried.

2.2. Phase analysis, morphology and elemental composition

The phases present in the PEO coatings were investigated by grazing incidence X-ray diffraction (GIXRD), using a PANalytical X'Pert-Pro-MPD (PW3050/60) instrument with a grazing angle of 5°, a step size of 0.03°, a scan range from 20–70° (in 2θ) and scan step time of 1 s. To determine the weight fractions of the t-ZrO₂ and m-ZrO₂, the PEO coatings were characterized by considering the intensities of the characteristic peaks of each oxide phase in the XRD patterns. The relative amounts of t-ZrO₂ and m-ZrO₂ in the PEO coatings are given in Table 1. This data is derived from the following formulae:where I_t(101) and I_m(1̄11) are the characteristic peak intensities of t-ZrO₂ and m-ZrO₂, respectively, K^t_m is denoted as a function of K_t and K_m, i.e.K_t and K_m are the RIR values of t-ZrO₂ and m-ZrO₂, respectively, which are referenced from the JCPDS cards.

Table 1 Phase composition, crystallite size and intensity value of highest intense (1̄11)_m and (101)_t planes obtained from XRD patterns of CV0–CV100 PEO coated alloys

Samples	Weight fraction (wt%)		Crystallite size (nm)		Intensity (a.u.) of peaks
	t-ZrO2	m-ZrO2	(101)t	(1̄11)m	(101)t	(1̄11)m
CV0	19.6	80.4	20.7	14.1	51.0	100.0
CV25	37.2	62.8	20.7	17.6	100.0	80.7
CV50	44.4	55.6	21.1	18.5	100.0	59.9
CV75	29.4	70.5	21.7	22.0	87.0	100.0
CV100	27.9	72.1	19.2	25.3	80.9	100.0

---

The crystallite size of t-ZrO₂ and m-ZrO₂ in all coatings were evaluated from the FWHM values of I_t(101) and I_m(1̄11) diffraction peaks (Fig. 1) by using Scherrer's equation:²⁶

where d is the average crystallite size in nm, λ is the radiation wavelength (0.154 nm), β denotes the corrected half-width at half intensity (FWHM), and θ is the diffraction peak angle. The obtained values are reported in Table 1.

Fig. 1 XRD patterns of the PEO coated alloys. The coatings were formed at different cathodic voltages. (a) CV0, (b) CV25, (c) CV50, (d) CV75, (e) CV100.

The surface and cross-sectional morphology, coating thickness, elemental composition and elemental mapping of the PEO coatings were examined by field emission gun scanning electron microscopy (FEG-SEM), using CamScan Apollo 300 instrument equipped with energy-dispersive X-ray spectroscopy (EDS) analysis facility. Due to the low electrical conductivity, all the PEO coated alloys were treated by carbon sputtering before SEM tests.

2.3. Electrochemical corrosion test

Electrochemical measurements were performed at room temperature on the bare Zr-4 alloy and coated alloys. A Zahner Ennium electrochemical workstation (Germany) was used for the measurements, employing a 3-electrode configuration, with a platinum plate as the counter electrode, a saturated calomel electrode as the reference electrode. The experimental solution was 1200 mg L⁻¹ H₃BO₃ + 2 mg L⁻¹ LiOH, which simulated the environment of pressurized water reactor (PWR) primary water.²⁷ Electrochemical impedance spectroscopy (EIS) measurements were performed in the frequency range between 100 kHz and 10 mHz at an AC amplitude of 10 mV (rms). Anodic potentiodynamic polarization curves were recorded at a scan rate of 1 mV s⁻¹ from the open circuit potential (OCP) to a final potential of 4 V.

3. Results

3.1. Phase composition

The XRD patterns of the CV0-CV100 PEO coated alloys formed at different cathodic voltages are shown in Fig. 1. The presence of both m-ZrO₂ (JCPDS card no. 83-0937) and t-ZrO₂ (JCPDS card no. 79-1764) can be confirmed in all coated alloys. A variation of the relative intensities between main t-ZrO₂ (101) peak (indexed by symbol “◆”) and main m-ZrO₂ (1̄11) peak (indexed by symbol “●”) can be observed from the XRD patterns in Fig. 1, indicating that the content of t-ZrO₂ significantly increases at first and then decreases with the increase of cathodic voltage. Table 1 gives the weight percentages and crystallite sizes of t-ZrO₂ and m-ZrO₂ calculated from the XRD results in Fig. 1. The weight percentage of t-ZrO₂ is found to be 19.6, 37.2, 44.4, 29.4 and 27.9 for CV0, CV25, CV50, CV75 and CV100, respectively. Both the Fig. 1 and Table 1 show that the most amount of t-ZrO₂ is stabilized in CV50 coating.

The crystallite sizes of t-ZrO₂ and m-ZrO₂ in coated alloys also influenced by the variation of cathodic voltage. It can be seen in Table 1, the crystallite size of m-ZrO₂ phase increases from 14.1 nm to 25.3 nm with the increasing cathodic voltage, as for t-ZrO₂ phase, its crystallite size distribute in a relatively narrow range, CV75 has the largest value, as 21.7 nm, following by CV50 as 21.1 nm, CV0 and CV25 as 20.7 nm, and CV100 exhibits the minimum as 19.2 nm.

3.2. Surface morphology and elemental composition

The surface morphology of CV0–CV100 coated alloys are shown in Fig. 2. It exhibits that the typical porous structures with micropores and microcracks are adjusted by the variation of cathodic voltage. With increasing cathodic voltage in the range of 0–75 V, the amount of microcracks reduces and the micropores become much less with their diameter being marginally enlarged. More interestingly, from the pore in CV75, we can find that in CV75 coating the long tortuous microcracks are replaced by the tiny dendritic microcracks which radially distributed around the central pore. When cathodic voltage is further increased to 100 V, these microcracks violently grow and become much greater in both size and quantity.

Fig. 2 SEM surface micrographs of the CV0–CV100 PEO coated alloys and pore in CV75 the higher magnification image of the pore with a red circle in CV75.

Fig. 3 shows EDS spectra of the surface of PEO coated alloys. The existence of Zr, O and electrolyte borne elements is revealed. The elemental composition obtained from the spectra are reported in Table 2. It can be observed from Fig. 3 and Table 2 that the coatings are mainly composed of ZrO₂, and electrolyte borne elements Ce and P. Here, C comes from the sputtered carbon coating. With increasing the cathodic voltage, the content of Ce increases, while the content of Zr and O relatively decreases. Fig. 4 shows the EDS element mapping images of CV50 coating. It can be seen that Zr, O and Ce are uniformly distributed over the surface of the coating.

Fig. 3 Surface SEM-EDS spectra of CV0–CV100 PEO coated alloys.

Table 2 Element concentration measured by SEM-EDS at the surface of PEO coated alloys shown in Fig. 2

Samples	Element (at%)
	O	Zr	Ce	C	P
CV0	53.7	28.1	4.4	10.7	3.1
CV25	52.3	27.8	5.9	9.9	4.1
CV50	49.4	26.9	8.8	11.3	3.6
CV75	48.4	23.1	12.1	12.1	4.3
CV100	46.0	19.1	13.6	15.7	5.6

---

Fig. 4 The EDS elemental mapping images of CV50 PEO coated alloy.

Cross-sectional micrographs of the PEO coated alloys are presented in the Fig. 5. All the coatings in the present study other than CV100 are dense and uniform. And their thickness values is found to be 5.9 ± 1.1 μm, 6.7 ± 0.8 μm, 7.6 ± 1.4 μm, 11.1 ± 2.1 μm and 30.3 ± 2.2 μm for CV0–CV100, respectively. For the CV0–CV50 coatings, their thickness shows an almost linear increase with a rate about 0.9 μm per 25 V, then the increase appears to be steep in CV75 and CV100 coatings. In accordance with the observation of surface morphology in Fig. 2, lots of microcracks exist in CV0 and CV100 coatings. The amount of microcracks is significantly reduced in CV25, CV50 and CV75 coatings, and the CV50 exhibits the most compact coating structure.

Fig. 5 Cross-sectional SEM micrographs of PEO treated CV0–CV100 coatings.

3.3. Electrochemical corrosion characteristics

In order to further understand the structure and corrosion properties of the PEO coatings, electrochemical tests were performed. Fig. 6 shows Nyquist and Bode plots of the uncoated and coated Zr-4 alloys. As displayed in Fig. 6(a), Nyquist curves of coated alloys have much larger diameter than that of uncoated Zr-4 alloy, so that Nyquist curves of uncoated Zr-4 alloy is hard to be noticed. Details of the Nyquist curve of uncoated Zr-4 alloy can be observed in Fig. 6(b). It shows one obvious capacitive loop on the large scale, as well as a small quarter circle in the range of 1500 Ω cm² to 5400 Ω cm². The phase angle of uncoated Zr-4 alloy in Fig. 6(d) shows one nearly symmetrical peak at low frequency and an increase with frequency at high frequency. Taking both Fig. 6(b) and (d) into consideration, uncoated Zr-alloy has two time constants, one at the low frequency corresponds to the natural oxide film, and the other at high frequency corresponds to the double electric layer. In contrast to uncoated Zr-4 alloy, the phase angles of all coated alloys in Fig. 6(d) exhibit maximums at both high and low frequency ranges. Except the time constant at the end of the curve which corresponds to the double electric layer, the existence of the other two time constants indicates that the coatings have a two-layer structure. This is in good accordance with the previous study.^12,28 For the two time constants, one appearing at higher frequency corresponds to the porous outer layer, and the other at lower frequency corresponds to the compact inner layer. Because of the further corrosion resistance provided by the PEO coatings, the impedance values of coated alloys in Fig. 6(c) are about two orders of magnitude higher than that of uncoated Zr-4 alloy at low frequency.

Fig. 6 Nyquist plots of (a) Zr-4 alloy and CV0–CV100 PEO coated alloys, (b) only Zr-4 alloy; (c) and (d) the Bode plots of Zr-4 alloy and CV0–CV100 PEO coated alloys.

Fig. 7 shows the results of anodic potentiodynamic polarization for the uncoated and coated Zr-4 alloy. The coated specimens reveal anodic current densities that are significantly lower than that of the uncoated Zr-4 alloy, indicating an improved protection of the alloy. Among all the specimens, PEO coating formed at cathodic voltage of 50 V has the lowest corrosion current density in the potential range of 1.5–4.0 V, indicating it provides the best protection.

Fig. 7 Anodic potentiodynamic polarization curves for the uncoated and coated Zr-4 alloy. The experimental solution was 1200 mg L⁻¹ H₃BO₃ + 2 mg L⁻¹ LiOH.

4. Discussion

The differences in electronic properties of the coatings are thought to be the major reason that gives rise to the variation of coating properties.

During the PEO process, t-ZrO₂ can be generated thanks to the high temperature (10³ °C to 10⁶ °C) at the discharge region.²⁹ However, due to the rapid cooling of electrolyte, t-ZrO₂ tends to transform to m-ZrO₂. As discussed in the previous study, the stabilization of t-ZrO₂ is influenced by several factors such as dopant, crystallite size and concentration of oxygen vacancies.³⁰ In the present study, as shown in Table 1, the crystallite size of t-ZrO₂ in all the coatings is smaller than 30 nm which is thought to be a critical size below which the t-ZrO₂ tends to be stabilized.³¹ As a result, t-ZrO₂ can be observed in all the coatings. The Ce³⁺ amount and concentration of oxygen vacancies contribute to the variation of the amount of stabilized t-ZrO₂. As Ce³⁺ is a low valency dopant cation, when it is introduced into the ZrO₂ lattice, oxygen vacancies are created for the charge balance. As discussed by S. Shukla,³² owing to the large size of Ce³⁺ relative to Zr⁴⁺, the generated oxygen vacancies tend to be associated with Zr⁴⁺, which causes the effective coordination number to be below 7. To maintain its effective coordination number close to 7, as dictated by the covalent nature of Zr–O bond, the ZrO₂ lattice assumed a crystal structure, such as tetragonal structure, which offers 8-fold (higher than 7) coordination number. With the increase of cathodic voltage, more Ce³⁺ participates in the process of coating growth (Table 2), thus resulting in the increase of t-ZrO₂ amount to maximize in CV50 coating.

In regard to the decrease of ZrO₂ amount in CV75 and CV100 coatings, we should focus on the reaction at both the anodic and cathodic cycles. As shown in Fig. 8, during the anodic cycle, anions like OH⁻ and the complex anion containing Ce³⁺ are poured into the discharge channels, which contributes to the coating growth and the generation of O₂ according to the reaction.³³

Zr + 2O²⁻ → ZrO₂ + 4e⁻ (4)

4OH⁻ → O₂↑ + 2H₂O + 4e⁻ (5)

Fig. 8 Anodic and cathodic processes in plasma electrolytic oxidation treatment.

The cathodic reaction is presumed to be primarily the reduction of H⁺ ions.³⁴

2H⁺ + 2e⁻ → H₂↑ (6)

H₂ generated in the cathodic cycle can promotes the expelling of O₂ and electrical charges accumulated in the anodic cycle. Besides, it provides a plasma-supporting gas medium in the vicinity of the oxide-electrolyte interface from the very beginning of the oxidation process.^10,34,35 Therefore, the presence of cathodic process is benefit to the coating growth and generation of O₂. And also, it give rise to the increase in the amount of Ce incorporated in the coating with increasing cathodic voltage.

However, when cathodic voltage increases to above 75 V, enormous OH⁻ are poured into the coating and excessive O₂ are generated in the discharge channels and the substrate–electrolyte interface. As a result, the oxygen vacancies would be partially annihilated. When the oxygen vacancy concentration is reduced to a critical magnitude below which the maintenance of tetragonal phase is impossible, t-ZrO₂ to m-ZrO₂ transformation occurs.³⁶

Both the phase composition and discharge behavior contribute to the variation of morphology in the CV0–CV100 coatings. Models had been built to describe the discharge behaviour. Hussein et al.³⁷ and Cheng et al.¹⁹ proposed a model that assumed three different types of discharges: metel–oxide interface discharges that caused by the dielectric breakdown in a strong electric field; oxide-electrolyte interface discharges at the gas attached to the coating surface; and discharges in the micropores below the surface. In this model, only metal–oxide interface discharges are considered to draw the discharge channel from the metal–oxide interface to the coating surface. However, in the recent work of Liu et al.,³⁸ it suggested that all the spark discharges are caused by the dielectric breakdown of PEO coating and every spark should correspond to a discharge channel occurring through the oxide layer. Based on this viewpoint, they proposed a model with two discharge types: type A discharges with small size correspond to the fine sparks, and type B discharges with large size correspond to the strong sparks. Considering the shorter PEO treatment time and thinner coating thickness of CV0–CV100 samples, the model of Liu et al. is more suitable to describe the discharge behaviour in the present study. As shown in Fig. 9, type A discharges are more likely to occur along microcracks and/or across micropores in the interior of coating. Type B discharges are stronger in energy, they tend to take place at the weak points with defects on the coating surface and cause intense dielectric breakdown. In Fig. 2, type A discharges can be distinguished by their wide distribution and obviously smaller pore diameter, while type B discharges are characterized by their crater-like features,³⁹ as indicated by arrows.

Fig. 9 Schematic illustration of the model describing the two types of plasma discharges.

Compared with the unipolar PEO treatment, more anodic current is consumed to generate gases in the bipolar PEO treatment. Therefore, with cathodic process, both the number and strength of strong type B discharges are reduced. By allowing enough time for the oxide to cool down before other pulses are initiated, cathodic process also provides longer sintering time, thus, a thick and hard coating with minimum porosity is produced.⁴⁰

In the present study, the increase of cathodic voltage results in the increase of coating thickness (Fig. 5) and, thus, the decrease in probability of intense dielectric breakdown. This will reduces the number of type B discharges and improves the energy of type A discharges. So that there will be a balance of discharge effect. Since large quantities of unstable low-energy discharges are replaced by small quantities of balanced high-energy discharges, the density of micropores and microcracks decreases (Fig. 2).^41,42 As for CV50 coating, it has the highest amount of stabilized t-ZrO₂, indicating that there is less volume expansion caused by the phase transformation. As a result, CV50 coating shows the most compact microstructure (Fig. 5). Whereas, with respect to CV100 sample, the sharp increase of coating thickness (Fig. 5) results in the intense increasing of discharge energy. With such ultrahigh-energy discharges as well as numbers of high-temperature gas bubbles, the thermal stresses in the coating dramatically increase. Therefore, the stable coating growth becomes impossible and more defects, like microcracks, are produced.^29,31

It is shown in Fig. 6 and 7 that all PEO samples generally show the improvement in corrosion resistance. From the details in Fig. 6(a), it can be noticed that the Nyquist plots of CV0–CV100 samples show some slight differences. The curve of CV50, which looks almost a straight line, is found to be the longest one, followed by CV75, CV25, CV0, and CV100. In accordance with the observation in Fig. 7, CV50 exhibits the lowest corrosion current density, followed by CV75, CV25, CV0, and CV100. These results indicate that, compared to the coating thickness, coating compactness plays a more crucial role in the improvement of corrosion resistance.

Based on the EIS plots, the equivalent circuits shown in Fig. 10 are used to model the electrical response of the coatings. Q_n, Q_e and Q_i model the electrical properties of the natural oxide film, outer layer, and inner layer, respectively. Q is a constant phase angle element (CPE), which was used to replace the pure capacitance. The CPE is defined as:

Z = 1/[Y₀(jω)ⁿ] (7)

where Y₀ is modulus of admittance, j is the imaginary unit , ω is the perturbing frequency and n is the exponent to characterize the deviation degree of real capacitance from its ideal value (0 < n ≤ 1). The coefficients Y₀ and n are expressed as Q − Y₀ and Q − n in Tables 3 and 4.

Fig. 10 Equivalent circuits of bare Zr-4 alloy (A) and coated samples CV0–CV100 (b).

Table 3 EIS fitting results of Zr-4 alloy substrate

Sample	Cdl (F cm^−2)	Rt (Ω cm^2)	Qn − Y0 (Ω^−1 cm^−2 s^−n)	Qn − n	Rn (Ω cm^2)
Zr-4	2.41 × 10^−10	3326	1.53 × 10^−5	0.933	3.75 × 10^5

---

Table 4 EIS fitting results of CV0–CV100 PEO coated alloys

Samples	Cdl (F cm^−2)	Rt (Ω cm^2)	Qe − Y0 (Ω^−1 cm^−2 s^−n)	Qe − n	Re (Ω cm^2)	Qi − Y0 (Ω^−1 cm^−2 s^−n)	Qi − n	Ri (Ω cm^2)
CV0	7.31 × 10^−8	635	8.14 × 10^−10	0.952	2028	2.28 × 10^−7	0.748	2.46 × 10^8
CV25	5.82 × 10^−10	1009	5.26 × 10^−8	0.983	1606	2.36 × 10^−7	0.737	3.16 × 10^8
CV50	4.26 × 10^−10	1072	1.08 × 10^−7	0.926	6352	2.52 × 10^−7	0.735	5.07 × 10^8
CV75	5.96 × 10^−10	1200	7.15 × 10^−8	0.960	615	2.81 × 10^−7	0.781	3.28 × 10^8
CV100	4.25 × 10^−8	460	1.63 × 10^−9	0.931	1131	3.81 × 10^−7	0.773	1.09 × 10^8

---

The corresponding R_n, R_e and R_i are the resistance of natural oxide film, outer and inner layer, respectively. R_s is the solution resistance, R_t is the charge transfer resistance paralleled with double electric layer capacitance (C_dl). The EIS plots are fitted with these equivalent circuits, and the fitting results are shown in Fig. 6 with solid lines. Tables 3 and 4 shows the fitting values. It can be observed in Table 3 that the R_n (3.75 × 10⁵ Ω cm²) is two orders of magnitude higher than R_t (3326 Ω cm²), indicating that the Zr-4 alloy has certain corrosion resistance resulting from the presence of passive film on the alloy surface. Furthermore, as exhibited in Table 4, the R_i are five or more orders of magnitude higher than the corresponding R_e and R_t. This implies that the compact inner layer makes a major contribution to the corrosion resistance. And it can be noticed, the highest value of R_i with 5.07 × 10⁸ Ω cm² is belong to CV50 coating. Together with the results in Fig. 6 and 7, we can deduce that the CV50 sample has the most excellent corrosion resistance.

5. Conclusion

A novel approach to regulate the phase composition and morphology of PEO coatings was achieved by adjusting the cathodic voltage. It was demonstrated that the phase composition, morphological characteristics, elemental composition and corrosion resistance of the PEO coatings show strong negative voltage dependencies. With increasing the cathodic voltage, both the coating thickness and the content of Ce incorporated into the coatings increased, the content of t-ZrO₂ first increased and then decreased with a peak at 50 V. Owing to the reduced volume expansion caused by the phase transformation of ZrO₂, CV50 exhibited less microcracks inside the coating and, thus, shown the most compact microstructure. Moreover, all the coatings exhibited the remarkable improvement in corrosion resistance compared with Zr-4 alloy, especially, the sample CV50 performed best in the electrochemical tests due to its most compact microstructure.

Acknowledgements

Haibin Zhang is grateful to the foundation by the Recruitment Program of Global Youth Experts and the Youth Hundred Talents Project of Sichuan Province. This work is supported by the National Natural Science Foundation of China (Grant No. 91326102), the Science and Technology Development Foundation of China Academy of Engineering Physics (Grant No. 2013A0301012), and Science and Technology Innovation Research Foundation of Institute of Nuclear Physics and Chemistry.

References

1. Y. Cheng, E. Matykina, R. Arrabal, P. Skeldon and G. E. Thompson, Surf. Coat. Technol., 2012, 206, 3230–3239.
2. H. Matsumiya, K. Yoshioka, T. Kikuchi, T. Sugita, S. Higuchi and N. Yoshida, Prog. Nucl. Energy, 2015, 82, 16–21.
3. S. Yeo, E. McKenna, R. Baney, G. Subhash and J. Tulenko, J. Nucl. Mater., 2013, 433, 66–73.
4. S. J. Zinkle, K. A. Terrani, J. C. Gehin, L. J. Ott and L. L. Snead, J. Nucl. Mater., 2014, 448, 374–379.
5. K. A. Terrani, S. J. Zinkle and L. L. Snead, J. Nucl. Mater., 2014, 448, 420–435.
6. K. Yueh and K. A. Terrani, J. Nucl. Mater., 2014, 448, 380–388.
7. M. Sandhyarani, N. Rameshbabu, K. Venkateswarlu, D. Sreekanth and C. Subrahmanyam, J. Alloys Compd., 2013, 553, 324–332.
8. M. Uchida, H. M. Kim, T. KoKubo, K. Tanaka and T. Nakamura, J. Ceram. Soc. Jpn., 2002, 110, 710–715.
9. G. Wang, X. Liu and C. Ding, Surf. Coat. Technol., 2008, 202, 5824–5831.
10. R. Arrabal, E. Matykina, T. Hashimoto, P. Skeldon and G. E. Thompson, Surf. Coat. Technol., 2009, 203, 2207–2220.
11. Y. Cheng, E. Matykina, P. Skeldon and G. Thompson, Electrochim. Acta, 2011, 56, 8467–8476.
12. X. L. Zhang, Y. F. Zhang, L. M. Chang, Z. H. Jiang, Z. P. Yao and X. W. Liu, Mater. Chem. Phys., 2012, 132, 909–915.
13. M. Sieber, T. Mehner, D. Dietrich, G. Alisch, D. Nickel, D. Meyer, I. Scharf and T. Lampke, Surf. Coat. Technol., 2014, 240, 96–102.
14. M. Shokouhfar, C. Dehghanian and A. Baradaran, Appl. Surf. Sci., 2011, 257, 2617–2624.
15. M. Shokouhfar, C. Dehghanian, M. Montazeri and A. Baradaran, Appl. Surf. Sci., 2012, 258, 2416–2423.
16. F. Jin, P. K. Chu, G. Xu, J. Zhao, D. Tang and H. Tong, Mater. Sci. Eng., A, 2006, 435–436, 123–126.
17. R. O. Hussein, D. O. Northwood, J. F. Su and X. Nie, Surf. Coat. Technol., 2013, 215, 421–430.
18. M. R. Bayati, A. Z. Moshfegh and F. Golestani-Fard, Electrochim. Acta, 2010, 55, 2760–2766.
19. Y. Cheng, Z. Peng, X. Wu, J. Cao, P. Skeldon and G. E. Thompson, Electrochim. Acta, 2015, 165, 301–313.
20. C. Ying, X. Nie and D. O. Northwood, Surf. Coat. Technol., 2010, 205, 1774–1782.
21. Y. Yan, Y. Han, D. Li, J. Huang and Q. Lian, Appl. Surf. Sci., 2010, 256, 6359–6366.
22. Y. Ya, F. Guo, P. Li, L. Liu and L. Hao, Rare Met. Mater. Eng., 2010, 39, 1443–1446.
23. J. Xu, F. Liu, J. Luo and L. Zhao, J. Mater. Sci. Technol., 2013, 29, 22–28.
24. Q. Li, J. Liang, Z. Peng and Q. Wang, Appl. Surf. Sci., 2014, 297, 176–181.
25. Z. F. Zou, W. B. Xue, X. N. Jia, J. C. Du, R. S. Wang and L. K. Weng, Surf. Coat. Technol., 2013, 222, 62–67.
26. B. D. Cullity, Elements of X-ray Diffraction, Addison-Wesley, MA, 1978.
27. S. Cissé, L. Laffont, B. Tanguy, M.-C. Lafont and E. Andrieu, Corros. Sci., 2012, 56, 209–216.
28. W. Xue, Q. Zhu, Q. Jin and M. Hua, Mater. Chem. Phys., 2010, 120, 656–660.
29. M. Sandhyarani, T. Prasadrao and N. Rameshbabu, Appl. Surf. Sci., 2014, 317, 198–209.
30. P. Li and I. W. Chen, J. Am. Ceram. Soc., 1994, 77, 118–128.
31. R. C. Garvie, J. Phys. Chem., 1965, 69, 1238–1243.
32. S. Shukla, S. Seal, R. Vij and S. Bandyopadhyay, Nano Lett., 2003, 3, 397–401.
33. L. O. Snizhko, A. L. Yerokhin, A. Pilkington, N. L. Gurevina, D. O. Misnyankin, A. Leyland and A. Matthews, Electrochim. Acta, 2004, 49, 2085–2095.
34. E. Matykina, R. Arrabal, P. Skeldon, G. E. Thompson, P. Wang and P. Wood, Surf. Coat. Technol., 2010, 204, 2142–2151.
35. A. L. Yerokhin, L. O. Snizhko, N. L. Gurevina, A. Leyland, A. Pilkington and A. Matthews, J. Phys. Chem. Solids, 2003, 36, 2110–2120.
36. X. Guo, J. Phys. Chem. Solids, 1999, 60, 539–546.
37. R. O. Hussein, X. Nie and D. O. Northwood, J. Phys. D: Appl. Phys., 2010, 43, 105203–105215.
38. R. Liu, J. Wu, W. B. Xue, Y. Qu, C. L. Yang, B. Wang and X. Y. Wu, Mater. Chem. Phys., 2014, 148, 284–292.
39. Z. Qiu, R. Wang, J. Wu, Y. Zhang, Y. Qu and X. Wu, RSC Adv., 2015, 5, 44149–44159.
40. R. O. Hussein, P. Zhang, X. Nie, Y. Xia and D. O. Northwood, Surf. Coat. Technol., 2011, 206, 1990–1997.
41. J. A. Curran and T. W. Clyne, Surf. Coat. Technol., 2005, 199, 168–176.
42. A. I. Slonova and O. P. Terleeva, Prot. Met., 2008, 44, 65–75.

---