Plasma Electrolytic Oxidation Treatment of Zr-2.5%Nb Alloy for Corrosion Resistance Enhancement

ELISABETA COACA1,2, ALEXANDRU MARIN1*, OANA RUSU1, VIOREL MALINOVSCHI3, VICTOR ANDREI4 1Institute for Nuclear Research Pitesti, 1 Campului Str., Mioveni, Arges, Romania 2Politehnica University of Bucharest, Physical Chemistry and Electrochemistry, Department of Inorganic Chemistry, 132 Calea Grivitei, 010737, Bucharest, Romania 3University of Pitesti, 1 Targul din Vale Str., 110040, Pitesti, Romania 4ELSSA Laboratory SRL, 26 Crinului Str., 110109, Pitesti, Romania

Although zirconium-based alloys are used in nuclear industry due to their low neutron absorption cross-section and favorable mechanical and electrochemical properties, enhanced oxidation and hydration chemical processes occur due to tremendous operating conditions found in water-cooled nuclear reactor environment [1][2][3][4][5][6][7][8][9]. In the case of nuclear pressure tubes made from Zr-2.5%Nb alloy, the black oxide layer formed on this commercial alloy protects the surface against oxidation and corrosion at high temperatures during in-reactor operation [10]. However, plasma electrolytic oxidation (PEO) coatings attracted attention in the last decade regarding the improvement of wear and corrosion resistance of zirconium nuclear alloys by growing ZrO 2 protective layers during electrochemical oxidation of metal surfaces [11,12]. Additionally, taking into account the limited available literature on PEO-treated Zr alloys, including Zy-2 [13,14], Zy-4 [15,16], Zr-1%Nb [17,18] and Zr-2.5%Nb [19,20], further work will be required to optimize and improve the coating layers in the approach of material-coolant interaction.
This paper assesses the enhanced anti-corrosion behavior of commercial Zr-2.5%Nb alloy. The purchased CANDU pressure tubes are always covered with a black oxide film obtained by autoclaving. We propose to prepare additional ceramic films based on ZrO 2 (zirconia) generated onto the surface of existing black oxide layer. For this, plasma electrolytic oxidation is performed by using Na 4 P 2 O 7 and Na 2 SiO 3 aqueous electrolytes. Then, the ZrO 2 films are characterized by X-ray diffraction, metallography of their cross-section and potentiodynamic polarization curves in LiOH aqueous solution.

Experimental part
Plasma electrolytic oxidation process was performed on samples cut out of a commercial Zr-2.5%Nb CANDU pressure tube without mechanically removing the initial black oxide coating, in order to probe another deposition approach as compared to our previous work [12].
Prior to PEO oxidation treatment, the substrates were ultrasonically cleaned in ethyl alcohol and acetone for 10 min. The samples were divided into two groups according to the electrolytic environment for PEO process (table 1). The compositions of the aqueous electrolyte solutions consist of 10 g/L Na 4 P 2 O 7 + 2 g/L KOH (E1 electrolyte) and 10 g/L Na 4 P 2 O 7 + 10 g/L Na 2 SiO 3 ·5H 2 O (E2 electrolyte), respectively. A pulsed unipolar power supply with 150 Hz frequency was employed in a constant voltage regime [21]. A full description of the preparation method was described elsewhere [12]. Table 1 shows the sample codes, applied voltage (U) and PEO process time (τ) in E1 and E2 electrolytes.
Microstructure of the deposited PEO coatings has been studied by X-ray diffraction and optical metallography and the corrosion behavior was evaluated using potentiodynamic polarization curves.
The diffraction patterns were recorded in a parallel beam geometr y using a Rigaku Ultima IV diffractometer equipped with a Cross Beam Optics system and a high precision vertical goniometer (θ-θ) with 285 mm radius. A CuKα radiation source operated at 40 kV and 40 mA was used. The samples were scanned from 10 to 80 o (2θ) in steps of ∆(2θ) = 0.05 o and 4 s step time. An Olympus GX71 optical microscope was employed to observe the multilayered structure of the PEO coatings. The samples were prepared by cutting strips using a diamond saw (Isomet4000) and embedding them in thermosetting resin with the METAPRESS-A (Metkon) mounting press. The mechanical cleaning process including grinding and polishing was performed with abrasive papers of 1200 and 2400-4000 grits, respectively. The chemical attack solution contains 45% HNO 3 , 10% HF and 45% demineralized water.
The corrosion resistance of the oxide coatings was assessed by potentiodynamic polarization measurements carried out on a PARSTAT-2273 potentiostat/galvanostat (Advanced Electrochemical System, Princeton Applied Research) in a three-electrode set-up, including the working electrode (tested sample), a platinum counterelectrode and a saturated calomel reference electrode (0.244 V vs. SHE at 25°C). Polarization curves were recorded at a scan rate of 2 mV/s, starting from -0.5 V to +2 V, in an 0.2 M LiOH aqueous solution.
, and I(hkl) are the integrated experimental intensities in the diffraction patterns for different orientation planes.
Also, the volumic fraction of t-ZrO 2 phase was estimated from the equation: (2) For this purpose, the diffraction lines (-111) and (111) of monoclinic ZrO 2 and diffraction line (101) of tetragonal ZrO 2 , respectively, were successfully indexed and the calculated integrated intensities were refined using the pseudo-Voigt analytical function. The obtained integrated intensities have been used to estimate m-ZrO 2 and t-ZrO 2 phase volume fractions involving equations (1) and (2) [12].
From table 2, it can be seen a dominant monoclinic ZrO 2 phase (~97%) associated with an inhibited tetragonal phase of zirconia (~3%), totally in contrast to our previous results reported in [12] where either NaAlO 2 (pH = 13.0) or NaAlO 2 and NaOH (pH = 13.2) aqueous solutions were used in PEO process. Although identical potentiostatic operating regime was used, crystallographic phase modifications occur and could be related to the substrate surface chemistry or the chemical nature of electrolytic solution employed. However, it is interesting to note that changing the electrolyte environment has no strong influence on the crystallographic behavior of ZrO 2 coating on Zr-2.5%Nb alloy, as evidenced in table 2.  By applying Toraya method [22], the monoclinic and tetragonal phase volumes of zirconia were estimated. In order to do this, volume fraction of m-ZrO 2 (V m ) can be calculated using the equation (1): (1) where X m has an expression:  Figure 2 displays the optical examination of the PEOtreated alloys revealing the layered structure of the coatings. It can be seen that a bi-layered coating was obtained on Zr-2.5%Nb alloy during electrolytic oxidation in E1 electrolyte solution (figs. 2a,b). The layers thicknesses increased with the increase of treatment time for the same electrolytic bath composition. Notably, a different behavior was observed after electrolysis process of Zr alloy in E2 electrolyte (figs. 2c,d). The influence of deposition time on film growth was illustrated by a three-layered structure ( fig. 2d). Apparently, although the coating is found to possess a larger thickness after a longer period of deposition, the total layer thickness generated in E2 electrolyte was approximately the same for both samples ( fig. 2c,d). However, a contribution of commercial black oxide layer cannot be ruled out.
The experimental potentiodynamic polarization curves for commercial Zr-2.5%Nb alloy, M1, M2, M3 and M4 samples were shown in figure 3. Table 2 summarizes calculated values of corrosion potential (E corr ), corrosion current density (j corr ) and corrosion rates (V corr ) for the nontreated and PEO-treated samples. The above electrochemical parameters were estimated by Tafel method with PowerSuite corrosion analysis software. Table 3 shows the key measurements regarding the corrosion resistance of M1, M2, M3 and M4 PEO coatings as compared to Zr-2.5%Nb covered with black oxide. As can be seen, the corrosion current density of the black oxide coating is about 94 x 10 -3 µA/cm 2 , a value with one order of magnitude higher then plasma electrolytic oxidized samples indicating an increased corrosion protection for all PEO-treated samples. Moreover, M1 sample yielded best corrosion resistance showing that the corrosion current density decreased with two orders of magnitude as compared to the commercial Zr-2.5%Nb alloy. One possible explanation for this outcome could be the formation of a more compact and dense structure of about 100 µm thickness layer observed onto the surface during E1 electrolyte deposition, confirmed by metallographic measurements (fig. 2a).

Conclusions
Using constant voltage deposition regime, the ZrO 2 films were prepared on commercial autoclaved Zr-2.5%Nb alloy (covered with black oxide) from two aqueous electrolytes: 10 g/L Na 4 P 2 O 7 + 2 g/L KOH and 10 g/L Na 4 P 2 O 7 + 10 g/L Na 2 SiO 3 ·5H 2 O by means of plasma electrolytic oxidation. The coating layers are uneven, possessing thicknesses between 75-175 µm, depending on the employed electrolyte. XRD patterns indicated the presence of both However, all coatings exhibit increased corrosion resistance with one order of magnitude higher than the autoclaved Zr-2.5%Nb alloy while M1 sample (PEO prepared from 10 g/L Na 4 P 2 O 7 + 2 g/L KOH, 200 V, 5 min) presents a two order of magnitude increase in corrosion protection compared to the black oxide. Both electrolytic environments favor corrosion protection of Zr-2.5%Nb alloy.