Influence of incremental rate of anodising current on roughness and electrochemical corrosion of oxide film on titanium alloy Ti–10V–2Fe

Abstract

In this study, the influence of incremental rate of anodising current V_IR on the crystal structure, surface roughness and corrosion resistance of anodic oxide films on Ti–10V–2Fe–3Al was studied using Raman spectroscopy, atomic force microscopy (AFM), electrochemical impedance spectroscopy (EIS) and potentiodynamic polarisation. The results showed that the surface roughness, crystal structure and electrochemical corrosion behaviour of anodic oxide films were remarkably affected by different V_IR with regard to the change in the rate of anodising voltage. In addition, results of EIS and potentiodynamic polarisation revealed that the film that formed at V_IR of 0··50 A min⁻¹ had the highest corrosion resistance. Quantitative AFM characterisation showed good correlation between surface roughness and EIS results. Furthermore, results of Raman spectroscopy showed that the relative intensity of anatase and rutile had a tendency to increase with a decrease in V_IR before reaching 0··50 A min⁻¹.

Keywords

Titanium Film Roughness Corrosion Anodising current Impedance

Introduction

Titanium alloy Ti–10V–2Fe–3Al has widespread use in aerospace applications because of its excellent mechanical properties combined with satisfactory corrosion resistance behaviour.1 Titanium alloys exhibit a tendency to form a natural oxide layer on their surfaces. This layer protects against the harmful effects of harsh and aggressive environments and is responsible for the good corrosion resistance behaviour of the materials.2 However, there are several weak spots in the natural oxide layer that permit corrosive ions to penetrate the layer and cause material degradation. Anodic oxidation on titanium substrate is an often used surface treatment that increases the thickness and enhances the performance (resistance to wear, adhesion and corrosion) of the oxide film.

The electrochemical parameters of anodic oxidation significantly affect the growth behaviour and properties of the anodic oxide films that form on titanium alloys.3 ^– 7 Parameters that significantly affect the characteristics of the oxide film include electrolyte solution (concentration of reagents) and temperature as well as electrical parameters. The electrical parameters can alter the process of anion transfer during the process of anodisation. Moreover, electrical parameters determine the thickness, surface morphologies and microstructure of anodic coatings.8 Although several studies have been carried out to evaluate the effect of electrical parameters on the growth of anodic films following pulse anodic oxidation,9, 10 these studies have not paid much attention to the method of imposing the anodic current.

In a previously reported work by these authors, thick, uniform and non-transparent anodic oxide films were fabricated on the titanium alloy Ti–10V–2Fe–3Al using a pulse galvanostatic method.11 The present work aims to investigate the effect of the incremental rate of anodising current V_IR on the growth, surface roughness and electrochemical corrosion resistance behaviour of the anodic oxide film that formed on the titanium alloy Ti–10V–2Fe–3Al. Correlations among V_IR, film growth, surface roughness, crystal structure and electrochemical corrosion resistance behaviour of anodic oxide film were recognised on the basis of the results.

Experimental

A forged block of titanium alloy Ti–10V–2Fe–3Al was cut into sheets measuring 100×30×3 mm. The nominal composition of the titanium alloy Ti–10V–2Fe–3Al is given in Table 1. Each sample was abraded with silicon carbide paper of successive grades from 300 to 1500 grit, then rinsed with acetone and deionised water and, finally, air dried.

Table 1

Chemical components of Ti–10V–2Fe–3Al, wt-%

V	Fe	Al	C	N	O	Ti
9··3	1··7	3··2	<0··03	<0··01	<0··130	Balance

Anodic oxidation was carried out in a cell with a thermostat controlled water bath and a magnetic stirring apparatus. The prepared sample was used as anode, and a 1Cr18Ni9Ti stainless steel plate was used as cathode. An aqueous solution of sodium tartrate, which was prepared using analytical grade chemicals and deionised water, was the electrolyte for the reaction.

A pulse galvanostatic power supply, WMY-IV, was used, and the output mode of the power supply is shown in Fig. 1. The pulse of the power supply was unidirectional and exhibited a square wave. The ultimate value of the anodising current was set at 10 A. T_LR denotes the time that the anodic current needs to increase to the specified value and was set at 2, 5, 10, 20 and 30 min; V_IR was 5··0, 2··0, 1··0, 0··50 and 0··33 A min⁻¹ and was changed in correlation with the duration of the reaction. Details of the fabrication conditions have been previously published.11

Figure 1

Output mode of power supply: T_LR time anodic current needs to increase to specified value; T₀ working period

The surface morphologies of the inner and outer anodised layers were evaluated using a scanning electron microscope (SEM, CS3400; Camscan). The surface roughness of the anodic oxide film was observed using a Dimension Icon atomic force microscope (AFM; Veeco, USA). The crystal structure of the anodic oxide film was determined by a Raman spectroscope (Raman; Yvon Jobin Horiba-HR800, using an He–Ne laser without filter at 633 nm).

Electrochemical measurements were carried out using a potentiostat/galvanostat (Parstat 2273; Princeton Applied Research) with a conventional three-electrode electrochemical cell, wherein the samples functioned as working electrode, a platinum plate was used as counter electrode and a saturated calomel electrode (SCE) was used as reference electrode. All experiments were carried out in 0··5 mol L⁻¹ H₂SO₄ solution at 25±3°C.

The anodised electrodes were immersed in H₂SO₄ solution in open circuit conditions for 30 min before the electrochemical impedance spectroscopy (EIS) or potentiodynamic polarisation tests were conducted. The alternating current impedance spectra of the samples were obtained at the open circuit potential, with an amplitude of 10 mV and scan frequencies ranging from 100 kHz to 10 mHz (10 points/decade). Nyquist and Bode plots were obtained by curve fitting using a commercial software package called the Electrochemistry Power Suite. The EIS spectra were simulated using the ZSimpWin simulation software. The scanning rate of the potentiodynamic polarisation curve was 2 mV s⁻¹, with a scanning range from −0··7 to 1··2 V(SCE).

Results

Development of film

Figure 2 shows the change in anodising voltage as a function of anodising time at different V_IR. Details of the results of the anodising voltage obtained at different V_IR are shown in Table 2. The anodising process was classified into three steps for all of the samples. In the first step (step 1), the anodising voltage increased linearly with time up to the maximum anodising voltage. The maximum value and incremental rate of anodising voltage decreased when V_IR decreased during this stage in the process. In the second step (step 2), the anodising voltage decreased proportionally with time of anodisation. The decreasing amplitude and declining rate of the anodising voltage decreased with the reduction of V_IR. In the third step (step 3), the anodising voltage was approximately maintained at a steady value. The ultimate value increased when V_IR decreased before reaching 0··50 A min⁻¹. The steady time decreased with a decrease in V_IR in step 3. This indicates that the ultimate value was greatest at V_IR of 0··50 A min⁻¹.

Figure 2

Anodising forming voltage with anodising time relation curves at different V_IR

Table 2

Results of anodising forming voltage obtained at different V_IR

V_IR/A min⁻¹	Maximum value/V	Ultimate value/V	Decreasing amplitude/V	Change rate
V_IR/A min⁻¹	Maximum value/V	Ultimate value/V	Decreasing amplitude/V	Incremental rate in step 1/V min⁻¹	Decline rate in step 2/V min⁻¹	Steady time in step 3
5··0	124	86	38	17··7	3··8	18
2··0	108	92	16	15··4	2	19
1··0	106	92	14	9··6	1··8	14
0··5	99	95	4	4··7	0··7	8
0··33	92	90	2	3··1	0··5	4

Atomic force microscopy characterisation

The AFM three-dimensional topographic representations for the samples at different V_IR are depicted in Fig. 3. Figure 4 shows the variation of root mean square roughness R_q and average roughness R_a derived from the AFM analysis of the recorded area (20×20 μm). Values of R_q and R_a decreased with a decrease in V_IR. The results indicated that the surface of the anodised samples gradually became smooth as V_IR decreased before reaching 0··50 A min⁻¹.

Figure 3

Images (AFM) of three-dimensional anodic oxide films at different V_IR: a 5··0 A min⁻¹; b 2··0 A min⁻¹; c 1··0 A min⁻¹; d 0··50 A min⁻¹; e 0··33 A min⁻¹

Figure 4

Surface roughness of anodic oxide films at different V_IR

Raman spectroscopic characterisation

Figure 5 presents the Raman spectra of anodic oxide films formed on titanium alloy Ti–10V–2Fe–3Al at different V_IR. Results showed that the films had the same crystal structure12, 13 (anatase and rutile). The intensity of the peaks increased with a decrease in V_IR before the maximum V_IR of 0··5 A min⁻¹ was reached. Thereafter, the intensity decreased with a decrease in V_IR.

Figure 5

Raman spectra of anodic oxide films on alloy at different V_IR

Electrochemical impedance spectroscopy

The Bode plots of samples at different V_IR are shown in Fig. 6. The |Z| value, first, increased with decreasing V_IR before it reached the maximum at 0··50 A min⁻¹. Thereafter, the |Z| value at V_IR = 0··33 A min⁻¹ was approximately the same as that at V_IR = 0··5 A min⁻¹ at low frequency. It consisted of two time constants in the selected range of frequency (10⁵–10⁻²) that were associated with inner and outer layers respectively. The inner layer was compact, and the outer layer was porous, as shown in Fig. 7.

Figure 6

Bode plots of anodic oxide films at different V_IR

Figure 7

Images (SEM) of a inner and b outer anodised layers

On the basis of these facts, the obtained spectra were interpreted in terms of an ‘equivalent circuit’, as shown in Fig. 8. The depicted model assumes that the oxide film on the titanium alloy comprises a compact inner layer and a porous outer layer (Fig. 7), with R_S as the resistance of the solution, R_pL as the polarisation resistance of the porous layer, Q_pL as the capacitance of the porous layer, R_bL as the polarisation resistance of the barrier layer, Q_bL as the capacitance of the barrier layer and W as the warburg impedance. A constant phase element (CPE), representing a shift from the ideal capacitor, was used instead of the capacitance. The impedance of the CPE Z_CPE was defined by equation (1) as (1) All the impedance parameters are shown in Table 3. These impedance parameters have confirmed the conclusions of the qualitative analysis of EIS results. The equivalent circuit of the EIS for the anodic oxide films shows the base structure of the films, and the fitting values indicate the changing regularity of the structure with a decrease in V_IR.

Figure 8

Equivalent circuit diagram for samples at different V_IR

Table 3

Alternating current impedance parameters for specimens at different V_IR

V_IR/A min⁻¹	R_s/Ω cm²	R_p/Ω cm²	Q_p/F cm⁻²	n _p	R_b/Ω cm²	Q_b/F cm⁻²	n _b	W/μ Ω⁻¹ cm⁻²s^−1/2
5··0	14··90	77··19	1··513×10⁻⁶	1	2··462×10⁴	7··116×10⁻⁵	0··5839	3··774×10⁻⁴
2··0	17··33	85··43	1··364×10⁻⁶	1	4··786×10⁴	5··094×10⁻⁵	0··6223	2··006×10⁻⁴
1··0	18··69	82··30	1··545×10⁻⁶	0··9833	5··667×10⁴	4··122×10⁻⁵	0··6573	1··012×10⁻⁴
0··50	15··54	7··404	3··191×10⁻⁵	0··6829	5··862×10⁴	3··854×10⁻⁷	1	4··741×10⁻⁵
0··33	13··14	5··464E4	1··794×10⁻⁵	0··7258	7··405×10⁴	1··418×10⁻⁶	0··9285	1··797×10⁻⁵

Potentiodynamic polarisation

Figure 9 shows the results of experimental potentiodynamic polarisation of samples in a 0··5 mol L⁻¹ H₂SO₄ solution at different V_IR. The corrosion current density i_corr, corrosion potential E_corr and anodic/cathodic Tafel slopes b_a and b_c were derived from the data obtained. On the basis of the approximate linear polarisation behaviour near the open circuit potential, the polarisation resistance R_p was determined using the Stern–Geary equation14 (2) The results of potentiodynamic polarisation are listed in Table 4. These data clearly show that the corrosion resistance of the samples increased with a decrease in V_IR before a V_IR of 0··50 A min⁻¹. The R_p value reached the maximum when V_IR was 0··50 A min⁻¹. The sample that formed an oxide layer at V_IR of 0··50 A min⁻¹ showed corrosion potential (0··4102 V) that was significantly more positive than that at V_IR of 5··0 A min⁻¹ (0··1996 V), which facilitated deduction of the thermodynamic tendency for corrosion. Furthermore, the sample that formed at V_IR of 0··50 A min⁻¹ showed lower corrosion current density (one order less) than that at V_IR of 5··0 A min⁻¹, which facilitated a deduction in the rate of corrosion.

Figure 9

Potentiodynamic polarisation curves of anodic oxide films at different V_IR

Table 4

Results of potentiodynamic corrosion tests of samples at different V_IR

V_IR/A min⁻¹	b_a/mV	b_c/mV	E_corr/V	i_corr/A cm⁻²	R_p/Ω
5··0	285··11	150··38	0··1996	3··347×10⁻⁶	1··277×10⁴
2.0	381··79	168	0··2215	2··924×10⁻⁶	1··732×10⁴
1··0	254··98	171··34	0··2431	2··833×10⁻⁶	1··571×10⁴
0··50	204··15	154··33	0··4102	8··110×10⁻⁷	4··764×10⁴
0··33	226··32	132··53	0··2533	9··566×10⁻⁷	3··794×10⁴

Discussion

In general, it is assumed that the behaviour for the growth of anodic oxide film is characterised by the electrochemical dynamics between the ability for the formation of oxide film and the oxide dissolution rate determined by the nature of the electrolyte. The current density increased significantly more rapidly at V_IR of 5··0 A min⁻¹ than at 0··33 A min⁻¹. Therefore, when V_IR was 5··0 A min⁻¹, the oxide thickness and voltage also increased rapidly. In this situation, the rate of oxide film formation overwhelmingly exceeded the rate of chemical dissolution. As a result, the incremental rate of the voltage in step 1 was much faster at V_IR of 5··0 A min⁻¹ (17··7 V min⁻¹) than at V_IR of 0··33 A min⁻¹ (3··1 V min⁻¹). This might be associated with the relatively high growth rate that led to the low compactness of the inner layer of oxide film at V_IR of 5··0 A min⁻¹ in step 1. In addition, the rate of decline in step 2 was higher at V_IR of 5··0 A min⁻¹ (3··8 V min⁻¹) than at V_IR of 0··33 A min⁻¹ (0··5 V min⁻¹). The rate of decline in step 2 might be related to the uniformity of the outer layer of the oxide film. In step 3, a dynamic equilibrium between the rate of oxide formation and the rate of chemical dissolution was established. As shown in Table 2, the equilibrium in step 3 was maintained for only 4 min at V_IR of 0··33 A min⁻¹ and might be related to the thickness of the outer layer.

The surface roughness, crystal structure and electrochemical impedance of the anodic oxide film were strongly dependent on V_IR in terms of the changing rate of the anodising voltage.

The crystal structure of the anodic oxide films can be explained by the ultimate voltage. In general, at low applied voltages, the anodic oxide film is amorphous. With increasing voltages, the structure of the oxide film changes from an amorphous to a crystalline oxide. The ultimate voltage value is maximum at V_IR of 0··50 A min⁻¹. The intensity of the peaks corresponds to the ultimate voltage value.

From the EIS results, it can be seen that the solution resistance R_s increased with a decrease in V_IR before 1··0 A min⁻¹ and decreased thereafter, which might be associated with the change of the surface state of the anodic oxide films (such as porosity or real surface area). This concurred well with the porosity of the anodic oxide films (Fig. 3). The R_p values are much smaller than the R_b values, indicating that the outer porous layer exhibits a smaller resistance than the inner barrier layer (Table 3); furthermore, the R_b values increase when V_IR decreases, indicating that the compact inner layer has a higher ability to protect from corrosion. The lower the incremental rate is in step 1, the higher is the corrosion protection ability of the inner layer before reaching a V_IR of 0··50 A min⁻¹.

As V_IR decreases, the value of n_b increases before V_IR = 0··50 A min⁻¹, which indicates that the ‘scattered effect’ of Q_bl is weakened and the capacitive interface is strengthened. However, when V_IR was 0··33 A min⁻¹, the value of n_b decreased, and the capacitive interface was weakened. This was attributed to the fact that the inner layer that formed at V_IR of 0··33 A min⁻¹ may be less compact than that formed at V_IR of 0··50 A min⁻¹. The compactness of the inner layer increases as V_IR decreases before V_IR = 0··50 A min⁻¹. This result is consistent with the analysis of the incremental rate in step 1.

In addition, according to Liu,15 the empirical exponent of the CPE is related to the fracture dimension D_F of the coarse surface. The change in n_p is mainly attributable to the roughness of the surface, and D_F is a physical variable that corresponds to the roughness. The relation between n_p and D_F is shown in equation (3) as follows16 (3) From equation (3), it can be seen that the value of D_F decreased as V_IR decreased before a V_IR of 0··50 A min⁻¹, which is consistent with the results of surface roughness obtained from AFM characterisation.

Furthermore, the electrode is a circle flat with an area of 1 cm²; therefore, there may be a semi-infinite length diffusion step during the process at the electrode. The Warburg impedance decreases when V_IR decreases. The diffusing resistance of the reactants is mainly attributed to the inner layer.

Based on the EIS and potentiodynamic polarisation results discussed above, the sample formed at V_IR of 0··50 A min⁻¹ has the best corrosion resistance, which could be attributed to the special structure of the inner compact layers and outer porous layers of the anodic oxide films. The special structure of the anodic oxide film could be attributed to the changing rate of anodising voltage at different V_IR.

Conclusion

The influence of the incremental rate of anodising current on the structure, surface roughness and electrochemical corrosion of the anodic oxide film on Ti–10V–2Fe–3Al alloy has been investigated in this work, and the following conclusions were obtained.

The relational curves of the anodising voltage and anodising duration at different V_IR can be classified into three steps. The surface roughness, crystal structure and behaviour of electrochemical corrosion of the anodic oxide film were remarkably affected by the different V_IR with regard to the changing rate of anodising voltage in the three steps of the reaction.

The EIS results and the equivalent circuit analysis have indicated that the base structure of the films consists of a compact inner layer and a porous outer layer, and the structure changed proportionally as V_IR decreased. For the inner layer, the lower the incremental rate was in step 1, the higher was the corrosion protection ability maintained in the layer. For the outer layer, the lower the rate of decline in step 2, and the longer the steady time in step 3, the higher the corrosion protection ability maintained.

The uniformity of the surface and the relative intensity of anatase and rutile tended to increase with a decrease in V_IR before 0··50 A min⁻¹.

When V_IR was 0··50 A min⁻¹, the anodic oxide film was the most uniform, as shown by the least surface roughness, and the amount of crystal phase of the film was the largest. In addition, EIS and potentiodynamic polarisation results have shown that the sample formed at V_IR of 0··50 A min⁻¹ had the best corrosion resistance.

References

Raghunathan

, Dashwood

, Jackson

, Vogel

, Dye

: Mater. Sci. Eng. A, 2008, A488, 8–15.

Fovet

, Gal

, Chemla

: Talanta, 2001, 53, 1053–1063.

Matykina

, Skeldon

, Thompson

: Surf. Eng., 2007, 23, 412–418.

Montazeri

, Dehghanian

, Shokouhfar

, Baradaran

: Appl. Surf. Sci., 2011, 257, 7268–7275.

Shokouhfar

, Dehghanian

, Baradaran

: Appl. Surf. Sci., 2011, 257, 2617–2624.

Liu

, Yi

, Li

, Yu

, Wu

: J. Appl. Electrochem., 2010, 40, 1545–1553.

Sinebryukhov

, Gnedenkov

, Khrisanfova

, Gnedenkov

: Surf. Eng., 2009, 25, 565–569.

Verdier

, Boinet

, Maximovitch

, Dalard

: Corros. Sci., 2005, 47, 1429–1444.

Park

, Lee

: Surf. Interface Anal., 2011, 43, 1030–1037.

10.

Yao

, Jiang

, Sun

, Xin

, Li

: Thin Solid Films, 2004, 468, 120–124.

11.

Liu

, Yi

, Li

, Yu

, Xu

: Int. J. Miner. Metall. Mater., 2009, 16, 96–100.

12.

Ohsaka

, Izumi

, Fujiki

: J. Raman Spectrosc., 1987, 7, 321–324.

13.

Ragalevičius

, Stalnionis

, Niaura

, Jagminas

: Appl. Surf. Sci., 2008, 254, 1608–1613.

14.

Stern

, Geary

: J. Electrochem. Soc., 1957, 104, 56–63.

15.

Liu

: Phys. Rev. Lett., 1985, 55, 529–532.

16.

Zhang

, Yang

, Shi

, Guo

: J. Chin. Ceram. Soc., 2003, 31, 127–132.

Influence of incremental rate of anodising current on roughness and electrochemical corrosion of oxide film on titanium alloy Ti–10V–2Fe–3Al

Abstract

Keywords

Introduction

Experimental

Results

Development of film

Atomic force microscopy characterisation

Raman spectroscopic characterisation

Electrochemical impedance spectroscopy

Potentiodynamic polarisation

Discussion

Conclusion

References