Effects of Heat Treatment on Microstructures and Properties of Cold Rolled Ti-0.3Ni Sheets as Bipolar Plates for PEMFC

Abstract

As promising materials for bipolar plates substrate, as-cold rolled Ti-0.3Ni (wt.%) sheets were heat treated with three different processes in this work. As-cold rolled sheets consist of α matrix and dispersed Ti₂Ni intermetallic precipitates, and typical Widmanstatten microstructure can be observed after heat treatment. Lamellar Ti₂Ni precipitates inside the colonies. Elongation of as-cold rolled sheets equals less than 7% while this value rises up to around 20%, and tensile strength decreases by more than 47% after heat treatment. Open circuit potentials of as-cold rolled sheets treated at 950 °C for 1 h followed by wind cooling (950 °C/1 h/WC), sheets aged at 500 °C for 3 h followed by air cooling (950 °C/1 h/WC + 500 °C/3 h/AC), and sheets treated at 950 °C for 1 h followed by furnace cooling (950 °C/1 h/FC) equals −0.536 V, −0.476 V, −0.486 V, −0.518 V, respectively. A potentiodynamic polarization test reveals that all of the specimens exhibit typical active–passive transition behavior. Sheets treated at 950 °C/1 h/WC possess the lowest corrosion current density (155.4 μA·cm⁻²). Results of electrochemical impedance spectroscopy (EIS) show that 950 °C/1 h/WC treated sheets possess the largest polarization resistance (R_pol), 122.6 Ω·cm². Moreover, steady-state current densities (I_ss) increase in the order of 950 °C/1 h/WC, 950 °C/1 h/WC + 500 °C/3 h/AC, 950 °C/1 h/FC according to the results of potentiostatic polarization. This can be attributed to various amounts of Ti₂Ni precipitation caused by different cooling rates.

1. Introduction

Proton exchange membrane fuel cells (PEMFCs) are an energy conversion device with promising development potential due to the low operating temperature, fast start-up, long service life, high energy conversion efficiency, and stable operation [1,2,3,4,5]. As a component of PEMFCs, bipolar plates account for around 45% of total cell cost and 78% of total cell weight [6], undertaking the responsibility of conducting current, transmitting heat, distributing reaction gas, and removing water in a fuel cell.

Nowadays, bipolar plates made of graphite are widely commercialized. However, high manufacturing cost and inferior mechanical properties confined their further application. In comparison with graphite bipolar plates, metallic bipolar plates have attracted the attention of many researchers because of their lower manufacturing cost, superior formability as well as enough mechanical strength [5,7,8,9,10,11,12,13]. However, poor corrosion resistance becomes the main problem that metallic bipolar plates need to be settled. Thus, a lot of work for improving the corrosion resistance of metallic bipolar plates by surface modification has been carried out. However, defects such as pinholes often occur during coating manufacturing, and corrosive electrolytes will react with the underlying substrate [5,14]. Consequently, the corrosion resistance of the substrate largely determines the longevity of bipolar plates.

At present, stainless steel is the most widely used metallic bipolar plate substrate [7,9,10]. Nevertheless, high density and inferior corrosion resistance in PEMFCs operating conditions become disadvantages of bipolar plates made of stainless steel. As a prospective candidate material for bipolar plate substrate, titanium alloys possess superior corrosion resistance and low density, which can prolong the longevity and reduce the weight of bipolar plates. However, taking cost, corrosion resistance, and formability into consideration, most titanium alloys are not suitable for bipolar plate substrate.

As a common method that improves the corrosion resistance of titanium in reducing acid, alloying is a low-cost, simple, and effective way. Tomashov et al. [15] proposed several ways to enhance the corrosion resistance of titanium. One of the methods is cathodic alloying by adding elements including platinum-group metals (PGMs) or Ni into titanium. Sedriks et al. [16] studied the electrochemical behavior of titanium and Ti-Ni alloys in acid NaCl solution at ambient temperature and boiling point, finding that introduction of Ti₂Ni intermetallic compounds enhanced corrosion resistance of titanium by increasing the efficiency of the cathodic process. Bonilla et al. [17] improved corrosion resistance of titanium in 0.1 M H2SO4 at 80 °C via surface alloying with Ni, Ni-Mo, or Pd. Wang et al. [18] compared the corrosion behavior of Ti-0.3Mo-0.8Ni, Ti-0.2Pd, and pure titanium in 25 °C fluoride-containing sulfuric acids. By using mixed potential theory, it was found that Ti-0.3Mo-0.8Ni and Ti-0.2Pd alloys were more corrosion resistant than pure titanium due to the acceleration effect of Ni and Pd on the cathodic process. Robert [19] systematically investigated the electrochemical behavior of TiCode-12 alloy in hydrochloric acid and found that alloy passivity resulted from a galvanic coupling between intermetallic Ti₂Ni and Ti matrix.

To our knowledge, there are few studies on the corrosion behavior of titanium-nickel alloys in the PEMFC’s operating environment. In the present work, cold-rolled binary Ti-0.3Ni (wt.%) sheets were manufactured. The content of Ni addition is controlled to less than 1% to avoid the adverse influence of elements on the formability of the alloys. In order to obtain various microstructures, as-cold rolled sheets were heat-treated in different processes. Microstructures of as-cold rolled and heat-treated sheets were observed. Corrosion behavior and tensile properties of as-cold rolled and heat-treated sheets were investigated.

2. Materials and Methods

Materials used in this work were 0.7 mm thick as-cold rolled Ti-0.3Ni (wt.%) sheets. The as-cold rolled sheets were heat-treated in electric chamber furnace by three processes, including solid solution treatment at 950 °C for 1 h followed by wind cooling (denoted as 950 °C/1 h/WC), solid solution treatment at 950 °C for 1 h followed by wind cooling, then the specimens were aged at 500 °C for 3 h followed by air cooling (denoted as 950 °C/1 h/WC + 500 °C/3 h/AC), solid solution treatment at 950 °C for 1 h followed by furnace cooling (denoted as 950 °C/1 h/FC).

10 mm × 10 mm square specimens were cut from each sheet by electrical discharge machining for microstructural investigation and electrochemical experiments. The specimens were ground with SiC emery paper up to 3000 grit, and subsequently mechanical polished with 0.25 μm diamond. Then, each specimen was ultrasonically cleaned successively with acetone, ethanol, and distilled water. For microstructure observation, the polished specimens were etched in Kroll’s reagent, i.e., 10 vol.% HF, 10 vol.% HNO₃, 80 vol.% H₂O. Microstructure observation of specimens was conducted by scanning electron microscopy (FEI, Hillsboro, OR, USA). Area fractions of Ti₂Ni intermetallic compounds in specimens were calculated using Image-Pro Plus software (6.0, Media Cybernetics Inc., Rockville, MD, USA).

Electrochemical experiments were carried out with CHI 760E electrochemical workstation. A conventional three-electrode cell was used, which consisted of Pt foil counter electrode, saturated calomel reference electrode (SCE) connected to the cell via a salt bridge with a luggin capillary, and the specimen as the working electrode. The working electrodes were embedded in epoxy resin with 1 cm² exposed area. Unless otherwise stated, potentials in this work are provided with respect to SCE.

Open circuit potential (OCP) measurements were performed for 3600 s. Potentiodynamic polarization test was conducted from −1.0 V_SCE to +1.5 V_SCE at a scan rate of 0.5 mV/s. Electrochemical impedance spectroscopy (EIS) analysis was carried out at OCP using a sinusoidal potential perturbation of 10 mV in a frequency ranging from 10 mHz to 100 kHz. Fitting of the EIS data was carried out by Zview software (3.1, Scribner Associates Inco., Southern Pines, NC, USA). All of the measurements were at least triplicated in simulated PEMFC solution, i.e., naturally aerated 0.5M H₂SO₄ solutions with 5 ppm F⁻. Potentiostatic polarization measurements at +0.6 V were conducted for 21,600 s (6 h). Temperatures of electrolyte were maintained at 80 °C via thermostatic water bath during tests.

Tensile testing at room temperature was performed using Instron 5569 universal testing machine (Instron, Norwood, MA, USA) at the strain rate of 5 × 10⁻⁴ s⁻¹. The tensile direction was respectively parallel to rolling direction (RD) and transverse direction (TD) of each heat-treated sheet. Strain extensometer was used in order to guarantee the accuracy of strain measurement.

3. Results and Discussion

3.1. Microstructure

Figure 1 exhibits photographs of Ti-0.3Ni alloys treated by various processes. Some pits can be seen in high magnification photos due to corrosion caused by the etchant. On the basis of the Ti-Ni binary phase diagram, eutectoid microstructure composed of α-Ti and Ti₂Ni intermetallic compounds will be obtained in the form of the following reaction at 765 °C [19,20]:

Β→α + Ti₂Ni

(1)

For as-cold rolled sheets, dispersed Ti₂Ni in the α-Ti matrix and a large number of fine grains due to the cold rolling process can be observed. In contrast, a typical Widmanstatten microstructure can be found after being heat-treated at 950 °C. The microstructure evolution of heat-treated sheets is described as follows. During the heating of specimens, α grains transform into β grains via the reaction α→β at α/β transform temperature. Moreover, the soluting of Ni atoms into Ti lattice reduces the hindrance of grain growth. Consequently, the size of β grains significantly increases. In the period of cooling, the α phase firstly forms on the grain boundary of β grains when temperature decreases to β/α transform temperature. Then α grains grow into β grains with temperature further decreasing. When temperature declines to 765 °C, i.e., the eutectoid temperature of Ti-0.3Ni alloy according to Ti-Ni binary phase diagram [16], a eutectoid reaction takes place, and β phase transforms into α phase and Ti₂Ni intermetallic compounds.

As the highest cooling rate in the present study, in the case of 950 °C/1 h/WC, the temperature of the specimen decreases so rapidly that only a small amount of Ni atoms precipitate in the form of Ti₂Ni. Thus, less Ti₂Ni is observed in Figure 1c,d. Moreover, a high cooling rate results in the formation of jagged grains, as marked in Figure 1d. As for specimens treated by 950 °C/1 h/WC + 500 °C/3 h/AC, more Ti₂Ni precipitates on account of the lower cooling rate, as depicted in Figure 1e. Additionally, jagged grains still exist, but the number of them becomes rather smaller. As for the specimen treated at 950 °C/1 h/FC, the cooling rate is further declined, and the time for Ti₂Ni precipitation is prolonged, making Ti₂Ni precipitates more sufficient. This can be evidenced that the amount of Ti₂Ni in Figure 1g is more than Figure 1c,e. The widths of Ti₂Ni laths are approximately 1 μm, larger than other specimens. Typical (α-Ti + Ti₂Ni) eutectoids with clear boundaries instead of jagged grains can be observed.

3.2. Mechanical Properties

Figure 2 presents engineering stress-strain curves of specimens treated by different processes. The results of the room temperature tensile experiment are listed in

Table 1. Mechanical properties of as-cold rolled and heat treated sheets.

Specimen	Tensile Strength /MPa	Yield Strength /MPa	Elongation /%
As-cold rolled-RD	733.6	610.2	6.12
As-cold rolled-TD	748.8	564.0	2.01
950 °C/1 h/WC-RD	374.3	269.2	19.1
950 °C/1 h/WC-TD	376.8	281.0	22.3
950 °C/1 h/WC +500 °C/3 h/AC-RD	385.1	291.2	20.8
950 °C/1 h/WC +500 °C/3 h/AC-TD	375.3	282.4	20.3
950 °C/1 h/FC-RD	356.3	261.5	15.7
950 °C /1 h/FC-TD	341.5	256.0	21.0

. According to Figure 2 and Table 1, as-cold rolled specimens possess the highest ultimate tensile strength (UTS) of 733.6 MPa and 748.8 MPa in RD and TD, respectively. However, elongations of as-cold rolled specimens equal only 6.12% and 2.01% in RD and TD, respectively.

While UTSs of the heat-treated specimens significantly differ from the as-cold rolled ones. Anisotropy of the specimens reduces by observing the curves after heat treatment. Moreover, the plasticity of the heat-treated specimens is enhanced, observed by the apparent increase in elongation. For 950 °C/1 h/WC treated specimens, UTSs in RD and TD are 374.3 MPa and 376.8 MPa, respectively. Moreover, elongations of the specimens in RD and TD are 19.1% and 22.3%. In the case of heat-treated specimens, no discernable difference can be noted among them. Additionally, under the action of aging at 500 °C, more Ti₂Ni precipitates, resulting in slightly lower elongation, i.e., 20.8% and 20.3% in RD and TD, respectively. In the case of specimens treated at 950 °C/1 h/FC, UTSs and elongations decrease to 261.5 MPa and 256.0 MPa, 15.7% and 21.0% in RD and TD, respectively. As the previous investigation stated, higher content of the intermetallic phase in the ductile matrix will deteriorate mechanical properties, especially the plasticity of the alloys [21]. Herein, the hard, brittle Ti₂Ni phase existing in the alloy matrix is the dominant factor that affects the mechanical properties of the heat-treated Ti-0.3Ni sheets. Moreover, the amount of Ti₂Ni precipitation in heat-treated sheets increases in the order of 950 °C/1 h/WC, 950 °C/1 h/WC + 500 °C/3 h/AC, 950 °C/1 h/FC. Hence, elongation decreases in the same order.

3.3. Open Circuit Potential Measurements

Figure 3 shows the variation of OCPs with immersion time for as-cold rolled and heat-treated Ti-0.3Ni sheets in simulated PEMFC solution. For all of the specimens, OCPs turn out to be a quick decrease followed by an abrupt decline to the lowest values due to chemical dissolution of native oxide formed on the surface by the following reaction: [22]

TiO₂ + 6F⁻ + 4H⁺→TiF₆²⁻ + 2H₂O

(2)

After decreasing to the lowest values, a small rebound of curves can be observed. Finally, the OCPs relatively stabilize at a certain value, and no evident fluctuation is noted.

As-cold rolled specimen possesses the most negative OCP (−0.536 V). The heat-treated sheets exhibit nobler potential in different degrees compared with as-cold rolled ones. OCPs of sheets treated at 950 °C/1 h/WC and 950 °C/1 h/WC + 500 °C/3 h/AC are −0.476 V and −0.486 V, which are respectively 0.06 V and 0.05 V nobler than the as-cold rolled one. With respect to sheets treated at 950 °C/1 h/FC, their OCP (−0.518 V) is nobler than as-cold rolled ones and more negative than 950 °C/1 h/WC as well as 950 °C/1 h/WC + 500 °C/3 h/AC treated sheets.

3.4. Potentiodynamic Polarization Measurements

Potentiodynamic polarization curves for as-cold rolled and heat-treated sheets in naturally aerated 0.5 M H₂SO₄ + 5 ppm F⁻ solution are displayed n in Figure 4. According to the curves, apparent anodic peak and passive zone can be observed for all of the specimens, demonstrating typical active–passive behavior of specimens in simulating PEMFC solution. Moreover, corrosion potentials (E_corr) of the specimens increase in the order of as-cold rolled, 950 °C/1 h/WC, 950 °C/1 h/WC + 500 °C/3 h/AC, 950 °C/1 h/FC, and their values equal −0.481 V, −0.437 V, −0.454 V, −0.491 V, respectively. Corrosion current densities (i_corr) are obtained through Tafel extrapolation method and the results of fitting equal 446.3 μA·cm⁻², 155.4 μA·cm⁻², 1286 μA·cm⁻², 482.1 μA·cm⁻² for as-cold rolled, 950 °C/1 h/WC, 950 °C/1 h/WC + 500 °C/3 h/AC, 950 °C/1 h/FC, respectively. Attention should be paid to the work by Sedriks et al. [16], where the conclusion that Ti₂Ni could enhance the corrosion resistance of titanium was drawn. However, in the present work, the introduction of Ti₂Ni does not appear to have an obvious positive effect on the corrosion resistance of titanium. The contradiction between these results can be ascribed to the different content of Ni in these two investigations. The Ni contents used in the former work were more than 2 wt.%. Nonetheless, Ni contents of the alloys equal 0.3 wt.% in this work. As stated above, Ni content should be controlled at a lower level so as to decrease the adverse effect of Ti₂Ni on the plasticity of the sheets used for bipolar plates.

According to relevant research, the cathodic reaction of titanium alloys in acid solution is mainly a hydrogen evolution reaction (HER). In the anodic branch of the curves, there are several regions, including the active dissolution region, active–passive transition region and passive region existing. In the active region, titanium dissolves as the following steps [23,24]. The detailed curves of each specimen near the active region are shown in Figure 4b.

$${\mathrm{Ti}+\mathrm{H}}_2\mathrm{O}\underset{ }{\leftrightarrow }\mathrm{Ti}{\left({\mathrm{H}}_2\mathrm{O}\right)}_{\mathrm{ad}}$$

(3)

$$\mathrm{Ti}{\left({\mathrm{H}}_2\mathrm{O}\right)}_{\mathrm{ad}}\underset{ }{\leftrightarrow }\mathrm{Ti}{\left({\mathrm{OH}}^{-}\right)}_{\mathrm{ad}}{+\mathrm{H}}^{+}$$

(4)

$$\mathrm{Ti}{\left({\mathrm{OH}}^{-}\right)}_{\mathrm{ad}}\underset{ }{\leftrightarrow }{\left(\mathrm{TiOH}\right)}_{\mathrm{ad}}{+\mathrm{e}}^{-}$$

(5)

$${\left(\mathrm{TiOH}\right)}_{\mathrm{ad}}\underset{ }{\leftrightarrow }{\left(\mathrm{TiOH}\right)}_{\mathrm{ad}}^{+}{+\mathrm{e}}^{-}$$

(6)

$${\left(\mathrm{TiOH}\right)}_{\mathrm{ad}}^{+}\underset{ }{\to }{\left(\mathrm{TiOH}\right)}^{2+}{+\mathrm{e}}^{-}$$

(7)

$${\left(\mathrm{TiOH}\right)}^{2+}{+\mathrm{H}}^{+}\underset{ }{\leftrightarrow }{\mathrm{Ti}}^{3+}{+\mathrm{H}}_2\mathrm{O}$$

(8)

Within the period of the active–passive transition, as shown in Figure 3, the following reactions occur, and the active dissolution of titanium will be inhibited [23,25,26]. The decreasing current density should be associated with the formation of TiO₂ in reaction 13.

$${\left(\mathrm{TiOH}\right)}_{\mathrm{ad}}^{+}\underset{ }{\leftrightarrow }{\left(\mathrm{TiOH}\right)}_{\mathrm{ad}}^{2+}{+\mathrm{e}}^{-}$$

(9)

$${\mathrm{H}}_2\mathrm{O}+{\left(\mathrm{TiOH}\right)}_{\mathrm{ad}}^{ }\underset{ }{\leftrightarrow }{\left[\mathrm{Ti}{\left(\mathrm{OH}\right)}_2\right]}_{\mathrm{ad}}^{2+}{+\mathrm{H}}^{+}+{\mathrm{e}}^{-}$$

(10)

$${\mathrm{H}}_2\mathrm{O}+{\left(\mathrm{TiOH}\right)}_{\mathrm{ad}}^{ }\underset{ }{\to }{\left[\mathrm{Ti}{\left(\mathrm{OH}\right)}_2\right]}_{ }^{2+}{+\mathrm{H}}^{+}+{\mathrm{e}}^{-}$$

(11)

$${\left[\mathrm{Ti}{\left(\mathrm{OH}\right)}_2\right]}_{\mathrm{ad}}^{2+}\underset{ }{\to }{\left[\mathrm{Ti}{\left(\mathrm{OH}\right)}_2\right]}_{ }^{2+}$$

(12)

$${\left[\mathrm{Ti}{\left(\mathrm{OH}\right)}_2\right]}_{\mathrm{ad}}^{2+}\underset{ }{\leftrightarrow }{\mathrm{TiO}}_2{+2\mathrm{H}}^{+}$$

(13)

Eventually, the curves enter a passive region ranging from 0 V–1.5 V. In the region, due to the valve metals properties of titanium [27,28,29,30], anodically-grown passive film forms upon the surface of specimens and current densities decline at a relatively low rate with the applied potential scanning to nobler direction according to Figure 4.

3.5. Electrochemical Impedance Spectroscopy Measurements

EIS measurements of as-cold rolled and heat-treated specimens in naturally aerated 0.5 M H₂SO₄ + 5 ppm F⁻ solution were presented through Bode plots and Nyquist plots, respectively, in Figure 5a,b. According to Figure 5a, each specimen possesses two peaks in medium and low frequency. Moreover, the maximum phase angle of each specimen is lower than 60°. The phase angles of 950 °C/1 h/WC and as-cold rolled sheets are relatively lower than other sheets. Two capacitive loops can be observed in the Nyquist plot of each specimen. According to relevant research, the phenomenon observed above is an echo of a porous layer formed on the sheets due to the adverse effect of fluoride ions on passive film [18,31,32,33]. Radii of low-frequency capacitive loops decrease in the order of 950 °C/1 h/WC, as-cold rolled, 950 °C/1 h/FC as well as 950 °C/1 h/WC + 500 °C/3 h/AC.

Results of EIS were investigated by the equivalent circuit exhibited in Figure 6, which was applied extensively for Ti alloys to fit EIS data with two-time constants [18,31,33]. In the equivalent circuit, R_f and Q_f correspond to the resistance and capacitance of the porous layer, respectively. R_ct and Q_dl are charge transfer resistance and capacitance of the electrical double layer, respectively. Moreover, R_s is the resistance of solution between passive film/solution interface and the tip of the luggin probe [26,34]. To substitute the ideal capacitance due to the distribution of relaxation times as a result of the non-uniform surface of the alloy in the solution, a constant phase element (CPE) is employed. The impedance of CPE is defined as the following equation:

Z_CPE = [C(jω)ⁿ]⁻¹

where j² = −1, ω = 2πf. The exponent n is an adjustable parameter that lies between −1 and 1, corresponding to a perfect inductor and perfect capacitor, respectively.

Parameters obtained from the equivalent circuit model are listed in

Table 2. Equivalent circuit parameters for as-cold rolled and heat treated sheets in naturally aerated 0.5 M H₂SO₄ + 5 ppm F⁻ solution.

Specimen	Rs / Ω·cm2	Qdl / mS·cm−2·sn	ndl	Rct / Ω·cm2	Qf / mS·cm−2·sn	nf	Rf / Ω·cm2	χ2 /10−3
Process 1 *	7.06	0.50	0.91	18.0	50.4	0.84	25.9	0.56
Process 2	9.82	0.48	0.89	30.5	29.5	0.86	90.9	0.94
Process 3	6.67	1.31	0.87	10.4	91.2	0.91	20.1	0.67
Process 4	13.64	0.56	0.92	16.6	54.2	0.86	25.4	0.31

*: Process 1 represents cold rolling. Process 2 represents 950 °C/1 h/WC treatment. Process 3 represents 950 °C/1 h/WC + 500 °C/3 h/AC treatment. Process 4 represents 950 °C/1 h/FC treatment.

. The validity of the obtained parameters is evaluated by the quite small chi-square values (χ²) [31]. The fitted data is in good accord with experiment data based on the table. In the present study, due to the porous structure, the layer cannot provide sufficient protection to the substrate from interacting with the solution. Based on Table 2, R_ct of as-cold rolled sheets and the ones treated at 950 °C/1 h/WC, 950 °C/1 h/WC + 500 °C/3 h/AC, 950 °C/1 h/FC equal 18.0 Ω·cm², 30.5 Ω·cm², 10.4 Ω·cm², 16.6 Ω·cm², respectively. In order to further evaluate the corrosion resistance of sheets treated by different processes, polarization resistance (R_pol) is calculated using the following expression for the equivalent circuit in Figure 6:

R_pol = R_ct + R_f

Figure 7 shows polarization resistance of as-cold rolled and heat-treated sheets in naturally aerated 0.5 M H₂SO₄ + 5 ppm F⁻ solution. According to the plot, sheets treated at 950 °C/1 h/WC possess the highest polarization resistance of 121.4 Ω·cm². The polarization resistance of as-cold rolled sheets is 77.5 Ω·cm² lower than 950 °C/1 h/WC treated ones. Additionally, polarization resistances of both sheets mentioned above are apparently higher than 950 °C/1 h/WC + 500 °C/3 h/AC and 950 °C/1 h/FC treated sheets, suggesting passive film formed on these sheets are more resistant to corrosion.

3.6. Potentiostatic Polarization Measurements

In order to simulate the loading process of PEMFC, potentiostatic polarization of as-cold rolled and heat-treated sheets was carried out. Attention should be paid that in this section, only heat-treated sheets were tested owing to poor ductility and inferior corrosion resistance of as-cold rolled sheets. The results of potentiostatic polarization are exhibited in Figure 8. Figure 8b is the magnification of the dashed rectangular region marked in Figure 8a. As suggested by a previous investigation, steady-state current density (I_ss) is the current density obtained after the specimen is potentiostatic polarized for a period of time. According to Figure 8a, current densities decrease abruptly at the beginning of the experiment, which is observed in the case of all tested sheets. This is due to the formation of the passive film [35,36]. After that, decreasing rates of current densities gradually slow down, and current densities eventually stabilize at around 1 μA·cm⁻², as is shown in Figure 8b. Steady-state current densities of 950 °C/1 h/WC, 950 °C/1 h/WC + 500 °C/3 h/AC as well as 950 °C/1 h/FC treated specimens are 0.68 μA·cm⁻², 0.86 μA·cm⁻², 1.03 μA·cm⁻², respectively.

On the basis of microstructure observation mentioned above, different I_ss obtained from specimens treated in various processes can be attributed to the amount of Ti₂Ni precipitation. After heat treatment at 950 °C and cooling at various rates, the amount of precipitation differs. The more Ti₂Ni intermetallic compounds precipitate, the more defects may be introduced into the passive film of titanium alloy sheets, which does harm the integrity of the passive film [21,37]. Consequently, I_ss of the sheets increased in the order of sheets treated at 950 °C/1 h/WC, 950 °C/1 h/WC + 500 °C/3 h/AC, and 950 °C/1 h/FC.

Figure 9 displays steady-state current densities of specimens polarized at 0.6 V and Ti₂Ni area fractions as functions of different heat treatment processes. It can be observed that the current densities of the specimens increase with the area fraction of Ti₂Ni precipitation. Based upon investigation by Osório et al. [21] on the electrochemical behavior of binary Ti-Cu alloys with different Cu contents, nobler Ti₂Cu intermetallic compounds serve as a cathode in the matrix. Moreover, more Ti₂Cu precipitate after increasing Cu contents, and passive current densities increase with Cu contents, which is a similar tendency to the present work.

4. Conclusions

Binary Ti-0.3Ni sheets were manufactured by cold rolling, and the effects of heat treatment on microstructures and properties were investigated. The following conclusions are reached:

(1)

Dispersed Ti₂Ni intermetallic compounds are observed in as-cold rolled sheets. In contrast, typical Widmanstatten microstructures can be seen in the case of heat-treated sheets. Amounts of Ti₂Ni precipitations increase in the order of as-cold rolled, 950 °C/1 h/WC, 950 °C/1 h/WC + 500 °C/3 h/AC, and 950 °C/1 h/FC treated sheets.

(2)

Mechanical properties of as-cold rolled and heat-treated specimens vary greatly. Elongations of as-cold rolled specimens increase by more than 110% while tensile strengths reduce by more than 47% after heat treatment, improving the formability of the sheets.

(3)

Specimens treated at 950 °C/1 h/WC possess the noblest corrosion potential. In contrast, the steady-state current density of sheets treated at 950 °C/1 h/FC is higher than other sheets. This can be attributed to the adverse effect of Ti₂Ni precipitations on the integrity of the passive film.