Corrosion of Mechanically Alloyed Nanostructured FeAl Intermetallic Powders

1 Facultad de Cenćıas Quı́micas e Ingenieŕıa, P.E. I.M. Universidad Autónoma del Estado de Morelos, Avenida Universidad 1001, Col. Chamilpa, 62210 Cuernavaca, MOR, Mexico 2 Instituto de Investigaciones Metalurgicas, UMSNH, Edificio U., Ciudad Universitaria, 58000 Morelia, MICH, Mexico 3 Centro de Investigacion en Ingenieŕıa y Ciencias Aplicadas, UAEM, Avenida Universidad 1001, Col. Chamilpa, 62210 Cuernavaca, MOR, Mexico 4 Facultad de Quı́mica, Instituto de Ingenieria, UNAM, Circuito Exterior S/N, Cd. Universitaria, 04510, Mexico


Introduction
In the past, intermetallic materials based on aluminum with transition metals such as Ni, Ti, Co, and Fe have been extensively investigated [1][2][3][4][5][6][7][8][9].Particularly, the Fe 40 Al 60 system has been very attractive due to several characteristics such as low cost, good mechanical performance at elevated temperatures, low density, and excellent oxidation-corrosion resistance.The Fe 40 Al 60 develops an Al 2 O 3 protective film, which provides good properties of corrosion resistance at high temperature [10,11].It has also been reported that Fe 40 Al 60 nanostructured material improves in general their physical properties [12].In this sense, the mechanical alloying is an effective technique to achieve very fine crystal sizes (nanometer), thus getting better material properties [13].Concerning Fe 40 Al 60 intermetallic with nanosized particles (100 nm) obtained by mechanical alloying, there is no information related to their corrosion behavior in aqueous solution at long exposure times.This is due to the absence of a technique or equipment specifically designed for these purposes.However, in the present investigation an innovative electrochemical cell arrangement is employed to evaluate the Fe 40 Al 60 (% at) intermetallic corrosion behavior by means of EIS and LPR conventional electrochemical techniques.Furthermore, the aim of the present work is the application of such electrochemical techniques in Fe 40 Al 60 intermetallic nanocrystals to monitor the aqueous corrosion performance for long periods of time.

Experimental
Fe 40 Al 60 (% at.) nanosized intermetallic powders were produced by mechanical alloying according to the procedure reported by Rosas et al. [14].Figure 1 shows schematically the electrochemical cell experimental setup used for the EIS and LPR tests.The cell arrangement consists of a platinum Advances in Materials Science and Engineering wire as a working electrode with an exposed area of 0.2 cm 2 , a saturated calomel electrode (SCE) as reference electrode and a graphite auxiliary electrode.The electrodes were connected to a Gill AC (ACM instruments) potentiostat controlled by a personal desktop computer.Test solutions were prepared with sodium hydroxide and sulfuric acid analytical grade reagents and distilled water.The pH of the test solutions was adjusted prior to each test to pH values of 3, 7, and 11.The pH measurements were made with an Oakton digital pH meter, previously calibrated.In all tests 0.125 g of Fe 40 Al 60 intermetallic were used in 30 mL of aqueous test solutions.During every test the solution was agitated in turbulent regime.The electrochemical tests were carried out at 27 • C and 70 • C during 2 and 5 days, respectively.The Fe 40 Al 60 powders were characterized by scanning electron microscopy (SEM) and transmission electron microscopy (TEM), before and after the electrochemical tests.
For the EIS tests the AC Gill potentiostat was calibrated according to the ASTM G106-89 Standard.The EIS test conditions were AC current signal with amplitude of 32 mV at open circuit potential (OCP) in a frequency range of 30,000 to 0.01 Hz.The EIS data was recorded every 2 hours; |Z| impedance modulus was employed by monitoring corrosion behavior.
The LPR tests were conducted following the recommendations of the ASTM G 59-91 Standard, between −100 and +100 mV every 8 and 16 hours; in addition polarization curves were done at 60 mV/min sweep rate.

Results and Discussions
3.1.Microstructural Characterization. Figure 2(a) shows a Z-contrast TEM micrograph of the Fe 40 Al 60 intermetallic processed by mechanical alloying.In this micrograph, intermetallic particles with size range of 100 and 200 nm can be observed.Figure 2(b) shows a high-resolution TEM image of individual intermetallic particle with an interplanar spacing of 2.06 Å, consistent with the (110) planes of a BCC structure.Figure 3 shows the EDS pattern of the particles before the corrosion tests illustrating the presence of Fe and Al, as well as small amounts of oxygen.Cu element in the EDS spectrogram belongs to the grid material.These results confirm the presence of the phase of interest for their corrosion evaluation.

Electrochemical Tests.
(2) Both equations show the R s and R p dependence of |Z| when f → ∞ and f → 0, respectively, where f is the alternating current (AC) frequency.These relationships clearly explain the experimental results.In the present case the R p values tend to decrease and the R s would have a major influence on the Fe 40 Al 60 |Z| values due to the constant variation in the resistivity of the solution, as showed in Figure 4(a).The electrochemical explanation to this behavior is that aluminum oxidation reaction ((3), anodic reaction) transfers Al +++ ions into the test solution by a diffusion mechanism, continuously modifying their ionic conductivity properties: It has been previously reported that the ionic species diffusion into the corrosion solution occurs in a chemical equilibrium regime that depends on the corrosion solution thermal energy [15].Figure 4(e) shows EIS Nyquist representation of the Fe 40 Al 60 corrosion behavior at room temperature and 70 • C, respectively.The corrosion mechanisms presented at both temperatures were principally controlled by diffusion mass transport in the intermetallic surface particle.As mentioned the intermetallic corrosion particles were controlled by diffusion mechanisms.This is clearly evidenced by tendency of the Nyquist curves to form a straight line approximately 45 • with respect the Z axis.This mechanism was presented during the 48-hour testing at both temperatures.
Figure 4(f) shows the Fe 40 Al 60 polarization curves in test solution with pH3 at room temperature, 70 • C, and at low potential scanning ranges.The Fe 40 Al 60 corrosion potentials (E corr ) were extracted from these curves being −180 mV SCE (61 mV SHE ) and −261 mV SCE (−20 mV SHE ), respectively.Both correspond to Al E corr in the intermetallic particles, because Al is more active and prone to corrosion than Fe.This was confirmed by the EDS results obtained after the electrochemical tests (Figure 7).Transferring the Al E corr values to their corresponding Pourbaix diagrams [16] at pH3, a continuous process without any passive behavior was observed.Solutions with pH ≤ 3 induced water dissociation with the consequent evolution of hydrogen dissolved as trivalent ionic aluminum (Al +++ ).Under certain special conditions not well established yet, the dissolution also occurs with the formation of (Al + ) monovalent ions [16].
From R p values and the Stern Geary equation, the corrosion current density (i corr ) data was estimated.We observed an increase in the (i corr ) values during the first 18 hours of test from 0.0014 to 0.0027 A/cm 2 .Near the end of the test the i corr values were stabilized around ≈0.0028 A/cm 2 .The stabilization of the i corr maybe indicates that the corrosion diffusion mechanism has been diminished in the interface particles solution.This can be attributed to the monovalent reaction previously mentioned.One of the principal factors that participate in the evolution of this reaction is the ionic exchange at chemical equilibrium in the metal-solution interface: Evidence of this mechanism at 70 • C can be seen from the Nyquist plot in Figure 4(e).This corroborates the previous results by comparison with the FeAl Nyquist plot at room temperature.The room temperature |Z| limit at low frequencies is greater than 70 • C.This may be due to the increase in temperature originating a catalytic effect on the corroding system increasing the corrosion rate.These results were corroborated by the polarization curves obtained at room temperature and 70 • C (Figure 4(f)).As the R p and |Z| decrease at 70 • C, the E corr was more negative indicating more susceptibility for corrosion at this temperature than at room temperature.Moreover, the i corr and |Z| tend to stabilize after 24-hour testing at ≈0.027 A/cm 2 and 300 to 400 Ω • cm 2 , respectively, which eventually occur through the mechanism proposed in (4).
It was observed in EIS tests with pH7 during the first 24 hours that the |Z| values range from 1023 to 1880 Ω • cm 2 for room temperature and 2200 to 11000 Ω • cm 2 for 70 • C (Figures 5(a) and 5(c)).After this period of time and toward the end of the test, there is a |Z| values tendency to increase steadily from 1092 to 1430 Ω • cm 2 (room temperature) and from 2200 to 9000 Ω•cm 2 (70 • C).At this stage, the corrosion mechanism is controlled by a diffusion process as evidenced by their corresponding Nyquist plots on Figure 5(e).
The E corr = −108 mV SCE (133 mV SHE ) obtained at room temperature and estimation from the polarization curves are showed in Figure 5(f).Introducing this value in their Pourbaix diagram with a pH 7 for pure water, can be observed that it is forms a layer known as bayerite.
Al 2 O 3 • 3H 2 O.After corrosion test, evidence for the formation of bayerite phase on the surface of Fe 40 Al 60 particles was previously presented in HRTEM images and from EDS analysis (Figures 2 and 3).However, this layer is not protective or contributes to decreasing the corrosion rate at the particle surfaces due to its high solubility in pure water.
Furthermore, the bayerite dissolution could reduce the pH solution [16].This fact was confirmed by the results of pH measurements obtained directly from the test solution at 70 • C, as shown in Figure 5(c).After 48 hrs of testing, the pH-solution measurement decreased from 7 to 6.68.This indicates the presence of bayerite at surface particles even at this temperature.Thus, the randomness in the |Z| values at early stages of the corrosion test would depend on the variation in R p values originated by the formation and dissolution of this oxide nonprotective layer.
Subsequently, the impedance modulus values for both temperatures tend to increase constantly until the end of the tests corrosion.Thus, (3) as the test results of LPR (Figures 5(b) and 5(d)) indicates that this behavior in the impedance modulus is a function of the R p which steadily increases, and consequently the i corr decreases.These results could indicate the transformation of the bayerite to the formation of a film of Al(OH) 3 on the surface particles.This takes place when Al reacts with the hydroxyl OH − ions formed for the potential levels (E corr ) obtained in both temperature conditions, according to their potential-pH equilibrium water diagram.Such formations of aluminum oxide are known as aging [16] and are a function of exposure corrosion time and pH.
Eventually, the OH − formation increases with the increment in temperature, which induces the pH increment to 9.7.The pH was measured directly from the corrosion test solution after 5-day testing (Figure 5(d)).In this way, the increment of OH − ions activates the Al(OH) 3 formation on the surface particles increasing R p values.It is not clear if the film has passive properties at the particle surfaces, but the results indicate that at least reduce their corrosion rates considerably at both temperatures, showing i corr values of 0.0016 (room temperature) and 0.0006 A/cm 2 (70 • C).
For a pH 11 the scenario was quite different.Figure 6(a) shows the Nyquist plots obtained at room temperature and 70 • C. The Nyquist plots show a partial semicircle that does not reach the limit at low frequencies.This behavior is characteristic when the relationship resistance-capacitance (RC) became large.The mechanism associated with control of this corrosion performance is mainly by charge transfer and can occur in passive metals or in metal which forms a porous surface conductive film with high capacitance [13].This could indicate that Fe 40 Al 60 particles form at their surface a Al(OH) 3 layer due to Al that tends to form under alkaline solutions this type of unstable precipitate [16].This was confirmed by the TEM images shown in Figure 6(b), in which two different phases were presented.In these TEM images, the more intensive zones indicate a phase with more atomic weight (metallic phase); meanwhile the zones with minor contrast show the corrosion products generated.Figure 6(c) is a high-resolution TEM image of the isolated corroded phase.The interatomic distances from the image indicate a value 2.55 Å which corresponds to the plane (104) of this hydrated phase.A postcorrosion EDS analysis showed in Figure 7 revealed that the regions of lower intensity correspond to the minor contrast zones illustrated in Figure 6(c), which reveals O and Al in a high percentage.This indicates corrosion products at the particle surfaces, formed after the corrosion electrochemical tests.
For 48 hours of corrosion test performed at room temperature, the impedance modulus (Figure 8(a)) presents random values with oscillating resistance between 1300 and 2300 Ω • cm 2 .However, their corresponding R p values (Figure 8(b)) have a tendency to increase from 3500 to 8500 Ω • cm 2 .This means that due to charge transfer of the formed corrosion layer, the variation of conductivity in R s was the determining factor of |Z| data behavior according to the relationship showed by (3).
After the solution was treated for 24 hours at 70 • C, the |Z| values show a tendency to increase gradually (Figure 8(c)).As previously mentioned, this is related to the catalytic effect of the temperature.In this case, there is accelerating and homogenizing of the electronic exchange between charges produced during the transfer in the metalsolution interface.As consequence of this an increase of the corrosion rates of the particles was obtained.This was confirmed by analyzing the R p data, which are decreasing in their maximum R p values (∼6000 Ω • cm 2 ) at 70 • C (Figure 8(d)) in comparison to the R p values exhibit at room temperature, as well as, the decrement in the Z and Z values in Nyquist plot, as shown in Figure 6(a).
Finally, the Fe 40 Al 60 polarization curves for the pH11 solution at room temperature and 70 • C are observed in Figure 8(e).The E corr values were −40 and 48 mV SCE at both temperature conditions.These values were higher than those obtained at the pH3 and pH7 solutions.This means large times required to reach equilibrium between cathodic and anodic reactions.Accordingly, this might influence the large   values obtained from the I corr , reaching maximum values of 0.006 mA/cm 2 .

Summary
The corrosion behavior of the Fe 40 Al 60 nanostructured intermetallic particles was strongly affected by the temperature and different types of oxides formed at the particles surfaces.
The oxide type formed can vary according to its solution solubility that depends on whether the solutions were acid/basic of pure water.In any case, the particles corrosion does not show localized corrosion, a passive behavior, or pitting, presenting principally uniform corrosion along the particles surfaces areas.The corrosion potentials obtained from Fe 40 Al 60 nanostructured do not correspond to any bulk pure Al or Al-alloys mostly reported at −1700 mV SCE .This suggests that the Fe 40 Al 60 nanoparticles are responsible for their new corrosion potentials values reported in this work.This may be due to a high number of superficial atoms developed by the nanostructures formed by mechanical alloying, compared with conventional macro-Al-alloys.The nanocrystals increase the time required to reach electrochemical equilibrium between anodic and cathodic partial reactions, leading to more positive corrosion potentials values.However, further corrosion studies with different size of Fe 40 Al 60 nanoparticles will be carried out to corroborate these results.(2) The nanoparticle corrosion behavior depends on testing temperature, time, and their corrosion products formation principally in the form of aluminum oxides.The oxide solubility in the different testing solutions proves to be also an important factor to determine their corrosion behavior.

Figure 4 (
a) shows the impedance modulus |Z| for 48-hour monitoring of the Fe 40 Al 60 intermetallic corrosion performance in a solution with pH = 3 at room temperature.These results show a rather irregular behavior with values oscillating between 2120 and 1600 Ω • cm 2 .However, the LPR tests (Figure 4(b)) show the start values of R p = 2 kΩ • cm 2 which decrease steadily during the elapsed time of the test.After 18 hours of LPR test the R p values reach a stable behavior with fluctuating values around 9000 Ω • cm 2 toward the end of the test.The previously behavior can be explained based on the relationships between the solution resistance (R s ), polarization resistance (R p ), and the impedance modulus |Z| as described by the following:

Figure 4 : 6 AdvancesFigure 5 :Figure 6 :
Figure 4: Electrochemical tests plots in the pH 3 solution, (a) |Z| at room temperature for 48 hours, (b) R p at room temperature for 48 hours, (c) |Z| at 70 • C for 48 hours, (d) R p at 70 • C for 48 hours, (e) EIS Nyquist plot at room temperature and 70 • C, and (f) polarization curves at room temperature and 70 • C.

( 1 )
The possibility of the Fe 40 Al 60 nanoparticles corrosion monitoring in suspension was demonstrated by conventional bulk electrochemical techniques and setups such as those employed by EIS, LPR, and electrochemical polarization techniques, for long periods of time.
•C indicate |Z| values that decrease significantly in comparison of the tests performed at room temperature.In these tests the maximum and minimum |Z| values observed were 900 Ω • cm 2 and 300 Ω • cm 2 (Figure4(c)).The same behavior was noted for the R p , where the maximum and minimum values were