ELECTROCHEMICAL CORROSION OF Al-Pd ALLOYS IN HCl AND NaOH SOLUTIONS

The corrosion performance of as-solidified Al–Pd alloys in HCl and NaOH aqueous solutions is investigated in this work. Four different alloys (Al88Pd12, Al77Pd23, Al72Pd28 and Al67Pd33, element concentrations are given in at.%) were prepared from high purity Al and Pd lumps by arc–melting in Ar. Subsequently, the alloy microstructure and phase occurrence were investigated by a combination of scanning electron microscopy and room-temperature powder X–ray diffraction. The assolidified Al–Pd alloys were found to consist of several single-phase microstructure constituents with various structures and chemical compositions, including structurally complex intermetallic phases. The polished surfaces of the Al–Pd alloys were subjected to electrochemical polarization in aqueous HCl and NaOH solutions (0.01 mol.dm-3) at 21±2°C. The corrosion experiments were conducted in a standard 3–electrode cell controlled by potentiostat. The corrosion potentials and corrosion current densities were determined by Tafel extrapolation of the experimental polarization curves. Phase dissolution has been observed on the alloy surfaces and some of the phases were preferentially corroded. The effects of the alloy microstructure and the phase occurrence are evaluated. The local nobility of individual intermetallic compounds is discussed. Finally, the conclusions for the alloys corrosion performance in acidic and basic solutions are provided.


Introduction
Binary Al-Pd alloys with compositions close to 27 at.%Pd belong to the complex metallic alloy (CMA) family.The CMAs are predominantly composed of structurally complex intermetallic phases of either crystalline or quasicrystalline nature [1].The former type of phases consists of giant unit cells comprising hundreds or even thousands of atoms.In the latter type of phases, atoms are arranged non-periodically.Because of their complex crystal structure, the structurally complex intermetallic phases exhibit markedly different physical properties from those observed in traditional alloys [1].Al-based CMAs have been utilized as scratch-resistant coatings [2].Furthermore, possible applications of these materials in catalysis [3], hydrogen generation [4] and metalmatrix composites [5,6] have been reported.
Studies dealing with chemical properties and corrosion behaviour of CMAs are limited.Only few investigations have been reported so far and these include corrosion studies of Al-Co, Al-Cu-Fe, Al-Cr-Fe and Al-Cu-Fe-Cr CMAs in various electrolytes [14][15][16][17][18][19][20][21][22][23].The electrochemical behaviour of rapidly solidified Al-Pd alloys has been studied in a NaCl aqueous solution only (1 mol.dm -3 [24]).A selective dissolution of Al (de-alloying) from intermetallic phases in these alloys has been found.Electrochemical de-alloying is a phenomenon primarily originating from corrosion [25].It has attracted much attention in recent years as it has been found to be efficient in creating nanoporous metal networks with a three-dimensional bi-continuous interpenetrating ligament-channel structure at the nanometer scale [26,27].Interactions between coexisting phases play an important role during dealloying of a double-phase alloy.These interactions are, in principle, controlled by the diffusivity of the noble element (Pd), the curvature-dependent undercritical potential dissolution, and chemical reaction between the noble element and the chlorine ion [24].
To the best of our knowledge, the corrosion behaviour of Al-Pd complex metallic alloys in HCl and NaOH solutions has not been investigated yet.In industrial applications, the materials are often exposed to harsh environments including saline solutions and/or acid rains.The corrosion resistance of Al and its alloys in alkaline as well as acidic and saline environments is often very low due to protective oxide scale dissolution [28,29].Therefore, the electrochemical corrosion behaviour of Al-Pd alloys in HCl and NaOH electrolytes (0.01 mol dm -3 ) has been studied in the present work through the potentiodynamic polarization.Surface morphology and phase compositional analyses were also conducted to provide the interpretation of the obtained polarization behaviour.After solidification, the samples were mounted in epoxy resin, ground with SiC papers of various granularities and polished with diamond suspension down to 1 µm surface roughness.The polished sample surfaces were subjected to an electrochemical testing in a standard three electrode cell at 21±2°C.The Ag/AgCl electrode suspended in a saturated KCl solution (saturated silver chloride electrode, SSCE) was used as the reference electrode.A platinum sheet was used as the counter electrode.The corrosion experiments were carried out in aqueous HCl and NaOH solutions (0.01 mol dm -3 ).The solutions were prepared by dissolving the weighted amount of HCl and NaOH respectively in de-ionized water.The electrochemical experiments were controlled by a PGU 10 V-1A-IMP-S potentiostat /galvanostat from Jaissle Electronic Ltd. (Waiblingen, Germany).The open circuit potential (OCP) of the alloys was measured during first 30 min of the sample immersion in the electrolyte.After stabilization, the electrochemical polarization experiments were carried out using a sweeping rate of 1 mV/s.The potential range for each measurement was chosen according to the OCP of the alloy and covered a minimum of 500 mV around the OCP on both anodic and cathodic sides.This corresponded to potentials between -1700 and -300 mV (SSCE) for NaOH and -1300 and +400 mV (SSCE) for HCl solutions.The resulting polarization curves were analyzed by the Tafel extrapolation.

Material and methods
The alloy microstructures before the electrochemical testing were studied by a scanning electron microscope (SEM) JEOL JSM-7600F, using the back-scattered electron imaging at 20 kV.For the energy-dispersive X-ray spectroscopy (EDX), an Oxford Instruments X-max 50 spectrometer working with INCA software was used.To obtain mean values of metal concentrations, at least ten measurements per constituent/phase were done.Similarly, average volume fractions were determined from a minimum of ten measurements.The microstructures of corroded samples were documented by a light microscope NEOPHOT 32.The powder X-ray diffraction (XRD) was carried out by using a Panalytical Empyrean PIXCel 3D diffractometer with Bragg-Brentano geometry, working with an iron filtered CoKα1,2 radiation.The X-ray scattering was performed for 2θ angle between 20° and 60°.The step size was 0.0131°, and the exposure time was 98 seconds per step.
Two sets of samples were investigated in this work, corresponding to as-solidified (i.e.original) and corroded (i.e. after the electrochemical testing) conditions.

Results and Discussion
All the as-solidified microstructures, given in Fig. 2, were observed to consist of two microstructure constituents.Each of the constituents was identified as a single-phase area.The alloys Al 88 Pd 12 and Al 77 Pd 23 were found to comprise (Al) and ε n (ε 6 +ε 28 ).In the alloys Al 72 Pd 28 and Al 67 Pd 33 , ε n (ε 6 +ε 28 ) and δ were identified respectively.To perform the phase-toconstituent assignment, metal compositions available in Table 1 were compared with both SEMmicrographs (Fig. 2) and XRD-patterns (Fig. 3).This procedure led to the unambiguous structure characterization of the as-solidified alloys.In Fig. 3, only two characteristic XRD patterns are documented, each for one pair of the as-solidified alloys, showing the same phase constitution (see above).As the backscattered electron imaging was used in the SEM observation, the constituents/phases exhibiting brighter appearance were revealed to contain more Pd (Fig. 2).Alloys with the same phase constitution were found to differ from one another in phase volume fractions (Table 1).On the other hand, each phase showed only negligible changes in metal composition across the set of investigated alloys (Table 1).The phase-to-constituent assignment is presented in Fig. 2 for each the as-solidified alloy.
The corrosion performance of the as-solidified Al-Pd alloys has been studied by potentiodynamic polarization in aqueous HCl and NaOH solutions at 21±2°C.At the beginning a steady-state open circuit potential (free corrosion potential) has been measured for 30 min of sample immersion in the electrolyte.The variation of the OCP with time is illustrated in Figs. 4 and 5.The measurement of Al is also included for comparison.The OCP values after 30 minutes of sample immersion increase in the following order in both electrolytes: The OCPs of the alloys increase with increasing Pd concentration in both solutions.This observation is in accordance with expectations since Pd has a higher standard electrochemical potential compared to Al [30].The OCPs of the alloys Al 72 Pd 28 and Al 67 Pd 33 were nearly identical and significantly higher compared to alloys Al 88 Pd 12 and Al 77 Pd 23 (Figs. 4 and  5).Since these two alloy groups are composed of different phases (Table 1), the OCPs are not only determined by the bulk chemical composition, but are also influenced by a combination of structural factors including phase occurrence.
The OCP of aluminium was stable at the beginning of measurement; however, it eventually began to decrease in HCl solution as a result of corrosion.The OCPs of the Al-Pd alloys, on the other hand, were found to be relatively stable over time.Table 1.Volume fractions and metal compositions of phases/constituents identified in as-solidified Al-Pd alloys before electrochemical testing.For phase-to-constituent assignment see Fig. 2.
studied in NaCl solution previously [24].These oscillations indicate that the Cl -1 present in the electrolyte probably interacts with the alloy phases and this interaction influences the open circuit potential.
The OCP values of the alloys measured in 0.01 M NaOH solution are considerably smaller compared to HCl (Tables 2 and 3).This observation shows a higher susceptibility towards corrosion in alkaline environments.This behaviour is in agreement with equilibrium E-pH diagram of Al.In this diagram, lower potentials of Al in equilibrium with ionic Al species in solution are found in alkaline environments [31].In 0.01 M NaOH, the OCPs of alloys Al 77 Pd 23 , Al 72 Pd 28 and Al 67 Pd 33 have been found to slightly increase with time (Fig. 5).For the Al 88 Pd 12 alloy and       a preferential corrosion attack of ε n has also been observed (Figs.9c and 9d).The corrosion attack has been much more pronounced compared to the 0.01 M HCl solution.The electrochemical activity of phases in 0.01 M NaOH thus decreases in the following order: Since Al alloys are prone to localized corrosion in Cl-containing electrolytes, the susceptibility of the Al-Pd alloys towards pitting in HCl has been further investigated by cyclic polarization.The cyclic polarization curves are presented in Fig. 10.For the alloys, a positive hysteresis loop has been found upon reverse polarization in 0.01 M HCl (Fig. 10).This observation is indicative of localized corrosion [32].The physical origins behind the positive hysteresis lie in the connection between the critical crevice solution (CCS), the stability of pits, and the actual competition between diffusion and dissolution at localized corrosion sites [32].During the initial portion of the reverse scan, the driving force for dissolution is decreasing, but CCS has not had sufficient time to diffuse away.As the reverse scan continues, the driving force for dissolution continues to decrease, and eventually the CCS cannot be maintained against diffusion.At this point, the localized corrosion sites re-passivate, and the anodic current drop is observed (Fig. 10).Throughout the scan, the areas on the surface that were not dissolving at high rate were having the passive films covering them thicken.Thus the re-passivation potential is a convolution of the loss of the CCS from the localized corrosion site and the raising of the corrosion potential of the surrounding passive surface (ennoblement) due to a decrease in anodic kinetics [32].
For the Al-Pd alloys in 0.01 M NaOH, a negative loop occurs, i.e. the current densities on the reverse scan are smaller than those on the forward scan at the same potential (Fig. 11).On the reverse scan, the currents are lower at all potentials because of the thickened passive film.In these cases the pitting corrosion can be ruled out.
Four alloys, Al 88 Pd 12 , Al 77 Pd 23 , Al 72 Pd 28 and Al 67 Pd 33 (element concentrations are given in at.%) were investigated in the present work.The alloys nominal compositions were chosen to stay close to the ε n (~Al 3 Pd) chemical composition.The experimental Al-Pd alloys were prepared by arc melting of high purity (99.95 wt.%) Al and Pd lumps under argon atmosphere.The melt homogeneity was improved by repeated re-melting.A rapid solidification of the alloys was done by quenching the melted drops on a water-cooled Cu mould.

Figure 3 .
Figure 3. XRD-patterns of alloys Al 88 Pd 12 (upper) and Al 67 Pd 33 (lower) linked with dotted lines corresponding to angular peak positions (2Theta).In the intermediate window, diffracting planes of particular phases are specified and assigned to peaks occurring in both XRDpatterns.Arrows with upwards/downwards oriented tips show the classification of corresponding diffracting plane to upper/lower patterns, respectively.

Figure 4 .
Figure 4. Open circuit potentials of as-solidified Al-Pd alloys and Al in 0.01 M HCl as a function of time.

Figure 5 .
Figure 5. Open circuit potentials of as-solidified Al-Pd alloys and Al in 0.01 M NaOH as a function of time.

Figure 6 .
Figure 6.Polarization curves of as-solidified Al-Pd alloys and Al in 0.01 M HCl.

Figure 7 .
Figure 7. Polarization curves of as-solidified Al-Pd alloys and Al in 0.01 M NaOH.
Small potential oscillations have been found for alloys Al 72 Pd 28 and Al 67 Pd 33 respectively.Similar oscillations have been observed for alloys Al 70 Pd 30 and Al 77 Pd 23

Table 3 .
Electrochemical parameters of as-solidified Al-Pd alloys and Al after corrosion in 0.01 M NaOH at 21±2°C.

Table 2 .
Electrochemical parameters of as-solidified Al-Pd alloys and Al after corrosion in 0.01 M HCl at 21±2°C.