Elsevier

Electrochimica Acta

Volume 52, Issue 1, 5 October 2006, Pages 86-93
Electrochimica Acta

Understanding aluminum behaviour in aqueous alkaline solution using coupled techniques: Part I. Rotating ring-disk study

https://doi.org/10.1016/j.electacta.2006.03.076Get rights and content

Abstract

Rotating disk electrode (RDE) and rotating ring-disk electrode (RRDE) measurements have been undertaken to study the behaviour of pure aluminum electrodes in alkaline media. The measurements did consist of linear sweep voltammetry from anodic to cathodic potentials on 4N, 5N or 5N5-aluminum samples in 4 M aqueous potassium hydroxide solution. In the potential range studied (−0.7 V versus NHE to −2.5 V versus NHE) the aluminum undergoes oxidation/dissolution into aluminates anions at high electrode potential while it yields strong hydrogen evolution at low potentials. Thanks to the RRDE technique, we show that hydrogen starts to evolve from the aluminum electrode even above the open circuit potential. Also, the oxidation state of superficial aluminum varies according to the electrode potential: whereas non-conducting aluminum oxides are present above the open-circuit potential hindering hydrogen evolution reaction (HER), they tend to disappear below the ocp, due to the strong hydrogen evolution, following the probable porous oxide layer blow up induced by the hydrogen bubbles formation. In consequence at very low potential, HER occurs on bare aluminum, HER kinetics being much faster than on oxide-covered aluminum.

Introduction

Aluminum, the third most abundant elements in the earth crust, is a very attractive material for energy storage and conversion. Its low atomic weight and trivalence result in high capacity per unit weight (2980 Ah kg−1) and per unit volume (804 Ah L−1) [1]. Moreover, the very negative value of aluminum standard electrode potential (E° = −1.66 V/NHE) makes it a very attractive anode material for high energy density power sources [2].

However, in most aqueous solutions, aluminum is covered by a thick oxide or hydroxide superficial film, which is not electron-conducting and results in high anodic dissolution overvoltage [3], [4], [5], [6]. In consequence, aluminum anodic dissolution is only efficient (and interesting for energy production) in media allowing the dissolution of its oxide film, e.g. concentrated alkaline solution. Unfortunately, even in such media, great voltage losses are measured on aluminum electrodes, mainly resulting from the competition between the numerous processes that take place on aluminum surface in concentrated alkaline solutions: formation and/or dissolution of a corrosion layer, three-electron charge transfer yielding AlIII species, formation of corrosion products (such as Al(OH)4 or Al(OH)3) and evolution of hydrogen either concomitant with aluminum corrosion or directly produced at low potentials [7], [8], [9], [10]. One of the consequences of the existence of so many phenomena, sometimes occurring simultaneously at a given potential, is that the comprehension of aluminum behaviour in alkaline solution remains tricky.

Now, the high corrosion rate of aluminum in concentrated alkaline media coupled with the concomitant hydrogen evolution on its surface renders any study of its behaviour in concentrated alkaline solution difficult with classical techniques. Nevertheless, basic electrochemistry experiments have been used for this purpose so far. Among them, one of the simplest is to plot polarization curves using aluminum electrodes in different cell configurations: batch cell, rotating disk electrode or rotating cylindrical electrode, or even channel flow cell [7], [8]. However, determining the potentials where the different processes take place solely from polarization curves is nearly impossible: indeed, the currents measured are the sum of various algebraic currents, all resulting from contributions of different electrochemical or chemical reactions that can occur simultaneously and/or proceed in an electroless-like process. Another technique consisted of electrochemical impedance spectroscopy [9], [10], but achieving steady-state and linearity conditions in alkaline media on an aluminum electrode is nearly hopeless, following aluminum strong corrosion with the hydroxide anions from the electrolyte. As a result, none of the pre-cited studies [3], [4], [5], [6], [7], [8], [9], [10] provide a complete knowledge of the various mechanisms occurring at aluminum in strong alkaline media.

Given the classical electrochemical techniques insufficiencies and considering the need for a technique yielding more information about the phenomena occurring at an aluminum electrode in alkaline medium, we have considered rotating ring-disk electrode measurements (RRDE). In the second part on the present paper (Understanding aluminum behaviour in aqueous alkaline solution using coupled techniques—Part II. Acoustic emission study), we will consider the coupling of the acoustic emission technique with electrochemical measurements and direct hydrogen gas volumetry. Both these techniques enable aluminum behaviour in situ monitoring while the electrode is immersed in concentrated potassium hydroxide solution, which should provide interesting information about the mechanism occurring at aluminum in strong alkaline media.

In the first part of the present work, we focussed on the in situ measurement of the amount of hydrogen gas evolved from an aluminum electrode during linear sweep voltammetry from anodic to cathodic potentials. In order to simplify the analysis of the experimental results, the electrolyte was a 4 M aqueous potassium hydroxide solution, while we did choose to consider solely the behaviour of pure (non-alloyed) aluminum electrodes. Thus, our electrolyte solutions did not contain aluminum corrosion inhibitors, such as stannates, which would be very useful for the effective utilisation of aluminum as the anode of an aluminum-air battery [3], [4], [5], [6], [11], [12], [13], [14]. Also, they contained very little amount of corrosion products: all solutions were freshly prepared before each experiment and the cell volume was large enough so that we could consider the electrolyte bulk composition did not vary significantly in the time of each experiment.

Section snippets

Experimental

The aluminum used was generally 99.99 (4N), 99.999 (5N) or 99.9995% (5N5) pure. However, we did study several samples with other aluminum purities: 99.5% (1050), and 99.998 (4N8). Except for the 1050 grade, all the pure aluminum samples did exhibit rather identical behaviour in 1 M alkaline solution (Fig. 1). In the course of the present paper, only results for 4N, 5N or 5N5 will be considered, with special emphasis for the 5N grade. In addition, we will use the term pure aluminum in the case of

Rotating disk experiments

Following the very good experimental reproducibility for the polarisation curves plotted on pure aluminum electrodes in the 4N, 5N or 5N5 grades, the classical voltammogram relative to a 4 M potassium hydroxide solution displayed on Fig. 3a for a 5N aluminum electrode is representative of all the current/potential curves plotted on pure aluminum in the present study. Only the value of the open-circuit potential (ocp) slightly decreases in the sequence 4N < 4N8 < 5N < 5N5, all other features of the

Rotating disk measurements and electrode micrographs

Fig. 1, Fig. 3 show that understanding the complex phenomena occurring on the aluminum disk electrode is totally hopeless solely from the voltammograms relative to the disk: for example around ocp conditions, aluminum dissolution, aluminum oxide formation and hydrogen evolution reaction (HER) occur simultaneously, and it is possible by no mean to evaluate the magnitude of each phenomena from the current density/potential curve. However, the photographs from Fig. 1b reveal that the aluminum

Conclusions

In the first part of this study, we used for the first time the RRDE technique to evaluate the mechanisms occurring at an aluminum electrode in 4 M potassium hydroxide solution. Thanks to this technique, we have clarified some of the features observed (but unexplained) on the aluminum potential/current density response. Above aluminum equilibrium potential (ocp): −1.9 V/NHE for 5N or 5N5 aluminum samples, the main process occurring at the electrode is aluminum dissolution. However, hydrogen

Acknowledgements

We thank SAGEM-SA for its financial and technical support, Delphine Tigréat, Jean-Luc Bergamasco and Xavier Boutin for their interest in this fundamental study and Michel Vaujany without the technical skills of whom the RRDE study would not have been possible.

References (24)

  • Q. Li et al.

    J. Power Sources

    (2002)
  • D. Chu et al.

    Electrochim. Acta

    (1991)
  • K.C. Emregül et al.

    Corros. Sci.

    (2000)
  • H.B. Shao et al.

    J. Electroanal. Chem.

    (2003)
  • M. Pourbaix, Atlas d’Equilibres Electrochimiques, Gauthier-Villars, Paris,...
  • D.S. Keir et al.

    J. Electrochem. Soc.

    (1967)
  • D.S. Keir et al.

    J. Electrochem. Soc.

    (1969)
  • A.R. Despic et al.

    J. Appl. Electrochem.

    (1976)
  • J. Albert et al.

    J. Appl. Electrochem.

    (1989)
  • S. Real et al.

    J. Electrochem. Soc.

    (1988)
  • R.P. Hamlen et al.

    Handbook of Batteries

  • D.E. Sargent, US Patent 2,554,447...
  • Cited by (40)

    • Eco-friendly rosin-based 6-dehydroabietic acylamino sodium as corrosion inhibitor for AA2024-T3 in alkaline solution by experimental and theoretical studies

      2021, Journal of Molecular Liquids
      Citation Excerpt :

      AA2024-T3 is a well-known metal alloy in applications associated with the aerospace, marine, and chemical industries owing to its high strength and low cost [1–8]. Under a wide variety of industrial process conditions, the natural Al2O3 film on the alloy surface has many defects or pores and can be easily destroyed by corrosive OH−, especially in alkaline aqueous solutions [9,10]. Currently, various corrosion inhibition measures exist, such as the anode oxidation method [11], surface coating method [12], addition of inhibitors [13], and plasma electrolytic oxidation [14,15].

    View all citing articles on Scopus
    View full text