L. A. Pawlick and R. G. Kelly
Keywords:
Abstract:
Because you are not logged-in to the journal, it is now our policy to display a 'text-only' version of the paper. This version is obtained by extracting the text from the PDF or HTML file, and it is not guaranteed that the text will be a true image of the text of the paper. The text-only version is intended to act as a reference for search engines when they index the site, and it is not designed to be read by humans!
If you wish to view the human-readable version of the paper, then please Register (if you have not already done so) and Login. Registration is completely free.
JCSE Volume 1 Paper 4 Submitted 15 July 1995, revised version submitted 29 November 1995 An Electrochemical Investigation of the Corrosion Behavior of Aluminum Alloy AA5052 in Methanolic Solutions L. A. Pawlick, R. G. Kelly Center for Electrochemical Science and Engineering Department of Materials Science and Engineering University of Virginia Charlottesville, VA 22903 §1 ABSTRACT The electrochemical behavior of aluminum alloy AA5052 in methanolic solutions containing low concentrations of acid, chloride, sulfate and water has been studied. In all solutions investigated, the alloy exhibited spontaneous passivity. The addition of 1 mM acid increased the corrosion potential dramatically, whereas 1 mM chloride decreased the pitting potential and 1mM sulfate had no measurable effects. Water appears to decrease the pitting potential due to the enhancement of aluminum ion hydrolysis at incipient pits. No important second order interactions were observed. The effects of acid on the corrosion potential and rate are explained in terms of mixed potential theory. §2 INTRODUCTION As the use of alcohol in automotive fuels increases due to environmental concerns, so does the need to characterize and understand the corrosion of metals and alloys in alcoholic solutions. Of particular importance are the effects of low levels of the impurities commonly expected in these fuels such as acid, chloride, sulfate, and water. In recent work on iron in methanolic solutions, Brossia and Kelly first developed a quantitative characterization of the electrochemical phenomenology of iron in methanolic solutions including the effects of common impurities [1] which led to a description of the mechanisms underlying the accelerating effects of acid and the inhibitory effects of water on the open circuit corrosion rate [2]. One of the key elements of that work involved the use of statistical experimental design [3] to allow a quantitative analysis of the main (first-order) effects of the various solution additions, second-order interactions and an estimate of the experimental error. The error estimation permitted the statistical significance of each effect to be quantitatively assessed. §3 Brossia et al. [2] were able to explain the two primary empirical observations concerning the effect of acid on the corrosion of iron in methanol [4-6]: large increases in corrosion rate (by a factor of 15 due to a 1 mM acid addition [4]) and the corrosion potential (by 200 mV due to a 1 mM acid addition [5]). Brossia et al. [2] showed that the effects of acid were due to two separable phenomena; acid activates the originally passive iron surface, and proton reduction has substantially faster kinetics than oxygen reduction in acidified methanol. This latter effect dominates, leading to an increase in the corrosion potential with a subsequent increase in corrosion rate. §4 The addition of water inhibits the corrosion of iron in acidic methanol dramatically [1-6]. Brossia et al. [2] determined that proton reduction was under substantial mass transport control at the corrosion potential of iron. They then showed that the addition of water to acidified solutions reduced the corrosion rate predominantly by inhibiting the mobility of the proton, thus reducing the proton diffusivity and hence the diffusion limited current density of the dominant cathodic reaction in acidified methanol. This reduction in diffusion limited current density led directly to the decrease in the corrosion rate. The reduction in proton mobility is due to the preferential solvation of protons by water relative to methanol [7]. At low concentrations of water, this limits the proton hopping that is important for conductivity. §5 In an extension of that work, the electrochemical behavior of aluminum alloy AA5052 has been studied and analyzed. By studying a valve metal such as aluminum which passivates with a thick film, information on the effect of the nature of the passive film on the corrosion of materials in methanol can be gained. In addition, AA5052 is being considered as a construction material for future automobile fuel tanks. Thus, an assessment of the effects of impurities on its corrosion behavior would have practical importance as well. §6 Limited previous work on the corrosion of aluminum alloys in methanolic solutions has been performed. Mansfeld [8] and Palit et al. [9] focussed on the corrosion of aluminum alloys in strongly acidified methanolic solutions. Upon anodic polarization of AA6061 above 0 V(SCE) in 0.1 N sulfuric acid, Mansfeld observed pitting with gas evolution from the pits. He proposed that the pitting was due to the presence of the sulfate ion. Palit et al. [9] also observed pitting of pure Al in acidic methanol in solutions open to air. Hronsky [10] measured weight loss for pure Al in 0.3 M HCl. Severe pitting was observed under open circuit conditions. Wing et al. [11] and Lash [12] also used coupon testing to investigate the behavior of various cast aluminum alloys in M85 (15% fuel, 85% methanol with 1.1mM acid, 0.1% water, and 0.07mM chloride). They found substantial weight losses and pitting of the cast Al alloys. §7 In order to better understand the corrosion of aluminum alloys in methanol, the present study was conducted. The corrosion and electrochemical behavior of AA5052 in a variety of methanolic solutions was studied and the effects of the addition of several impurities were quantified and explained in terms of the previous work on iron [2]. Previous observations in this area are also rationalized within the framework of mixed potential theory. §8 EXPERIMENTAL Materials - Aluminum alloy AA5052 was supplied by Kobe Steel in the form of 0.50 mm thick sheet. Its composition was (in wt.%) 2.46 Mn, 0.22 Fe, 0.19 Cr, 0.1 Si, 0.02 Ti, remainder Al. Disk electrodes were cut from the sheet and wet polished with successively finer silicon carbide paper to a 600 grit finish. §9 Solutions - All test solutions were based on spectrophotometry grade, "Photrex" reagent methanol (J.T. Baker). All solutions contained 0.1 M anhydrous sodium perchlorate (Aesar) as a supporting electrolyte. Other solution additions included water, chloride, sulfate, and/or acid. The chloride was added as anhydrous lithium chloride (Fisher), the sulfate as anhydrous sodium sulfate (Aldrich), and acid as sulfuric acid. The water content of the solutions was measured via Karl Fischer titration (Mettler DL-18). A full factorial design was used to investigate the effects of the various impurities. Two levels of concentration were used, zero and 1 mM, except for water. The inherent water level of the methanol and that added from perchlorate led to a minimum water content for solutions of <0.06 wt.%. The effects of water were studied by the addition of water to a concentration of wt.%. No efforts were made to remove dissolved molecular oxygen as this would tend to evaporate large quantities of methanol, changing the concentrations of the dissolved substances. Previous work has identified the relevant cathodic reactions at occurring in the different solutions. §10 Testing Procedures - All experiments were conducted at room temperature. Electrochemical measurements were conducted with an E.G.&G. Princeton Applied Research (PAR) Versastat(TM) controlled by the PAR Model 352 software. A silver/silver chloride (SSC) electrode immersed in a compartment containing methanol with 0.1 M LiCl and 1.5 mass% water was used as the reference electrode. The reference electrode was separated from the working electrode solution by a Vycor(TM) frit. The SSC electrode has a potential of -29 mV vs. an aqueous saturated calomel electrode. The SSC was found to be very stable with respect to time and minimized the liquid junction potential [13]. Cyclic polarization scans were conducted at a scan rate of 0.5 mV/s starting at an initial potential of -0.8 V(SSC). A vertex current density of 5 mA/cm2 was used. Automatic current interruption was used for on-line correction for ohmic drop. §11 Analysis Procedures - Each curve was analyzed for corrosion potential, corrosion current density and pitting potential. The corrosion current density was determined by use of PARcalc(TM) and checked manually by Tafel extrapolation. To quantitatively determine the significance of the effects of added impurities and the experimental error, the results of the full factorial design were analyzed with Number Cruncher Statistical Software (NCSS) ver.5.03 (licensed by J. L. Hintze). Performing duplicate experiments for each condition and comparing the results of all experiments containing the solution species of interest to all experiments that did not contain that species allowed for the separation of effects due to single impurities as well as any synergistic effects between species. Upon completion of the tests, each effect was statistically analyzed to determine its significance as well as to determine the experimental error. §12 RESULTS Figure1: Cyclic polarization curve for AA5052 in the base solution (anhydrous methanol containing 0.1 M sodium perchlorate). Figure 2: Photomicrograph of AA5052 surface after polarization scan shown in Figure 1. Note the corrosion pits in an otherwise as-polished surface. In the base solution, AA5052 did not exhibit an active-passive transition as shown in Figure 1. This lack of an active-passive transition near the corrosion potential can be termed spontaneous passivity, and this term is used throughout this manuscript to refer to such behavior. This passive region was limited at +0.35 V(SSC) by pitting as shown in the photomicrograph of Figure 2. Figure 3: Polarization behavior for AA5052 in (a) the base solution and upon the addition of individual impurities: (b) 1 mM acid or (c) 1 mM sulfate or (d) 1 mM chloride or (e) 0.5wt.% water. Note the lack of an effect of sulfate, while acid, water and chloride all lead to a decrease in the pitting potential. The effects on the polarization behavior of the addition of individual impurities (1 mM acid or 1 mM sulfate or 1 mM chloride or 0.5 wt.% water) are shown in Figure 3. Acid increased the open circuit potential and decreased the pitting potential of the AA5052, and increased the open circuit corrosion rate slightly. The addition of water led to a decrease in the pitting potential, while sulfate had little effect. Chloride decreased the pitting potential. §13 A summary of the first order effects of the impurities on the electrochemical parameters that characterize the corrosion process in both low and high water contents is shown in Figures 4 and 5. Figure 4: Summary of the effects of the impurities (a) alone and (b) in combination on the corrosion potential and open circuit corrosion rate for AA5052. All solutions contained 0.1 M sodium perchlorate and <0.06 mass% water. Figure 5: Summary of the effects of the impurities (a) alone and (b) in combination on the corrosion potential and open circuit corrosion rate for AA5052. All solutions contained 0.1 M sodium perchlorate and 0.5 mass% water.