Application of X-Ray Computed Tomography to Assess the Effect of Elevated Temperature on Atmospheric Corrosion of Aluminum Alloy 6063 in Contact with FeCl₃ Droplets

Aluminum is used extensively in urban and marine environments at elevated temperatures or over temperature gradients. The material is chosen for its high thermal conductivity, durability, and competitive price with the typical application as heatsink in radiators or air conditioning units.¹  These applications, for example, result in the aluminum being maintained at elevated temperatures for longer periods of time to distribute or dissipate heat. This usage can result in contact with higher-strength steel-based connectors and fasteners. Exposures over longer time periods generally result in the formation of both, aluminum and steel-based corrosion products, particularly in humid air or marine environments. In particular, the appearance of a red-to-brown steel corrosion product occurs where exposure to higher humidity results in the formation of iron oxides and oxyhydroxides.

Exposure to challenging marine environments can result in the corrosion of aluminum which is usually obscured by the presence of corrosion products on the aluminum surface. This creates a dynamic localized environment that can potentially result in the acceleration or deceleration of corrosion. Steels corrode to form ferrous ions which are then transformed to ferric ions by the oxygen present. This results in the formation of FeCl₂ and FeCl₃ when undergoing corrosion in the presence of oxygen and chlorides, with both ferrous and ferric chloride being highly soluble in lower pH aqueous environments. The corresponding oxides, oxyhydroxides, and hydroxides typically form more insoluble corrosion products, with the pH of the environment playing a crucial role. Aluminum exposed to droplets of FeCl₃ results in the formation of AlCl₃ and a variety of iron oxyhydroxides, providing an electrochemical driving force for the dissolution of aluminum.^2-3 

A recent study investigating the effect of FeCl₃ on aluminum corrosion clearly showed the formation of Fe-rich deposits within the corrosion product layers, often resulting in a coherent layer of iron oxyhydroxides and oxides directly in contact with the aluminum surface.⁴  The observed corrosion product cap was very porous and could not prevent further corrosion to occur. The porosity and volume of the corrosion products were affected by the hydration periods, with samples exposed to multiple droplets showing greater levels of corrosion than those exposed to single droplets.³ 

The role of temperature on corrosion varies on the material and the reactive agents present. This behavior is because the temperature has a variety of influences on various components such as the oxygen solubility, electrochemical half-potentials, and precipitation rates in a corrosion system. Higher exposure temperatures reduce the oxygen solubility in aqueous solutions slowing the rate of the cathodic reaction (if driven by oxygen reduction) and the formation of a protective surface film.⁴  This behavior can result in a maximum corrosion rate at a given temperature.⁵  The changes in the pH of water, for example, becoming more acidic at higher temperatures would play a role as well as the formation of inhibitive surface films reducing corrosion rates. When aluminum is exposed to pure water and the temperature is increased, the corrosion rate increases, with a maximum corrosion rate reached at around 80°C.³  Al-2A02 exposed to marine environments with NaCl results in a maximum corrosion rate at temperatures of around 60°C. This peak in corrosion rate is related to the solubility of oxygen and the formation of an Al(OOH) protective layer.⁶ 

Similar behavior is also observed with Mg-alloy AM50 showing a strong temperature dependence, caused by the inhibitive effect of CO₂ at low temperatures as its solubility increases with decreasing temperature.⁷  This inhibitive effect of CO₂ is also the case for aluminum treated with NaCl, which corrodes rapidly in CO₂ free air due to the enhanced anodic dissolution of aluminum, here caused by the formation of high pH areas due to the cathodic reduction of oxygen.⁸  The chemical and electrochemical reactions of high-purity aluminum alloys in chloride-containing environments typically become faster as the temperature increases.⁹ 

This paper sets out to investigate the exposure of aluminum alloy 6063-T5 to Fe-containing solutions at elevated temperatures up to 50°C, here simulated in our experiments by using FeCl₃ droplet exposures. The corrosion product formation is explored using a quasi in situ x-ray computed tomography (XCT) approach,¹⁰  allowing for a detailed analysis of the development of corrosion in a nondestructive manner.^10-11  A combination of Raman spectroscopy, scanning electron microscopy (SEM), and energy dispersive x-ray analysis (EDX) is used to analyze corrosion product formation, with a series of techniques combined to provide detailed information on the degradation mechanism.¹² 

Coupons of AA6063-T5 with dimensions of 20 mm × 20 mm × 8 mm (l × w × t) were prepared from a block of extruded aluminum alloy, with a chemical composition of (wt%) 0.56% Si, 0.03% Mn, 0.20% Fe, 0.02% Ti, 0.02% Zn, and 0.53% Mg and Al (bal.). The material had a hardness of 92 HV₁. All exposed surfaces were first ground to 4000 grit, followed by a diamond polish to 1 µm finish, and then electropolished in a 20:80 (vol%) perchloric acid-ethanol solution. The solution was kept in an ice bath to keep the temperature below 5°C. Electropolishing involved an application of 20 V for 1 min, followed by immersion in distilled water, and drying in hot air.

A single 35 µL droplet of aqueous FeCl₃ was applied to each electropolished sample surface, with the exposure regimes summarized in Table 1. The rectangular coupons were placed in airtight plastic containers, with one container of each set placed in an oven where the temperature was kept at 50°C. The other container was kept at room temperature. The humidity was maintained close to 75% relative humidity (RH) using NaCl saturated salt solutions and monitored inside the containers using EL-USB-2+ humidity and temperature loggers. The RH of saturated NaCl at room temperature varies between 75% and 75.6% with the RH at 50°C being slightly lower at 74.4%.¹¹  The high RH results in the presence of liquid FeCl₃ droplets at the aluminum surface, which over time then convert into solid corrosion products due to changes in droplet chemistry.

The equilibrium concentration of FeCl₃ at 75% RH is approximately 2.5 mol/kg.¹³  The different droplet concentrations result in changes in the overall droplet volume, with droplet concentrations below the equilibrium concentration reducing their volume via water evaporation, and concentrations above the equilibrium absorbing water from the surrounding environment, thereby growing in volume. It must also be noted that the maximum concentration of moisture that the air can support varies significantly between 20°C and 50°C. At 20°C the maximum water content in the air is only 17.3 ×10⁻³ kg/m³, whereas at 50°C the maximum water content reaches up to 83 × 10⁻³ kg/m³.¹⁴  As such, the droplet volume would change from the initial 35 µL to its equilibrium volume with respect to the set RH and exposure temperature. If the droplet had a concentration of 5 M it would equilibrate to 70 µL or if the initial concentration was 0.5 M it would equilibrate to 7 µL in volume.

The first set of samples (1a, 2a) were removed after 14 d of controlled humidity exposure, and then XCT scans were performed. During the XCT scans, all samples were kept under normal room temperature conditions without humidity control. Directly after completing the first scan, a second droplet was applied to the same sites and the samples immediately placed in their airtight containers back in their respective environmental conditions. One container was again placed in an oven at 50°C and the second container was kept at room temperature for another 14 d. All samples were scanned in sequence so that the only difference between them was the droplet FeCl₃ concentration and exposure temperature.

After terminating exposure experiments, Raman spectroscopy of samples was performed using an inVia^† micro-Raman microscope. A 633 nm laser was used, with an 1,800 L/mm grating, exposure time of 10 s, an objective lens of 20× and a laser power of 5%. Raman was used to identify the iron-containing compounds formed on the aluminum surface to provide an understanding of the corrosion reactions that are occurring.

SEM and EDX analyses were performed on a Zeiss Merlin^†. EDX maps were obtained with a resolution of 2,048 pixel, 60 frames, and a pixel dwell time of 5 µs. The scan was performedt at 5 keV with a working distance of 4.5 mm. The technique provided detailed information on the corrosion product morphology and elemental breakdowns helping to identify the various aluminum and iron compounds formed.

Interestingly, aluminum exposed to FeCl₃ at 50°C experiences significantly less corrosion regardless of the concentration of exposure. This behavior continues even after the second droplet exposure of FeCl₃. These experiments suggest that the temperature is playing a significant role in reducing the corrosion rates. The data also show that the corrosion rates at these higher temperatures are significantly lower, one of the factors at play may also be related to the time of wetness, with higher temperatures resulting in faster water evaporation to reach the droplets equilibrium volume.¹⁵  This would be dependent, however, on the time taken for the droplet to reach its equilibrium concentration, with droplets exposed to samples at 50°C would reach their equilibrium concentrations significantly faster.

Another key result of Figure 2 is that all samples showed similar changes in the volume corroded with increasing FeCl₃ concentration when the sample was exposed to the same temperature and droplet cycle. The exposure of the sample to a second droplet results in a very large increase in the corrosion rate of the aluminum regardless of the temperature. This is due to an increase in the number of wet dry cycles and materials for corrosion to occur. The corrosion products formed on the surface have a volume many times greater than the volume of aluminum dissolved. This is due to the nature of the corrosion products formed which are very porous and have a low density and consist of a mixture of different corrosion products such as Al(OH)₃, Al₂O₃, FeOOH, Fe₂O₃, etc. When the products are exposed to a second droplet a rapid increase in the corrosion products similar to the increase in corroded volume is not observed. This is potentially due to the pores formed in the corrosion products being filled with a new corrosion product thereby becoming denser and slightly less porous as has been observed.¹⁶ 

Figure 7 shows the distribution of iron-rich compounds (green) being influenced by the temperature. As discussed, at high temperatures we can see that these iron compounds are scattered across the aluminum surface, with only 0.38 mm³of Fe compounds in contact with the surface (scan 1). In contrast, the first droplet exposure at room temperature results in a clear separation of iron-rich deposits. These separate from the deposits of aluminum-based corrosion products that form above the aluminum. At high temperatures, the corrosion products appear to form around these iron-rich deposits. These act as a barrier separating the iron-containing compounds from the aluminum substrate.

Also, there is a possibility of corrosion-driven drying reactions occurring where some of the soluble species are converted into solid products, ultimately resulting in a reduction of the FeCl₃. This reduction may then cause the droplet to lose water to maintain the equilibrium with the humidity, with such processes expected much faster at higher temperatures resulting in lower corrosion in general.

This occurs because Fe³⁺ would reduce to Fe²⁺ in such an acidic environment. Then other cathodic reactions would take over so that the Fe²⁺ forms a variety of ferrous compounds (hydroxides to then oxides) driving multiple reactions. As the iron-rich deposits still consist of goethite (α-FeO(OH)) shown in Figure 16, it indicates that the iron is still in a temporary phase where it has the potential to form iron oxides and, in particular , magnetite/maghemite. However, the corrosion potential is reduced due to the distance from the aluminum of these iron deposits. This results in a reduction in the amount of corrosion occurring. The formation of spheres of aluminum nestled into iron surroundings suggests that the higher temperatures drive the reaction of the aluminum to favor these sphere-like structures and thereby cause the dispersal of iron. In Figure 6 (images [a] and [b]), we observe white dry products formed on the surfaces and the corrosion products having a dark brown color suggesting that it is composed of iron oxides (maghemite) that have reacted onwards.²³  In contrast, Figure 6 (image [b]) also shows the corrosion products closest to the surface having that dark brown color. However, the raised regions have a bright orange color with many white regions at the edges showing that a smaller portion of the iron undergoes reactions with the aluminum at higher concentrations due to lack of contact with the aluminum. These correspond to what is being observed in the Raman where we see the formation of the goethite (iron hydroxide) thereby having the potential to react further to form iron oxides.²⁴  Under standard conditions at room temperature iron oxides (dark brown color) are observed in contrast at 50°C iron hydroxides are mostly observed.

This paper provides evidence of the role of temperature in the corrosion of aluminum when exposed to FeCl₃. The key observations from experiments performed when aluminum samples were exposed are:

• Lower rates of corrosion were observed at higher temperatures, with less penetration into the aluminum alloy for samples exposed to the higher temperatures.

• Less iron/goethite was present on the aluminum surface when exposed to higher temperature. Potentially a reduction in corrosion-driven drying processes.

• Goethite was formed on the aluminum alloy surface, and aluminum oxide was formed above it.

• For samples exposed to 50°C, 2 M droplet exposures show platelets of iron or iron-containing compounds that were deposited across the aluminum surface and that was not in direct contact with the aluminum base, and the samples exposed to 5 M droplet exposures showing the formation of an iron-rich layer which is raised above the aluminum surface and pushed up within the corrosion products formed.

NGN EPSRC Grant number EP/L015390/1, Sellafield Ltd. for financial funding. We acknowledge the Engineering and Physical Science Research Council (EPSRC) for funding the Henry Moseley X-Ray Imaging Facility which has been made available through the Royce Institute for Advanced Materials through Grants EP/F007906/1, EP/F001452/1, EP/I02249X, EP/M010619/1, EP/F028431/1, EP/M022498/1, and EP/R00661X/1.