Corrosion in artificial defects. II. Chromate reactions

Abstract

Artificial defects, in the form of slots, were milled through a chromate-containing protective paint system on AA2024-T3 and exposed to neutral salt spray (NSS). Proton induced X-ray emission (PIXE), scanning electron microscopy (SEM) with energy dispersive X-ray analysis (EDS), electron microprobe analysis (EMPA), and Raman spectroscopy were used to characterise the primer and the alloy surface. Chromate was released by the primer to form a 40 μm depletion zone around the edge of the slot. Within the depletion zone, the chromate was reduced but not completely removed. Chromate was detected in the runoff from the slots and was also found to have reacted with the exposed alloy surface. Chromate was found to react with intermetallic particles, smears formed by the milling process, and pits.

Introduction

Chromate inhibition of corroding surfaces has been a topic of ongoing investigation for many years. Numerous studies have focused on the inhibitive effect of chromate in solution on corrosion of aluminium alloys [1], [2], [3], [4], [5], [6], [7], [8]. Others have focussed on the protection offered by chromate conversion coatings [9], [10], [11], [12], including investigations into the inhibitive effect on unprotected aluminium provided by chromate released from conversion coatings [12], [13]. While there have been studies on the leaching of chromate from inhibited primers [14], [15], [16], [17], [18], [19], there are very few studies [20] available on the inhibitive effect of chromate released from inhibited primers during neutral salt spray (NSS) testing.

In part I of this two-part series [21], the development of corrosion on AA2024-T3 was examined in artificial defects (slots) milled through the paint system. The paint system included a polyurethane topcoat, a chromate-inhibited epoxy polyamide primer, and a chromate conversion coating. Milling produced features on the surface not observed on polished surfaces, such as fragmentation of intermetallic particles and smearing of the aluminium matrix over the surface, which were observed to be preferential sites for corrosion.

In previous work, such as that of Howard et al. [22], the protection offered by the cut edge of the primer system is achieved through diffusion of chromate through solution in constant immersion experiments. In service, however, defects in the paint system generally experience cyclic conditions such as wet and dry cycles as well as temperature cycles. Immersion offers more corrosion protection than cyclic environments as the immersion solution generally has an increasing inhibitor concentration whereas the surface is continually washed in the cyclic conditions [23].

In the experiments presented here, exposure to neutral salt spray (NSS) goes part way towards mimicking in-service cyclic conditions as the condensed salt fog continually washes the surface. The NSS environment has been characterised in detail elsewhere [23]. Droplet formation, growth, and removal are dynamic processes that lead to variability in the corrosion within the slots. Wider slots have a higher probability of having isolated droplets in the middle of the slot which are not diffusionally connected to droplets adjacent to the primer. Those droplets in contact with the primer will contain chromate leached from the primer. Those droplets not diffusionally connected to primer will not contain any inhibitor. As droplets form, grow, and are removed, areas of the alloy surface may experience fluctuations in the amount of inhibitor in the droplets and some areas may not be exposed to any inhibitor.

As described in part I [21], exposure of artificial defects (slots) to NSS resulted in S-phase particles and smears on the milled surface being preferential sites for corrosion. In the former case, dealloying of the S-phase particles was similar to that observed on polished surfaces [24]. In the latter case, there was no analogue for the smears on polished surfaces. There were other corrosion sites on the surface that could not be correlated with the underlying microstructure of the alloy. Occasional pitting was observed at Cu–Fe–Mn–Al intermetallic particles, but trenching was not a feature of the corrosion associated with these particles [25]. One possible explanation was that the milling process smeared the aluminium matrix over this class of intermetallic, thus providing a ‘protective’ layer.

In this paper, which forms part II of this series, the leaching of chromate during neutral salt spray testing from the cut edge of the primer system into an artificial defect (milled slot) and its reaction with the milled aluminium surface have been investigated. Leaching of chromate from the primer has been studied using scanning electron microscopy (SEM) in conjunction with energy dispersive X-ray spectroscopy (EDS), electron microprobe analysis (EMPA), proton induced X-ray emission (PIXE), and Raman spectroscopy. The reaction of chromate with the milled surface has been examined using SEM/EDS and Raman spectroscopy.