Web http://www.jcse.org/

Volume 2 Paper 36


Filiform Corrosion of Aluminium Alloy 3003 H14 under Humid and Immersed Conditions

Z. Marsh, J. Marsh* and J.D. Scantlebury

Corrosion and Protection Centre, UMIST, Manchester 60 1QD, United Kingdom.
*CAPCIS Ltd, CAPCIS House, 1 Echo Street, Manchester M1 7DP, United Kingdom.

Abstract

Epoxy coated aluminium 3003 H14alloy panels were examined for susceptibility to filiform corrosion under both humid and immersed conditions. The panels were not artificially damaged, but were artificially contaminated with sodium chloride. Significant filiform corrosion was found to occur at 84% relative humidity (RH). However, the extent of filiform corrosion was greatly reduced at 100% RH. Extensive filiform corrosion was also noted for samples immersed in 1% NaCl solution, especially during the early stages of exposure, and in glycerol:water mixtures. If a cross-cut specimen is examined under immersion conditions, severe filiform corrosion rapidly occurs, before steadily converting to general under film corrosion, especially in the vicinity of the cross-cut. The results clearly show that aluminium 3003 H14 alloy can undergo filiform corrosion under immersion conditions.

Keywords: Filiform Corrosion, Aluminium 3003 H14 Alloy.

Introduction

One of the most common forms of corrosion occurring on coated aluminium and aluminium alloys is filiform corrosion. While this form of corrosion rarely, if ever, leads to a working failure, the consequences can be severe for any decorative function[1]. In appearance, filiform corrosion on aluminium alloys resembles long thin thread like filaments, often up to several centimeters long, but less than 1 milimeter wide. The filaments are generally white, with a grey head. This head forms the active corrosion site, with corrosion propagation and filament growth taking place here.

Filiform corrosion is often induced by the presence of chloride salt contamination on the aluminium surface. As described by Slabaugh et al[2], the head of the filament usually contains significant quantities of chloride ions, with little or no free chloride in the long tail. Any chloride present in the tail is tied up as an aluminium oxychloride. The chloride containing electrolyte appears to move forward with the head as the filiform corrosion progresses across the aluminium substrate.

Two primary filiform corrosion mechanisms are put forward in the literature[2,3,4,5]. These consist of a cathodic delamination mechanism, discussed by Funke[3], and an anodic undermining mechanism, as proposed by Ruggeri and Beck[4]. Lenderink[5] believes that the first mechanism is generally thought to be more important for steel, and the second for aluminium.

Aluminium 3000 (Al-Mn) alloys are most often used in the manufacture of cooking utensils and in the canning industry[6], although some limited use as an architectural material also occurs. These are two areas where a retention of the visual and cosmetic properties of the coated product is extremely important.

It has been generally accepted that filiform corrosion only occurs under humid conditions. However, the authors have attended a number of public presentations where speakers described observing "filament like" structures under immersion conditions on lacquered aluminium surfaces, especially with respect to food and drink cans. However, the speakers were extremely reluctant to describe this as filiform corrosion because "filiform corrosion does not occur under immersion conditions". This publication examines the tendency of epoxy coated aluminium alloy 3003 H14 to undergo filiform corrosion under a variety of humid and immersed conditions.

Experimental

The composition of the aluminium alloy 3003 H14 (Q-Panel, U.K.) is given in Table 1. The alloy is one of the 3000 series Al-Mn alloys.

Table 1 Composition of Alloy 3003 H14

Aluminium

Copper

Iron

Manganese

Silicon

Zinc

Other

96.8-97.4

.05-0.2

0.7

1.0-1.5

0.6

0.1

0.15

The as received sample panels were rinsed with de-ionised water and degreased with acetone. They were then contaminated at the centre of the panel with a 5μl droplet of 0.1g/L sodium chloride solution using a micro pipette. This gives 0.5μg of contamination. Upon drying the contaminated zone was of approximately 2mm diameter. The panels were then coated with a 50μm wet thickness coating of a 2-pack epoxy/polyamide clear lacquer coating (Epilife/H Marcel Guest, Manchester, U.K.) using the draw bar technique. The curing regime consisted of 48 hours at room temperature followed by 48 hours at 60oC. The resultant coating was of 20-25μm thickness.

Coated samples were exposed to relative humidities of 84 and 100% at 28-29oC. Samples were also immersed in 1% sodium chloride solution, 1% hydrogen chloride solution or 1:10 glycerol:water mixtures. The samples were generally undamaged, but some cross cut specimens were also examined in 1% sodium chloride and 1% hydrogen chloride solutions.

Results

84% RH

Figure 1 shows examples of epoxy coated aluminium 3003 H14 panels exposed to an environment of 84%RH. Only a very limited amount of filiform corrosion can be observed, with only a small number of filaments growing from the contamination site.

100% RH

Little or no evidence of filiform corrosion was observed at this humidity. Some general corrosion and blistering occurred in the vicinity of the contamination spot.

Immersion in 1:10 water:glycerol

Figure 2 shows an example of a sample exposed under immersion to 10:1 water:glycerol. An area of general corrosion surrounding the central contaminated area is clearly visible after 7 days. However, beyond this region, filiform corrosion is clearly observed to have occurred at 14 days. An area of general corrosion has lead to the initiation of filiform corrosion.

Immersion in 1% sodium chloride solution - intact specimen

Figure 3 shows an example of a specimen immersed in 1% sodium chloride solution. Here, filiform corrosion initiates around the general corrosion in the contamination zone. However, with time further general corrosion initiates, spreading across the area affected by filiform corrosion. Filiform corrosion was far more extensive than that observed on any of the samples exposed to humidity.

Immersion in 1% sodium chloride solution - cross-cut specimen

Figure 4 shows an example of a cross-cut specimen immersed in 1% sodium chloride solution. The cross-cut is surrounded by a region of general corrosion. This region is surrounded by extensive filiform corrosion after 35 days of exposure, covering most of the available exposed area.

Immersion in 1% hydrogen chloride solution – scribed specimen

Samples were partly immersed in 1% hydrogen chloride solution. Within the immersed region, general corrosion rapidly initiated, with undermining of the coating occurring within 48 hours. However, filiform formation occurred at the scribed defect in the region above the immersion line. Figure 5 shows a region of a specimen immediately above the immersion line after 21 days of exposure. General corrosion is beginning to advance from the scribe cut into the region where filiform corrosion has occurred.

A summary of the results for each system is given in Table 2.

Exposure Conditions

Observation

84% Relative Humidity

Small number of slow growing filaments

100% Relative Humidity

Little or no evidence of filiform corrosion

Immersion in 10:1 Water:Glycerol – intact

Larger number of faster growing filaments initiating from area of general corrosion in vicinity of contamination.

Immersion to 1% sodium chloride solution - intact

Initially severe filiform corrosion, changing to general corrosion after several weeks of exposure.

Immersion in 1% sodium chloride solution - scribed

General corrosion at scribe, surrounded by severe filiform corrosion.

Part immersion in 1% HCl

Severe general corrosion in immersed zone. Filiform corrosion above immersion line with general corrosion advancing from the scribe with time.

Table 2 A summary of the exposure conditions and observations for epoxy coated aluminium 3003 H14 alloy samples exposed to a variety of conditions.

Discussion

It is not the purpose of this publication to give a detailed mechanistic interpretation of filiform corrosion, or to give anything more than a general indication of corrosion rates in the environments examined. The primary objective is to establish whether filiform corrosion occurs under immersion, as opposed to humid conditions. From specimens exposed to 84% humidity, it is noticeable that there is only a limited level of filiform corrosion, even after 8 weeks of exposure. When the relative humidity is raised to 100%, little or no filiform corrosion is present. This would seem to be in agreement with the theories generally put forward for the initiation and propagation of filiform corrosion, where a humidity range of 60-95% RH is suggested[2-5]. The indication would be that filiform corrosion would not occur under immersion conditions, as these would be representative of 100% RH, when a liquid water film would form on the sample surface. However, the authors have attended a number of public presentations where speakers described observing "filament like" structures under immersion conditions on lacquered aluminium surfaces, especially with respect to food and drink cans. However, the speakers were extremely reluctant to describe this as filiform corrosion because "filiform corrosion does not occur under immersion conditions".

When the corrosion of these panels is examined under immersion conditions it is apparent that the conventional theory has some flaws. For intact panels immersed in both 1% sodium chloride solution and 10:1 water:glycerol, general corrosion occurs in the vicinity of the salt contamination. However, beyond this region extensive filiform corrosion is observed. The level of filiform corrosion is substantially greater than that noted for any of the specimens exposed to humidity. When the scribed panels immersed in 1% sodium chloride solution are examined, general corrosion occurs at the scribe, but severe filiform corrosion can be noted in the surrounding zone. Filiform corrosion clearly can occur for this aluminium alloy under immersion conditions where the water activity is close to unity, and under conditions where the water activity is rendered less than unity by the use of water:glycol mixtures.

Conclusions

Filiform corrosion can occur on epoxy coated aluminium alloy 3003 H14 under both immersed and humid conditions.

Also, filiform corrosion is far more extensive under the immersion conditions examined, compared to the humid conditions.

References

    1. S. Wernick, R. Pinner and P.G. Sheasby, "The Surface Treatment and Finishing of Aluminium and its Alloys", 5th Ed., Vol 1, p14 (1987), ASM International-Metals Park-Ohio/Finishing Publications Ltd-Teddington-Middlesex.
    2. W.H. Slabaugh, W. Dejager, S.E. Hoover and L.L. Hutchinson, J. Paint Technol. 44 (1976) 76.
    3. W. Funke, Prog. Org. Coatings, 9 (1981) 29.
    4. R.T. Ruggeri and T.R. Beck, Corrosion-NACE, 39 (1983) 452
    5. H.J.W. Lenderink, PhD Thesis, Delft University of Technology (1995).
    6. S. Wernick, R. Pinner and P.G. Sheasby, "The Surface Treatment and Finishing of Aluminium and its Alloys", 5th Ed., Vol 1, p5 (1987), ASM International-Metals Park-Ohio/Finishing Publications Ltd-Teddington-Middlesex.

Figures

Figure 1 Filiform corrosion of epoxy coated aluminium alloy 3003 H14, after 8 weeks exposure to 84% RH (Left - 20μm coating, Right - 40μm coating).

Figure 2 Filiform corrosion of epoxy coated aluminium alloy 3003 H14, after exposure to 10:1 water:glycerol (Left - 7 days, Right - 14 days).

Figure 3 Filiform corrosion of epoxy coated aluminium alloy 3003 H14, after exposure to 1% sodium chloride solution (Left-14 days, Right-35 days).

Figure 4 Filiform corrosion of cross-cut epoxy coated aluminium alloy 3003 H14, after exposure to 1% sodium chloride solution for 35 days.

Figure 5 Filiform corrosion of dual circle cut epoxy coated aluminium alloy 3003 H14, after exposure to 1% hydrogen chloride solution for 21 days. The sample was partly immersed. The section shown is just above the immersion line.