Web http://www.jcse.org/

Volume 2 Paper 26


Blistering and Delamination Processes on Coated Steel

D. Greenfield & J. D. Scantlebury
Corrosion and Protection Centre, UMIST, PO Box 88, Manchester M60 1QD

Abstract

The alkaline nature of the environment at the paint/metal interface has been addressed.  A pH indicator has been incorporated into an alkyd coating and the alkalinity of the environment at the interface monitored as a function of time using time lapse photography.

Keywords: Blistering, Delamination, Alkyd, Marine Environment.

Introduction

When a coating is exposed to an aggressive medium, such as that found in a marine environment, the associated failure mechanisms of blistering and delamination are amongst the most important to be considered.  These two modes of failure are often treated separately, but have so many common features that it could be argued that the differences are a matter of degree rather than type.  There is a general consensus over the integrity of a typical coating and it is fair to assume that one will find defects, due to porosity or damage in service.  Therefore the following considers the situation of an organically coated steel structure containing a defect, which is subjected to aggressive, immersed conditions.

Due to the presence of imperfections in the coating, the steel substrate is directly exposed to its surroundings.  This initiates a corrosion process, with the anodic reaction occurring at the defect, this reaction follows that shown in equation (1)

 

Fe → Fe2+ + 2e

(1)

In order to maintain electroneutrality within the system, this reaction is balanced by a cathodic reaction.  In most naturally occurring situations, this reaction will be oxygen reduction, as illustrated in equation (2).  These two reactions initially take place adjacent to each other but separate as the process continues with the cathode moving under the coating.

 

2 H2O + O2 + 4e → 4OH

(2)

The ferrous ions produced in equation (1) go into solution and react to produce electrically neutral compounds through combination with cations in the medium.  This leaves us with a charge imbalance.  The environment has a surfeit of positive charge, in the form of whatever cations are present and the environment at the cathode is producing hydroxyl ions, resulting in an excess of negative charge.  In the case of uncoated steel, the route for the counter-cations is straightforward.  However, when the steel is coated the situation becomes more complicated.  The path from the exposure environment to the cathodic site is either restricted or blocked completely. 

It was shown by Mayne [1] that coatings were so permeable to water and oxygen that their rate of arrival at the cathodic region was greater than that required for corrosion to proceed.  Figure 1 shows a schematic representation of the results of the corrosion reactions, along the lines proposed by Schwenk [2].  It can be seen that the cathodically produced hydroxyl is present at both the blistering and the delamination sites.  The alkaline nature of this resultant solution at these sites is considered to be a major contributory factor in the failure of the coating.

Figure 1: An illustration of the blistering and delamination processes after [2]

The reason this alkalinity causes such failure has been variously ascribed to saponification of the coating [3], dissolution of the oxide layer at the interface [4], alteration of the ionic resistance of the film [5, 6].

One particular feature that has been identified by a number of workers, is a delay time or initiation period between a coated substrate first being exposed to a corrosive environment and the start of the blistering or delamination process taking place.  Leidheiser [7] ascribes this delay as due to the time required to set up a steady state diffusion in the film; whether this diffusion was of water, oxygen or ions was unclear.  Nguyen [8], on the other hand, suggests that the important route for diffusion is that of the cations along the paint/metal interface.  Both interpretations provide explanations for the observed phenomena.

The work described here is concerned with blistering and delamination of an alkyd coating. Part of the essential quality of this research is to use, wherever possible, non-destructive test methods.  To this end, some work has been carried out to determine the feasibility of incorporating pH indicators into the organic coating under investigation. 

The initial tests, described below, were carried out on an alkyd resin.  The hypothesis of the tests was that cathodic areas under a clear resin coating may be identified by the high pH generated due to the presence of OH- produced by the oxygen reduction reaction; it should therefore be possible to track the delamination front as it proceeds along the interface. 

Experimental

Specimen preparation

The substrates used for all the experiments were cold rolled, mild steel Q Panels.  The panels were solvent de-greased and dried with paper towels before being coated in a dust free, ventilated, coating cabinet.  The coatings were cured at room temperature for 48 hours and then stoved at 65-70oC for a further 48 hours.  

In order to provide an indication of the degree of protection offered by this coating, a control experiment was carried out to demonstrate the effect of immersing an unprotected substrate in the test solution of 0.5M NaCl.  This was recorded with time lapse photography and the result is available as video1 in MPEG format in either medium (2.6 Mb) or small (0.9 Mb) size. [Editor's note: there is always a compromise between file size and image quality for video clips - these are the best we can do at present - note that it will take about 1 hour to retrieve 10 Mb using a V90 modem].

In order to track the development of alkaline and therefore cathodic regions under the film, phenolphthalein pH indicator was incorporated into the wet coating, as detailed below. 

After testing various preparations of the indicator for incorporation into the coating, the optimum solution for the alkyd coating was determined to be 1g phenolphthalein in 100ml EtOH + 100 ml water.  Three panels were selected at random to test if the indicator would leach out of the coating.  A drop of pH 13.5 buffer solution was placed on the surface of the coating.  Within a minute, the droplet had turned pink as seen in figure 2.


Figure 2 Effect of a phenolphthalein loaded coating upon a droplet of pH 13.5 buffer solution

The average DFT of the final coatings was 22mm.  A 2mm hole was introduced into the cured coatings and the panels were fitted with a 30mm diameter perspex cylinder, 50mm high, which was filled with a 0.5M NaCl solution.  The panels were left at ambient temperature for 10 days and were photographed at regular intervals.  On the second run of the experiment, time-lapse photography was employed using a digital camera linked to a PC.  Computerised videos were produced showing the progress of the breakdown of alkyd coatings, stills from these videos are included in the result section.

Results

Figure 3 shows the progression of the alkalinity of a typical panel.  Note that after 10 days the alkaline front has progressed under the cylinder and along the panel for a short way.


Figure 3: Stages in blistering process as indicated by phenolphthalein

In all cases, the movement of the pink areas followed the lines of the rolling marks on the surface of the panel.  Although this method provided a picture of the process as it occurred, it was felt that the use of time-lapse photography would produce data that would illustrate the picture more clearly, especially in terns of whether the blistering was in deed affected by the topography of the substrate.

In the first experimental trial, the coating was applied across the striations of the panels and it was felt that this might have had an effect upon the ability of the coating to adequately wet the surface.  Therefore, when the experiment was repeated, the coating was applied along the rolling marks with all other conditions remaining the same.

10mm, single coat alkyd:

This coating failed completely within 30 hours, the progression of its deterioration is shown in figure 4 and as time-lapse video as video2 in MPEG format in medium (9 Mb) and small (3.3 Mb) size. 

rim00001.jpg (172032 bytes)

rim00041.jpg (172032 bytes)

t=0 t=13 hr
rim00090.jpg (172032 bytes) rim00135.jpg (172032 bytes) rim00180.jpg (172032 bytes)
t=30 hr t=45 hr t=60 hr

 Figure 4:  Breakdown of thin alkyd coating (click the image for a larger view)

Once the exposure had finished, the coating was scraped off the substrate to reveal the state of the metal beneath, shown in figure 5.

Figure 5: 10mm coated panel at the end of the exposure trial before and after the removal of the coating

The steel underneath the coating, although corroded, had no corrosion products attached to it; these were on the outside of the film.  This result is in line with the proposition by Mayne [9] that coatings acquire a negative charge when immersed and therefore allow the passage of positive ions.

30mm two coat alkyd.

This coating fared better, the period to breakdown was greater and the mode of failure was different.  Rather than the substrate suffering underfilm corrosion, the coating failed due to delamination and for the period of the experiment, the underlying steel was protected by the alkaline conditions present at the interface.

0000001.jpg (172032 bytes) 0000081.jpg (172032 bytes) 0000150.jpg (172032 bytes) 0000156.jpg (172032 bytes)
t=0 t=27 hr t=50 hr t=53 hr
0000213.jpg (172032 bytes) 0000243.jpg (172032 bytes) 0000300.jpg (172032 bytes) 0000375.jpg (172032 bytes)
t=71 hr t=81 hr t=100 hr t=125 hr

Figure 6: The progress of the blistering and delamination process  (click the image for a larger view).  

The progress of the breakdown of the coating is illustrated in figure 6 and as two sections of time-lapse video as video3a and 3b in MPEG format in medium (3a, 19 Mb and 3b, 26 Mb) or small (3a, 6.8 Mb and 3b, 9.4 Mb) size. Note that there is  magnification change on going from 3a to 3b. There is an initiation period where there is no sign of any activity under the coating, the progression as a function of time is as follows:

At the end of the exposure period, the photographs were examined and a plot constructed of the percentage area of the exposed surface that was indicated as being alkaline.  The results of this plot are given in figure 7, below.

Figure 7: Percentage area of the exposed panel covered by blistering and delamination.

Discussion

The results presented here show quite clearly that, for the coating considered, not only do blistering and delamination processes share many common features but that they may be considered as constituent parts of one overall process. 

The process starts with the initiation of blisters around the region of the point of damage.  These blisters become larger as time progresses until they eventually coalesce.  Once this has occurred, the delamination process takes over and the front progresses rapidly.  As the coating delaminates from the fault, the pH at the interface behind it falls as the process continues.

Figure 7 indicates that the processes occurring under the film are cyclic in nature, starting with an initiation time where there is no apparent activity.  Of course, the indicator used only functions at higher pHs so there would be a delay in its response.  Notwithstanding this, the initial delay period coupled with the apparent inactivity indicated by the plateau at about 100 hours does provide evidence of a cyclic process in line with that proposed by Scantlebury [10].

Conclusions

Blistering and delamination act in conjunction with each other in the failure of organic coatings. 

As blisters develop and grow, they coalesce to form a large disbonded region in the vicinity of a fault in the coating.  This disbonded area provides the site for the delamination process to initiate and propagate.

The initial delay time is repeated in a cyclic nature once the delamination reaches a critical point.  At this point, the pH of the solution under the film at the disbonded area falls as the oxygen reduction reaction moves away from the fault.  At this point, the delamination process gives way to further blistering under the intact coating.  One could view the delamination front needing the blisters to act as stepping-stones in order to progress.

References

1.      Mayne J.E.O. “The Mechanism of the Protective Action of an Unpigmented Film of Polystyrene”.  J.O.C.C.A. Vol. 32, No. 352, Oct 1949, pp 481-487.

2.      Schwenk W. “Adhesion Loss for Organic Coatings, Causes and Consequences for Corrosion Protection” Corrosion Control by Organic Coatings.  Ed H. Leidheiser Jr. 1981,  pp 103-110

3.      Castle J.E. & Watts J.F. “Cathodic Disbondment of Well Characterised Steel/Coating Interfaces”.  Corrosion Control by Organic Coatings.  Ed H. Leidheiser Jr. 1981,  pp 78-86

4.      Ritter J.J. “Ellipsometric Studies on the Cathodic Delamination of Organic Coatings on Iron and Steel”.  J. Coat. Tech. Vol. 54, No. 695, 1982, pp 51-57

5.      Mayne J.E.O. & Mills D.J. “The Effect of the Substrate on the Electrical Resistance of Polymer Films” J.O.C.C.A. Vol. 58, 1975, pp 155-159

6.      Skar J.I. & Steinsmo U. “Cathodic Disbonding of Paint Films — Transport of Charge”.  Corrosion Science Vol. 35, Nos. 5-8, 1993, pp 1385-1389.

7.      Leidheiser H. Jr. Wang W & Igtoft L “The Mechanism for the Cathodic Delamination of Organic Coatings from a Metal Surface” Prog. Org. Coat. Vol. 11, 1983, pp 19-40

8.      Nguyen T. Hubbard J.B. & McFadden G.B. “A Mathematical Model for the Cathodic Blistering of Organic Coatings on Steel Immersed in Electrolytes” J. Coat. Tech. Vol. 63, No. 794, 1991, pp 63-52.

9.      Mayne J.E.O. “The Mechanism of the Inhibition of the Corrosion of Iron and Steel by Means of Paint” Official Digest, Feb 1952, pp 127-136.

10.    Scantlebury J.D. “The Dynamic Nature of Underfilm Corrosion” Corrosion Science Vol. 35, Nos. 5-8, 1993, pp 1363-1366.