Test Methods to Predict the Microbial Attack of Waterbased Coatings [MAWC] - Project Update

Note:The following summaries are provided for dissemination purposes to those who may have an interest in the work but not be directly involved in the projects. They may however contain references to other project reports containing full experimental details, access to which is restricted to Industrial Advisory Group Members.

Overall Summary for Project MDE Dl to the End of the Eleventh Quarter

Introduction

Structural materials are frequently coated with polymeric paints and varnishes to afford protection from environmental degradative processes. The organic nature of such coatings make them susceptible to spoilage by microorganisms, eg algæ bacteria and fungi. Infestation is progressive until a point is reached when the coating must be replaced either for æsthetic reasons or because an invading organism has damaged the coating to an extent that diminishes its protective properties.

Predicting the ability of a coating to resist microbial spoilage is of concern to the coating manufacturers and their raw material suppliers, as well as applicators and end-users of the coating product. The requirement for accurate predictive test methods is of current importance because of the increased emphasis on environmentally friendly, but potentially susceptible formulations, of which water-based systems are a particular example.

Spoilage prediction for coating products is traditionally based on short-term laboratory procedures (eg BS 3900 G6), augmented by several months of field trials. The former are favoured on the grounds of time, although test conditions may be unrealistic. The latter approach has the disadvantages of being longer-term, and the results obtained are known to be sensitive to the local conditions of the selected field location. Inter-correlations between tests and with real lifetime records are frequently poor. There is limited understanding of much of the basic science underpinning present trial protocols, a fact that is most certainly responsible for the poor predictive responses from current test methods in dealing with initially "resistant" coatings.

Recognising these shortcomings and the importance of reliable test methodology to industry, the DTI have instigated a programme to develop novel fungal spoilage test protocols for applied coatings. Particular attention is to be given to those studies which enhance understanding of the science that underlies any test procedure.

A research contract was awarded to the PRA and an Industrial Advisory Group (IAG) formed to oversee progress.

The Work Programme

The programme of work being undertaken by the PRA in fulfilment of the project contract (see The Proposal and Quarterly Report 1) has involved:

Statistical Design and Analysis

In view of the complexity of the systems to be studied, the PRA has sought expert advice and guidance in the use of experimental design and statistical analysis. Although the statistical techniques required are well established, the details needed for use of design theory and analysis were generally less familiar to the members of the IAG. Introducing the experimental design and analysis concept to the IAG (see QR 2 and subsequent reports) is an achievement of the dissemination programme which had not been foreseen in the proposal. Use of this approach has clearly increased the efficiency and effectiveness of the field trials and is allowing novel, and unexpected, findings to be extracted from the data.

As the project approaches its concluding stages increasing effort is being applied to the data analysis. The data acquisition for the project will effectively cease in September 1999.

Local Climate Weather Monitoring

Intensive local climate (and more important microclimate) monitoring is a procedure not typically found on industrial coating exposure testing sites. An achievement of the project has been the installation of automatic climate monitoring units at each exposure site, and the collecting and processing of the data. The results to date show there can be significant variation in microclimates on test panels exposed at three sites, located over a relatively small geographical area. (QR 4 and 5).

In the ninth quarterly report, summary tables of the available climate data were published. It is believed that the project's weather records form a useful data set in their own right which will be of interest elsewhere in the scientific community.

While intuition suggests that the local climate should influence the rate of colonisation and growth of microorganisms, and indeed seasonal and site variations are found, the statistical analysis of the results had until recently failed to establish any strong correlations between rate and extent of growth and the particular climate features selected for study. The work is ongoing and new methods to probe for subtle relationships are being examined.

In the Eleventh Quarterly Report a break-through is disclosed. An apparent, good, correlation between a high number of hours of wetness at one of the exposure sites, and a period of concurrent explosive microbial growth is reported.

The Exposure Trials

The field exposure trials have allowed investigation of coatings on three typical constructional substrates, wood, metal and an exterior board. The experimental design has involved phased exposure, which is allowing seasonal effects as well as inter-site factors to be probed. Additionally, the exposure rack (the design of which was developed for the project) has allowed exposure direction and angle to be investigated. Aspect has proved to be a significant factor in the spoilage and one which will be of importance in the formulation of future test methods. ( See for example the design of new test cabinets in QR 9 & 10).

Analysis of all the effects of the various factors studied is being undertaken and these are being identified and ranked by significance. While the main effects of site, formulation etc, are clearly observed and quantifiable, at an early stage it become apparent that interactions are also of great importance (see QR 5 for example). It has been shown, for example, that a relationship exists between the exposure angle and the type of biocide present in the coating. Differential leaching due to weathering may provide an explanation for this result.

In the current Report ( 11) it is statistically confirmed that coatings on a wood substrate behave distinctly, at all sites, in comparison with coatings on metal or plaster board. Also reported is an indication that nine months into the exposure programme, the rural site alone began to exhibit a dramatic increase in activity.

Subjective v Objective Assessment

The subjectivity of the traditional method of spoilage assessment, by a visual rating on zero to five scale is questionable. The project has developed and tested electronic image capture and analysis methods as a means for more objective measurement. This approach has the additional benefit of producing achievable and retrievable pictorial records. The task has proved more difficult than expected, (QR Nos 2 and 4). An automated image capture and analysis method has, however, been developed.

This assessment method is producing ratings which may be more closely paralleling the microbial ecology assay studies than the normal visual assessment. It is probable the technique is monitoring the life and death cycles of the organisms in a way not discernable by visual observation. At the current stage of the project it is not possible to say whether this approach will provide a more effective prediction of the test materials life time spoilage pattern.

Early Biochemical Detection Methods

Work undertaken by the subcontractor (CABI-Bioscience) has probed potential biochemical methods for detecting and quantifying early activity of the microorganisms on the surface of exposed coated panels. A number of avenues have been explored with initially limited success. (QR 8 and 9). The more recent work, testing for ergosterol levels have given encouraging results (QR 10).

An additional (not within the original project scope) piece of work is reported in Quarterly Report 11. A molecular biology approach is being applied to amplify the DNA profiles and hence detect and fingerprint each of the species present in a contact sample taken from the surface coatings.

Microbial Ecology Assay

As well as quantifying the rate and magnitude of spoilage, the project plans to catalogue the microbial ecology changes on test panels at the various sites over time. It has been noted that the relative dominance of the various species change both between sites and with time, which may be evidence of serial replacement rather than late colonisation.

The ninth quarterly report, for example, notes that algal growth is now being detected, but only at the rural site and then only on North facing samples.

The previous Quarterly Report ( 10) gave extensive details of the micro-ecology evaluations over a twelve month period. The ecological assay approach is probing features of spoilage in a way distinct from the other observational techniques. The preponderance of the various organisms with respect to experimental parameters are now being analysed. It is apparent that the method of assay significantly affects results and the feature will be the subject of further investigation

In the current report (Report 11) the variation in type and count of species and their relative dominance under different conditions is shown to be a very complex issue.

Laboratory Test Procedures

Using the knowledge gained in the field studies new approaches to short-term test procedures have been formulated (QR 8)and are to be tested. The report gives details of the construction of special test cabinets. Of note was the need for incorporation of exposure angle and test panel preconditioning (artificial weathering) into the experimental design.

Report 10 gave details of initial results. The finding that, with the exception of resin type, the laboratory experiments fail to detect the variability seen under natural exposure is significant and serious. It may be that the pre-conditioning, carried out on the test panels, was too extreme. Further experiments are in hand.

The Eleventh Report states a conclusion that while the existing standard laboratory test protocol (BS 3900 : G6) does induce rapid colonisation of susceptible coatings, its predictive power is questionable. It is demonstrated that the test procedure, as it stands, is unable to discriminate between coatings which in field trials display variation in the time they remain resistant to microbial spoilage.

Conclusion and Outlook

The programme continues to progress well and is providing novel results in all the areas of activity. The large volume of data generated has created some backlog in analysis and interpretation. Detailed analysis had been confined to tests involving the wood substrate. These deficiencies are being eroded as increasing attention is given to analysis and the data acquisition programme draws to a close.

The programme investigating new short-term laboratory tests, designed using the knowledge gained from the field trials, had in the initial phase yielded unexpected results and additional experiments are being carried out.

The apparent absence of a clear correlation between weather factors and growth pattern has prompted a search for more subtle relationships from within site climate records. Natural exposure data, and climate recordings in excess of twelve months field exposure, are now becoming available and these should increase the chances of identifying causes of seasonal variation.

The microbial assay studies suggest that such method are assessing features distinct from those which affect the classical visual ratings, which are typically used in current predictive testing.

Dr R Holman
Project Manager

August 1999

Note: The Funding Programmes "Degradation of Materials in Aggressive Environments" (DME) and "Materials Degradation in Aggressive Environments" (MDE) form parts of the UK Government's, Department of Industry (DTI) support for manufacturing industries.