Year: 1994
Type: report
Abstract: Over the last two decades, the organic coatings industry has undergone rapid technological and structural changes.These changes have been induced by legislative actions such as restrictions pertaining to hazardous chemicals, toxic effluents, and volatile organic compounds.The consequence of these changes has been the displacement of almost all commercially-important, well-established coatings (largely high- solvent coatings) by newer systems, the formulation and application of which arc based on different chemistries and technologies.Unlike the displaced coatings, however, the new coatings do not have performance histories and the only accepted method for generating performance data is through an extensive outdoor exposure program.Since outdoor exposure results typically take five years to obtain, a desperate need exists for a methodology which is capable of generating timely, accurate, and reliable service life estimates of a coating system.This report reviews the attributes of the service life prediction problem which are common to all materials, components, and systems in an effort to establish a set of criteria for assessing the adequacy of existing or proposed service life prediction methodology for coating systems.The current durability methodology and the reliability-based methodology are then evaluated against these criteria.The proposed cnteria include the ability to 1) handle high variability in the time-to-failure data for nominally identical coated panels exposed in the same service environment, 2) analyze multivariate and censored time-to-failure data, 3) establish a connection between laboratory and field exposure results, and 4) quantitatively predict the service life of a coating system exposed in its intended service environment.The current durability methodology was developed prior to the recognition of the proposed criteria and, as such, was not designed to satisfy these criteria.Efforts to correct its deficiencies have been made over the last 80 years, but success has been elusive.The failure of the current methodology has generally been ascribed to inadequacies in laboratory-based aging tests, specifically, the inability to simulate weathering conditions in the laboratory.However, we suggest that the failure of the current methodology can be attributed to its being based on faulty premises, inadequacies in experimental design, and the lack of reproducibility of the weather over any time scale.An alternative, reliability-based methodology is reviewed and assessed.This methodology has a strong theoretical basis plus a history of successful applications in the electronics, medical, aeronautical, and nuclear industries.A number of experiments with coatings has already been conducted using this methodology.The results indicate that this methodology can be applied in predicting the service life of coating systems and that it satisfies the proposed service life prediction criteria.Implementation of a reliability-based methodology will require substantial changes in the current experimental procedures.These changes result from the quantitative nature of the service life data and will include 1) more systematic characterization of the initial properties of a coating system, 2) quantitative characterization of each of the weathering variables comprising the in-service environment, 3) quantification of macroscopic degradation and relating submacroscopic to macroscopic measures of degradation, 4) utilization of experimental design techniques in planning and executing short-term laboratory-based experiments, and 5) development of computerized techniques for storing, retrieving, and analyzing collected data.These changes will be justified in view of the greater reliability of the results.
DOI: 10.6028/nist.bss.172
Authors: Jonathan W Martin, Sam C Saunders, F Louis Floyd, John P Wineburg
Language: English
Subfield: Safety, Risk, Reliability and Quality
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W637938763