Objective - Develop a U.S. resource for service life prediction by delivering a suite of measurement science tools including publically available data sets, statistical models, validation data, and geographically specific predictions based on validated statistical models. This resource will be the publically-available source for weathering exposure conditions and material performance data for two major polymer chemistries.
What is the new technical idea? The new technical idea is to utilize the NIST Accelerated Weathering Laboratory to build a U.S. resource for indoor and outdoor weathering data, establish characterization methodologies, and statistical and physical models for predicting property changes. This resource will serve as an engagement mechanism for stakeholders and support the development of standards for service life prediction. The development of service life prediction tools from accelerated weathering is difficult due to the myriad of differences between outdoor and indoor exposure conditions, such as ultraviolet (UV) spectrum, UV intensity, and moisture/temperature variability. NIST has developed the foundational science for accelerated weathering metrologies that predict outdoor performance based on cumulative dosage. This methodology utilizes the SPHERE exposure device and a complex design of experiments to identify intensity-dependent degradation modes in order to develop a cumulative damage model. Upon completing that research, service life prediction efforts at NIST have focused on development of comprehensive characterization methods for chemical degradation pathways and microstructure changes in multicomponent polymeric materials and composites. This prescriptive process is not conducive to performance standards and may lead to limited correlations between accelerated and outdoor service life performance, thus hindering service life prediction models.
This project builds on the current state-of-the-art that correlation between indoor exposure and outdoor exposure can deliver a verified service life prediction model with known predictive capability and will collect data for series of major polymer chemistries used in industry for correlation calculations. In this case, full UV exposure of samples will be conducted on the NIST SPHERE and monitored exposures will be conducted in MD and FL. Databases of controlled indoor and monitored outdoor exposure, changes in photo-oxidation products and industrial relevant performance properties will be compiled. This data will be mined in order to develop a statistical chemical/mechanical property change vs. exposure dosage. Existing physical models available in the literature will also be collected to guide this effort. The loss of performance as a function of dosage will be mapped to a response surface for the material that will be validated against outdoor exposures to prove reciprocity is obeyed and determine the predictive capability of the statistical model. If predictions are successful, the model will be extended to multiple U.S. geographic locations and stakeholders will be engaged to supply additional outdoor weathering data for further validation. This methodology will support technology transfer of standards for weathering to stakeholders. This effort will establish baseline performance and demonstrate the potential of these methods for these classes of materials. To focus this body of research, samples will be prepared without the additional complications of fillers, additives or assembly found in conventional complex systems. Future studies can build on this foundation by examining how the models are changed by these additional factors.
What is the research plan? The research plan will produce a methodology to quickly develop service life prediction models from validated exposure data. This methodology involves four steps:
- Identify critical polymeric materials and important chemistries.
- Establish the characterization methods for photo-oxidation and performance tracking.
- Generate weathering data indoor and outdoor.
- Develop models based on the indoor data and validate the model against the outdoor data.
Determine high use commercial polymers.
Most multicomponent engineered systems fail because critical materials properties change in ways that are not anticipated by the engineering design. In order for end-users to have the most benefit with the results of this project, the selection of materials to study will be driven by a combination of end-user interest, scientific merit, and feasibility. Initial discussions with end-users indicate that two types of polymeric materials are relevant:
- Glassy, cross-linked polymers: Glassy materials are typically irreversibly cured into a final use state. They are identified as having a higher strength and brittle fracture. An amine epoxy will be used to represent this class of polymers, because this is the major chemistry that controls long-term properties. These materials are used as protective coatings, adhesives, and composites (global production 2.6M tons/yr).
- Thermoplastic polymers: Polymers that are typically solid at use-temperatures, but can be heated to a molten state and reformed into specific shapes. These materials are often semi-crystalline and their performance is strongly related to microstructure. The olefinic backbone is generally stable, but will degrade reducing the molecular weight and decreasing performance. Polyethylene (60M tons/yr) of medium density molecular weight will be used for this class of materials. These materials are used as house wrap, geomembranes, piping, and composites.
This project will select one material for degradation modeling. It will also aggregate and analyze prior NIST exposure data to build service life prediction models and expand publically available data. This will enable this project to focus on developing validated statistical predictions for a large array of stakeholders that increase impact.
Establish characterization methods - Leveraging previous accelerated weathering efforts at NIST allows for the use standardized characterization methods for photo-oxidation and mechanical performance. Performance characterization methods that will be used in this study will be selected based on consultation with our industrial partners. Generating the indoor and outdoor data.
The most time consuming aspect of this project is generating the validation data from outdoor exposures. Outdoor exposure will occur in at least two locations as soon materials are identified. It is important for validation that there be at least two locations, Gaithersburg Maryland and at a commercial exposure site in Florida. A larger number of exposure sites increases the validation of the model predictions for the entire United States. These two sites are selected due to the easy access and milder weather in Gaithersburg and the availability of commercial weathering sites in a site that has higher temperature, UV intensity, and humidweather in Florida. The SPHERE (indoor exposures) will continue throughout the life of the project.
Develop models, validate data - The project will work with the Statistical Engineering Division (SED) to provide the statistical models and validation. There is a long standing cooperative relationship with SED that has led to predictive models for sealants. Continued cooperation with SED is critical for model development, but the project will regularly identify new mechanisms to support the modeling effort. Once a validated prediction is obtained, a geographic map of validated property change will be produced.