Coatings Technology at NIST Christopher C. White FEDERAL LABS AS BEACONS FOR INNOVATION AND ENTREPRENEURSHIP FLC Mid-Atlantic Regional Meeting Maritime Institute of Technology and Advanced Graduate Studies (MITAGS) Linthicum Heights, MD November 18-19, 2014

Opportunity Conventional test methods for the service life prediction of polymeric materials in outdoor exposures do not generate reliable or repeatable results. They are expensive and time consuming. Goal: Develop test methods which have the ability to accurately, precisely and reliably predict the in-service performance of polymeric materials designed for specific outdoor exposure in less than real time.

Durability testing now: Threshold testing – Accelerated, typically machine based. – Based on historical performance. – Not typically correlated to in-service performance. – Equipment is based around a specific test. Long term outdoor exposure, again threshold. – Slower, requires years. – Typically, not repeatable. – Not predictive.

Current SLP problem… I did an accelerated/ threshold test, how long will the product last in service? The same as: If I tell you the tensile strength of steel, tell me how big of a bridge I can build… Currently, we have prescriptive tests. Very difficult to correlate to real world performance.

Manufacturer’s Dilemma: Time to Profit Coatings > 15 yr product introduction IC Chip ~ 4 months Increased Liability Fire Retardant Plywood Polybutyldiene Pipe Moisture Resistant Coatings EFIS Automotive Clear Coats (4-6 $B/yr. In warranty costs.) Class Action Lawsuits Increase Time to Profit or Increase Liability Exposure Current Service Life Methodology: Inability to establish durability creates a barrier to innovation.

Service Life Prediction Outdoor exposure Laboratory exposure or “ This brings us face-to-face with one of the most perplexing problems concerned with outdoor weathering, that the weather does not duplicate itself. How can one ever expect a laboratory method to duplicate the weather when the weather can never duplicate itself” [Grinsfelder, 1967] “ Successful laboratory simulation of the effects of weather on coatings, plastics and other materials has eluded scientists for over fifty years. Published literature report hundreds of attempts to duplicate and accelerate weathering effects and conclude that there is no substitute for natural weathering “ [Dreger, 1973] Current Reality: The Standard is Outdoor Weathering “ Current estimates of Service Life Prediction are Crude and there is Little or no Correlation between Laboratory and Field Exposure.” Rilem State of the Art Report, 1999.

How do we get to Reliability Based SLP? Elements of a Standard Methodology for Service Life Prediction: Characterize the service environment- Dose Characterize the material - Damage Identify the degradation mechanisms – Develop a model for predicting the rate of degradation Define the failure criterion Using the model, calculate the time to failure Prepare a report of the results in standard format stating clearly the assumptions made ASTM E632 Protocol

Creating Controlled Exposure SPHERE: Simulated Photodegradation by High Energy Radiant Exposure Lots of Light Stable to limit of detection Uniformity >95 % Temperature ± 0.1 C Humidity ± 0.2% 32 Chambers, 17 samples/chamber >500 samples at a time. Use design of experiments for all correlations of Temp, Humidity, UV, & Load

Equipment Need to be able to transfer all of this equipment to the public. – Commercially viable versions of our SPHERE. Independent control of light, temperature and humidity. Data should be equivalent to our SPHERE or better. Guidance from customers, suppliers to help shape this device.

Reliability-Based SLP Methodology Temperature Material Response Laboratory Cumulative Dosage Model Specific SLP Estimate Damage Mechanism Instrumented Outdoor Exposure Controlled Separable High Volume High Flux Environmental Exposure Moisture UV

Cumulative UV Dosage Models  dt )101)(,()( 0 )( max min  t dtEtD A ototal    D total (t) = Damage to material. min and max = minimum and maximum photolytically effective wavelengths E o (,t) = spectral UV irradiance from light source (1-10 -A( ) ) = spectral adsorption of specimen  ( ) = spectral quantum yield of specimen A( ) = adsorption at wavelength DOSE

Lab: 25°C,100% ND (16) 35°C,4RH,4WL, 100% (64) 45°C,2RH,4WL, 100% (32) 55°C/75%RH,60% ND (16) Outdoor: G4-17 (56) Changes from Current Practice- Predictive Models: Field and Laboratory Failure Mechanisms are the Same

Model, based on Laboratory data, successfully predicts damage occurring outdoors Using Model-Free Heuristic Approach

Modulus

Opportunity Conventional test methods for the service life prediction of polymeric materials in outdoor exposures do not generate reliable or repeatable results. They are expensive and time consuming. Goal: Develop test methods which have the ability to accurately, precisely and reliably predict the in-service performance of polymeric materials designed for specific outdoor exposure in less than real time.

Coatings Technology at NIST Christopher C. White FEDERAL LABS AS BEACONS FOR INNOVATION AND ENTREPRENEURSHIP FLC Mid-Atlantic Regional Meeting Maritime Institute of Technology and Advanced Graduate Studies (MITAGS) Linthicum Heights, MD November 18-19, 2014