U.S. Department of Energy

Physics of Reliability: Evaluating Design Insights for Component Technologies in Solar

The SunShot Physics of Reliability: Evaluating Design Insights for Component Technologies in Solar (PREDICTS) program is taking a physics- and chemistry-based approach to identifying failure modes for solar products.

Launched in October 2013, the PREDICTS program is supporting four projects working in two distinct topic areas:

  • Topic 1 awardees focus on the identification, evaluation, and modeling of intrinsic failure mechanisms in photovoltaic (PV) and concentrating solar power (CSP) sub-systems and system components.

  • Topic 2 awardees focus on the development of standard reliability tests for microinverters and microconverters, both in stand-alone and module-integrated configurations.

Awardees

Arizona State University (Topic 1)

  • Location: Tempe, AZ
  • Amount: $1,417,100
  • Cost Share: $390,000
  • Project Summary: ASU will develop a tool to study the fundamentals behind performance and metastabilities of CdTe thin-film photovoltaics (PV). The core of the solution will be a multi-level solver that combines macroscopic diffusion-reaction equations describing sub-systems of point defects combined with the global Poisson equation. This will form a closed system that can be solved in time domain and quasi-3D space utilizing cylindrical grains.

GE Global Research (Topic 1)

  • Location: Niskayuna, NY
  • Amount: $1,933,138
  • Cost Share: $483,285
  • Project Summary: Leveraging internal models developed by GE, this research program focuses on predicting the reliability of hybrid gas bearing (HGB) and dry gas seal (DGS) components in the turboexpander of a supercritical CO2 turbine. The Bayesian model is to include phase changes, low cycle fatigue/high cycle fatigue, dynamic instabilities, and corrosion processes.

National Renewable Energy Laboratory(Topic 1)

  • Location: Golden, CO
  • Amount: $1,556,250
  • Cost Share: $525,000
  • Project Summary: NREL will develop the capability to predict measurable characteristics from models of fundamental physico-chemical processes driving intrinsic degradation and failure mechanisms of advanced materials and coatings targeted for use in concentrating solar power (CSP) systems. We propose to use novel physics- and chemistry-based modeling techniques to predict the time evolution of reflector properties under both controlled and real-world conditions through the creation and application of validated, predictive models that use the measured behavior of individual materials to provide predictions for lifetimes of complete multi-layer reflectors.

Sandia National Laboratories (Topic 2)

  • Location: Albuquerque, NM
  • Amount: $1,387,500
  • Cost Share: $377,021
  • Project Summary: SNL will conduct a comprehensive research and development program to create technically sound, vendor-neutral, and technology-neutral reliability standards for stand-alone and module-integrated PV microinverters and microconverters. The approach synthesizes domain expertise, failure mode and effects analysis, physics-of-failure determination, physical modeling, field stressor data, field reliability statistics, and accelerated laboratory testing to inform draft standards that will reduce technology risk and guide future product development.

Stanford University (Topic 1)

  • Location: Stanford, CA
  • Amount: $1,165,500
  • Cost Share: $297,675
  • Project Summary: Stanford will investigate coupled intrinsic thermo-mechanical and photo-chemical degradation mechanisms that determine the reliability and operational lifetimes for concentrated photovoltaic (CPV) technologies. The research approach will result in quantitative metrologies to characterize degradation processes during PV operation, detailed kinetic models of degradation mechanisms, and reliability benchmarking of materials and interfaces in CPV devices and modules. The approach is based on fundamental materials science and reliability physics for improved materials and development of accelerated testing protocols for more accurate lifetime predictions.