There has been steady progress over the years in the development of a Physics of Failure (PoF) understanding of the effects that various stress drivers have on semiconductor performance and wearout.
Early investigators sought correlations between the degradation of single device parameters (Vth, Vdd or Isub) and the degradation of parameters related to circuit performance such as the delay between read and write cycles. However, as modeling and simulation capabilities progressed, it was determined that device performance hinged on the degradation of a broad range of parameters instead of just one.
Today, new generations of electronic devices require improved tools for reliability prediction. Simulations provide a wide range of predictions, starting from the lower-level treatment of PoF mechanisms up to high-level simulations of entire devices. In fact, current reliability simulations are commonly based on a combination of PoF models, empirical data and statistical models in order to investigate new manifestations of existing failure mechanisms like Negative Bias Temperature Instability (NBTI), Electromigration (EM), Hot Carrier Injection (HCI) and Time Dependent Dielectric Breakdown (TDDB).
With all simulations, primary questions need to be answered regarding the accuracy of simulation results applied in real world scenarios, and the overall confidence level achieved by the simulations. Reliability data generated from field failures best represents the electronic circuit reliability in the context of the target system or application because field failure rates exemplify competing failure mechanisms’ effects and include actual stresses, in contrast to standard industry accelerated life tests.
Therefore, single failure mechanism modeling and degradation simulations are giving way to system level reliability simulators. The latest circuit design tools have integrated reliability simulators that reflect the needs of real world end users. These simulators model the most significant physical failure mechanisms and help designers address the lifetime performance requirements. Likewise, they simplify the collection of field data semiconductor manufacturers use to validate and calibrate simulations so subsequent lab testing accurately determines device reliability.
Taking a step back from calibration and calibration to design, Sherlock Automated Design Analysis™ Software allows failure to be predicted earlier in the process and for reliability to be designed into to the product. Only Sherlock offers the unique combination of easy-to-use software and industry-specific Physics of Failure analysis for accurate reliability prediction and feedback that happens in hours — not weeks. Faster feedback results in more efficient design revisions, reduced physical testing cycles and speeds introduction to market.
Advancements in reliability testing underscore the importance of applying PoF prediction simulations to electronic devices, and working with experts like DfR Solutions strengthens their effectiveness. To learn more about PoF and its impact on Designing for Reliability, download our Introduction to Physics of Failure Reliability Methods webinar slides below.