Defect Detection for Lithium Ion Battery Manufacturing via Automated Image Processing of Pulse Thermography

Peter R O'Regan, Purdue University


Lithium ion batteries offer some of the highest power and energy densities of commercially available secondary batteries, but their manufacturing requires tight tolerances to achieve this performance with high reliability. Pulse infrared thermography has been developed to provide on-line monitoring of battery electrodes during manufacturing so that defects in battery cell anode and cathode films can be detected before final cell assembly. With this technology, composition differences as small as 2% and thickness gradients on the order of 10 &mgr;m are demonstrated to be detectable. Custom image processing algorithms are developed to be used with pulse thermography, leading to automated detection, classification, and quantification of defects on electrode films. Imperfections such as gradients, bubbles, streaks, and mixing issues are efficiently identified on lab-created and simulated electrode films. This process can be readily adapted to manufacturing operations.




Caruthers, Purdue University.

Subject Area

Chemical engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server