• Article

Understanding chromaticity shifts in LED devices through analytical models

This paper demonstrates that chromaticity shifts in light-emitting diode (LED) devices arise from multiple mechanisms that produce chemical changes in the materials used to construct the LED devices. Each chromaticity change is shown to proceed over a finite period of time, and there is a limit on the impact of each shift. For example, chromaticity shifts in LED devices usually start with a fast-acting component that quickly reaches a maximum value, followed by one or more slower acting component(s). This behavior can be modeled analytically with a bounded exponential component to describe the fast-acting component, followed by one or more generalized logistic models. These analytical models contain several key parameters, including the limiting value of each chromaticity shift (A for the upper asymptote and L for the lower asymptote) and the rate of the change k.

This approach to chromaticity modeling is demonstrated with analytical models of the chromaticity shifts caused by the irreversible degradation of phosphors. These analytical models provide insights into the kinetic processes responsible for green and red chromaticity shifts caused by phosphor degradation. A green shift is produced by the surface oxidation of the nitride phosphor that changes the emission profile to lower wavelengths. As the surface oxidation reaction proceeds, surface reactants are consumed thereby slowing the reaction rate, and the bulk oxidation processes become more prevalent. A red chromaticity shift can arise from quenching of the green phosphor which shift the emission in the red direction. This paper concludes by discussing the implications of these models for projecting chromaticity for different operational conditions.

Citation

Davis, J. L., Mills, K. C., Bobashev, G., Rountree, K. J., Lamvik, M., Yaga, R., & Johnson, C. (2018). Understanding chromaticity shifts in LED devices through analytical models. Microelectronics Reliability, 84, 149-156. DOI: 10.1016/j.microrel.2018.03.023