Researchers from the University of Twente and Philips Lighting have devised a more predictive modeling method for describing absorption and scattering of light that stays inside an LED. White LEDs use a combination of a blue light emitter and a phosphor to turn the blue into a yellowish white light. The phosphor will scatter some of the light and also absorb some of it. This scattering and absorption are difficult to predict.
One issue that is particularly difficult for current models to predict occurs when the phosphor scatters the light absorbs some of it and then re-emits some of it in another wavelength (color). Maryna Meretska and her colleagues at the University of Twente found that using theory from astronomy helps in predicting the scattering and absorption.
Instead of using the often used diffusion theory, which is only valid in weakly absorbing or thin phosphor layers, they utilized the radiative transfer equation. According to the researchers, the use of the radiative transfer equation can produce a full description of light propagation inside and outside the phosphor plates.
The group found that compared to a description using diffusion theory, using the radiative transfer equation showed absorption levels up to 30 percent higher. They also found that it is considerably closer to the numerical approach of the Monte Carlo check. This difference and the closeness of the predictions to the “Monte Carlo” check shows how potentially inaccurate using the diffusion theory is compared to using the radiative transfer equation.
University of Twente comparison of optical scattering and absorption modeling methods. Diffusion Method -Blue, Radiative Transfer Equation-Red, Monte Carlo Check -Black
Also, the method was shown to be about 17 times faster than the numerical approach, which employs what is known as a “Monte Carlo” check to make predictions about absorption and scattering. The Monte Carlo Check uses randomly generated values of certain variables with a specific distribution of values. Then the random test is completed over and over again until a good approximation of the variable’s actual distribution is reached. While particularly accurate overall, the Monte Carlo check method takes a lot of time and computing power and is generally thought of more as a brute force and inelegant mathematical model.
The research was performed with the Complex Photonic Systems group of UT’s MESA+ Institute for Nanotechnology, together with Philips Lighting in Eindhoven.
The findings appear online in Optical Express.
Reference
M. L. Meretska, R. Uppu, G. Vissenberg, A. Lagendijk, W. L. Ijzerman, and W. L. Vos, “Analytical modeling of light transport in scattering materials with strong absorption,” Optics Express, Vol. 25, Issue 20, pp. A906-A921 (2017). https://doi.org/10.1364/OE.25.00A906
