Optical modeling and light extraction of an LED with surface roughening and sharpening Ching-Cherng Sun and Chao-Ying Lin Institute of Optical Sciences, National Central University, Chung-Li 320, Taiwan Tel: +886-3-4276240, Fax: +886-3-4252897, Email : ccsun@ios.ncu.edu.tw ABSTRACT We have presented our study on optical models for simulating the light distribution of an LED and for calculating light extraction efficiency of an LED chip. In the former, we have precisely predicted the light distribution of an LED by a lens. In the latter, we has proposed to introduce a periodic sharpening structure on the interface between the sapphire and the n-GaN of a GaN-based LED to obtain as high as 73% in light extraction efficiency. Keywords: optical model, light extraction efficiency, sharpening structure, inverse pyramid, GaN 1. INTRODUCTION Light emitting diodes (LED) are regarded as the most important light sources in solid-state lighting for their advantages in energy efficiency, long life, vivid colors, high reliability, environmental protection, safety and multiple applications. [1-3] Therefore, LEDs have been extensively studied in the past decades for lighting proposes. In developing solid-state light based on LED, two optical issues are important. One is the optical model of the LED, and the other is the light extraction analysis of the LED chip. The former is the basis of the LED when LEDs are applied to the light sources in any kind of application. The latter is quite important when one tries to increase the external quantum efficiency of an LED. In this paper, we will present our studies on optical models for simulating the light distribution of LEDs and for calculating the light extraction efficiency of a GaN-based LED with introducing surface variation with a periodic structure. Third International Conference on Solid State Lighting, edited by Ian T. Ferguson, Nadarajah Narendran, Steven P. DenBaars, John C. Carrano, Proc. of SPIE Vol. 5187 (SPIE, Bellingham, WA, 2004) · 0277-786X/04/$15 · doi: 10.1117/12.512296 100