Premium
Uniformity investigation of MOCVD‐grown LED layers page 30–40 Haisheng Fang et al.
Author(s) -
Jiang Zhimin,
Yan Han,
Liu Sheng,
Zhang Zhi,
Gan Zhiyin,
Fang Haisheng
Publication year - 2016
Publication title -
crystal research and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.377
H-Index - 64
eISSN - 1521-4079
pISSN - 0232-1300
DOI - 10.1002/crat.201670001
Subject(s) - susceptor , wafer , metalorganic vapour phase epitaxy , materials science , optoelectronics , graphite , enhanced data rates for gsm evolution , composite material , layer (electronics) , epitaxy , telecommunications , computer science
The image shows a graphite susceptor, on which GaN wafers were grown. Uniformity of the produced wafers is at the heart of high‐power‐LED industries. The main purpose of the paper is to propose a direction to optimize the MOCVD reactors by comprehensively analyzing uniformities of the wafers. To achieve this goal, growth of GaN compounds in a specific MOCVD reactor, capable of a one‐time production of 36 × 2″ wafers of nitrides, has been investigated. To examine growth uniformity across a wafer and from wafer to wafer, the reactor is divided into Zone A, Zone B and Zone C according to distance to the center of the graphite susceptor. Comparative analysis of each zone offers a straightforward view of the mean excitation wavelength, luminous intensity, film thickness and their standard deviations. Conformity of the growth uniformity in each zone is further checked carefully through averaging across‐wafer and wafer‐to‐wafer variables and their standard deviations. Zone B is found to retain excellent wavelength uniformity, since it is located at the middle of the susceptor with weaker effects of the susceptor edge and of the inlet gas flow. Zone A, at the center of the reactor, has the best mean intensity and thickness uniformities due to a well control of the infrared temperature measurement during the growth. Zone C is the worst in the uniformities, and should be the main focus when optimizing the reactor. The above experimental analysis reveals the principles common to the MOCVD technique, and provides a basic for further optimization of the process window to improve the cycles with considerable reduction of the costs. (Picture: Haisheng Fang et al., 10.1002/crat.201500135 pp. 30–40, in this issue)