| With the development of the high and new technology re, semiconductor lasers (LDs) as a kind of important light source are widely used in optical fiber communication, optical sensor, information memory, medical treatment, pumping solid lasers and optical fiber amplification . The quality and reliability of semiconductor lasers is imperative in all the applications. The study on the reliability of LDs and its regularity is very significant to improve their specific property and use them rightly. At present, the usual method of screen is electric aging by increasing the temperature and aging current to measure LDs threshold, quantum efficiency and output power. These devices with high variance of output optic power at some current or high variance of the drive current at some output power will be screened, and the reliability of the devices will be classified according to the value of variance. The result of aging is statistic, and the time is long. During the aging, the lifetime of the reliable devices is also affected because of the hot and electric impulsion. It can cause batches of devices damage when some accidence happens, such as the uncontrolled system or power cut. On the other hand, electric aging is not sensitive for potential defects, for example, some devices after electric aging screening which are reliable are usually found fail rapidly during using. Thus, it is significant to find a kind of a fast, nondestructive, convenient method to test the devices and estimate their reliability.When the semiconductor laser is operating in a constant bias current, the terminal voltage of the device often fluctuates. The fluctuation that is named electric noise indicates the imperfects of material or device, especially; it is very sensitive for various reliability-dependent defects than other parameters. The noise test may estimate the quality and the reliability of the single device. Moreover, the voltage and current added on the device is near or under the normal stress level of the device, so that it is nondestructive to the device. So the electric noise is becoming a useful tool to characterize device quality and reliability. Thus, it is very significant for the preparing, manufacturing, using the devices to apply the fast, sensitive, nondestructive method to the test the devices and estimate their reliability.First, the electric noise characteristic of semiconductor lasers is reviewed, including the kind, character, produce mechanism, ect. Secondly, the measuring methods of noise in semiconductor lasers are introduced. Based on direct measuring and cross-spectrum measuring methods, we construct the measuring system of noise in semiconductor lasers. Next, the work on noise used to characterize quality and reliability of semiconductor lasers is first studied. The main research contents of this paper are shown hereinafter:1. The devices used in the experiment are 808nm GaAs/GaAlAs semiconductor laser diodes. Their epitaxial layer are grown by MOCVD technique. The electric derivative and the low frequency electric noise of devices are measured and the aging test is made. The results indicate the noise at the low injecting current reflects the inner defects of the device, which is related with the quality and reliability of the device. The aging current is increased gradually during aging and the noise is measured after aging at a time. By contrasting the results, we find the turning point of the noise value from the descending to the ascending, namely the maximum of the bias current added on the device when aging. Using the maximum may avoid inducing the new defects which effect on the quality and reliability of the device during aging. At the same time, 1/f noise, g-r noise and burst noise are studied under the condition of the accelerating aging. The results indict that g-r noise and burst noise is related with the value of the bias current added on the device when aging, and foretell that g-r noise and burst noise are possibly the evident symbol of the device degeneration.2. Wavelet is used to ana...
|
PDF Full Text Request