DocumentCode :
789289
Title :
A Prediction and Selection System for Radiation Effects in Planar Transistors
Author :
Poch, W. ; Holmes-Siedle, A.G.
Author_Institution :
RCA Astro-Electronics Division Princeton, New Jersey
Volume :
15
Issue :
6
fYear :
1968
Firstpage :
213
Lastpage :
223
Abstract :
Radiation tests on a wide range of planar silicon bipolar transistors under high-energy radiation indicate that the degradation of gain due to the ionizing component of radiation is due to a "surface effect" namely, an increase in the surface recombination velocity near the emitter-base junction. This effect is dominant in many real-life operating environments encountered by space electronic systems, and can produce quite serious losses of gain. The authors have devised a practical prediction technique for the surface effect and applied it to a wide range of space-approved bipolar transistor devices. Specifications of the expected levels of surface damage, devised from such data, have only rarely been exceeded in subsequent sampling tests. The prediction technique requires data from a Cobalt-60 gamma-ray test of 10 to 20 samples of each device type. Progress in using this prediction technique is reported. During testing of many devices under Cobalt-60 gamma radiation, infrequent cases have been noted where the surface damage levels exceed those predicted. Some devices are anomalously sensitive, with damage levels falling well outside the normal statistical distribution of sensitivity. It is thus important to discover a method of detecting these rare, but excessively sensitive units. The authors describe a detection or "preselection" method for such anomalously sensitive devices. The conclusion of the investigation is that the "irradiate-anneal" form of component preselection cycle can produce, from a commercial batch of planar transistors, a set of devices possessing a uniform, minimized sensitivity to ionizing radiation.
Keywords :
Bipolar transistors; Degradation; Gamma rays; Ionizing radiation; Radiation effects; Sampling methods; Silicon; Spontaneous emission; Statistical distributions; Testing;
fLanguage :
English
Journal_Title :
Nuclear Science, IEEE Transactions on
Publisher :
ieee
ISSN :
0018-9499
Type :
jour
DOI :
10.1109/TNS.1968.4325050
Filename :
4325050
Link To Document :
بازگشت