An FM transistor radio owned by one of us (L. E.) since the early 1960s still works beautifully. Reasoning that this was testament to the performance of its single Esaki diode, we tested the effects of storage on some of these germanium devices made in 1960.
The Esaki diode (L. Esaki Phys. Rev. 109, 603–604; 1958) was the first quantum-electron device. Unlike the mechanism that powers most semiconductor devices, current flows through the diode as a result of quantum-mechanical electron tunnelling across a potential barrier.
Semiconductor transport devices are extremely stable, so their shelf-life should be infinite if they are stored at room temperature. But the Esaki diode's tunnel current is very sensitive to its enormous built-in electric field in the junction region (E. Spenke Electronic Semiconductors 232; McGraw-Hill, 1958), which could affect its long-term performance.
As the most likely indicator of any small structural changes in the device, we re-measured the peak current in 20 devices and discovered that it had fallen by an average of just 3.3% over 50 years, corresponding to a junction widening of only 0.25%.
This very tiny shift in electronic characteristics is probably down to inbuilt impurities and imperfections within the structure. A gratifying confirmation of the diode's longevity, nonetheless.
Author information
Authors and Affiliations
Additional information
Contributions to Correspondence may be submitted to correspondence@nature.com . Please see the Guide to Authors at http://go.nature.com/cMCHno . Science publishing issues that may be of interest to authors are regularly featured at our blog Nautilus ( http://blogs.nature.com/nautilus ), where we welcome comments and debate.
Rights and permissions
About this article
Cite this article
Esaki, L., Arakawa, Y. & Kitamura, M. Esaki diode is still a radio star, half a century on. Nature 464, 31 (2010). https://doi.org/10.1038/464031b
Published:
Issue Date:
DOI: https://doi.org/10.1038/464031b
This article is cited by
-
Smooth statistical modeling of bivariate non-monotonic data by a three-stage LUT neural system
Neural Computing and Applications (2018)