SI CHINA     SI JAPAN
Login  |  Register          Free Newsletter Subscription
Subscribe
Email
Print
Reprint
Learn RSS

AIGaAs Finds an Oxide

Ruth DeJule, Associate Editor -- Semiconductor International, 8/1/1999

The IC manufacturing community takes SiO2 for granted. But in the 1950s, some of the first diffused silicon transistors and p-n-p-n switches built at Bell Labs used no oxides. Wafers were diffused using dry gases, and in the process many were destroyed. In the spring of 1955, during one such diffusion, an accident caused hydrogen to flash back into the furnace, forming water vapor, and from the tube emerged silicon wafers, green and pink with grown oxides. Ironically, a similar accident in 1990 resulted in the discovery of an oxide for III-V semiconductors.

Nick Holonyak Jr., the John Bardeen Professor of Electrical and Compu ter Engineering and Physics at the University of Illinois, and graduate student John Dallesasse were investigating the effects of moisture degradation on crystals containing aluminum gallium arsenide (AlGaAs) layers. They observed that at high aluminum compositions, >60%, and at high humidities, the Al structures crumbled. In an effort to push the hydrolization process, temperatures and humidity were raised using a simple quartz tube placed in a furnace. But at temperatures above 400°C, instead of destroying the crystal, a smooth, solid 'native oxide' was formed. It was later confirmed in ceramics textbooks that a phase boundary had been crossed. 'Prior to this discovery, there was no known method for forming useful oxides in AlGaAs or similar III-V materials,' Holonyak said.

A cross section of an AIGaAs quantum well heterostructure laser diode, image (a) shows a 2.5 m n oxide-defined pair of buried apertures. Applied current cannot penetrate the oxidized regions (Ox), thus providing a high degree of lateral optical confinement in the QW active region. Inset (b), a top view, shows that the oxidized region -- oxidized from the left and the right edges -- appears translucent.

These native oxides are homoepitaxial; they can be formed on top of or inside layers. Therefore, the oxides can be grown edgewise, forming a buried oxide layer. Fortuitously, instead of expanding and forming cracks, the native oxide contracts a bit. Reliability issues appear to occur with abrupt or step changes. Therefore, grading the Al composition of the AlGaAs layer by even a couple of percent can reduce stress at the interface and limit dark line defect formation that can lead to catastrophic failure.

Holonyak and his graduate students first used this oxide in edge-emitting laser diodes, forming a buried linear aperture. But the most popular application has been in vertical-cavity, surface-emitting lasers (VCSEL), which can serve as an optical interconnect for high-speed data communication, a $2B market. Unlike conventional edge-emitting laser diodes (the kind used in compact disc players and laser pointers, for example), a VCSEL's optical beam is perpendicular to the chip surface. This simplifies device fabrication and testing, and creates smaller structures, thus consuming less power. 'Research performed in various labs has shown that the U. of I. oxidation process makes the smallest, most efficient and highest performance VCSELs to date,' Holonyak said.

The strength of the oxidation process, Holonyak said, is its ability to selectively oxidize layers of AlGaAs buried deep within the device structure, creating an insulating 'collar' around a VCSEL's conducting cavity. The oxide collar very effectively defines the electromagnetic field and confines the current within the aperture. The collar also controls the geometry of the optical beam, making it easier to couple the light into optical fibers for data transmission.

Inherent in VCSEL technology is the ability to make arrays, opening the possibility of forming integrated structures. Essentially any structure grown on a GaAs substrate containing high percentage AlGaAs layers can be oxidized. 'But it's now a question of the imagination of people and the time they are willing to invest,' said Holonyak. 'I don't foresee a revolution, rather an evolution based on development and study.'

Email
Print
Reprint
Learn RSS

Talkback

We would love your feedback!

Post a comment

» VIEW ALL TALKBACK THREADS

Related Content

Related Content

 

By This Author

SPONSORED LINKS



 
Advertisement
SPONSORED LINKS

More Content

  • Blogs
  • Podcasts
  • Videos

Blogs

Videos

Advertisements





NEWSLETTERS
Plug in and get the latest SI news, trends and industry updates delivered free, directly to your inbox!

SI NewsBreak and Special Reports (Weekdays)
Wafer Processing Report (Monthly)
Lithography Report (Monthly)
Metrology Report (Monthly)
Clean Processing Report (Monthly)
Packaging Report (Twice Monthly)
©2008 Reed Business Information, a division of Reed Elsevier Inc. All rights reserved.
Use of this Web site is subject to its Terms of Use | Privacy Policy
Please visit these other Reed Business sites