Printer Friendly

Silicon-germanium for electronics and optoelectronics.

Germanium has given silicon a new boost, enabling transistors to work at transition frequencies as high as 75 GHz

Silicon has been a singularly successful semiconductor for electronics. Its intrinsic chemical property, namely the ability to form a multi-purpose stable insulating oxide, had been a central advantage in the development of the "planar process" which has advanced to a level such that several million complex transistors can be "written into" a tiny piece of silicon today.

Although the pace of silicon integrated circuits development had been incredible in the past four decades, it has received a new boost from an "old defeated rival", namely germanium. It had recently been shown (by researchers from IBM) |1~ that a transistor incorporating a silicon-germanium alloy can work at transition frequencies as high as 75 GHz (and |is greater than~ 90 GHz at 77K). This high a range of frequencies was thought to be well beyond the reach of silicon.

The advent of Silicon-Germanium offers numerous interesting possibilities involving heterostructures since the alloy has a smaller bandgap compared with silicon. Thus for the first time "bandgap engineering" is possible in silicon-based technology. In view of the smaller bandgap of SiGe, one can hope to build various infra-red radiation detectors.

Although there appears to be evidence for weak light emission from silicon-based structures most recently, it has yet to be studied in detail. Thus the silicon-germanium based devices offer very promising features for high-speed electronics and optoelectronics.

The research at Waterloo in the area of silicon-germanium devices has been concentrated on the development and study of novel low-temperature deposition techniques and high-dose ion implantation to create SiGe alloy in a localized region. The first SiGe transistors made at the IBM Corporation were realized using expensive molecular beam epitaxy, but since then a number of interesting new techniques have been developed, such as the ultra-high vacuum chemical vapor deposition (UHV/CVD) |2~ and Rapid Thermal CVD |3~.

We have developed at Waterloo a high-dose germanium ion implantation to create a novel SiGe-channel MOSFET transistor |4~. The transistors realized with this high-dose germanium ion implantation have shown significant gains. The main advantage of this implantation approach is that one can create the new SiGe-channel transistor on the same chip containing conventional Si-channel transistors. The schematic processing steps are shown in Fig. 1.

The observed transconductance gain improvements of the SiGe transistors having the same mask dimensions as the Si-channel transistors on the same chip were in the range of 40 to 70%. The experimentally observed transistor characteristics are shown in Fig. 2.

Several issues about the recrystallization of the implanted region and their effect on the devices are being studied. For example, it has been experimentally observed that higher quality material can be realized in smaller areas than in larger areas. The germanium profile can influence the recrystallization. Secondary ion mass spectrometry had been used for assessing the distribution of germanium within silicon. The oxide grown on SiGe layers have somewhat poorer quality in terms of the interface state densities, and methods of improving it are being studied. A second approach we are investigating is a low-temperature vacuum evaporation technique for the realization of SiGe layers. The technique of ion-beam-assisted deposition seems to offer significant potential. We have been able to grow good quality SiGe layers using this technique. Since the low energy bombardment of the growing SiGe film improves the epitaxial alignment, it is an attractive low temperature technique.

The films grown in Waterloo are analyzed using Raman Spectroscopy, X-ray diffraction and Rutherford Backscattering spectra. Most recently obtained Rutherford Backscattering spectra indicate clear evidence of enhanced epitaxial alignment. Since our emphasis had been on devices, we attempted to make a heterostructure diode and we observed near-ideal exponential characteristics over four decades of current.

Apart from these experimental studies, D. Roulston and A. Nathan, electrical and computer engineering department have been investigating theoretical issues such as calculating mobility, bandstructure and transistor optimization |5,6~.

Reference

1. G.L. Patton et al, "75-GHz fT SiGe-base heterojunction bipolar transistors," IEEE Electron Device Lett. 11, 175 (1990).

2. B.S. Meyerson, "Low Temperature Silicon Epitaxy by Ultra High Vacuum Chemical Vapor Deposition," Appl.Phys. Lett., 48(12), 797-799 (1986).

3. J.F. Gibbons, C.M. Gronet and K.E. Williams, "Limited Reaction Processing: Silicon Epitaxy", Appl. Phys. Lett., 47, 721-723 (1985).

4. C.R. Selvakumar and B. Hecht, "SiGe-Channel n-MOSFET by Germanium Implantation", IEEE Electron Device Lett., 12(8) (1991).

5. T. Manku and A. Nathan, "Energy-band structure for strained p-type Si1-xGex" Phys. Rev. B., 15(15), 12634-12637 (1991).

6. J.M. McGregor, T. Manku, and D.J. Roulston, "Bipolar Transistor Base Grading for Minimum Delay", Solid-St. Electron., 34, 421 (1991).
COPYRIGHT 1992 Chemical Institute of Canada
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Selvakumar, C.R.
Publication:Canadian Chemical News
Date:Nov 1, 1992
Words:781
Previous Article:Chemical sensing with zeolite materials.
Next Article:Turning molecules into metals: 20th century alchemy.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters