Project home page: http://www.sp.phy.cam.ac.uk/~dp109/SIQUIC.html
The prime objective of the project is to use a present silicon CMOS production facility to fabricate quantum devices which have the potential for ultra-high speed and functional operation. The devices are resonant tunnelling diodes (RTDs) and velocity modulation transistors (VMTs).
The diminutive sizes in the direction of carrier transport of RTDs allows ultra-high speed operation. RTDs in the III-V systems have already shown oscillations in sub-millimetre wave frequencies (710GHz). Negative differential conductivity allows a large number of high speed functional devices to be designed. For example, signal processing circuits with significantly reduced number of devices (and hence smaller real-estate on chips) and multiple valued memory cells using RTDs have been proposed and demonstrated in III-V systems. These functional applications are highly promising since RTDs with their simple structure and small size may easily be integrated with conventional devices such as field effect transistors and bipolar devices. We propose to demonstrate RTDs in Si/SiGe and also one of the simplest functional uses, a cascade RTD memory. A novel gating technique will be investigated to electrostatically squeeze RTDs, creating single electron transistors.
Velocity modulation transistors (VMTs) are novel, double-layer systems where manipulation of the wave function allows electrons to be switched between spatially separated channels of different mobilities, potentially offering ultra high speed operation as the switching speed is not limited by the gate source capacitance. This is a device which offers one possible method of surmounting the ultimate capacitive speed limitation imposed on all field effect transistors. Front- and back-gates are used to move the electrons from one channel to the other without charging. The electrons have either high or low velocity depending on the mobility of the quantum well they are in and this provides the two distinct states for transistor action. The VMT is essentially a normal MODFET on top of an inverted MODFET with a difference in mobility between the two structures and a small energy barrier between the two quantum wells.
The results from the manufactured devices combined with theoretical modelling of such structures will demonstrate the potential for implementation of each device on present silicon production lines. A four pronged collaborative approach is proposed with work on
If successful, it would allow industry to increase performance by changing the architectures of the chips in the present silicon fabrication plants rather than scaling the present architectures.
Dr. Douglas Paul
University of Cambridge
Cavendish Laboratory - Dept. of Physics
UK - Cambridge - CB3 0HE
e-mail: (E-mail removed)
1 January 97
Duration: 36 months
Welcome to the new
Information Society Technologies Programme (IST)
Our new activities in the IST Programme:
Future & Emerging Technologies
This document is located at /esprit/src/22987.htm
It was last updated on 1 July 1999, and is maintained by (E-mail removed)