Transcript Slide 1

Single-electron Devices
Speaker: Qiaoyan Yu
[email protected]
ECE423
12-16-2006
Agenda
Milestone
 Motivations
 New characteristics
 Single-electron transistor (SET)
 Challenges
 Conclusions

2
Milestone
In 1909, Millikan first illustrated the
manipulation of single electrons.
 In 1985, Dmitri Averin & Konstantin
Likharev proposed single-electron
transistor.
 In 1987, Theodore Fulton & Gerald Doald
at Bell lab fabricated and demonstrated
SET.

3
Motivation

Power consumption challenges high
performance and high density chip design


Faster and more information processing
resulting in generating more heat flux
Reduce the corresponding charge per bit!
Single-electron Device!!!
4
SED new characteristics

Exploit the quantum effect of tunneling


Charge doest not flow continuously


Control and measure the movement of single
electron
Instead in a quantized way
Extremely high charge sensitivity

High precise for charge measurement
5
Single-electron transistor (SET)
Architecture
 Equivalent circuit
 Operation principle
 I-V curve
 SET VS. MOSFET

6
Architecture
Tunnel junctions
q
island
Source
+q1
Vb
V2
V1
-q1
+q2
C1
gate
Drain
-q2
C2
Cg
Vg
7
Equivalent circuit
8
Operation principle
9
Operation principle
10
Operation principle
11
I-V curve
12
SET vs. MOSFET

Structure


Size


Periodic vs. not periodic
Sensitivity


Coulomb blockade vs. electron diffusion
Threshold voltage & source-drain current


Extremely small vs. large (although scaled down)
Main physical principle


Two tunneling barrier vs. inversion channel
High vs. low (10000X)
Power

Low vs. high
13
SED challenges
Difficult to precisely control the device
implementation
 Characteristic of the device varies
significantly from location to location
 Suffer from “offset charges”


Randomly fluctuation of control signal
14
Conclusions





Because of natural small dimension, SED is a
potential solution for continue silicon scaling.
It is not clear that SET replaces FET.
It is certain that quantum properties of electrons
will be crucial in the design of electron devices.
Electron beam lithography and scanning probe
techniques offer the best prospects for the future.
Some more esoteric techniques based on atomic
particle deposition and colloid chemistry may also
provide some benefits.
15
The End
Thank You!
Any questions?
16