AC Performance of Nano - Electronics
AC Performance of Nano - Electronics
Burke
Peter J. Burke
Department of Electrical Engineering
Department of Electrical Engineering
and Computer Science
and Computer Science
University of California, Irvine
University of California, Irvine
AC performance
AC performance
of nano
of nano
-
-
electronics
electronics
Funding: ARO, ONR, NSF, DARPA
Students:
Zhen Yu
Shengdong Li
Chris Rutherglen
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 2 2
Outline
Outline
Motivation
Motivation
Theory of nanotubes/nanowires
Theory of nanotubes/nanowires
Synthesis
Synthesis
AC performance
AC performance
Towards
Towards
integrated nanosystems
integrated nanosystems
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 3 3
Motivation
Motivation
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 4 4
Moores law:
Moores law:
Top down approach
Top down approach
From Gordon Moore 2003 ISSC conference talk
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 5 5
Length scales
Length scales
1 A 1 nm 10 nm 100 nm 10 m 1 m
Atom
DNA Protein Virus Cell
Nanotube Smallest
lithographic
resolution:
Electron beam
lithography (10 nm)
2003
Si transistor
(0.13 m)
1970
Si transistor
(10 m)
T
o
p
d
o
w
n
B
o
t
t
o
m
u
p
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 6 6
Highest resolution lithography:
Highest resolution lithography:
Expensive
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 7 7
Theory of nanotubes/nanowires
Theory of nanotubes/nanowires
Theory of carbon nanotubes
Theory of carbon nanotubes
? R =
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 9 9
What is 1d?
What is 1d?
2 options
2 options
Geometry:
Geometry:
Length >> Width
Length >> Width
Quantum
Quantum
mechanics:
mechanics:
Width ~
Width ~
de de Broglie Broglie
In semiconductors,
Fermi
~ 1-10 nm
d ~
Fermi
In metals,
Fermi
~ 0.1 nm
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 10 10
Ballistic vs. diffusive transport
Ballistic vs. diffusive transport
2
12
2
h
R k
e
=
W
W
l
l
L
Ballistic
Diffusive
2
L
R
W
=
2
1 1
12
2
h
R k
N e N
=
Single mode
Multi mode
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 11 11
Landauer
Landauer
formula:
formula:
2
2e
G n
h
=
If the leads are not perfect injectors into each channel then:
2
2
n
e
G T
h
=
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 12 12
Resistance quantum
Resistance quantum
right goers
e
I dE
h
=
Ballistic conductor
Right
contact
Left
contact
right goers
e
I dE
h
=
right goers left goers
e
I dE dE
h
=
E
k
x
=2n
x
/L
x
eV
Fermi energy
[ ]
2
( )
F F
e e
I E eV E V
h h
= + =
2
quantum
h
V I IR
e
= =
2
25
quantum
h
R k
e
= =
2
12.5
2
quantum
h
R k
e
= =
With spin:
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 13 13
Quantum point contact
Quantum point contact
B.J. van Wees et al. (1988), Phys. Rev. Lett., 60, 848.
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 14 14
Graphite: 2d semiconductor
Graphite: 2d semiconductor
Bond length = 0.142 nm
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 15 15
Graphite band structure
Graphite band structure
E
k
x
k
y
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 16 16
Nanotube:
Nanotube:
Zig
Zig
-
-
zag
zag
vs. armchair
vs. armchair
x
y
Armchair
Zig-zag
In between:
Chiral
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 17 17
Nanotube: (
Nanotube: (
n,m
n,m
) description
) description
x
y
a
1 a
2
6 a
1
+ 2 a
2
(n,m) = (6,2) In this example:
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 18 18
k
k
-
-
vector
vector
Graphite:
Graphite:
Arbitratrary
Arbitratrary
k
k
x x
,
,
k
k
y y
allowed
allowed
Nanotube:
Nanotube:
(
(
)=(+2)
)=(+2)
k
k
perp perp
spaced by 2/d
spaced by 2/d
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 19 19
Allowed k
Allowed k
-
-
vectors:
vectors:
k
x
k
y
2/d
NT axis
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 20 20
k
k
-
-
space
space
k
x
k
y
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 21 21
(9,0) armchair nanotube
(9,0) armchair nanotube
k
x
k
y
2/d
NT axis
All armchairs are metallic.
2 2
2 4
n
e e
G T
h h
= =
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 22 22
Semiconducting nanotube
Semiconducting nanotube
k
x
k
y
2/d
NT axis
1
[ ]
eV
bandgap
d nm
=
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 23 23
Electrical properties
Electrical properties
All armchair metallic
All armchair metallic
33% of
33% of
zig
zig
-
-
zag
zag
metallic
metallic
Semiconducting tubes:
Semiconducting tubes:
Gap = 0.9
Gap = 0.9
eV/d[nm
eV/d[nm
]
]
Statistically, 9/10 metallic
Statistically, 9/10 metallic
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 24 24
AC nanosystems?
AC nanosystems?
R
R
quantum
quantum
= h/e
= h/e
2
2
= 25 k
= 25 k
Z
Z
quantum
quantum
(
(
)= ?
)= ?
Inductor
Inductor
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 25 25
Scaling down to nano
Scaling down to nano
1 1 Oscillator Q Oscillator Q
Good Good Moderate Moderate Poor Poor
10e 10e- -8 A 8 A 2 2 Current Current
10e 10e- -21 J 21 J 3 3 Electrostatic energy Electrostatic energy
10e 10e- -8 W/K 8 W/K 1 1 Thermal conductance Thermal conductance
10e 10e- -13 s 13 s 2 2 Thermal time constant Thermal time constant
10e 10e- -21 J/K 21 J/K 3 3 Heat capacity Heat capacity
>10e18 >10e18 rad/s rad/s - -1 1 Elect. Elect. oscill oscill. frequency . frequency
0 0 Capacitive time constant Capacitive time constant
<10e <10e- -16 s 16 s 2 2 Inductive time constant Inductive time constant
10e 10e- -15 H 15 H 1 1 Inductance Inductance
10e 10e- -23 N 23 N 4 4 Magnetic force Magnetic force
10e 10e- -6 T 6 T 1 1 Magnetic field Magnetic field
10e 10e- -20 F 20 F 1 1 Capacitance Capacitance
10 ohm 10 ohm - -1 1 Resistance Resistance
10e 10e- -12 N 12 N 2 2 Electrostatic force Electrostatic force
1 V 1 V 1 1 Voltage Voltage
100 m/s 100 m/s - -3/2 3/2 Thermal speed Thermal speed
1 1 Wear life Wear life
2 2 Frictional force Frictional force
10e6 N/m^2 10e6 N/m^2 - -1 1 Viscous stress Viscous stress
10e19 W/m^3 10e19 W/m^3 - -1 1 Power density Power density
10e 10e- -8 W 8 W 2 2 Power Power
10e 10e- -12 to 10^ 12 to 10^- -9 s 9 s - -1 1 Motion time Motion time
10e3 m/s 10e3 m/s 0 0 Stress Stress- -limited speed limited speed
10e13 10e13 rad/s rad/s - -1 1 Vibrational Frequency Vibrational Frequency
10e15 m/s^2 10e15 m/s^2 - -1 1 Acceleration Acceleration
10e 10e- -24 kg 24 kg 3 3 Mass Mass
10e 10e- -11 m 11 m 1 1 Deformation Deformation
1000 N/m 1000 N/m 1 1 Stiffness Stiffness
10e 10e- -8 N/m^2 8 N/m^2 2 2 Force Force
10e 10e- -18 m^2 18 m^2 2 2 Area Area
Bad Bad Typical magnitude Typical magnitude Exponent Exponent Quantity Quantity
adapted from:
Nanosystems: Molecular Machinery,
Manufacturing, and Computation
K. Eric Drexler
Wiley, 1992
Different classical
models fail at different
length scales, with the
most dramatic failures
appearing in AC
electrical circuits.
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 26 26
1 channel spinless electrons
1 channel spinless electrons
RF circuit model:
RF circuit model:
A transmission line
A transmission line
Geometry
Geometry
h
d
L
kinetic
L
magnetic
C
quantum
C
electrostatic
An RF Circuit Model for Carbon Nanotubes, P.J. Burke IEEE Transactions on Nanotechnology 2(1), 55-58 (2003)
Luttinger Liquid Theory as a Model of the GHz Electrical Properties of Carbon Nanotubes, P.J. Burke, IEEE Transactions
on Nanotechnology 1(3), 129-144 (2002)
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 27 27
Quantum impedances:
Quantum impedances:
Characteristic
Characteristic
impedance =
impedance =
Sqrt(L/C) =
Sqrt(L/C) =
h/2e
h/2e
2 2
= 12.5 k
= 12.5 k
Wave velocity =
Wave velocity =
Sqrt(1/LC) =
Sqrt(1/LC) =
v
v
Fermi Fermi
= 8 10
= 8 10
5 5
m/s
m/s
~c/100
~c/100
Kinetic inductance = 16 nH/
Kinetic inductance = 16 nH/
m
m
Quantum capacitance = 100 aF/
Quantum capacitance = 100 aF/
m
m
Electrostatic capacitance = 50 aF/
Electrostatic capacitance = 50 aF/
m
m
If we neglect electrostatic capacitance:
If we neglect electrostatic capacitance:
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 28 28
Carbon nanotube RF circuit
Carbon nanotube RF circuit
without e
without e
-
-
e interactions:
e interactions:
Four channels in
Four channels in
parallel:
parallel:
x2 for spin
x2 for spin
x2 from band
x2 from band
structure
structure
Neglect e
Neglect e
-
-
e
e
interaction
interaction
neglect
neglect
electrostatic
electrostatic
capacitance
capacitance
L
kinetic
C
quantum
L
kinetic
C
quantum
L
kinetic
C
quantum
L
kinetic
C
quantum
spin up a spin up a
spin down a spin down a
spin up b spin up b
spin down b spin down b
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 29 29
Carbon nanotube RF circuit
Carbon nanotube RF circuit
with e
with e
-
-
e interactions:
e interactions:
Include e
Include e
-
-
e
e
interaction
interaction
electrostatic electrostatic
capacitance capacitance
4 channels
4 channels
coupled
coupled
Spin/charge
Spin/charge
separation:
separation:
Common mode = Common mode =
Charge mode Charge mode
Differential modes (3)= Differential modes (3)=
Spin modes Spin modes
spin up a spin up a
spin down a spin down a
spin up b spin up b
spin down b spin down b
L
kinetic
C
quantum
L
kinetic
C
quantum
L
kinetic
C
quantum
L
kinetic
C
quantum
C
electrostatic
An RF Circuit Model for Carbon Nanotubes, P.J. Burke IEEE Transactions on Nanotechnology 2(1), 55-58 (2003)
Luttinger Liquid Theory as a Model of the GHz Electrical Properties of Carbon Nanotubes, P.J. Burke, IEEE Transactions
on Nanotechnology 1(3), 129-144 (2002)
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 30 30
Spin charge separation:
Spin charge separation:
spin up a spin up a
spin down a spin down a
spin up b spin up b
spin down b spin down b
L
kinetic
C
quantum
L
kinetic
C
quantum
L
kinetic
C
quantum
L
kinetic
C
quantum
C
electrostatic
Wave velocity
Wave velocity
= Sqrt(1/LC)
= Sqrt(1/LC)
Charge mode:
Charge mode:
C
C
-
-
1
1
= C
= C
quantum
quantum
-
-
1
1
+ C
+ C
electrostatic
electrostatic
-
-
1
1
Spin mode:
Spin mode:
C =
C =
C
C
quantum
quantum
Contact resistance
Contact resistance
= 0.5 h/e
= 0.5 h/e
2 2
per end
per end
(Buttiker calls charge
(Buttiker calls charge
relaxation resistance)
relaxation resistance)
L
kinetic
L
magnetic
C
quantum
C
electrostatic
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 33 33
Synthesis
Synthesis
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 34 34
Synthesis:
Synthesis:
The current paradigm
The current paradigm
substrate
Oven 1000 C
feedstock gas, e.g.:
CH
4
for carbon nanotubes
SiH
4
for Si nanowires
catalyst on a chip
gas
nanoparticle catalyst
Carbon nanotube
Systems are commercially available now, e.g.:
Nanodevices, Atomate
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 35 35
Materials systems
Materials systems
Nanotubes
Nanotubes
B
B
Semiconductor nanowires
Semiconductor nanowires
Si
Si
III
III
-
-
V
V
II
II
-
-
VI
VI
Metal nanowires
Metal nanowires
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 36 36
Fabrication techniques:
Fabrication techniques:
Two options
Two options
Define catalysts before growth
Define catalysts before growth
In
In
-
-
situ electrical contact possible
situ electrical contact possible
A
)
-4 -3 -2 -1 0
Vds (V)
Gate = + 7.7 V
Gate = -5 V
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 57 57
Operating principles:
Operating principles:
Javey, Guo, Wang, Lundstrom, Dai, Ballistic Carbon Nanotube Field-Effect
Transistors, Nature, 424, 2003.
Ballistic wires:
Schottky barrier modulation
Appropriate for short carbon nanotubes
Diffusive wires:
Charge density modulation
Appropriate for long carbon nanotubes,
semiconducting nanowires?
Many details still not known:
Electrostatics
Quantum 1d effects
Finite frequency
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 58 58
Depletion Curve of a
Depletion Curve of a
Semiconducting CNT
Semiconducting CNT
5
4
3
2
1
0
C
o
n
d
u
c
t
a
n
c
e
(
S
)
-8 -6 -4 -2 0 2 4 6 8
Gate Voltage (V)
300
250
200
150
100
50
0
G
(
n
S
)
1.0 0.8 0.6 0.4 0.2 0.0
Back gate voltage (V)
-2
0
2
C
u
r
r
e
n
t
(
n
A
)
-10 0 10
Voltage (mV)
Work of Sheldon Li
T = 4 K T = 300 K
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 59 59
Depletion Curve of a
Depletion Curve of a
Metallic CNT
Metallic CNT
4
3
2
1
0
G
(
S
)
-6 -4 -2 0 2
Back gate voltage (V)
-50
0
50
C
u
r
r
e
n
t
(
n
A
)
-10 0 10
Voltage (mV)
Work of Sheldon Li
T = 4 K
gate (back)
source
drain
insulator
gate (top)
L
oxide
Typical geometry
Typical geometry
S.J. Wind, J. Appenzeller, R. Martel, V. Derycke, Ph. Avouris, Vertical Scaling
of Carbon Nanotube Field-Effect Transistors Using Top Gate Electrodes,
Applied Physics Letters, 80(20), 3817-3819 (2002).
A. Javey, H. Kim, M. Brink, Q. Wang, A. Ural, J. Guo, P. Mcintyre, P. McEuen, M. Lundstrom, H.
Dai, High-K dielectrics for advanced carbon-nanotube transistors and logic gates, Nature
Materials, 1, p. 241-246, (2002).
DC performance demonstrated in:
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 61 61
Top gated device
Top gated device
T= 4 K
S.Li, Z. Yu, P Burke, Journal of Vacuum Science and Technology B: Nanotechnology (in press, 2004).
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 62 62
Geometry variations
Geometry variations
gate (back)
source drain
insulator
gate (top)
oxide
gate (back)
source drain
ins
gate
oxide
gate (back)
source drain
ins
gate
oxide
ins
gate
G1 G2 G3
Source Drain
N
a
n
o
w
i
r
e
N
a
n
o
w
i
r
e
N
a
n
o
w
i
e
Nanowire
Radial and longitudinal band gap engineering
recently demonstrated.
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 63 63
Work still needed on:
Work still needed on:
Electrostatics
Quantum 1d effects
Finite frequencies
p type n type
-
-
-
-
-
-
+
+
+
+
+
+
Electrostatics example: p-n junction
p type n type
Traditional
Nano p-n
+
-
Fringe field lines extend way out.
See, e.g, Leonard, Tersoff, Novel Length Scales in
Nanotube Devices, Physical Review Letters, 83 (1999).
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 64 64
Advantages
Advantages
Small diameter ~ 1 Small diameter ~ 1- -3 nm 3 nm
Demonstrated Demonstrated
- ->10 >10
12 12
/cm /cm
2 2
density (long term goal) density (long term goal)
Not listed in ITRS 03 Not listed in ITRS 03
Interconnects Interconnects
Demonstrated Demonstrated
Ultra Ultra- -low parasitic capacitance (10 low parasitic capacitance (10
- -16 16
F/ F/ m) m)
Useful for molecular, single electron memory Useful for molecular, single electron memory
Conductivity for SWNT 10x higher than copper demonstrated Conductivity for SWNT 10x higher than copper demonstrated
Not listed in ITRS 03 Not listed in ITRS 03
Speed/mobility Speed/mobility
Research topic Research topic
High High- -mobility: demonstrated, interpret with caution mobility: demonstrated, interpret with caution
Ballistic transport: demonstrated, why advantage? Ballistic transport: demonstrated, why advantage?
Listed in ITRS 03 Listed in ITRS 03
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 65 65
Disadvantages
Disadvantages
Current synthesis paradigm not scalable
Current synthesis paradigm not scalable
\
|
|
|
.
|
\
|
=
|
|
.
|
\
|
d
g
gs m
d
g
v
v
v g
i
i
0
0 0
g g
v i ,
d d
v i ,
s s
v i ,
|
|
|
.
|
\
|
|
|
|
.
|
\
|
=
|
|
|
.
|
\
|
g
d
s
g
d
s
v
v
v
i
i
i
. . .
. . .
. . .
In general:
Y-matrix has 9 elements, but once you
know 4 you know them all because:
g d s
i i i + =
Only need:
|
|
.
|
\
|
|
|
.
|
\
|
=
|
|
.
|
\
|
d
g
d
g
v
i
h h
h h
i
v
22 21
12 11
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 69 69
Semi-empirical predication for f
t
Frequency, GHz
G
a
i
n
,
d
B
0
20
1
10
Predicted h
21
f
T
= 3 GHz
1 1 10 5 .5
Cutoff frequency
GHz 3
2
(measured) 1
d) (calculate aF 50
) / ln(
2
= =
=
= =
c
g
f
S g
d h
c
T
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 70 70
Measurement techniques
Measurement techniques
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 71 71
S
S
-
-
matrix
matrix
gate
drain
source
+
1
V
) (
2
) (
2 2
) , (
t kx i t kx i
e V e V t x V
+
+ =
) (
1
) (
1 1
) , (
t kx i t kx i
e V e V t x V
+
+ =
1
V
+
2
V
2
V
|
|
.
|
\
|
|
|
.
|
\
|
=
|
|
.
|
\
|
+
+
2
1
22 21
12 11
2
1
V
V
S S
S S
V
V
Work of Chris Rutherglen
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 72 72
h parameters
h parameters
( )( )
( )( )
( )( )
( )( )
( )( )
( )( )
21 12 22 11
21 12 22 11
'
22
21 12 22 11
12
'
21
21 12 22 11
12
'
12
21 12 22 11
21 12 22 11
'
11
1 1
1 1
1 1
2
1 1
2
1 1
1 1
S S S S
S S S S
h
S S S S
S
h
S S S S
S
h
S S S S
S S S S
h
+ +
+
=
+ +
=
+ +
=
+ +
+ +
=
22 21
12 11
S S
S S
S-Parameter
Work of Chris Rutherglen
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 73 73
Mobility/Speed
Mobility/Speed
2
T
g
f
C
=
( )
th g
c V V
g
L
=
2
( )
2
th g
T
V V
f
L
=
Diffusive, low electric field:
2
sat
T
v
f
L
=
Diffusive, high electric field:
2
100
[ ]
T
THz
f
L m
=
For Fuhrers 200,000 cm
2
/V-s
For Liebers 1,300 cm
2
/V-s
2
650
[ ]
T
GHz
f
L m
=
130
[ ]
T
GHz
f
L m
=
Ballistic transport
~ 60 g S
Numerical modeling by Purdue group,
in J Guo, et al, Applied Physics Letters, 80,3192 (2002).
100
[ ]
T
GHz
f
L m
=
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 74 74
Substrate
Thin film
electrode
Typical fringe fields
Gap, width = 1 micron => C ~ 1E-16 Farads
P.J. Burke, AC Performance of Nanoelectronics: Towards a THz Nanotube Transistor, Solid State Electronics, (2004).
Parasitics
Parasitics
Semiconducting AC circuit model
Semiconducting AC circuit model
(CNT Field Effect Transistor)
(CNT Field Effect Transistor)
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 76 76
gate
drain
source
C
gs
g
m
v
gs
C
gs
~ 1-10 aF
g
m
~ 10 S
C
gs
AC circuit model:
AC circuit model:
P.J. Burke, AC Performance of Nanoelectronics: Towards a Ballistic Nanotube Transistor, Solid State Electronics, (2004).
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 77 77
gate
drain
source
C
gs
g
m
v
gs
C
gs
~ 1-10 aF
g
m
~ 10 S
C
p
~ 100 aF
C
gs
C
g
s
,
p
a
r
a
s
i
t
i
c
C
gd,parasitic
AC circuit model:
AC circuit model:
P.J. Burke, AC Performance of Nanoelectronics: Towards a Ballistic Nanotube Transistor, Solid State Electronics, (2004).
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 78 78
gate
drain
source
C
gs
g
m
v
gs
R
S
=
h
/
8
e
2
=
3
k
R
D
= h/8e
2
= 3 k
C
g
s
,
p
a
r
a
s
i
t
i
c
C
gd,parasitic
g
d
C
gs
~ 1-10 aF
g
m
~ 10 S
C
p
~ 100 aF
AC circuit model:
AC circuit model:
P.J. Burke, AC Performance of Nanoelectronics: Towards a Ballistic Nanotube Transistor, Solid State Electronics, (2004).
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 79 79
gate
drain
source
C
gs
g
m
v
gs
R
S
=
h
/
8
e
2
=
3
k
R
D
= h/8e
2
= 3 k
C
g
s
,
p
a
r
a
s
i
t
i
c
C
gd,parasitic
( )
p gd p gd gs
m
p gd D S
T
C C C
g
C R R
f
, , ,
1
) (
2
1
+ + + + =
( )
p gd p gd gs D S
m
d
C C C R R
g
g
, ,
) ( + + + +
~ 10 GHz
g
d
C
gs
~ 1-10 aF
g
m
~ 10 S
C
p
~ 100 aF
AC circuit model:
AC circuit model:
P.J. Burke, AC Performance of Nanoelectronics: Towards a Ballistic Nanotube Transistor, Solid State Electronics, (2004).
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 80 80
gate
drain
source
C
gs
g
m
v
gs
R
S
=
h
/
8
e
2
=
3
k
R
D
= h/8e
2
= 3 k
C
g
s
,
p
a
r
a
s
i
t
i
c
C
gd,parasitic
( )
p gd p gd gs
m
p gd D S
T
C C C
g
C R R
f
, , ,
1
) (
2
1
+ + + + =
( )
p gd p gd gs D S
m
d
C C C R R
g
g
, ,
) ( + + + +
m
gs
T
g
C
f
=
2
1
~ 10 GHz
g
d
C
gs
~ 1-10 aF
g
m
~ 10 S
C
p
~ 100 aF
AC circuit model:
AC circuit model:
P.J. Burke, AC Performance of Nanoelectronics: Towards a Ballistic Nanotube Transistor, Solid State Electronics, (2004).
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 81 81
gate
drain
source
C
gs
g
m
v
gs
R
S
=
h
/
8
e
2
=
3
k
R
D
= h/8e
2
= 3 k
C
g
s
,
p
a
r
a
s
i
t
i
c
C
gd,parasitic
( )
p gd p gd gs
m
p gd D S
T
C C C
g
C R R
f
, , ,
1
) (
2
1
+ + + + =
( )
p gd p gd gs D S
m
d
C C C R R
g
g
, ,
) ( + + + +
m
gs
T
g
C
f
=
2
1
~ 10 GHz
g
d
C
gs
~ 1-10 aF
g
m
~ 10 S
C
p
~ 100 aF
RC times:
RC times:
~ 1 THz
If parasitics minimized:
P.J. Burke, AC Performance of Nanoelectronics: Towards a Ballistic Nanotube Transistor, Solid State Electronics, (2004).
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 82 82
Performance projections: Speed
Performance projections: Speed
f
f
T T
(
G
H
z
)
(
G
H
z
)
1
2
2
C= =50 aF/ m (calculated)
ln(h/d)
g ~ 10 S measured
~ 60 S calculated
Nanotubes? Nanotubes?
Gate Length ( Gate Length ( m) m)
0.1 0.1 0.2 0.2 1 1 10 10 2 2 5 5 0.5 0.5 0.05 0.05
10 10
100 100
500 500
200 200
20 20
50 50
Si Si
GaAs GaAs
InP InP
Sze, book Sze, book
Fujitsu 50 nm Fujitsu 50 nm
HEMT (2000) HEMT (2000)
Estimate, Estimate,
speculative speculative
2
T
g
f
C
=
Ref: 1. A. Javey, et al, Nature Materials, 1,241 (2002).
2. J Guo, et al, Applied Physics Letters, 80,3192 (2002).
P.J. Burke, AC Performance of
Nanoelectronics: Towards a THz
Nanotube Transistor,
Solid State Electronics, (2004).
Digital switching speed not yet modeled,
but CV/I metric looks promising.
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 83 83
Semiconducing
Semiconducing
CNT:
CNT:
2.5 GHz demonstration
2.5 GHz demonstration
-12
-10
-8
-6
-4
-2
0
2
S
1
1
(
d
B
)
10 8 6 4 2 0
Frequency (GHz)
Experiment
RLC curve fit
-12.3
-12.2
-12.1
S
1
1
(
d
B
)
8 6 4 2 0
Time (s)
V
g
a
t
e
Fit: L= 40 nH, C=0.1 pF
R=27 k
Li, Burke, et al, Nano Letters, (2004).
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 84 84
-14
-12
-10
-8
-6
-4
-2
0
2
S
1
1
(
d
B
)
15 10 5 0
Frequency (GHz)
Experiment
RLC fit
-9.2
-9.1
-9.0
S
1
1
(
d
B
)
8 6 4 2 0
Time (s)
V
g
a
t
e
Metallic
Metallic
CNT:
CNT:
No field effect at dc or GHz
No field effect at dc or GHz
Fit: L= 14 nH, C=0.1 pF
R=1.7 k
Li, Burke, et al, Nano Letters, (2004).
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 85 85
Ring Oscillator
Carbon
nanotube
Ring Oscillator circuit
design
A more straight forward
measurement of f
t
The frequency of a ring oscillator is given by
f= 1/(2
p
N)
N: the number of stages
p
: the propagation delay in voltage switching
between successive inverter stages, easier to be
measured
V
d
d
p
n
p
n
p
n
V
o
u
t
Work of Jenny Yu
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 86 86
Non
Non
-
-
linear devices
linear devices
Mixers
Mixers
Detectors
Detectors
Plasma wave
Plasma wave
electronics
electronics
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 87 87
Long tubes:
Long tubes:
Electrical properties
Electrical properties
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 88 88
A
u
A
u
SWNT
We can make them LONG
We can make them LONG
0.4 cm
S. Li, Z. Yu, C. Rutherglen, P Burke, Nano Letters, (2004).
1 mm
Conductivity 10x larger than copper!
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 89 89
Carbon Nanotube:
Carbon Nanotube:
Contact resistance or bulk?
Contact resistance or bulk?
1kOhm
10kOhm
100kOhm
1MOhm
10MOhm
100MOhm
R
e
s
i
s
t
a
n
c
e
(
O
h
m
)
100nm 1m 10m 100m 1mm 10mm
Length (m)
UC Irvine (2004, Nano Letters)
Maryland (2004, Nano Letters)
Stanford (2004, Nature)
Ballistic limit
(lowest contact resistance allowed
by quantum mechanics)
Scaling 6 k/m
R = R
contact
+ L 6 k/m
A
u
A
u
Nanotube
0.4 cm
1 mm
Conductivity larger than copper!
Shengdong Li, Zhen Yu, Christopher Rutherglen, Peter J. Burke
"Electrical properties of 0.4 cm long single-walled carbon nanotubes"
Nano Letters, September (2004).
Nanotube transistors:
Does gate modulate bulk resistance or contact resistance?
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 90 90
Challenges
Challenges
Fabrication
Fabrication
Diameter control (
Diameter control (
NWs
NWs
,
,
NTs
NTs
)
)
Chirality
Chirality
control (
control (
NTs
NTs
)
)
Integration
Integration
chemical
chemical
electrical
electrical
A lot of room for improvement
A lot of room for improvement
AC performance of nano-electronics
EE Department Seminar EE Department Seminar
UCSD October 20, 2004 UCSD October 20, 2004 Peter J. Burke Peter J. Burke 94 94
Integration solutions so far
Integration solutions so far
Fluid flow (Lieber)
Fluid flow (Lieber)
Electrical field manipulation
Electrical field manipulation