Quantum Transport:

From Atoms to Transistors

Supriyo Datta
Purdue University

Outline

• Equilibrium
• Current flow as a balancing act
• Broadening
• Charging
• Coulomb blockade
• Capacitance
• Conductance
• Transistors
• Summary: Basic equations
 Nanoscale Transistor
L

OXIDE

----------
µ1 CHANNEL µ2
----------

OXIDE

SOURCE GATE DRAIN

Vg

→I V

Zero bias, V = 0
µ1 µ2

V≠0
µ1
qV

µ2
µ1
qV

µ2
 Equilibrium

OXIDE

----------
Ef CHANNEL
----------
Ef
OXIDE

SOURCE GATE DRAIN

Vg

Vacuum
Level

Work
function
DEVICE
Ef Ef
SOURCE DRAIN
 Current flow as a “balancing act”

----------
µ1 CHANNEL µ2
----------

Vg V
→ I

0.2

0.15

0.1
E-Ef (eV) --->

0.05

0 0.2

-0.05

-0.1
µ1 0.15

0.1
E-Ef (eV) --->

-0.15 0.05

0
-0.2
0 0.5 1
f --->
-0.05
µ2 -0.1

-0.15

-0.2
0 0.5 1
f --->

µ1 = E f + (qV / 2) µ 2 = E f − (qV / 2)
 One-level Model
γ1 /h
µ1 γ 2 /h

ε µ2

N1 = 2 f1 (ε) N2 = 2 f2 (ε)
γ γ
I1 = q 1 (N1 − N) I 2 = q 2 (N − N 2 )
h h

γ1N1 + γ 2N 2
N =
γ1 + γ 2

q γ1γ 2
I = I1 = I 2 = (N1 − N 2 )
h γ1 + γ 2
2 q γ1γ 2
=
h γ1 + γ 2
[f1 (ε) − f2 (ε)]
 One-level Model:
Current (I) vs. Voltage (V)

-5
x 10
2.5

1.5

1
Current (A) --->

0.5

-0.5

-1

-1.5

-2

-2.5
-1 -0.5 0 0.5 1
Voltage (V) --->

µ1 ε µ2

µ1 = E f + ( qV / 2 ) ,
µ 2 = E f − ( qV / 2 ) ,
E f − ε = 0.25 eV ,
µ1 γ1 = γ 2 = 0.1 eV
ε
µ2
 Broadening

γ1 /h0.2 γ 2 /h
0.15

0.1

µ1 µ2
ε
0.05

E (eV) --->
0

-0.05

-0.1

-0.15

-0.2
0 5 10 15
g --->

γ = γ1 + γ 2
γ / 2π
D(E ) =
(E − ε − ∆ ) 2 + (γ / 2) 2

2 q γ1γ 2
I =
h γ1 + γ 2
[f1(ε) − f2 (ε)]
+∞
γ1γ 2
I =
2q
∫
h −∞
dE D(E )
γ1 + γ 2
[f1 (E ) − f2 (E )]

γ1f1 (ε) + γ 2f2 (ε)

N =
γ1 + γ 2
+∞
γ1f1 (E ) + γ 2f2 (E )
N = ∫ 2π
dE
D(E )
γ1 + γ 2
−∞
 One-level Model:
Current (I) vs. Voltage (V)
-5
2.5 x 10

2
Discrete
1.5

1
Broadened
Current (A) ---> 0.5

-0.5

-1

-1.5

-2

-2.5
-2 -1 0 1 2
Voltage (V) --->
E f − ε = 0.25 eV ,
γ1 = γ 2 = 0.1 eV
-4
1.2 x 10

1
Discrete
Conductance (S) --->

0.8 Conductance
quantum
0.6

0.4

0.2 Broadened
0
-2 -1 0 1 2
Voltage (V) --->
 Charging

ε → ε + Uscf (N)
γ / 2π
D(E ) =
(E − ε − Uscf − ∆ ) 2 + (γ / 2) 2

+∞
γ1f1 (E ) + γ 2f2 (E )
(1) N = ∫ 2π
dE
D(E )
γ1 + γ 2
−∞

Uscf → N, Eq.(1)

N → Uscf , Eq.( 2)

(2) Uscf (N) = U a + U (N − N eq )

 One-level Model:
Current (I) vs. Voltage (V)

-5
2.5 x 10 U=0
2 Discrete
U = 1 eV
1.5

1
Broadened
Current (A) --->

0.5

-0.5

-1
E f − ε = 0.25 eV ,
γ1 = γ 2 = 0.1 eV
-1.5

-2

-2.5
-2 -1 0 1 2
Voltage (V) --->

γ1 /h
0.2 γ 2 /h
0.15

0.1

µ1 µ2
ε
0.05
E (eV) --->

-0.05

-0.1

-0.15

-0.2
0 5 10 15
g --->

Uscf (N) = U a + U (N − N eq )
 Coulomb Blockade

γ1 /h γ 2 /h

ε
µ1 µ2

ε↓

U
µ1 µ2

ε↑

ε↑ = ε + U N↓
ε↓ = ε + U N↑
 One-level model:
Restricted vs. Unrestricted
-6
x 10
5

2
Unrestricted
Current (A) --->

1
Restricted
0

-1

-2

-3

-4

-5
-2 -1 0 1 2
Voltage (V) --->

Unrestricted
ε↑ = ε + U N↓ ε↓ = ε + U N↑

Restricted
( )
Current (A) --->

ε↓ , ↑ = ε + U N↓ + N↑
 Capacitance
L

µ ----------
CHANNEL
----------
µ
Vg

 ∂N  1
C ≡ q  =
∂VG  V= 0 1
+
1
CQ C E
+∞
N = ∫ dE D(E ) f0 (E − µ)
−∞
∂N
CQ ≡ q2 = q 2 D0
∂µ

+∞  ∂f (E − µ) 
D0 ≡ ∫ dE D (E ) − 0 ∂ E 
−∞   0.2

0.15

0.1
E-mu (eV) --->

0.05

Ef -0.05
0

-0.1

-0.15

-0.2
0 0.1 0.2 0.3 0.4
(- df / dE) --->
 Conductance
L

µ1 µ2
----------
CHANNEL
----------

Vg V
→ I

T(E )
+∞
γγ
I =
2q
∫
h −∞
dE D(E ) 1 2
γ1 + γ 2
[f1(E ) − f2 (E )]

−1  ∂I  2q 2
R ≡   = T0
∂V  fixed V h
G

+∞  ∂f (E − µ) 
T0 ≡ ∫ dE T (E ) − 0 ∂ E 
−∞  
 Transistors: n- and p- channel

VG

1.5 1.5 1.5

1 1 1

µ
0.5 0.5 0.5
E --->

0 0 0

-0.5 -0.5 -0.5

-1 -1 -1

-1.5 -1.5 -1.5

0 0.5 1 1.5 2 2.5 0 0.5 1 0 0.5 1 1.5 2 2.5
D (E), arb. units---> D (E), arb. units---> D (E), arb. units--->

n- channel
Source Drain
p- channel
1.5 1.5 1.5

1 1 1

µ
0.5 0.5 0.5
E --->

0 0
0

-0.5 -0.5 -0.5

-1 -1 -1

-1.5 -1.5 -1.5

0 0.5 1 1.5 2 2.5 0 0.5 1 1.5 2 2.5
D (E), arb. units---> 0 0.5 1 D (E), arb. units--->
D (E), arb. units--->
“Good” transistors and “bad” transistors

VG V

1.5

1.5
1

1
0.5 1.5
µ1
E --->

0.5
0 ⇒ ⇒ 1
0
0.5
µ 2 = µ1 − qV
-0.5
-0.5
-1
Source 0
-1 Channel
-1.5 -0.5
0 0.5 1 1.5 2 2.5
D (E), arb. units---> -1.5
0 0.5 1
D (E), arb. units--->
-1
Drain
-1.5
0 0.5 1 1.5 2 2.5
D (E), arb. units--->
(a) Good Transistor
1.5

1 1.5

0.5 1 1.5

µ1
E --->

0 0.5 1
⇒ ⇒
-0.5 0 0.5
µ 2 = µ1 − qV
-1 Source -0.5 0

-1.5
0 0.5 1 1.5 2 2.5
-1 Channel -0.5
D (E), arb. units--->

-1.5
0 0.5
D (E), arb. units--->
1 -1
Drain
-1.5
(b) Bad Transistor 0 0.5 1 1.5
D (E), arb. units--->
2 2.5
 “Good” transistor: On - current

1.5

1.5
1

1
0.5 1.5
µ1
E --->

0.5
0 ⇒ ⇒ 1
0
0.5
µ 2 = µ1 − qV
-0.5
-0.5
-1
Source 0
-1 Channel
-1.5 -0.5
0 0.5 1 1.5 2 2.5
D (E), arb. units---> -1.5
0 0.5 1
D (E), arb. units--->
-1
Drain
-1.5
0 0.5 1 1.5 2 2.5
D (E), arb. units--->

v
I ON = C (VG − VT ) Lundstrom, EDL (1997)
L

∞
I ON =
2q
h ε
∫ dE f1 (E )
0

∞
C (VG − VT ) = qN eq = q ∫ dE D(E ) f1(E )
ε0
∞
⇒ q ∫ dE π h2vL(E ) f1(E )
ε0

L
v(E ) ⇒
π h D (E )
 One-level model:
Basic Equations

+∞
N = ∫ dE [D1(E ) f1(E ) + D2 (E ) f2 (E )]
−∞
+∞
I =
2q
∫
h −∞
dE T(E ) [f1 (E ) − f2 (E )]

2 π D1 (E ) = G γ1 G*
2 π D2 (E ) = G γ 2 G*
T(E ) ≡ γ1 G γ 2 G*

1
G =
E − ε − Uscf (N) − (∆ − iγ / 2)

∆ ≡ ∆1 + ∆ 2 , γ ≡ γ1 + γ 2
 One – level à Multi-level
Numbers à Matrices

µ0.151 = E f + (qV / 2)
0.2

0.1

µ0.22 = E f − (qV / 2)
E-Ef (eV) --->

0.05

-0.05

-0.1
µ1 0.15

0.1

E-Ef (eV) --->

-0.15 0.05

-0.2 0
0 0.5 1
f --->
-0.05
µ2 -0.1

-0.15

-0.2
0 0.5 1
f --->

∆1 − i γ1
ε → [H] ∆ 2 − i γ2
Uscf → [Uscf ]
→ [Σ1(E )] → [Σ2 (E )]

N → [ρ] , Density matrix

2πD (E ) → [A(E )] Spectral function
 Multi-level Model:
Basic equations

+∞
[ρ] = ∫ dE { [A1(E )] f1(E ) + [A2 (E )] f2 (E ) }
−∞
+∞
I =
2q
∫
h −∞
dE T(E ) [f1 (E ) − f2 (E )]

A1 (E ) = G Γ1 G+
A2 (E ) = G Γ2 G+

[
T(E ) = Trace Γ1 G Γ2 G+ ]
−1
G = [EI − H − Uscf ([ρ]) − Σ1 − Σ2 ]
 Summary

OXIDE

µ1 CHANNEL µ2
OXIDE

SOURCE GATE DRAIN

Vg

[Σ1 ] [H] [Σ 2 ]

[Uscf ] { [ρ] }