Conductance of molecular wire

2. Molecular Electronics
•Molecular-scale electronics
•Molecular materials for electronics
•Molecular wire
•Diode, rectifier
•Molecular switches
•Molecular memory
•Sensors
•Optics and optical switches
•Displays
•Electrochemical devices
•Molecular heterostructure and quantum well devices
Molecular Electronic Switching Devices
• Electric-field controlled molecular switching devices
including quantum-effect molecular electronic
devices
• Electromechanical molecular electronic devices
• Photoactive/photochromic molecular switching
devices
• Electrochemical molecular devices
Eb>kT
Bistability
강의 순서
•
•
•
•
•
•
•
•
•
Information transfer
Interconnections
Electron transport
Molecular wire
Molecular rectifier
Molecular switches
Molecular RTD
Molecular memory
Atom relay
Computational Limits
• Rolf Landauer
Nonreversible computer performing Boolean
logic operation requires minimum energy for a
bit operation,
P = nkBT ln2
Ex) energy required to add two 10 digit decimal numbers ?
P ~ 100kT ln2 ~ 3x10-18 joule per additions
1018 additions per joule
This is roughly the equivalent
of 109 Pentiums !
Why Molecular electronics ?
ENIAC, 1947
HP Jornada, 2000
17,468 vacuum tubes
60,000 pounds
16,200 cubic feets
174 kilowatts (233 horsepower)
•Popular Mechanics (1949, March issue) predict that
someday ENIAC would contain only 1500 vacuum tubes
•Now, Improve power efficiency by 108 and shrink by 108
•Their prediction was based wrong foundation(vacuum-tube
technology)
Technology and Biology
Technology
• Artificial
• Si-based
• Manufactured
• Short history
• Function: logical
numercal operation
• Room temp
Biology
Natural
C-based
Self-assembled
Long history
Replication, adaptation…
Room temp
Microprocessor vs Brain
MPU
• # of MOSFET ~ 107
• # of switches ~ 107
• # of connections ~107
• Wiring: fixed
• Architecture: serial
Brain
# of cells ~ 1012
# of switches ~ 1011 neurons
# of connections ~ 1015 synapses
flexible
parallel
DRAM vs DNA
DRAM
DNA
Advantages of Molecular Electronics
•Nano-scaled structures with identical size
•Ultra High density: 106 times denser than Si logic
circuits
•Very cheap
Critical issues on Molecular Electronics
•What device types can provide bistable operation ?
•How can these devices be organized into high
density 2D and 3D arrays ?
•How can these devices be connected in large number
to input/output lines?
Dynamic Random Access Memory (DRAM)
W1
W2
CMOS FET
D1 D2 (bit line)
Interconnection Dilemma
• Today’s chip densities are such that the wires
consume some 70% of the real estate
→ they cause some 70% of the defects that
lower chip yield
• The rate of defects in a chemically fabricated
nanocircuit : ppb level
→ million defects in a system containing 1015
components
It is impossible to tolerate such a
level of defects!!
Several Fascinating Possibilities
Future computing architecture
should be highly tolerant defects!!
• Defect-tolerant computing architecture:
Heath, and Stoddart at UCLA, teamed up with Williams at HP
• Nanocell:
Tour at Rice, Reed at Yale teamed up with Penn State
in Nov. 1999
• Switching with nanotubes:
Lieber at Harvard University
• DNA assembly, computation:
Seeman at New York University
Defect-Tolerant Computing Architecture
•Fat tree architecture enable one to route around and avoid the defect.
•Manufacturing by chemical assembly is feasible.
HPL Teramac
1THz multi-architecture computer
•Tera: 1012 operations per sec
•Mac: Multiple architecture computer
•106 gates operating at 106 cycle/sec
•Largest defect-tolerant computer
•Contains 256 effective processors
•Computes with look-up tables
•220,000 (3%) defective components
Nanotube Interconnects
•Molecular scale wire with atomic perfection
•Interconnect at high density
•High thermal conductivity
•Very stiff materials
Lieber et al
Wire conductance vs. Electron transfer
Molecular wire
Potential
Energy
Intramolecular
electron transfer
Ef
R(DA)
P(D+A- )
Eg
Obsevable
current
rate constant
Continuum
electrode
vibronic levels
Process
electron tunneling
electron tunneling
Theory
I = (2pe/h)∫dE T2(E,V)
[fD(E)(1-fA(E+eV))
k =(2p/h)T2DA(FC)
Conductance : Landauer formula
m2
eVmol
m1
eV
m2
m1
I = (2e/h)∫dE T(E,V) [f(E-m1)-f(E- m2)]
•F(E): Fermi function
•T(E,V): transmission function
molecular energy levels and their coupling
to the metallic contacts
m1, m2 : electrochemical potentials
m1= Ef- eVmol
m2 =Ef+eV- eVmol
W. Tian et al, J. Chem. Phys. 109, 2874(1998)
Intramolecular Electron Transfer
D Bridge
A
Eg
R(DA)
P(D+A- )
RDA
k = (2p/h)V2DA(FC)
VDA = aexp(-bRDA), b= -(1/a)ln(2t/Eg)
VDA : electronic coupling through direct and
superexchange
FC: Frank-Condon factor
RDA : DA distance
a: length between electronic basis function in
the bridge structure
T: matrix elements between those bridge structure
Eg : energy gap
Chap1 and 2 in Molecular Electronics ed. J. Jortner and M. Ratner
Fabrication: Self-Assembled Monolayer (SAM)
Fabrication: Langmuir-Blodgett Technique
Setup for Langmuir-Blogett Deposition
Transferring Monolayers
&
Multilayer Film
P-Area Isotherms
http://www.public.iastate.edu/~miller/nmg/lbfilms.html
Conductance of molecular wire: STM
•Tip bias voltage = 1V,
•Tunnling current = 10 pA
L. A. Bumm et al , Science 271, 1705 (1996).
Conductance of Molecular Wire: MBJ
•Mechanically controllable break junction
•Energy gap Gap: 0.7 eV
•Conductance = 0.45 mS (R=22 MW)
MA REED,C Zhou, CJ Muller, TP Burgin, JM Tour, Science 278, 252 (1997)
Structural Effects on Conductance:Theory
•Resistance increases exponentially with the # of the rings
•Relative orientation of the rings
•The bonding between them.
MP Samanta et al, Phys. Rev. B53, R7626 (1996)
Temperature Effects : Theory
0
90
•Unusual temperature induced large shift (~1eV) in is due to:
- The rotation property of the NO2 group
- Different symmetry of the states localized on the NO2 group
with respect to the orbitals of the carbon ring
M Di Ventra, SG Kim, ST Pantelides, ND Lang, PRL86, 288(2001)
Comparison of Conductivity
1,4 benzene
ditiol
polyphenylene
wire (3ring)
App. Vol
1
1
1
Current (A)
2x10-8
3.2x10 -5
1x10 -7
Cross section
(nm2)
~0.05
~0.05
~3.1
r=1nm
4x1012
2x1011
Current density 2x1012
(e/sec-nm2)
carbon
nanotube
copper wire
2x10–3
(10cm)
1
~3.1x1012
r=1mm
2x106
Molecular Rectifier
I
A. Aviram and M.A. Ratner,
Chem. Phys. Lett. 29, 277 (1974)
electron donor
V
Forward
bias
electron acceptor
p
n
-
B
+
TTF
Tetrathiofulvalene
+
+
+
P
-
TCNQ
tetracyanoquinodimethane
πD*
πD
πA*
πA
EF
Energy Levels of Molecular Orbitals
DELUMO =ELUMO (D)- ELUMO (A)
Vacuum
f
Unoccupied
Ef
Occupied
No bias
Off resonance
Transmitted
electron
eVb
Forward bias
In resonance
Electrical Rectification of LB films
R. M. Metzger et al, J. Am. Chem. Soc. 119, 10455 (1997)
Energy levels
Electron
affinity
Ionization
potential
LUMO
HOMO
• Ionization potetials ID for D end must be small and match as closely as
possible work function (f1) of metal layer (M1).
- If ID is too low, the molecule would oxidize in air
• Electron affinity AA for A end must be as large as possible, match with
the work function metal layer M2(f2): this is not easy !
Molecular Switches
1. Chemical switching
2. Electrochemical switching
3. Photochemical switching
V.Balzani et al, Acc. Chem. Res. 31, 405 (1998)
Molecular Switches and Gates: Rotaxane
4PF6-
N+
N+
+N
N+
CH2OH
•Closed at reducing voltage(-2V): current flow due to resonant tunneling
•Open at oxidizing voltage(>0.7V): irreversible
C.P. collier et al, Science 285, 391(1999)
Electron Transport in Single molecule
B1
B2
LUMO
Ef
M
M
Ef
V
HOMO
V
I
I
V=VLUMO
V=0 VOLT
I
Ef
V
DELUMO-HOMO
I
V
V=-VHOMO
Electron Transport in Single molecule
Molecule switches
M  M*
V
Ef
I
V = -VLOMO
.. ..
V = -VSWITCH
.. ..
Then, reset at the opposite bias
M  M*
.. ..
.. ..
V = -VRESET
I
DELUMO-HOMO
V
Configurable Molecular AND Gate
A
C
B
A
A
0
1
0
1
C
B
R
V+
B
0
0
1
1
C
0
0
0
1
•Difference between high
and low current levels:
15 ~30
C.P. Collier, E.W. Wong et al.
Current (10-9 Amps)
4
+V
L
high
2
A
B
out
low
0
A= 0
B= 0
0
1
1
1
0
1
AND Gate Address Levels
Reversible Molecular Switches: [2]Catenane
Collier et al, Science, 289,1172 (2000)
Ti/Al : top electrode
Si
Ox.
LB monolayer
n-type ploy Si film:
bottom electrode
SiO2
Red.
Close
Open
•Upon oxidation, the TTF become positively charged, the Coulombic repulsion
between TTF+ and the tetracationic cyclophane causes to circumrotate.
•Reversible switching :Opened >+2V, closed <-1.5V
Molecular Field Effect Transistor
R.A. Reed and J.M. Tour, Sci. Amer. 282,86 (2000)
Source Gate Drain
Reversible Molecular Switch with NDR Effect
•Negative Differential Resistance ~ 400 MWcm2
•Peak current denisty: 50A/cm2
•Peak to valley ratio = 1030:1 (typical device=30:1)
•Temperature induced shift : rotation of ligand (JACS,122,3015 (2001))
J Chen, MA Reed, AM Rawlett, JM Tour, Science 286, 1550 (1999)
NDR Effect (continued)
Q=0
-1
-2
A
I
B
B
A
C
V
+
•Anion conduction state
•ON
•Dianion insulating state
•OFF
C
Molecular Diode
•The prominent rectifying behavior is due to the asymmetry of
the molecular heterostructure.
•The barrier from the bottom electrode is higher than the barrier for
electrons from the top T electrode
C. Zhou, M. R. Deshpande, M. A. Reed, L. Jones II, and J. M. Tour, Appl. Phys.
Lett., 71, 611 (1997).
Lowest Unoccupied Molecular Orbital (LUMO)
No LUMO states on the ring
Molecular RAM
On
Off
•15min hold time DRAM at room temperature
•Reversible molecular memory
•Over one billion cycles and counting with no degradation
M.A. Reed et al, Appl. Phys. Lett. 78, 3735 (2001),
Z.J. Donahuer et al, Science 292, 2305 (2001)
Molecular Resonant Tunneling Diode (MRTD)
Unoccupied
Off
Occupied
1nm
On
•Peak to valley ratio = 1.3 :1
•Electrically active device by molecular orbital engineering
M.A. Reed, Proc. IEEE , Volume: 87, 652 (1999)
Synthesis of Molecular Devices
Source
Drain
Gate
•Nano-scaled structures with identical size and shape
•High density and low power
J.A Tour, Acc. Chem. Res. 33, 391 (2000)
Logic Gate : OR
A
C
B
R
A
C
B
V-
Donor
CH2
Acceptor
wire
CH2
Acceptor
Long HC
Donor
Conjugated aromatic organic molecules
A
0
1
0
1
B
0
0
1
1
C
0
1
1
1
Electromechanical Molecular Electronics
1. Single molecule electromechanical amplifier:
Joachim and Gimzewski Chem Phys. Lett. 265, 353 (1997)
- conductance modulation due to
electromechnical deformation of C60 cage
2. Atom relay transistor
Y. Wada, JVST A17, 1399 (1999)
•Upon charging a gate, a mobile switching atom move into
a line of atomic wire
Inroganic-Organic Hybrid Circuits: Nanocell
• Molecular wire having alligator clip
INPUT
Metallic nanoparticles
which are connected
by functional
molecules
Gate
OUTPUT
1mm
Further challenges
•
•
•
•
•
•
•
•
Combining individual devices
Mechanisms of conductance
Nonlinear I/V behavior
Energy dissipation
Necessity of gain in molecular electronic circuits
Slow speed
New computer architecture
Synthesis of new molecules
숙제 1.
Read Feyman’s lecture “There is plenty
of room at the bottom” listed at
www.zyvex.com/nanotech/feynman.html
공고
•금주 20일 (목요일): 휴강
•스케쥴 변경
week 7 :
10.16-10.18 Macromolecular nanostructures (김상율 교수)
week 8 : 10.20-26 중간고사
switch
ㅇ 강의실:
자연과학동 2124호
ㅇ 참고문헌 : Nanotechnology Research Directions: IWGN Workshop Report(1999)
http://itri.loyola.edu/nano/IWGN.Research.Directions/
ㅇ 강의내용: 기능성 나노구조의 합성, 제조, 물리 화학적 성질과 나노구조에 대한 특성분석
방법 등을 다루고 나노구조를 응용한 나노센서 및 나노소자의 개념을 소개함.
------------------------------------------------------------------------------------------
week 1 :
9. 4
Introduction (김세훈 교수)
week 1/2 : 9.6-9.11
Nano-characterization (김세훈 교수)
week 2/3 : 9.13-9.18
Nano dvice I (김세훈 교수)
week 3/4 : 9.20-9.25
Template based nanostructures (유룡 교수)
week 4/5 : 9.27-10.4
Template based nanostructures (유룡 교수)
week 6 :
10.9-10.11
Macromolecular nanostructures (김상율 교수)
week 7 :
10.16-10.18 휴강
week 8 : 10.20-26
Macromolecular nanostructures (김상율 교수)
week 9 :
Nano-fabrication and nano-lithography (김진백 교수)
10.30-11.1
week 10 : 11.6-11.8
Nano-quantum chemistry (이윤섭 교수)
week 11 : 11.13-11.15 Nano-thermodynamics (이억균 교수)
week 12 : 11.20-11.22 Nano-sensor and nano-device II (곽주현 교수)
week 13:
11.27-11.29 Nano-sensor and nano-device II (곽주현 교수)
week 14:
12.4-12.6
week 15:
12.11-12.13 Nanoparticles and nanowires (천진우 교수)
week 16 : 12.15-12.21
Nanoparticles and nanowires (천진우 교수)
기말고사