1. VIII. Molecular Electronics
and Nanoscience

2. Molecular Electronics and Nanoscience
Why Molecular Electronics
Moore’s Law
Devices: Top-down and Bottom-Up
Fabrication
Single Molecule Systems and Materials
Many-Molecule Systems and Thin Films
DNA Computing

5. •Role of contacts

6. •Molecular device
ContactContact
Molecule
 what type of behaviour we can expect for a
•complete system?

7. •Ballistic conductor
•Contacts of finite transparency
•P1
•P2
 The model:
•– Electrons tunnel with some probability (contact
•transparency) into the channel
•– Transport is coherent
•– Contacts are ”reflectionless”
 The question:
 What is the resistance of the channel? Where the heat is
•dissipated?

8. •Ballistic conductor model
 Electrons moving from left to right have
•potential µ1, from right to left µ2.

9. Why Molecular Electronics

10. Moore’s Law

14. Silicon and Moore’s Law
• Heat dissipation.
– At present, a state-of-the-art 500 MHz microprocessor with 10
million transistors emits almost 100 watts--more heat than a stove-
top cooking surface.
• Leakage from one device to another.
– The band structure in silicon provides a wide range of allowable
electron energies. Some electrons can gain sufficient energy to hop
from one device to another, especially when they are closely packed.
• Capacitive coupling between components.
• Fabrication methods (Photolithography).
– Device size is limited by diffraction to about one half the wavelength
of the light used in the lithographic process.
• ‘Silicon Wall.’
– At 50 nm and smaller it is not possible to dope silicon uniformly. (This
is the end of the line for bulk behavior.)

15. Silicon and Moore’s Law
• Moore’s second law.
– Continued exponential decrease in silicon device size is
achieved by exponential increase in financial
investment. $200 billion for a fabrication facility by
2015.
• Transistor densities achievable under the present and
foreseeable silicon format are not sufficient to allow
microprocessors to do the things imagined for them.

16. Moore’s “Second Law"
Plant cost Mask cost
generation
X1000$

17. Why Molecular Electronics?
• If current trend continues, it will reach molecular scale in
two decades.
• There are many molecules with interesting electronic properties.
Semiconductor devices shrink to the nano-scale
Year
1950 1960 1970 1980 1990 2000 2010
1 cm
1 m
100 nm
1-5nm
TransistorSize

18. Devices: Top-down and
Bottom-up Fabrication

19. Electronics Development Strategies
• Top-Down.
– Continued reduction in size of bulk semiconductor devices.
• Bottom-up (Molecular Scale Electronics).
– Design of molecules with specific electronic function.
– Design of molecules for self assembly into supramolecular structures with
specific electronic function.
– Connecting molecules to the macroscopic world.

20. Bottom-Up (Why Molecules?)
• Molecules are small.
– With transistor size at 180 nm on a side, molecules are some 30,000 times
smaller.
• Electrons are confined in molecules.
– Whereas electrons moving in silicon have many possible energies that will
facilitate jumping from device to device, electron energies in molecules and
atoms are quantized - there is a discrete number of allowable energies.
• Molecules have extended pi systems.
– Provides thermodynamically favorable electron conduit - molecules act as
wires.
• Molecules are flexible.
– pi conjugation and therefore conduction can be switched on and off by
changing molecular conformation providing potential control over electron
flow.
• Molecules are identical.
– Can be fabricated defect-free in enormous numbers.
• Some molecules can self-assemble.
– Can create large arrays of identical devices.

21. • Top-down Synthesis
1. Lithography

22. Developing
Positive Negative
Etching and
Stripping
Polymer Resist
Thin Film
Substrate
Resist Resist
Exposing
Radiation
Figure 1.1. Schematic of positive and negative resists.

23. Figure 1.6. Schematic of a focused ion beam system.

24. 1m
400nm
300nm
200nm
160
nm
120nm
100nm
80nm
60nm
100 nm
Carbon Nanotubes/Nanocones with Various Catalyst Patterning Dimensions
by E-beam Lithography

25. Molecular Self-Assembly
• Self-Assembly on Metals
– (e.g., organo-sulfur compounds on
gold)
• Assembly Langmuir-Blodgett Films
– Requires amphiphilic groups for
assembly
• Carbon Nanotubes
– Controlling structure

26. Figure 2.1. The process of forming a self-assembled monolayer. A
substrate is immersed into a dilute solution of a surface-active
material that adsorbs onto the surface and organizes via a self-
assembly process. The result is a highly ordered and well-packed
molecular monolayer. (Adapted from Ref. 9 by permission of
American Chemical Society.)

29. Cyclic Peptide Nanotubes as Scaffolds for Conducting Devices
Hydrogen-bonding interactions promote stacking of cyclic peptides
Pi-systems stack face-to-face to allow conduction along the length of the tube
Cooper and McGimpsey - to be submitted
CYCLIC BIOSYSTEMS

30. Spontaneous self-directed chemical growth allowing
parallel fabrication of identical complex functional structures.

31. Self-assembly

32. .Characterization and Handling of
Ultra-small particles or Assemblies
–a. Optical Tweezers
–b. Electromagnetic tweezers
–c. In nanotemplates
–d. Structural Analysis by TEM,
SEM, X-ray, etc.

36. Ballistic Nanotube MOS Transistors (Chen,Hastings)
W
d
D
L
SWNTSWNT
SiO2
Source
Al-Gate
Ti
HfO2
Drain
L
L~20L~20 nmnm
Placement of Nanotubes by E-Field
(The first-demo)
Nanotube Field-Effect Transistor(FET)
E-Beam Lithography

37. • Measurements

38. Figure 3.1. Schematic showing all major components of an STM. In
this example, feedback is used to move the sensor vertically to
maintain a constant signal. Vertical displacement of the sensor is
taken as topographical data
Coarse approach
mechanism
S
c
a
n
n
e
r
Sensor
Sample
Reference
-
Signal
feedback
data
Figure 3.1. Schematic showing all major
components of an SPM. In this example,
feedback is used to move the sensor
vertically to maintain a constant signal.
Vertical displacement of the sensor is taken
as topographical data.

39. •1980’s
Single Molecule Detection.
How to image at the molecular level.
How to manipulate at the molecular
level.
• Scanning Probe Microsopy.
STM (IBM Switzerland, 1984)
AFM
Molecules as Electronic Devices:
Historical Perspective

41. Major equipment
• Focused Ion Beam System (FIB) (scheduled for installation in mid 2007)
• Atomic Layer Deposition System (ALD)
• Rapid Thermal Processing System (RTP)
• Plasma Enhanced Chemical Vapor Deposition System (PECVD)
• Standard Resolution Electron Beam Lithography (EBL)
• Atomic Force Microscope for Nanopatterning, and Manipulation (AFM)
• Atomic Force Microscope for Atomic Resolution Imaging (AFM)
• Quartz Crystal Microbalance (QCM)
• 4-furnace bank of 3-zone oxidation, dopant diffusion, and annealing furnaces
• Class 100 Clean Room
• Spin-Coating Station
• Photolithography System
• Surface Profiler
• Chemical Treatment Station (cleaning, etching, and functionalization)
• Ion Milling System
• Plasma Cleaning/Oxidation System
• Gas Cabinet Bank
• Experimental Materials Thermal Evaporator
• Standard Materials Thermal Evaporator
• Electron-Beam Evaporator
• Multi-target Sputtering System
• Probe Station and Device Characterization System
• Four-Point Resistance Measurement System
• Ellipsometer
• Optical Microscopes
• Dicing Saw
• Equipment Cooling Systems (3)
• Inductive Coupled Plasma (ICP) Etching System (scheduled for installation in Feb. 2006)
• Experimental materials sputtering system (scheduled for installation in mid 2006)
• Ultra-High Resolution EBL and SEM System
Clean Room
Photolithography
Rapid Thermal
Processing
Quartz MicroBalance
Plasma
Enhanced
Chemical
Vapor
Deposition
Reactive Ion Etching
Atomic
Layer
Deposition

44. IV. Nanotemplates
• G. Inorganic
• H. Organic

46. Fig.2 (a) Nanostructure of anodically formed Al2O3 template. (b) its cross-section,
(c) catalyst deposited at the bottom of the pores, (e) vertically aligned nanotubes, and (f) TEM
image of a nanotube.

47. (Chen, Singh, DeLong, Saito, Yang, Bhattacharyya, and Sumanasekeras)
Nano-scale Material Research
(a)
(b)
(c)
Catalyst
200nm
200 nm
Vertically aligned MWNTs
embedded in AAO insulator
(d)
Si substrate
SiO2
SiO2
Carbon nanotubes
AlSiO2
Hexagonal Cells
Nano-template
Horizontally
aligned
The first vertically aligned nanotubes on silicon substrates using templates

48. • Fig. 3 Schematic representing the helix-coil transitions within the pore of a
Poly-L-Glutamic Acid functionalized membrane (a) random-coil formation at
PH > 5.5 , (b) helix formation at low pH ( <4 ).

57. Single Molecule Devices

58. Using a Single Molecule

59. Cornell group

60. What Might Single Molecule
Devices Do?

61. Single-Electron Memory Cell
Fe+3
Fe+2
Heme group
e-
e-
Au

62. Molecular Abacus
The “bead” can be reversibly
switched between two positions
by pH.
Ashton et al. JACS, 120, 11932(1998)

63. Some Fancy Molecules
Rotaxane
Catenane Pretzelane Handcuffcatenane

64. Synthesis of a Rotaxane Molecule
Amabilino and Stoddart, Chemical Reviews 95, 2725 (1995).

65. Angew. Chem., Int. Ed. 2000, 39, 3284-3287.
Molecular Motor
Molecular Oscillator

66. Molecular Sensor
K+
K+
Molecular Recognition: A capability that Si lacks
K +
Ag
+ K+
K+ K +
Ag
+
Crown ether
• A small difference in the diameters of the K+ and Ag+
can cause a huge difference in the binding capacity

67. Heath and Ratner, Physics Today, May 2003, p. 43
Many Ideas for Single Molecule Devices

68. Nano-switch

69. Single Molecule Systems and
Materials

70. Molecular conduction
molecule

71. Molecules as Electronic Devices:
Historical Perspective
•1970’s: Single Molecule Devices?
• In the 1970’s organic synthetic
techniques start to grow up
prompting the idea that device
function can be combined into a
single molecule.
• Aviram and Ratner suggest a
molecular scale rectifier. (Chem.
Phys. Lett. 1974)
• But, no consideration as to how
this molecule would be
incorporated into a circuit or
device.

72. Molecular Rectifiers
Arieh Aviram and Mark A. Ratner
IBM Thomas J. Watson Research Center, Yorktown Heights, New
York 10598, USA
Department of Chemistry, New York New York University, New
York 10003, USA
Received 10 June 1974
Abstract
The construction of a very simple electronic device, a
rectifier, based on the use of a single organic molecule is
discussed. The molecular rectifier consists of a donor pi
system and an acceptor pi system, separated by a sigma-
bonded (methylene) tunnelling bridge. The response of such
a molecule to an applied field is calculated, and rectifier
properties indeed appear.

73. Acceptor Donor

75. Single Molecule Systems and
Materials

76. Elastic
Inelastic
V
h/e V
h/e V
h/e V
IdI/dVd
2
I/dV
2
h/e
Finding a true molecular signature:
Inelastic Electron Tunnelling Spectroscopy (IETS)

77. Towards Single Molecule Electronics
Can a single molecule behave like a
diode, transistor (switch), memory ?
If that’s possible, how long will the molecule last ?
First, let’s look at many molecules acting in parallel.
Nitzan and Ratner, Science 300, 1384 (2003); Heath and Ratner, Physics Today, May 2003, p. 43

78. Single-Molecule Conductivity
L ELECTRODE R ELECTRODEMOLECULE

79. L ELECTRODE R ELECTRODEMOLECULE
Fermi
energy
Molecular
Orbitals

80. eV
V
L ELECTRODE R ELECTRODEMOLECULE
I
Molecular
Orbitals

81. Molecular Electronics:
Measuring single molecule conduction
Kushmerick et al. PRL 89
(2002) 086802
Cross-wire
Wang et al.
PRB 68 (2003) 035416
Nanopore STM Break Junction
B. Xu & N. J. Tao Science (2003) 301, 1221
Electromigration
H. S. J. van der Zant et al.
Faraday Discuss. (2006)
131, 347
Nanocluster
Dadosh et al. Nature
436 (2005) 677
Scanning Probe
Cui et al.
Science 294
(2001) 571
Reichert et al. PRL 88
176804
Mechanical Break
Junction

82. Mechanically-Controlled Break Junction
Resistance is a few megohms.
(Schottky Barrier)
Molecular Junction

83. A schematic representation of Reed and Tour’s molecular junction containing a
benzene-1,4-dithiolate SAM that bridges two proximal gold electrodes.
Break Junctions
At the beginning of single molecule
electronics, break junctions were very
popular: Just crack a thin Au wire open
in a vice and adjust the width of the
crack with piezos (as in STM). Then
pour a solution of molecules over it.
Alternatively, one can burn out the
thinnest spot of a thin Au wire by
running a high current density through
it (using the effect of electromigration).
These days, many try to achieve a
well-defined geometry using a STM
or AFM, with a well-defined atom at
the end of the tip and another well-
defined atom at the surface as con-
tacts to a single molecule.

84. Nanotube conductivity is quantized.
Nanotubes found to conduct current ballistically and do not dissipate heat.
Nanotubes are typically 15 nanometers wide and 4 micrometers long.
Carbon Nanotubes
Gentle contact needed

85. Using a Few Molecules
Observe tunneling through 1, 2, 3, 4, 5 alkanethiol molecules
Cui et al., Science 294, 571 (2001)

86. “The resistance of a single
octanedithiol molecule was 900 50
megaohms, based on measurements
on more than 1000 single molecules.
In contrast, nonbonded contacts to
octanethiol monolayers were at least
four orders of magnitude more
resistive, less reproducible, and had a
different voltage dependence,
demonstrating that the measurement
of intrinsic molecular properties
requires chemically bonded
contacts”.
Cui et al (Lindsay), Science
294, 571 (2001)


87. Dynamics of current voltage
switching response of single
bipyridyl-dinitro oligophenylene
ethynylene dithiol (BPDN-DT)
molecules between gold contacts.
In A and B the voltage is changed
relatively slowly and bistability
give rise to telegraphic switching
noise. When voltage changes
more rapidly (C) bistability is
manifested by hysteretic behavior
Lortscher et al (Riel), Small, 2, 973 (2006)

88. Chem. Commun., 2006, 3597 - 3599, DOI: 10.1039/b609119a
Uni- and bi-directional light-induced switching of diarylethenes on gold
nanoparticles
Tibor Kudernac, Sense Jan van der Molen, Bart J. van Wees and Ben L.
Feringa
“In conclusion, photochromic
behavior of diarylethenes
directly linked to gold nanoparticles
via an aromatic spacer has
been investigated. Depending on the
spacer, uni- (3) or bidirectionality
(1,2) has been observed.”
Switching with light

89. Current–voltage data (open circles) for (a) open
molecules 1o and (b) closed molecules 1c
Nanotechnology 16 (2005) 695–702
Switching of a photochromic molecule on gold electrodes: single-molecule
measurements
J. He, F. Chen, P. Liddell, J. Andr´easson, S D Straight, D. Gust, T. A. Moore,
A. L. Moore, J. Li, O. F Sankey and S. M. Lindsay

90. Conductance through a C60 Molecule
Distance dependence tells whether it is tunneling (exponential decay)
or quantum conductance through a single or multiple orbitals (G0).
Kröger et al., J. Phys. Condend. Matter 20, 223001 (2008)

91. (a) Structures of the long and short linked cobalt coordinated terpyridine thiols used
as gate molecules. (b) A topographic AFM image of the gold electrodes with a gap.
(c) A schematic representation of the assembled single atom transistor.
A Molecular Transistor

92. Many-molecule Systems
and Thin Films

93. mNDR = molecular Negative Differential Resistance
Measured using a conducting AFM tip
Negative Differential Resistance
One electron reduction provides a charge carrier.
A second reduction blocks conduction.
Therefore, conduction occurs only between the
two reduction potentials.

94. Voltage-Driven
Conductivity Switch
Applied perpendicular field favors
zwitterionic structure which is planar
Better pi overlap, better conductivity.

95. Dynamic Random Access
Memory
Voltage pulse yields
high conductivity
State - data bit stored
Bit is read as high
in low voltage region

96. Device is fabricated by sandwiching a layer
of catenane between an polycrystalline layer of n-doped
silicon electrode and a metal electrode. The switch is
opened at +2 V, closed at -2 V and read at 0.1 V.
Voltage-Driven
Conductivity Switch

97. High/Low Conductivity Switching Devices
Respond to I/V Changes
Voltage-Driven
Conductivity Switch

98. n-type
Voltage-Driven
Conductivity Switch

99. Other Molecular Switches
Chen et al., Science 286, 1550 (1999)
Large On-Off Ratios

100. Data Storage via the Oxidation State of a Molecule
Electrochemistry

101. 40 nm line width, 40 Gbit/inch2
HP Molecular Memory

102. Output:
Stored
Data
Input:
Address
Molecular Memory
MRAM
(Magnetic Random Access Memory)
Crossbar Memory
Architecture
DRAM
1
0

103. HP Molecular Memory
The blue ring can shuttle back
and forth along the axis of the
rotaxane molecule, between
the green and red groups.
Rotaxane molecules switch
between high and low resis-
tance by receiving a voltage
pulse.

104. Collier et al., Science 289, 1172 (2000).
(Many Molecules)
HP Molecular Memory
Change the resistance between
low and high by voltage pulses.
Is the resistance change really due
to the rotaxane ring shuttling back
and forth? Other molecules exhibit
the same kind of switching.
One possible model is the creation
and dissolution of metal filaments
which create a short between the
top and bottom electrodes. (Some-
thing like that happens in batteries).

105. Robert F. Service,
Science 302, 556
(2003).

106. Quantum
Dot
Molecular Switch
Self-Organizing Memory + Data Processor
Heath et al., Science
280, 1716 (1998)
People have been thinking about
how to combine memory with logic
(= a microprocessor) in a molecular
device.
Self-assembly is the preferred
method. It generates errors, though.
They need to be absorbed by a
fault-tolerant architecture (e.g. in the
HP Teramac)

108. DNA Computing

109. DNA Computing
I believe things like DNA computing will
eventually lead the way to a “molecular
revolution,” which ultimately will have a
very dramatic effect on the world.
L. Adleman

110. What is DNA?
• All organisms on this planet are made of the same type of
genetic blueprint.
• Within the cells of any organism is a substance called DNA
which is a double-stranded helix of nucleotides.
• DNA carries the genetic information of a cell.
• This information is the code used within cells to form proteins
and is the building block upon which life is formed.
• Strands of DNA are long polymers of millions of linked
nucleotides.

111. Graphical Representation of inherent bonding properties of
DNA

112. Double Helix shape of DNA
The two strands of a DNA molecule are anti
parallel where each strand runs in an opposite
direction.
GC base pair and AT base pair

116. Adleman’s Experiment
• Hamilton Path Problem
(also known as the travelling salesperson problem)
Perth
Darwin
Brisbane
Sydney
Melbourne
Alice Spring
Is there any Hamiltonian path from Darwin to Alice Spring?

117. Adleman’s Experiment (Cont’d)
• Solution by inspection is:
Darwin  Brisbane  Sydney  Melbourne  Perth 
Alice Spring
• BUT, there is no deterministic solution to this
problem, i.e. we must check all possible
combinations.
Perth
Darwin
Brisbane
Sydney
Melbourne
Alice Spring

119. Adleman’s Experiment (Cont’d)
1. Encode each city with complementary base -
vertex molecules
Sydney - TTAAGG
Perth - AAAGGG
Melbourne - GATACT
Brisbane - CGGTGC
Alice Spring – CGTCCA
Darwin - CCGATG

120. Adleman’s Experiment (Cont’d)
2. Encode all possible paths using the
complementary base – edge molecules
Sydney  Melbourne – AGGGAT
Melbourne  Sydney – ACTTTA
Melbourne  Perth – ACTGGG
etc…

121. Adleman’s Experiment (Cont’d)
3. Merge vertex molecules and edge molecules.
All complementary base will adhere to each other to
form a long chains of DNA molecules
Solution with
vertex DNA
molecules
Solution with
edge DNA
molecules
Merge
&
Anneal
Long chains of DNA molecules (All
possible paths exist in the graph)

122. Adleman’s Experiment (Cont’d)
• The solution is a double helix molecule:
CCGATG – CGGTGC – TTAAGG – GATACT – AAAGGG – CGTCCA
TACGCC – ACGAAT – TCCCTA – TGATTT – CCCGCA
Darwin Brisbane Sydney Melbourne Perth Alice Spring
Darwin
Brisbane
Brisbane
Sydney
Sydney
Melbourne
Melbourne
Perth
Perth
Alice Spring

123. Basics And Origin of DNA Computing
• DNA computing is utilizing the property of DNA for massively
parallel computation.
• With an appropriate setup and enough DNA, one can
potentially solve huge problems by parallel search.
• Utilizing DNA for this type of computation can be much faster
than utilizing a conventional computer
• Leonard Adleman proposed that the makeup of DNA and its
multitude of possible combining nucleotides could have
application in computational research techniques.

124. Problems with Adleman’s Experiment
• The researchers performed Adleman’s Experiment and the
results obtained were inconclusive.
• The researchers state that “At this time we have carried out
every step of Adleman’s Experiment but have not gotten an
unambiguous final result.”
• The problem is because of the underlying assumption that the
biological operations are error-free.

125. Problem Instance
• There are 2 problems with extraction:
– The removal of strands containing the sequence in not 100% efficient.
– May at times inadvertently remove strands that do not contain the
specified sequence.
• Adleman’s did not encounter problems with extraction
because only a few operations were required.
• However, for a large problem instance , the number of
extractions required may run into hundreds or even
thousands.

126. • Time- Adleman talked of a week of work in lab, but tuning
such an experiment could take one month work
• Contradictory results- We do not know a lot of experiments
like Adleman’s, nor Adleman’s trials of repeating the
experiment.
Problems Contd.

127. Advantages of a DNA Computer
• Parallel Computing- DNA computers are massively parallel.
• Incredibly light weight- With only 1 LB of DNA you have more
computing power than all the computers ever made.
• Low power- The only power needed is to keep DNA from
denaturing.
• Solves Complex Problems quickly- A DNA computer can solve
hardest of problems in a matter of weeks

128. Disadvantages of DNA Computer
• High cost is time.
• Occasionally slower-Simple problems are solved much faster
on electronic computers.
• It can take longer to sort out the answer to a problem than it
took to solve the problem.
• Reliability- There is sometime errors in the pairing of DNA
strands

129. DNA Chip
Source: Stanford Medicine Magazine, Vol 19, 3 Nov 2002
http://mednews.stanford.edu/stanmed/2002fall/translational-dna.html

130. Chemical IC
Source: Tokyo Techno Forum 21, 21 June 2001
http://www.techno-forum21.jp/study/st010627.htm

131. The Smallest Computer
• The smallest programmable DNA computer
was developed at Weizmann Institute in Israel
by Prof. Ehud Shapiro last year
• It uses enzymes as a program that processes
on on the input data (DNA molecules).
• http://www.weizmann.ac.il/mathusers/lbn/new_pag
es/Research_Biological.html

132. References
• “Molecular Computation of Solutions to Combinatorial
Problems”, L.M. Adleman, Science Vol.266 pp1021-1024,
11 Nov 1994
• “Computing With Cells and Atoms – an introduction to
quantum, DNA and membrane computing”, C.S. Calude and
G. Paun, Taylor & Francis, 2001
• “The Cutting Edge Biomedical Technologies in the 21st
Century”, Newton, 1999
• “Human Physiology: From Cells to Systems 4th Ed.”, L.
Sherwood, Brooks/Cole, 2001