nanotechnology

Primary Contributor: S. Tom Picraux

ARTICLE
from the
Encyclopædia Britannica

nanotechnology,  the manipulation and manufacture of materials and devices on

the scale of atoms or small groups of atoms. The “nanoscale” is typically measured in
nanometres, or billionths of a metre (nanos, the Greek word for “dwarf,” being the source of the
prefix), and materials built at this scale often exhibit distinctive physical and chemical properties
due to quantum mechanical effects. Although usable devices this small may be decades away
(see microelectromechanical system), techniques for working at the nanoscale have become
essential to electronic engineering, and nanoengineered materials have begun to appear in
consumer products. For example, billions of microscopic “nanowhiskers,” each about 10
nanometres in length, have been molecularly hooked onto natural and synthetic fibres to impart
stain resistance to clothing and other fabrics; zinc oxide nanocrystals have been used to create
invisible sunscreens that block ultraviolet light; and silver nanocrystals have been embedded in
bandages to kill bacteria and prevent infection.

Possibilities for the future are numerous. Nanotechnology may make it possible to manufacture
lighter, stronger, and programmable materials that require less energy to produce than
conventional materials, that produce less waste than with conventional manufacturing, and that
promise greater fuel efficiency in land transportation, ships, aircraft, and space vehicles.
Nanocoatings for both opaque and translucent surfaces may render them resistant to corrosion,
scratches, and radiation. Nanoscale electronic, magnetic, and mechanical devices and systems
with unprecedented levels of information processing may be fabricated, as may chemical,
photochemical, and biological sensors for protection, health care, manufacturing, and the
environment; new photoelectric materials that will enable the manufacture of cost-efficient solar-
energy panels; and molecular-semiconductor hybrid devices that may become engines for the
next revolution in the information age. The potential for improvements in health, safety, quality
of life, and conservation of<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388670/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009140764"></script> the
environment are vast.

At the same time, significant challenges must be overcome for the benefits of nanotechnology to
be realized. Scientists must learn how to manipulate and characterize individual atoms and small
groups of atoms reliably. New and improved tools are needed to control the properties and
structure of materials at the nanoscale; significant improvements in computer simulations of
atomic and molecular structures are essential to the understanding of this realm. Next, new tools
and approaches are needed for assembling atoms and molecules into nanoscale systems and for
the further assembly of small systems into more-complex objects. Furthermore, nanotechnology
products must provide not only improved performance but also lower cost. Finally, without
integration of nanoscale objects with systems at the micro- and macroscale (that is, from
millionths of a metre up to the millimetre scale), it will be very difficult to exploit many of the
unique properties found at the nanoscale.

Overview of nanotechnology
Nanotechnology is highly interdisciplinary, involving physics, chemistry, biology, materials
science, and the full range of the engineering disciplines. The word nanotechnology is widely
used as shorthand to refer to both the science and the technology of this emerging field.
Narrowly defined, nanoscience concerns a basic understanding of physical, chemical, and
biological properties on atomic and near-atomic scales. Nanotechnology, narrowly defined,
employs controlled manipulation of these properties to create materials and functional systems
with unique capabilities.

In contrast to recent engineering efforts, nature developed “nanotechnologies” over billions of

years, employing enzymes and catalysts to organize with exquisite precision different kinds of
atoms and molecules into complex microscopic structures that make life possible. These natural
products are built with great efficiency and have impressive capabilities, such as the power to
harvest solar energy, to convert minerals and water into living cells, to store and process massive
amounts of data using large arrays of nerve cells, and to replicate perfectly billions of bits of
information stored in molecules of deoxyribonucleic acid (DNA).

There are two principal reasons for qualitative differences in material behaviour at the nanoscale
(traditionally defined as less than 100 nanometres). First, quantum mechanical effects come into
play at very small dimensions and lead to new physics and chemistry. Second, a defining feature
at the nanoscale is the very large surface-to-volume ratio of these structures. This means that no
atom is very far from a surface or interface, and the behaviour of atoms at these higher-energy
sites have a significant influence on<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388676/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009140039"></script> the properties
of the material. For example, the reactivity of a metal catalyst particle generally increases
appreciably as its size is reduced—macroscopic gold is chemically inert, whereas at nanoscales
gold becomes extremely reactive and catalytic and even melts at a lower temperature. Thus, at
nanoscale dimensions material properties depend on and change with size, as well as
composition and structure.

Using the processes of nanotechnology, basic industrial production may veer dramatically from
the course followed by steel plants and chemical factories of the past. Raw materials will come
from the atoms of abundant elements—carbon, hydrogen, and silicon—and these will be
manipulated into precise configurations to create nanostructured materials that exhibit exactly
the right properties for each particular application. For example, carbon atoms can be bonded
together in a number of different geometries to create variously a fibre, a tube, a molecular
coating, or a wire, all with the superior strength-to-weight ratio of another carbon material—
diamond. Additionally, such material processing need not require smokestacks, power-hungry
industrial machinery, or intensive human labour. Instead, it may be accomplished either by
“growing” new structures through some combination of chemical catalysts and synthetic
enzymes or by building them through new techniques based on patterning and self-assembly of
nanoscale materials into useful predetermined designs. Nanotechnology ultimately may allow
people to fabricate almost any type of material or product allowable under the laws of physics
and chemistry. While such possibilities seem remote, even approaching nature’s virtuosity in
energy-efficient fabrication would be revolutionary.

Even more revolutionary would be the fabrication of nanoscale machines and devices for
incorporation into micro- and macroscale systems. Once again, nature has led the way with the
fabrication of both linear and rotary molecular motors. These biological machines carry out such
tasks as muscle contraction (in organisms ranging from clams to humans) and shuttling little
packets of material around within cells while being powered by the recyclable, energy-efficient
fuel adenosine triphosphate. Scientists are only beginning to develop the tools to fabricate
functioning systems at such small scales, with most advances based on electronic or magnetic
information processing and storage systems. The energy-efficient, reconfigurable, and self-
repairing aspects of biological systems are just becoming understood.

The potential impact of nanotechnology processes, machines, and products is expected to be far-
reaching, affecting nearly every conceivable information technology, energy source, agricultural
product, medical device, pharmaceutical, and material used in manufacturing. Meanwhile, the
dimensions of electronic circuits on semiconductors continue to shrink,<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388673/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009140045"></script> with
minimum feature sizes now reaching the nanorealm, under 100 nanometres. Likewise, magnetic
memory materials, which form the basis of hard disk drives, have achieved dramatically greater
memory density as a result of nanoscale structuring to exploit new magnetic effects at
nanodimensions. These latter two areas represent another major trend, the evolution of critical
elements of microtechnology into the realm of nanotechnology to enhance performance. They
are immense markets driven by the rapid advance of information technology.

Milestones in the development of nanotechnology

Visionaries

In a lecture in 1959 to the American Physical Society, “There’s Plenty of Room at the Bottom,”
American Nobelist Richard P. Feynman presented his audience with a vision of what could be
done with extreme miniaturization. He began his lecture by noting that the Lord’s Prayer had
been written on the head of a pin and asked,

Why cannot we write the entire 24 volumes of the Encyclopædia Britannica on the head of a
pin? Let’s see what would be involved. The head of a pin is a sixteenth of an inch across. If you
magnify it by 25,000 diameters, the area of the head of the pin is then equal to the area of all the
pages of the Encyclopædia Britannica. Therefore, all it is necessary to do is to reduce in size all
the writing in the Encyclopædia by 25,000 times. Is that possible? The resolving power of the
eye is about 1/120 of an inch—that is roughly the diameter of one of the little dots on the fine
half-tone reproductions in the Encyclopædia. This, when you demagnify it by 25,000 times, is
still 80 angstroms in diameter—32 atoms across, in an ordinary metal. In other words, one of
those dots still would contain in its area 1,000 atoms. So, each dot can easily be adjusted in size
as required by the photoengraving, and there is no question that there is enough room on the head
of a pin to put all of the Encyclopædia Britannica.

Feynman was intrigued by biology and pointed out that

cells are very tiny, but they are very active; they manufacture various substances; they walk
around; they wiggle; and they do all kinds of marvelous things—all on a very small scale. Also,
they store information. Consider the possibility that we too can make a thing very small which
does what we want—that we can manufacture an object that maneuvers at that level!

He also considered using big tools to make smaller tools that could make yet smaller tools,
eventually obtaining nanoscale tools for directly manipulating atoms and molecules. In
considering what all this might mean, Feynman declared,

I can hardly doubt that when we have some control of the arrangement of things on a small scale
we will get an enormously greater range of possible properties that substances can have, and of
different things that we can do.

Perhaps the biggest barrier to following these prophetic thoughts was simply the immediate lack
of tools to manipulate and visualize matter at such a small scale. The availability of tools has
always been an enabling aspect of the advance of all science and technology, and some of the
key tools for nanotechnology are discussed in the next section, Pioneers.

Starting with a 1981 paper in the Proceedings of the National Academy of Sciences and
following with two popular books, Engines of Creation (1986) and Nanosystems (1992),
American scientist K. Eric Drexler became one of the foremost advocates of nanotechnology. In
fact, Drexler was the first person anywhere to receive a Ph.D. in molecular nanotechnology
(from the Massachusetts Institute of Technology). In his written works he takes a molecular view
of the world and envisions molecular machines doing much of the work of the future. For
example, he refers to “assemblers,” which will manipulate individual atoms to manufacture
structures, and “replicators,” which will be able to make multiple copies of themselves in order
to save time dealing with the billions of atoms needed to make objects of useful size. In an article
for Encyclopædia Britannica’s 1990 Yearbook of Science and the Future, Drexler wrote:

Cells and tissues in the human body are built and maintained by molecular machinery, but
sometimes that machinery proves inadequate: viruses multiply, cancer cells spread, or systems
age and deteriorate. As one might expect, new molecular machines and computers of subcellular
size could support the body’s own mechanisms. Devices containing nanocomputers interfaced to
molecular sensors and effectors could serve as an augmented immune system, searching out and
destroying viruses and cancer cells. Similar devices programmed as repair machines could enter
living cells to edit out viral DNA sequences and repair molecular damage. Such machines would
bring surgical control to the molecular level, opening broad new horizons in medicine.
Drexler’s futurist visions have stimulated much thought, but the assembler approach has failed to
account for the strong influence of atomic and molecular forces<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388668/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009179862"></script> (i.e., the
chemistry) at such dimensions. The controversy surrounding these popularizations, and the
potential dangers of entities such as intelligent replicators (however remote), have stimulated
debate over the ethical and societal implications of nanotechnology.

Pioneers

A number of key technological milestones have been achieved by working pioneers. Molecular
beam epitaxy, invented by Alfred Cho and John Arthur at Bell Labs in 1968 and developed in the
1970s, enabled the controlled deposition of single atomic layers. This tool provided for
nanostructuring in one dimension as atomic layers were grown one upon the next. It
subsequently became important in the area of compound semiconductor device fabrication. For
example, sandwiching one-nanometre-thick layers of nonmagnetic-sensor materials between
magnetic layers in computer disk drives resulted in large increases in storage capacity, and a
similar use of nanostructuring resulted in more energy-efficient semiconductor lasers for use in
compact disc players.

In 1981 Gerd Binnig and Heinrich Rohrer developed the scanning tunneling microscope at
IBM’s laboratories in Switzerland. This tool provided a revolutionary advance by enabling
scientists to image the position of individual atoms on surfaces. It earned Binnig and Rohrer a
Nobel Prize in 1986 and spawned a wide variety of scanning probe tools for nanoscale
observations.

The observation of new carbon structures marked another important milestone in the advance of
nanotechnology, with Nobel Prizes for the discoverers. In 1985 Robert F. Curl, Jr., Harold W.
Kroto, and Richard E. Smalley discovered the first fullerene, the third known form of pure
carbon (after diamond and graphite). They named their discovery buckminsterfullerene
(“buckyball”) for its resemblance to the geodesic domes promoted by the American architect R.
Buckminster Fuller. Technically called C60 for the 60 carbon atoms that form their hollow
spherical structure, buckyballs resemble a football one nanometre in diameter (see

). In 1991 Sumio Iijima of NEC Corporation in Japan discovered carbon

nanotubes, in which the carbon ringlike structures are extended from spheres into long tubes of
varying diameter. Taken together, these new structures surprised and excited the imaginations of
scientists about the possibilities of forming well-defined nanostructures with unexpected new
properties.

The scanning tunneling microscope not only allowed for the imaging of atoms by scanning a
sharp probe tip over a surface, but it also allowed atoms to be “pushed” around on the surface.
With a slight bias voltage applied to the probe tip, certain atoms could be made to adhere to the
tip used for imaging and then<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388672/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009180713"></script> to be released
from it. Thus, in 1990 Donald Eigler spelled out the letters of his company’s logo, IBM, by
moving 35 xenon atoms into place on a nickel surface. This demonstration caught the public’s
attention because it showed the precision of the emerging nanoscale tools.

Properties at the nanoscale

At nanoscale dimensions the properties of materials no longer depend solely on composition and
structure in the usual sense. Nanomaterials display new phenomena associated with quantized
effects and with the preponderance of surfaces and interfaces.

Quantized effects arise in the nanometre regime because the overall dimensions of objects are
comparable to the characteristic wavelength for fundamental excitations in materials. For
example, electron wave functions (see also de Broglie wave) in semiconductors are typically on
the order of 10 to 100 nanometres. Such excitations include the wavelength of electrons,
photons, phonons, and magnons, to name a few. These excitations carry the quanta of energy
through materials and thus determine the dynamics of their propagation and transformation from
one form to another. When the size of structures is comparable to the quanta themselves, it
influences how these excitations move through and interact in the material. Small structures may
limit flow, create wave interference effects, and otherwise bring into play quantum mechanical
selection rules not apparent at larger dimensions.

Electronic and photonic behaviour

Quantum mechanical properties for confinement of electrons in one dimension have long been
exploited in solid-state electronics. Semiconductor devices are grown with thin layers of
differing composition so that electrons (or “holes” in the case of missing electron charges) can be
confined in specific regions of the structure (known as quantum wells). Thin layers with larger
energy bandgaps can serve as barriers that restrict the flow of charges tocertain conditions under
which they can “tunnel” through these barriers—the basis of resonant tunneling diodes.
Superlattices are periodic structures of repeating wells that set up a new set of selection rules
which affect the conditions for charges to flow through the structure. Superlattices have been
exploited in cascade lasers to achieve far infrared wavelengths. Modern telecommunications is
based on semiconductor lasers that exploit the unique properties of quantum wells to achieve
specific wavelengths and high efficiency.

The propagation of photons is altered dramatically when the size and periodicity of the transient
structure approach the wavelength of visible light (400 to 800 nanometres). When photons
propagate through a periodically varying dielectric constant—for example, semiconductor posts
surrounded by air—quantum mechanical rules define and limit<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1371336/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009188497"></script> the
propagation of the photons depending on their energy (wavelength). This new behaviour is
analogous to the quantum mechanical rules that define the motion of electrons through crystals,
giving bandgaps for semiconductors. In one dimension, compound semiconductor superlattices
can be grown epitaxially with the alternating layers having different dielectric constants, thus
providing highly reflective mirrors for specific wavelengths as determined by the repeat distance
of layers in the superlattice. These structures are used to provide “built-in” mirrors for vertical-
cavity surface-emitting lasers, which are used in communications applications. In two and three
dimensions, periodic structures known as photonic crystals offer additional control over photon
propagation.

Photonic crystals are being explored in a variety of materials and periodicities, such as two-
dimensional hexagonal arrays of posts fabricated in compound semiconductors or stacked loglike
arrays of silicon bars in three dimensions. The dimensions of these structures depend on the
wavelength of light being propagated and are typically in the range of a few hundred nanometres
for wavelengths in the visible and near infrared. Photonic crystal properties based on
nanostructured materials offer the possibility of confining, steering, and separating light by
wavelength on unprecedented small scales and of creating new devices such as lasers that require
very low currents to initiate lasing (called near-thresholdless lasers). These structures are being
extensively investigated as the tools for nanostructuring materials are steadily advancing.
Researchers are particularly interested in the infrared wavelengths, where dimensional control is
not as stringent as at the shorter visible wavelengths and where optical communications and
chemical sensing provide motivation for potential new applications.

Magnetic, mechanical, and chemical behaviour

Nanoscale materials also have size-dependent magnetic behaviour, mechanical properties, and
chemical reactivity. At very small sizes (a few nanometres), magnetic nanoclusters have a single
magnetic domain, and the strongly coupled magnetic spins on each atom combine to produce a
particle with a single “giant” spin. For example, the giant spin of a ferromagnetic iron particle
rotates freely at room temperature for diameters below about 16 nanometres, an effect termed
superparamagnetism. Mechanical properties of nanostructured materials can reach exceptional
strengths. As a specific example, the introduction of two-nanometre aluminum oxide precipitates
into thin films of pure nickel results in yield strengths increasing from 0.15 to 5 gigapascals,
which is more than twice that for a hard bearing steel. Another example of exceptional
mechanical properties at the nanoscale is the carbon nanotube, which exhibits great strength and
stiffness along its longitudinal axis.

The preponderance of<script

src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388674/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009190424"></script> surfaces is a
major reason for the change in behaviour of materials at the nanoscale. Since up to half of all the
atoms in nanoparticles are surface atoms, properties such as electrical transport are no longer
determined by solid-state bulk phenomena. Likewise, the atoms in nanostructures have a higher
average energy than atoms in larger structures, because of the large proportion of surface atoms.
For example, catalytic materials have a greater chemical activity per atom of exposed surface as
the catalyst is reduced in size at the nanoscale. Defects and impurities may be attracted to
surfaces and interfaces, and interactions between particles at these small dimensions can depend
on the structure and nature of chemical bonding at the surface. Molecular monolayers may be
used to change or control surface properties and to mediate the interaction between
nanoparticles.

Surfaces and their interactions with molecular structures are basic to all biology. The intersection
of nanotechnology and biotechnology offers the possibility of achieving new functions and
properties with nanostructured surfaces. In this surface- and interface-dominated regime, biology
does an exquisite job of selectively controlling functions through a combination of structure and
chemical forces. The transcription of information stored in genes and the selectivity of
biochemical reactions based on chemical recognition of complex molecules are examples where
interfaces play the key role in establishing nanoscale behaviour. Atomic forces and chemical
bonds dominate at these dimensions, while macroscopic effects—such as convection, turbulence,
and momentum (inertial forces)—are of little consequence.

Nanotechnology research
Nanomaterials

As discussed in the section Properties at the nanoscale, material properties—electrical, optical,

magnetic, mechanical, and chemical—depend on their exact dimensions. This opens the way for
development of new and improved materials through manipulation of their nanostructure.
Hierarchical assemblies of nanoscale-engineered materials into larger structures, or their
incorporation into devices, provide the basis for tailoring radically new materials and machines.

Nature’s assemblies point the way to improving structural materials. The often-cited abalone
seashell provides a beautiful example of how the combination of a hard, brittle inorganic
material with nanoscale structuring and a soft, “tough” organic material can produce a strong,
durable nanocomposite—basically, these nanocomposites are made of calcium carbonate
“bricks” held together by a glycoprotein “glue.” New engineered materials are emerging—such
as polymer-clay nanocomposites—that are not only strong and tough but also lightweight and
easier to recycle than conventional reinforced plastics. Such improvements in structural materials
are particularly important for the transportation industry, where reduced weight directly
translates into<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388675/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009191788"></script> improved fuel
economy. Other improvements can increase safety or decrease the impact on the environment of
fabrication and recycling. Further advances, such as truly smart materials that signal their
impending failure or are even able to self-repair flaws, may be possible with composites of the
future.

Sensors are central to almost all modern control systems. For example, multiple sensors are used
in automobiles for such diverse tasks as engine management, emission control, security, safety,
comfort, vehicle monitoring, and diagnostics. While such traditional applications for physical
sensing generally rely on microscale sensing devices, the advent of nanoscale materials and
structures has led to new electronic, photonic, and magnetic nanosensors, sometimes known as
“smart dust.” Because of their small size, nanosensors exhibit unprecedented speed and
sensitivity, extending in some cases down to the detection of single molecules. For example,
nanowires made of carbon nanotubes, silicon, or other semiconductor materials exhibit
exceptional sensitivity to chemical species or biological agents. Electrical current through
nanowires can be altered by having molecules attached to their surface that locally perturb their
electronic band structure. By means of nanowire surfaces coated with sensor molecules that
selectively attach particular species, charge-induced changes in current can be used to detect the
presence of those species. This same strategy is adopted for many classes of sensing systems.
New types of sensors with ultrahigh sensitivity and specificity will have manyapplications ; for
example, sensors that can detect cancerous tumours when they consist of only a few cells would
be a very significant advance.

Nanomaterials also make excellent filters for trapping heavy metals and other pollutants from
industrial wastewater. One of the greatest potential impacts of nanotechnology on the lives of the
majority of people on Earth will be in the area of economical water desalination and purification.
Nanomaterials will very likely find important use in fuel cells, bioconversion for energy,
bioprocessing of food products, waste remediation, and pollution-control systems.

A recent concern regarding nanoparticles is whether their small sizes and novel properties may
pose significant health or environmental risks. In general, ultrafine particles—such as the carbon
in photocopier toners or in soot produced by combustion engines and factories—have adverse
respiratory and cardiovascular effects on people and animals. Studies are under way to determine
if specific nanoscale particles pose higher risks that may require special regulatory restrictions.
Of particular concern are potential carcinogenic risks from inhaled particles and the possibility
for very small nanoparticles to<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388669/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009191794"></script> cross the
blood-brain barrier to unknown effect. Nanomaterials currently receiving attention from health
officials include carbon nanotubes, buckyballs, and cadmium selenide quantum dots. Studies of
the absorption through the skin of titanium oxide nanoparticles (used in sunscreens) are also
planned. More far-ranging studies of the toxicity, transport, and overall fate of nanoparticles in
ecosystems and the environment have not yet been undertaken. Some early animal studies,
involving the introduction of very high levels of nanoparticles which resulted in the rapid death
of many of the subjects, are quite controversial.

Biomedicine and health care

Drug delivery

Nanotechnology promises to impact medical treatment in multiple ways. First, advances in

nanoscale particle design and fabrication provide new options for drug delivery and drug
therapies. More than half of the new drugs developed each year are not water-soluble, which
makes their delivery difficult. In the form of nanosized particles, however, these drugs are more
readily transported to their destination, and they can be delivered in the conventional form of
pills.
More important, nanotechnology may enable drugs to be delivered to precisely the right location
in the body and to release drug doses on a predetermined schedule for optimal treatment. The
general approach is to attach the drug to a nanosized carrier that will release the medicine in the
body over an extended period of time or when specifically triggered to do so. In addition, the
surfaces of these nanoscale carriers may be treated to seek out and become localized at a disease
site—for example, attaching to cancerous tumours. One type of molecule of special interest for
theseapplications is an organic dendrimer. A dendrimer is a special class of polymeric molecule
that weaves in and out from a hollow central region. These spherical “fuzz balls” are about the
size of a typical protein but cannot unfold like proteins. Interest in dendrimers derives from the
ability to tailor their cavity sizes and chemical properties to hold different therapeutic agents.
Researchers hope to design different dendrimers that can swell and release their drug on
exposure to specifically recognized molecules that indicate a disease target. This same general
approach to nanoparticle-directed drug delivery is being explored for other types of nanoparticles
as well.

Another approach involves gold-coated nanoshells whose size can be adjusted to absorb light
energy at different wavelengths. In particular, infrared light will pass through several centimetres
of body tissue, allowing a delicate and precise heating of such capsules in order to release<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388673/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009194944"></script> the
therapeutic substance within. Furthermore, antibodies may be attached to the outer gold surface
of the shells to cause them to bind specifically to certain tumour cells, thereby reducing the
damage to surrounding healthy cells.

Bioassays

A second area of intense study in nanomedicine is that of developing new diagnostic tools.
Motivation for this work ranges from fundamental biomedical research at the level of single
genes or cells to point-of-care applications for health delivery services. With advances in
molecular biology, much diagnostic work now focuses on detecting specific biological
“signatures.” These analyses are referred to as bioassays. Examples include studies to determine
which genes are active in response to a particular disease or drug therapy. A general approach
involves attaching fluorescing dye molecules to the target biomolecules in order to reveal their
concentration.

Another approach to bioassays uses semiconductor nanoparticles, such as cadmium selenide,

which emit light of a specific wavelength depending on their size. Different-size particles can be
tagged to different receptors so that a wider variety of distinct colour tags are available than can
be distinguished for dye molecules. The degradation in fluorescence with repeated excitation for
dyes is avoided. Furthermore, various-size particles can be encapsulated in latex beads and their
resulting wavelengths read like a bar code. This approach, while still in the exploratory stage,
would allow for an enormous number of distinct labels for bioassays.

Another nanotechnology variation on bioassays is to attach one half of the single-stranded

complementary DNA segment for the genetic sequence to be detected to one set of gold particles
and the other half to a second set of gold particles. When the material of interest is present in a
solution, the two attachments cause the gold balls to agglomerate, providing a large change in
optical properties that can be seen in the colour of the solution. If both halves of the sequence do
not match, no agglomeration will occur and no change will be observed.

Approaches that do not involve optical detection techniques are also being explored with
nanoparticles. For example, magnetic nanoparticles can be attached to antibodies that in turn
recognize and attach to specific biomolecules. The magnetic particles then act as tags and
“handlebars” through which magnetic fields can be used for mixing, extracting, or identifying
the attached biomolecules within microlitre- or nanolitre-sized samples. For example, magnetic
nanoparticles stay magnetized as a single domain for a significant period, which enables them to
be<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388671/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009192724"></script> aligned and
detected in a magnetic field. In particular, attached antibody–magnetic-nanoparticle
combinations rotate slowly and give a distinctive magnetic signal. In contrast, magnetically
tagged antibodies that are not attached to the biological material being detected rotate more
rapidly and so do not give the same distinctive signal.

Microfluidic systems, or “labs-on-chips,” have been developed for biochemical assays of

minuscule samples. Typically cramming numerous electronic and mechanical components into a
portable unit no larger than a credit card, they are especially useful for conducting rapid analysis
in the field. While these microfluidic systems primarily operate at the microscale (that is,
millionths of a metre), nanotechnology has contributed new concepts and will likely play an
increasing role in the future. For example, separation of DNA is sensitive to entropic effects,
such as the entropy required to unfold DNA of a given length. A new approach to separating
DNA could take advantage of its passage through a nanoscale array of posts or channels such
that DNA molecules of different lengths would uncoil at different rates.

Other researchers have focused on detecting signal changes as nanometre-wide DNA strands are
threaded through a nanoscale pore. Early studies used pores punched in membranes by viruses;
artificially fabricated nanopores are also being tested. By applying an electric potential across the
membrane in a liquid cell to pull the DNA through, changes in ion current can be measured as
different repeating base units of the molecule pass through the pores. Nanotechnology-enabled
advances in the entire area of bioassays will clearly impact health care in many ways, from early
detection, rapid clinical analysis, and home monitoring to new understanding of molecular
biology and genetic-based treatments for fighting disease.

Assistive devices and tissue engineering

Another biomedical application of nanotechnology involves assistive devices for people who
have lost or lack certain natural capabilities. For example, researchers hope to design retinal
implants for vision-impaired individuals. The concept is to implant chips with photodetector
arrays to transmit signals from the retina to the brain via the optic nerve. Meaningful spatial
information, even if only at a rudimentary level, would be of great assistance to the blind. Such
research illustrates the tremendous challenge of designing hybrid systems that work at the
interface between inorganic devices and biological systems.
Closely related research involves implanting nanoscale neural probes in brain tissue to activate
and control motor functions. This requires effective and stable “wiring” of many electrodes to
neurons. It is exciting because of the<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388676/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009194172"></script> possibility of
recovery of control for motor-impaired individuals. Studies employing neural stimulation of
damaged spinal cords by electrical signals have demonstrated the return of some locomotion.
Researchers are also seeking ways to assist in the regeneration and healing of bone, skin, and
cartilage—for example, developing synthetic biocompatible or biodegradable structures with
nanosized voids that would serve as templates for regenerating specific tissue while delivering
chemicals to assist in the repair process. At a more sophisticated level, researchers hope to
someday build nanoscale or microscale machines that can repair, assist, or replace more-complex
organs.

Information technology

Semiconductor experts agree that the ongoing shrinkage in “conventional” electronic devices
will inevitably reach fundamental limits due to quantum effects such as “tunneling,” in which
electrons jump out of their prescribed circuit path and create atomic-scale interference between
devices. At that point, radical new approaches to data storage and information processing will be
required for further advances. For example, radically new systems have been imagined that are
based on quantum computing or biomolecular computing.

Molecular electronics

The use of molecules for electronic devices was suggested by Mark Ratner of Northwestern
University and Avi Aviram of IBM as early as the 1970s, but proper nanotechnology tools did
not become available until the turn of the 21st century. Wiring up molecules some half a
nanometre wide and a few nanometres long remains a major challenge, and an understanding of
electrical transport through single molecules is only beginning to emerge. A number of groups
have been able to demonstrate molecular switches, for example, that could conceivably be used
in computer memory or logic arrays. Current areas of research include mechanisms to guide the
selection of molecules, architectures for assembling molecules into nanoscale gates, and three-
terminal molecules for transistor-like behaviour. More-radical approaches include DNA
computing, where single-stranded DNA on a silicon chip would encode all possible variable
values and complementary strand interactions would be used for a parallel processing approach
to finding solutions. An area related to molecular electronics is that of organic thin-film
transistors and light emitters, which promise new applications such as video displays that can be
rolled out like wallpaper and flexible electronic newspapers.

Nanotubes and nanowires

Carbon nanotubes have remarkable electronic, mechanical, and chemical properties. Depending
on their specific diameter and the bonding arrangement of their carbon atoms, nanotubes exhibit
either metallic or semiconducting behaviour. Electrical conduction within a perfect nanotube is
ballistic (negligible scattering), with low thermal dissipation. As a result, a wire<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388670/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009241626"></script> made from a
nanotube, or a nanowire, can carry much more current than an ordinary metal wire of
comparable size. At 1.4 nanometres in diameter, nanotubes are about a hundred times smaller
than the gate width of silicon semiconductor devices. In addition to nanowires for conduction,
transistors, diodes, and simple logic circuits have been demonstrated by combining metallic and
semiconductor carbon nanotubes. Similarly, silicon nanowires have been used to build
experimental devices, such as field-effect transistors, bipolar transistors, inverters, light-emitting
diodes, sensors, and even simple memory. A major challenge for nanowire circuits, as for
molecular electronics, is connecting and integrating these devices into a workable high-density
architecture. Ideally, the structure would be grown and assembled in place. Crossbar
architectures that combine the function of wires and devices are of particular interest.

Single-electron transistors

At nanoscale dimensions the energy required to add one additional electron to a “small island”
(isolated physical region)—for example, through a tunneling barrier—becomes significant. This
change in energy provides the basis for devising single-electron transistors. At low temperatures,
where thermal fluctuations are small, various single-electron-device nanostructures are readily
achievable, and extensive research has been carried out for structures with confined electron
flow. However, room-temperature applications will require that sizes be reduced significantly, to
the one-nanometre range, to achieve stable operation. For large-scale application with millions of
devices, as found in current integrated circuits, the need for structures with very uniform size to
maintain uniform device characteristics presents a significant challenge. Also, in this and many
new nanodevices being explored, the lack of gain is a serious drawback limiting implementation
in large-scale electronic circuits.

Spintronics

Spintronics refers to electronic devices that perform logic operations based on not just the
electrical charge of carriers but also their spin. For example, information could be transported or
stored through the spin-up or spin-down states of electrons. This is a new area of research, and
issues include the injection of spin-polarized carriers, their transport, and their detection. The
role of nanoscale structure and electronic properties of the ferromagnetic-semiconductor
interface on the spin injection process, the growth of new ferromagnetic semiconductors with
nanoscale control, and the possible use of nanostructured features to manipulate spin are all of
interest.

Information storage

Current approaches to information storage and retrieval include high-density, high-speed, solid-
state electronic memories, as well as slower (but generally more spacious) magnetic and optical
discs (see computer memory). As the minimum feature size for electronic processing approaches
100 nanometres, nanotechnology<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388671/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009133634"></script> provides ways
to decrease further the bit size of the stored information, thus increasing density and reducing
interconnection distances for obtaining still-higher speeds. For example, the basis of the current
generation of magnetic disks is the giant magnetoresistance effect. A magnetic read/write head
stores bits of information by setting the direction of the magnetic field in nanometre-thick
metallic layers that alternate between ferromagnetic and nonferromagnetic. Differences in spin-
dependent scattering of electrons at the interface layers lead to resistance differences that can be
read by the magnetic head. Mechanical properties, particularly tribology (friction and wear of
moving surfaces), also play an important role in magnetic hard disk drives, since magnetic heads
float only about 10 nanometres above spinning magnetic disks.

Another approach to information storage that is dependent on designing nanometre-thick

magnetic layers is under commercial development. Known as magnetic random access memory
(MRAM), a line of electrically switchable magnetic material is separated from a permanently
magnetized layer by a nanoscale nonmagnetic interlayer. A resistance change that depends on the
relative alignment of the fields is read electrically from a large array of wires through cross lines.
MRAM will require a relatively small evolution from conventional semiconductor
manufacturing, and it has the added benefit of producing nonvolatile memory (no power or
batteries are needed to maintain stored memory states).

Still at an exploratory stage, studies of electrical conduction through molecules have generated
interest in their possible use as memory. While still very speculative, molecular and nanowire
approaches to memory are intriguing because of the small volume in which the bits of memory
are stored and the effectiveness with which biological systems store large amounts of
information.

Communications

Nanoscale structuring of optical devices, such as vertical-cavity surface-emitting lasers

(VCSELs), quantum dot lasers, and photonic crystal materials, is leading to additional advances
in communications technology.

VCSELs have nanoscale layers of compound semiconductors epitaxially grown into their
structure—alternating dielectric layers as mirrors and quantum wells. Quantum wells allow the
charge carriers to be confined in well-defined regions and provide the energy conversion into
light at desired wavelengths. They are placed in the laser’s cavity to confine carriers at the nodes
of a standing wave and to tailor the band structure for more efficient radiative recombination.
One-dimensional nanotechnology techniques involving precise growth of very thin epitaxial
semiconductor layers were developed during the 1990s. Such nanostructuring has enhanced the
efficiency of VCSELs and reduced the current required for lasing to<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1371336/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;kvchannel=SCIENCE;misc=1300009124196">
</script> start (called the threshold current). Because of improving performance and their
compatibility with planar manufacturing technology, VCSELs are fast becoming a preferred
laser source in a variety of communicationsapplications.
More recently, the introduction of quantum dots (regions so small that they can be given a single
electric charge) into semiconductor lasers has been investigated and found to give additional
benefits—both further reductions in threshold current and narrower line widths. Quantum dots
further confine the optical emission modes within a very narrow spectrum and give the lowest
threshold current densities for lasing achieved to date in VCSELs. The quantum dots are
introduced into the laser during the growth of strained layers, by a process called Stransky-
Krastanov growth. They arise because of the lattice mismatch stress and surface tension of the
growing film. Improvements in ways to control precisely the resulting quantum dots to a more
uniform single size are still being sought.

Photonic crystals provide a new means to control the steering and manipulation of photons based
on periodic dielectric lattices with repeat dimensions on the order of the wavelength of light.
These materials can have very exotic properties, such as not allowing light within certain
wavelengths to be propagated in a material based on the particular periodic structure. Photonic
lattices can act as perfect wavelength-selective mirrors to reflect back incident light from all
orientations. They provide the basis for optical switching, steering, and wavelength separation on
unprecedented small scales. The periodic structures required for these artificial crystals can be
configured as both two- and three-dimensional lattices. Optical sources, switches, and routers are
being considered, with two-dimensional planar geometries receiving the most attention, because
of their greater ease of fabrication.

Another potentially important communications application for nanotechnology is

microelectromechanical systems (MEMS), devices sized at the micrometre level (millionths of a
metre). MEMS are currently poised to have a major impact on communications via optical
switching. In the future, electromechanical devices may shrink to nanodimensions to take
advantage of the higher frequencies of mechanical vibration at smaller masses. The natural
(resonant) frequency of vibration for small mechanical beams increases as their size decreases,
so that little power is needed to drive them as oscillators. Their efficiency is rated by a quality
factor, known as Q, which is a ratio of the energy stored per cycle versus the energy dissipated
per cycle. The higher the Q, the more precise the absolute frequency of an<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388674/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;kvchannel=SCIENCE;misc=1300009124202">
</script> oscillator. The Q is very high for micro- and nanoscale mechanical oscillators, and
these devices can reach very high frequencies (up to microwave frequencies), making them
potential low-power replacements for electronic-based oscillators and filters.

Mechanical oscillators have been made from silicon at dimensions of 10 × 100 nanometres,
where more than 10 percent of the atoms are less than one atomic distance from the surface.
While highly homogeneous materials can be made at these dimensions—for example, single-
crystal silicon bars—surfaces play an increasing role at nanoscales, and energy losses increase,
presumably because of surface defects and molecular species absorbed on surfaces.

It is possible to envision even higher frequencies, in what might be viewed as the ultimate in
nanomechanical systems, by moving from nanomachined structures to molecular systems. As an
example, multiwalled carbon nanotubes are being explored for their mechanical properties.
When the ends of the outer nanotube are removed, the inner tube may be pulled partway out
from the outer tube where van der Waals forces between the two tubes will supply a restoring
force. The inner tube can thus oscillate, sliding back and forth inside the outer tube. The resonant
frequency of oscillation for such structures is predicted to be above one gigahertz (one billion
cycles per second). It is unknown whether connecting such systems to the macro world and
protecting them from surface effects will ever be practical.

Nanofabrication

Two very different paths are pursued. One is a top-down strategy of

miniaturizing current technologies, while the other is a bottom-up strategy of building ever-
more-complex molecular devices atom by atom. Top-down approaches are good for producing
structures with long-range order and for making macroscopic connections, while bottom-up
approaches are best suited for assembly and establishing short-range order at nanoscale
dimensions. The integration of top-down and bottom-up techniques is expected to eventually
provide the best combination of tools for nanofabrication. Nanotechnology requires new tools for
fabrication and measurement.

Top-down approach

The most common top-down approach to fabrication involves lithographic patterning techniques
using short-wavelength optical sources. A key advantage of the top-down approach—as
developed in the fabrication of integrated circuits—is that the parts are both patterned and built
in place, so that no assembly step is needed. Optical lithography is a relatively mature field
because of the high degree of refinement in microelectronic chip manufacturing, with current
short-wavelength optical lithography techniques reaching dimensions just below 100 nanometres
(the traditional threshold definition of the nanoscale). Shorter-wavelength sources, such<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388675/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009126677"></script> as extreme
ultraviolet and X-ray, are being developed to allow lithographic printing techniques to reach
dimensions from 10 to 100 nanometres. Scanning beam techniques such as electron-beam
lithography provide patterns down to about 20 nanometres. Here the pattern is written by
sweeping a finely focused electron beam across the surface. Focused ion beams are also used for
direct processing and patterning of wafers, although with somewhat less resolution than in
electron-beam lithography. Still-smaller features are obtained by using scanning probes to
deposit or remove thin layers.

Mechanical printing techniques—nanoscale imprinting, stamping, and molding—have been

extended to the surprisingly small dimensions of about 20 to 40 nanometres. The details of these
techniques vary, but they are all based on making a master “stamp” by a high-resolution
technique such as electron-beam lithography and then applying this stamp, or subsequent
generations of it, to a surface to create the pattern. In one variation a stamp’s surface is coated
with a very thin layer of material (the “ink”) that can then be deposited (“inked”) directly onto
the surface to reproduce the stamp’s pattern. For example, the controlled patterning of a
molecular monolayer on a surface can be achieved by stamping an ink of thiol functionalized
organic molecules directly onto a gold-coated surface (molecules that contain a sulfur end group,
called a thiol, bond strongly to gold). In another approach the stamp is used mechanically to
press the pattern into a thin layer of material. This surface layer is typically a polymeric material
that has been made pliable for the molding process by being heated during the stamping
procedure. Plasma etching can then be used to remove the thin layer of the masking material
under the stamped regions; any residual polymer is thus removed, and a nanoscale lithographic
pattern is left on the surface. Still another variation is to make the relief pattern out of photoresist
on a silicon wafer by optical or electron-beam lithography and then pour a liquid precursor—for
example, polydimethylsiloxane, a form of silicone—over the pattern and then cure it. The result
is a rubbery solid that can be peeled off and used as a stamp. These stamps can be inked and
printed as described above, or they can be pressed to the surface and a liquid polymer allowed to
flow into the raised regions of the mask by capillary action and cured in place. A distinction for
this latter approach is<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388669/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009126682"></script> that the stamp
is flexible and can thus be used to print nanoscale features on curved surfaces.

These nanoscale printing techniques offer several advantages beyond the ability to use a wider
variety of materials with curved surfaces. In particular, such approaches can be carried out in
ordinary laboratories with far-less-expensive equipment than that needed for conventional
submicron lithography. The challenge for all top-down techniques is that, while they work well
at the microscale (at millionths of a metre), it becomes increasingly difficult to apply them at
nanoscale dimensions. A second disadvantage is that they involve planar techniques, which
means that structures are created by the addition and subtraction of patterned layers (deposition
and etching), so arbitrary three-dimensional objects are difficult to construct.

Bottom-up approach

Bottom-up, or self-assembly, approaches to nanofabrication use chemical or

physical forces operating at the nanoscale to assemble basic units into larger structures. As
component size decreases in nanofabrication, bottom-up approaches provide an increasingly
important complement to top-down techniques. Inspiration for bottom-up approaches comes
from biological systems, where nature has harnessed chemical forces to create essentially all the
structures needed by life. Researchers hope to replicate nature’s ability to produce small clusters
of specific atoms, which can then self-assemble into more-elaborate structures.

A number of bottom-up approaches have been developed for producing nanoparticles, ranging
from condensation of atomic vapours on surfaces to coalescence of atoms in liquids. For
example, liquid-phase techniques based on inverse micelles (globules of lipid molecules floating
in a nonaqueous solution in which their polar, or hydrophilic, ends point inward to form a hollow
core, as shown in the ) have been developed to produce size-selected
nanoparticles of semiconductor, magnetic, and other materials. An example of self-assembly that
achieves a limited degree of control over both formation and organization is the growth of
quantum dots. Indium gallium arsenide (InGaAs) dots can be formed by growing thin layers of
InGaAs on GaAs in such a manner that repulsive forces caused by compressive strain in the
InGaAs layer results in the formation of isolated quantum dots. After the growth of multiple
layer pairs, a fairly uniform spacing of the dots can be achieved. Another example of self-
assembly of an intricate structure is the formation of carbon nanotubes under the right set of
chemical and temperature conditions.

DNA-assisted assembly may provide a method to integrate hybrid heterogeneous parts into a
single device. Biology does this very well, combining self-assembly and self-organization
in<script
src="http://adserver.adtechus.com/addyn/3.0/5308.1/1388668/0/170/ADTECH;target=_blank;gr
p=741;key=false;kvqsegs=D;kvtopicid=962484;misc=1300009262097"></script> fluidic
environments where weaker electrochemical forces play a significant role. By using DNA-like
recognition, molecules on surfaces may be able to direct attachments between objects in fluids.
In this approach, polymers made with complementary DNA strands would be used as intelligent
“adhesive tape,” attaching between polymers only when the right pairing is present. Such
assembly might be combined with electrical fields to assist in locating the attachment sites and
then be followed by more-permanent attachment approaches, such as electrodeposition and
metallization. There are several advantages of DNA-assisted approaches: DNA molecules can be
sequenced and replicated in large quantities, DNA sequences act as codes that can be used to
recognize complementary DNA strands, hybridized DNA strands form strong bonds to their
complementary sequence, and DNA strands can be attached to different devices as labels. These
properties are being explored for ways to self-assemble molecules into nanoscale units. For
example, sequences of DNA have been fabricated that adhere only to particular crystal faces of
compound semiconductors, providing a basis for self-assembly. By having the correct
complementary sequences at the other end of the DNA molecule, certain faces of small
semiconductor building blocks can be made that adhere to or repel each other. For example, thiol
groups at the end of molecules cause them to attach to gold surfaces, while carboxyl groups can
be used for attachment to silica surfaces. Directed assembly is an increasingly important
variation of self-assembly where, in quasi-equilibrium environments, parts are moved
mechanically, electrically, or magnetically and are placed precisely where they are intended to
go.