A guide to nanotechnology
David Briddock reveals the immense potential of a world filled with nanotechnology
Nanotechnology is already here and far more widespread than you might imagine. It’s used by the manufacturing, chemical and power generation industries. It can be found in clinics and hospitals. And it’s increasingly found in everyday consumables and household products.
Washing powder manufacturers have used nanoparticles for years to enhance dirt removal and create whiter whites. Emulsion paint is likely to contain nanoparticles to add durability and reduce colour fading in UV light. And cars use nanoparticles for tougher protective paint coatings and engine emission catalytic cleaning.
Every day new nanotechnology research is performed by physicists, chemists, biologists and engineers. This research has the potential to revolutionise many aspects of industry and society including health and diet, home and work environments, transport and electronic gadgets.
However, in this particular article, I’m going to focus on nano applications in the field of electronics and their impact on our future computing technology.
What is nanotechnology?
Nano comes from the Greek word ‘nanos’, meaning dwarf, and indicates something is incredibly tiny. How tiny? Well, the generally accepted criteria is up to 100 nanometers (nm) in size. To put this in perspective, a billion one nanometer particles stacked end-to-end would only be a metre long. For comparison, a DNA helix has a diameter of about 2nm and proteins range in size from 3nm to 20nm, while a typical bacterium is around 1,000nm long.
We are constantly surrounded by nanoparticles. They are emitted by wood fires, diesel engine exhausts, vacuum cleaners, laser printers and many other sources. Nevertheless, great care needs to be exercised, as nanoparticles are small enough to float freely in the atmosphere, enter the lungs and pass through cell membranes around the body - including the brain.
One of the primary reasons nanoparticles are so interesting is they exhibit quite different properties and capabilities compared to the very same material in bulk form, just as predicted by the visionary scientist Richard Feynman in his famous 1959 lecture ‘There’s Plenty of Room at the Bottom’.
For an example of just how strange a nano-sized particle can be, let’s consider gold. Depending on its size, and therefore the number of atoms it contains, a gold nanoparticle will appear to change colour. The colour we perceive is caused by light scattering off the surface, but with so few atoms, this scattering is radically altered. Secondly, in its bulk form, gold is classed as an inert (or non-reactive) material, yet a 5nm gold particle has quite different properties and acts as a catalyst to other chemical reactions. And while a gold ring will melt at a toasty 1,024ºC, a 2nm gold nanoparticle has a melting point of around 500ºC.
The aim of nanotechnology research is to take full advantage of these size-dependent properties and functions to create novel, innovative and often ground-breaking applications.
As a nanoparticle is in essence simply a small collection of atoms, it can be constructed from just about any solid element in the periodic table. Carbon, iron, platinum, gold, silver, silica and silicon and titanium dioxide have all received considerable attention from scientists and engineers. Each element has its own very distinct set of nano-scale properties and capabilities, and therefore its own specific set of potential application scenarios.
A high percentage of nanotechnology research effort and funding to date has been focused on carbon-based particles. Carbon is a vital ingredient to life and a key component in a wide range of natural compounds and man-made materials.
Carbon’s ability to form long molecular chains and its willingness to readily bond to many other elements, has raised intense interest in its nano-scale potential.
One of the first carbon nanoparticles to be discovered was carbon buckyballs (also known as fullerenes). Identified in 1985 by Richard Smalley, Harry Kroto and Robert Curl at the Rice University, a buckyball links each carbon atom to three adjacent carbon atoms in a spherical pentagon and hexagon formation - reminiscent of an inflated soccer ball.
A buckyball can be built from 20, 60, 70 or more carbon atoms. The 60- and 70- atom variants are the most commonly found and are often referred to as C60 and C70. These spherical buckyballs can be added during the fabrication process to create a nanocomposite materials with increased strength and robustness. Buckyballs also exhibit interesting electrical properties, which has interested researchers aiming to increase the efficiency of solar cells.
Naturally occurring carbon nanotubes were another significant scientific discovery, although this time the ‘who’ and ‘when’ is hotly disputed. A carbon nanotube has a similar lattice structure to a buckyball, but this time the carbon-to-carbon bonds form a cylinder. These nanotubes can be as small as 1nm in diameter. The ends can either be open or closed, the closed version having a domed, half-buckyball type of termination.
Carbon nanotubes are strong and resilient, with an ability to ‘spring’ back to their original shape after bending. It’s also possible to fabricate multi-walled nanotubes for even greater strength and resilience. Adding nanotubes to composite materials will significantly increase their strength-to-weight ratio - a tactic that’s frequently employed by organisations like Nasa, the military and F1 racing teams.
However, it’s a nanotube’s electrical conductor properties, which are superior to copper, that particularly attracts computer scientists. The specific nature of these electrical properties depends on the orientation of the hexagons in the nanotube wall, and when other molecules attach themselves to a nanotube the electrical resistance changes significantly. Unsurprisingly, this unusual behaviour is of interest to researchers working with nano-scale electronic components, sensors and circuit board designs.
In 2004, at the University of Manchester, Andre Geim and Konstantin Novoselov successfully produced sheets of graphene, for which they were awarded the 2010 Nobel Prize for Science. A two-dimensional graphene sheet is the thickness of a single carbon atom, namely 0.22nm, so nearly five million graphene layers are required to achieve a thickness of one millimetre. Amazingly, though, the discovery process centred around a lump of naturally occurring graphite, some sticky tape and a pair of tweezers.
In structure, a graphene layer is rather like a flattened nanotube and makes for an incredibly strong material with a high tensile strength (meaning it can be stretched without breaking). For example, experiments suggest a multi-layer sheet of graphene the thickness of a piece of clingfilm could support the weight of a fully grown elephant. When mixed in with other materials, say in an epoxy resin, the resulting composite will be far stronger than a similar material infused with buckyballs or nanotubes.
Yet once again it’s graphene’s electrical properties that particularly excite researchers. The speed of electron flow across a graphene sheet is superior to any other known material and a full one thousand times faster than copper. One goal is to replace the de facto standard silicon wafer technology with graphene sheets to deliver incredibly fast switching transistors and high-speed integrated circuitry.
Modern microprocessors contain hundreds of millions of field-effect transistors (FET). These FETs act as tiny electronic switches using a substrate-insulator-gate sandwich (see image). A reduction in FET size means higher processing speeds combined with lower power consumption, plus the potential to squeeze millions more onto a single integrated circuit die.
CPU manufacturers apply great ingenuity to constantly miniaturise these transistors and have kept pace with the infamous Moore’s Law prediction of chip performance doubling every two years. In fact, to date it’s invariably doubled every 18 months. The FETs in today’s processors are already tiny. Today’s 32nm or 22nm structure sizes already fall well within the sub-100nm nanotechnology range.
To maintain this momentum, manufacturers are now investigating a modified FET structure that rises from the substrate layer like a fin (see image), hence the moniker ‘finFET’. Chips constructed with finFET transistors should keep the Moore’s Law roadshow alive for a little while longer, with transistor sizes of 14nm a distinct possibility.
However, even finFET technology will only take us a little further. With the photolithography techniques currently used to design and create the integrated circuits, we’re about to hit the boundaries of what’s physically possible. Current leakage between the ever thinner sandwich layers is just one in a long list of issues.
Nanowire technology is seen by many as the next big step. With nanowires, the FET is in the form of a cylinder, with the insulator and gate wrapped around the central nanowire core (see image). This construction allows greater control of the voltage than a finFET component, and their vertical orientation allows them to be densely stacked together, creating a nanowire FET forest numbered in billions.
Memory chips are critical components in any computing technology. Prices have dropped dramatically over the last few years, but it’s still a significant part of the overall cost for many devices - especially for under £50 ebook tablets and music players.
Once again, nanowire technology is likely to have a big impact in the next generation of memory circuitry. Researchers in Hewlett Packard’s labs have been working with nanowires coated with titanium dioxide. When another nanowire is laid in a perpendicular fashion over a group of these coated nanowires, a ‘memister’ device is created at each junction. The switch formed at each memister junction can represent a binary one or zero value.
Meanwhile, IBM researchers are using magnetic nanowires made from an iron and nickel alloy. They plan to place hundreds of millions of these nanowires onto a silicon substrate to increase memory storage capacities. And researchers at the Rice University are layering 5nm diameter silicon dioxide nanowires in a triple-layer sandwich to achieve dramatic increases in information storage.
Other labs are trying a different approach by fabricating arrays of magnetic particles called ‘nanodots’, each around 6nm in diameter. Many billions of nanodots will be needed for a single memory device. The magnetic charge level of each dot is read to determine if it represents a binary zero or one value.
Everyone of these organisations is in a billion-dollar race to find the best solution, yet developing ground-breaking nanotechnology isn’t enough. To be successful it must be backed up by equally innovative manufacturing techniques and technologies, able to fabricate billions of high-density, low-cost memory chips.
Researchers at various universities and some Asia-based manufacturing companies have already demonstrated their ability to use nanowires as the electrodes in organic light-emitting diode (OLED) displays. The next generation of OLEDs will include robust nanowire technology to create extremely thin, lightweight screens. Importantly, these displays would require far less power to display the same levels of brightness, contrast and colour saturation.
Unlike the previous OLED components, nanowires exhibit an ability to flex while continuing to operate. If the nanowires are deposited onto plastic sheets, the whole display will be able to bend. Flexible displays would be a revolutionary technology for gadget designers, who could envisage all kinds of novel and innovative scenarios. Foldable e-readers, bendy iPads and screens that wrap around your wrist are just some of the possibilities. Also, flexible devices will be far less likely to suffer damage after a heavy impact or drop.
Unfortunately, at present this advanced display technology requires expensive manufacturing techniques. However, it’s only a matter of time before engineers crack the mass production issues, and our handheld and pocketable devices take on new and futuristic appearance.
One of the best funded and most promising areas of nanotechnology research involves improving power storage. When applied to battery technology, it will lead to gadgets, portable computers and electric cars that weigh less, run for longer and charge much more rapidly.
So what does nanotechnology offer in this area? Well, as the particle size decreases, the number of particles within a specific volume increases. Greater numbers of anode and cathode particles means an increase on effective surface area and a more efficient battery. In other words, adding nanowires, nanotubes and other nanoparticles to batteries will deliver greater storage capacities or smaller and lighter batteries with the same capacity.
Research reports suggest it should be possible to achieve a tenfold increase in power density. This kind of progress would dramatically decrease battery size and radically change the look of our laptops, tablets and mobile phones.
Rapid recharge times are just as important as charge capacity. Owning portable devices where the battery doesn’t last all day or an electric car that has a range of just 100 miles is far more acceptable if the battery can be recharged in a minute or two.
Fast recharge technology research looks to be well advanced. In August 2012, there were news reports that Korean scientists have created a nanoparticle-enhanced lithium-ion battery, which can be recharged 30 to 102 times quicker than a typical lithium-ion battery. Such reports suggest the 60-second battery recharge scenario is a distinct possibility.
Capacitors act a little like batteries in that they store electrical charge. However, unlike batteries, capacitors can be charged extremely quickly and have very long usable lifetimes. Unfortunately, to store reasonable amounts of charge with current capacitor technology requires a prohibitively large and bulky component.
When nanotube technology is used to increase the internal surface area, the capacitor suddenly becomes a much more interesting prospect. Nano-enhanced capacitors could replace batteries in certain situations - possibly even in rapidly rechargeable electric cars - and a pocket-size nano-enhanced capacitor would certainly be useful as a top-up power source for our power-hungry mobile gadgets.
Efficient solar cells
An often advocated solution to power-hungry gadgets is to cover them with solar cells. Yet today’s first- and second-generation solar cell technology doesn’t have either the efficiency or robustness to make that a practical proposition. However, utilise buckyball and nanotube particles and it’s possible to create considerably more robust and efficient solar cells.
These nano-enhanced solar cells could start to appear in a whole range of consumer goods, including hi-tech clothing and fabric gadget cases. Some pundits go much further and envisage solar-cell-coated road surfaces, able to generate enough energy to power whole communities.
Certain types of nanoparticle make excellent sensors. Carbon nanotubes, zinc oxide nanowires and palladium nanoparticles all have their own specific sensor capabilities. Many are able to detect tiny amounts of chemical vapours in the air or hazardous bacterial molecules. When these molecules come into contact with other molecules, the tiny changes in their electrical resistance or capacitance can be measured.
At present, nanosensor manufacturing costs are far too high. The hope is that innovative research will lead to more efficient manufacturing processes that will slash prices, just as has happened with integrated circuits, which can cost pennies to produce.
You might be asking what relevance these sensors have to computing technology. Remember the medical tricorder from Star Trek? Well, with nanosensor technology your smartphone could become a handheld health-check device, able to detect, analyse and display heart rate, blood pressure, glucose levels and a whole host of other medical information.
A similarly equipped smartphone could ‘scan’ food products and kitchen work surfaces for potentially harmful bacteria. Also, it would be possible to produce handheld devices that can monitor air quality, detect deadly carbon monoxide gas leaks and even ‘sniff out’ explosives or drugs just as effectively as a trained dog.
To keep up with Moore’s Law, each new generation of integrated circuitry must be crammed with more and more transistors. In fact, the prediction suggests we will see integrated circuits with single-atom transistors around 2020. However, every advance in transistor miniaturisation significantly increases the research challenges and manufacturing costs.
Photolithography technology uses a photo-etching process where light is shone through intricate masks and reduction lenses to create fantastically complex integrated circuit designs. With current photolithography technology, manufacturers can produce integrated circuits with structures sized at the 22nm level. These are amazing technological achievements. At 22nm, it takes the movement of less than 10 individual electrons to cause a switching event.
To attain the necessary resolution, manufacturers use light with very small wavelengths. An extreme ultraviolet light beam has a wavelength in the 70nm to 10nm range, while x-rays are better still, with a range from 10nm right down to 0.1nm. Using these light sources, photolithography is expected to deliver structures at the 14nm level over the next year or so.
Nevertheless, even these advanced techniques have their limitations, which will stop Moore’s Law in its tracks quite soon. Advanced nanotechnology, however, offers a way forward.
Instead of the subtractive photolithography method, which takes previously deposited layers of materials and strips them away in intricate patterns, nano-structures can be assembled via an additive or ‘bottom-up’ building process.
Nanoimprint lithography is gaining ground as a viable alternative. With nanoimprinting the layers of component material are deposited with a suitably ‘inked’ nano-scale stamp. Lab-based nanoimprint technology has already stamped out structures as small as 10nm. However, more research is required to develop processes that can fabricate the complex multi-layered structures necessary for the next generation of integrated circuits.
Nanomanufacturing and mechanosynthesis technologies and tools offer even more potential. They are engineered to build structures particle-by-particle or even atom-by-atom. Essentially a transistor, sensor, circuit or other nano-device is formed in a similar manner to building a house or Lego model, namely brick-by-brick.
The ultimate scenario of building components atom-by-atom means nature’s elements can be fabricated from scratch. For example, it would be possible to assemble ‘diamondoid’ components. Diamondoid has around 50 times the strength-to-weight ratio of high-strength steel and would be perfect for a wide range of application - including a virtually indestructible mobile phone case.
There are numerous approaches to this kind of intricate bottom-up fabrication, but here I’ll describe just two. One approach involves building specialised tools that can place individual particles in precise positions. Way back in 1981, IBM demonstrated the scanning tunnelling microscope (STM), which had the capability to move individual atoms. When you consider a typical atom is only 0.2nm, it’s no surprise that IBM’s Heinrich Rohrer and Gerd K Binning received the Nobel Prize for Physics in 1986 for inventing the STM.
STMs and similar technologies, including the atomic force microscope (AFM), have advanced to the point where quite intricate structures can be created with a range of atomic elements. Unfortunately, while it offers the ultimate in fabrication resolution, this technology is far too slow and expensive to use in a mass production environment.
To address this mass production problem, some researchers are experimenting with ‘growing’ nanoscale structures. To achieve this, they devise a series of carefully controlled chemical reactions, which deposit the various nanoparticles on a suitable surface. Once perfected, these processes can be scaled up to an industrial scale, to simultaneously assemble millions of low cost yet highly advanced nanodevices.
However, while nanotubes, quantum dots and low-cost nanosensors are well suited to the ‘growing’ approach, it’s not yet clear if this type of solution can be refined sufficiently to assemble complex multi-layered structures required for a new generation of integrated circuits.
More surprises to come
We have only taken the first few steps on a vast nano-world adventure, and we may discover new nano-scale particles. Many nanoparticle functions and properties have yet to be explored, and the potential range of applications is almost unlimited.
Soon we may have a super computer inside every PC, carry rolled-up screens in our pockets, take nano-engineered drugs and own electric drive cars that are recharged in minutes and driven on solar-cell-embedded roads.