Universitas Helsingiensis

The quarterly of the University of Helsinki

A Nano Lift into Space?
Carbon nanotubes are peculiar molecules that scientists are planning to use to build collision-proof cars, space lifts and superspeed computers.

Henrikki Timgren

Back to autumn issue 2002

In 1991, while peering with an electron microscope at amorphic carbon - i.e. soot - Japanese researcher Sumio Iijima observed strange, supremely long threads a few atoms wide. These `threads' peeking out from among the soot turned out to be hollow tubes formed by hexagonal carbon molecules; owing to their extreme thinness and tubular shape scientists began to call them `nanotubes'. A nanometre is a billionth of a metre, which is roughly the size of a nanotube's diameter. A row of about four atoms can fit in a single nanotube.

In 2002, carbon nanotubes are more relevant than ever. Incredibly strong, stable and electrically conductive, it has been predicted they will become the new superstrong space-age material for cars and aeroplanes. The US National Aeronautics & Space Administration NASA probably has the wildest vision: they have seriously considered anchoring satellites that float at a height of 36,000 kilometres to the ground with a rope made of carbon nanotubes. In theory, the project is feasible. The tensile strength of carbon nanotubes is a hundredfold that of steel, which according to calculations is more than enough to construct `a space rope' or even a space lift.

Information technology revolution?

In the short run, it is the electronics industry that has the greatest expectations for carbon nanotubes. Since the 1960s, the vertiginous development of the calculation capacity of computers has been based on semiconductor technology, in which silicon guides the electric impulses going through the computer's circuits. The smaller the circuits can be made, the more efficiently and quickly the computer operates. Owing to the physical characteristics of the element, however, silicon chips cannot be made endlessly smaller. In today's computers, the silicon components have been reduced to a size of some one hundred nanometres, but according to theorists, it is enormously difficult to make them as small as twenty nanometres, let alone smaller. With the current speed of development, we will reach a dead end around the year 2010.

Nanotubes, however, are only about two nano-metres wide. According to some estimates, nanotube electronics could be as much as a thousand times faster than the current electronics based on silicon and metals. And to top it all, the electric conductivity of nanotubes can be customised simply by choosing a tube with a certain helicity. A straight tube behaves like metal, while a twisted one acts as a semiconductor. Therefore, nanotubes could replace both the semiconductors used in transistors, such as silicon, and the metal wires that connect transistors, such as copper.

There are so far two basic problems in the way of the nanotube revolution. Firstly, the manufacturing of nanotubes is very expensive: a gram of nanotubes costs about a hundred US dollars. That is worth ten grams in gold! Secondly, the construction of complex computer circuits requires new technologies so that the electrical and mechanical characteristics of straight `basic' nanotubes can be easily altered. It must also be possible to join nanotubes together to form more complex constructions.

Simulation - a gate
to experimental research

Nanotechnology in general and carbon nanotubes in particular are becoming a major business. Crossing the frontier of research, however, does not always require huge early investments. The nanotube research group at the Accelerator Laboratory, the University of Helsinki, is an excellent example of how it is possible to participate in the nanoscientific avant-garde with meagre financial resources.

"We use computer simulations to see what happens to nanotubes when they are irradiated with ions or atomic clusters because ion irradiation is a good tool to study and modify properties of nanotubes. In simulations, we can study these dynamic processes in real time, which is not possible in experimental conditions, as the processes in reality only last femtoseconds, which is a quadrillionth, or million billionth, of a second," says Kai Nordlund, PhD, Academy Fellow, who is in charge of the computational work in the nanotube group. The nanotube group is part of the ion beam group of the Accelerator Laboratory which is led by Professor Juhani Keinonen. Besides Nord-lund, the nanotube group includes postdoctoral researcher Arkady Krasheninnikov and Emppu Salonen, MSc, who has just completed his doctoral thesis.

Bombarding imaginary nanotubes with imaginary ions may sound like futile tinkering, but quite the contrary. Through the simulations, it is possible to find solutions such as how to `weld' single nanotubes together with ion beams to form the shape of an X or an even more complex construction. These are the methods that the electronic industry badly needs.

"At their best, the simulations can be used to predict new processes, which means that we can also meet with surprises in the simulations. Of course, that's the fun part of this work. A successful simulation can also lead to the launch of experimental research. Who knows, in a couple of years our group may well be experimenting with nanotubes," says Nordlund.

In fact, 70 per cent of the research conducted at the Accelerator Laboratory is experimental. Nanotubes are bombarded virtually owing to the scarce availability and minuscule size of the research material, but for the bombarding of larger surfaces, formed by millions of atoms, the Laboratory has a real ion accelerator. It has been used to target materials such as graphite, a close relative of nanotubes. Nanotubes can be considered extremely narrow, single layers of graphite that have been wrapped up into cylinders.

All tricks are known

Carbon is a thoroughly known element, which means that basically any physicist with a powerful computer could start to simulate the impact of ion bombardment on carbon nanotubes. The mathematical models used in the simulations were invented well before the discovery of carbon nanotubes.

"It's true that almost any scientist could perform the same simulations. But there are so many theoretical and practical problems connected with the simulations that it can easily take a year to solve them. The competition is so tough today that a year is too long," says Krasheninnikov, who has specialised in nanotube modelling.

"In a way it's a question of capturing area. When you are a researcher in some really hot field and want to stay in front, you must know what the other trail blazers are doing right now," says Emppu Salonen, the postgraduate of the group.

"Our speciality is a particular problem: the modelling of the ion irradiation of nanotubes. When we started some eighteen months ago, we were the first in the world. Our group is a lucky combination of know-how from different fields. Arkady was familiar with nanotubes and quantum mechanical computation, while we others have been working with irradiation for a decade and know all the tricks. Even the models become unstable if you don't know what you're doing," says Nordlund.

Silicon sticks to its guns

Ultimately the nanotube revolution is a question of money. Even if scientists succeeded in developing and modifying nanotubes, their production would become the bottleneck. If NASA really wants to construct a 36,000-kilometre nanotube rope for the price of a hundred dollars per gram, the final costs will be fairly high.

Today, there are about half a dozen companies producing nanotubes in the world. Nordlund doubts, however, whether the price will ever be lowered enough. "Nanotubes have great potential in electronics, but we should remember that since the 1960s there have been promises about brand new materials that will beat silicon during the next decade. Yet silicon has been sticking to its guns. I think that nanotubes will find their own niche in the electronics industry, but I very much doubt they will completely topple silicon chips.''

More moderate scientists have indeed been warning the business world of premature triumph. So far only few commercial applications have succeeded with nanotubes. The American car-manufacturing giant General Motors has mixed nanotubes in the plastic parts of cars. During painting, an electric current is conducted into the plastic mixed with nanotubes, which makes the paint stick harder to the car's surface. On the other side of the globe, Korean Samsung has already begun serial production of flat computer monitors where carbon nanotubes have been used. The idea is that when a very weak electric current is conducted into the nanotubes, they begin to transmit large quantities of electrons from their tips. When hitting the surface of the phosphorised screen, the electrons activate the phosphorus, and the monitor is lit.

There are also many who believe that nanotube compounds will soon replace carbon fibre. Who knows, in the near future we might speed up with unbreakable nanotube ski sticks and beat the tennis ball (or court) with an unbreakable tennis racket. And maybe some day, after a century or two, take a nanotube lift into orbit.

What are nanotubes?

Nanotubes are a form of carbon, discovered in 1991. Other known forms of carbon are fullerene, diamond, graphite and amorphic carbon, or soot.

Nanotubes consist of chains of pure carbon, the diameter of which is about two nanometres, or two billionths of a metre. Their length can be several thousand nanometres. If you think of an even, hexagonal sheet of graphite as chicken wire mesh, the nanotube is like a rolled up piece of that mesh.

Owing to their molecular structure, nanotubes are extremely strong - they endure bending better than the strongest carbon fibres and their tensile strength is a hundredfold that of steel. The density of nanotubes, however, is only a sixth of that of steel. Therefore, it is hoped nanotubes will provide new materials for such as cars and aeroplanes.

By adjusting the size and structure of nanotubes, it is possible to make them either conduct electricity like metals or act as silicon and other semiconductors, that is, to conduct electricity only under the influence of an outside electromagnetic or light pulse. Both metals and semiconductors are needed in advanced electronics.

The electrical conductivity of nanotubes has been estimated to be a billion amperes per square centimetre. Copper wire burns and breaks, if an electrical current of more than a million amperes passes a square centimetre thick wire.

Nanotubes remain stable in temperatures up to 3,000 degrees Centigrade, while the metal wires in microchips melt in temperatures of 600 to 1,000 degrees centigrade.

Nanotubes are still extremely difficult to manufacture. Their current price is about a hundred US dollars per gram.

What is materials physics based on ion beams?

In an ion beam, the material under study is irradiated with single, electrically charged atoms, molecules or atomic clusters: ions. The ion beam is created with an ion accelerator. Materials and applied physics based on ion beams have been the main line of research at the Accelerator Laboratory of the University of Helsinki.

The ion beam method has been used for purposes such as developing extremely thin diamond coatings that can be used to cover artificial joints; the diamond coating protects the patient's body from metal compounds dissolving from the artificial joint.

Traditionally, ion beam based physics has been carried out by firing single ions at large surfaces to enable the modification and study of the characteristics of the irradiated area. Ion beams are also used in the characterisation of the qualities of various materials.

The current trend in materials physics is to study and modify nanosize atomic structures, that is, structures the size of a billionth of a metre, such as carbon nanotubes.

The equipment at the Accelerator Laboratory is so far insufficient for bombarding with ion beams objects the size of a few nanometres, such as single carbon nanotubes. With the experience gained in ion irradiation of larger areas, however, the nanotube research group of the Laboratory has developed powerful computer simulations of the ion irradiation of nanotubes.

With the nanocluster accelerator, which will probably be taken into use at the end of 2002, it is possible to irradiate nanosize clusters of ten to a million atoms.

Back to autumn issue 2002