Tiny tubes and thinnest strips of carbon are promising candidates for new types of electronic components. Researchers at Empa, working with their German colleagues, have developed methods to precisely create specific carbon structures – a step on the path from micro-electronics to nano-electronics.

Tiny pillars of colourful hexagons all pointing upwards, adorn the cover of the renowned British journal Nature, issued on 7 August 2014, – perfectly formed carbon nanotubes which embody a long-awaited research success. For the first time, physicists at Empa, working with German chemists, were able to grow carbon nano-tubes that all have a single definite structure – a feat that scientists had been trying to achieve for the past 25 years without success. In fact, it had been possible to produce large quantities of the so-called nanotubes, but this always led to a mixture of up to 50 different types of tubes with various different properties.

The tiny tubes are made of a honeycomb structure of carbon atoms. The material is extremely stable, ultra-lightweight and flexible. Tubes with a single-layer wall have a diameter of the order of a millionth of a millimetre (nanometre), and behave according the laws of quantum mechanics. “Depending on how a tube of this kind is rolled, it will conduct electricity either like a metal or like a semi-conductor”, explains Prof. Roman Fasel, head of the Empa “nanotech@surfaces” department and adjunct professor at the University of Bern. "If you are interested only in the mechanical properties of the nanotube, this is not important. But if you are building an electronic circuit, it is crucial." Only a semi-conducting material would be suitable for that, whereas a metallic one would cause a short-circuit.

Origami with molecules

It is in fact possible to sort out the different types of tube from the mixture that was produced in the conventional way, but none of these time-consuming and wasteful methods work very well. So it is not surprising that Nature praised the new method as an “outstanding breakthrough in the synthesis of carbon single-walled nanotubes”. The starting material for the process is a flat molecule that chemists produced at the Max Planck Institute for Solid State Research in Stuttgart: a hydrocarbon consisting of 150 atoms. Roman Fasel's team created a three-dimensional, dome-like shape from this two-dimensional object. “It's like origami” explains the physicist, taking up a piece of paper. “We have stuck the molecule together at the edges like with a sheet of paper, and folded it like this to make a cap.”

For this trick to work, the researchers had to place the molecule on a platinum surface in a vacuum and heat it. This caused hydrogen atoms to split away from the edge of the hydrocarbon. “This is how we made our glue – 'unlucky' carbon atoms that are looking for a partner to join with, and which then combine with other carbon atoms along the edge”, explained Roman Fasel. “This produces end caps that are all identical.” In a second process, these seeds grow to form tubes. For this, a gas is needed which supplies more carbon pairs by catalytic decomposition on the platinum surface. These collect on the open edge between the cap and the metal, and cause the structure to grow upwards like little finger cots. Because the “seeds” are identical, only one single type of tailor-made tube is formed in this way – just as desired.

Most of the nanotubes grow vertically like the bristles on a brush, and will probably be about five to ten nanometres long. But there are some tubes of up to 300 nanometres lying across them. The researchers still do not know how to control the growth in length. “We have demonstrated that this method works in principle”, says the Empa researcher. But although more than 100 million nanotubes per square millimetre grow on the platinum surface, the process is still inefficient. “We are still a long way off from mass production”, the expert says.

Tailor-made narrow graphene strip

Single-walled carbon nanotubes are not the only candidate for new applications in electronics, however. Roman Fasel and his team are also working on another “super-material”, graphene. When one cuts a single-walled carbon nanotube along its length and flattens it out, the result is graphene. This material, too, consists of just a single honeycomb layer of carbon, which is just one atom thick. Graphene is harder than diamond, extremely tear-proof, and is an excellent conductor. For electronic applications, however, the material has the disadvantage that it is neither a metal nor a semi-conductor, but something in between: a semi-metal without the so-called “band gaps”. These are what the physicists call an energy range where no electrons can reside.

Semi-conductors like silicon or gallium arsenide have a band gap, unlike metals which do not. A band gap is needed if one wants to switch an electronic component on and off, for example when one wants to build a transistor. In the case of graphene, a band gap is formed by a quantum mechanical effect if the material is shaped in thin strips. “This was predicted in theory long ago”, explains Roman Fasel, “but the difficulty is that the strips have to be really narrow, only a few nanometres wide, and the edges must not have any irregularities.” This means that one can not simply cut a graphene layer into small strips. However, in 2010 the Empa researchers, in collaboration with the Max Planck Institute of Polymer Research in Mainz, did manage to produce the narrow graphene strips, by building them up in a controlled manner from hydrocarbon molecules – in a similar way to the carbon nanotubes.

The narrower the strips are, the bigger the band gap. “Therefore in the past few years the aim has been, firstly, to refine the production process, and secondly to improve the properties of the graphene nano-strips”, says Roman Fasel. This is where the researchers made a significant advance in 2014. They were able to produce more complex structures, in which the strips contained various segments. The starting material for this – molecules that are fitted with nitrogen atoms at certain positions – were again produced by the chemists in Mainz, who sent them to Dübendorf. If, on the graphene strips, the segments doped with nitrogen alter nate with undoped ones, what is known as hetero-transitions are formed. The researchers were able to show that these behaved in a very similar way to the material transitions in semi-conductor crystals between regions of different doping. “Such semi-conductor transitions, where the doping changes from positive to negative, are found in every mobile phone or other modern electronic device”, says Roman Fasel.

Sensitive detectors, efficient solar cells

In future, when one wants to manufacture electronic components from graphene nano-strips, one can use the nitrogen doping and the bandwidth to select the best values for the important parameters. “So we have two knobs to turn”, the physicist sums up. But practical applications for nanoelectronics are still far off. Nevertheless, the Swiss and German research cooperation is collaborating with the BASF chemical group. In this way, the industrial partner has already secured six patents that could pay off in future.

Because the graphene nano-strips react very sensitively to their environment, one can perhaps at some later stage produce not only transistors but also sensors. And the narrow strips could also be suitable for use in solar cells. Because they absorb visible light so well that one only needs a few layers to suck up all the light. “We know this from our daily experience”, says Roman Fasel. “Graphite is black, and graphite is nothing other than layers of graphene stacked on top of each other.” By adjusting the width of the strips to an atomic degree of precision, it is even possible to enormously increase the specific light absorption for certain wavelengths. Nobody is sure whether this technology will in fact conquer the market in the next ten years, though. The experts are not willing to speculate, either: “At present we are still at the stage of fundamental research, and there is certainly a lot that still needs to be done”, says Roman Fasel.