Building transistors today is done with lithography, which is a “top-down” process that uses patterning to create the complex layers that make up the transistor structure. It’s a bit like exposing a negative on photographic paper to get the pattern you want and then using this pattern as a template to place each material – metal, insulator or semiconductor – in exactly the right location. This process has worked successfully since the 1950s, and our scientists have even demonstrated the first working test chips with features approaching 7 nanometers, the equivalent of placing more than 20 billion tiny switches on chips the size of a fingernail. But as we get to ever-smaller dimensions, new approaches to building nanoscale devices will be required.
At IBM’s T. J. Watson Research Center we use a technique called self-assemblyto grow and directly control nanostructures that could one day form parts of integrated circuits. Self-assembly looks at chip building from the other end of the spectrum: a “bottom-up” approach that builds nanostructures in a way that is dictated by physics rather than by an imposed pattern. In some ways it’s like farming, in that that you plant seeds to grow a crop, and then support the growth with the right conditions to get the result you want.
But exploring self-assembly doesn’t mean we are ready to throw away today’s approach; instead, we want to use top-down strategies that we have already learned over many years, and combine them with new tricks that use self-assembly.
Think of water splashing onto a pane of glass. It spontaneously forms little hemispheres. The droplets are hemispherical because surface tension pulls the water molecules into this shape to minimize the surface area and energy of each droplet. But there is no reason for the droplets to form in any particular location or to be any particular size, so their positions and sizes are random. The spontaneous formation of the hemispherical shape is an example of self-assembly, but other aspects of the process (position, size) are not controlled.
Now imagine there is a scratch on the glass. Water droplets form on the scratch, because it is a good, low energy place for the water molecules to stick. We have now combined self-assembly – “make a hemispherical droplet on this surface” – with an imposed pattern – “make a droplet on this part of the surface by using carefully placed scratches.” The result is that we can build more complicated patterns. Flexible, customized patterns like this water example, but on the nanoscale, help us build integrated circuits.
The more precisely we can direct this self-assembly, the more versatility we can achieve. We can choose different materials for our nanostructures, build them with different sizes, and control their chemical compositions in ways that allow them to be tuned to have the properties we need. The properties of some nanomaterials could include the ability to do the job of a transistor but with less power, or at extreme temperatures beyond what silicon can handle.
How to direct a nanowire
In order to direct self-assembly, we have to understand the physical stimuli that influence atoms to assemble in a certain way as they form a nanostructure. The particular nanostructures we find most interesting are called nanowires. These are long thin crystals whose amazing length-to-width ratio could help create very densely packed transistors. Using a combination of imposed patterning and self-assembly, we can grow nanowires spontaneously using the help of catalytic particles. And we can watch the nanowires as they grow, recording the process on video using a one-of-a-kind Ultra High Vacuum Transmission Electron Microscope in our lab.
We load a flat substrate into the microscope, place catalytic particles onto it (this is the directed part of the process), then heat it and add some reactive gases. We watch what happens to the catalytic particles (this is the self-assembly part of the process) by magnifying the image by 50,000 times or more. The reaction can be slow – it takes hours for the whole experiment to be finished – but the videos show how the nanostructures grow, one layer of atoms after another. Recording videos, for example at different temperatures or with different added gases, is central to understanding every step of the nanowires’ growth. We get to see cause and effect when the conditions change, so we can work out the laws of physics that control the growth.
Here is a description/caption Frances sent me: Video recorded in the ultra high vacuum transmission electron microscope showing a silicon nanowire growing from a gold-based catalyst. The silicon is the vertical post and the catalyst is the liquid droplet on the tip. The diameter of the nanowire is 70 nanometers; in other words, about 1/1000th of the diameter of a human hair. The stripes in the silicon and the zigzag shape along its edge provide information about the atomic arrangement within the nanowire. This movie was recorded at 500 degrees C and is speeded up by 30 times.
This video was recorded in an ultra high vacuum transmission electron microscope
showing a silicon nanowire growing from a gold-based catalyst.
Recently, we have become especially interested in growing nanowires made of gallium arsenide that form with the help of catalysts made of gold nanoparticles. For this we need two reactive gases, trimethylgallium and arsine. We chose these because they supply the two components needed to build the nanowire, gallium and arsenic. When we record our movies, the first reaction we see is between gallium and gold. This reaction turns the original gold nanoparticles into hemispherical liquid gold-gallium droplets. As we continue to watch, gallium and arsenic combine within each droplet to start growing a gallium arsenide nanowire beneath the droplet.
Gallium arsenide nanowires grown this way are particularly special because it is possible to change the way the gallium and arsenic atoms stack up within each nanowire. Two arrangements of the atoms are possible, and we can change from one to the other simply by altering the temperature of the reaction or even just varying the ratio of the two gases as they flow past the catalysts. The videos show how these changes in growth conditions modify the way the atoms arrange themselves at the junction between the nanowire and the catalyst. And that causes a change in how the atoms eventually stack up when they form the nanowire. We still have the same material, gallium arsenide, but the two possible arrangements of the atoms lead to different electrical properties for the whole nanowire.
Understanding what drives atoms to take up one arrangement versus another gives us a better chance of growing nanowires that have the particular electrical properties that are needed for a device such as a nano-transistor. It’s akin to having more colors on your palette so that you can paint a better picture.
These special nanowires, composed of regions with different atomic arrangement, have applications in photonics or single electron transistors, both important building blocks for electronic circuits. And simply knowing that we can control the crystal arrangement in a nanowire will open up the microprocessor community’s imagination for new devices. In particular, optoelectronics, where light and electricity are combined in photonics structures, is a good bet. But that’s just the “tip of the crystal.”
Our latest results, “Interface dynamics and crystal phase switching in Gallium Arsenide nanowires”, will be published in this week’s Nature.