Existing methods of making microelectronic components are essentially two-dimensional. But a third dimension will soon be added

CUT through a computer chip and you will reveal something that looks like a microscopic layer cake. Chips are built up a piece at a time by a process called microlithography. This involves repeatedly depositing stuff on the chip's surface and then etching part of it away to leave the transistors and other devices that do the actual computing. The process is time-consuming and error-prone. Even if there is a 99% chance that an individual layer will be deposited and etched without a mistake, only 90% of ten-layer chips will be usable. Add another ten layers and the figure drops to 82%.

On top of this, microengineering's requirements are increasing. There is rising demand for so-called micromechanical devices, which have moving parts built on to the chip. These frequently require undercuts and rotating elements that can be hard, or even impossible, to make using conventional microlithography. And pure microelectronics is now gradually giving way to optoelectronics, which uses photons (the particles of light) as well as electrons to carry data. Some optoelectronic processes need devices that have significant thickness, as well as length and breadth. If these were built by conventional methods, they would need hundreds, rather than dozens, of layers. That would make the error rate enormous, and the time involved prohibitive.

These optoelectronic devices also require not just microfabrication (which creates components with dimensions measured in microns, or millionths of a metre), but nanofabrication (where the components are nanometres, or billionths of a metre, across). The hunt is therefore on for ways to build a new generation of "thick" three-dimensional (3D) components in a single step, preferably using nanofabrication. Three research groups, employing three different methods, are now on the case.

One-step wonders

The most straightforward approach is that of Shoji Maruo and Koji Ikuta, at Nagoya University in Japan. They have taken an existing technique called stereolithography, and miniaturised it. Stereolithography works by steering laser beams through a photosensitive liquid that solidifies when it is exposed to enough light. One way to control this process is to use two beams that, when they cross, double the amount of light in a spot, and thus cause the liquid to solidify faster. When this scanning is complete, the 3D structure is released by pouring out the remaining liquid.

The problem with stereolithography for small things is that the liquid tends to solidify even when it is not at the most intense part of the beam. To get round this, Dr Maruo and Dr Ikuta picked a material whose molecules solidify only when hit simultaneously by two photons from different lasers.

Using this technique, the two researchers have managed to build 3D structures with features as small as a wavelength of light (a few hundred nanometres). Electronic and optical applications are restricted by the properties of the material used, but there should be jobs to do in micromechanics. At a microelectromechanical systems conference that is taking place at Interlaken, Switzerland, from January 21st-25th, the researchers will describe their latest creations, which include moving microparts that are driven by light.

At Oxford University, meanwhile, Andrew Turberfield and Bob Denning are using crossed laser beams to create so-called photonic crystals. These are employed to trap photons so that they can be exploited efficiently in, for example, low-power lasers. They consist of two interlaced materials that have different optical properties—in particular, different refractive indices, a measure of how much a substance bends light. The boundaries between the materials create reflective surfaces within the crystal. These can be used to shepherd photons around and keep them under control. Such crystals might serve as components of optical switches which would, among other things, fulfill the role played by transistors in conventional electronics.

In order to make a photonic crystal, though, the materials have to be organised into structures no more than a wavelength of light apart. And layer upon layer of such organisation is required. The technique that Dr Turberfield and Dr Denning are using to do this has several stages. First, a block of material whose chemistry can be altered by light is made. Four laser beams are shone into this block from different directions. Where the beams overlap, they interfere with one another. This is because, in the wacky world of quantum physics, light is wavy as well as particulate. Where the crests or troughs of the waves of two or more beams coincide, bright spots are created; where a crest falls on a trough, a dark spot results.

The material responds to the pattern of spots, undergoing chemical changes at the bright ones. The block is then bathed in a liquid that dissolves the chemically altered part of the material in a way reminiscent of conventional microlithography, while leaving the rest intact. If the second material of the crystal is intended to be air, that is enough. Otherwise, it can be bathed in a liquified form of the second material until all of the voids created by the laser are filled in.

A third approach to 3D nanofabrication is being pioneered by a collaboration between Junichi Fujita at NEC's Fundamental Research Laboratories and Fujio Shimizu of the University of Tokyo. In this case, too, interference patterns are the tool of choice. However, it is not waves of light that are made to interfere with one another, but waves of matter.

Just as light can be thought of as both particles and waves, so can atoms. That means that holograms - 3D "photographs" usually made with light - can, in principle, be made with atoms. And making a 3D structure at the atomic scale is what nanofabrication is about.

In conventional holography, two beams of light interfere with one another. One beam is reflected from the object that is having its picture taken; the other is a "clean" reference beam. When the two meet, the interference pattern produced, if captured on photographic film, has the ability to change a new reference beam into a new object beam. Thus an image of the object can be reconstructed in light.

Since holography was invented in 1948, there have been two major developments in the field. The first was the realisation that a hologram could be synthesized mathematically in a similar way to the process used to generate computer graphics. It does not, in other words, need to be the image of an actual object. The second was the realization that holograms do not have to be recorded on film. In the right circumstances, they can be created by manipulating the refractive indices of materials in the way that Dr Turberfield and Dr Denning are doing. Or, in the case of beams of atoms, by manipulating electric fields to achieve the same effect.

Dr Fujita and Dr Shimizu have combined these two advances to create a holographic projection system for beams of atoms. By passing such beams through a screen of tiny electrodes that act as holographic "pixels" (the equivalents of the grains in photographic film) a 3D image, in atoms, of the desired object can be built up. If the structure is too complicated to create using a single hologram, then the pattern of charge on the electrodes can be changed to add extra features.

This system, which is still in its earliest stages, has several drawbacks: not least, that not all atoms can easily be patterned in this way, and that the whole process has to take place at very low temperatures. But if it could eventually be industrialised, it would be a triumphant demonstration that even the weirdest scientific arcana can sometimes have practical applications.

(The Economist)

ЧТО И КАК ПИСАЛИ О МОДЕ В ЖУРНАЛАХ НАЧАЛА XX ВЕКА Первый номер журнала «Аполлон» за 1909 г. начинался, по сути, с программного заявления редакции журнала...

Живите по правилу: МАЛО ЛИ ЧТО НА СВЕТЕ СУЩЕСТВУЕТ? Я неслучайно подчеркиваю, что место в голове ограничено, а информации вокруг много, и что ваше право...

Что будет с Землей, если ось ее сместится на 6666 км? Что будет с Землей? - задался я вопросом...

Что способствует осуществлению желаний? Стопроцентная, непоколебимая уверенность в своем...

Не нашли то, что искали? Воспользуйтесь поиском гугл на сайте: