An innovation in chip fabrication, as Epson unveils its micro-liquid process breakthrough, a liquid silicon spray that could revolutionise semiconductor chip manufacturing.
Spraying in the name of science
In a breakthrough that will impact the way semiconductor chips are made, Epson has developed a practical method of creating liquid silicon. The material can easily be sprayed or spin-coated onto a substrate to form a thin film. John Boyd, freelance writer based in Yokohama discusses the new silicon, developed jointly with JSR Corporation steming from research commissioned by Japan's New Energy and Industrial Technology Development Organization (NEDO).
Silicon composes one-quarter of the earth's crust and is the bedrock of the chip industry, a material scientists have long been seeking to replace with something less expensive and more manageable. Extracting silicon and refining it into the ultra-pure form required to make semiconductors entails great effort and costs. In addition, much of the silicon is lost in the slicing, grinding, polishing, etching, cleaning, and dicing that takes place during the convoluted process of fabricating chips. While many researchers around the globe have focused on creating materials that could take the place of silicon, an Epson team decided to tackle the problem with a different approach. Rather than seeking to substitute silicon with something exotic, Epson's Corporate R&D Division began with the idea of creating a new kind of silicon, using an innovative chemical solvent process that would render the normally solid material into a solution. Epson researchers surmised that in liquid form, silicon could be applied just like ink via adapted inkjet heads.
This method of application had the potential to eliminate waste, as only the exact amount of silicon required would need to be applied when fabricating transistors, the fundamental on-off switching elements that control the flow of electrons in chips. "It was an unconventional idea inspired by the understanding that silicon is the king of semiconductors," one of Epson's research manager comments. At the beginning of the project, some eight years ago, most scientists considered this approach a lost cause. They believed that the solvents required in creating a silicon solution would inherently contaminate it, making it impractical to use. Yet if the Epson researchers were to succeed, they would have to use precisely such a process of chemical synthesis. The only alternative method would be to use heat, which was out of the question, given that a temperature of 1,414ºC is required to render silicon molten, making it impossible to work with.
Consequently, the Epson research engineers invited JSR chemical scientists to join them in the challenge. Following much trial and error, the combined team focused in on a compound of silicon and hydrogen known as cyclopentasilane, which is comprised of five silicon atoms and 10 hydrogen atoms forming a closed ring. Although this silicon compound is liquid at room temperature and therefore seemed ideal for this purpose, when heated, it evaporates before it can turn into silicon film and leaves nothing on the substrate. Needing to take a different approach to the problem, the researchers decided to exploit cyclopentasilane's sensitivity to ultraviolet (UV) light. By irradiating the material with UV light, they opened up the atom rings, which then re-formed into long molecular polymer chains that took on the characteristics of a viscous oil or solid. This made it far less prone to evaporation. The resulting polysilane material has actually been known about for quite some time, but it had been mostly neglected by scientists, because it was thought that there was no way of purifying it. After experimenting with the irradiation process, however, the researchers were able to produce a solution that, with further treatment and heating, could be transformed into a pure amorphous silicon film.
Reordering the crystalline make-up
In the ranking of semiconductor materials, amorphous silicon lies midway between lowperforming organic materials and highperformance polysilicon. To improve the structure of the amorphous silicon, the team further refined it using a high-intensity laser beam. This process reorders its crystalline make-up to quicken the flow of electrons. The result is a liquid material with the characteristics of polysilicon and is the first instance of a high performance silicon film to be produced using chemical synthesis. "No hightemperature metallurgical refining is needed," states one Epson researcher, though he is quick to add that producing the film is not without its challenges. These include the need for further refinement and achieving a consistency of composition during the production process. The team also has to take a number of precautions against oxygen contamination. "Just as we have to make sure we remove all the oxygen from the film—for even a tiny amount will spoil the entire batch" he states. Along with the relatively low cost of production, Epson is most excited about the liquid silicon's printability using the inkjet technology it calls micro-liquid process. Besides eliminating the wastage of material, direct patterning with inkjets requires none of the lithography or expensive masks used in forming conventionally fabricated transistors and more complex electronic devices. This, in turn, reduces the number of process steps, speeds up fabrication, and eliminates the need for some types of manufacturing equipment. The manufacturing area is also reduced, so the dimensions of the overall plant can be smaller. But before all this can be achieved, a lot more work is required to refine the micro-liquid process. "Right now we have produced just one transistor to show the feasibility of the process," says the project manager. "The goal, of course, is to make millions of them in a uniform manner and without defects."
The researchers are also endeavouring to produce liquid versions of other materials that can be used as conductors and insulators, the basic components of all transistors. Another task facing them is learning how to dope the liquid silicon, so as to modify its electronic properties. Given the relatively large size of the thin-film transistors used to turn the pixels in LCD images on and off, Epson intends to target such displays as the first practical demonstration of a product that can be manufactured with micro-liquid process technology. Other possible areas for its use are the manufacture of organic light emitting diode (OLED) displays and solar cells.