The making of a chip is an extremely sophisticated manufacturing process. The basic point, in a nutshell, is to position on a silicon wafer the right atoms in the right places to make transistors, resistors and connections among them. Doing this requires a series of steps each one finely tuned to the previous one. To make sure that you are positioning the atoms at the right place on the wafer you create a mask that protect those areas on the wafer where there shouldn't be any deposition of atoms, a process that takes place by diffusing a gas over the surface.
The masking is obtained through photoresists, a layer that is covering the whole wafer and that is selectively removed in those places where one wants the gas to hit the wafer. To remove the wafer in just the right spots one uses a sort of photographic process: you illuminate the photoresist and that changes its characteristics becoming soluble to special solvents. By rinsing the wafer in the solvent one removes the photoresist and exposes the underline wafer that can now be affected by the gas.
The exposure of the photoresist to light is made by illuminating a mask (produced by a computer) having the details magnified. A focalisation of the light beam shrinks the details to the desired scale. The shrinking depends on the wavelength of the light beam being uses, the shorter the wavelength the smaller the details that can be illuminated on the photoresists. For chip manufacturing the light beam is made by ultraviolet light, in the wavelength range of 248-193 nm. This wavelength is too "long" to scale down to the 10nm of the next chip generation and engineers have to move to extreme ultra violet light (EUV) in the wavelength of 13.5nm.
The problem is that present photoresists are not sensitive to that short wavelength, hence the need to find newer photoresists. Although it might seem strange, scientists so not have a complete understanding on how a photoresists really works and this is a stumbling block in progressing any further.
This is where this news come in. Scientists at Berkeley, teaming up with engineers at Intel, have come up with a new type of photoresist (in short - resist) and in order to do that they have reached a much better understanding on what is a resist and how the fundamental chemistry processes work.
The new resist combines two already used resists to achieve better sensitivity to light (hence being responsive to the shorter wavelength) and better mechanical stability (which is essential as you shrink size). Actually one of the resist had very good sensitivity but low mechanical stability whilst the other had the required mechanical stability (cross-linking) but poor light sensitivity. By understanding the fundamental chemistry the team has been able to capitalise from the strength of each.
They expect to see the new resist being used in the manufacturing process as early as 2017.