Infineon achieves breakthrough - soon glass instead of silicon?
Researchers at the semiconductor manufacturer Infineon have achieved a breakthrough in the very promising research branch of carbon nanotubes (CNT). A method that has been tried and tested in the semiconductor industry has been modified so that CNTs are grown in predefined locations on 6-inch wafers for the first time.
The properties of CNTs, which are highly interesting for many applications, such as current densities of up to 1010 amperes per square centimeter and a thermal conductivity that is almost twice as high as that of diamonds can be used for the first time in wafer-compatible processes for IC development. Numerous other features make CNTs the top material of future semiconductor technology with significantly higher reliability and considerable potential for increasing chip clock rates.
Probably the most interesting property of CNTs for the semiconductor industry is their extremely high electrical conductivity. Due to the so-called ballistic electron transport, Ohm's law is invalidated and the electrical resistance is a constant that is almost independent of length. This remarkable property of CNTs enables current densities of up to 1010 amperes per square centimeter. This is an enormously high value when you consider that copper begins to melt at a current density of around 107 A/cm2. In view of the expert opinion that semiconductor chips will have to cope with a current density of 3.3 x 106 A/cm2 in about ten years, the importance of this CNT property cannot be overestimated. This is hardly possible with conventional current conductors, at least not without extreme heat generation.
Because the current flows in CNTs with practically no 'friction', there is no excess heat that needs to be dissipated. When CNTs are used as current conductors, heat only develops at the contact points with other materials. But here another helpsfavorable property of CNTs: their extremely high thermal conductivity. It even surpasses that of diamond, the material with the highest known thermal conductivity to date.
These few properties alone are enough to make CNTs the top material in future semiconductor technology. The only problem up to now has been that the manufacturing processes for CNTs - laser vaporization or arc discharge - are difficult to reconcile with semiconductor technology. Infineon has now changed that and one can rightly speak of a major step forward in CNT technology. The Infineon team has succeeded in growing CNTs in a highly parallel batch process on a 6-inch wafer at predefined locations. The Infineon scientists successfully modified a deposition process widely used in microelectronics ..
As the first possible application of the technological Innovation comes into question via vias, i.e. contact bridges between two metal layers in ICs. Conventional vias tend to deform at higher current densities due to the development of heat and thus impair the functionality of the chip. This risk would no longer exist if CNTs were used as vias, because they cope with far greater current densities and, moreover, have a much greater mechanical stability. 'In the next step, it would be conceivable that with the knowledge we have gained, we will be able to replace all metallic conductor tracks in the chip with CNTs,' says Dr. Franz Kreupl. This would ultimately lead to a considerable increase in the chip clock rates.
However, the assumption of the conductor path function by CNTs is only one possible application of this versatile material. Another important feature of the tubes is the possibility that they can also be produced and doped semiconducting. In this way, active switching elements such as field effect transistors can also be formed with CNTs.In the case of semiconducting tubes, the energy gap can be dimensioned by setting the diameter. It is typically 1 electron volt at a diameter of 1 nm, which roughly corresponds to the ratios for silicon transistors. The Infineon researchers are working on applying the semiconducting CNTs to a wafer in a targeted manner - using the same catalytic deposition process. “The whole subject goes far into the future. It is entirely conceivable that this technology could be used as an overall substitute for silicon-based semiconductor technology, ”says team leader Dr. Hönlein. In that case, the relatively expensive silicon could be replaced as a substrate, for example by glass. But that's not all. Infineon visionaries are already running through scenarios to use CNTs to expand today's planar microcircuits into real 3D technology.