An article published in the journal “Science” describes the creation of a new type of transistor with a working gate of one nanometer, a billionth of a meter, the smallest ever made. A team led by Ali Javey of the USA’s Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab) used carbon nanotubes and molybdenum disulfide as materials to create this transistor instead of silicon.
The length of the gate, one of the parts of a transistor, is considered the one that defines the size of the transistor. The ongoing miniaturization of these semiconductor devices crucial in electronics is reaching the limits of silicon’s possibilities. The next year microprocessors produced with a 10 nanometer process are expected to be on the market and various experiments have been conducted to produce even smaller transistors.
The physical limit for a silicon transistor is 5 nanometers because below that the electrons start moving in an uncontrolled manner through the gate preventing its operation. Precisely for this reason the manufacturers of microprocessors and not only have been exploring alternative materials for some time.
Carbon nanotubes are considered among the most promising possibility to build very small transistors but the Berkeley Lab team used them together with molybdenum disulfide, part of the dichalcogenide group. Those are transition metals with characteristics that are intermediate between those of two-dimensional materials such as graphene and those of three-dimensional materials such as silicon. For this reason, among other things they can reach extremely small size.
The normal lithographic techniques used to produce silicon-based transistors don’t work when sizes are so small. This is the reason that led researchers to use carbon nanotubes to build the gate which, together with the elements called source and drain, made of molybdenum disulfide, form the transistor.
A test of the transistor produced showed that it works. It’s a laboratory test, a proof-of-concept quite different from its practical application. Ali Javey is the first to point out that there’s much to do before a chip with this type of transistor can be built. However, the fact that it works means that it’s possible to build microprocessors more powerful than those possible with silicon-based technologies.