The atomic calculator

With electronics shrinking according to Moore’s Law, designers fear the day that their devices become so small that quantum behavior will dominate the response. PhD student Jan A. Mol (MSc), at Applied Sciences, turned the logic around: If eventually you’ve got to deal with the atomic scale, why not use atomic quantum behavior to your benefit? And that’s exactly what he has done with a binary adder based on atomic energy levels.

Physically, the heart of the Single Atom Transistor is a single arsenic atom, buried a few atom layers deep in silicon. On top of it, three electrodes join to apply the gate and the bias voltage. Depending on the voltages applied, the current tunneling through the arsenic atom is either zero, minus 2 or minus 6 nano-amperes.

This feature can be used to build an adding unit in which 0 + 0 = 0; 1 + 0 = 1; 0 + 1 = 1 and 1 + 1 = 2 in numerical values. In binary digits, the value 2 is written as 10, whereby the ‘1’ is carried to the next level. Mol uses another single electron transistor or SET to make this numerical to binary conversion.

In doing so, he only needs four atomic transistors to make a full adder (two inputs, two outputs), whereas the industry standard is 28 transistors for a full binary adding device.

For Mol, the beauty of the concept is that by making use of single-atom energy levels, he was able to make something not only smaller than any other transistor, but also more powerful in the sense that it has a multiple value output instead of just ‘0’ and ‘1’.

In principle the Single Atom Transistor could drastically reduce the size of computing devices, but for the moment there are a number of impractical issues. The first is the location of the dopant atoms in the silicon: currently they are ‘shot’ into the substrate without much control over where they land. Clearly that’s not the best way to (mass) produce devices.

The other problem is that so far the experiments have been done at very low temperatures, in order to keep thermal noise from drowning the signal. Using other materials however may make the system more robust. Replacing the arsenic atom with selenium for example will increase the difference in energy levels and amplify the signal, making it less vulnerable to thermal noise.

Mol will continue his research as a postdoc at the University of South Wales, in Sydney, Australia, where he will follow the trail of his PhD supervisor, Professor Sven Rogge. Mol will start by exploring the behavior of ensembles of single atoms. “I like doing fundamental research” he explains.

–> Mol, Jan A.: Single Atom Electronics, 14 September 2012, PhD supervisor Prof. Sven Rogge (University of South Wales, Australia).