About 35 years ago, IBM invented a scanning probe microscope. This opened the door for scientists to exam materials down to the atoms. After years of learning how the atoms are assembled to create the material, one scientist has been using the scanning probe microscope to build things atom by atom.
Bob Wolkow, a physicist at the University of Alberta has the goal of creating microprocessors using this technique. His process involves breaking down the processor into smaller circuits and assembling them on a substrate atom by atom. This method gives the assembler a more precise control over circuit placement. This method will produce a chip that is more efficient and will operate at faster speeds.
The way CPUs work are by manipulation electrons. There several circuits built into the CPU, some are used to move the data in a sequential fashion. The data is maintained as a combination of ‘1’s and ‘0’s, these are not actually 1s or 0s but they are voltage levels. It is these voltage levels that are manipulated by the circuits to process the data.
By having control over the placement of the circuits, designs can be facilitate the flow of data. Energy can be conserved by reusing the electrons. Under current designs, when the CPU needs to change a ‘1’ to a ‘0’, it grounds the circuit that is at the ‘1’ voltage level and the voltage drops to a ‘0’ voltage level. Wolkow’s design will move the voltage drop to a circuit that needs to change a ‘0’ to a ‘1’ so the voltage is not wasted by sending it to ground.
To give an example that simplifies the process. Assume we have two capacitors, when the charge is 5 volts it will be considered as a ‘1’, if the charge is 0 volts it will be considered as a ‘0’. Lets assume that one capacitor (A) is charged to 5 volts and the other (B) has no charge. To move the charge from A to B; one way is to short capacitor A to ground, this will bring capacitor A to 0 volts and apply a battery or power source to capacitor B to charge it to 5 volts. The other way is to connect the two capacitors and transfer the charge. The first way we wasted the energy by shorting the capacitor A to ground and we had to use more energy to charge capacitor B. The second way, we used no energy, it was just transferred.
Note, in my example if you connect the capacitors, the whole charge will not transfer. The charge on each capacitor will equalize and both will have about 2.5 volts. I just used this so it would be easier to visualize moving a charge.