When I read the news from the ICFO that in the Graphene project (a European Flagship project) a team of researchers from ICFO and ICREA have managed to create the fastest conversion from light to electricity I was not particularly impressed. It is no news that researchers manage to speed up a bit more what was already really fast just yesterday.
Then my eye caught what they meant by "fast": below 50 fs, that is below 50 femtoseconds. Now, in a femtosecond a ray of light, that we know runs pretty fast, moves for just 0.3µm, that is the size of a virus. Less than 50 fs is less than the time it takes light to move 15µm. It would take a ray of light 70 times as much to cover 1mm.
Now, that is really fast. And yet, 0.3µm is a distance where you can have 3,000 atoms one after the other. Since the conversion from light to electrical current requires just moving one electron there is still plenty of space to do this faster! As Richard Feynman said: "There is plenty of room at the bottom".
Yet, we are talking about conversion speeds that are 20 times faster than the fastest transistor (that in special cases have been clocked, in a lab, at 1THz).
The researchers have been able to achieve this speed by using as conversion point the edge where one graphene layer meets two graphene layers with impurities creating a difference in the Fermi energy of the carbon atoms at the edge. A short pulse of light (photons) is sufficient to tip the equilibrium balance and generate an electrical current. This opens the way to even faster optoelectronic switches that might be needed once we move from the Exabyte Internet to the Zettabyte Internet.
Still a few more years, something for the next decade.