New frontiers of femtosecond magnetisation dynamics

Modern electronics is getting closer and closer to fundamental limits of the currently employed technologies. Therefore the need of faster and faster electronics is driving the attention towards the ultrafast dynamics as a route to increase the computation speed. It has been shown that the magnetism of ferromagnetic materials like iron or nickel can be modified in as little as 1 picosecond, which is one trillionth of a second. This could lead to the development of ultrafast information storage with unprecedented speed. However the modification was believed to be due to a local process. In my previous work I have theoretically shown how ultrafast transport of magnetisation could happen and drive an important part of ultrafast magnetisation dynamics. This can be considered as the seminal work that laid the basis to the field of femtosecond spin transport. This has been a huge step forward, since the non locality of the process points towards the possibility of transferring information. The transfer of information is what distinguishes information storage from electronics. In the past few years many experiments confirmed my theoretical predictions and discovered new unexpected ultrafast phenomena. I will continue developing the theoretical model, to include other very important effects, like for instance the magnetisation transport in semiconductors (which are the basis of modern electronics), or the possibility of inducing a full reversal of magnetisation through spin injection. I will also couple my model to more sophisticated description of real materials, with the aim of providing more precise material-specific predictions.

The ability of producing ultrashort laser pulses (with a pulse duration of the order of femtosecond, one quadrillionth of a second) permitted the monitoring of extremely fast dynamics. These timescales are already of extreme relevance to a number of existing technological applications: as for instance solar cells. However, scientists are also exploring the possibility of critically increasing computation speed by utilizing a range of exotic processes on the femtosecond timescale. In particular the ultrafast generation and transport of spin currents has attracted huge attention for its potential application to electronics running thousands of times faster than modern present-day electronics. Together with others, I have pioneered the field of ultrafast demagnetization. Within the Lise Meitner project, in particular, we have now proposed, among others, a method to inject such ultrafast spin currents into semiconductors, building a fundamental bridge between this new promising carrier of information and the widely used semiconductor information technology.