No one can match these electrons when it comes to relaxing. Within a few hundred attoseconds – billionths of a billionth of a second – of being hit by an X-ray pulse, they are already back where they were, sitting calmly in a low-energy state.

  Experiments that have clocked them doing so have solved a decades-old puzzle about the way electrons behave inside solids – and could improve the best X-ray lasers.

  In 2008, Eleftherios Goulielmakis at the Max Planck Institute for Quantum Optics in Garching, Germany, and his colleagues generated what were then the shortest pulses of light ever achieved: extreme ultraviolet bursts just 80 attoseconds long. Last year, the team created brief pulses of visible light too, each 380 attoseconds long.

Now, Goulielmakis and his colleagues have used a combination of these pulses to probe the behaviour of virtual particles called excitons.

  We know that an electron inside an atom can become excited, as physicists put it, when it absorbs energy from a photon of light. In doing this, it leaves a positively charged hole in its former, non-excited place – and the hole and excited electron together form an exciton.

   But physicists can’t agree whether excitons form in all such situations. In particular, if an atom inside a solid such as a silica wafer is hit by an X-ray photon and a hole forms, it’s not clear that the excited electron hangs around to form an exciton. It might instead escape the atom and move freely through the wafer. It doesn’t help that surrounding electrons are attracted to the hole and destroy it very quickly. This means that even if an exciton forms, it would be too short-lived for anyone to observe it – until now, that is.

  “No scientist has been able to do the necessary experiments,” says Goulielmakis. What was needed was laser pulses short enough to photograph any excitons before they “relax”, or dissipate. “We can now attack physical questions that have tortured the physics community for decades,” he says.

Fastest relaxation

   Goulielmakis and his colleagues zapped a 125-nanometre-thick wafer of silica with a 200-attosecond-long pulse of low-energy X-rays. Immediately afterwards, they used similarly short pulses of visible light to take “snapshots” of the system, capturing visual evidence that excitons really do form inside such solids – although some of them disappear again in just 750 attoseconds.

  “This is the fastest relaxation process we have ever seen in a natural solid system,” says Goulielmakis.

It’s a beautiful example of the power of this sort of ultra-fast spectroscopy, says Martin Schultze, also at the Max Planck Institute in Garching but not involved in this work. “Now we can study how their build-up and relaxation evolves.”

   Mackillo Kira at the University of Michigan in Ann Arbor says the results are truly exciting. Even though X-ray photons carry much more energy than photons in the visible spectrum, the study shows that the way they interact with electrons produces exciton-like behaviour after all. Finding out more could have implications for designing and building new forms of laser that work at wavelengths beyond those that conventional lasers operate at.