Rapid response: the first direct visualization of electron movement in an atom. Credit: M. DRESCHER ET AL.

“It has been a fantastic year,” beams Ferenc Krausz. His team at the Vienna Institute of Technology has used some of shortest laser pulses ever generated to track electrons moving within an atom. This motion is so quick that the vibrations of atoms — themselves among the fastest events subjected to scientific scrutiny — seem sluggish in comparison. Fundamental insights into atomic behaviour are likely to follow, as well as ideas for developing new kinds of laser.

For more than a decade, pulses of laser light just a few hundred femtoseconds (10−15 s) long have been used to follow the movement of atoms and molecules during chemical reactions. But tracking electrons requires even shorter pulses. An electron orbits a hydrogen atom in just 24 attoseconds (10−18 s), for example. Methods for producing suitably short pulses have emerged in the past couple of years1,2, and 2002 saw their first real application.

To get a sufficiently short laser pulse, Krausz bombarded neon gas with pulses of visible light 7 femtoseconds long, which sparked collisions between the neon atoms' nuclei and their electrons. This activity generated bursts of X-ray radiation about 100 attoseconds long, which Krausz focused and combined to make a single 650-attosecond pulse.

Using such pulses, Krausz was able to track the movement of electrons around the nucleus of a krypton atom for the first time. A 650-attosecond pulse was used to initiate the Auger process — the rearrangement of electrons around an atom that follows the removal of an electron from an inner orbit. By applying their pulse, the researchers knocked an inner electron out of its orbit, leaving an unstable hole. This caused an electron from an outer orbit to fall into the hole, which in turn displaced another electron — the Auger electron — from its outer orbit.

The researchers tracked the Auger electron using a femtosecond laser pulse, which allowed them to work out the time between the initial X-ray pulse and the expulsion of the Auger electron. They were able to pin the length of this process down to about 7.9 femtoseconds (ref. 3).

Other methods had already produced similar estimates by indirect measurements, but Krausz's experiment confirmed that sub-femtosecond pulses could be used to study the process by manipulating the electrons themselves. “This is the dream of attosecond science,” says Paul Corkum, a physicist at the Steacie Institute for Molecular Sciences in Ottawa, Canada. He suggests that insights into atomic behaviour could also come from studying the way in which electrons are scattered in the collisions that create the short X-ray pulses4.

X-rays can also be generated when electrons move between different atomic orbitals. Krausz hopes that future attosecond studies will improve our sketchy knowledge of the processes involved. If we understood these, he argues, it might be possible to build a compact X-ray laser to study molecular structures in materials science and biology.