Attosecond Physics Has Arrived

An Austria-Canada-Germany collaboration (Ferenc Krausz, Vienna Institute of Technology, 011-43-1-58801-38711, ferenc.krausz@tuwien.ac.at) reports that it has produced and detected, for the first time, isolated x-ray pulses lasting on the scale of attoseconds, where one attosecond is a billionth of a billionth of a second (10^-18 s). The reported pulses, lasting approximately 650 attoseconds (as) and residing in the soft x-ray part of the electromagnetic spectrum, subsequently provided attosecond-scale measurements of a physical phenomenon (specifically, the detachment of an electron from an atom by an x-ray photon).

With these observations, and several earlier ones by other groups, attophysics becomes the short-timescale frontier of physics. It replaces femtochemistry, the production of light pulses at the 10^-15 s (femtosecond) scale, in this regard. Just as a strobelight can yield stop-action photographs of a falling water drop, femtosecond pulses can capture the ultrafast steps of a chemical reaction between multiple atoms or molecules. But attosecond pulses are better equipped to capture the even speedier motions of electrons within atoms.

If light can be imagined as a wave of peaks and valleys, a one-second visible light pulse is a train of roughly 600 trillion peaks and valleys in length. The researchers report an attosecond pulse just 200 nanometers long, carrying just over a dozen peaks and valleys. Therefore, the duration of a light pulse can be thought of as the length along its direction of travel. A 1.28-second pulse can stretch from an earthbound laboratory to the moon; a 650-attosecond pulse would barely span the length of two typical viruses.

Previous experiments have reported evidence of trains of attosecond pulses following each other roughly every 1 fs (Papadogiannis et al., Physical Review Letters, 22 November 1999, covered in Update 467; Paul et al., Science, 1 June 2001). Other experiments (Bartels et al., Nature, 13 July 2000) have, according to their authors, already accessed attosecond-scale physics with femtosecond pulses (Christov et al., Physical Review Letters, 11 June 2001). But the new experiment, according to Krausz and his colleagues, represents the first detection and measurement of isolated attosecond pulses. Such isolated pulses, Krausz states, are important for taking attosecond-resolution snapshots of electron motion in atoms.

To accomplish their feat, the researchers first prepared an intense fsec pulse and aimed it at neon gas. The interaction between the neon gas and the fsec pulse created an attosecond-scale pulse in the soft x-ray range. According to a helpful theoretical picture (Corkum, Physical Review Letters, 27 September 1993), the fsec pulse ejects electrons from neon atoms, and the resulting oscillations of the electrons in the bath of fsec light produce an even shorter-duration soft-x-ray pulse.

Producing attosecond light is only half the battle. The researchers then had to measure its duration. By adjusting the delay between the times at which the x-ray pulse and a fsec visible pulse hit a gas of krypton atoms, the researchers affected the spectrum of energies in the electrons liberated from the atoms. Such modulations in the observed energy spread served as evidence for an x-ray pulse of 650 attoseconds. Henry Kapteyn of JILA/University of Colorado (303-492-8198, kapteyn@jila.colorado.edu), a member of a competing group, claims that the evidence is ambiguous as to whether the collaboration detected isolated attosecond pulses or trains of attosecond pulses. In reply, Krausz has invited Kapteyn to submit a formal critique of the experiment to a scientific journal so as to provide an opportunity for his team to respond in a similar fashion. This web page will post references to such comments if and when they become available.

However this debate pans out, attosecond metrology has arrived, and it will doubtlessly lead to some staggering physics experiments never before possible. (Hentschel et al., Nature, 29 November 2001; some other papers leading to this accomplishment are Sartania et al., Optics Letters, October 15, 1997; Schnürer et al., Physical Review Letters, 26 July 1999; and Drescher et al., Science, 9 March 2001; other related articles are Richard Fitzgerald, Physics Today, September 2000 and Corkum, Optics and Photonics News, 6 March 1995.)