Developing attosecond spectroscopy for lasers with longer pulse durations

Gary Boas, gary.boas@photonics.com

The development of high-speed photography in the 19th century demonstrated to equestrians and other interested parties that all four of a horse’s hooves leave the ground when it runs, presumably settling bar bets from Saratoga, N.Y., to Louisville, Ky. Similarly, in the 1980s, the emergence of laser pulses lasting only a few femtoseconds (10^—15 s), enabled researchers to record the motion of atoms in molecules – with the ultrafast pulses acting like the shutter of a camera.

Over the past decade, researchers have developed technology with which to record electron processes occurring on the timescale of attoseconds, including subfemtosecond laser pulses. Shown is the first source of light pulses shorter than 1 fs, generated by exposing neon atoms to 0.3-mJ, 5-fs, near-infrared laser pulses. The fluorescence emission originates from ionizing neon atoms streaming from the interaction volume (thin metal tube). Photonics Institute, Vienna University of Technology, courtesy of Dr. Gabriel Tempea. Used with permission of attoworld.de.

Questions remained, however, with respect to the motion of electrons in atoms as well as in molecules and solids. Electron processes occur not just on a much smaller spatial scale than molecular dynamics but on a considerably faster timescale: from tens to hundreds of attoseconds (10^—18 s). Much shorter laser pulses therefore were needed.

The first isolated attosecond laser pulses were reported in 2001. They proved too dim, however, to be used as both the “pump” and the “probe” in time-resolved spectroscopy of processes occurring on the attosecond scale. To address this, investigators developed an approach that uses a few-cycle infrared laser pulse with a well-controlled electric-field waveform instead of the attosecond probe pulse. Practical attosecond spectroscopy thus became a reality.

To date, researchers have demonstrated pulses as short as 80 attoseconds, and, using these, have achieved temporal resolution of close to 24 attoseconds, the atomic unit of time.

Attosecond spectroscopy holds considerable promise for a wide range of applications – from probing the microscopic origins of disease to developing electronic circuits in which currents can be switched on and off with the electric field of light – but the technology used to perform the measurements often is both fragile and complex, calling for spectral lasers that only a few labs know how to build and that require a significant amount of skill and training to operate.

The next step in developing the method, therefore, may be to make the technology accessible to a larger user base. “In order for attosecond lasers to be effective as a tool outside a few specialized laboratories,” said Zenghu Chang, an investigator at Kansas State University, “turnkey sources of strong attosecond pulses need to be developed.”

In the Oct. 30, 2009, issue of Physical Review Letters, Chang and colleagues reported a generalized double optical gating technique with which to produce single attosecond pulses with 20- to 28-fs lasers. They previously had described double optical gating for generating single pulses with approximately 10-fs lasers, combining two-color and polarization gating, but there was an upper limit on how long the input laser pulse could be.

Researchers have reported a method called generalized double optical gating that enables attosecond spectroscopy with lasers with longer pulse durations than were previously used with the technique.

The single attosecond pulse is generated within one of 10 cycles of the femtosecond laser pulse used, Chang said. “The difficulty is to avoid complete ionization of the target atom by the laser field of the cycles before the attosecond pulse is produced. In the generalized double optical gating method, the unwanted ionization is suppressed by transforming the field before the generation cycle into elliptical polarization.” The researchers demonstrated this method by producing single isolated pulses of 260 attoseconds with 20-fs laser pulses from a hollow-core fiber and pulses of 148 attoseconds with 28-fs amplifier pulses.

Being able to use lasers with longer pulse durations could open up the field to a much larger number of research groups. Many labs already have 25-fs lasers, for example; 5-fs lasers can be difficult to construct – and difficult to operate, even if they are available. Also, optical components such as waveplates and mirrors are considerably less expensive for 25-fs lasers, helping to drive down the overall costs of systems used for attosecond spectroscopy.