Hello --

To clarify: when you ask "why do we use few cycle laser pulses ...", are you referring to the infrared driving laser pulses? Most techniques for the generation of isolated attosecond pulses (rather than a train of attosecond pulses) require very short driving laser pulses, so this is why few cycle laser pulses are used so often in attosecond spectroscopy experiments.

I'm not sure what you mean in terms of "trying to compress the laser pulse from fs to attosecond". Are you asking about how attosecond pulses are generated?

I have attached two publications from members of our group that will hopefully help satisfy your curiosity.

~Eric

Article Sub-cycle Oscillations in Virtual States Brought to Light

Article XUV Time-Domain Spectroscopy Using Isolated Attosecond Pulse...

Three different approaches are used in pump-probe measurements aimed at attosecond time domain spectroscopy. Two of them rely on the direct use of attosecond light pulses. Because of low intensity of attosecond pulses produced so far, a combination of synchronized fs (IR) and as (VUV) pulses is commonly used instead of a couple of as pulses. (i) In the methods such as attosecond streak camera, the as pulse is used to trigger the process of interest, whereas a streaking of the released electron by the IR field is used as a probe. (ii) Just the opposite, in the attosecong absorption spectroscopy, the as pulses are used to probe the dynamics driven by the IR pulse. In both cases, a single as pulse is usually more useful than the as pulse train. The most direct way to produce an isolated as pulse is to use high-harmonic generation driven by a few-cycle IR pulse. (iii) The third group of methods, which is used to monitor the ultrafast molecular dynamics, provides attosecond measurements without attosecond light pulses. In these methods, the molecular target is probed by its own electron, which was released by the IR pulse and later reencounters the parent ion (for the reference see, e.g., the review [M. Lein, J. Phys. B 40, R135 (2007)]. This approach also benefits from using few-cycle IR pulses, since this ensures that the molecular target has not been changed significantly during the measurement.

Hi,

Attosecond pulses are generated through the non-linear excitation of electrons in plasmas using short optical pulses. Each cycle of the impinging field generates waves in the X-UV energy range. According to the duration-energy relationship delta t . delta omega ≥ 1/2, there must be very many frequency components in order to reach the attosecond regime. But also these created frequencies must obey an ordered phase relationship.

The ideal situation for generating attosecond pulses is therefore to use one or less cycle visible light pulses. The addition of cycles in the exciting pulse blurs the phase of the emitted fields.

Hope it's clear?