6
A. Rouzée et al.
The interest in the use of these novel XUV/X-ray light sources in atomic and
molecular physics can be rationalized both in the time and frequency domain.
Viewed in the time domain, the inherently short optical periods of XUV/X-ray light
(τ optical = λ/c, where λ is the wavelength and c is the speed of light), allows the synthesis of pulses with unprecedented pulse durations, accessing the sub-femtosecond
i.e. attosecond domain [17, 18, 21]. Such pulses are ideal for the investigation of
electron dynamics on its natural timescale. The generation of attosecond laser pulses
requires the availability of a process that generates light in the XUV/X-ray regime
over a large enough bandwidth (E ≥ 5 eV) and with an appropriate phase relationship between the different frequency components contained within the pulse. This
is precisely what the HHG process does, given the one-to-one relationship between
the ionization time within the optical cycle of the driving infrared laser, the kinetic
energy at the time of the electron-ion re-collision and the photon energy produced.
Under typical HHG conditions, XUV/X-ray bandwidths in excess of 20 eV are easily achieved, and the pulse duration is determined by the chirp that is generated in
the HHG process. The shortest pulses reported to date are about 80 attoseconds long
[22], and it is to be expected that the existence of even shorter pulses will soon be
demonstrated. So far, pulses obtained at XUV/X-ray FELs are still in the femtosecond domain, but ideas exist that would allow to significantly shorten the pulses [23].
At LCLS, X-ray laser pulses with a pulse duration below 10 fs have already been
achieved [24].
Viewed in the frequency domain, the short wavelength and thus intrinsic high
photon energy of XUV/X-ray light sources creates the ability to produce high energy
photoelectrons. As we will discuss, this allows configuring molecular pump-probe
experiments where photoelectrons are produced with kinetic energies where the de
Broglie wavelength becomes comparable to or smaller than the internuclear distances in the molecule, so that the angular distribution of the ejected photoelectron
encodes information on the molecular structure. Mentioning the time domain, attosecond science context is highly relevant here, since the intensive and widespread
efforts to develop and characterize attosecond light pulses have largely been responsible for the emergence of the experimental protocols that need to be used when
MFPADs are to be measured using XUV/X-ray light generated by HHG. Motivated
by the requirements for attosecond science experiments, it has become possible to
develop interferometrically stable multi-color pump-probe setups, with appropriate
optics that can be used to image, focus, split and recombine the XUV/X-ray light
beam. An example of such a setup is shown in Fig. 1.2 and corresponds to the setup
that is in operation at the Max Born Institute (MBI) in Berlin.
If one wishes to time-resolve the evolution of internuclear distances in a molecule
(in other words, make a “molecular movie”) using photoelectrons that are ejected
from the molecule using XUV/X-ray light, then it is imperative that the photoelectron angular distribution is observed in the molecular frame. One way to do this is
by making use of a so-called reaction microscope [25], where the 3D momentum
of ejected photoelectrons is measured in coincidence with the 3D momentum of
fragment ions that are formed, and where in the axial recoil approximation the lat-
A. Rouzée et al.
The interest in the use of these novel XUV/X-ray light sources in atomic and
molecular physics can be rationalized both in the time and frequency domain.
Viewed in the time domain, the inherently short optical periods of XUV/X-ray light
(τ optical = λ/c, where λ is the wavelength and c is the speed of light), allows the synthesis of pulses with unprecedented pulse durations, accessing the sub-femtosecond
i.e. attosecond domain [17, 18, 21]. Such pulses are ideal for the investigation of
electron dynamics on its natural timescale. The generation of attosecond laser pulses
requires the availability of a process that generates light in the XUV/X-ray regime
over a large enough bandwidth (E ≥ 5 eV) and with an appropriate phase relationship between the different frequency components contained within the pulse. This
is precisely what the HHG process does, given the one-to-one relationship between
the ionization time within the optical cycle of the driving infrared laser, the kinetic
energy at the time of the electron-ion re-collision and the photon energy produced.
Under typical HHG conditions, XUV/X-ray bandwidths in excess of 20 eV are easily achieved, and the pulse duration is determined by the chirp that is generated in
the HHG process. The shortest pulses reported to date are about 80 attoseconds long
[22], and it is to be expected that the existence of even shorter pulses will soon be
demonstrated. So far, pulses obtained at XUV/X-ray FELs are still in the femtosecond domain, but ideas exist that would allow to significantly shorten the pulses [23].
At LCLS, X-ray laser pulses with a pulse duration below 10 fs have already been
achieved [24].
Viewed in the frequency domain, the short wavelength and thus intrinsic high
photon energy of XUV/X-ray light sources creates the ability to produce high energy
photoelectrons. As we will discuss, this allows configuring molecular pump-probe
experiments where photoelectrons are produced with kinetic energies where the de
Broglie wavelength becomes comparable to or smaller than the internuclear distances in the molecule, so that the angular distribution of the ejected photoelectron
encodes information on the molecular structure. Mentioning the time domain, attosecond science context is highly relevant here, since the intensive and widespread
efforts to develop and characterize attosecond light pulses have largely been responsible for the emergence of the experimental protocols that need to be used when
MFPADs are to be measured using XUV/X-ray light generated by HHG. Motivated
by the requirements for attosecond science experiments, it has become possible to
develop interferometrically stable multi-color pump-probe setups, with appropriate
optics that can be used to image, focus, split and recombine the XUV/X-ray light
beam. An example of such a setup is shown in Fig. 1.2 and corresponds to the setup
that is in operation at the Max Born Institute (MBI) in Berlin.
If one wishes to time-resolve the evolution of internuclear distances in a molecule
(in other words, make a “molecular movie”) using photoelectrons that are ejected
from the molecule using XUV/X-ray light, then it is imperative that the photoelectron angular distribution is observed in the molecular frame. One way to do this is
by making use of a so-called reaction microscope [25], where the 3D momentum
of ejected photoelectrons is measured in coincidence with the 3D momentum of
fragment ions that are formed, and where in the axial recoil approximation the lat-