4
A. Rouzée et al.
ergies are needed than the photon energy of the equivalent X-rays. The de Broglie
wavelength of an electron is λ DeBroglie (a.u.) = π
√
2/E kin (a.u.), where E kin is the
electron kinetic energy. A de Broglie wavelength of ∼ 1 Angström (which, as a
laser wavelength would imply the use of 12.4 keV photons!), is already achieved
for electrons with a kinetic energy as low as ∼ 0.15 keV. It follows that it is significantly easier to prepare the short pulse electrons that are needed for a timeresolved electron diffraction experiment with atomic resolution, than it is to prepare
the short pulse X-rays that are needed for a time-resolved X-ray diffraction experiment.
Short electron pulses with kinetic energies in the 0.1–1000 keV range can be
generated externally to a molecule on a photo-cathode that precedes a small accelerator. Using such a technique impressive results have been achieved by Zewail
and co-workers [7–9] and by Miller and co-workers (see Fig. 1.1b) [10]. Applications have included studies of halo-ethane elimination reactions and ring opening of
cyclic hydrocarbons [7], phase transitions in cuprate semiconductors [9], the transition from a monoclinic to a final tetragonal phase in crystalline vanadium dioxide
[8], and laser-induced melting [10]. Already, these experiments can be performed
with a time resolution of approximately 100 femtoseconds. It remains to be seen if
pump-probe experiments with ca. 10 femtosecond time-resolution will become possible using this technique, although proposals to push the time resolution into the
attosecond domain have already been put forward [11].
In the last few years, our research team has started working on a number of
alternative methods that allow the generation of electrons with kinetic energies
in the 0.1–1 keV range, two of which will be detailed in this book chapter (see
Fig. 1.1c). First of all, in experiments performed at extreme ultra-violet (XUV)/Xray FELs like the FLASH free electron laser in Hamburg (the pre-cursor of the
European XFEL, which generates radiation down to 4 nm) and at LCLS, we have
explored the generation of fast electrons by XUV/X-ray photo-ionization as a means
to study time-resolved molecular dynamics. Like the time-resolved X-ray diffraction studies mentioned above, this work may be seen as a natural continuation
of earlier synchrotron-based experiments, where ideas to use XUV/X-ray radiation for “illuminating a molecule from within” were developed about a decade
ago [12, 13]. A progress report on the extension of these ideas to the time domain will be presented below. Secondly, in experiments performed at the midinfrared free electron laser FELICE (Free Electron Laser for Intra-Cavity Experiments) in the Netherlands, we have investigated strong-field ionization at wavelengths ranging between 4 and 40 µm. Under these conditions, considerable ponderomotive acceleration of the electrons that are freed in the ionization event sets
the stage for laser-driven re-collisions with the target from which the electrons
are ionized, allowing the experimental measurement of photoelectron holograms
that encode both molecular structure and dynamics [14]. These experiments are
discussed in the present chapter as well. We note that in future we are furthermore planning experiments where 0.1–1 keV electrons that can encode molecular structures will be ejected from (time-evolving) molecules by means of a collision of the molecule with a 100 keV electron beam that is similar to the elec-
A. Rouzée et al.
ergies are needed than the photon energy of the equivalent X-rays. The de Broglie
wavelength of an electron is λ DeBroglie (a.u.) = π
√
2/E kin (a.u.), where E kin is the
electron kinetic energy. A de Broglie wavelength of ∼ 1 Angström (which, as a
laser wavelength would imply the use of 12.4 keV photons!), is already achieved
for electrons with a kinetic energy as low as ∼ 0.15 keV. It follows that it is significantly easier to prepare the short pulse electrons that are needed for a timeresolved electron diffraction experiment with atomic resolution, than it is to prepare
the short pulse X-rays that are needed for a time-resolved X-ray diffraction experiment.
Short electron pulses with kinetic energies in the 0.1–1000 keV range can be
generated externally to a molecule on a photo-cathode that precedes a small accelerator. Using such a technique impressive results have been achieved by Zewail
and co-workers [7–9] and by Miller and co-workers (see Fig. 1.1b) [10]. Applications have included studies of halo-ethane elimination reactions and ring opening of
cyclic hydrocarbons [7], phase transitions in cuprate semiconductors [9], the transition from a monoclinic to a final tetragonal phase in crystalline vanadium dioxide
[8], and laser-induced melting [10]. Already, these experiments can be performed
with a time resolution of approximately 100 femtoseconds. It remains to be seen if
pump-probe experiments with ca. 10 femtosecond time-resolution will become possible using this technique, although proposals to push the time resolution into the
attosecond domain have already been put forward [11].
In the last few years, our research team has started working on a number of
alternative methods that allow the generation of electrons with kinetic energies
in the 0.1–1 keV range, two of which will be detailed in this book chapter (see
Fig. 1.1c). First of all, in experiments performed at extreme ultra-violet (XUV)/Xray FELs like the FLASH free electron laser in Hamburg (the pre-cursor of the
European XFEL, which generates radiation down to 4 nm) and at LCLS, we have
explored the generation of fast electrons by XUV/X-ray photo-ionization as a means
to study time-resolved molecular dynamics. Like the time-resolved X-ray diffraction studies mentioned above, this work may be seen as a natural continuation
of earlier synchrotron-based experiments, where ideas to use XUV/X-ray radiation for “illuminating a molecule from within” were developed about a decade
ago [12, 13]. A progress report on the extension of these ideas to the time domain will be presented below. Secondly, in experiments performed at the midinfrared free electron laser FELICE (Free Electron Laser for Intra-Cavity Experiments) in the Netherlands, we have investigated strong-field ionization at wavelengths ranging between 4 and 40 µm. Under these conditions, considerable ponderomotive acceleration of the electrons that are freed in the ionization event sets
the stage for laser-driven re-collisions with the target from which the electrons
are ionized, allowing the experimental measurement of photoelectron holograms
that encode both molecular structure and dynamics [14]. These experiments are
discussed in the present chapter as well. We note that in future we are furthermore planning experiments where 0.1–1 keV electrons that can encode molecular structures will be ejected from (time-evolving) molecules by means of a collision of the molecule with a 100 keV electron beam that is similar to the elec-