Proceedings Article | 30 March 2020
The electron current dynamics occurring during the interaction of an intense laser pulse with a dielectric or a semiconductor material generates radiation up to the extreme ultraviolet (XUV) regime [1-3]. This phenomenon, called high-order harmonic generation (HHG), is a coherent process and is a result of the laser driven electron current oscillating at petahertz frequencies emitting photons upon the recollision of the electrons with the atom cores [4]. Since the excursion takes place within a crystal the harmonics are strongly influenced by the crystal structure and can exhibit fundamentally different behaviours compared to the well-known HHG in noble gases. In an theoretical approach this can be understood in a classical three step model where, during the interaction with a strong electric field, the electron is first tunnelled in the conduction band, is accelerated there and recombines after a certain time. Based on this concept two HHG driving mechanisms can be identified: The intraband HHG due to the oscillation of the electron in the Brillouin zone and the interband HHG due to the recombination to the valence band [5, 6].
In order to understand these contributions and fully characterize the electron trajectory a measurement of the attosecond dynamics is a crucial step. This information can be extracted from the HHG spectral phase, accessible using the RABBIT (Reconstruction of attosecond beating by interference of two photon transitions) technique. It is using two photon ionisation of a noble gas by one XUV and one fundamental photon. Since this process allows two channels for the same resulting electron energy, a beating signal at the even order harmonic positions is generated when scanning the XUV pulse with the fundamental pulse [7, 8]. The recorded photoelectron signal allows the reconstruction of the spectral phase and, therefore, direct experimental insight into the temporal attosecond structure of the XUV pulse train.
The RABBIT technology is well known and established for the characterization of HHG in noble gases. As a RABBIT target commonly also noble gases are used providing a lowest possible ionisation potential of 12.13eV for Xenon. Since we achieved HHG up to 25eV in magnesium oxide a RABBIT can work in a similar fashion as for HHG from noble gases. However, the contribution of inter- and intraband harmonics is different for each energy. This generates the need to reduce the Ionisation energy of the RABBIT target to observe the effects of both contributions, as well as, enabling studies on materials with lower HHG energy cut-offs (below 12eV typically) such as zinc oxide or gallium arsenide. For these reasons, we will implement a novel RABBIT setup, replacing the noble gas with alkali metals such as potassium (Ip=4.34eV), allowing a full characterization of the spectral phase down to the UV spectral range.
Ideally, we would like to resolve the electron dynamics involved in many semiconductors and dielectrics during strong field interactions. This knowledge is relevant for applications of attosecond pulses e.g. for the metrology of optoelectronic switches operating in the petahertz regime. In addition, exact knowledge hopefully allows the control of the generated pulses in time and space by enhancing or discriminating inter- or intraband processes or modulating the atto-phase of the pulse.
References
[1] Ghimire, S. et al., Nature Physics 7, 138 (2011)
[2] Luu, T. T. et al., Nature 521, 498 (2015)
[3] Ndabashimiye, G. et al., Nature 534, 520 (2016)
[4] You et al., Nature Physics 13, 345 (2017)
[5] Ghimire et al., Phys. Rev. A 85, 043836 (2016)
[6] Vampa et al., Phys. Rev. Lett. 113, 073901 (2014)
[7] P. M. Paul et al., Science 292, 5522 (2001)
[8] Y. Mairesse et al., Science 302, 5650 (2003)
