Sunday, 24 Sep 2017
You are here: Home Research Attosecond XUV spectroscopy

Attosecond time resolved XUV spectroscopy

Electronic processes in atoms take place on timescales that can reach into the attoseconds. To trace or trigger such dynamics one can use attosecond extreme ultra-violet (XUV). In current state-of-the-art experiments, the precisely controlled electric field of a few-cycle near-infrared (NIR) laser pulse first generates the attosecond XUV pulse in a process called high-harmonic generation (HHG). Subsequently, the XUV and the NIR pulse impinge on a system and trigger dynamics. Depending on the question to be asked, either of them can be the pump or the probe pulse.

Spectroscopic measurement techniques typically only provide partial insight into the complex dynamics. For instance, time resolved photo-electron spectroscopy (TRPES) measures the spectrum of kinetic energies of released electrons for variable time delays between pump and probe laser beams. This may give information about the electronic structure of the sample under investigation. However, no information on the created ions is provided and it is not possible to identify the origin of the emitted electrons, e.g., the part of a complex molecule where the electron originated from or the charge state of the emitting atom. On the other hand, time resolved photo-ion spectroscopy (TRPIS) measures the mass to charge ratio of the produced ionic fragments by means of a mass spectrometer. This gives a wealth of information about fragmentation, but not the underlying electronic structure. Due to the fact that ionization of the sample is required for both of these techniques, none of them provides information about non-ionized excited states. The technique of attosecond transient absorption spectroscopy (ATAS) is able to investigate these states by measuring the absorption of the XUV probe pulse for different delays with respect to the infrared pulse. In our experiment, the combination of all three techniques leads to significantly enhanced insight into the different dynamics in systems such as atoms, molecules or solids.

Experimental Setup

In Fig. 1 a schematics of the experimental setup and a model of the spectroscopy chamber are shown.


Fig. 1: Left: Schematics of the experimental setup. A chirped pulse amplifier (CPA) produces 25fs NIR laser pulses, which are spectrally broadened in a Neon filled capillary and temporally compressed down to ~5fs in a chirped mirror compressor. These few-cycle NIR pulses are focused in a Neon gas jet to produce attosecond XUV pulses by HHG. Right: Model of the spectroscopy chamber. (1) Filter to separate XUV and IR. (2, 8 & 9) Filter, mirror and lens, used for alignment and optimization. (3) Focusing mirror. (4) Sample gas jet. (5) time-of-flight (TOF) spectrometer. (6) Entry-slit of the XUV spectrometer, which consists of a XUV grating (7) and an XUV charged-coupled-device (XUV CCD). TS1-3 are motorized translation stages to move components in the beam path. XYZ is a motorized 3D translation stage to align the gas jet with the mirror focus.

High Harmonic Generation (HHG)

In HHG, a phase controlled NIR few-cycle pulse is focused into a gas jet. Due to the high intensity of the few-cycle pulse, XUV radiation is generated in the gas jet and co-propagates with the few-cycle pulse. Since the divergence of the XUV radiation is smaller than that of the generating near-infrared beam, it is possible to separate the XUV pulses from the near-infrared light by placing a small filter (1) in the center of the beam. By using a suitable filter one can select an isolated burst of XUV radiation per laser pulse. When the near-infrared and XUV beams are focused onto the gas target, the delay between the XUV and near-infrared pulses is controlled by shifting the central part of the focusing mirror using a piezo-actuator (3).

Time resolved spectroscopy

In TRPES experiments the XUV pulses ionize the atoms or molecules under investigation in the gas target (4). Depending on the delay between the NIR and the XUV pulse, the kinetic energy of the emitted electrons changes, and the final kinetic energy of the electrons is measured using a time-of-flight (TOF) spectrometer (5). When the electron kinetic energy is measured for a whole series of delays, it is possible to reconstruct the excitation and relaxation dynamics of the target under investigation. In TRPIS experiments the extraction voltages of the TOF are changed, so that instead of the energy of the emitted electrons the mass to charge ratio of the produced ions is measured. This way, knowledge of the temporal evolution of charge states, molecular orbits and fragmentation channels can be obtained.

In contrast to TRPES and TRPIS, in ATAS experiments instead of electrons or ions the intensity of light that passes through the target versus its wavelength is measured with an XUV spectrometer (6, 7, CCD). Important for this is the concept of resonance: The coupling of two quantum mechanical states is strongest when the energy that drives the transition matches the energy difference between the states. Therefore, photons whose energy match the energy difference between two electronic states of a target are absorbed stronger than others. By subtracting the absorption spectrum from a reference spectrum the electronic structure can be reconstructed. The main advantage of this technique is that binding excited states are accessible.

Example of attosecond transient absorption spectroscopy

As an example we show ATAS measurements on Krypton in Fig. 2. The left panel shows the static absorption spectrum that is obtained without the near-infrared laser pulse. It shows the absorption lines due to excitation from the Krypton 3d level to the 5p through the 8p levels. The peaks come in pairs due to the energy splitting of the Krypton 3d level. The right panel shows the absorption spectrum as a function of delay between XUV and near-infrared pulse in the range from -10fs to 10fs in steps of 0.1fs. Data analysis revealed modulations of the intensity of the 5p and 5p' absorption lines depending on the time delay between XUV and IR pulses. The origin of these modulations and whether the remaining peaks exhibit similar time delay dependence are currently subject to further measurements and data-analysis.


Fig. 2: Left: Static absorption spectrum of Krypton for a photon-energy range from 90 to 100eV. Absorption lines due to excitation from the Krypton 3d level to the 5p through 8p levels are visible. Dashed energy levels are due to energy splitting. Right: ATAS spectrum of Krypton from 90eV to 100eV for time delays of -10fs to 10fs in steps of 0.1fs.


Last update May, 2017