CORDIS - EU research results
CORDIS

Extreme ultraviolet and X-ray spectroscopy to understand dynamics beyond the Born Oppenheimer Approximation

Final Report Summary - XBEBOA (Extreme ultraviolet and X-ray spectroscopy to understand dynamics beyond the Born Oppenheimer Approximation)

The interaction among electrons and between electrons and nuclei is described by a many particle Schrödinger equation, which is in general too complex to be solved exactly. The Born-Oppenheimer approximation simplifies the electron-nuclei interaction. This approximation breaks down if two electronic states cross each other in a conical intersection (CI). CIs are of enormous importance in nature and are thought to be responsible for a wide variety of excited state chemical processes such as DNA photoprotection, light harvesting, vision, and atmospheric chemistry. Understanding the details of the molecular electronic dynamics in CIs means shedding light on the most fundamental processes of our very existence and can improve strategies in clean energy research. Dynamics related to the breakdown of the -Born-Oppenheimer approximation is happening on an ultrafast timescle. Thus ultrafast spectroscopic tools are applied where generally one femtosecond pulse (termed "pump") launches a photochemical or photophysical process. A second ultrashort pulse (or a series of pulses) probes the reaction in a second interaction and ion yield, absorption, photoelectrons or fluorescence are recorded as a function of the pump-probe time delay. However, the separation of the electronic transition from the nuclear wavepacket dynamics in the experimental transient is not per se possible.

The project "Extreme ultraviolet and X-ray spectroscopy to understand dynamics beyond the Born Oppenheimer Approximation" (XBeBOA) aimed at the development of tools to study ultrafast electronic and nuclear dynamics in metal-organic complexes that go beyond the Born-Oppenheimer approximation. These tools use high harmonic generation (HHG) as a novel, emerging source of femtosecond extreme ultraviolet light pulses (XUV, photon energies of 10-100 eV). Experimentalsetups were developed during the first part of the project , namely an XUV transient grating spectrometer and an XUV photoelectron spectrometer for gas-phase studies. During the second phase, we comissioned a beam line for time-resolved photoelectron spectroscopy of liquid solutions. In addition, XUV and x-ray (photon energies >100 eV) studies of molecular samples were performed at synchrotrons and free electron lasers (FEL). These three sources for medium and high photon energy light pulses have complementary strengths in terms of spectral and time resolution, machine parameters, accessibility and flexibility and enable us to address distinct questions about molecular photochemistry.

Using the XUV transient grating technique, the insulator-to-metal phase transition in the strongly correlated material vanadium dioxide, VO2, was investigated. Our experiments allowed for a clear distinction of spectral features due to electronic and nuclear dynamics - which is without doubt one of the major challenges in ultrafast spectroscopy. Further improvements of the setup will enable us to track the very first step of this process that take place on the few tens of fs time-scale in the future.

Importantly, we used the XUV transient grating setup to demonstrate signal heterodyning at wavelengths of a few tens of nanometers. Heterodyning is a powerful way to amplify weak signals but exceedingly difficult to implement experimentally at short wavelengths because of the need for phase stability between the signal and the heterodyne field. We achieved this by structuring the sample with a periodic pattern (i.e. imprint a permanent grating), which creates a heterodyne wave that is intrinsically phase stabilized with the signal wave since they are generated on the same element, namely the sample. We demonstrated the feasibility of heterodyning in the XUV and plan to bring it to even shorter wavelength, with transient grating experiments at free electron lasers.

We also implemented a HHG based photoelectron spectroscopy (PES) setup for gas phase samples. This system employs a combination of an indium filter with an aluminum mirror to separate the 9th harmonic (photon energy 14 eV) from the XUV quasi-continuum without temporal broadening of the pulse. High collection efficiency is obtained with a magnetic bottle type time-of-flight spectrometer. A particular asset of this setup is its ability to record XUV and multiphoton PES under otherwise identical conditions. This feature was used later to illustrate how changes observed in the UV pump – multiphoton IR probe PES are dominated by artifacts that originate from the intermediate state involved in the ionization process and do neither reflect electronic nor nuclear dynamics of the excited state under investigation.
We finalized the liquid-microjet photoelectron setup at the Laboratory of Ultrafast Spectroscopy (LSU). A few groups worldwide are seeking to run this experimentally complex but extremely promising technique. We commissioned the time-preserving monochromator which can selectively provide either high spectral or high temporal resolution. Remarkably, we were able to resolve the approx. 20meV spacing in the vibrational progression on the water 1b2 photoelectron peak at 18eV binding energy. First time-resolved experiments show the significant improvement of the data quality owing to the much higher photon energies, flux and laser repetition rates routinely achieved in the LSU setup as compared to other ones.

In addition, we investigated the static XUV PES spectra of iron pentacarbonyl, Fe(CO)5, and chromium hexacarbonyl, Cr(CO)6, in vicinity of the 3d ← 3p resonance. A so-called Fano line shape analysis of the Fe(CO)5 PES enabled us to determine the amount of 3d metal character of the valence molecular orbitals. Using the Fano formalism, this crucial parameter can be easily accessed experimentally from a simple measurement of relative cross-sections.

Finally, we used the LCLS free electron laser to study the response of C60 fullerene to intense ultrashort x-ray pulses. Owing to the large photoabsorption cross-section of carbon 1s electrons at 485 eV photon energy, we could reach conditions in which each atom in a C60 molecule absorbs multiple photons during x-ray pulses of 4–90 fs duration. Such will be the conditions in imaging of proteins and viruses at FELs and C60 served as model system for future investigations of biomolecular samples. As our investigation showed, the Coulomb explosion of the molecule - which will limit the ability for undistorted imaging - can be described by classical mechanics. The inclusion of the photoelectric effect, secondary ionization, recombination, as well as molecular Auger processes and chemical bonding in the model produced excellent agreement between experiment and calculation.

Overall, we successfully implemented several novel experimental strategies and demonstrated their unique powers for the study of transition metal compounds in gaseous, solid and liquid phase. Owing to the numerous collaborations that have been established during this fellowship, it promoted knowledge transfer not only between the two host groups, but within a broad network of researchers.