Atomic-Scale Nonlinear Optics

As part of the DFG-funded CRC 1375 - Nonlinear Optics down to Atomic scales (NOA), we investigate the interaction of ultrashort laser pulses with metallic, semiconducting, and dielectric films and particles on the scale of only a few Angströms. Tailoring of materials at this scale gives rise to novel nonlinear phenomena, related to quantum confinement, enhanced surface nonlinearities and individual defect sites. We study effects, such as Harmonic generation in thin, dense ensembles of metallic nanoparticles and the interaction of novel materials with classic guided wave structures.

(right) Ultrathin metallic films and dense particle arrays are an intricate platform for nonlinear optics down to the atomic scale. (right) The study of laser-based plasmas requires stringent and precise laser pulse metrology methods.

One size fits all for the World's fastest Zoo: The Common Pulse Retrieval Algorithm COPRA

Femtosecond laser pulses are among the shortest man-made events. They are invaluable for many scientific and industrial applications, which is acknowledged by 2018's Physics Nobel Prize for Mourou and Strickland.

Their ultrashort duration is what makes the unique but also challenging. They cannot be characterized with conventional opto-electronics.

Several techniques tackle this problem - most of them known by their creative acronyms. Probably the most influential are frequency-resolved optical gating (FROG) and spectral phase interferometry for direct electric field reconstruction (SPIDER). They have inspired numerous adaptations and extensions.  Many techniques are conceptually similar: they measure spectra after a nonlinear process. A second measurement dimension is introduced, either by tuning a previous linear interaction or by changing the nonlinear process.

This led us to develop a common mathematical formalism, called parametrized nonlinear process spectra (PNPS) measurements, describing FROG, d-scan, time-domain ptychography (TDP), and MIIPS, among many others. Using PNPS we developed a common pulse retrieval algorithm (COPRA) that can be applied to all of these measurements [1], independent of the specific experimental implementation. Besides its universality it has the advantage of being fast and robust against noise commonly encountered in experiments.

Pulse retrieval from different PNPS measurements using COPRA: a)-d) Synthetic measurement traces with added Gaussian noise (σ = 1%). e)-h) The retrieved pulses (blue: intensity, orange: phase) and the original pulse (in black). The retrieval error ε quantifies the retrieval accuracy. The test pulse has a time-bandwidth product of 2. Taken from [1].

We have published a fully documented Python package that implements pulse retrieval algorithms including COPRA for many measurement techniques, available as open-source on github. It can be used to design pulse measurement experiments, retrieve pulses and to reproduce our COPRA paper [1]. Let us know if you are using it in your research and also if you have questions or comments.

Towards the World's Fastest 3D-Camera

3D-Vision is the next big thing in computer vision. It will help machine to interact more safely, efficiently and helpful with humans in our everyday life. While many approaches to develop 3D-vision systems have been develop, only few rival the micro-level precision and completeness of active pattern projection techniques. In active pattern projection a series of patterns is projected onto an object, from which the shape of the objects surface can be reconstructed with up to micrometer precision. Naturally the recording of multiple pictures takes time and therefore highly-dynamic processes, as can be encountered in traffic, sports or otherwise quickly moving systems are a big challenge for this technique.

We propose to combine active pattern projection with our expertise in spatiotemporal femtosecond pulse shaping to create a femtosecond high-precision 3D-camera system. Fields of research include the development of experimental setups for spatiotemporal pattern projection systems, hyperspectral real-time imaging systems, broadband laser sources, and the adaption of software for fringe projection systems. In this project we collaborate closely with the Fraunhofer Institute for Applied Optics and Precision Engineering IOF, with the clear aim to identify applications for our novel approach.

3D-Point-Cloud of a flat object recorded with a simplified version of a femtosecond 3D-camera.

Joining us/Collaborating

If you are interested in working with us, please feel free to contact us. We have several opportunities for students to work in the field of pulse metrology, either numerically or experimentally. The CRC 1375 operates a structured application portal, which can be found here.

Also, we are very interested in collaborating on solving challenging laser pulse measurement problems. Feel free to contact us.


[1] N. C. Geib, M. Zilk, T. Pertsch, F. Eilenberger, "Common pulse retrieval algorithm: a fast and universal method to retrieve ultrashort pulses," accepted in Optica, pre-print available at

Border Bottom