Mobile version: On
Computational Radiation Physics
We model, simulate and visualise the dynamics of particles and radiation phenomena that are of interest when investigating the physics of laser particle acceleration.
We aim to create models for innovative compact sources of radiation that make the best use of the ultra-strong electromagnetic fields that are created in the interaction of light and matter at relativistic intensities.
We work closely together with the experimental group from the laser particle acceleration division, developing realistic models and simulations to foster planing and analysis of experiments performded at the high-power laser systems DRACO and PEnELOPE. Working in close collaboration with the local experimental groups is one of our big strengths.
Advanced Radiation Sources
Our main research topic is the theory of laser-driven radiation sources. We study these sources by building analytic models and complex simulations to better understand and control the properties of these sources.
In order to increase our understanding of these sources, we have to be able to gain insight into the dynamical behaviour of large many-particle systems. High-intensity lasers can ionise matter, forming a new state of matter called a plasma. This means that they can rip matter apart, separating the positively charged atomic nuclei, so called ions, from their negatively charged electrons. This charge separation can create strong electric fields which in turn can be used to accelerate charged particles.
Physical Models of such a plasma need to take into account a variety of physical processes that can occur during the evolution of the plasma. Important processes include the ionisation of atoms by the laser pulse or by collisions with other particles but also radiation emitted by charged particles and collisions between particles.
These processes can happen on various time and length scales. This challenges both simulation techniques and theoretical models of these processes. One of our projects is PIConGPU, an implementation of the Particle-in-Cell algorithm for GPU clusters.
Our current research interests are
- Laser-driven acceleration of ion beams
- Laser-driven acceleration of electron beams
- Laser-driven X-ray sources
Complex, large-scale simulations
Realistic simulations of the interaction of high-power laser pulses with matter require to compute the trajectories of several million to a few billion charged particles in electromagnetic fields that strongly vary in space and time. It is therefore important, to use high performance computers with several hundred to thousands of processors. For this we are developing new parallel computing schemes to reduce the simulation time and thus the time the scientists have to wait for results.
Besides our physics research program we thus deal with information technology topics such as
- Massively-parallel simulations on new compute hardware such as GPUs
- 3D visualization and interaction for data analysis (3D caves, Kinect)
Combining accelerator physics and laser physics
Besides working on new acceleration schemes using lasers, we are interested in making use of standard accelerator techniques in combination with these new sources. For many applications, transport or focusing of laser-driven particle beams is essential.
Our work currently focuses on the following topics in accelerator physics:
- Laser cooling of relativistic ion beams
- Compact beam transport systems (permanent magnet quadrupoles, pulsed magnets)
Our research in laser-driven ion acceleration focuses on the interaction of a laser pulse with new targets. By changing the material, form and structure of the target as well as the way they interact with the laser pulse it is possible to control the properties of the resulting ion beam such as its maximum energy and its spectrum. Good control over these properties is important for future applications such as tumour therapy with laser-driven ion beams.
Currently it is not possible to accelerate ions directly by the field of the laser pulse. Instead one must accelerate the electrons in the target in such a way that the resulting charge separation between electrons and ions creates fields strong enough to accelerate the ions.
Theoretical research in our group is closely related to the experiments performed in the group of Dr. Stephan Kraft.
Beams of high-energy electrons can be created by focusing a high-intensity laser pulse into a gas. Inside the gas, electrons are accelerated collectively, creating local changes in the density of electrons in the gas, thereby forming electron density waves.
Under the right conditions these density waves which often have a wavelength of only a few microns can travel distances of centimetres through the gas without being perturbed. If one injects electrons into these waves, one can accelerate them to very high energies.
One of our research topics is to understand how this mechanism can be used to repeatedly accelerate electrons.
Powerful sources of X-ray radiation with small bandwidth are an important tool to study the atomistic structure of matter. At HZDR experiments on Thomson scattering of high-intensity laser pulses on relativistic electron beams are pursued.
Our group tries to answer the question, how this scattering process can be optimised and if it is possible to create laser-like pulses of X-ray radiation with it. For this we develop analytic models and use various simulation techniques.
One of our most important research tools are massively-parallel simulations. It is our goal to enhance the quality and extend the applicability of the physical models our simulations are based on. We furthermore want to make optimum use of the most innovative computation hardware when running our simulations.
For this we use new programming techniques such as Domain Specific Embedded Languages and Metaprogramming as well as high performance computers and new accelerator hardware such as GPUs. For more details please visit the project website of PIConGPU, a powerfull implementation of the Particle-in-Cell algorithm on GPUs.
We also need to understand the simulation results, so we experiment using new data analysis techniques using data warehouses or gesture control of visual data analysis.
At future accelerator facilities such as FAIR in Darmstadt cooling of highly-relativistic ion beams will be necessary to conduct precision experiments. A promising new cooling method is laser cooling. Laser Cooling is normally used in ion traps, for example when studying quantum computing.
We want to use laser cooling to create ultra-cold ion beams in storage rings and study their properties.
Laser-driven radiation sources are usually more compact than standard accelerators. In order to make good use of them, they have to be coupled to some sort of beam transport system.
We try to find new ways to shrink these beam transport systems in size while making them more powerful and versatile. This allows us to built very compact combinations of laser-driven sources and transport beam optics.
Our current interest lies in permanent magnet quadrupole lenses and pulsed magnet beam optics.
In the News
- China-Germany Symposium for Laser Cooling at Storage Rings
- Dr. Thomas Kluge wins HZDR prize for his dissertation
- Inauguration of Hypnos HPC System at HZDR
- Shedding new Light on Laser Heating (DISCOVERED 2/2012)
- CUDA in Action Spotlight
- Transtec Edition12
- Dresden ENLITE 2012 at MPI-PKS
- Wissenschaft als visuelles Erlebnis
- Technische Universität Dresden named a CUDA Center of Excellence
- Laser Heating – New Light Cast on Electrons Heated to Several Billion Degrees
- Part-Time Scientists: HZDR’s First Summer Student Program Launched
- TU Dresden and HZDR at Supercomputing conference SC'11 presenting PIConGPU
- Insider 3/2011: ELBE, DRACO & Co. virtuell erlebbar
- Scientists from TU Dresden and HZDR team up with graphic cards manufacturer
- Ultra-short, high-density electron pulses for advanced x-ray sources
- World of Physics Podcast "Laser Particle Acceleration" (German version only)
- Laser accelerated protons to the highest energies so far