Helmholtz awards Doctoral Prize

Celia Dobersalske

Deutsches Krebsforschungszentrum

Our understanding of the interaction between the immune system and brain tumors remains insufficient to fully harness the potential of immunotherapy. Despite numerous approaches, the most aggressive brain tumor – glioblastoma – remains incurable to this day. As part of her translational doctoral research, Celia Dobersalske, together with her colleagues, initially characterized immune stem and progenitor cells within tumor tissue, leading up to the discovery of tumor-reactive, cytotoxic CD8+ T lymphocytes in the skull bone. These cells may play a critical role in the body’s defense against cancer. Innovative imaging studies in patients, correlated with disease progression, provided initial indications of the therapeutic potential of this discovery. Future research will explore how these unique cells can be specifically leveraged to improve treatment strategies.

“Realizing that we had discovered something completely new was an unforgettable moment. Our finding that the human skull bone harbors a previously overlooked immune reservoir next to brain tumors added a crucial piece to the evolving understanding of the brain, which was long considered immunologically isolated. That moment marked a pivotal turning point for me.”

Clara Vázquez García

Max Delbrück Center

In her thesis, Clara Vázquez García developed a new technology called SWIBRID – short for SWItch-joint Breakpoint Repertoire IDentification. This method analyzes small DNA scars, that naturally occur during the production of antibodies. At first, SWIBRID was used to study how fragments of DNA get inserted into antibody genes in both humans and mice. But the project soon shifted toward a translational focus: using SWIBRID to identify immune and DNA repair deficiencies in individuals. Using SWIBRID, Clara Vázquez García was able to detect DNA repair defects in immune cells from mice with over 90% accuracy. Today, her method can detect human immune defects with 99% accuracy, and she continues to advance this work. SWIBRID is highly versatile, as the immune and DNA repair systems are central regulators across various body systems. This makes the technology a strong candidate for contributing to the future of personalized medicine.

“The moment I saw in a plot we could distinguish between DNA repair defects and healthy controls, I realized the potential of SWIBRID, the technology I developed. At that moment, I knew it needed to reach the clinic so that people could benefit from it. Since then, we’ve been focused on identifying the most clinically relevant applications to the technology.”

Lars Grundhöfer

Deutsches Zentrum für Luft- und Raumfahrt (German Aerocspace Center)

Lars Grundhöfer has made significant contributions to the development of alternative navigation systems, particularly for the terrestrial R-Mode systems. R-Mode, short for “Ranging Mode,” is a ground-based alternative navigation system for maritime applications that leverages existing maritime radio infrastructure. It uses medium-wave transmitters to calculate distances between the station and the vessels and thus determine their position – especially as a backup in case of failures in global satellite navigation systems like GPS. Lars Grundhöfer significantly advanced the development of the R-Mode system. Among other things, he developed theoretical foundations for coverage estimation and signal design. He also created a software-defined radio receiver capable of receiving R-Mode signals and obtained the first positon in the open sea.

 “When we determined the position on a ship in the middle of the Baltic Sea, I realized that my approach and implementation worked, and that we could navigate without GNSS*.”

*GNSS stands for Global Navigation Satellite System, an umbrella term for all satellite navigation systems that enable global positioning and navigation.

Laura Helleckes

Forschungszentrum Jülich

How can we develop biotechnological processes faster, more precisely, and more sustainably? This was the central question driving Laura Helleckes throughout her doctoral research. At Forschungszentrum Jülich, she explored how laboratory automation and machine learning can work hand in hand to make biotechnological processes, such as those used to produce enzymes or other proteins, more efficient. By integrating automated laboratories, statistical models, and custom-built software, she not only accelerated experiments but also made them more targeted and strategic. This approach saves time and resources while opening new paths toward a sustainable bioeconomy. Her research brings us one step closer to self-driving bioprocesses – systems that operate intelligently, autonomously, and guided by data.

“In my PhD, I discovered that my training in biotechnology and my work in machine learning allow me to connect two disciplines that rarely speak the same language. This interdisciplinary connection, bringing biotechnology and machine learning closer together, has been my passion ever since.”

Stephan Hilpmann

Helmholtz-Zentrum Dresden-Rossendorf

A deep understanding of microbial processes is crucial for a comprehensive safety analysis for the long-term storage of radioactive waste. The aim of Stephan Hilpmann’s doctoral thesis was to investigate the interactions of the anaerobic, sulfate-reducing bacterium Desulfosporosinus hippei DSM 8344ᵀ with uranium and europium. This bacterial genus has been detected in various environments considered as potential final repository sites. D. hippei DSM 8344ᵀ is capable of immobilizing uranium(VI), which can contribute to the retention of uranium in the near-field environment of a repository. Stephan Hilpmann was the first to identify uranium(V) as an intermediate product of microbial reduction, using synchrotron-based X-ray spectroscopy. His experiments enhance our understanding of biogeochemical processes in complex environmental systems and support the development of improved safety concepts for future radioactive waste repositories.

 “Initially, I wanted to detect the formation of uranium(IV) in the system using relatively simple UV-Vis methods. However, I kept seeing signals that I couldn't identify. It wasn't until I switched to synchrotron spectroscopy that I realized the process and formation of uranium(V) were present, but invisible to conventional methods. Sometimes science just needs the right ‘glasses.’”

Monica Keszler

Forschungszentrum Jülich

With increasing demand for a more sustainable, circular economy, direct recycling offers promising ways to reuse materials without breaking them down through complex chemical processes. In her thesis Monica Keszler focuses on using a specialized sintering technique, known as field assisted sintering technology/spark plasma sintering (FAST/SPS), to directly recycle two distinct material waste streams: hot-deformed Nd-Fe-B magnets and high-speed steel swarf. Nd-Fe-B magnets are key components in wind turbines, e-mobility devices, and consumer electronics. High-speed steel often contains critical elements, such as chromium and tungsten. However, direct recycling methods have not been fully developed for these waste streams due to their specialized microstructure or contaminants. Using FAST/SPS, Monica Keszler successfully created new magnets entirely from Nd-Fe-B scrap with strong magnetic performance. The process was also adapted to turn contaminated steel shavings into new 120 mm cutting disks, showing how industrial waste can be turned back into useful products without complex reprocessing.

 “Often, uneven powder and residuals would damage or break my sintering dies. However, the answer to this problem ended up being quite unusual; I ended up using barista tools and paintbrushes to make sure my powder was level and my dies were clean. Sometimes the best solutions are the most unexpected!”

Marvin Carl May

Karlsruhe Institute of Technology

In wake of an ever increasing complexity the desire to move towards intelligently controlling operations is amplified in manufacturing, most notably in complex semiconductor manufacturing. Here, operational excellence is the key to success. A major concern in controlling complex semiconductor wafer fabrication is the presence of time constraints that limit the transition time of products between two, mostly successive, processes. Adhering to these product specific time constraints is of utmost importance as violations result in scrapping the violating product. To overcome the industrial, manual approach, Marvin Carl May developed as a part of his doctoral thesis a novel, real-time data based approach for intelligently controlling production control for time-constrained complex job shops. He used  real world data both uni-, multi-variate time series models and a digital twin to obtain violation predictions. As a second step, based on the violation probability derived from ML uncertainty, he derived the production control and successfully validated it.

“The moment I realized that industry is already performing very well in this problem, so well that there is a huge bias that renders traditonal approaches not applicable. As a result I shifted my perspective from end-to-end and big data AI to including and quantifying the uncertainty, a model based digital twin and only deriving the decision making at a later stage.”

Vanessa Stenvers

GEOMAR Helmholtz Centre for Ocean Research Kiel

For her doctoral research, Vanessa Stenvers investigated adaptations in pelagic invertebrates, both in the short-term, in response to environmental stress, and in the long-term on evolutionary timescales. Here, she focused on the effects of global warming and deep-sea mining on a pelagic jellyfish. While a growing topic of concern with mining is the effect of discharged sediment in the water column, experimental data was lacking. She found that exposure to plumes comes at high energetic costs, but also has a more severe effect than the most extreme warming scenario. In addition, she revealed that knowledge of symbiotic behaviour is crucial to understanding camouflage and visual adaptations in a group of crustaceans. This knowledge will help to predict ecosystem resilience, as pelagic communities and their interactions are likely to shift under environmental change.

“The deep ocean and its inhabitants do not exist in isolation from life on land. Pelagic animals help regulate our climate, cycle nutrients and sustain fisheries. If and how these animals adapt is a mounting concern and understanding their responses to change is critical to managing a healthy planet.”

 

Hanna Trzesniowski

Helmholtz-Zentrum Berlin für Materialien und Energie

Hydrogen is seen as an important building block for the energy system of the future and is also needed in large quantities as a raw material for the chemical industry. With the help of catalysts hydrogen can be produced by electrolysis of water. During her doctoral research, Hanna Trzesniowski investigated nickel-based electrocatalysts for water splitting. Under alkaline conditions, nickel–iron catalysts have emerged as a promising alternative to rare and expensive materials such as iridium for hydrogen production. A key finding of her research was her elucidation of the electronic structure of nickel-iron oxide catalysts in their catalytically active state. Additionally, she was the first to use spectroscopy to observe processes at the electrochemical interface, i.e., precisely where water splitting occurs. Overall, Hanna Trzesniowski’s work contributes to a deeper understanding of alkaline water electrolysis and paves the way for the development of more efficient and stable catalysts.

“I was sitting at the beamline in the middle of the night, slightly disappointed because at first glance, nothing seemed to be happening at the interface between the catalyst and the electrolyte. But then, a closer analysis revealed something remarkable: electrolyte ions slip between the catalyst layers. And they do so just moments before the catalyst begins its work of splitting water, almost as if the ions want to be right there when things start to get exciting.”

Benedikt Wagner

CISPA Helmholtz Center for Information Security

Benedikt Wagner’s doctoral thesis aims to strengthen the cryptographic foundations of practical digital signature schemes, such as those used in electronic voting systems, blockchains, and distributed systems. His work improves the security and efficiency of these methods, particularly in scenarios involving strong attackers. His work addresses the limitations of existing methods and provides new theoretical and practical tools for trustworthy digital infrastructures.

“Shortly after starting my thesis, I already had my first result, which I was very proud of. But instead of writing it up and uploading it to the ePrint server, my supervisor and I worked on many details and improved it further. After a while, however, we realized that another team had published a similar result in the meantime. From this, I learned to write down results quickly at a certain point and publish them before someone else does.”

Tim Ziegler

Helmholtz-Zentrum Dresden-Rossendorf (HZDR)

In his dissertation, Tim Ziegler investigates optimization methods for laser-driven plasma acceleration of protons—a promising technology for compact, powerful particle sources. Tim Ziegler's work combines precise characterization and control of ultrashort laser pulses with innovative target concepts. The methods developed significantly improved the quality of laser-accelerated proton beams and enabled a new energy record of 150 MeV. The results mark a technological breakthrough and establish robust standards for future plasma experiments with high-intensity laser systems. The work thus makes a significant contribution to the further development of laser-based accelerators – from basic research to applications in medicine and materials science.

“We spent months trying to find the ideal combination of parameters. When the laser pulse was finally fired and our detectors showed a new energy record, there was huge cheering in the control room. At that moment, everyone realized that the concept worked better than we had hoped.”