News

Quantum chemist Werner Dobrautz joins CASUS

The DRESDEN-concept research alliance attracted early-career scientist Werner Dobrautz to establish his first research group in Saxony. Werner is a quantum chemist focusing on complex simulations to answer open research questions in e.g. catalysis or superconductivity. Coming from Chalmers University of Technology in Gothenburg, Sweden, he started his DRESDEN-concept Research Group “AI4Quantum” in December. The group is a joint initiative of the Center for Scalable Data Analytics and Artificial Intelligence (ScaDS.AI Dresden/Leipzig) and the Center for Advanced Systems Understanding (CASUS) at the Helmholtz-Zentrum Dresden-Rossendorf (HZDR).

Welcome to Görlitz, Werner! You have spent the first days in our city and with our team. What are your first impressions?

I have never been in Görlitz before and my first impression of the city is very positive. It seems like a very picturesque little town with a lot of history but at the same time modern amenities. For example, I did not expect to find so many delicious vegan/vegetarian meal options. And, of course, having the winter market directly in front of the office doors at CASUS in the old town city center motivates to get the day’s work done quickly to have a hot punch with the colleagues. Speaking of which, I especially enjoyed the warm welcome I received from everybody there.

Press contact:

Dr. Martin Laqua

Officer Communications, Press and Public Relations Center for Advanced Systems Understanding (CASUS) at HZDR

m.laqua@hzdr.de

We are very happy you decided to come to Saxony to start your first own group. “AI4Quantum” is a DRESDEN-concept Research Group. Could you please tell us more about the positioning of your group?

The DRESDEN-concept Research Groups address innovative research topics and draw on the expertise of two or more partner institutions of the 2010 established alliance. Planning my next career step after my second postdoc phase, I identified ScaDS.AI – one of the six German Competence Centers for Artificial Intelligence – as one of the prime host institutions. I even got more excited when I learned about CASUS and the possibility to apply for a joint position that combines the benefits of both institutions.

That means the artificial intelligence (AI) aspect of ScaDS.AI and the complex systems aspect of CASUS … ?

Exactly. My research aims to understand groundbreaking problems in physics and chemistry, such as bio-catalysis for efficient ammonia production and unconventional superconductivity, with the aid of simulations and computers. In my PhD, I developed and used ‘conventional’ high-performance computing approaches. However, these approaches hit a so-called ‘exponential wall’, as the computational resources grow exponentially with the problem sizes for the really hard ones. This is why, during my two postdocs, I switched into the field of quantum computing, as it promises a potential computational advantage compared to conventional approaches. However, I realized that as it currently stands, both approaches alone can still not tackle the computationally hard problems. Solving such problems in physics and chemistry will require a synergistic approach combining conventional high-performance computing, novel quantum computing, and AI approaches. I believe that the AI expertise of ScaDS.AI combined with CASUS’ focus on understanding complex systems, aided by the interinstitutional DRESDEN-concept research alliance, will boost my research.

As mentioned before, you will now start your own research group. Besides being among the DRESDEN-concept Research Group leaders, you are now also a CASUS Young Investigator. How does that feel?

Honestly, a bit overwhelming! This is a huge opportunity, but of course also a great responsibility. ScaDS.AI will provide me with funds to hire two PhD students, and CASUS will provide the funds to finance another two researcher positions, either PhD students or postdoctoral researchers. After positive evaluation after three years, there is the potential for additional funding for another 3 years. Benefitting from the experiences of other junior group leaders both at ScaDS.AI and CASUS would certainly be a great thing.

You have been a European Postdoctoral Fellow at Chalmers University in Gothenburg, Sweden, financed through a Marie Skłodowska-Curie Action before. What are the experiences you gained at Chalmers?

The Marie Skłodowska-Curie call is very competitive because the funding is very generous and flexible, especially concerning travel and networking-related funds. I chose to move from Stuttgart, where I did my PhD, to Gothenburg specifically because Chalmers and Prof. Martin Rahm were the perfect hosts for my proposed research project QC-SQUARED that aimed to develop novel quantum computing approaches to simulate complex quantum chemical problems. Chalmers coordinates the ‘Wallenberg Centre for Quantum Technology’, a 12-year, 86 million euro research program that aims to take Swedish research and industry to the forefront of quantum technology by developing and building superconducting-qubit quantum devices. Consequently, Chalmers is a worldwide hotspot for quantum technology that is heavily involved in the European Quantum Technology Flagship projects OpenSuperQ and OpenSuperQPlus, which aim to build a versatile 1,000-qubit quantum-computing system made in Europe as well as the NordIQuEst project that aims to build a Nordic-Estonian e-infrastructure combining high-performance computing and quantum computing from across Norway, Sweden, Denmark, Finland and Estonia. So overall, my experiences have been really great and I could achieve all that I hoped for. As I will stay involved with the research efforts at Chalmers as well as WACQT, OpenSuperQPlus and NordIQuEst, ScaDS.AI and CASUS will now have another access to a quantum computing system.

You have briefly mentioned your scientific interests. Could you tell us something more about the research you did in the past??

During my PhD studies, I was able to develop and implement two major novel contributions to an established computational method (full configuration interaction quantum Monte Carlo – FCIQMC), that extended its applicability to a wider and more complicated set of problems. Specifically, I implemented FCIQMC in a spin-adapted basis and enabled the use of an explicitly correlated method (transcorrelation). Both those novelties were highly sought-after, provided real utility for relevant quantum chemical problems and are still both actively developed and used.

The spin-adapted implementation has, among others, allowed the accurate study of iron-sulfur clusters that are relevant to enzymatic nitrogen fixation. Transcorrelation, on the other hand, has severely reduced the necessary computational resources to perform highly accurate calculations. Thus, larger and more relevant quantum mechanical problems can be studied.

Being a postdoctoral researcher at Chalmers, I am proud to have extended this transcorrelation method to quantum computing algorithms, enabling one of the first accurate calculations – within so-called chemical accuracy – of quantum chemical systems despite the current noisy quantum hardware.

And what are the projects you have in mind for your time in Saxony?

My long-term goal is to develop efficient computational methods and algorithms to help solve and understand fundamental problems in chemistry and physics, especially those relevant to the green energy transition. To achieve these goals, I focus on studying nano-scale quantum mechanical systems with strong electronic correlations that are responsible for groundbreaking phenomena like high-temperature superconductivity, nitrogen fixation, and artificial photosynthesis. To understand these processes, I aim to develop a synergistic computational toolset that combines high-performance computing, quantum computing, artificial intelligence, and deep machine learning approaches.

Press contact:

Dr. Martin Laqua

Officer Communications, Press and Public Relations Center for Advanced Systems Understanding (CASUS) at HZDR

m.laqua@hzdr.de