“We have to move forward step by step”
Developing quantum algorithms for chemical research
Initially not overly enthusiastic about his physics studies, Ludwig Nützel eventually discovered his fondness for research. Today, he develops algorithms for quantum computing. Never stuck on one thing, but always having a wide range of interests and curiosity about all the funny opportunities that everyday life has to offer, a fun idea can turn into a side career as a ‘QuantenSchafkopf’ expert.
By Maria Poxleitner
A group of young scientist huddles around two beer tables. Some are wearing Indian saris, others Japanese kimonos, still others a caftan. It is the traditional Bavarian evening at the Nobel Laureate Meeting in Lindau at Lake Constance, to which the young guests from all over the world are invited to dress in their local costumes. Among them sits Ludwig Nützel in Lederhosen and a blue vest, explaining QuantenSchafkopf. The 25-year-old physicist has added new rules to the traditional Bavarian card game “Schafkopf” that take up phenomena of quantum physics. And because this fits so perfectly into a Bavarian evening of a conference focusing on quantum physics this year, he was invited by the Bavarian State Ministry of Science and the Arts to present his game in the supporting program – public premiere for QuantenSchafkopf.
“I was pleasantly surprised by the level of interest in the game,” Ludwig recalls the evening in Lindau. The biggest hurdle was that people had to understand the not-so-simple classical rules of Schafkopf. The quantum variant then results from a few simple additional rules, says its inventor. For example, in QuantenSchafkopf, it is allowed to play two cards at the same time, which takes up the phenomenon of superposition. The additional rules allow for new tactics, but Ludwig doesn't claim that the quantum variant is more fun than the traditional game: “For me, it's mainly a form of public outreach. The game makes the basic concepts of quantum mechanics tangible in a fun way.”
From a fun idea to QuantenSchafkopf
For this reason, the physicist is now also working on a project on how QuantenSchafkopf could be used in physics lessons at school. Furthermore, he was recently invited to the first “Weltschafkopftag” to present his extension. But he never intended for it to become "big", says Ludwig. “There are things I find funny, and then I just do them.” A few years ago, he also found it funny to think about how to describe a trick in Schafkopf mathematically. Not that such a mathematical description would have been necessary – “There's no point, it was just a weird consideration,” says Ludwig and laughs. This initial consideration had nothing to do with quantum physics, but later inspired his idea of QuantenSchafkopf. He finds it funny that so many people are now interested in what was originally just a joke.
Ludwig Nützel, 25
Position
Ph.D. student
Institute
FAU – Chair for Quantum Theory
THEQUCO
Degree
Physics
Ludwig develops algorithms for quantum computing and tries to make them as efficient and error-resistant as possible.
Finding things funny – that doesn't seem to be difficult for Ludwig. The regular squirrel visits to the office window, for example, he finds very amusing. The “squirrel mix” that he and his colleagues keep ready in the office looks dangerously similar to trail mix. The 25-year-old also has to laugh when he thinks back to a 100-kilometer “walk” he and a colleague took from Erlangen to Würzburg. After walking for hours in the dark through overgrown and unrecognizable forest paths in the rain, they had the misfortune of another heavy downpour during the day: “I've never seen anything like it, it was a real deluge,” Ludwig remembers, laughing. The last hour was quite a torture. Now he is curious to see what a 100-kilometer walk will be like in better conditions. A second time is therefore already planned.
“With our algorithm, we want to predict how a molecule will behave in a chemical reaction”
Not necessarily 100 kilometers, but going for a walk also helps Ludwig in his daily research: “When I realize that I can't find a solution, I go outside. Walking often gives me ideas.” He works at the Institute for Quantum Theory at the Friedrich-Alexander-Universität in Erlangen. As part of his Ph.D., he is researching algorithms that can be used to solve problems on quantum computers, and is trying to make these algorithms as error-resistant and efficient as possible. His latest publication describes an algorithm for chemical research, one of the most promising application areas for quantum computing. “With our algorithm, we want to predict how a molecule will behave during a chemical reaction,” explains the doctoral student. The publication specifically examines a metal chelate, which consists of a metal ion and a chelating agent. Chelating agents play a role in many areas, for example in households as water softeners or in agriculture as fertilizers. In the chemical industry, quantum computers could help make the development of new or better compounds more efficient in the future, explains the physicist: “Before expensively producing and testing compounds in the laboratory, you can simulate the behavior.”
A crucial step to be able to predict the reaction behavior of a molecule is determing its ground state, that is, the state in which the molecule has the lowest energy. This can be done using the so-called “Variational Quantum Eigensolver (VQE)”, an algorithm that can run on already existing hardware because it works on NISQ computers and does not require an error-corrected quantum computer. “You have a quantum circuit, the result of which describes the quantum state of the molecule,” Ludwig begins to outline how the algorithm works. You then try to adjust parameters in the circuit in such a way that when you run the circuit again, you get a state with lower energy, he explains further. “And you go through this loop again and again to get the energy as low as possible.” This search for the best possible parameters is also referred to as training the algorithm. However, this is precisely where the major weakness of the VQE lies, Ludwig points out: “There is a huge problem with the trainability.” If you change the parameters, the energy changes only minimally and the larger and more complex the molecule, the more the parameter landscape flattens out, the doctoral student describes the problem. You can imagine it a bit like trying to hike the deepest valley in a hilly landscape, but incline and decline are so small that you can't feel whether you're going up or down.
Ludwig and his colleagues also had a problem with the trainability, but they found a way around it. The twist was not simply to apply the standard VQE, explains the physicist, but to cleverly replace part of it with classical methods. Only with this adaptation was it possible to accurately determine the ground state energy of the metal chelate – and thus of a molecule that is significantly more complex than other molecules whose behavior had previously been simulated on quantum computers. This was a great success, but the doctoral student emphasizes: “I'm not saying that our algorithm scales.” Difficulties will continue to arise with larger or more complex compounds. But perhaps another adaptation or a new algorithm will be found, says Ludwig: “We have to move forward step by step.”
On a typical university day, Ludwig is always in the office by seven in the morning. This allows him to finish work in the late afternoon, if meetings and the like permit. Having enough time for hobbies, friends and family is important to him. The tuba takes up a large part of his free time. Especially in the warmer months, when the folk festival season begins, he is much in demand as a tuba player in a traditional brass band: “In the summer, my tuba is always in the trunk of my car.” It's not worth lugging the massive instrument back to the apartment between rehearsals and the many gigs, Ludwig says. Unexpectedly, he came to the tuba not in his Franconian homeland, but in California. During his time at a high school near San Francisco, there were plenty of trumpets in the school band: “So I just switched to tuba.” After his stay abroad, he used his first summer job to buy his own tuba, which has accompanied him ever since.
The bachelor's thesis sparks his enthusiasm for research
Studying physics was not Ludwig's plan from the beginning. After he had to abandon his original plan to become a pilot for various reasons, physics became his Plan B. In the first few semesters, the course did not really arouse his enthusiasm. He seriously considered doing something else: “I thought about becoming an instrument maker or a brewer, for example.” But with only a year to go before graduation, he decided to see it through. In retrospect, it was the right decision, because: “Then came the bachelor's thesis, and that was something completely different.” Instead of lectures and tutorials, you do real research, Ludwig says. “Maybe you discover something that nobody else has discovered before you – and that's how it was, and it was really cool.”
After completing his master's degree in Würzburg, Ludwig decided to return to Erlangen for his doctorate: “My current field is particularly well represented there.” He also appreciates the good networking that comes with being part of Munich Quantum Valley. This ultimately led to the collaboration with researchers from Innsbruck, on whose ion trap quantum computer they were able to experimentally implement their algorithm. “Our algorithm can basically run on any hardware platform,” adds Ludwig, “but we take advantage of the fact that each qubit can interact directly with every other qubit.” Ion trap hardware, which has this ‘all-to-all connectivity’, was therefore better suited than, for example, superconducting hardware.
He really likes the free scientific work and the topics at his chair. Although he still sometimes thinks he would have liked to do an apprenticeship and learn a trade, looking back he thinks that everything worked out perfectly: “I think it's cool to explore the unknown. I'm happy now, the way it is.”
Published 31 January 2025; Interview 26 November 2024