100 years of quantum chemistry: A personal perspective 

Prof. Frank Neese reflects on past and future of quantum chemistry on occasion of the United Nations’ International Year of Quantum Science and Technology

Prof. Frank Neese reflects on quantum chemistry's past and future, emphasizing computational power, challenges, and AI's transformative potential.

What has quantum chemistry achieved in nearly a century, and where is it headed? Frank Neese, Managing Director at the Max-Planck-Institut für Kohlenforschung, shares his personal reflections in a perspective article published to mark the United Nations’ International Year of Quantum Science and Technology.

In 1929, Paul Dirac, one of the founders of quantum mechanics, made a bold prediction: the fundamental laws of chemistry were already known, but solving them exactly would lead to equations too complicated to handle. In this view, chemistry’s questions could in principle all be answered once enough numerical precision and computing power became available. In some ways, this prophecy has come true. Over the decades, quantum chemistry has grown from simple models of chemical bonds into highly sophisticated algorithms that can now calculate systems with thousands of atoms.

Yet Neese emphasizes that sheer computational power is not the whole story. “In 2013 it became possible to calculate an entire protein with hundreds and even thousands of atoms using quantum chemical methods. This was a major milestone,” he notes, “but it did not teach us anything new about the protein itself.” In real chemistry, solvents, entropy, and the flexible shapes of molecules are just as important as electronic energies and often harder to capture. Thus, dealing with conformational complexity is one of the grand challenges in the field now that the problem of how to solve the Schrödinger equation for a large-molecules has, to a good approximation, been solved. 

Equally central is the balance between numbers and insight. Accurate predictions are valuable, but they cannot replace understanding. Simplified models, even if imperfect, remain vital because they give chemists a language to think creatively, design experiments, and develop new ideas.

The future of quantum chemistry in the age of artificial intelligence

Artificial intelligence and machine learning are advancing quickly and have the potential to reshape the field. While they cannot yet, or maybe ever, fully replace physics-based methods, they are developing at remarkable speed. One limitation is that machine learning models often perform well within the range of their training data but become unreliable outside it. For this reason, Neese foresees a mixed landscape where traditional and AI-based approaches coexist and enforce each other. In the area of code generation, however, the role of AI could become even more transformative. It is conceivable, Neese suggests, that future quantum chemistry programs may one day be designed and optimized entirely by AI, fundamentally changing how scientific software is developed.

“All these new methods and powerful tools are extremely exciting but we must never forget the scientific basis of what we are doing,” Neese reflects. “We must remain critical, question results, and be aware of limitations. With this in mind, theoretical chemistry will continue to be a vibrant and rapidly evolving field that is indispensable for modern science.”