How does a synthetic receptor selectively recognise a guest molecule? How does a responsive polymer change its structure in response to temperature, pH or charge state? From individual binding sites to collective self-organisation in micelles and microgels, we address these questions using various approaches (QM, MM, SQM; MC and MD) with established software packages and software developed in-house. The simulation protocol itself is also the subject of development: we are increasingly automating screening workflows, conformer searches and the analysis of large data sets using agent-based, LLM-supported methods, applying the same standards of traceability and human verification as we do to any other computational method.
Supramolecular recognition
How accurate must a calculation or the chosen method be in order to make a reliable statement about binding affinity or selectivity, and how long are we allowed to wait for this? Artificial and natural receptors bind their guest molecules through a combination of many weak interactions (van der Waals interactions, hydrogen bonds), none of which is decisive on its own. Even for clearly defined, low-molecular-weight systems, the calculations are computationally intensive. We therefore systematically benchmark fast methods against expensive reference calculations (DLPNO-CCSD(T)) and seek the point at which additional computation time no longer yields any further insight.
Macromolecular Chemistry
At what point does a single polymer chain behave differently from the collective formed by thousands of them? Above a critical concentration, chains assemble into ordered aggregates, and the way in which they accommodate a guest molecule depends not on a single binding site, but on the architecture of the entire chain and the external conditions: temperature, pH and charge state. Monte Carlo and molecular dynamics simulations link this question to the structural investigations carried out by the Plamper research group at the atomic level.
Methods and software
The range of methods extends from established quantum chemistry (DFT, WFT, semi-empirical methods such as GFN2/GFN1-xTB) through force fields (GFN-FF) to modern sampling approaches: metadynamics and its variants for conformer search, and Monte Carlo simulations for the self-organisation of entire chains. For productive QM calculations, we rely on established external programmes, primarily ORCA and CP2K, and integrate them into automated workflows using our own tools. For GFN calculations, we rely on our own implementation in curcuma. Machine learning is increasingly being incorporated, particularly as an explainable, uncertainty-aware model for the small, heterogeneous datasets typical of this branch of chemistry.
Software
curcuma – Quantum chemistry, force fields and conformer search in a C++ tool. Native GFN2/GFN1-xTB and GFN-FF implementations with GPU acceleration via CUDA and ROCm.
qurcuma – Qt-based visualisation and interaction for curcuma, including the basis for molecular visualisation in VirtMolSim.
SupraFit – Determination of binding constants from titration experiments (NMR, ITC), statistically validated. Global and local fitting for any binding models (1:1, mixed models, dimerisation). Monte Carlo and F-test-based confidence intervals, cross-validation and reduction analysis check whether a model supports the data at all, not just how well it fits.
ALIMA – RAG-based system for AI-assisted, source-verified text analysis: A language model selects only from actual, verified matches found, rather than inventing them. Developed for subject indexing at the Freiberg University Library, as an example of agent-based workflows.
Projects
- VirtMolSim
VirtMolSim is an educational project that allows users to experience molecules in real time with haptic feedback. Designed for use on consumer VR hardware as well as in the Mixed Reality Space at the University Library. Funded by the Foundation for Innovation in Higher Education Teaching (Freiraum 2026, 04/2026–03/2028), in collaboration with the “Georgius Agricola” University Library.
- EdUp (2025)
- TWINS project on collaborative teaching DE-UA
Teaching collaboration between Dnipro University of Technology and TUBAFdigital
Chemistry Summer Camp
- At the Chemistry Summer Camp, pupils can gain an insight into studying chemistry at TUBAF
A brief biography of Conrad Hübler
- 2025 Participation in the Saxon Library Prize for “ALIMA – Train-the-Trainer”
- since 2023 Postdoctoral researcher, Plamper Research Group (Physical Chemistry), TU Bergakademie Freiberg
- since 2022 Subject Specialist for Natural Sciences, University Library of TU Bergakademie Freiberg
- 2021–2022 Research Assistant, Straßner Research Group (Physical Organic Chemistry), TU Dresden
- 2021 PhD (Dr. rer. nat., summa cum laude), Freiberg University of Mining and Technology
- 2016 Georgius Agricola Medal, Freiberg University of Mining and Technology: awarded for “outstanding academic achievement during the course of study”
- 2015–2020 Research Assistant, Mazik’s Research Group (Organic Chemistry), TU Bergakademie Freiberg
- 2015 State postgraduate scholarship
Courses
- Molecular Modelling and Quantum Chemistry (lecture and seminar), annually in the summer term
- Theoretical Concepts of Molecular and Electronic Structure of Chemical Compounds (lecture), annually in the winter semester
- Literature Research and Artificial Intelligence (lecture), Module: Scientific Information Media, annually in the winter semester
- Microcredential: Modelling Interactions of High Molecular Weight Compounds, in collaboration with Dnipro University of Technology and TUBAFdigital (2025)
The Team
- Dr rer. nat. Conrad Hübler
- M. Sc. Lena Bussmann Fuentes
- B. Sc. Gerd Gehrisch
- Anja Voigt
Theses
Topics for theses (Bachelor’s and Master’s theses, as well as project and course papers, etc., for students of Chemistry and Applied Natural Sciences) are available on a variety of subjects. For example
- Analysis of biologically relevant supramolecular systems using various theoretical methods
- Parameterisation of coarse-graining potentials for polymers
- Further development of the GFN-FF for chemical reactions
- Software development and machine learning support in SupraFit
Ongoing work:
- Gerd Gehrisch (Master’s thesis): Machine learning prediction model for complex formation constants of artificial carbohydrate receptors
- Anja Voigt (Diploma thesis): Monte Carlo simulation of the folding of diblock copolymers (in collaboration with Felix Plamper)
Completed theses:
- Gerd Gehrisch (Bachelor’s thesis, 2024): Calculation of dipole moments from partial charges in curcuma
- Anja Voigt (student research project, 2024): Monte Carlo simulation of the folding of diblock copolymers (in collaboration with Felix Plamper)
- Wiebke Hadwich (project work, 2023): Monte Carlo simulation of the conformation and flexibility of a polymer chain (in collaboration with Felix Plamper)
- Eric Franke (undergraduate project, 2020): Methods for the analysis of supramolecular interactions (at the Institute of Organic Chemistry)
Publications (selection)
- Hübler, C. "ALIMA – A RAG-based system for LLM-supported subject indexing: prototype development and initial practical experience." Library Research and Practice 2026, 2, pp. 197–216. 10.1515/bfp-2026-0014
- Voigt, A.; Strauch, C.; Höfken, T.; Wahab, M.; Hadwich, W.; Barner-Kowollik, C.; Hübler, C.; Schneider, S.; Plamper, F. A. “Following the Formation of Single-Chain Nanoparticles Generated by Interblock Crosslinking within Diblock Copolymers”, Soft Matter 2026, 22, 3393–3402. 10.1039/D5SM01153A
- Neubert, R.; Schumann, E.; Schildknecht, V.; Lehmann, A.; Rosenfeldt, S.; Hübler, C.; Plamper, F. A. “Structural Changes of Flexible, Small-Sized Polycationic Microgels Upon Interaction With Hexacyanoferrate(III)”, Macromolecular Chemistry and Physics 2026, 227(4), e00493. 10.1002/macp.202500493
- Fuhrmann, B.; Hübler, C.; Mazik, M. “Use of Halogenated Units for the Construction of Artificial Carbohydrate Receptors”, Molecules 2026, 31(8), 1237. 10.3390/molecules31081237
- Hübler, C. “Analysing binding stoichiometries in NMR titration experiments using Monte Carlo simulation and resampling techniques”, PeerJ Analytical Chemistry 2022, 4, e23. 10.7717/peerj-achem.23
- Hübler, C. “SupraFit – An Open Source Qt-Based Fitting Application to Determine Stability Constants from Titration Experiments”, Chemistry Methods 2022. 10.1002/cmtd.202200006
- Köhler, L.; Hübler, C.; Seichter, W.; Mazik, M. “Binding modes of methyl α-D-glucopyranoside to an artificial receptor in crystalline complexes”, RSC Advances 2021, 11(36), 22221–22229. 10.1039/d1ra03390e
