RNRasmus Nilsson

Rasmus Nilsson

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Rasmus Nilsson has a total of 21 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Does the Si/Al-order in K-feldspars influence their hydration structure?

    by Franziska Sabath, Rasmus V. E. Nilsson, Tobias Dickbreder, Hans-Georg Stammler, Ralf Bechstein, Angelika Kühnle, Bernhard Reischl

    No significant impact of Si/Al-order on hydration layer structure of K-feldspars is observed in AFM experiments, but atomistic MD simulations indicate differences in hydration layer dynamics.

  • Open Access

    Ice nucleation on microcline (001) in the absence of active sites

    by Florian Schneider, Rasmus Väinö Erik Nilsson, Ralf Bechstein, Hans-Georg Stammler, Bernhard Reischl, Thomas Koop, Angelika Kühnle, Tobias Dickbreder

    Of the many different ice-nucleating mineral particles, dust from feldspar mineral microcline stands out as being particularly active. However, why microcline outperforms other particles remains a puzzle. Here, we use nanoscale imaging to show that the (001) surface of microcline can induce ice nucleation even in the absence of active sites. In contrast, for the closely related feldspar sanidine, ice nucleation is dominant at step edges as expected. Atomistic simulations suggest that the ice nucleation is induced by a lattice match between the higher-index (10.4) plane of hexagonal ice and microcline. Our findings provide a nanoscopic explanation for microcline’s exceptional ice nucleation ability and demonstrate the importance of taking higher-index surfaces into account.


  • Open Access

    Benchmarking Ice-Water Equilibria Exhibited by Foundation Neural Network Potentials

    by Rasmus Nilsson, Golnaz Roudsari, Mária Lbadaoui-Darvas, Bernhard Reischl, Stephen Ingram

    Herein we report the melting points of ice exhibited by some recently published foundation Neural Network Potentials (NNPs) using the direct coexistence method: SO3LR, Orb v3, MACE-MP, FENNIX-Bio1, NEP89 and PET-MAD as well as the first generation model ANI-2x. Using our previously published LICH-TEST algorithm, we classify the structural motifs adopted by individual molecules over time, finding that most models correctly exhibit ice growth and shrinkage via interfacial hexagonal structures. However, the observed melting temperatures vary from the experimental value significantly, with the lowest and highest nearly 150 K apart. The growth rates of ice below the melting points were also found to vary significantly, in some cases exceeding classical water models by one decade. SO3LR was the only NNP exhibiting an accurate value, and represents the best trade-off of speed and accuracy for the simulation of ice nucleation in (bio)organic systems. Disconcertingly the MACE, Orb and ANI models overestimate the melting point to such a degree that liquid water is effectively under deep supercooling when simulated at standard conditions. By comparing variants of the latter two potentials, we infer that an accurate description of dispersion interactions during training and/or evaluation improves the water density isobar and leads to slightly better melting temperatures, albeit with a substantial speed penalty if added during inference. We conclude with some general recommendations for training the next generation of foundation models, in order to improve their description of ice-water interactions.

  • Open Access

    Ice nucleation on microcline (001) in the absence of active sites

    by Florian Schneider, Rasmus Nilsson, Ralf Bechstein, Hans-Georg Stammler, Bernhard Reischl, Thomas Koop, Angelika Kühnle, Tobias Dickbreder

    Heterogeneous ice nucleation is at the heart of a wide range of technological and natural fields ranging from cryopreservation 1 and the development of anti-icing coatings 2–4 to the glaciation of clouds 5–11 . The latter process has a profound impact on Earth’s climate as it changes major physical properties, such as the albedo and precipitation efficiency 6,7 . Of the many different ice-nucleating mineral particles, dust from feldspar mineral microcline stands out as being particularly active 11–17 . However, why microcline outperforms other particles remains a puzzle. Here, we use nanoscale imaging to show that the (001) surface of microcline is able to induce ice nucleation in the absence of active sites. In contrast, for the closely related feldspar sanidine, ice nucleation is dominant at step edges as expected. Atomistic simulations suggest that the ice nucleation is induced by a lattice match between the higher index (10.4) plane of hexagonal ice and microcline. Our findings provide a nanoscopic explanation for microcline’s exceptional ice nucleation ability and demonstrate the importance of taking higher-index surfaces into account. We expect that the latter insight will enhance our understanding of nucleation processes in nature and our ability to create materials with outstanding ice nucleating abilities.

  • Open Access

    Nanoscopic insights on ice nucleation on feldspar microcline and sanidine

    by Bernhard Reischl, Florian Schneider, Rasmus Nilsson, Ralf Bechstein, Thomas Koop, Angelika Kühnle, Tobias Dickbreder

    Ice and mixed-phase clouds can form at moderate supercooling on seed particles through heterogeneous ice nucleation. Feldspar particles constitute a significant fraction of mineral dust in the atmopshere and have been identified as good ice nucleating particles. However, they can exhibit different chemical composition and crystal structure, which affects their ice nucleation activity [1], and the details of the ice nucleation mechanism(s) remain unknown.Here, we present atomic force microscopy images of ice crystals growing from the vapor phase at low temperature on the (001) surfaces of two types of feldspar crystals: the highly ordered microcline, and the disordered sanidine. In contrast to the prevailing view of active sites such as step edges or cracks being responsible for ice nucleation on feldspar [2], we observe ice growth at random positions on the bare terrace of feldspar microcline (001). For the closely related feldspar sanidine, ice nucleation is only observed at step edges, as previously reported.Our observations underscore the exceptional ice nucleating ability of microcline as it demonstrates ice nucleation even in the absence of surface defects and raise important questions regarding the different ice nucleation mechanisms on these two feldspar mineral surfaces, which are investigated using atomistic simulations.[1] Harrison, A. D., et al., Atmos. Chem. Phys., 16, 10927-10940, 2016.[2] Kiselev, A., et al., Science, 355, 367-371, 2017.

  • Open Access

    Using New Generation Neural Network Potentials to Benchmark Ice-Water Equilibria

    by Rasmus Nilsson, Golnaz Roudsari, Mária Lbadaoui-Darvas, Bernhard Reischl, Stephen Ingram

    Ice and mixed phase clouds in the earth's atmosphere form predominantly through heterogeneous nucleation on seed particles, such as mineral dust and organics. Determining the atomistic ice-nucleation mechanism on these particles is challenging for experiments and simulations. When simulating ice nucleation using Molecular Dynamics (MD), one typically relies on classical empirical potentials (force fields) to describe interactions between atoms in the particle surface and water. However, due to the large number of different materials ice-nucleating particles can consist of, accurate classical empiric potentials are not available for all systems, leading to heavy computational costs for creating and testing new ones. In recent years, foundation neural network potentials (NNPs), trained on large sets of quantum chemical data, aim to enable simulations of any system, thus circumventing the issue of creating new potentials. These NNPs would ideally combine accuracies of Density Functional Theory (DFT) with simulation speeds of classical MD. To determine the viability of using foundation models in MD simulations of heterogenous ice nucleation, we have benchmarked the ice-water equilibria of four NNPs: SO3LR (Kabylda et al. 2025), Orb v3 (Rhodes et al. 2025), Fennix-Bio1 (Plé et al. 2025) and ANI-2x (Devereux et al. 2020). We determined their melting points and, where not available in the literature, the water density isobars they exhibit in the temperature range 250-300 K. We have used the coexistence method: A system initially containing hexagonal ice and liquid water is simulated in the NPT ensemble, and the melting point is determined as the temperature at which the number of ice-like water molecules (counted using the classification algorithm LICH-TEST) does not change over time. The SO3LR potential was the only one of the four displaying a melting point close to 273 K. Fennix-Bio1 underestimated the melting point by 20 K, while both Orb v3 and ANI-2x overestimated it by over 75 K. By comparing variants of the latter two models, we can infer that inclusion of dispersion interactions during either training or inference improves the water density isobar, which in turn leads to a more accurate melting point. In addition, we find that while the NNPs are in theory reactive models, no Grotthus-like proton transfers were observed in the simulations. Kabylda et al.: Molecular Simulations with a Pretrained Neural Network and Universal Pairwise Force Fields, ChemRxiv, 2025. Rhodes et al.: Orb-v3: atomistic simulation at scale, https://arxiv.org/abs/2504.06231, 2025.Plé et al.: A Foundation Model for Accurate Atomistic Simulations in Drug Design, ChemRxiv, 2025. Devereux et al.: Extending the Applicability of the ANI Deep Learning Molecular Potential to Sulfur and Halogens, Journal of Chemical Theory and Computation, 16, 4192–4202, 2020.

  • Open Access

    Hydration layer structure and ice nucleation ability of K-feldspar surfaces investigated using molecular dynamics and machine learning

    by Rasmus Nilsson, Patrick Rinke, Hanna Vehkamäki, Bernhard Reischl

    K-feldspar mineral dust particles have been observed to nucleate ice heterogeneously at higher supercooling compared to other atmospheric minerals. There is experimental and computational evidence pointing to the importance of high energy (100) crystallographic planes, mostly exposed in surface cracks, but the exact atomistic ice nucleation mechanism remains unknown (Kiselev et al., 2017). Recent atomistic molecular dynamics simulations did not exhibit spontaneous ice nucleation on flat (100) K-feldspar microcline surfaces (Soni and Patey, 2019). This could have been caused by inaccurate force fields, insufficient sampling time, or considering too simple surfaces that do not present active sites for ice nucleation. We try to shed new light on the phenomenon by combining molecular dynamic simulations with machine learning models. To validate the force field used in our simulations, we compare the calculated hydration layer structures with recent 3D AFM experiments at the feldspar-water interface and find good agreement between the two (Dickbreder et al., 2024). Using non-equilibrium molecular dynamics, we determine the onset freezing temperature on a large sample of K-feldspar (100) surfaces with different termination and topographical features, such as step edges, defects, and strained lattices, by looking for a potential energy jump as the simulation temperature is decreased. Machine learning models are then trained on this data set to predict the onset freezing temperature based on the characteristics of the surface, and by using feature analysis we will determine which surface characteristics enable higher onset freezing temperatures (Fitzner et al., 2020). Dickbreder, T., Sabath, F., Reischl, B., Nilsson, R. V. E., Foster, A. S., Bechstein, R., and Kühnle, A.: Atomic structure and water arrangement on K-feldspar microcline (001), Nanoscale, 16, 3462-3473, 2024. Fitzner, M., Pedevilla, P., and Michaelides, A.: Predicting heterogeneous ice nucleation with a data-driven approach, Nat. Commun., 1-9, 2020. Kiselev, A., Bachmann, F., Pedevilla, P., Cox, S. J., Michaelides, A., Gerthsen, D., and Leisner, T.: Active sites in heterogeneous ice nucleation-the example of K-rich feldspars, Science, 355, 367-371, 2017. Soni, A. and Patey, G. N.: Simulations of water structure and the possibility of ice nucleation on selected crystal planes of K-feldspar, J. Chem. Phys., 150, 214501, 2019.

  • Open Access

    Searching for the atomic scale mechanism of ice nucleating particles: hydration layer structures on K-Feldspar microcline surfaces from a combination of atomistic simulation and atomic force microscopy

    by Bernhard Reischl, Rasmus Nilsson, Adam Foster, Franziska Sabath, Tobias Dickbreder, Ralf Bechstein, Angelika Kühnle

    Ice and mixed-phase clouds can form at moderate supercooling on seed particles through heterogeneous ice nucleation, but despite numerous experimental and computational investigations, understanding heterogeneous ice nucleation remains one of the great challenges in atmospheric science. While feldspar mineral dust particles have been identified as particularly good ice nucleating particles, they can exhibit different chemical composition and crystal structure, making it difficult to determine the atomistic details of the ice nucleation mechanism, both experimentally, and computationally. Here, we present systematic atomistic molecular dynamics studies of hydration layer structures at the interfaces of K-feldspar maximum microcline (001), (010), and (100) surfaces and water, at room temperature and moderate supercooling. Simulations on the fully hydroxylated α-terminated (001) cleavage plane reveal a complex lateral structure in the first water layer and a less ordered second layer. At room temperature, water exchange within the first hydration layer and between the first and second hydration layers occurs on a sub-nanosecond timescale. We also observe that surface potassium ions can go into solution and return to vacant surface sites on a timescale of tens of nanoseconds, but this causes surprisingly minor perturbations within the first hydration layer if the sampling time is sufficient. Hydration layer structures from simulation are in very good agreement with 3D atomic force microscopy data recently obtained for the first time on a freshly cleaved microcline surface in pure water (Dickbreder et al., 2024) – validating the accuracy of the atomistic model and providing an interpretation of the experimental data. However, the simulated hydration layer structures on the low energy (001) or (010) surfaces do not exhibit a lattice match with faces of cubic or hexagonal ice. Only the higher energy (100) surface with slightly strained lattice parameters can stabilize an ice interface at moderate supercooling in the simulations. Our results confirm previous findings (Kiselev et al., 2017; Soni and Patey, 2019) and indicate that the good ice nucleating properties of feldspars likely result from more complex active sites, possibly involving changes in surface chemistry, or topographic features such as defects, strained lattices, or step edges, which we are currently investigating.Dickbreder, T., Sabath, F., Reischl, B., Nilsson, R. V. E., Foster, A., Bechstein, R. and Kühnle, A.: Atomic structure and water arrangement on K-feldspar microcline (001), accepted in Nanoscale, DOI:10.1039/d3nr05585j, 2024.Kiselev, A., Bachmann, F., Pedevilla, P., Cox, S. J., Michaelides, A., Gerthsen, D., and Leisner, T.: Active sites in heterogeneous ice nucleation—the example of K-rich feldspars, Science, 355, 367–371, 2017.Soni, A. and Patey, G. N.: Simulations of water structure and the possibility of ice nucleation on selected crystal planes of K-feldspar, J. Chem. Phys., 150, 214501, 2019.

  • Open Access

    Atomic structure of pristine and water-covered microcline (001) – A prerequisite for understanding the ice nucleation mechanism on feldspar mineral dust particles

    by Tobias Dickbreder, Franziska Sabath, Bernhard Reischl, Rasmus V. E. Nilsson, Adam S. Foster, Ralf Bechstein, Angelika Kühnle

    The aggregate state of water in clouds has a fundamental impact on the clouds’ properties such as reflectivity and lifetime. Consequently, it is crucial for the development and improvement of climate models to understand the mechanism of ice nucleation under atmospheric conditions. Most atmospheric ice nucleation is heterogeneous caused by the interaction between water droplets and ice nucleating particles. Under mixed-phase cloud conditions, one of the most important ice nucleating particles are feldspar minerals. Recent scanning electron microscopy studies have shown that ice nucleation on cleavage planes of K-rich feldspars predominantly takes place at step edges and pores (Kiselev, 2017). This has also been confirmed by video and atomic force microscopy on the micrometer scale (Holden, 2019). However, experimental insights into the atomic-scale structure of the most ice-nucleation active K-feldspar microcline are still missing, and, thus, the mechanism behind ice nucleation on feldspar minerals remains elusive. Here, we present high-resolution atomic force microscopy (AFM) data revealing the atomic structure of the microcline (001) surface in its pristine state and in contact with water (Dickbreder, 2024). AFM images of the pristine microcline (001) surface kept under ultrahigh-vacuum conditions, reveal features consistent with a hydroxyl-terminated surface. This finding suggests that water in the residual gas readily reacts with the surface highlighting the high reactivity of the as-cleaved surface. Indeed, corresponding density functional theory calculations confirm a dissociative water adsorption. Three-dimensional AFM measurements performed at the mineral-water interface unravel a layered hydration structure with two features per surface unit cell. Comparison with MD calculations suggest that the structure observed in AFM corresponds to the second hydration layer rather than the first water layer. We are convinced that the combination of structural information of the pristine and water-covered microcline (001) surface will contribute to uncovering the atomic-scale mechanism behind the exceptional ice-nucleation activity of feldspar minerals. References:Atkinson, J. D., Murray, B. J., Woodhouse, M. T., Whale, T. F., Baustian, K. J., Carslaw, K. S., Dobbie, S., O’Sullivan, D., Malkin, T. L., Nature, 498, 355-358, 2013.Dickbreder, T., Sabath, F., Reischl, B., Nilsson, R. V. E., Foster, A., Bechstein, R. and Kühnle, A., Nanoscale, DOI:10.1039/d3nr05585j, 2024.Holden, M. A., Whale, T. F., Tarn, M. D., O’Sullivan, D., Walshaw, R. D., Murray, B. J., Meldrum, F. C., Christenson, H. K., Science Advances, 5, 4316, 2019.Kiselev, A., Bachmann, F., Pedevilla, P., Cox, S. J., Michaelides, A., Gerthsen, D., and Leisner, T., Science, 355, 367–371, 2017.

  • Open Access

    Atomic structure and water arrangement on K-feldspar microcline (001)

    by Tobias Dickbreder, Franziska Sabath, Bernhard Reischl, Rasmus V. E. Nilsson, Adam S. Foster, Ralf Bechstein, Angelika Kühnle

    The properties of clouds, such as their reflectivity or their likelihood to precipitate, depend on whether the cloud droplets are liquid or frozen. Thus, understanding the ice nucleation mechanisms is essential for the development of reliable climate models. Most ice nucleation in the atmosphere is heterogeneous, i.e., caused by ice nucleating particles such as mineral dusts or organic aerosols. In this regard, K-feldspar minerals have attracted great interest recently as they have been identified as one of the most important ice nucleating particles under mixed-phase cloud conditions. The mechanism by which feldspar minerals facilitate ice nucleation remains, however, elusive. Here, we present atomic force microscopy (AFM) experiments on microcline (001) performed in an ultrahigh vacuum and at the solid–water interface together with density functional theory (DFT) and molecular dynamics (MD) calculations. Our ultrahigh vacuum data reveal features consistent with a hydroxyl-terminated surface. This finding suggests that water in the residual gas readily reacts with the surface. Indeed, the corresponding DFT calculations confirm a dissociative water adsorption. Three-dimensional AFM measurements performed at the mineral–water interface unravel a layered hydration structure with two features per surface unit cell. A comparison with MD calculations suggests that the structure observed in AFM corresponds to the second hydration layer rather than the first water layer. In agreement with previous computation results, no ice-like structure is seen, questioning an explanation of the ice nucleation ability by lattice match. Our results provide an atomic-scale benchmark for the clean and water-covered microcline (001) plane, which is mandatory for understanding the ice nucleation mechanism on feldspar minerals.