Thermodynamic properties of liquid water as well as hexagonal (Ih) and cubic (Ic) ice are predicted based on density functional theory at the hybrid-functional level, rigorously taking into account quantum nuclear motion, anharmonic fluctuations and proton disorder. This is made possible by combining advanced free energy methods and state-of-the-art machine learning techniques. The ab initio description leads to structural properties in excellent agreement with experiments, and reliable estimates of the melting points of light and heavy water. We observe that nuclear quantum effects contribute a crucial 0.2 meV/H2O to the stability of ice Ih, making it more stable than ice Ic. Our computational approach is general and transferable, providing a comprehensive framework for quantitative predictions of ab initio thermodynamic properties using machine learning potentials as an intermediate step.

Machine-learning models have emerged as a very effective strategy to sidestep time-consuming electronic-structure calculations, enabling accurate simulations of greater size, time scale, and complexity. Given the interpolative nature of these models, the reliability of predictions depends on the position in phase space, and it is crucial to obtain an estimate of the error that derives from the finite number of reference structures included during model training. When using a machine-learning potential to sample a finite-temperature ensemble, the uncertainty on individual configurations translates into an error on thermodynamic averages and leads to a loss of accuracy when the simulation enters a previously unexplored region. Here, we discuss how uncertainty quantification can be used, together with a baseline energy model, or a more robust but less accurate interatomic potential, to obtain more resilient simulations and to support active-learning strategies. Furthermore, we introduce an on-the-fly reweighing scheme that makes it possible to estimate the uncertainty in thermodynamic averages extracted from long trajectories. We present examples covering different types of structural and thermodynamic properties and systems as diverse as water and liquid gallium.

We use extensive first-principles quantum mechanical calculations to show that, although the static lattice and harmonic vibrational energies are almost identical, the anharmonic vibrational energy of hexagonal ice is significantly lower than that of cubic ice. This difference in anharmonicity is crucial, stabilizing hexagonal ice compared with cubic ice by at least 1.4 meV=H 2 O, in agreement with experimental estimates. The difference in anharmonicity arises predominantly from molecular O-H bond-stretching vibrational modes and is related to the different stacking of atomic layers.

Ice is one of the most extensively studied condensed matter systems. Yet, both experimentally and theoretically several new phases have been discovered over the last years. Here we report a large-scale density-functional-theory study of the configuration space of water ice. We geometry optimise 74,963 ice structures, which are selected and constructed from over five million tetrahedral networks listed in the databases of Treacy, Deem, and the International Zeolite Association. All prior knowledge of ice is set aside and we introduce “generalised convex hulls” to identify configurations stabilised by appropriate thermodynamic constraints. We thereby rediscover all known phases (I–XVII, i, 0 and the quartz phase) except the metastable ice IV. Crucially, we also find promising candidates for ices XVIII through LI. Using the “sketch-map” dimensionality-reduction algorithm we construct an a priori, navigable map of configuration space, which reproduces similarity relations between structures and highlights the novel candidates. By relating the known phases to the tractably small, yet structurally diverse set of synthesisable candidate structures, we provide an excellent starting point for identifying formation pathways.

Nuclear Magnetic Resonance (NMR) spectroscopy is particularly well-suited to determine the structure of molecules and materials in powdered form. Structure determination usually proceeds by finding the best match between experimentally observed NMR chemical shifts and those of candidate structures. Chemical shifts for the candidate configurations have traditionally been computed by electronic-structure methods, and more recently predicted by machine learning. However, the reliability of the determination depends on the errors in the predicted shifts. Here we propose a Bayesian framework for determining the confidence in the identification of the experimental crystal structure, based on knowledge of the typical error in the electronic structure methods. We also extend the recently-developed ShiftML machine-learning model, including the evaluation of the uncertainty of its predictions. We demonstrate the approach on the determination of the structures of six organic molecular crystals. We critically assess the reliability of the structure determinations, facilitated by the introduction of a visualization of the of similarity between candidate configurations in terms of their chemical shifts and their structures. We also show that the commonly used values for the errors in calculated 13 C shifts are underestimated, and that more accurate, self-consistently determined uncertainties make it possible to use 13 C shifts to improve the accuracy of structure determinations.