Session K20: Matter at Extreme Conditions: Material Response and Phase Transitions

Phase transition and melting of titanium at high pressure-temperature conditions

Due to exceptional mechanical and thermal properties, behaviors of titanium (Ti) under high pressure-temperature conditions have been extensively studied. Upon compression, multiple phases of Ti are occurred via the martensitic phase transitions. The phase diagram up to 200 GPa and 3000 K was established. However, there is a lack of understanding of their stability and melting at elevated temperature-high pressure regime. Using quantum molecular dynamics simulations, we show that applying moderate temperatures on several Ti phases can result in either reconstructive or displacive phase transitions. Stronger heating leads to the formation of a new state before the melting takes place.   

Coexistence of Superionicity and Superconductivity in Li2MgH16

Recently, we and collaborators develop a stochastic path-integral approach (SPIA) to study ion-mediated superconductivity in general systems. The method makes no assumptions of the nature of ion motion, and thus allows studies beyond harmonic solids. We implement the method based on density-functional theory (DFT). Applications predict superconducting liquid hydrogen  and verify the strong effect of anharmonicity in hydride sulfide.

In this work, we study superconductivity in the superionic phase of the clathrate hydride Li₂MgH₁₆, where hydrogen ions diffuse among the lattice formed by lithium and magnesium ions. By employing the SPIA, we non-perturbatively take into account the effects of quantum diffusion and anharmonic vibrations. Our calculations show strong electron-ion coupling (λ(0)) and a high superconducting transition temperature (T_c) of 277 K under 260 GPa, at which the material is still superionic.  T_c is significantly suppressed compared with the result T_c=473 K obtained from the conventional approach based on the harmonic approximation. Our study, based on a first-principles approach applicable to superionic systems, predicts that the superconductivity and superionicity can coexist in Li₂MgH₁₆.   

Predicting Crystal Structures from Molten Phases at Extreme Pressures

We will present an approach for predicting solid phases under specific pressure and temperature conditions using liquid informed structure searches (LISS). Machine learning potentials and atomistic simulations are employed to study the emergence of finite-temperature crystal structures from molten phases under compression. The method will be illustrated with applications to the phase diagrams of several systems, including Mg and Sn.   

Prediction of Pressure-induced Polymerization and Recovery of a Dense and Energetic CO-O2 Phase

We report results from first principles calculations on the pressure-induced polymerization of carbon monoxide-oxygen (CO + O₂) mixtures at finite temperature. Our simulations predict the formation of a dense and energetic amorphous phase at pressures significantly lower than where the appearance of an amorphous CO₂ solid has been previously reported. Our analysis indicates that this dense polymer can be recovered to ambient conditions. A characterization of its structural and energetic properties will be presented and compared to known CO₂ solid phases.    

Time resolved X-ray diffraction studies on the non-equilibrium structural dynamics of silicate glasses at ultrahigh pressures

Shock experiments give a unique insight into the behavior of the matter subjected to extreme conditions and are key to modeling material failure and deformation dynamics under ballistic impact. To understand the role of glass network structure on the dynamic response of silicates, we performed laser-driven shock compression experiments on soda-lime glass (SLG) and borosilicate glass (BSG). In-situ VISAR and XRD were used to determine the pressure and structural phases, respectively. Following laser shock compression, between 40-90 GPa, SLG was seen to undergo a poly-amorphic transition from its ambient 4-fold coordinated amorphous structure to a 6-fold coordinated high-density amorphous (HDA) structure. Above 150 GPa, SLG transforms into a high-density melt. Time-resolved in-situ XRD also revealed that a small volume of crystalline d-niccolite phase was formed in the SLG during release between 55 - 80 GPa. In contrast, BSG was seen to transition from its ambient 4-fold amorphous structure to a crystalline stishovite structure in the 40 to 60 GPa range. A mixed phase was observed between 60-70 GPa with HDA and crystalline d-niccolite phases. Above 70 GPa, BSG transforms to HDA followed by high-density melt like that observed in SLG. A pressure-temperature and pressure-time phase diagram was created to describe the structural phases observed. Our in-situ XRD studies were able to directly resolve the structural transformation in silicate glasses and clarify the role of composition.   

Accurate Chemically-Sensitive-Space Factors for Enhanced High-Pressure Crystal Structure Prediction: A Leap beyond Machine Learning

We developed a hierarchical iterative random sampling statistics (HIRSS) method for accurate and efficient high-pressure structure prediction. The HIRSS can effectively explore chemically sensitive space and identify crystal structures that satisfy certain chemical constraints, namely stoichiometry, coordination numbers, bond lengths, and angles. Other factors, such as crystal densities, symmetries of space groups, lattice systems, and neighbouring relations of chemical functional groups, are used to define the boundaries of the chemical space and guide the search. Thus, accurate factors of the chemically sensitive space can be used as a robust fingerprint of random sampling statistics to extract crystal structures with three-dimensional atomic coordinates. Our results indicate that predicted crystal structures using chemically sensitive space factors through HIRSS have significantly reduced sampling errors, thereby achieving accurate and efficient high-pressure structure prediction for a small number of samples. For complex dense silanes SH₄ at high pressure, HIRSS yielded the two lowest enthalpy structures from only 20 crystal structure samples. The predicted structures matched results from the machine learning method using data-derived potentials trained from more than 6,500 structure samples. Additionally, HIRSS structure predictions revealed a new structure with comparable enthalpies that was not discovered by the previously mentioned machine learning method.   

In-situ x-ray diffraction of bismuth loaded to 4.4 GPa via shock reverberation

Static and dynamic high-pressure studies of bismuth have yielded conflicting phase diagrams, especially regarding the existence of the incommensurate host-guest Bi-III phase. Prior shock compression experiments may not have provided adequate loading conditions (stemming from uncertainties in the bismuth equation of state) and/or may not have provided sufficient loading duration to enable bismuth to transition to Bi-III. To drastically reduce loading condition uncertainties and to monitor phase transition kinetics for more than a few ns, we performed in-situ x-ray diffraction measurements of bismuth subjected to shock reverberation loading up to 4.4 GPa at the Dynamic Compression Sector of Argonne National Laboratory's Advanced Photon Source. Machine learning-informed molecular dynamics analysis and traditional Rietveld refinement of collected diffraction patterns indicate that, under steady loading conditions commensurate with Bi-III, bismuth exhibits a mixed state that evolves over 100s of ns.   

Mechanical and dynamical stability of nitrides at high pressure.

The structure of the Na-N, K-N, P-N, and Fe-N solids was predicted using evolutionary algorithms and density functional theory (DFT). The pressure dependence of elastic moduli of the predicted nitrides is analyzed. Calculations of phonon dispersion important for understanding dynamical stability are accompanied by estimates of mechanical stability based on elastic constants, Born conditions, and direct calculations of eigenvalues of the stiffness matrix. All the studied structures are found to be mechanically stable indicating that elastic mechanical stability condition is the least critical compared with thermodynamical stability based on the construction of a convex hull for the enthalpy of formation and dynamical stability, based on phonon dispersion. For example, PN3 crystal with Immm symmetry group previously predicted to be stable above 200 GPa is metastable at 30 and 50 GPa according to convex hull construction, but PN3 crystal is found to be mechanically and dynamically stable at these pressures.   

Time-Resolved Transformation Kinetics of the BCC to HCP Transition in Iron Using Piezo- Driven Compression

Understanding of the effect of kinetics on pressure induced phase transformations is important in interpreting experiments, but has historically been challenging due to the need for relatively high compression rates and time resolution. Use of dynamic diamond anvil cells (dDAC) and fast X-ray detectors allows us to overcome these historical limitations in a study of the BCC to HCP transition in iron. This transformation near 13 GPa has been extensively studied by both dynamic and static compression techniques. We examine the effects of compression rate, from 10^-3 to 10² GPa/s. In addition, we investigate the effects of stress state and particle size by performing experiments on cells loaded without pressure media and on iron samples with a smaller mean particle diameter.

 

The results are examined in an Avrami framework, where phase evolution is determined by the interplay of nucleation, growth, and impingement. We assess the suitability of the Avrami equation for the high-pressure phase transformation in iron, discuss what differentiates high-pressure experiments from more conventional studies, and develop a new model which allows for non-standard impingement processes that is also more suited to high-pressure experiments. With the development of a kinetic law for high-pressure transformations, future studies may more directly constrain kinetic parameters, such as nucleation rate, that are currently difficult to assess experimentally.   

Speciation of carbonates in water under pressure: the effect of counterions

We address a problem relevant to the Earth deep carbon cycle: the speciation of carbonates in

water under pressure, as a function of the counterions present in the solution. Experimentally,

it is challenging to characterize the properties of species dissolved in aqueous solutions at

extreme conditions. Here, we adopt a first-principle computational approach, and we study the

speciation of carbonates in the presence of platinum using ab initio molecular dynamics and

the Qbox code (http://qboxcode.org/). We predict the structural and vibrational properties of

carbonates in in the presence of cations (K+, Na+, Ca2+, and Pt2+) and we compare our results

with limited experimental data and with geochemical models. Our study gives insights into the

speciation of platinum in aqueous carbonates which are critical to the understanding of

platinum deposits and ore formation.   

Towards a Rate-Dependent Material Model for a Polyetherimide Copolymer

Polyimides are a robust class of materials that are an industry staple in aerospace and electronics manufacturing for applications involving extreme conditions such as rapid temperature changes, irradiation, and dynamic mechanical loads including shocks. We perform all-atom molecular dynamics simulations of a common polyimide, i.e., poly (4-4'-oxydiphenylene-pyromellitimide), to predict and understand its response to high-rate uniaxial loads, isotropic compression, and thermal cycling. We find that the pressure-temperature-volume response is highly rate- and path-dependent when the material is subjected to rapid heating and compression rates typical of shocks. At low degrees of polymerization, the flow response to ultrafast uniaxial tension and compression is predicted to be initially strain hardening. Rate dependent structural changes are identified for large deformations, including apparent strain softening and failure due to the formation of voids under tension. These simulations are used to identify factors governing material equation of state and strength needed for accurate constitutive models at larger length scales.   

Solute influences on void nucleation in magnesium alloys studied by transition interface sampling

Precipitation hardening, a commonly used method for strengthening of alloys, results in materials microstructures that interfere with dislocation slip. Previous studies have shown that interactions between solute atoms and vacancies promotes agglomeration of vacancies into voids. However, this nucleation may involve mutli-stage processes that are not well described by classical nucleation theory. Simulations performed in a Mg-Al system exhibit intermediate metastable states associated with solute-defect clusters. To further explore the hypothesis that solute-vacancy interactions play an important role in promoting void formation, we replace aluminum with yttrium, which is not favored to bind to vacancies, unlike aluminum. Using the Replica Exchange Transition Interface Sampling methodology and molecular dynamics simulations, we have explored the phase space of the void nucleation transition by sampling dynamical trajectories to obtain the activation barrier of the process. The comparison with the Mg-Al alloy barrier is presented and analyzed. We discuss the implications for developing a more accurate void nucleation model that takes solute effects into account.   

Modeling Fracture in Protective Aluminum Surface Materials Exposed to Crystal Defects

Aluminum metal reacts readily with the environment to form protective oxide or hydroxide surface films, which inhibit further corrosion by shielding the bare metal underneath. Fracturing of the protective film due to stress corrosion cracking can expose the metal, accelerating corrosion chemistry. Film fracture is further complicated by the presence of defects that may be present in the protective layer. Determination of the fracture tendency of these materials subjected to defects can be leveraged by component-scale predictive lifetime models to make more refined predictions about net reaction kinetics. In this work, we use molecular dynamics with a reactive force field to predict the response of bulk γ-Al₂O₃ and γ-Al(OH)₃ under tensile loading up to the point of fracture for a range of high strain rate conditions. The effects of crystal defects such as grain boundaries and porosity are also considered. We find that while defects in γ-Al₂O₃ can dramatically increase fracture tendency, they are typically insufficient to reduce fracture strains to those observed for perfect single crystal γ-Al(OH)₃.

 

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.