We determine the phase diagram for dense carbon-oxygen mixtures in white dwarf (WD) star interiors using molecular dynamics simulations involving liquid and solid phases. Our phase diagram agrees well with predictions from Ogata et al. and from Medin and Cumming and gives lower melting temperatures than Segretain et al. Observations of WD crystallization in the globular cluster NGC 6397 by Winget et al. suggest that the melting temperature of WD cores is close to that for pure carbon. If this is true, our phase diagram implies that the central oxygen abundance in these stars is less than about 60%. This constraint, along with assumptions about convection in stellar evolution models, limits the effective S factor for the 12C(α,γ)16O reaction to S(300)≤170  keV b.

Nonuniform neutron-rich matter present in both core-collapse supernovae and neutron-star crusts is described in terms of a semiclassical model that reproduces nuclear-matter properties and includes long-range Coulomb interactions. The neutron-neutron correlation function and the corresponding static structure factor are calculated from molecular dynamics simulations involving 40,000 to 100,000 nucleons. The static structure factor describes coherent neutrino scattering which is expected to dominate the neutrino opacity. At low momentum transfers the static structure factor is found to be small because of ion screening. In contrast, at intermediate momentum transfers the static structure factor displays a large peak due to coherent scattering from all the neutrons in a cluster. This peak moves to higher momentum transfers and decreases in amplitude as the density increases. A large static structure factor at zero momentum transfer, indicative of large density fluctuations during a first-order phase transition, may increase the neutrino opacity. However, no evidence of such an increase has been found. Therefore, it is unlikely that the system undergoes a simple first-order phase transition. Further, to compare our results to more conventional approaches, a cluster algorithm is introduced to determine the composition of the clusters in our simulations. Neutrino opacities are then calculated within a single heavy nucleus approximation as is done in most current supernova simulations. It is found that corrections to the single heavy nucleus approximation first appear at a density of the order of 10 13 g/cm 3 and increase rapidly with increasing density. Thus, neutrino opacities are overestimated in the single heavy nucleus approximation relative to the complete molecular dynamics simulations.

Neutrinos in core collapse supernovae are likely trapped by neutrino-nucleus elastic scattering. Using molecular dynamics simulations, we calculate neutrino mean free paths and ion-ion correlation functions for heterogeneous plasmas. Mean free paths are systematically shorter in plasmas containing a mixture of ions compared to a plasma composed of a single ion species. This is because neutrinos can scatter from concentration fluctuations. The dynamical response function of a heterogeneous plasma is found to have an extra peak at low energies describing the diffusion of concentration fluctuations. Our exact molecular dynamics results for the static structure factor reduce to the Debye Huckel approximation, but only in the limit of very low momentum transfers.

Nucleosynthesis, on the surface of accreting neutron stars, produces a range of chemical elements. We perform molecular dynamics simulations of crystallization to see how this complex composition forms new neutron star crust. We find chemical separation, with the liquid ocean phase greatly enriched in low atomic number elements compared to the solid crust. This phase separation should change many crust properties such as the thermal conductivity and shear modulus.

The nuclear pasta-a novel state of matter having nucleons arranged in a variety of complex shapes-is expected to be found in the crust of neutron stars and in core-collapse supernovae at subnuclear densities of about 10 14 g/cm 3 . Owing to frustration, a phenomenon that emerges from the competition between short-range nuclear attraction and long-range Coulomb repulsion, the nuclear pasta displays a preponderance of unique low-energy excitations. These excitations could have a strong impact on many transport properties, such as neutrino propagation through stellar environments. The excitation spectrum of the nuclear pasta is computed via a molecular-dynamics simulation involving up to 100,000 nucleons. The dynamic response of the pasta displays a classical plasma oscillation in the 1-to 2-MeV region. In addition, substantial strength is found at low energies. Yet this low-energy strength is missing from a simple ion model containing a single-representative heavy nucleus. The low-energy strength observed in the dynamic response of the pasta is likely to be a density wave involving the internal degrees of freedom of the clusters. PACS number(s): 26.60.+c, 24.10. Lx, 25.30.Pt Baryonic matter is organized as a result of short-range nuclear attraction and long-range Coulomb repulsion. Often the corresponding nuclear and atomic length scales are well separated, so nucleons bind into atomic nuclei that are themselves segregated into a crystal lattice. However, at the enormous densities present in astrophysical objects-densities that exceed that of ordinary matter by 14 orders of magnitudethese length scales become comparable and complex new phenomena emerge. Complexity arises because it is impossible for the constituents to be simultaneously correlated from nuclear attraction and anticorrelated from Coulomb repulsion. Competition among these interactions plays a fundamental role in the organization of matter and results in Coulomb frustration. Frustration-a ubiquitous behavior in complex systems ranging from magnetism to protein folding to neural networks-develops from the inability of a system to simultaneously satisfy all of its elementary interactions. For example, the Ising antiferromagnet on a triangular lattice is frustrated because not all of the nearest neighbor spins can be antiparallel to each other. Frustrated systems have unusual dynamics owing to the preponderance of low-energy excitations [1].At subnuclear densities of about 10 14 g/cm 3 (normal nuclear matter saturation density is 2.5 × 10 14 g/cm 3 ) Coulomb frustration is expected to promote the development of complex shapes. These shapes follow from the competition between surface tension and Coulomb energies. Whereas surface tension favors spherical shapes, Coulomb interactions often * favor nonspherical configurations. Therefore, a variety of complex structures with a diversity of shapes-such as spheres, cylinders, and plates-have been predicted. The many phases of nuclear matter displaying this variety of shapes are known collectively as nuclear ...

The formation of complex nonuniform phases of nuclear matter, known as nuclear pasta, is studied with molecular dynamics simulations containing 51 200 nucleons. A phenomenological nuclear interaction is used that reproduces the saturation binding energy and density of nuclear matter. Systems are prepared at an initial density of 0.10 fm −3 and then the density is decreased by expanding the simulation volume at different rates to densities of 0.01 fm −3 or less. An originally uniform system of nuclear matter is observed to form spherical bubbles ("swiss cheese"), hollow tubes, flat plates ("lasagna"), thin rods ("spaghetti") and, finally, nearly spherical nuclei with decreasing density. We explicitly observe nucleation mechanisms, with decreasing density, for these different pasta phase transitions. Topological quantities known as Minkowski functionals are obtained to characterize the pasta shapes. Different pasta shapes are observed depending on the expansion rate. This indicates non equilibrium effects. We use this to determine the best ways to obtain lower energy states of the pasta system from MD simulations and to place constrains on the equilibration time of the system.

Nuclear pasta, with non-spherical shapes, is expected near the base of the crust in neutron stars. Large scale molecular dynamics simulations of pasta show long lived topological defects that could increase electron scattering and reduce both the thermal and electrical conductivities. We model a possible low conductivity pasta layer by increasing an impurity parameter Qimp. Predictions of light curves for the low mass X-ray binary MXB 1659-29, assuming a large Qimp, find continued late time cooling that is consistent with Chandra observations. The electrical and thermal conductivities are likely related. Therefore observations of late time crust cooling can provide insight on the electrical conductivity and the possible decay of neutron star magnetic fields (assuming these are supported by currents in the crust).

We present a series of high-resolution, three-dimensional hydrodynamics simulations of a gravitationally unstable solar nebula model. The influences of both azimuthal grid resolution and the treatment of thermal processes on the origin and evolution of gravitational instabilities are investigated. In the first set of simulations, we vary the azimuthal resolution for a locally isothermal simulation, doubling and quadrupling the resolution used in a previous study; the largest number of grid points is (256, 256, 64) in cylindrical coordinates (r, ', z). At this resolution, the disk breaks apart into a dozen short-lived condensations. Although our previous calculations underresolved the number and growth rate of clumps in the disk, the overall qualitative, but fundamental, conclusion remains: fragmentation under the locally isothermal condition in numerical simulations does not in itself lead to the survival of clumps to become gaseous giant protoplanets. Since local isothermality represents an extreme assumption about thermal processes in the disk, we also present several extended simulations in which heating from an artificial viscosity scheme and cooling from a simple volumetric cooling function are applied to two different models of the solar nebula. The models are differentiated primarily by disk temperature: a high-Q model generated directly by our self-consistent field equilibrium code and a low-Q model generated by cooling the high-Q model in a two-dimensional version of our hydrodynamics code. Here, '' high-Q '' and '' low-Q '' refer to the minimum values of the Toomre stability parameter Q in each disk, Q min ¼ 1:8 and 0.9, respectively. Previous simulations, by ourselves as well as others, have focused on initial states that are already gravitationally unstable, i.e., models similar to the low-Q model. This paper presents for the first time the numerical evolution of an essentially stable initial equilibrium state (the high-Q model) to a severely unstable one by cooling. The additional heating and cooling are applied to each model over the outer half of the disk or the entire disk. The models are subject to the rapid growth of a fourarmed spiral instability; the subsequent evolution of the models depends on the thermal behavior of the disk. The cooling function tends to overwhelm the heating included in our artificial viscosity prescription, and as a result the spiral structure strengthens. The spiral disturbances transport mass at prodigious rates during the early nonlinear stages of development and significantly alter the disk's vertical surface. Although dense condensations of material can appear, their character depends on the extent of the volumetric cooling in the disk. In the simulation of the high-Q model with heating and cooling applied throughout the disk, thin, dense rings form at radii ranging from 1 to 3 AU and steadily increase in mass; later companion formation may occur in these rings as cooling drives them toward instability. When heating and cooling are applied only over the outer radia...