Model interaction potentials for real materials are generally optimized with respect to only those experimental properties that are easily evaluated as mechanical averages ͓e.g., elastic constants ͑at Tϭ0 K͒, static lattice energies, and liquid structure͔. For such potentials, agreement with experiment for the nonmechanical properties, such as the melting point, is not guaranteed and such values can deviate significantly from experiment. We present a method for reparametrizing any model interaction potential of a real material to adjust its melting temperature to a value that is closer to its experimental melting temperature. This is done without significantly affecting the mechanical properties for which the potential was modeled. This method is an application of Gibbs-Duhem integration ͓D. Kofke, Mol. Phys. 78, 1331 ͑1993͔͒. As a test we apply the method to an embedded atom model of aluminum ͓J. Mei and J.W. Davenport, Phys. Rev. B 46, 21 ͑1992͔͒ for which the melting temperature for the thermodynamic limit is 826.4Ϯ1.3 K-somewhat below the experimental value of 933 K. After reparametrization, the melting temperature of the modified potential is found to be 931.5Ϯ1.5 K.

We present a new algorithm for isothermal-isobaric molecular-dynamics simulation. The method uses an extended Hamiltonian with an Andersen piston combined with the Nosé -Poincaré thermostat, recently developed by Bond, Leimkuhler, and Laird ͓J. Comp. Phys. 151, 114 ͑1999͔͒. This Nosé -Poincaré -Andersen ͑NPA͒ formulation has advantages over the Nosé-Hoover-Andersen approach in that the NPA is Hamiltonian and can take advantage of symplectic integration schemes, which lead to enhanced stability for long-time simulations. The equations of motion are integrated using a generalized leapfrog algorithm ͑GLA͒ and the method is easy to implement, symplectic, explicit, and time reversible. To demonstrate the superior stability of the method we show results for test simulations using a model for aluminum and compare it to a recently developed time-reversible algorithm for Nosé -Hoover-Anderson. In addition, an extension of the NPA to multiple time steps is outlined and a symplectic and time-reversible integration algorithm, based on the GLA, is given.

No abstract

A novel computational treatment of dense, stiff, coupled reaction rate equations is introduced to study the nucleation, growth, and possible coalescence of cavities during neutron irradiation of metals. Radiation damage is modeled by the creation of Frenkel pair defects and helium impurity atoms. A multi-dimensional cluster size distribution function allows independent evolution of the vacancy and helium content of cavities, distinguishing voids and bubbles. A model with sessile cavities and no cluster-cluster coalescence can result in a bimodal final cavity size distribution with coexistence of small, high-pressure bubbles and large, low-pressure voids. A model that includes unhindered cavity diffusion and coalescence ultimately removes the small helium bubbles from the system, leaving only large voids. The terminal void density is also reduced and the incubation period and terminal swelling rate can be greatly altered by cavity coalescence. Temperature-dependent trapping of voids/bubbles by precipitates and alterations in void surface diffusion from adsorbed impurities and internal gas pressure may give rise to intermediate swelling behavior through their effects on cavity mobility and coalescence. IntroductionIrradiation of metals has long been known to culminate in macroscopic property changes including void swelling [1]. Characteristic stable voids and steady volumetric swelling develop for a range of temperatures and fluxes, independent of whether radiation bombardment damage occurs as disseminated Frenkel pairs or as small defect clusters. This can occur whether or not helium is generated along with atomic displacements. In either case, small, unstable voids, loops, and other defect clusters will develop almost immediately within the irradiated material. Their subsequent evolution determines the fluence required to create stable voids and achieve 1 arXiv:0803.3829v1 [cond-mat.mtrl-sci]

No abstract