For the general installation instruction please refer to the documentation of LAMMPS and to SPH-USER package:

• http://lammps.sandia.gov/doc/Section_start.html
• http://lammps.sandia.gov/doc/USER/sph/SPH_LAMMPS_userguide.pdf

The original code is from https://github.com/slitvinov/lammps-sph-multiphase where the remaining of the README is based on.

Implementation

We add the following extension to USER-SPH package:

atom_style meso/multiphase

This is data structures which provides

• position
• velocity
• extrapolated velocity (vest)
• forces
• SPH density (rho)
• time derivative of SPH density (drho)
• internal energy per particle (e)
• time derivative of internal energy per particle (de)
• color gradient vector (colorgradient)
• per-particle heat capacity (cv)

This data structure can be activated by

atom_style meso/multiphase

pair_sph_colorgradient

A pair_style to calculate a color gradient

pair_style         sph/colorgradient
pair_coeff         I J     ${h} ${alpha}

Here, I and J are the types of SPH particles for which a color gradient is calculated, alpha is a surface tension coefficient, h is a cutoff.

pair_sph_surfacetension

A pair_style to calculate surface tension

pair_coeff         I J     sph/surfacetension ${h}

Here, I and J are the types of SPH particles for which a surface tension is calculated, h is a cutoff. Note that surface tension coefficient is included into color gradient.

pair_sph_heatconduction_phasechange

A modified heat conduction equation to use for phase change model. Has to forms. Simple form is equivalent to the heat conduction equation from USER-SPH package.

pair_coeff         I J  sph/heatconduction/phasechange  ${D_heat_ld}

Here, I and J are the types of SPH particles which interact and D is a heat diffusion coefficient.

Full form of the pair style is

pair_coeff         I J  sph/heatconduction/phasechange  ${D_heat_ld} TI TJ

where TI and TJ are temperatures for corresponding particles in I and J interactions.

NULL can be used as a placeholder to indicate that normal temperate should be used for corresponding particle

pair_coeff         I J  sph/heatconduction/phasechange  ${D_heat_ld} TI NULL

fix_phase_change

Fix which adds a phase change

fix     fix_ID group_ID phase_change &
             ${Tc} ${Tt} ${Hwv} ${dr} ${mass_v} &
	     ${pcutoff} ${l_type} ${v_type} ${insert_every} 123456 ${prob} region

fix_ID and group_ID are described in LAMMPS documentation. TC is critical temperature of the phase change, =TT= is transition temperature for the algorithm (should be set above TC), dr a characteristic distance for a new particle position, mass a mass of a new particle, h cutoff of the interaction, from_type and to_type types of the particles involved in phase transition, N frequency of the check for phase transition algorithm, seed a seed for random number generator, prob probability of the phase transition if all criteria are met (0<prob<1), region a region where algorithm checks for potential phase transition.

Examples

See examples/USER/sph/