The NRL Dynamically Shielded Dust code (DSD) is a particle simulation code
developed to study the behavior of strongly coupled, dusty plasmas. The model includes the
electrostatic wake effects of plasma ions flowing through plasma electrons, collisions of dust and plasma
particles with each other and with neutrals. The simulation model contains the short-range strong
forces of a shielded Coulomb system, and the long-range forces that are caused by the wake. It also
includes other effects of a flowing plasma such as drag forces. Magnetic fields may also be
included in the formalism.
In order to model strongly coupled dust in plasmas, we make use of the
techniques of molecular dynamics simulation, PIC simulation, and the "particle-particle/particle-mesh"
(P3M) technique of Hockney and Eastwood. We also make use of the dressed test particle
representation of Rostoker and Rosenbluth. Many of the techniques we use in the model are common to all PIC
plasma simulation codes. The unique properties of the code follow from the accurate
representation of both the short-range aspects of the interaction between dust grains, and
long-range forces mediated by the complete plasma dielectric response. If the streaming velocity is zero,
the potential used in the model reduces to the Debye-Huckel potential, and the simulation is identical
to molecular dynamics models of the Yukawa potential. The model basically represents the dust
as simulation particles interacting via the dressed potential. The plasma appears only
implicitly through the plasma dispersion function, so it is not necessary in the code to resolve the
fast plasma time scales.
The following studies will be investigated using the DSD particle simulation
code:
CRYSTAL STRUCTURES
We will determine the crystal structures which occur under both gravity and
microgravity conditions, as a function of system parameters such as plasma density, neutral
gas pressure, Te, Ti, dust grain size, ion flow velocity, location in the discharge. In full gravity,
structures will occur in the sheath edge region where the ion flow velocity is on the order of the Bohm
velocity. In microgravity, the structures will be formed in the presheath where the flow
velocity is less than Bohm. In many cases, more than one crystal structure is possible and the free
energy difference between structures is very small. Thus we will determine any metastable crystal
states as well, by starting from very low dust temperature. We will determine the free energy
differences between states. We will attempt to categorize and explain the different crystal
structures seen in various regions of the discharges, and in particular to explain the differences between
the typically 2-D structures seen in terrestrial experiments and the 3-D structures which occur in
some regions of microgravity experiments.
LATTICE EXCITATIONS AND MELTING TRANSITION
We will determine the lattice vibration spectra of the various configurations
and compare this to experimental measurements by Goree et al in terrestrial experiments, and by
Morfill et al in microgravity experiments. As the dust temperature, Td, is increased, we will
also look at the percentage of condensed defects and isolated dislocations in the lattice, and
these results will be compared directly to the experimental results of Goree based on Delaunay
triangulation. We can reproduce the diagnostics used by the experimenters, and in addition we can
increase understanding by using techniques that are not available in experiments, e.g. we can
deliberately introduce lattice defects of various types, and study their stability, how they spread and merge,
and how they affect lattice vibrations. We will study in detail the microscopic processes that occur
as Td increases further and the lattice melts. We will focus on the role played by defects and
by vibrational modes, and by break-up into domains separated by flowing regions. We will determine
whether the melting transition is a first- or second-order phase transition, and in the
former case will determine directly the specific heat of melting. We will compare results with the KTHNY
melting theories, and later theories (Glaser (1993)), to determine quantitatively which if any of
the existing theories are adequate to describe dusty plasma crystal melting in 2-D and 3-D situations.
LIQUID STATE
We will study the structure factors governing short-range order in the dust
liquid state, such as the pair correlation function and orientational correlation functions, and compare
the results to measurements by the Goree group, and by the Morfill group in microgravity. We
will determine the excitation spectrum in the liquid state. We will study the liquid-solid
phase transition from the liquid side, to develop understanding of the basic dynamics of crystal
formation. Because plasma streaming leads to a partially attractive force between dust grains, it is
possible that there is a firstorder liquid-gas phase transition in the dust plasma. (In an ordinary electron-ion
plasma, there is no such phase transition, but only a gradual and smooth transition from liquid to
gas.) We will determine whether there is such a phase transition in any of the parameter
regimes accessible to terrestrial or microgravity experiments.
Joyce, G., Lampe, M., Ganguli, G., DSD - A Particle Simulation Code for Modeling Dusty Plasmas, Fifth Microgravity Fluids Physics and Transport Phenomena Conference, NASA Glenn Research Center, Cleveland, OH, CP-2000-210470, pp. 1140-1142, August 9, 2000.