Crystallisation Experiments with Complex Plasmas

1
Crystallisation Experiments with Complex Plasmas
M. Rubin-Zuzic1, G. E. Morfill1, A. V. Ivlev1, R.
Pompl1, B. A. Klumov1, W. Bunk1, H. M. Thomas1,
H. Rothermel1, O. Havnes2, and A. Fouquét3

• Max-Planck-Institut für extraterrestrische
  Physik, 85740 Garching, Germany
• 2. University of Tromsø, Department of Physics,
  9037 Tromsø, Norway
• 3. Institut Polytechnique de lUniversité
  dOrléans 14, ESPEO, 45067 Orléans Cedex 2,
  France

2
Outline

• Objectives
• Experimental setup and procedure
• Observation of crystal growth fronts
• Identification of different states
• Identification of detailed growth process
• Comparison with numerical simulations
• Summary

3
Objectives for our experiments
Study of dynamics of single particles during
crystallisation in real time without changing the
plasma parameters Questions
What are the self-organisation principles
governing crystal growth? What is the resultant
surface structure and its temporal
evolution? What is the microscopic (kinetic)
structure of interfaces?
4
PKE-Nefedov (PK3) - Experimental Setup
5
Formation and Growth of Plasma Crystals
Experimental parameters Particle diameter 1,28
µm 0,056 µm Particle number 107 Gas
Argon Gas pressure p 0.23 mbar Laser sheet
thickness 80-250 µm Images 1028 772 Pixel
Intensity values 8 bitImage rate 15
images/sec 4030 mm overview camera6.44.8 mm
high resolution camera Experimental procedure A
large vertically extended crystal (80 µm lattice
distance) is created (no horizontal layers!) The
system is disturbed by decreasing the ionization
voltage from 0.88 V down to 0.39 V. The
recrystallisation is investigated.
Overview
High resolutioncamera
High resolution
6
Experimental observation color codedmovie
7
Experimental observation color codedmovie
Particles fall down

The crystal dissolves from top to bottom
The crystallisation process starts at the bottom
A crystallisation front is observed
The propagation velocity of the crystallisation
front slightly decreases
Domains of different lattice orientation form
below the front
At the interface the thermal velocity of the
particles is higher
6.44.8 mm, 15 Hz, superposition of 10
consecutive image
8
Discovery of interfacial melting
16 sec later
Discovery of different crystal domains a stable
region of interfacial melting (a few lattice
thicknesses) is located between two lattice
domains. Similar phenomena have also been
observed in colloidal systems.

9
Comparison of structures - before voltage
decrease Triangulation

• No horizontal crystal layers (no influence of
  electrodes)
• Plasma crystal is oriented in an arbitrary angle
  towards the plane of the laser sheet
• No information about 3d structure

Lattice distance 80 mm
10
Comparison of structures after
recrystallisation Triangulation
Lattice distance 75 mm
11
Numerical Results Crystal Growth
2 D simulation box (molecular dynamics
simulation, gravity, shielded Coulomb potential,
neutral gas damping, Ar, Q3000e, initial
velocity is Gauss distributed with 3cm/sec,
parabolic potential). Fast dropping particles
disturb the upper part of the crystal. They
exchange their energy through Coulomb collisions
. Energy dissipation shock-and compressional
waves).
Boris Klumov
12
Sedimentation after power variation
40 sec later
40 sec later
Particles 1.28 mm Ueff (top)22.7 V, Ueff
(bottom)22.8 VURF (forward)0.39 V, URF
(backward)0.018 V Pressure 0.25
mbar Experimental procedure Voltage is
increased (from 40 to 140 levels) and then
quickly decreased back. ? Vertical extension and
particle distance decrease with time.
13
Cooling - Numerical result
fcc, hcp and a small amount of bcc structure is
present. The final ground state (fcc) is reached
much slower than predicted by neutral gas damping
(fcc/hcp volume ratio increases with time). A
reason might be that particles have a small size
(charge) variation. This allows a large number of
possible crystalline states. During the
sedimentation the particles slowly rearrange to
the state with lowest potential energy (very slow
process driven by thermal motion In
experiment the cooling is slower - additional
heating?.
Boris Klumov
14
Particle positions
Velocities in the yellow region
Mean velocity Growth velocity
Velocity distribution
15
Particle Velocity Variation Numerical Resultat
fixed time
3D simulation Yukawa System crystallises from
bottom upward (due to gravitational compression,
box no periodic conditions). The particles
thermal velocity increases upward and reaches a
local minimum at the position of the growing
crystal front.
Boris Klumov
16
Quantitative phase separation
Overlap technique - In the crystalline state
particles overlap almost completely in
consecutive images, in the disturbed (liquid)
state they do not.
- Superposition of n images Determination of
ratio of overlapping particle area in all n
images/particle area in the first image 1
particle is stationary 0 particle has moved
further than its image size
17
The overlap technique
Particle 1 (Frame 1)
Particle 2 (Frame 1)
18
The overlap technique
Particle 1 Frame 12 Ratio 0.1
Particle 2 Frame 12 Ratio 0.9
This yields a quantitative measure of particle
kinetic energy
19
Phase separation
20
Discovery of nanocrystallites and
nanodroplets during crystal growth
crystallite
droplet
Rubin-Zuzic et al. (Nature Physics,2006)
21
Fractal dimension of crystallisation front
Determination of the (linear) fractal dimension
of the crystallisation front to obtain a
quantitative measure for the variation of the
interface front during the growth process
1
2
Ln n? (length of measuring rod) L length of
crystallisation front
22
Fractal dimension of (1D) crystallisation front
log (L)
2.200
-gt surface structure is scale-free
2.100
a - 0.2 -gt D 1.19
2.000
2?
0.3
0.5
0.7
0.9
1.1
Log (n?)
23
Fractal dimension of the crystallisation front
Rough surface
Smooth surface
crystal growth follows a universal
self-organization pattern at the particle level
24
MD simulations of the crystallization front
Growth velocity crucial role of the
dimensionality in strongly coupled systems only
the 3D simulations provide quantitative agreement
with the experimental data (open circle)
Particle positions and thermal energies
Thermal velocities front has a complex structure
with a transition layer and transient
temperature islands
Boris Klumov
25
Summary (crystallisation experiment)
Crystallisation starts mostly from bottom,
because there the compression is higher (due to
gravity) than on top Crystal is build up with
particles from the gaseous state located above.
During crystallisation the particles in the
liquid state lose energy through collisions
with neighbours. Energy is dissipated by waves,
which are propagating through the crystal medium
(see numerical results). Interfacial melting
has been observed between domains, the layer is
only 2-4 lattice planes wide. The (transient)
energy source could be the latent energy of
converting an excited lattice state (hcp) into a
lower (ground) state (fcc). The transition
region is characterised by numerous droplets (in
the crystal regime) and crystallites (in the
fluid regime) and oscillating variaton in
roughness. The crystallisation front obeys a
universal fractal law down to the (minimum)
lattice spacing. Next steps new camera (higher
temporal and spatial resolution), fast 3D scans
during crystallisation, identification of 3D
crystal structure variation during crystal
growth, Future Investigation on the ISS
26
Thank You!