|
The brochure of the John von Neumann Institute for Computing
is available in English and in German. It can be ordered at the NIC secretariat
(nic@fz-juelich.de).
deutsche Broschüre (pdf) | English brochure (pdf)
 Polymers
Not only do all organisms consist of largely soft matter (biopolymers such as DNA, proteins,
lipids, which form cell membranes, etc.) plus water, but also many materials in our daily life
contain polymers. Applications range from simple commodities such as plastics for
yogurt cups to
high-tech materials for optical lenses or electronic applications. However, what is
special and
what makes these systems soft? Soft matter molecules are huge compared to the typical
size of an
atom. They contain thousands or even millions of atoms. This results in huge conformational
freedom, meaning strongly fluctuating molecular shapes. Fluctuations are governed by thermal
activation where the thermal energy, kBT, is the relevant
energy scale. Since this is much smaller
than a typical bond energy (a carbon-carbon bond energy is about
80 kBT), the interaction energy
density is very small and the material is soft (as a first approximation
elastic constants are given
by the energy density). This leads to characteristic difficulties for
theoretical and computational
modeling. Due to the huge range of relevant length scales from local
chemical bonds all the way
up to the mesoscopic shape fluctuations, and, even more important,
many orders of magnitudes in
time (up to 10 or even more) have to be considered. As a consequence,
investigations of system
properties span a wide range of methods and levels of description.
This also is reflected in the
scientific activities at NIC. Here a few examples are given. The topics considered cover basic
fundamental questions concerning the generic universal behavior of macromolecular systems in
the limit of very long chains all the way up to predictions of material properties.
The first aspect
lacks any technical applications but is essential for a basic understanding and
in many cases
forms the starting point for more specific investigations. Typical examples are
"lattice animals"
as models for branched polymers. Polymeric systems with random branches (lattice animals) are
of great interest. They pose special simulation problems, since it is very
difficult to equilibrate
objects large enough to study asymptotic scaling laws. Although Monte Carlo calculations along
these lines have been performed for many years only recent developments in the group of P.
Grassberger at NIC allowed systems of more than 10,000 monomers to be efficiently simulated
and analyzed. Algorithms along these lines will also be of use for simple
protein models or for
branched structures which are not entirely randomly branched.
Other important systems are biological membranes. It is only
possible to study small systems in
full atomistic detail, thus restricting the investigations to
local interactions and leaving out
essentially all relevant membrane fluctuations. When it comes to interactions with membrane
proteins both microscopic interactions as well as global membrane properties are equally
important. To study the latter, the group of F. Schmid in Bielefeld
investigates models of rather
idealized amphiphiles. This allows them to treat large systems and
to study the interaction of a
double layer with a model membrane protein. Eventually, when this
is connected to more detailed
models, a more complete understanding of characteristic phenomena linked to membrane
proteins is expected.
In addition to bulk properties, interfaces or surfaces are of
special relevance. Surface coating is
common in many areas of our daily life. However, painting a wall
with dispersion paint is already
a rather sophisticated procedure. A way to produce very thin coatings is to graft polymers
chemically or physically with one end to a surface. Such polymer layers can be used, for
example, for biocompatibilization of surfaces. Experimentally,
they sensitively respond to the pH
value of the surrounding liquid and can either collapse or swell. This problem can be mapped
onto the situation of varying the solvent quality from good to poor, as the contribution of L.
Wenning and M. Müller from Mainz shows. Under good solvent conditions, the brushes
completely cover the surface while in poor solvents a characteristic
surface pattern of coagulating
polymer beads occurs and, because of this, leaves random patches of the surface "unprotected".
Even closer to practical application is another study from the group of F. Schmid.
They looked at
polymeric liquid crystals attached to a surface. Liquid crystals play an
important role in many
applications for displays. One problem is to control the orientation of the
mesogenic units with
respect to the surface. At a hard wall, small molecule liquid crystals are
oriented parallel to the
wall, which is unfavorable for many applications. By linking the mesogenes
into a polymer, they
can be oriented differently. The example shows the orientation of the
mesogenes with respect to a
surface to which the polymers are grafted. Depending on the grafting density,
one can adjust the
orientation continuously from parallel to almost perpendicular to the surface.
As in previous years, the examples show the variety of soft matter studies performed at NIC.
They give an impression of the breadth of the questions encountered when dealing with soft
matter, although they only cover a very small part of what is of interest nowadays. Because
analytical theory only can treat simplified and limiting cases, and experiments
typically deal with
much more complicated, often even poorly characterized systems, computer simulations, as the
examples show, are an indispensable intermediate between these two other means of research.
(Kurt Kremer, Max Planck Institute for Polymer Research, Mainz)
Although Monte Carlo methods have been used in statistical
physics for more than half a century,
developing more efficient Monte Carlo algorithms is still a
very active field. This is particularly
true for polymer physics, where topological constraints
tend to make standard algorithms
inefficient. The figure shows a randomly branched polymer
(more precisely, a "lattice animal" on
the bcc lattice) of 16,000 monomers, generated with a
newly developed algorithm. Such
simulations allow us to estimate, for example, the scaling
behavior of the diameter of branched
polymers and of their configurational entropy. But the basic
strategy they use can then be applied
to many more problems, ranging from the folding of proteins
to fluctuation effects in simple
chemical reactions.
(Hsiao-Ping Hsu, Walter Nadler, Peter Grassberger,
NIC Research Group "Complex Systems", Jülich)
Lipid (fat) molecules are basic constituents of cell membranes.
They are "amphiphilic", i. e., they
contain water-loving and water-hating parts. In a water environment,
under appropriate
conditions, they assemble spontaneously into sheetlike structures:
They build bilayers such that
the water-hating ends are shielded from the water. Such bilayers form
the skeleton of a biological
membrane, which is then filled by numerous other functional biomolecules.
The structure,
organization, and function of the latter depends to a large extent on
their local lipid environment.
We try to understand the interactions between lipids and proteins by
computer simulations.
Essential properties of lipid bilayers can be reproduced by simplified
models, which just account
for the amphiphilic character of the molecules. By simplifying the
protein in a similar way, we
can study basic physical interaction mechanisms.
The snapshots show an idealized membrane-protein system in two different membrane phases
("liquid" and "gel"). The protein shown here is designed in such a way that it represents a
transmembrane protein with a single alpha-helix. The lipid-protein interactions depend
sensitively on the state of the membrane.
(Olaf Lenz, Friederike Schmid, Department of Physics, University of Bielefeld)
For many technological applications, polymers are used as surface coatings.
As one example, it is
possible to make materials biocompatible which would otherwise be attacked by our immune
system. As another example, one can use polymer coatings to reduce friction between hard
sliding surfaces. One way to stabilize such a surface coating is to attach one
end of the polymer
chains chemically to the surface.
One interesting property of polymer chains is their sensitivity to
environmental changes. A single
polymer chain in solution can, for example, if the pH of the solution is altered, change its
conformation from a loose random coil structure into a compact, dense globule. This behavior
gives rise to the term "smart polymers". When such polymer chains are attached to a surface at
one end and when there is a sufficiently high density of grafting points and they are immersed
into a liquid that is a good solvent for the polymers, they will stretch
out, making the surface look
like a polymer brush. If a liquid is allowed to flow past the polymer
brush it will collapse, if there
is any change in the composition or pH of the liquid - thus creating a sensor.
To advance such technological applications it is necessary to develop a fundamental scientific
understanding of the structure of a polymer brush under varying environmental conditions (here
we use the temperature as a control parameter for the structure of the brush) and also of the
influence of processing conditions (for instance the density and regularity of
the positions of the
anchoring points). Our project tries to find answers to these questions using
computer simulation
methods. The following figures show examples of the configurations of a polymer brush in a
solvent whose quality induces a collapse of the brush since it deteriorates with decreasing
temperature.
The three figures show configurations of a polymer brush in a poor
solvent with a small density
of grafting points. For the temperatures of T=1.8, 1.5, 1.2
(in units of the strength of the attractive
interaction between the monomers) one can observe a collapse of
the brush and the development
of holes. At the highest temperature, T=1.8 (figure on the top), the
holes are uniformly distributed. At
T=1.5 (figure in the middle) they have coalesced into long stripes but
the brush still forms a continuous
network. At the lowest temperature, T=1.2 (figure on the bottom), the
empty stripes have grown together
and the brush has broken up into isolated clusters.
|
|
(Ludger Wenning, Marcus Müller, Institute of Physics, University of Mainz)
In LCD technology, surfaces are used to orient liquid crystals. Here we explore ways of
designing surfaces such that the liquid crystals are oriented with an arbitrary,
pre-defined tilt
angle. The idea is to graft liquid crystalline chains onto a surface which favors parallel
orientation. If these chains touch each other, they are forced to stand up,
and the competition of
the two effects leads to a finite tilt angle. The tilt angle can be controlled
through the grafting
density. We use simulations of systems of ellipsoids to explore this effect.
The snapshots illustrate the mechanisms. The snapshot on the left shows the
liquid crystal in contact
with the bare substrate. The particles are on average oriented parallel
to the surface. The
snapshot in the middle shows the same liquid crystal (pink) in contact with a surface
decorated with liquid
crystalline chains (yellow). The particles are now tilted with respect
to the surface. The
figure on the right shows the same snapshot with transparent liquid crystal particles,
in order to illustrate the
conformations of the chains.
(Harald Lange, Friederike Schmid, Department of Physics, University of Bielefeld)
|
