Mesoscopic and Atomistic Modelling of Systems and Processes

Computational simulation and modelling of systems and modelling of processes is a very modern field, the importance of which increases strongly. This is due to the enormous progress in computer technology as well as in the progress of theoretical concepts and models. In quite a few cases, computational simulations are on the same level of accuracy and reliabilty as experiment. The advantage of computational simulations is that all the parameters are well defined and their influence on the results can be analyzed. By that, computational-theoretical modelling becomes a powerful tool for predicting yet unknown properties, and suggesting experiments. Therefore, strong collaborations with experimental groups are a natural consequence, which offers the student and scientist the almost unique opportunity to get experience with both technqiues, the experiment and the computer simulation.

In our group, two basic concepts are pursued. One concept consists in truly atomistic simulations by solving the Schrödinger equation for a many body quantum system of atoms, which interact in terms of chemical bonding. No empirical information is needed, every property - which can be derived from the wave functions solving the Schrödinger equation - is calculated directly on the computer in a truly ab initio way. The crucial point is the treatment of the many body interactions which reflect all the quantum properties, such as Pauli's principle and quantized states. We do this in terms of density functional theory (DFT) which is the fundamental theory most widely used nowadays in many simulations of solid state science and materials science.

The other basic concept is the modelling of polymer systems. Quantum mechanical calculations from first principles are too exhaustive in this case, as length scales of structural properties as well as time scales of dynamic properties are spread over several orders of magnitude. Thus, mesoscale simulation techniques (combining several atoms or monomers to segments and using more simple potentials) are used as well as force field based atomistic classical dynamics for smaller systems.

In the subgroup of G. Zifferer numerical investigations comprising simulations of polymer systems as well as theoretical and numerical investigations of polymerization processes are performed.
Thus, the projects may be divided into two main areas, i.e. (1) modeling of polymer and oligomer systems based on atomistic molecular dynamics and mesoscale simulation techniques and (2) theoretical and numerical aspects of the kinetics of polymer processes. A further project, not connected to polymers, refers to the atomistic simulation of amorphous ice phases under high pressure.

Detailed investigations of the concentration dependence of characteristic polymer properties have been undertaken supplemented by analytical models. An important feature was the calculation of the thermodynamic shielding factor of termination reactions between polymer radicals, which served as the basis of investigations on free radical polymerization considering a chain-length dependent termination rate coefficient. Recently, the shielding concept has been extended to various types of reactions including Z-RAFT star polymerization and surface initiated polymerization (part of FWF Projects P20124 and P23142)

Numerical and analytical modeling of pseudostationary polymerization processes yielded a sound theoretical basis for the pulsed laser polymerization method which in the meantime is used all over the world and is an IUPAC recommended benchmark method for the determination of kinetic coefficients; nevertheless, further improvements are still in progress.

A lot of efforts have also been made to study the properties of a variety of polymer systems by use of Monte Carlo methods: linear as well as branched and ring shaped chains have been examined. Isolated chains, pairs of chains, fundamental bulk properties, surface and interface properties have been investigated for homo- and copolymers.

In recent time we extended our interest to off-lattice mesoscale simulations and to fully atomistic simulations, e.g. glass transitions of polymers, small polymer chains in solution, adsorption of oligomers at metal oxide surfaces and super cooled water. For these latter (atomistic) investigations the commercial program package Materials Studio (© Accelrys) and the open source program GROMACS (http://www.gromacs.org) is used, while for the other cases the necessary programs have been developed by ourselves.

For details see homepage.univie.ac.at/gerhard.zifferer. As an example a snapshot of a star-branched polymer (with different colors for segments of different arms) is shown.

The subgroups of P. Herzig and R. Podloucky, see www.tssc.univie.ac.at,
homepage.univie.ac.at/peter.herzig and
homepage.univie.ac.at/renate.eibler
apply DFT approaches for solving materials properties.

*P. Herzig* works on transition metal and rare earth hydrides which show
interesting physical properties, like metal-insulator transitions depending on
hydrogen concentration ("switchable mirrors"). Our investigations reveal how
the structural properties (hydrogen vacancies and the huge atomic relaxations
induced by them) are related to the electronic and optical properties (band
gaps) of these hydrides.

In the figure the electron density for a particular
state in an exceptionally stable hydrogen double vacancy in almost
stoichiometric LaH3 is shown. Further work is done in boride systems where NMR
spectra are simulated and compared to single-crystal measurements by Prof.
Zogal in Wroclaw and in Li-transition metal nitrides which are remarkable for
their fast Li+ ion diffusion.

In the group of *R. Podloucky* several scientific and technological aspects are under
considerations as can be seen from the funded projects, which deal with e.g.
nanoscience on surfaces (oxide surfaces and nanostructure), surface and
adsorption properties (complex adsorption of atomic layers), precipitations in
metals and alloys, nanocoating of materials (hardening of materials), properties of compounds
(magnetism, bonding, phase stabilities, superconductivity).

As an example, the figure (by courtesy of C. Franchini)
shows the results of a simulation of
thin films of manganese oxide as grown on a palladium substrate with different
coverages. The top row in grey shows
calculated scanning-tunneling microscopy (STM) images which can be compared to
the experimental STM image coloured in brown (in the center). The middle row
is a topview of the calculated atomic geometry (with oxygen in red and
manganese in blue),
the bottom row is a sideview cutting through the film with the top layer being
the surface. By closer inspection with an experienced eye, all the shown
calculated structures are found in the experimental STM image, which indicates
how sensitive the manganese oxide layer is to small changes of physical
parameters. In this particular case for both, theory as well as experiment it
is difficult to get well-defined structures, and both need each other to
derive conclusive results.

The research of *R. Stadler* and his co-workers focuses on single-molecule
electronics, which has become a vibrant area of nano-electronics,
because the predictions of Moore's law of a continuous rise in the
performance of digital devices due to their ongoing miniaturisation,
cannot be upheld down to the atomic scale based on Silicon devices. For
realising the potential of this field, it is necessary to design
realistic devices by theoretical means, where the active part of the
circuit would be performed by a single molecule junction, i.e. a single
molecule sandwhiched between two metal electrodes. Ideally one would
like to combine two things: i) A device scheme developed and justified
by theoreticians should be so simple that it can be applied by
experimentalists without significant theoretical knowledge; ii) the
device scheme should be reliable enough, which means that its validity
has to be derived from and assessed by first principles calculations.
Both i) and ii) have been achieved recently, where a graphical scheme
has been established, which can predict the occurrence or absence of
quantum interference (QI) effects in relation to the molecular
structure. The findings have been verified by density functional theory
(DFT) calculations and provide an important tool for the design of data
storage elements as well as logic gates based on single molecules. This
work has been published in Nano Lett. 10, 4260 (2010) and highlighted in
a recent Uni:View contribution.
Another aspect of this field is that in spite of the recent progress in
experimental and theoretical research on single molecule conductivity in
a ultra-high vacuum setup at cryogenic temperatures, the latter two
conditions impose severe limits on any practical applications.
Experimental studies in an electrochemical environment offer a new
perspective notably at room temperature but for a better understanding
of electron transport in such an environment it is essential to arrive
at a clear theoretical picture of its mechanism based on first
principles calculations.
This is a formidable task. Not only are the transition metal complexes
which need to be investigated rather large and potentially problematic
in terms of an accurate description of the localization of charges, but
also the influence of solvent and substrate further complicates the
picture. A comprehensive approach within the framework of the
semi-classical Marcus theory needs to be developed for electron transfer
based on vibrationally induced electron hopping but also the theoretical
description of the competing coherent tunneling mechanism requires an
adjustment of the oxidation state of the central redox system, which
poses significant methodological challenges for DFT. This research is
embedded in close collaborations with experimentalists at IBM Zurich and
Imperial College London.

Our group member *I. Schnöll-Bitai* passed away in December, 2008. An obituary may be found in dieUniversitaet-online. Her research interests were focused on two main topics, namely polymerization kinetics and analysis of polymers which were carried out in national and international cooperations and in the frame of IUPAC projects. The method of pulsed laser polymerization was developed experimentally, backed up by the corresponding theoretical calculations and its versatile applicability was demonstrated for polymerization in homogenous (bulk, solution) and heterogeneous (microemulsion) systems for some monomer and comonomer systems. Several methods based on the concepts of pseudostationary and quenched instationary polymerization were developed as well. For the analysis of the molecular weight distribution (MWD) obtained by size-exclusion chromatography (SEC) it is necessary to take into account the phenomenon of band broadening (BB), thus several methods to determine its extent were developed and tested experimentally. The ultimate goal of correcting for the deviations from the true MWD due to BB was achieved (at least partially) for some selected types of information deduced from chromatograms with the aid of equations based on theoretical considerations and simulations. Lately, Irene was engaged in the use of matrix assisted laser desorption / ionization mass spectroscopy as a complementary technique (to SEC) which offers an alternative route to gain insight into BB phenomena of SEC.

- Towards Advanced Functional Materials and Novel Devices

PW-7 "Advanced polymeric materials - From calculation to application":

Edyta Wawrzynska, visiting student from 15.10.2012 to 15.08.2013

G. Zifferer:

R. Stadler:

G. Zifferer:

Offenlegung nach MedienG §25:

Medieninhaber: Universität Wien / Fakultät für Chemie / Institut für Physikalische Chemie

1090 Wien, Währinger Straße 42