** Next:** Why boron nitride polymers?
** Up:** Material design from first
** Previous:** Material design from first

Recent advances in synthesis and nanofabrication technologies have
dramatically broadened the range of materials that can be designed
with desired and controlled
characteristics. In parallel with these
developments, improvements in computer simulation algorithms and the
availability of powerful computers offer a complementary way to probe
the properties of potentially interesting materials, even before (or
without) making them in the laboratory. In particular,
first-principles methods (whose input parameters consist only of a
list of the atoms in the system, and which then solve the
Schrödinger equation for the
interacting electrons in the potential of the nuclei) are a
powerful and unbiased tool for predicting the behaviour of new
materials at the atomistic level.
Among first-principles techniques, density-functional theory (DFT),
which is a modern reformulation of quantum mechanics in terms of the
electron density [1,2], offers a favorable ratio between
accuracy and
computational cost, which makes it suitable for (relatively) large
scale calculations. Its success in describing structural and
electronic properties of real materials has been recognized by the
award of the 1998 Nobel Prize in Chemistry to its founder,
Walter Kohn [3].
It has been combined with molecular dynamics, thus allowing
the simulation of systems at finite temperature [4,5].
Because of the absence of empirical parameters,
DFT is suitable
for applications in very diverse fields, from materials
science to biochemistry [6,7].
In particular,
various properties of polymers have been studied with such a method
(see e.g. Refs.
[8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25] ).
DFT accuracy has a price, which nowadays
limits size and simulation time to a few hundreds atoms and a few
picoseconds. On the one hand, this is sufficient to study
many properties successfully, as testified by
the thousands of published papers on DFT applications [7];
on the other, it is spurring on the development of
new methods for overcoming the scale and size limitations [26],
as well as
other pitfalls mainly due to the description of the troublesome
electron exchange and correlation.
In this paper, we present an example of how state-of-the-art DFT
calculations can be used to analyze the properties of a series of
hypothetical systems, designed using BN polymers as building
blocks [27]. For a preliminary screening,
making these polymers in a ``virtual matter laboratory'' [28]
is
easier, cleaner and less
dangerous than trying to make them in a real
laboratory.

** Next:** Why boron nitride polymers?
** Up:** Material design from first
** Previous:** Material design from first
Peter D. Haynes
2002-10-28