Chemistry Reference
In-Depth Information
and running the software. If you prefer to ''do it yourself,'' most of the codes
provide detailed enough instructions to install the software on your own if you
have a little patience. If you obtain the source code, you can in principle com-
pile and run standalone versions of the software on a desktop machine. The
Mac OS X operating system runs native Unix, so installation on this architec-
ture is just like that on a Linux machine as long as you have installed the
appropriate compilers. For Windows machines, a compiler (or programming
environment like Visual C
) will be required to carry out the installation if
the code does not come with precompiled binaries. Alternatively, the Linux
emulation package Cygwin (http://www.cygwin.com/) can be installed to trick
your Windows machine into thinking it is a Linux box, and then regular Linux
installation instructions can be followed. A tutorial on high-performance com-
puting in computational chemistry was published in Volume 6 of this topic
series
142
and provides much useful information for the beginner.
รพรพ
SUMMARY AND OUTLOOK
Seven years ago, there were no examples of classical simulations being
used to study ionic liquids. Currently, a new publication on this topic appears
about every week. The tremendous growth in ionic liquids simulation is due in
part to the current popularity of ionic liquids, but the main driver for such stu-
dies is the fact that these calculations are generating useful results that are
helping us understand the physical chemistry of ionic liquids. Simulations
have given us the first detailed pictures of ionic liquid structure; they have
shown how the ions organize in the liquid phase, how solutes interact with
the ions, and that nanoscale segregation among polar and nonpolar groups
occurs. All of these predictions have been subsequently confirmed experimen-
tally. Simulations first predicted the enthalpies of vaporization of ionic liquids,
and subsequent experimental work confirmed these predictions. Simulations
have been used to study solvation dynamics and the agreement with experi-
mental spectroscopic studies has been excellent. Simulations have been used
to predict a wide range of ionic liquid properties including densities, heat capa-
cities, self-diffusivities, viscosities, electrical and thermal conductivities, and
solubilities. Many of these latter calculations were postpredictive, meaning
that they were carried out on systems for which the experimental results
were known. In general, there has been good agreement with the experiment
results, thereby establishing confidence in the simulation methods. There have
also been property calculations of ionic liquids that have not yet been made;
these predictions are still awaiting experimental confirmation. Used in this
mode, simulations are helping drive the discovery process for new ionic liquids
having properties tuned for particular applications. It is imperative that ato-
mistic simulations and experimental studies become complementary tools in
the search for both fundamental understanding and practical application of
