196
M. H. Köhler et al.
important to note that in terms of using nanostructured membranes in desalination
technology, such as RO systems, these advances open the door to work with a new
paradigm of membrane permeability and selectivity.
2 Computer Simulations
Before computers took over every laboratory—and even our daily life—science was
based on the assumption that we could only model the natural world through the
lenses of experiments and purely theoretical works. Take the example of the Dutch:
they built the Netherlands as we know it today with the help of the Delta Works,
a set of megastructures that hold back the ocean. To put together this enormous
project they had to build the Waterloopkundig Laboratorium, a massive concretebased hydrological laboratory conceived after World War II, in the pre-computer age,
where water could be guided into and out of large-scale trial models. Contemporarily
it is almost impossible to think of a world where things, from a small pen to a huge
hydroelectric plant, are not designed on a computer.
During an experiment, a set of results can be obtained directly from measuring
the properties of a system. Alternatively, a mathematical description could be used
to create a model, which in turn can be validated by its ability to describe some
physical behavior. Today, we have another tool to probe a physical system: computer
simulations. A model is provided by theorists but the calculations can be carried by
machines following some recipe. In this way, computer simulations unlocked the
possibility to study more complex and realistic systems, becoming a bridge between
theoretical models and real-world experiments [18].
There is no doubt that computer simulations play an important role in contemporary science development. Several different computational approaches can be used
to study physical systems. For example, when we aim to investigate nanoscaled
structures the interactions between atoms are the core of the simulation. Often,
empirical interatomic potentials (as Lennard-Jones potentials) are fitted to reproduce a given experimental property using van der Waals systems as a basis. Thereby,
molecular dynamics (MD) simulations are used to obtain the temporal evolution of
different systems. But if the interest resides on the electronic structure of a strongly
covalent material, electronic correlations are very important and we need to use
ab initio methods such as density functional theory (DFT). We can even merge both
approaches and use DFT to parameterize Lennard-Jones potentials that will be further
used in MD simulations. Method suitability will depend on which kind of properties we are interested in, from infrared spectra to dynamical and thermodynamical
information.
Just like a puzzle that fits piece by piece, MD simulations adapted perfectly to
the study of physical and chemical properties of nanofluidic systems. They have
allowed us to probe a wide range of microscopic behaviors that otherwise would be
tremendously difficult to access at nanoscale. Simulations within MD machinery are
usually performed in a feasible timescale with high accuracy. In other words, the MD
M. H. Köhler et al.
important to note that in terms of using nanostructured membranes in desalination
technology, such as RO systems, these advances open the door to work with a new
paradigm of membrane permeability and selectivity.
2 Computer Simulations
Before computers took over every laboratory—and even our daily life—science was
based on the assumption that we could only model the natural world through the
lenses of experiments and purely theoretical works. Take the example of the Dutch:
they built the Netherlands as we know it today with the help of the Delta Works,
a set of megastructures that hold back the ocean. To put together this enormous
project they had to build the Waterloopkundig Laboratorium, a massive concretebased hydrological laboratory conceived after World War II, in the pre-computer age,
where water could be guided into and out of large-scale trial models. Contemporarily
it is almost impossible to think of a world where things, from a small pen to a huge
hydroelectric plant, are not designed on a computer.
During an experiment, a set of results can be obtained directly from measuring
the properties of a system. Alternatively, a mathematical description could be used
to create a model, which in turn can be validated by its ability to describe some
physical behavior. Today, we have another tool to probe a physical system: computer
simulations. A model is provided by theorists but the calculations can be carried by
machines following some recipe. In this way, computer simulations unlocked the
possibility to study more complex and realistic systems, becoming a bridge between
theoretical models and real-world experiments [18].
There is no doubt that computer simulations play an important role in contemporary science development. Several different computational approaches can be used
to study physical systems. For example, when we aim to investigate nanoscaled
structures the interactions between atoms are the core of the simulation. Often,
empirical interatomic potentials (as Lennard-Jones potentials) are fitted to reproduce a given experimental property using van der Waals systems as a basis. Thereby,
molecular dynamics (MD) simulations are used to obtain the temporal evolution of
different systems. But if the interest resides on the electronic structure of a strongly
covalent material, electronic correlations are very important and we need to use
ab initio methods such as density functional theory (DFT). We can even merge both
approaches and use DFT to parameterize Lennard-Jones potentials that will be further
used in MD simulations. Method suitability will depend on which kind of properties we are interested in, from infrared spectra to dynamical and thermodynamical
information.
Just like a puzzle that fits piece by piece, MD simulations adapted perfectly to
the study of physical and chemical properties of nanofluidic systems. They have
allowed us to probe a wide range of microscopic behaviors that otherwise would be
tremendously difficult to access at nanoscale. Simulations within MD machinery are
usually performed in a feasible timescale with high accuracy. In other words, the MD