The thermal effects of external electromagnetic (e/m) fields on matter, e.g. macromolecular systems such as proteins or lipid bilayers, arise from the heat generated by the absorption of e/m energy. However, the precise mode of action of non-thermal field effects on matter is unknown . This gap in our fundamental understanding exists at both the quantum (sub-nuclear) level, e.g. on how e/m fields affect proton exchange in hydrogen-containing systems, and at the molecular and macromolecular level, where changes in chemical reactivity and gross structure, e.g. hydrogen bonding, can be stimulated by the application of e/m fields . Although the mechanisms of non-thermal e/m field effects are not well understood, it is believed that fields of specific frequency and intensity can excite certain vibrational modes of a macromolecule, which alters its conformation. The effects of these non-thermal modes on proteins, for instance, are quite surprising, including reversible changes in activity . It has been speculated that understanding the effect of e/m fields on proteins could be of use in designing molecules with specific properties . The issue of non-thermal effects on protein stability, and hence on function, in far IR and microwave fields has been the subject of extensive debate, particularly in relation to mobile communication signals on human health . E/m and electric field effects on proteins and other molecules of biological interest have been studied using molecular simulation , and this has opened up many possibilities to use simulation to gauge e/m and electric field effects at the atomistic and molecular level.
Electric fields have a significant influence on water, a compound of major biological relevance. Ice crystallisation can be triggered by an electric field , a process known as electrofreezing. At the nanoscale, the strong electric dipole of water molecules can be oriented to form chains or strips . MD simulation methods have been developed to incorporate e/m fields, and these have been applied to the study of bulk water and other aqueous and dipolar systems [9, 10] in the microwave to far IR region, studying dipolar response and the increase in hydrogen bond kinetics due to imparted rotational motion from dipolar alignment .
The impact of electric and e/m fields on electroporation , electrophoresis  and electro-osmosis  in nano-scale geometries, and the optimisation of nanoscale devices at the interface of biology  are also exciting avenues of current research to which molecular simulation can make important contributions and add insight. Static and alternating electric fields as a means of detecting DNA sequences in a de facto nanoporous capacitor have been investigated , and this has demonstrated the utility of simulation for realistic nanoscale device design, and the use of molecular simulation in the design of optimal nano-channel geometries and materials for electrophoresis and electro-osmosis applications is now a very real possibility [12, 13]. Storage of sets of ions in electromagnetic traps has led to several significant developments in atomic and cold-molecular physics, with applications in frequency standards , quantum information , and even clocks for deep space navigation . At a more mesoscopic level, dusty plasmas offer interesting physics for the two-dimensional Coulomb problem constrained by external fields .
In the past decade or so, molecular simulation has contributed much to our insights of basic science of electric and e/m field effects on matter and systems of biological interest. As will be described further in ‘State of the Art’ (vide infra), simulation offers the opportunity to investigate molecular mechanisms in atomistic detail, which facilitate the study of dipolar and translational response, frequency and intensity effects and other highly relevant areas of research. These findings can be used to good effect in guiding protein design or nanoscale device design which exploit external field effects.
In recent years, computer simulations have contributed greatly to the molecular-level understanding of biologically relevant molecules. An important example is MD simulation of protein function and folding. Initial MD simulation work focussed on the characterisation of folding intermediates, in conjunction with experiment, leading to the development of phenomenological models for folding. More recently, there has been a greater emphasis on the statistical study of energy landscapes of the protein, where folding pathways emerge as ensembles of protein conformations with characteristic signals to experimental probes. The most successful application of MD have been, unsurprisingly, to small, fast-folding proteins, e.g. hen egg white lysozyme, where advances have been allied with recent experimental developments to provide complimentary insights .
Monte Carlo methods have been tailored to better deal with the presence of external electric fields by a perturbative introduction of the electric field term, with applications to simulation of peptides and water clusters . Non-equilibrium MD simulation techniques have been developed to examine both the thermal and non-thermal effects of far IR and microwave radiation on matter for a variety of systems, including water , clathrate hydrates, metal oxides, ionic liquids, incipient denaturation of solvated proteins and water self-diffusion in carbon nanotubes through phospholipid bilayers . It has been found by various groups [6, 9, 18], that electric field intensities much lower than 0.01 V/Å do not tend to exhibit significant effects in terms of dipolar alignment of water or proteins. Although this field intensity is still several orders of magnitude larger than those applied typically in industry or experimentally [1-5], the limited sampling amenable from molecular simulation limits somewhat the range of applicable field intensities to obtain tangible simulation results.
As mentioned previously, water’s strong dipole moment makes it particularly responsive to electric and e/m fields. Although electric and e/m field effects on bulk water have been studied, including its dipolar response, heating, diffusion and increase in hydrogen bond kinetics, and electrofreezing observed due to dipole alignment in static fields [7, 9], the study of field effects on water in lower-dimensional wires and clusters has also been a fruitful and interesting area of research in recent years . However, the application of electric and e/m fields to ionic species (or those which are at least not electroneutral, e.g. charged proteins and polyelectrolytes)  has highlighted the importance of translational as well as dipolar response, where the charge-to-mass ratio is important, as well as conformational changes in proteins  and polyelectrolytes .
Non-equilibrium MD in both static and alternating electric fields has been applied extensively by Aksimentiev and co-researchers to investigate translocation and stretching of DNA through nanopore capacitor membranes . The voltage threshold required to induce the stretching and translocation through the membranes was studied in detail. In alternating field studies , it was shown that the hysteresis in the back-and-forth motion of the DNA sequences is sequence-specific. These exciting studies underline the many possibilities that molecular simulation has to offer to nanoscale device design.
Another challenging, but exciting, aspect of non-equilibrium simulation in electric and also magnetic fields is that of use as a predictive design tool for storage and controllable confinement of ionic species (e.g., ions, or perhaps polyectrolytes) in Paul and Penning traps. This cutting-edge area could have important implications for device design in quantum computation . In two dimensions, the few-body Coulomb problem can be studied using similar models in dusty plasmas . In these fields, experimental difficulties make molecular simulation highly valuable.
Computational terahertz spectroscopy is also an emerging area where both equilibrium [24,25] and direct non-equilibrium MD in e/m fields can be used to investigate spectroscopic response, leading to the advent of molecular simulation ‘experiments’ for complex, molecular systems. This has been used to compute the THz response behaviour of bulk water  and water molecules in the solvation layers of proteins .
An important issue for molecular simulation is the need for accurate interaction models that describe correctly the chemical response of the system to the electric or e/m field. Of particular interest is the treatment of polarisation forces and their interplay with other intramolecular (covalent) forces. Improving existing force fields appears to be necessity, which should be achieved with the help of dedicated electronic structure studies on the role of external e/m fields on chemical bonds. The polarisable TIP4P-FQ water model has been validated as being superior to some of the more popular, fixed charge water models in the presence of microwave and IR fields , but more refined polarisable potentials are now available, and reparameterisation work may yield better results in external fields. Ab initio simulation could be expected to give particularly valuable insights into these developments; the recent ab initio simulation of Choi et al  on electric field effects on water clusters is an important development in this respect.