Amorphous silica (SiO2) is an inorganic material commonly used in semiconductor circuits to isolate different conducting regions. Due to its mechanical resistance, high dielectric strength, and selectivity for chemical modification, amorphous silica has also become a key material in microelectronics and chromatography. Because of its unique properties, silica is quintessential for a broad range of applications: chips, optical fibers, and telescope glasses are manufacture on silica. Furthermore, molecular biologists employ silica in resins and optical beads to study the biomacromolecules.

In recent years, the synergy between molecular biology and nanotechnology has opened up opportunities for many applications that involve macromolecules and silica, such as nanoelectronics, self-assembly of nanostructures, microfluidics, DNA microarray technology and nanopore sensors. The picture on the left side shows one of such nanodevices, a MOS nanopore manufactured on a poly-Silicon-Silica-Silicon membrane and a single-stranded DNA molecule (blue) translocating through it. It has been proposed that nanopores could be used to sequence DNA with single base resolution, leading to a fast and cheap DNA sequencing technology, which promises to have a enormous impact in life sciences and personal medicine. Therefore, an atomic level understanding of the interactions between biomolecules and silica is now central for further development of bionanotechnology applications.

Currently, no experimental technique is yet sensitive enough to resolve atomic-scale dynamics at the amorphous interface. Molecular dynamics simulations can be tailored to study those systems, becoming unique imaging tools. However, modeling systems that combine amorphous silica and biomolecules imposes a variety of challenges to modelers.

The first compulsory step to perform molecular dynamics simulations is to specify the coordinates and connectivity of each atom of the system. However, due to the amorphous nature of silica, there are no X-ray or NMR structures that can be used as templates. To obtain amorphous models, crystalline structures are randomized using a set of MD annealing steps that reproduce structural features of actual amorphous silica. The procedure to construct an amorphous silica nanodevice is squematically pictured in the figure on the right side and briefly described in the next paragraph.

First, a cristoballite unit cell is build and replicate it to fill a cubic volume big enough to contain the nanodevice. Then, the crystalline cube is annealed by increasing the temperature to 8000 K and then cooling it back slowly to 300 K. The initial high temperature assures the randomization of the crystalline structure. After that, the nanodevice is sculpted by removing atoms from the amorphous cube. To mimic real amorphous silica surfaces, the nanodevice is annealed further, allowing the rearrangement of the surface atoms. Finally, the bonding connectivity is calculated using a cutoff distance of 0.2 nm. The resulting structure can be refined to produce chemically-modified silica forms, by introducing hydroxyl groups or organic molecules at the exposed surface (blue atoms).

There has been extensive research on the rectification phenomenon in which case the asymmetric behavior has been attributed to various factors, such as the surface charge, the ionic double layer confined in the small nanopore space and the different electrostatic distribution due to structural/chemical asymmetry at the two openings. Several models to explain rectification have been proposed based on continuous macroscopic approaches, refined to include the peculiarities of the surface system, the latter represented as a continuous surface charge. However, the applicability of continuum descriptions at the scale of a few nanometers is a subject of debate, since continuous approaches do not take into account the discreteness of the solvent nor the influence of the surface topography.