Background Silica continues to represent an intriguing topic of fundamental and applied research across various scientific fields, from geology to physics, chemistry, cell biology, and particle toxicology. The pathogenic activity of silica is variable, depending on the physico-chemical features of the particles. In the last 50 years, crystallinity and capacity to generate free radicals have been recognized as relevant features for silica toxicity. The ‘surface’ also plays an important role in silica toxicity, but this term has often been used in a very general way, without defining which properties of the surface are actually driving toxicity. How the chemical features (e.g., silanols and siloxanes) and configuration of the silica surface can trigger toxic responses remains incompletely understood. Main body Recent developments in surface chemistry, cell biology and toxicology provide new avenues to improve our understanding of the molecular mechanisms of the adverse responses to silica particles. New physico-chemical methods can finely characterize and quantify silanols at the surface of silica particles. Advanced computational modelling and atomic force microscopy offer unique opportunities to explore the intimate interactions between silica surface and membrane models or cells. In recent years, interdisciplinary research, using these tools, has built increasing evidence that surface silanols are critical determinants of the interaction between silica particles and biomolecules, membranes, cell systems, or animal models. It also has become clear that silanol configuration, and eventually biological responses, can be affected by impurities within the crystal structure, or coatings covering the particle surface. The discovery of new molecular targets of crystalline as well as amorphous silica particles in the immune system and in epithelial lung cells represents new possible toxicity pathways. Cellular recognition systems that detect specific features of the surface of silica particles have been identified. Conclusions Interdisciplinary research bridging surface chemistry to toxicology is progressively solving the puzzling issue of the variable toxicity of silica. Further interdisciplinary research is ongoing to elucidate the intimate mechanisms of silica pathogenicity, to possibly mitigate or reduce surface reactivity.

Amorphous silica is widely employed in pharmaceutical formulations both as a tableting, anticaking agent and as a drug delivery system, whereas MCM-41 mesoporous silica has been recently proposed as an efficient support for the controlled release of drugs. Notwithstanding the relevance of this topic, the atomistic details about the specific interactions between the surfaces of the above materials and drugs and the energetic of adsorption are almost unknown. In this work, we resort to a computational ab initio approach, based on periodic Density Functional Theory (DFT), to study the adsorption behavior of two popular drugs (aspirin and ibuprofen) on two models of an amorphous silica surface characterized by different hydrophilic/hydrophobic properties due to different SiOH surface groups' density. Particular effort was devoted to understand the role of dispersive (vdW) interactions in the adsorption mechanism and their interplay with H-bond interactions. On the hydrophilic silica surface, the H-bond pattern of the Si-OH groups rearranges to comply with the formation of new H-bond interactions triggered by the adsorbed drug. The interaction energy of ibuprofen with the hydrophilic model of the silica surface is computed to be very close to the sublimation energy of the ibuprofen molecular crystal, accounting for the experimental evidence of ibuprofen crystal amorphization induced by the contact with the mesoporous silica material. For both surface models, dispersion interactions play a crucial role in dictating the features of the drug/silica system, and they become dominant for the hydrophobic surface. It was proved that a competition may exist between directional H-bonds and nonspecific dispersion driven interactions, with important structural and energetic consequences for the adsorption. The results of this work emphasize the inadequacy of plain DFT methods to model adsorption processes involving inorganic surfaces and drugs of moderate size, due to the missing term accounting for London dispersion interactions.

The atomistic details of the interaction between ibuprofen (one of the most common nonsteroidal anti-inflammatory drugs) and a realistic model of MCM-41 (one of the most studied mesoporous silica materials for drug delivery) were elucidated by quantum mechanical modeling inclusive of London forces. Calculations are based on periodic density functional theory adopting all-electron Gaussian-type basis functions of polarized double-ζ quality and the B3LYP hybrid functional. By docking the drug on different sites of the MCM-41 pore walls, we have sampled different local features of the potential energy surface of the drug–silica system, both for low and high loadings (one and seven drug molecules per unit cell, respectively). For all cases, ibuprofen adsorption in MCM-41 is exothermic (average ΔH = −99 kJ·mol–1) and exergonic (average ΔG = −33 kJ·mol–1), exclusively when London interactions are taken into account due to their dominant role in dictating all features of this system. The comparison between simulated IR and NMR spectra suggests that static disorder of the adsorbed ibuprofen due to surface sites heterogeneity can also be invoked together with the current interpretation based on a dynamic behavior of the adsorbed ibuprofen to interpret the spectral features. Analysis of H-bond patterns exhibited by the drug interacting with the MCM-41 surface silanol (SiOH) groups revealed the importance of cooperativity in the H-bond strength. The present work shows that large-scale all-electron full quantum mechanical simulations employing accurate hybrid functionals can soon become competitive over modeling studies based on molecular mechanics methods, both in terms of superior accuracy and absence of the problematic parametrization, due to organic/inorganic interface

No abstract

Fully ab initio treatment of complex solid systems needs computational software which is able to efficiently take advantage of the growing power of high performance computing (HPC) architectures. Recent improvements in CRYSTAL, a periodic ab initio code that uses a Gaussian basis set, allows treatment of very large unit cells for crystalline systems on HPC architectures with high parallel efficiency in terms of running time and memory requirements. The latter is a crucial point, due to the trend toward architectures relying on a very high number of cores with associated relatively low memory availability. An exhaustive performance analysis shows that density functional calculations, based on a hybrid functional, of low‐symmetry systems containing up to 100,000 atomic orbitals and 8000 atoms are feasible on the most advanced HPC architectures available to European researchers today, using thousands of processors. © 2012 Wiley Periodicals, Inc.