Size Matters: Free-Energy Calculations of Amino Acid Adsorption over Pristine Graphene.

There is still growing interest in graphene interactions with proteins, both for its possible biological applications and due to concerns over detrimental effects at the cellular level. As with any process involving proteins, an understanding of amino acid composition is desirable. In this work, we systematically studied the adsorption process of amino acids onto pristine graphene via rigorous free-energy calculations. We characterized the free energy, potential energy, and entropy of the adsorption of all proteinogenic amino acids. The energetic components were further separated into pair interaction contributions. A linear correlation was found between the free energy and the solvent accessible surface area change during adsorption (ΔSASAads) over pristine graphene and uncharged amino acids. Free energies over pristine graphene were compared with adsorption onto graphene oxide, finding an almost complete loss of the favorability of amino acid adsorption onto graphene. Finally, the correlation with ΔSASAads was used to successfully predict the free energy of adsorption of several penta-l-peptides in different structural states and sequences. Due to the relative ease of calculating the ΔSASAads compared to free-energy calculations, it could prove to be a cost-effective predictor of the free energy of adsorption for proteins onto nonpolar surfaces.

On the effects of induced polarizability at the water-graphene interface via classical charge-on-spring models.

Molecular models of the water-graphene interaction are essential to describe graphene in condensed phases. Different challenges are associated with the generation of these models, in particular π-π and dispersion interactions; thus quantum and classical models have been developed and due to the numerical efficiency of the latter, they have been extensively employed. In this work, we have systematically studied, via molecular dynamics, two polarizable graphene models, denominated CCCP and CCCPD, employing the charge-on-spring model of the GROMOS forcefield, both being compatible with the polarizable water models COS/G2 and COS/D2, respectively. These models were compared with non-polarizable graphene and SPC water models. We focused the study on the water-graphene interface in two distinct systems and under the influence of an electric field: one composed of graphene immersed in water and the other composed of graphene with a water droplet above it. In the former, the orientation of water close to the graphene layer is affected by polarizable graphene in comparison to non-polarizable graphene. This effect is emphasised when an electric field is applied. In the latter, carbon polarizability reduced water contact angles, but graphene retained its hydrophobicity and the computed angles are within the experimental data. Given the significant extra computational cost, the use of polarizable models instead of the traditional fixed-charged approach for the graphene-water interaction may be justified when polarizability effects are relevant, for example, when applying relatively strong fields or in very anisotropic systems, such as the vacuum-bulk interface, as these models are more responsive to such conditions.

Entropy deepens loading chemical potentials of small alcohols by narrow carbon nanotubes.

Small alcohol confinement within narrow carbon nanotubes has been extensively and systematically studied via rigorous free-energy calculations. Employing molecular dynamics simulations, thermodynamic integration and thermodynamic cycling, the loading process of methanol and ethanol from aqueous solution into (6,6), (7,7) and (8,8) single-walled carbon nanotubes was computed and decomposed into its entropic and energetic terms. For all tubes and alcohols, loading is favoured from infinite dilution in water; for the same alcohol, wider tubes allow for the formation of a collective dipole which is cooperative in terms of electrostatics and reduce the rotational freedom of the loaded particles; narrow tubes only permit the formation of dipole-dipole dimers instead, with a (rotational) entropic gain that compensates for the loss of long-range dipole-dipole interactions. The latter renders deeper loading chemical potentials for narrower tubes when partitioning small alcohols from aqueous solution and it is a clear example of an entropy-energy compensation phenomenon.