Search

One-dimensional nanomaterials receive much attention thanks to their advantageous properties compared to simple, bulk materials. A particular application of 1D nanomaterials is photocatalytic hydrogen generation from water. Such materials are studied not only experimentally, but also computationally. The bottleneck in computations is insufficient computational power to access realistic systems, especially with water or another adsorbed species, using computationally expensive methods, such as ab initio MD. Still, such calculations are necessary for an in-depth understanding of many processes, while the available approximations and simplifications are either not precise or system-dependent. Two-dimensional models as an approximation for TiO2 nanotubes with (101) and (001) structures were proposed by our group for the first time in Comput. Condens. Matter journal in 2018. They were developed at the inexpensive DFT theory level. The principle was to adopt lattice constants from an NT with a specific diameter and keep them fixed in the 2D model optimization, with geometry modifications for one of the models. Our previous work was limited to studying one configuration of a nanotube per 2D model. In this article one of the models was chosen and tested for four different configurations of TiO2 nanotubes: (101) (n,0), (101) (0,n), (001) (n,0), and (001) (0,n). All of them are 6-layered and have rectangular unit cells of tetragonal anatase form. Results of the current study show that the proposed 2D model is indeed universally applicable for different nanotube configurations so that it can be useful in facilitating computationally costly calculations of large systems with adsorbates. [ABSTRACT FROM AUTHOR]

• 1D nanotubes can be modelled with 2D periodic model approximation. • The 2D model is built at moderately costly method (DFT). • The model can be used with more costly methods, such as ab initio MD. • Nanotube curvature can be partially preserved. Currently a lot of attention is paid to 1D nanomaterials due to their advantages in comparison to bulk materials. They offer broad possibilities of application, including photocatalytic water splitting. Simulations of water adsorption on such materials with computationally costly theoretical methods, such as ab initio molecular dynamics, are needed for improvement of such photocatalysts' efficiency. Still, it is very problematic to treat a real-size nanotube at available computational power. The existing nanotube surface approximations are not accurate and universal enough. We have already proposed methods for 2D model construction out of TiO 2 nanotubes of (101) and (001) configuration at the moderately expensive DFT level. The idea behind was to provide a partial description of nanotubular strain by applying lattice constants from nanotubes to slab models, and preserving geometry motifs. We use water adsorption energy, valence band maximum and conduction band minimum positions, as well as DOS shape as criteria for model validation. Our previous work was limited only to specific variants of nanotubes and water adsorption. In this work we establish these novel approaches along a wide nanotube diameter range, in particular for water adsorption studies. We demonstrate that the 2D models do not impose critical compromises in terms of accuracy, and therefore allow calculations of much larger nanotubes than common approaches do. [ABSTRACT FROM AUTHOR]