High-throughput Structure Search Of Two-dimensional Titanium Boride And Theoretical Study Of Its Superconducting Properties

In recent years,two-dimensional(2D)materials have attracted much attention because of their novel structure and excellent performance.The electron deficiency makes boron materials unable to stabilized through the classic 2c-2e bond.Boron compounds tend to form multi-centre bonds to meet the eight-electron rule,which leads to its rich structural diversity.Theoretical and experimental results show that the two-dimensional boron material can be stabilized on the metal substrate.The structures of borophenes are the triangular lattice with a certain proportion of hexagonal vacancies,and the specific distribution of vacancies influences the properties of the system.When boron and transition metal atoms form two-dimensional compounds,the structure and property of the compounds are related to the metal atoms.Consequently,the two-dimensional transition metal boron compounds have good application prospects and can be used as a candidate system to study new physical phenomena.Structural diversity is the source of rich properties of boron compounds,but it also makes it difficult to predict the configuration of boron compounds theoretically and synthesize them experimentally.The potential energy surface of boron compounds has a large number of local minimum values,which increases the difficulty of structure search.The existence of multi-centre bonds makes the potential function hard to construct and lack an efficient way to cal-culate the energy of compounds.Taking the two-dimensional titanium boride structure as an example,the high-throughput first-principles calculation based on structure recognition is car-ried out in this dissertation.We apply the biased screening method by using prior knowledge of the structures,and structural stability and properties of 2D titanium boride are studied.We construct a series of initial configurations with structure generating method.The initial configuration can be distributed uniformly around the local minimum potential energy surface.The efficiency and accuracy are better than the unbiased screening method.In two-dimensional titanium boride,boron atoms tend to maintain a planar triangular grid.Different coordination numbers can appear in titanium atoms by introducing polygonal vacancies.In a given super-cell,a variety of reasonable initial configurations were considered by removing the duplicated structures.We release the open-source package Structures of Alloy Generation And Recognition(SAGAR)and own its software patent.The structural stability of 2D titanium boride was systematically studied and the calcula-tions and results are managed by using Aii DA.We confirm the previously reported titanium boride monolayers and find that the TiB₇monolayer with space group(Cmmm)have lower en-ergy.We emphasize that by adding prior structural knowledge,comparing to the global search algorithm the total number of structures generated and calculated is greatly reduced.The biased screening method has simple logic and is easier to combine with workflow tools for calculation management.The superconductivity of TiB₇,TiB₉monolayer and the AB-stacking TiB₇bilayer are cal-culated and analyzed.The monolayer Tib₇has the largest superconducting transition tempera-ture,8.3 K,and the superconducting transition temperature decreases with the increase of tensile stress.The superconducting temperature of AB-stacking TiB₇bilayer and TiB₉monolayer are5.5 K and 1.2 K,respectively.In searching two-dimensional titanium boride structures,we use the prior knowledge re-lated to materials and find that a biased screening method can effectively reduce the configu-ration search space,which saves the cost of computation remarkably.The efficient duplicated periodic structure removing method can extend to systems with specific structure patterns.It is easy to be combined with the high throughput workflow tools to calculate and analyze other systems.