Computational Design Of Several New Structural And Functional Materials Using Crystal Structure Prediction

Materials innovation is important for modern technology and industry.With the development of computational methods for predicting materials properties,computational materials discovery has emerged,and is now one of the main part of Materials Genome Initiative（MGI）project as first launched by the United States and then followed by other countries.The main goal of MGI project is to speed up the discovery and application of new structural and functional materials.Choosing elastic moduli,Vickers hardness and permittivity as target properties,by utilizing powerful evolutionary algorithm USPEX,combined with first-principles calculations,this thesis aims to discover new high-temperature transition metal carbides,high-hardness transition metal borides and low-permittivity inorganic materials.Design of new high-performace high-temperature structural ceramics are important for the development of hypersonic vehicles.By performing evolutionary crystal structure prediction,we first explore thermodynamically stable structures for nine kinds of transition metal（from group IVB to VIB）carbides.Several new carbides have been predicted.We then study mechanical properties including elastic moduli,Pugh’s ratio and Vickers hardness for all those known and the newly predicted transition metal carbides.Then,we investigate effects of chemical composition and crystal structure on the mechanical properties.Especially,we find that complex carbides with distorted metal sublattice have better elastic moduli and hardness than those of interstitial carbides.We also find that the distribution of carbon vacancies in interstitial carbides has great effect on the mechanical properties;both non-adajacent and grouped carbon vacancies are detrimental to the elastic moduli and hardness,but the latter are helpful to improve toughness.These results help guiding the design of new high-temperature ceramics.Discovery of materials with（ultra-）high hardness are very important.In this thesis,we focus on discovering transition metal borides with high hardness.Firstly,we suggest a reasonable way to check Vickers hardness of transition metal borides;for a given transition metal boride,it is likely to be a high hardness material if its simulated Vickers hardness（by Chen-Niu model）and ideal strength are both high.Then,we study effects of boron content and topology of the boron sublattice on Vickers hardness of a number of hafnium borides and reveal the common structural features in high hardness transition borides:high boron content（no less than that of TMB₂）,lack of TM-TM layers,with 2D boron layers（graphene-like and corrugated layers）or 3D boron frameworks.Based on these structural features,we select numerous known transition borides and analyze their Vickers hardness.A number of high hardness transition borides are suggested.Design of reliable low-permittivity materials for modern microelectronics industry is a very difficult task.The present widely used low-permittivity organic and hybrid organic-inorganic materials in ultra-high scale integration circuits all have permittivity around 2.5,while there is no solution to find applicable materials with permittivity approaching 2.0 or lower,especially for inorganic materials.In this thesis,we develop a scheme for rational design of inorganic materials with anticipated permittivity.We first put forward a semiempirical method for quickly evaluating the permittivity of materials.We then propose several effective rules for the design of low-permittivity inorganic materials:to design compounds composed of coordination polyhedra with low coordination number,low electronic polarizibility,low ionic oscillator strength,and high volume.Based on these rules,we design several new low-permittivity inorganic materials such as Mg F₂,Be F₂ and Si OF compounds with permittivity in the range of 2.5-3.0.The present research would be help for the design of other type of structural and functional materials.