Cluster-plus-Glue-Atom Model For BCC Solid Solutions And Its Application In Composition Design Of Ti-Cr-V-Based Hydrogen Storage Alloys

Hydrogen energy has great potential to replace fossil energy in future. For its application, hydrogen storage materials is vital, in which solid solution alloys with BCC structure exhibit superiority in the theoretical capacity of hydrogen, characterized by good absorption-desorption dynamics and high security of storage. Nevertheless, multi-component alloying brings difficulties to the design of solid solution with high hydrogen storage properties. And thus composition design mostly depends on laborious experimental trials.Based on the short-range-order between atoms in solid solution alloy, the cluster model can give a good picture of the interaction of each element. So the cluster theory in the amorphous alloys applies to the BCC solid solution, and "Cluster-plus-Glue-Atom" model is similarly built up in which clusters are close-packed with glue atoms filling interstitials. In this paper, the "Cluster-plus-Glue-Atom" model for BCC solid solution is studied in details, in which the spatial distribution patterns of the clusters are discussed deeply, and then the cluster composition formula is given as:[cluster] (glue-atom)x. Practically, the occupying of different atoms is judged by the mixing enthalpy between elements, which finally moves on to the specific alloy composition. With this model, the structure analysis and composition design of the Ti-Cr-V-based solid solution alloy for hydrogen storage produce a series of alloys with both good performance and low cost. The main conclusions are as follows:(1) The values of x are determined directly by the spatial packing density of the clusters, which can be noted by vectors connecting clusters. Lattice position of Crystallographic direction families(?)<3 3 1>,(?)<420>,(?)<422>,(?)<333>,(?)<511> meets the neighbored cluster position described by spheric period distribution from the research of amorphous alloys, so these vectors are defined as basic position vector set to describe the distribution of clusters. Clusters can be packed in term of periodic primitive cells, whose basic vectors can be choosed from the basic position vector set. The size of unit cell corresponds to the density of the cluster packing, and the number of glue atoms increases with the density decreasing, similarly the number of cell configuration grows, as 1 (x= 0),1 (x= 1),1 (x= 2), 2 (x= 3),2 (x= 4),5 (x= 5),7 (x= 6). (2) Basic position vector set is applied to locate the neighboring clusters surrounding a certain cluster and also the concept of superclusters is induced to get a straight relevance of short-range-packing cluster. Superclusters are composed by vectors from basic position vector set, in a dense and homogenous way. All the possible results can be obtained by program operation, resulting in 21 superclusters capable of being extended to the whole BCC space. Some supercluster matches a certain cluster in the outer layer of a cluster group, and then new clusters from the very supercluster can be added in the group, by which bigger cluster groups are born. Programming according to this logic, mathematical programs can work out a series of second-level and third-level supercluster configuration. By scientific stastic analysis of all the third-level superclusters, clusters with glue atoms x=1.5 and 6 is figured out to have the most possibility in all ratios of cluster via glue atoms, while the result of simple supercluser classification statistics reveals two superclusters appeared most times in total, which are comprised of shorter vector ((?)<3 3 1>, (?)<4 2 0>) and longer vector ((?)<4 2 2>, (?)<5 1 1>) separately. The results are only based on the topological geometrical calculation so the analysis from the aspect of physical energy is still necessary in the future research.(3) Furthermore, the 1:1 cluster model [cluster](glue atom)1 is specially emphasized that guarantee to the maximum dispersion of the glue-cluster and hence is considered the most efficient configuration. We can use this model to design the low-Vanadium solid solution alloy for hydrogen storage. We design two cluster formula [Ti1(Cr8Ti6)]Vxand [Fe1(Ti8Cr6)]Vx in the alloy system Ti-Cr-V and Fe-Ti-Cr-V separately. Solid solution alloys with lowest Vanadium content Ti7Cr8V1 and Fe1Ti8Cr6V1 achieve good hydrogen storage properties, with maximum capacity of hydrogen both over 3.8wt%, and 1.8wt%,1.6wt% disorption at 333K separately. Alloying modification based on these cluster formula doesn't give any improvement to the properties, which proves that the two formula has already reached the limit.