The Research On Sphere Random Packings And Packing Structure

A lot of biomaterials and new functional materials are prepared by bonding and compressing particles and possess porous structure. In order to obtain the morphological characteristics which are necessary for functional materials, the formation mechanism of microstructure of materials, especially the relationship between particle feature and morphological characteristics of materials which are made of particles, must be understood. Then, the structure-property design of whole material can be implemented effectively. In the process of developing biomaterials such as artificial bone, because of lack of corresponding theories, a certain kind of the porous materials, which can be beneficial to the growth of histiocytie and capillary vessel and possess sufficient strength, can only be got on the basis of experience or by reiterative experiments. The research and development of materials repeatedly suggest that how to characterize the microstructrue of porous materials scientifically and affect the properties of materials seems apparently important. Due to complex microstructure of porous materials, a simulation method, which is implemented on computer on base of some models describing porous materials, has been an effective way. Sphere random packing, which may describing the matrix phase and pore, is a very useful model. Although the research on sphere random packings has a long history, there are some unsolved problems, for example a lack of perfect simulation algorithm and not many research on polydisperse sphere packings. Therefore, the deep research on sphere random packing is valuable for resolving some current questions.The computer simulation method was used to research sphere random packing in order to overcome the limitation of physical experiment. The simulation algorithm on computer is one of the important contents of sphere packings. It is a foundation to design an algorithm similar to the practical sphere packing. It is necessary to study a new parameter describing the information of spheres contacting an aim sphere to analysize packing structure of non-equal sphere packings. Polydisperse sphere packing is a key point in the paper because the non-equal sphere packing have attracted researchers’interests, however, some relevant study is rarely done. In the end, the discrete Boltzmann method is used to simulate flow in sphere packing.To obtain the simulation packing similar to physical packing, a new method is forwarded in this paper. Sphere packing is implemented by reducing paking space in the method. There are no overlaps among spheres and the sphere size doesn’t change during packing in contrast with current collective rearrangement models. The packing density is more than the density obtained by using sequential addition.A new parameter, coordination radius, is developed in the paper. Coordination radius is the ratio of mean radius of all spheres contacting aim sphere with the radius of aim sphere. The parameter can be used to describe the information of spheres contacting an aim sphere. The information is useful to research non-equal sphere packing. The relationship between coordination radius and the ratio of large sphere with small sphere in all spheres contacting aim sphere. As the research shows, the spheres contacting small aim spheres are mostly big ones and these spheres contacting big aim spheres are still big ones. When the sphere radius is log-normal distribution and normal distribution, the relative big sphere almost contacts relative big spheres, relative small sphere contacts relative big spheres.The packing structure of different packing density of uniform sphere is studied. There is a different packing structure in different packing density. The sphere packings are disorder when the density is less than 0.5. The packings have local order structure while the density is large.As the research shows, the packing density of non-equal sphere is more than the one of uniform sphere. For binary spheres, the ratio of big sphere with small sphere and the ratio of the number of big sphere with the number of small sphere affect the final packing density. When the ratio of the number of big sphere with the number of small sphere is constant, the packing density increases with the ratio of big sphere with small sphere. At fixed the ratio of big sphere with small sphere does not change, the packing density arrives at max value while the ratio of the number of big sphere with the number of small sphere is about 0.7. When sphere radius is log-normal and normal distribution, the packing density increases with coefficient of variation. The packing density of log-normal distribution is more than one of normal distribution at the same coefficient of uniformity. There are more relative small spheres when sphere radius is log-normal distribution.As the research shows, pore size distribution can be well approximated by the half normal distribution while the packing density is more than 0.5. For the packing of binary sphere, log-normal distribution and normal distribution, the pore size distribution can be well approximatedby the half normal distribution too.The significance of sphere random packing is as follows. (1)The research on packing algorithm is a foundation of deep studying packings structure. The method can provide reliable samples for researching structure. (2) The research on the relationship between sphere size and packing configuration with packing density and pore size distribution provides a basis for designing and control morphological parameters of materials. (3) On the basis of the algorithm, the research on packings of non-spherical particle can be implemented. (4) Some problems about opposite phase, such as the formation and properties of pore structure, can be studied according to the simulation algorithm and theory of sphere packings.