Modelling of acoustic wave Propagation through cementitious materials

Abstract

Osteoporosis constitutes a world-wide health problem that affects mechanically integrity of bones structure leading to unexpected fractures, mainly in the spine. A common clinical procedure to treat these fractures is vertebroplasty, where a cement material is injected trough a biopsy needle into the damaged vertebra. Mechanical characteristics of bone cement can be obtained with a difficult and expensive experimental process. The ultrasounds technique is one of the methods used to determine these properties. Nevertheless, the measurements realization requires expensive materials and equipment. Finite element analysis can characterize materials from a theoretical manner and it is possible to study the material behaviour during its hardening process. Others parameters can be changed to improve the connexion with the experimental data. The purpose of this project is to approach an ultrasonic technique through a computed process with the help of Comsol Multiphysics. Two methods of calculation were used to compare the experimental and computed results. The results are relatively mixed: According to the results obtained by the first calculation method, this one appears no appropriate and no conclusive since they did not concur with the experimental ones. Indeed, human errors can be promoted because of the numerous steps required. It takes much time, between 2 and 3 days to treat data, without taking in account the time used for realizing one simulation (between 5 and 10 days). So, a second calculation method was initiated. The second calculation approach was similar to the first one, but easier, quicker and provided relatively good results -a graph given by experimentation could be represented/approached by the software - . Therefore, this method seems to answer to the will of approaching experimental results with the use of software. To conclude, it would be pertinent and useful to refine this method through further studies, in order to reduce errors and increase the agreement between reality and modelling.