Natural materials are often more efficient and tend to have a wider range and combination of properties than do present-day engineered materials. Biological materials are composed from a limited set of components, but are able to achieve great diversity in their properties. The variation in properties is largely due to the different arrangements of the materials components, which form unique structures. We believe that there are underlying structural design principles, relating material structure to material properties, that commonly appear in biological materials. Because nature itself achieves highly effective design solutions, the utilization of these natural design principles could similarly improve the effectiveness of engineered materials. Materials scientists need a way to abstract relevant structural design principles from the myriad of biological materials articles for the development of bioinspired materials.
This research involves the development of a data mining tool that will quickly identify potential structural design principles of biological materials with respect to a chosen material property or combination of properties. This paper presents the first stage of this process: information retrieval. An algorithm is developed to extract structural design principles key terms and relevant passages for specified material properties from a corpus of materials journal articles. The development of this search tool is explained beginning with the determination of search term categories and appropriate search terms and continuing to the refinement of the program algorithm. An evaluation of the tool is also described comparing the programs results to those of a manual search for the structure-property relationships. The program identified 98% of the manually found structural design principle key terms, although many unanticipated passages were returned as well. Finally, the future work needed to improve the program is presented.