In past years advances in electronic document management have dramatically increased the availability and accessibility of scientific knowledge. Researchers are benefiting from faster access and a broader range of materials directly available to them. Despite these advances, however, there are still obstacles that researchers face when gathering and analyzing information from research articles and other relevant sources. Currently theoretical elements are usually embedded in research documents. This makes it difficult to search for specific theoretical elements or relationships across a larger set of sources. Since theories capture the essence of scientific knowledge a more effective representation of theories and its elements has the potential to improve the research process. This paper presents a framework based on the Theory Markup Language (TheoryML) with a goal to address this issue and describes the development of a prototype implementation of a repository for theories.