Immersive virtual reality (IVR) offers significant transformative potential for science education by supporting learning experiences that deeply engage students and improve their understanding of scientific concepts. Despite considerable interest, research on the use of IVR in science education is still in its formative stage. Currently, there is a substantial gap in a tool that can help stakeholders evaluate key elements of immersive software for science education contexts. This research addresses this gap by conceptualising and applying a framework designed to assist educators, researchers, and designers in assessing essential components of an immersive science application. The framework highlights three key components: IVR technological affordances, the exploration of science within IVR, and scientific representations. These components are synthesised into the Immersive Representations Model (IRM). Employing screen capture methodology, we evaluated the application and significance of the IRM. This study pioneers a structured approach to evaluating immersive technologies in science education.