Developing Ai-driven Dynamic Models of Metallic Bonding in Teacher Education Programs

Abstract Book of the 9th World Conference on Research in Teaching and Education

Year: 2025

[PDF]

Developing Ai-driven Dynamic Models of Metallic Bonding in Teacher Education Programs

Juhye Park, Seounghey Paik

ABSTRACT:

Modeling is widely recognized as a core pedagogical practice for advancing conceptual understanding, scientific reasoning, and meta-modeling competence in science education. The electron sea model, however, remains the prevailing representation of metallic bonding in high school chemistry despite its significant limitations: it reduces electrons to particles, overlooks their wave nature, and neglects electrostatic interactions. These simplifications constrain both students’ and teachers’ understanding of metallic properties, such as conductivity, luster, and malleability. To address these issues, this study was conducted with in-service chemistry teachers enrolled in a graduate science education program. The teachers critically analyzed the canonical model’s shortcomings and, using artificial intelligence (AI)–driven simulation tools, developed dynamic models that incorporated alternative conceptual aspects, including the wave behavior and delocalization of electrons. Through this process, they explicitly identified the ignorance embedded in the traditional model and explored new perspectives for interpreting metallic properties more effectively. The findings highlight how embedding AI-based modeling into teacher education can strengthen teachers’ conceptual knowledge, meta-modeling competence, and technological literacy. Furthermore, this work suggests implications for designing teacher professional development programs that integrate educational technology, foster critical engagement with scientific models, and ultimately enrich classroom practice.

Keywords: Artificial Intelligence; Educational Technology; Meta-Modeling Knowledge; Professional Development; Science Education