Mastering Physical Chemistry Learning with Large Language Models in the Classroom

In the learning process, asking questions is essential for acquiring knowledge and understanding complex concepts. However, students in Hong Kong often hesitate to engage in questioning due to the traditional "spoon-fed" education system. The introduction of large language models (LLMs) offers a transformative approach to this challenge. These AI-powered tools can generate responses to a wide range of queries, providing opportunities for interactive and engaging educational experiences, particularly in Chemistry. In subjects like physical chemistry, which require the integration of mathematical and physical concepts, LLMs can help students better understand knowledge and develop problem-solving skills.In this project, we plan to implement LLM-assisted learning activities tailored to address specific challenges in physical chemistry, such as understanding mathematical modeling, interpreting abstract concepts like thermodynamics, and solving complex numerical problems. These activities include group discussions where students interact with LLMs to explore controversial topics related to lecture content, evaluate AI-generated responses, and develop questioning techniques. Additionally, students will practice numerical exercises with built-in intentional errors, guided by LLM support, to develop their ability to identify and rectify mistakes.To address the lack of hands-on lab experience caused by the pandemic, LLMs will also be used to help students prepare for experiments and organize their lab reports. Before lab sessions, students can use LLMs to understand experiment objectives, procedures, and safety precautions. While writing lab reports, LLMs will assist students in organizing data and insights, encouraging them to verify information and think critically about their findings. To prevent over-reliance on LLMs and promote academic integrity, students will be required to validate AI-generated content and include their own insights. A scoring system will reward students' efforts in responsible AI use, such as verifying information, paraphrasing outputs, and reflecting on the capabilities and limitations of LLMs.  Specific metrics, such as pre- and post-assessments, self-reflection tools, and student feedback, will evaluate the integration of LLMs in improving learning outcomes. By fostering critical thinking and effective questioning, this project aims to bridge gaps in traditional education and prepare students for technology-enhanced academic and professional environments.