While AI shows great potential in empowering physics experiments, the highly innovative and rapidly iterating nature of these experiments poses challenges for traditional Large Language Models (LLMs). Reasoning on specific methods typically requires full-context injection of local knowledge bases, resulting in excessive context consumption and high inference costs. This study proposes an intelligent agent framework for physics experiments integrating dynamic context-aware management and collaborative reasoning. Adaptable to existing LLMs, the framework dynamically extracts and injects high-value information into the context, significantly reducing inference costs and improving efficiency without compromising reasoning quality compared to full-context methods. By supporting rapid secondary development and diverse scenario adaptation, the framework effectively overcomes computational and cost bottlenecks, offering an efficient and flexible paradigm for the automation and intelligent transformation of physics experiments.

The oxygen adsorption can drastically alter the electronic properties of the two-dimensional (2D) materials, which is usually difficult to be removed. In this work, we report the ultraviolet (UV) illumination induced desorption of the O₂ molecules from the monolayer MoS₂ surface by using the atmosphere dependent transport measurement, Kelvin probe microscopy, photoluminescence spectroscopy and x-ray photoelectron spectroscopy. Obvious increasing of the conductivity, rising of the Fermi level, and red shift of the photoluminescence peaks of the MoS₂ were observed after the UV illumination in vacuum, indicating the elimination of the depletion effect from the oxygen adsorption. Such parameter changes can be reversibly recovered by the subsequent O₂ exposure. Furthermore, obvious decreasing of the oxygen concentration after the UV illumination was also observed by x-ray photoelectron spectroscopy. Thus the UV induced O₂ photodesorption effect is evidenced. The photo-excited charge transfer mechanism is proposed to account for the photodesorption effect.These results provide a nondestructive way to clean the MoS₂ surface and manipulate the performance of the MoS₂ based devices.