发表机构
ZJU-UIUC Institute, Zhejiang University; Department of Mechanical Science and Engineering, University of Illinois Urbana-Champaign; School of Materials Science & Engineering, Zhejiang University; School of Mechanical and Electrical Engineering, Liuzhou Polytechnic University; SVAC, Zhejiang Saiweike Optoelectronics Technology Co.（浙江大学中乌联合研究院; 伊利诺伊大学厄巴纳-香槟分校机械科学与工程系; 浙江大学材料科学与工程学院; 柳州职业技术大学机电工程学院; 浙江赛微科光电科技有限公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究利用大型语言模型智能体通过模式绑定控制磁控溅射系统，自动撰写X射线多层膜沉积配方，经反射率验证使相对周期差异从20%降至0.79%。
AI 中文摘要
用于X射线的周期多层膜要求层周期精确到约0.02纳米，该周期由实验室自身记录本中保存的沉积速率设定。一个大型语言模型智能体被赋予目标多层膜规格，并可读取实验室档案。沉积配方以磁控溅射系统的原生配方语言编写，并通过一个包含17个工具、分三层级的模型上下文协议桥提交，控制应用程序中执行强制约束。目标是一个30双层的Ru/C多层膜，周期为6.85纳米，选择该配对是因为每个周期由两层拟合而成。智能体只能在其自身的工作项目内写入，因此无法修改其读取的校准数据或用于评分的测量结果。多层膜通过X射线反射率与人工撰写的参考样品进行评分，相对周期差异的3%接受阈值在沉积前已固定。在四次评分的人工智能体撰写沉积中，相对周期差异从初始校准的20%下降至第一次测量反馈迭代后的2.26%，并在第二次迭代后降至0.79%。共识别出智能体面对接口的16种故障模式。结果仅限于所观察的周期、配对和腔室。
英文摘要
Periodic multilayer mirrors for X-rays require a layer period controlled to about 0.1 nm, set by deposition rates held in the laboratory's own notebook. A supervising model session gave a large language model agent a target multilayer specification and calibration constants derived from a laboratory archive. The agent authored recipes in the magnetron sputtering system's native recipe language and submitted them through a Model Context Protocol bridge of 17 tools in three tiers, with enforcement in the control application. The target was a 30-bilayer Ru/C multilayer with a period of 6.85 nm, a pair chosen because each period is fitted as two layers. The agent could read and write only its own working project, so it could alter neither the calibration record nor the measurements that scored it. We scored mirrors by X-ray reflectivity and fixed the 3% acceptance threshold on relative period difference before deposition. Across four scored depositions, the relative period difference fell from 20% on the calibration as given to 2.26% after one measured-feedback iteration and 0.79% after a second. We identified sixteen failure modes of the agent-facing interface. The result is bounded to the period, pair, and chamber observed.
