基于视频运动显微技术的阻塞态颗粒和胶体体系微弱动力学研究

Granular and colloidal systems can be jammed into amorphous solids, and the jamming transition problem is an intriguing issue in the field of soft matter physics. The dynamics of a high-density jammed system may become too subtle to be observed by using any of the currently available experimental techniques. Noticing that some recently-developed computer vision methods are able to magnify tiny motions in videos, we propose to measure the subtle dynamics of jammed granular and colloidal systems by using such video-magnification methods. Video-magnification skills for the experimental systems will be developed, and more accurate measurements on dynamics are expected to be achieved. Then the subtle dynamics of high density systems will be measured, as well as their subtle response to very weak external perturbations. The related findings will help to physically distinguish high-density jammed systems from continuous media. The above technique may also allow the dynamics of deeply jammed states, which have been previously predicted by theory, to be experimentally measured for the first time, and further study on the deeply-to-marginally jammed state transition of the systems will be performed. This proposal aims to obtain a further understanding of the physical properties of jammed states and the mechanism of the jamming transition.

颗粒和胶体体系通过阻塞转变形成非晶固体，阻塞转变的物理机制是目前软物质研究领域备受关注的前沿问题。阻塞体系的动力学可能会随着密度的升高而变得极其微弱，以至于无法用现有实验方法进行观测。最新的计算机视觉方法可以放大视频中极细微的运动，本项目提出利用此视频运动显微技术可以研究阻塞态颗粒和胶体体系的微弱动力学行为。通过发展适用于实验体系的相应技术，将为阻塞态动力学研究提供更精细的实验表征手段，使得高密度体系的极弱动力学行为能够被观测，并允许探索体系对外部极弱扰动的响应特性，从而帮助认识高密度阻塞体系和连续介质的区别。对微弱动力学的观测，还有望首次从实验上验证理论预言的深度阻塞态动力学特性，进而研究体系从深度阻塞态到边缘阻塞态的转变规律。本项目将有望促进对阻塞态物理特性和阻塞转变本质的深入理解。

动力学变慢是胶体和颗粒等软物质体系中广泛存在的一类重要现象，无论是阻塞转变还是玻璃态转变都伴随着动力学的变慢。对于一些高密度的体系，其动力学有可能变得非常慢，以至于无法直接用光学成像的方法进行观测。利用最新的运动放大技术，本项目构建了多体体系的运动放大分析方法，并将其应用于对各种胶体和颗粒实验体系的研究，实现了对相关体系中粒子极微弱运动的定量观测。依赖于运动放大分析对微弱动力学的准确表征能力，高关联度的粒子运动轨迹能够被提取，使得对体系振动模式及其空间分布的计算变得可能，从而能够建立体系结构和动力学之间的关联。根据此方法和研究思路，我们有两点新的发现：1.高密度胶体体系中存在新颖的应力结构与低频准局域振动模式之间的关联；2.颗粒体系崩塌行为的关键要素可以从其结构的振动特性中预知。