22nd SmoleQ Greeting Seminar (2026.8.7 at 17:00 (JST))

 

Speaker: 

Veronika Zadin

University of Tartu, Institute of Technology MATTER Lab (https://matter.ee/ )

Title: 

Multi-scale and multi-physics studies of nanoscale field emitters in CERN particle accelerators

Abstract:

Electric fields on the order of 100 MV/m or higher are common in particle accelerators, high-field devices, electron sources, and vacuum switches. These systems are often limited by vacuum breakdown: a spontaneous electrical discharge that is difficult to predict. One proposed initiation mechanism is the formation of field-enhancing protrusions or tips. Field-emission measurements suggest that such protrusions have diameters of several tens of nanometers and aspect ratios of up to 100. However, they have not yet been directly observed because of their nanoscale dimensions, macroscopic electrode surfaces and their rare enough appearance. One of the most promising theories attributes their formation to the electric-field-assisted biased diffusion. In this mechanism, the electric field acts on surface atoms not only through the normal Maxwell stress but also tangentially, owing to variations in polarizability determined by the local atomic environment.

To study atomic dynamics and tip formation at nanoscale, we have developed FEMOCS multi-physics multi-scale simulation framework. FEMOCS describes the material using molecular dynamics, evaluates the electric field using finite element method, calculates the emission currents and tracks the material heating by incorporating both Joule and Nottingham heating effects. It incorporates PIC based plasma model to ensure accurate space charge sheet formation and plasma-surface interaction in early stages of vacuum breakdown. The simulation framework is fully coupled and enables us to study the field emitter dynamics and stability even under complete melting conditions and ion bombardment form plasma.

Current talk focuses on providing overview of vacuum breakdown process, introduces FEMOCS together with its capacities for studying field emitters and early-stage breakdown and introduces machine learning methods for tip formation under field due to the electric field biased diffusion.