Electron cloud progress report - December 2004

Electron cloud formation and the associated effects of beam loss and instabilities via the strong coupling with the charged particle beam have been observed experimentally at several storage rings. The problem is of particular relevance to currently planned machines (or under construction), which typically make use of intense beams, closely spaced bunches and vacuum chambers with small transverse dimensions. The three main sources of primary electrons in an accelerator are: (1) photo-electric effect, (2) residual gas ionisation and (3) electrons produced by stray protons hitting the vacuum chamber at grazing angles. This primary electron cloud can then become significantly amplified by the process of secondary electron emission, which occurs when electrons strike the vacuum chamber. Under certain conditions of bunch intensity, chamber geometry, bunch fill pattern or longitudinal profile of the bunch and gap length, and secondary electron yield at the wall, a process of multipacting (cascade production of secondaries) can occur in the beampipe, with potentially detrimental effects on the beam stability. Crucial to the simulation studies of the electron cloud formation is a detailed knowledge of the secondary emission process from the chamber walls, which is usually described in terms of the secondary emission yield (SEY or δ ) and the emitted-energy spectrum of the secondary electrons (dδ/dE). The main purpose of the work described here is a comparison between several existing phenomenological models of the secondary electron emission process (and currently used in simulation codes), and the study of their effect on the dynamics of electron cloud formation and dissipation. A number of codes have been developed in laboratories worldwide, often based on alternative experimental inputs and with different underlying assumptions and physical models. As part of an inter-laboratory code-code and codeexperiment benchmarking effort that was re-launched after the ECLOUD’04 workshop [1], the present study explores the effect on simulation results of a reference case when using different descriptions of secondary emission properties for electrons impinging on the vacuum chamber walls. The two models currently used in the CERN-developed code ECLOUD [2, 3] and the American POSINST [4, 5] (and described respectively in [6] and [7]) have been compared, along with some tentative phenomenological fits of recent laboratory measurements obtained in surface science experiments realised at CERN [8].