The 3He neutron spin filter at ILL

Abstract

Neutron Spin-Filters (NSF) using gaseous polarized 3He have long been recognized as having enormous potential value in many polarized neutron scattering applications and, accordingly, ILL started a development program some years ago. This report gives an account of the present status of the project. Let us first briefly recall the reasons why the 3He-NSF is desirable as a generic polarization tool: Being a pure transmission device, NSF naturally decouples the neutron spin selection from other optical functions. Designing polarized instruments will be simplified having no geometrical constraints due to Bragg optics or beam deviation in the polarizer. Angular acceptance and energy band pass are much widened, making polarization handling much more compatible with focusing geometry and time of flight techniques. Gaseous NSF naturally achieves well-defined transmission functions. Quantitative interpretation of intensity measurements in polarization analysis mode will be much easier. Very homogeneous polarization will be achieved on large beam areas, a feature which is particularly attractive for precision measurements. Fully polarized 3He is meant to be an ideally effective NSF with 50 percent transmission of the incident unpolarized beam. Already with half 3He polarization, it seems that NSF can improve some existing instruments when polarizing crystals or super mirrors are not good reflectors. To understand this point, it is important to define and calculate a relevant quality factor in order to properly handle the tradeoff in between neutron polarization and neutron flux. As an example, we have been able to show [I] that the inverse standard deviation in the measurement of small magnetic structure factors is proportional to PdT, P being the neutron beam polarization and T the filter transmission. At optimum gas pressure, this quantity turns to be linearly proportional to the 3He polarization in such a way that we can say that present NSF are only a factor of two away from ideal. Improving the 3He polarization remains most important for polarization analysis work, where this quality factor is generally squared.