Neutron Spin Echo Spectroscopy for the European Spallation Source
rHigh-resolution neutron spin echo (NSE) spectrometers will be adapted to and optimized for the parameters of the ESS long-pulse source drawing on the experience gained with the first NSE which is operated at a pulsed spallation source at the SNS in Oak Ridge, USA (figure 1). This spectrometer has been built and is operated now by experts from the Forschungszentrum Jülich.
NSE spectrometers are unique in their ability to extend the effective energy resolution of neutron spectroscopy significantly beyond the 1 μeV limit of backscattering or very cold TOF instruments. NSE extends this limit down to the neV range, however, as Fourier method it yields data that represent the intermediate scattering function, i.e. the Fourier transform S(Q,t) of the spectral function S(Q,ω). The relevant primary resolution parameter therefore is the Fourier time, which may extend into the μs regime.
Using the neutron spin precessing in a magnetic field as individual time keeper for each neutron allows the necessary sensitivity to minute energy resp. velocity changes upon scattering at the sample. This allows the use of incoming neutron beams with a much wider velocity distribution, thus preserving intensity.
Maximizing intensity is an important issue in order to arrive at a well performing instrument. The ESS with its envisaged high power and pulse structure will allow a step beyond the effective intensity of all other existing NSE instruments. However, careful optimizations are needed to fully exploit the ESS neutrons.
The most efficient way to transport the moderator brilliance to the sample including efficient and broadband polarization will be studied with the help of analytical methods and Monte-Carlo simulations. The neutron transport comprises a frame-overlap chopper system that has to be adopted to the chosen distance parameters (figure 2).
Fig.2: Tentative layout of a 4 chopper system to select wavelength frames without overlap as obtained by our chopper layout toolbox programs. C1..4 are the chopper positions (distance from moderator), the detector for this example is positioned at 50m. Chopper opening schemes are indicated by the broken horizontal lines. The green area show the paths accessible for neutrons, tested for the wavelength range λ=0…80Å. The four graphs show settings (by selecting different chopper phases) for the first 4 frames, thereby covering wavelengths up to about 20 Å. As is immediately visible the frames are free from contamination of any wrong wavelength below 80 Å.
This 4 chopper solution is designed such that the first chopper has a maximum distance from the moderator. If closer positions of the first chopper are acceptable also very good 2 chopper solution are possible. This example study was done for 16 Hz repetition frequency.
It is also planned to study new concepts for enhancing resolution by combining the optimum position of correction coils and a main coil design with minimum field inhomogeneity. The latter always requires a compromise between overall dimensions, potential beam divergence, minimum stray field and maximum field integral. The combination of magnetic simulation and neutron optics simulation will help to find the best solution possible at the ESS, taking into account the possible parameters of the source and the geometrical boundary conditions (space including shielding). In addition to maximum intensity and resolution due to a high field integral with little and correctable inhomogeneity in the beam area, minimization of the background due to the scattering from functional components in the beam is also an important objective of R&D work. Suitable “figures of merit” will be formulated and applied in order to optimize and compare the usability of the derived instrument concepts for known and foreseeable “typical” measuring problems in ultrahigh-resolution spectroscopy. In a later stage, an analogous consideration of a wide-angle spin echo spectrometer with high to medium resolution will be carried out.
In addition, the potential of other Larmor precession methods such as SESANS (Spin-Echo-Small-Angle-Neutron-Scattering) or SERGIS (Spin-Echo-Resolved-Grazing-Incidence-Scattering) as add-ons for reflectometers (SERGIS) under the boundary conditions at the ESS will be studied. These methods to not encode a change in velocity, but rather the scattering angle in spins. They will be mainly performed in the framework of the embedding instruments like SANS or reflectometry.