Hyspec Instrument Components

• Overview
  • Instrument layout
  • Operational Quirks
• Components for the Primary Spectrometer, conventional direct geometry
  • Neutron Guide
  • T₀ chopper and curved guide
  • Fermi Chopper
  • Disc Choppers
  • Fixed Beam Monitor
• “Hybrid” Components of HYSPEC, Focusing arrays to sample
  • Pyrolitic Graphite Focusing Array
  • Heusler Focusing Array
  • Drum shield & Swinging Arm
  • Optic Rail
  • Shield box
  • Motorized Slits
  • Manual Slits
  • Manual Attenuators
  • Optional Beam Monitor
• Sample
  • Sample Environment
  • Sample Positioning for bottom loading systems
  • Sample Positioning for top-loading systems
  • 3D Coils
  • Unpolarized standard samples
  • Polarized standard samples
• Components: Final Flight Path
  • Oscillating Radial Collimator
  • Polarizing Supermirror array
  • Argon Flight Path
  • Detectors
• Polarization specific components
  • Guide fields
  • Mezei flipper pre-sample
  • RF flipper pre-sample
• Polarization modes
  • Quirks of polarization analysis: lower statistics due to attenuation of polarization filter optics
  • Depolarization, direct beam
  • Half polarized
  • Vertical-only, full polarization analysis
  • Separation of Spin-Incoherent scattering with vertical full polarization analysis
  • 3D polarization analysis

Overview

HYSPEC is shorthand for “Hybrid Spectrometer”, combining a fully functional direct geometry spectrometer with features found on triple axis spectrometers. The dilemmas solved by incorporating triple axis features are (1) delivering high flux at sample with high vertical divergence, (2) providing a straightforward mechanism for changing between unpolarized and polarized incident beam, and (3) improving access at and around the sample position, compared to pit-based systems. We achieve high flux at sample by emplying Bragg vertical focusing, and by employing a vertical guide trumpet to change from a moderate-size high divergence at the moderator, to a high-size lower divergence beam at the focusing array (the lower divergence improves transport). Changing between unpolarized and polarized incident beam is achieved by switching between the pyrolitic graphite (unpolarized) array and the Heusler (polarized) array. High-level specifications for HYSPEC are as follows:

Instrument layout

The table below presents distances between critical components of HYSPEC. The values L1, L2 and L3 are used along with moderator time profiles and Fermi chopper frequencies, to estimate energy (transfer) resolution.

Operational Quirks

Vertical focusing impacts both slit settings and Bragg peak structure on detector view

Because HYSPEC employs Bragg optics for focusing, we experience a W configuration effect, similar to what is observed on TAX’systems. This means that for positive detector vessel S2 angles, signal is reduced and Q resolution is improved.

Ikeda-Carpenter tail restricted by Bragg focusing optics

Due to the use of many components which ‘feel’ like TAX components, the HYSPEC instrument adopts similar nomenclature for a variety of motor systems. For example, M1 and M2 refer to the focusing array and drum shield rotation angle about the vertical axis, respectively. Similarly, S1 refers to one of the sample vertical axis rotation stages (one for bottom loading sample environments), and S2 refers to the middle of the scatter direction range of the detector vessel.

Unlike the other DGS instruments at the SNS, HYSPEC cannot reconfigure for a ‘white beam’ or ‘quasi-white beam’ configuration, due to the use of a focusing array, the limitation of the Fermi chopper which cannot stop in the open position, and a safety-motivated block on the 0 deg orientation of the drum shield.

For unpolarized measurements we do not employ a per-pixel normalization using vanadium. Due to the low vertical divergence, the detector response is remakably uniform across the height of the LPSD’s. When using the polarizing supermirror array, we employ per-pixel normalztion with TiZr in order to correct for variations in transmission of the polarizing supermirror array.

Components for the Primary Spectrometer, conventional direct geometry

Neutron Guide

The neutron guide reflects lower energy neutrons providing higher flux, at the expense of beam divergence, at the sample position. It starts in the Shutter and is trumpeted vertically, increasing the beam height from 13 to 15 cm, but lowering the divergence. Supermirror guide coatings are m=3 except for the inner curved surface which are m=2.

T₀ chopper and curved guide

The purpose of the T₀ chopper is to suppress the prompt pulse of fast neutrons produced when the proton beam strikes the target. This suppression is accomplished by having an ~0.20-m-thick piece of the alloy inconel in the beam when the proton pulse hits the target. This piece of inconel must be out of the beam in sufficient time for the 0.01 - 1 eV neutrons to pass. It only turns counter clockwise and can operate at rotational speeds between 30 Hz and 60 Hz in multiples of 30 Hz. The high level system for setting incident energy selects the appropriate rotational speed.

The guide system is curved as well to move the guided beam out of line-of-sight with respect to the moderator, also to suppress the prompt pulse of fast neutrons produced when the proton beam strikes the target.

Fermi Chopper

The Fermi chopper is the primary energy selection device on HYSPEC. It consists of a series of closely spaced neutron-absorbing blades (slit package) held together by a rotor that spins about a vertical axis in the path of the beam. The slit package is straight and short, which works well with the cold neutron spectra of HYSPEC. The slit package can be spun from 30 to 420 Hz in increments of 30 Hz, providing a trade-off between flux and resolution. The Table below provides the dimensions of the Fermi chopper package. When decreasing the rotation speed of the chopper, the high level system for setting incident energy uses increments of 60 Hz for safe and controlled speed changes. At 180 Hz the time burst has a FWHM of 50 /mu s. Your local contact can assist you in choosing the appropriate rotational speed for your experiment.

Disc Choppers

Both Disc Choppers operate at 60 Hz, but perform different functions. The upstream ‘Frame Overlap’ Chopper is positioned and sized to block neutrons from the previous spallation event from arriving at the Fermi chopper when open. The downstream ‘Order Suppression’ Chopper is positioned just before the Fermi chopper and, even when the Fermi chopper is at full speed at 420 Hz, blocks neutrons from accessing the Fermi chopper at openings which straddle in time the desired opening.

Fixed Beam Monitor

The first of two neutron beam monitors is located just downstream of the Fermi chopper. These two monitors are not used to determine the speed of the incident neutrons. To detect the neutrons, the beam monitors use a low pressure of ³He.

“Hybrid” Components of HYSPEC, Focusing arrays to sample

Pyrolitic Graphite Focusing Array

The HOPG array has a 1.2 deg FWHM mosaic and is employed for unpolarized measurements.

Heusler Focusing Array

The Heusler array has a 0.5 deg FWHM mosaic and is employed for polarized measurements. Exchange with the PG focusing array is achieved via a motorized elevator inside the drum shield.

Drum shield & Swinging Arm

The drum shield houses the vertical focusing arrays, and employs a simplified shielding system compared to most TAX drum shields. There is a large angular opening which accepts the incident beam, and an exit port with permanent magnet guide field and tertiary shutter. The tertiary shutter closes to protect users onsite from gamma rays from the activated Heusler array. The large angular opening is exposed to the experiment room floor at Ei > 7 meV, so a swinging arm shield closes to block that opening. This swinging arm is manually actuated onsite.

Optic Rail

HYSPEC employs an X-95 standard optic rail, with center _ mm above beam elevation. Several rails with different lengths and end-shapes are used for different sample environment and polarization configurations. The components which can be optionally mounted on this rail are listed here but described in more detail below.

Shield box

This box encloses the optic rail between Drum shield and sample for most configurations of HYSPEC. It has an inner liner of 1.5 mm thick Cd sheet on all sides but bottom, and the outside is coated with BN paint.

Motorized Slits

There are two sets of beam defining slits which may be mounted on the optic rail between drum shield and sample. Typically only the second set of slits is adjusted to limit the beam on the sample. Each set of slits consists of four hot pressed B₄C blades (74 – 77% B, 23 – 26% C)

Manual Slits

There are two holders of fixed beam-defining slits which may be mounted on the optic rail between drum shield and sample. These are used in place of the motorized slits for large uncompensated magnets which may interfere with the operation of slit motors, or for polarization modes when space for guide field and flipper components take up too much space along the rails (and then the motorized slits run out of room). The downstream slit is generally positioned just outside and downstream of the ‘shield box’ when in use.

Manual Attenuators

Two attenuator plates, located between the Fermi chopper and the variable aperture, may be independently placed in the incident beam. Each plate is 1 mm thick and is 2% borated aluminum.

Optional Beam Monitor

The second (rarely used) of two beam monitors can be mounted between the drum shield and the sample. These two monitors are not used to determine the speed of the incident neutrons. To detect the neutrons, the beam monitors use a low pressure of ³He.

Sample

Sample Environment

Details of sample environment mounting are provided on the sample environment website

Sample Positioning for bottom loading systems

Sample Positioning for top-loading systems

A separate ‘stick rotation’ motor is employed for rotation about the vertical axis, and is either Axis1 or Axis2 ; for autoreduction this motor needs to be mapped to the ‘Omega’ value. Tilt motors SGU and SGL are not used, but STU and STL are horizontal translation motors which may be used to ensure the sample is centered on the beam and visible through either radial collimator or supermirror array. S1 is generally not used and is best disabled during stick-rotation operation.

3D Coils

Unpolarized standard samples

We use a vanadium rod, 6.35 mm diameter, ~50 mm tall. We also have access to several Vanadium annuli which are sometimes employed to establish absolute normalization. Finally we also employ Alumina powder, mainly to establish S2 clocking.

Polarized standard samples

Our workhorse is the 12.7 mm diameter TiZr, which exhibits isotope-incoherent scattering.

Components: Final Flight Path

Oscillating Radial Collimator

The fine radial collimator overfills the detector bank acceptance, and has gadolinium oxide coated panes that span between 550 to 750 mm radii from the sample, with 40’ between panes.

Polarizing Supermirror array

Designed and constructed at the Paul-Scherrer Institut, an array of polarizing remanent supermirrors, with an m=3 polarizer coating (FeCoV) on both sides, optionally replaces the fine radial collimator via manual operation of an elevator on the detector vessel. The array has 960 supermirrors distributed over 60^o and employs a magnetic holding field of ≈60 Gauss. Final energy range is 3.8-25 meV.

Argon Flight Path

The detectors are located outside a flight path chamber that is back filled with helium. To reduce background, there are two thin aluminum windows between the sample and the detectors at rough radii of 0.79 m and 4.45 m.

Detectors

The detector array is an assembly of 1.2 m long by 25 mm diameter Linear Position Sensitive Detectors (LPSDs). The array has 160 detectors grouped into packs of 8 and located on four flat panels with a minimum radius of 4.5 m in the horizontal plane. The LPSDs are filled with ³He at a pressure of 10 Atmospheres (1.0 MPa).

An incoming neutron is converted through the nuclear reaction n⁰ + ³He → ³H + ¹H + 0.764 MeV into charged particles tritium (T or ³H) and protium (p or ¹H) which then are detected by creating a charge cloud in the stopping surrounding gas. The electrons from the ionized gas are collected at an anode wire running down the center of the detector tube. This wire is at ~1900 V above ground and has a high resistance so the proportion of charge seen at each end allows one to determine the position of the neutron detection event. The length of the detectors is divided into 128 pixels of ~1. mm length by the electronics. The pixels are binned together during reduction for *.nxspe files; discuss the current binning with your Local Contact. Each pixel has a timing resolution of 0.1 µs and saturates at no less than 70,000 n/s, but this is not a time-averaged rate. After saturation a tube is ready for measurement within 10 µs. Saturation is likely when observing direct beam or a strong Bragg peak at the elastic line, because the neutrons all arrive at about the same time within ~50 µs, 60 times each second. Therefore, over a Bragg peak covering ~4 tubes we expect saturation above count rates of ~600 cps.

Polarization specific components

Already described above are the optional Heusler focusing array, the 3D coil systems at sample position, and the optional polarizing supermirror array for the scattered beam. Here we also include additional components used specifically with polarization analysis

Guide fields

Mezei flipper pre-sample

RF flipper pre-sample

Polarization modes

Quirks of polarization analysis: lower statistics due to attenuation of polarization filter optics

We do take a hit in flux due to both filters, beyond the idealized 50%. To give you and idea of how much we lose, see the image below. What you are used to is the ‘PG Rad NoFlip’. The impact of the polarizer alone is ‘Hu Rad NoFlip’ and is used for half-polarized experiments. The impact of the Supermirror alone is ‘PG Mrr Flip’. The signals you would compare are the bottom two orange and green lines. You can’t see it there (linear scale), but the flipping ratio is ~10 for 20 meV. Don’t let the flux drop scare you TOO much; we also observe significant background suppression when doing polarization analysis.

Our supermirror array acts like horse blinders; the critical angle limits the angular range at the sample which gets reflected to the detector array. At 15-20 meV Ei, no more than 7 mm horizontal extent is ‘visible.’ So, if you’re starting with a large sample, expect a hit in flux. If you’re building a multi-crystal mount, realize that vertical stacking works better than side-by-side.

Depolarization, direct beam

Half polarized

Vertical-only, full polarization analysis

Separation of Spin-Incoherent scattering with vertical full polarization analysis

If your H-loaded sample has overwhelming spin-incoherent scattering, then we may not get the desired 1/3 2/3 scattering expected from the spin-incoherent scattering, due to multiple scattering. The implication would be that polarization analysis would suppress, but not eliminate, the spin-incoherent scattering. For a powder-like sample, we recommend both with an annulus, and with a stack of horizontal cadmium discs inside the annulus, to minimize multiple scattering. For a single crystal, if you have several single crystals to stack vertically then one could put cadmium between those crystals.

3D polarization analysis