Test of a new Multi-Grid detector by an ILL-ESS team at the LLB : December 2017

Multi-grid detectors, introduced at the ILL in 2010, have shown their great potential to replace Helium-3-based detectors, and as a result, they will equip several instruments of ESS. A Multi-Grid is made of one or several columns of grids mounted in a gas vessel; each grid is composed of thin aluminum blades coated on both sides with a 10B-isotope-enriched B4C thin film to convert neutrons into ionizing particles. The standard grid used in several of our prototypes contains 60 cells of 2 cm x 2 cm x 1 cm, hence one column of n grids is equivalent to 60 proportional counter tubes segmented in n independent voxels. Anode wires are mounted in the middle of these tubes for gas amplification, and grids and anode wires are connected to front-end electronics via High-Voltage decoupling capacitors for the localization of neutrons.

One of the prototypes developed at the ILL in collaboration with ESS is called the MG_IN6; it has been equipped with 2 grid sectors, each of them occupying the same sensitive volume. The first sector is made of 3 columns of 16 standard grids, and the second one is made of one single column of 16 grids with enlarged dimensions (figure 1). A significant gain in detection efficiency is expected with the enlarged grid: first, there is no dead space in the whole sector, and second, the reduced dimension of the cells allows increasing the number of convertor layers in the depth direction (48 instead of 30); finally, the radial blades, which are not B4C-coated in the standard design, are now coated in the new grids. Another advantage of the enlarged grid comes from its cylindrical geometry, which gives a better response uniformity.

In order to confirm the gain in detection efficiency provided by the new grid, an ILL-ESS team has tested the MG_IN6 prototype in December 2017 at the LLB on the G4.3 instrument. Two beam conditions were used: a narrow beam of 3 mm x 3mm, or a non collimated beam with a plexi attenuator to avoid signal saturation. The detection efficiency measured at 2.8 Å with the narrow beam centered in the middle of the cells was 51% with the standard grid, and 62% with the new one. This improvement comes only from the higher number of convertor layers in the new grid. An even more significant gain in detection efficiency is provided by the radial blades and the absence of dead zones; this is shown by comparing the counting rate measured in the 2 sectors with the beam of large section; at 2.8 Å the sector with the new grids counts 45% more than the sector with the standard grids.

This work has been financed by the EU BrightnESS project, an H2020 grant agreement 676548. Our great thanks are going to the LLB people who helped organizing the test on G4.3.