ALSA – Automatic Laue Sample Aligner

Petr Čermák1, Damián Wałoszek1, Kryštof Višňák1, Ivo Gold1,2

1Charles University (MGML, Dep. Condensed Matter Physics), 12116 Prague 2, Czechia

2Institute of Physics of the Czech Academy of Science, 182 21 Prague 8, Czechia

petr.cermak@matfyz.cuni.cz

While the evolution of the synchrotron brightness still follows exponential Moore’s law [1], the time-integrated flux of the neutron sources reached its maximum 50 years ago at ILL, and it will be only slightly surpassed by the upcoming ESS [2]. Therefore, scientists have been trying for decades to optimize optics, measurement strategies or build multi-detector systems to take advantage of every possible neutron to measure weaker fluctuations on smaller samples.

Another approach is to increase the sample size for inelastic neutron experiments by co-aligning more single crystals. This process is very time-consuming and often not very precise (e.g. [3]). The goal of ALSA (Fig. 1) is to change it. It will automatize the co-alignment process by using a state-of-the-art X-Ray Laue diffractometer (Photonic Science), robotized manipulators (Mecademic Meca500), real-time camera recognition (Basler), and bespoke neural network software for crystal placing and Laue pattern solving. The device ALSA will be a true game-changer in the field of inelastic neutron scattering because it will drastically speed up sample preparation.

To glue small crystals as close to each other as possible, we have developed an online algorithm for irregular polygon stacking; a series of benchmarking tests proved, that it is the most efficient online algorithm available. In this presentation, we will focus on the hardware and software design of the device, and we will discuss possibilities of using neutral networks for Laue patterns solving.

ALSA device

Figure 1. Sideview of the ALSA device. (1) Photonic Science CCD back reflection Laue X-ray Detector, (2) Mecademic Meca500 six-axis industrial robot arm, (3) Samples waiting for alignment, (4) Aluminium plate with aligned samples, (5a, b, c) CCD cameras for sample shape determination and alignment, (6) High brilliance X-ray generator, (7) Signalling for X-Ray source.

 

1. Assoufid, L. and Graafsma, H. MRS Bulletin 42, (2017), 418-423. doi:10.1557/mrs.2017.118

2. Rønnow H.M. et al. Swiss Academies Reports 16, (2021), 7. doi:10.5281/zenodo.4637660

3. Duan, C. Et al. Nature 600, (2021), 636–640. doi:10.1038/s41586-021-04151-5