Imaging Grains in a Polycrystal in 3D with Nanoacoustic Waves: Signal Processing Methods - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

Imaging Grains in a Polycrystal in 3D with Nanoacoustic Waves: Signal Processing Methods

Résumé

Picosecond laser ultrasonics methods provide non-contact, non-invasive imaging of &ₘu;m samples, serving in biology to map cell mechanics, likewise in material science to scan microstructures [1]. This work presents the 3D reconstruction of the cubic polycrystalline structure of <i>ceria</i>, with a pump-probe method: the time-domain Brillouin scattering (TDBS) [2]. In an opto-acoustic transducer (OAT), the <i>pump</i> fs laser pulse induces, via thermoelastic generation, an acoustic wave. When this wave is transmitted to the elastically anisotropic ceria, up to three mode-converted acoustic pulses could propagate. The delayed <i>probe</i> fs laser pulse is strongly reflected at the fixed OAT surface, and weakly reflected at the propagating acoustic wavefronts. The interferences between these reflections produce the so-called Brillouin oscillations which frequencies depend on the propagating acoustic pulses? velocities. Monitoring in time the Brillouin frequency changes of each acoustic mode, i.e. the velocity changes due to propagation in differently-oriented grains, give access to three different views of the 1D distribution of the grains along the <i>z</i>-direction (depth). The 3D reconstruction is obtained by gathering these 1D distribution views, combined with a 1 &ₘu;m step, 60 x 60 &ₘu;m<sup>2</sup> scan (<i>xy</i>-directions) of the sample. Each gathered transient reflectivity signal is processed using a synchronous detection technique to estimate the instantaneous Brillouin frequency, eventually giving the distribution of the velocities as a function of depth and of <i>xy</i>-directions. To better differentiate and extract shapes of the grains, Otsu's and alphashape image processing methods are applied to each slice (<i>xy</i> velocity distribution at a given depth). Gathering the informations from the three acoustic modes improves grains differentiation. The expected changes in the TDBS signal when the acoustic pulse is incident on an inclined grain boundary are theoretically addressed. [1] M.Khafizov et al., <i>Acta Materialia</i>, 112, 209-215 (2016), [2] V.E.Gusev, P. Ruello, <i>Appl. Phys. Rev.</i>, 5, 031101 (2018).
Fichier principal
Vignette du fichier
000548.pdf (464.45 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03240281 , version 1 (29-05-2021)

Identifiants

Citer

Théo Thréard, Elton de Lima Savi, Samuel Raetz, Sergey M. Avanesyan, Nikolay Chigarev, et al.. Imaging Grains in a Polycrystal in 3D with Nanoacoustic Waves: Signal Processing Methods. Forum Acusticum, Dec 2020, Lyon, France. pp.1897-1898, ⟨10.48465/fa.2020.0548⟩. ⟨hal-03240281⟩
30 Consultations
25 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More