Intelligent tracking of catalytic nanoparticles trajectories during in situ ETEM experiments - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Intelligent tracking of catalytic nanoparticles trajectories during in situ ETEM experiments

F. Khuram
  • Fonction : Auteur
T. Grenier
  • Fonction : Auteur

Résumé

Characterizing at high spatial resolution nanoparticles (NPs) involved in heterogeneous catalysis is a key step for understanding their potential efficiency in catalytic processes. Using Environmental Transmission Electron Microscopy (ETEM), it is now possible to study these nanocatalysts under reactive conditions, i.e. under gas and temperature, approaching experimental working conditions corresponding to their conditioning or activation. With such in situ or operando studies, the reasons for deactivation of the catalyst, for example the growth of the NPs either by Ostwald ripening or coalescence, can be investigated. Recording series of images at reasonable temporal frequencies allows tracking the dynamic evolution of a quite large population of NPs exposed to gas and heat stimuli. A meaningful and quantitative analysis of such observations can then be performed, assuming a correct control or possible damages induced by electron irradiation. To do so, tedious measurements are needed to detect the NPs and identify their trajectories when they possibly move and interact between them by diffusion processes. This is typically a multiple object tracking (MOT) problem, a computer vision approach which has been well-known and treated through automatic routines for a few decades for important societal subjects, such as pedestrians’ localization or traffic survey, see e.g. [1]. A robust and automatic pipeline has been derived to achieve this task on the basis of Machine Learning (ML) and MOT approaches applied to the calcination study of Pd NPs supported on delta-alumina, a well-known catalytic system for selective hydrogenation. Details of the ETEM study (FEI/TFS Titan G2 80-300 kV) were reported previously [2]. Sequences of STEM images were acquired during up to 3 hours up to 400°C under oxygen. For the detection of NPs in these micrographs, we used the well-known Unet neural network [3]. Excellent results are obtained once the network has been properly trained on realistic simulated micrographs for which all information is then known a priori. Synthetic images of a given population of spherical (for simplicity) NPs, with known but varying chemical composition and size, can be generated on a supporting media with its own morphological characteristics (variable thickness, rugosity and pore distribution). We also developed a random walk routine to simulate dynamic sequences by moving NPs while respecting expected size and intensity variations during coalescence, crossing or disappearance events. To fine-tune the network, slightly more sophisticated simulations were also performed. The second step of the approach concerns the NP tracking. We have implemented additional features to the algorithm developed by Milan et al. [1] initially dedicated to the tracking of pedestrian walks. It consists in a continuous energy minimization (CEM) which accounts for energy terms representing various events mainly unrealistic in the case of humans or vehicles: (i) fusion (coalescence) of two particles into a single one (ii) less probable but still possible, division of one particle into smaller ones (iii) crossing of particles lying on upper and lower surfaces of the substrate According to the above, size and intensity energies criteria were added to the CEM algorithm to identify and confirm fusion / division events. Results of the Unet-based identification of NPs and of the modified CEM tracking approach will be illustrated on both simulated and experimental dynamic sequences [4]. [1] A. Milan et al., PAMI, 38 10 (2016) 2054. [2] T. Epicier et al., Catalysis Today, 334, 15 (2019), 68. [3] 0. Ronneberger et al., Lect. Notes Comput. Sci. 9351 (2015) 234. [4] The authors thank the EUR SLEIGHT https://manutech-sleight.com/) and the CNRS Fedreation IngeLySE for financial support, CLYM (www.clym.fr) for the access to the ETEM and IFPEN (Solaize, F) for providing samples.
Fichier non déposé

Dates et versions

hal-03662979 , version 1 (09-05-2022)

Identifiants

  • HAL Id : hal-03662979 , version 1

Citer

F. Khuram, T. Grenier, C. Ducottet, Thierry Epicier. Intelligent tracking of catalytic nanoparticles trajectories during in situ ETEM experiments. 2022 MRS Spring Meeting, May 2022, Honolulu, United States. ⟨hal-03662979⟩
48 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More