Machine learning-assisted characterization of electroless deposited Ni–P particles on nano/micro SiC particles

• Published in: Ceramics international Vol. 49; no. 18; pp. 29849 - 29856
• Main Authors: Gyökér, Zoltán, Gergely, Gréta, Takáts, Viktor, Gácsi, Zoltán
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15-09-2023
• Subjects: Ceramic substrate; Electroless metal deposition; Machine learning-assisted image analysis; Nanoparticles; Specific surface area; Nanoparticles; Electroless metal deposition; Specific surface area; Ceramic substrate; Machine learning-assisted image analysis
• ISSN: 0272-8842; 1873-3956
• DOI: 10.1016/j.ceramint.2023.06.242

In this experiment, Ni–P nanoparticles were deposited (ED) on SiC micro- and nanoparticles with different parameters. Our goal was to successfully prepare metal deposits and develop an effective method for comparing and evaluating the various procedures.
During the experimental work, a three-step electroless Ni–P coating process was applied with different concentrations. The coated SiC particles were examined by scanning electron microscopy (SEM). The mass-specific surface area (SSA) of the coated SiC was measured by the Brunauer‒Emmett‒Teller (BET) method, while the volumetric-specific surface area (VSSA) was also calculated. The adhesion between the metal and the ceramic particle was analyzed by X-ray photoelectron spectroscopy (XPS).
An image processing macroprogram was created (with a machine learning-based Trainable Weka Segmentation algorithm) to segment the SEM images of the ED metal particles to calculate the specific surface area (SV).
[Display omitted]
•Electroless Ni–P particles were deposited on SiC micro- and nanoparticles with different parameters.•X-ray photoelectron spectroscopy (XPS) was used to analyze the adhesion between the metal and ceramic particles.•A machine learning-assisted image processing macro program was created to segment the ED metal particles.•SSA (m2/g), VSSA (m2/cm3), SV (m2/cm3) and distribution of metal particle diameter were compared.