An investigation of the PMEDM processing and surface characterizations of AZ61 matrix composites via experimental and optimization methods


Mustu M., Demir B., Aydin F., GÜRÜN H.

Materials Chemistry and Physics, vol.300, 2023 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 300
  • Publication Date: 2023
  • Doi Number: 10.1016/j.matchemphys.2023.127526
  • Journal Name: Materials Chemistry and Physics
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aerospace Database, Chemical Abstracts Core, Communication Abstracts, INSPEC, Metadex, Civil Engineering Abstracts
  • Keywords: AZ61 alloy, Mg matrix composites, GNPs, PMEDM, TiB2
  • Gazi University Affiliated: Yes

Abstract

In this study, AZ61/15 wt%TiB2 and AZ61/15 wt%TiB2-0.5 wt%GNPs composites were manufactured by hot pressing, and the machinability of the produced samples was carried out by powder mixed electrical discharge machining (PMEDM). The influence of PMEDM parameters, namely pulse on time, current and materials, were studied by surface roughness (SR) and material removal rate (MRR) using the Taguchi design. The microstructure and surface quality of the machined surfaces and cross-sections were investigated using 3D microscopy, scanning electron microscopy (SEM), energy dispersive spectrometry (EDS) and X-ray diffraction (XRD). Results showed that PMEDM produced a melting, white layer and transition zone direct proportionally to the processing parameters. The white layer did not differ from the metal integrity of the transition zone. Additionally, volcanic craters, holes, cracks, debris covered with molten metal, and reinforcement particles were observed at the machined surface and cross-section. A high amount of oxygen was detected on the machined surface as a result of the interaction between kerosene and generated heat changing proportionally to the amount of the EDM parameters. The analysis of variance (ANOVA) showed that the pulse on time and materials, with 65.46% and 40.86%, were the most significant parameters on the SR and MRR, respectively. For regression models, the determination coefficient (R2) for the prediction of SR and MRR was noted to be 0.98 and 0.85, respectively.