An investigation of the mechanical and physical issues of the adhesion of chemically vapor deposited (CVD) diamond coatings on cemented carbide cutting tool inserts (Cemented carbide tools, Chemical vapor deposition)

In this work, the role of physical and mechanical factors on the adhesion strength of chemically vapor deposited (CVD) diamond coatings on cemented carbide inserts was investigated. The changes that occurred in the substrate residual stress due to pretreatment was measured by x-ray diffraction. A drop in the stress values of the substrates was found to occur after chemical etching and annealing processes. It was shown that the initial compressive residual stress present in the substrate was always found to drop as a result of the CVD diamond deposition process as well.; A scraper testing fixture was designed to examine the adhesion strength of diamond coatings to cemented carbide substrates. The adhesion was found to be extremely poor when the surface of the substrate was mechanically polished or seeded with a layer of nano-crystalline sized diamond particles. Diamond coatings having a thickness higher than 12 μm immediately failed after completion of the CVD process. Glass blasting the substrate prior to CVD diamond deposition and HF etching improved the adhesion strength of the diamond coatings when tested by the use of a scrapper test. Qualitative results show that the diamond coatings deposited on the samples treated by glass blasting and HF-HNO₃ etching did not flake off during the scraping action.; The thermal residual stress in CVD diamond-coated carbide inserts was computed using a three dimensional finite element model. The model revealed that large thermal stresses existed along the flank and rake regions of the cutting inserts. It was observed that the thickness of the diamond coating affected the magnitudes of the stress components. The use of the indentation method as a quantitative adhesion test was shown not to be useful. The normal stress at the interface resulting from the indentation method were highly compressive leading to brittle cohesive failure rather than interfacial failure. The superposition of the total residual stresses and the stresses resulting from indentation did not yield a desired high tensile normal stress at the interface necessary for quantitative adhesion evaluation.