Abstract

In active brazing of diamonds, high brazing temperature and difference in the thermo-mechanical properties of the diamond grain, filler alloy, and substrate lead to the development of unfavorable residual stresses in the bonding region. This residual stress can potentially cause initiation of cracks within the interfacial reaction layer formed between diamond and filler-alloy, weakening the joint. Since the diamond particles used in diamond grinding tools are micro-sized and complex in shape, it is experimentally difficult to measure residual stress at the grit-filler interface. In the present investigation, a coupled thermo-mechanical three-dimensional finite element model has been developed to predict, map, and analyse the formation of residual stresses in diamond-bond-substrate systems. The presented model uniquely differs from those in literature, which are mostly 2-D type and did not consider the presence of interfacial reaction layer between diamond and filler. A Ni-Cr-Fe-B active brazing alloy has been selected as the filler and C45 steel as the substrate in the current simulation. Chromium carbide has been considered to be the interfacial compound. The numerical simulation suggests that the maximum value of residual stresses, predominantly tensile in nature, gets shifted from the diamond to the intermetallic zone at the bond level when the interfacial reaction compound is introduced between the diamond and filler. It also establishes that orientations of diamond during its placement on substrate for brazing have significant influence on the magnitude of induced residual stress. Approximately 40 % reduction has been observed in case of hexagon-base brazed diamonds when compared to that (1620 MPa) developed on square-base brazed diamonds. Raman spectral analysis of brazed specimen was conducted to experimentally validate the numerical simulation results. Residual stress was estimated from the peak shift of D-band in the spectra. The numerically computed residual stresses were found to deviate within the range of 3–15 % of the experimentally determined values.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call