Ductile-brittle transition detection in scratching of single crystal silicon using charged particle emissions

摘要

During machining of single-crystal silicon,material removal involves two modes ductile shear-based removal and brittle fracture-based removal.Ductile shear-based chip removal occurs when fracture is suppressed due to local stress conditions along with reduced chances of defect involvement and is desirable for achieving better surface integrity of the machined silicon wafer.In this work,we use charged particle emissions to identify mode of material removal(ductile or brittle)during scratching of a silicon wafer.Scratching tests were performed using a pin-on-disc tribometer setup with a conical diamond tip indenter,in which the wafer was held at an inclined position to achieve a varying-depth tapered scratch.The varying-depth scratch test was performed in such a manner that both ductile-to-brittle and brittle-to-ductile modes occur in a single scratch test.The charged particles emitted during the material deformation were collected using a Faraday plate mounted in the vicinity of the indenter and the intensity of the charged particles were measured using a sensitive femto/picoammeter.The scratch depth was measured using a 3D surface profiler and the mode of fracture was identified by examining crack density per unit length in a scanning electron microscope.These results were then correlated with the emission intensity signals.From the experimental results,a positive current intensity was observed for ductile mode of scratching and highly varying current intensity signal is observed during brittle mode of scratching.The results obtained were consistent over time and exhibited good repeatability.The present work indicates suitability of employing charge emission signals to detect mode of material removal during scratching of silicon.This work can be field-tested by conducting diamond turning experiments of silicon in real-time machining environment further testing the scope of use of charged particle emission to monitor real-time machining process.