Graphite nanoparticles are synthesized by comminution of coarse graphite particles over an extended period in a ball mill. The size reduction is modelled using artificial neural network to develop a predictive tool to minimize contamination due to attrition in the ball mill. The particle size analysis and microstructural characterization were carried out using X-ray diffraction, Scanning Electron Microscopy and Transmission Electron Microscopy. It was found that the strain stored in the graphite lattice reaches a saturation value and remains constant during extended milling. Thermal annealing at 600 degrees C for 1 h effectively relieves the residual stresses so that the nanosized graphite particles are stress free. This paper demonstrates that mechanical milling can be an effective tool to synthesize stress free nanosized graphite particles in bulk. (C) 2015 Elsevier B.V. All rights reserved.