The use of high-pressure grinding rolls (HPGR) prior to ball mills has become a common practice in cement clinker grinding due to significant energy savings in comparison to ball milling alone. The energy savings has been attributed to higher energy efficiency of HPGR at low reduction ratios, smaller particle top sizes in the ball mill feed, and the weakening of particles resulting from compression of particle bed by HPGR. This study attempts to show how the weakening effect changes the kinetic breakage parameters of the HPGR-treated clinker. For this purpose, batch ball mill experiments were conducted with three narrow-size fractions (−3.35 + 2.36, −1.70 + 1.18, and −0.85 + 0.60 mm) of the feed and product samples of an industrial HPGR, and the breakage rate and breakage distribution parameters of the samples were compared. The results show that the weakening effect leads to nonlinear breakage rates for all sizes, coupled with abnormal breakage at the two coarsest sizes. The weakening effect leads to increased breakage rates of the HPGR product relative to particles of the HPGR feed. The increased breakage rates are accompanied by coarser progeny distributions for the two coarsest size ranges relative to that of the HPGR feed. The progeny distribution of the finest size of the HPGR product, however, is finer than that of the HPGR feed. Ball milling of size-distributed feeds prepared from the HPGR product and feed samples for the same grinding time produces self-similar size distributions when the size is normalized with respect to the median size.

High-pressure grinding rolls (HPGR) are widely used in cement clinker grinding prior to ball milling. The efficiency of HPGR was previously related to two capabilities: (a) efficient stressing mechanism and (b) weakening of particles which lead to finer product sizes after subsequent ball milling. This study aimed to investigate the influence of HPGR on the impact breakage behavior of coarsely sized cement clinker. Drop weight tests were conducted on three single-size samples from both feed and product streams of an industrial HPGR, so as to obtain impact breakage parameters: breakage probability and impact breakage functions. Radiographic, Vickers micro-hardness, BET, SEM, XRD, XRF, and image processing methods were utilized to explain the variations in impact breakage parameters. The product samples were found to have less porosity than feed samples but they contained HPGR-induced micro-cracks. Those cracks are believed to be responsible for lower hardness and higher breakage probability of product particles than of feed ones under impact. However, fragments generated from the product samples by drop weight tests were coarser than those generated from feed samples. That was due to elimination of pores by HPGR action. At excess energy levels, both samples were broken to same extent regardless of structural differences.