Column runs with varied operation conditions of flowrate, feed concentration and particle size have been conducted to observe the sorption of nickel, copper and zinc by the chelating resin. They all show a typical constant-pattern, steep curve in the first part of the breakthrough curve and the slope would dramatically reduce at a breakpoint such that it would take extremely long time to reach 100% breakthrough level. ~ The Modified Bed Depth Service Time (BDST) model incorporated with a timedependent bed capacity has been used together with the Empty Bed Residence Time (EBRT) model to analyze the fixed bed performance under different operating conditions. The Homogeneous Surface Diffusion Model (HSDM) and the Shrinking Core Diffusion . Model (SCDM) have been applied to- model the fixed bed performance. Due to the dramatic change in the slope, the first and second parts of the breakthrough curves need to be modeled separately. Comparing the two models, SCDM is more appropriate to explain the sorption of metal ions into the chelating resin. The research has suggested the existence of a Na-loaded outer shell and a H-loaded inner core in the chelating resin. When the moving boundary progresses from the outer shell into the inner core, there is a remarkable change in the ion exchange process, resulting in different kinetics. This explains the sharp change in the slope of the experimental breakthrough curves. A new version of SCDM has been developed to model the progression of the moving boundary inside the resin for the successive resin layers along the whole resin column so as to predict the column' s solid and liquid phase concentration profiles. The new SCDM has. the flexibility of varying the portion of the outer shell so that the behavior of resins with different resin Na contents can be predicted.