An “experimental friendly” model for zone refining process is proposed which predicts effective zone length in each refining passes that would lead to maximal solute removal, thereby leading to ultrapurification of the material for use in high-end electronic applications. The effectiveness of the model is experimentally tested and validated by purifying gallium from 4N (99.99%) to 6N5 (99.99995%) purity level at 30% yield and ∼6 N at 70% yield with respect to targeted metallic impurities such as, Zn, Cu, Al, Ca, Bi, Si, Pb, Ni, Mn, and Fe, as analyzed by inductively coupled plasma optical emission spectrometry, graphite furnace atomic absorption spectrometry, and high resolution inductively coupled plasma mass spectrometry techniques. The distribution coefficient (k) of all the targeted impurities, detected in the purified gallium, was found to be less than 1. By comparing the experimentally obtained axial concentration profiles with the theoretical calculations, the k values of some detected impurities, such as Ca and Al, are determined to be ∼0.8, Pb and Bi to be 0.7, Cu to be 0.65, and Fe to be 0.68, which prove the efficiency of the proposed model in reducing the concentration of these vulnerable impurities significantly. Following the model and as evidenced from the theoretical predictions, degradation of material purification containing a mixture of impurities having k less than as well as greater than 1 was elucidated experimentally by zone refining of 4N6 indium. Only a 40% yield of 5N6 indium was obtained, thereby highlighting the intricacies and problem areas in ultrapurification of these types of material.