The potential feasibility of steel slag as a low cost removal agent for heavy metal ions Pb(II), Cu(II) and Cd(II) in acidic conditions was investigated in this study. The initial pH effect on heavy metal ion removal efficiency, the compounds formed after heavy metal ion removal, and the binding force of metals with the compounds were determined. The results showed that the efficiency of removing heavy metal ions by steel slag was low at low initial pH levels, yet it sharply increased and then became stable as the initial pH increased. The pseudo-second order model provided the best description for the removal of Pb, Cu, and Cd ions, indicating that the predominant heavy metal ion removal mechanism was chemisorption. The images obtained by the Fourier transform infrared spectroscopy (FTIR) and the X-ray diffraction (XRD) analysis indicated that the main compounds formed after the removal of Pb, Cu, and Cd ions by steel slag in an aqueous solution were heavy metal ferrites, silicates, carbonates, hydroxides and oxides. Sequential extraction experiments showed that these three heavy metals bond to the compounds mainly in the carbonate fraction (F-2), the Fe oxide bound fractions (F-3 (a) and F-3 (c)), and the residual fraction (F-4) in which F-2 corresponded to the carbonates, and F-3 (a), F-3 (c) and F-4 corresponded to the amorphous or crystalline ferrites and silicates, respectively. The F-3 (a), F-3 (c) and F-4 are relatively stable and do not tend to re-release metal ions in acidic solutions. However, F-2 and heavy metal hydroxides have relatively low stability and dissolve readily, re-leaching heavy metal ions into the acid solution. When these three heavy metal ion mixtures were removed by steel slag, the Pb, Cu and Cd deposits were at higher levels in the F-3 and F-4 fractions. Therefore, it was concluded that the co-existence of heavy metal ions in an aqueous solution is beneficial for their removal by steel slag in acidic conditions.