Pk have prepared intermetailic phases and mixtures of such phases in the Sn-Fe-C Gibbs triangle by mechanical alloying methods or by direct melting. This second paper in a three-part series focuses on the intermetallic phases in the binary Sn-Fe system, Sn2Fe, SnFe, Sn2Fe3, and Sn3Fe5. Using in situ X-ray diffraction and electrochemical methods, we study the reversible reaction of Li with these materials. Li/Sn-Fe cells made from annealed powders have reversible capacities of 600, 50, 20, and 60 mAh/g, respectively for Sn2Fe, SnFe, Sn2Fe3, and Sn3Fe5. Li/Sn-Fe cells made from the same materials, but after high-impact ballmilling, show reversible capacities of 650, 320, 200, and 150 mAh/g. Specific capacities of 804, 676, 582, and 557 mAh/g are expected for Sn2Fe, SnFe, Sn2Fe3, and Sn3Fe5 if ail compounds react fully with Li tb form Li4.4Sn and Fe. In situ X-ray diffraction experiments on the ballmilled materials confirm the formation of Li4Sn during discharge but also show that in the cases of SnFe, Sn2Fe3, and Sn3Fe5 at least 50% of the starting phase remains unreacted. Structural considerations suggest that as the Fe:Sn ratio increases, Fe atoms may form a impenetrable "skin" on the surface of particles or grains, as Li reacts with the. Sn-Fe compounds. This skin prevents the full reaction of the intermetallic with Li, leading to an observed capacity which is lower than expected. High-impacting reduces particle and grain: size, so the effect of the skin is less than far the annealed powders and higher capacities are obtained. As the Fe content in the Sn-Fe intermetallics increases, the cycle life of the materials improves, presumably because there is more Fe per Sn and because the formed Fe and residual starting material act BS a "matrix" to hold the Sn and Li-Sn alloys together during cycling. We give an example of a material with a volumetric capacity of 1200 mAh/cm(3) showing stable cycling for over 80 cycles.