Driven by the rapidly-growing of hybrid and electric vehicles, there is an increasing demand on lithium ion batteries (LIBs) with high energy density, long cycle life, and low cost. One of the most widely used anode materials in LIBs, graphite, only possesses a theoretical specific capacity of about 370 mAh/g, limiting the specific energy of LIBs.1 Recently, conversion-type of electrode materials, i.e. metal oxide and metal sulfide, has been regarded as one of the most promising anode materials for high performance LIBs, due to the low cost and exceptionally high specific capacity (>1000 mAh/g).2 However, these conversion-type electrode materials suffers from several issues, i.e. relatively low conductivity and rapid capacity fade due to irreversible phase change and particle pulverization during cycling. For example, the poor performance of tin oxide is mainly caused by its low intrinsic conductivity, formation of irreversible insulating Li2O phase during discharging, and large volume change.3 To overcome these disadvantages, various methods have been developed, like carbon/tin oxide composites, tin phosphide (Sn4P3). Tin phosphide has a layered structure and can be prepared by a mechanochemical or hydrothermal method. It has shown promising electrochemical performance as the anode in Li-ion batteries.4,5 As the lithium was inserted, Sn4P3 was first converted into lithium phosphides and Sn, followed by lithium-tin alloy formation. The advantage of Sn4P3 is its high capacity and Li+ conductivity of the Li3P. However, it suffers from similar drawbacks as tin oxide, i.e. large volume change and irreversible Li3P phase formation. In this work, we prepared Sn4P3 nanowires by hydrothermal reaction of red phosphorus and tin in ethylenediamine at 200 °C for 40 h. Then, graphite/Sn4P3 composites were prepared by high-energy ball milling for 2 h under Ar atmosphere. Graphite and Sn4P3 were uniformly distributed in the composites, shown in Figure 1a. When tested as the anode materials, the composites showed superior performance compared to Sn4P3 alone. The capacity of Sn4P3 rapidly decreased to around 200 mAh/g after 15 cycles. In comparison, graphite/Sn4P3 composites showed an initial specific capacity of about 1014 mAh/g and maintained a capacity of 782 mAh/g after 50 cycles. In addition, the coulombic efficiency was also much higher in the graphite/Sn4P3 composites, indicating a reversible Li insertion/deinsertion behavior and more stable SEI layer. The rate performance of graphite/Sn4P3 composites was also much better than pure Sn4P3. In conclusion, we report a facile preparation of graphite/Sn4P3 composites with excellent electrochemical performance as anode material in Li-ion batteries. Figure 1