The potential use of different iron phosphates as cathode materials in lithium-ion batteries has recently been investigated.1 One of the promising candidates is LiFePO4. This compound has several advantages in comparison to the state-of-the-art cathode material in commercial rechargeable lithium batteries. Firstly, it has a high theoretical capacity (170 mAh/g). Secondly, it occurs as mineral triphylite in nature and is inexpensive, thermally stable, non-toxic and non-hygroscopic. However, its low electronic conductivity (~;10-9 S/cm) results in low power capability. There has been intense worldwide research activity to find methods to increase the electronic conductivity of LiFePO4, including supervalent ion doping,2 introducing non-carbonaceous network conduction3 and carbon coating, and the optimization of the carbon coating on LiFePO4 particle surfaces.4 Recently, the Li doped LiFePO4 (Li1+xFe1-xPO4) synthesized at ARL has yield electronic conductivity increase up to 106.5 We studied electronic structure of LiFePO4 and Li doped LiFePO4 by synchrotron based soft X-ray emission (XES) and X-ray absorption (XAS) spectroscopies. XAS probes the unoccupied partial density of states, while XES the occupied partial density of states. By combining XAS and XES measurements, we obtained information on band gap and orbital character of both LiFePO4 and Li doped LiFePO4. The occupied and unoccupied oxygen partial density of states (DOS) of LiFePO4 and 5 percent Li doped LiFePO4 are presented in Fig. 1. Our experimental results clearly indicate that LiFePO4 has wide band gap (~; 4 eV). This value is much larger than what is predicted by DFT calculation. For 5 percent Li doped LiFePO4, a new doping state was created closer to the Fermi level, imparting p-type conductivity, consistent with thermopower measurement. Such observation substantiates the suggestion that high electronic conductivity in Li1.05Fe0.95 PO4 is due to available number of charge carriers in the material. Furthermore, Hall effect measurement on Li doped sample confirmed presence of free charge carriers, which are responsible for the observed electronic conductivity increase in Li doped LiFePO4. There is no evidence that Fe3+ valence is created by doping with excessive Li+ in Li1.05Fe0.95PO4, as shown by Fe-edge XAS. (Fig.2) Instead, charge-carrier holes reside primarily in unoccupied O 2p states, which compensate for the charge deficiency from Li+ substitution for Fe2+. The increased conductivity in Li1.05Fe0.95PO4 is attributed to the new charge carriers (doped holes) and the strong electron correlation between O 2p and Fe 3d states.
