起始试剂Li2CO3, Fe2O3, NiO, Cr2O3 and TiO2 (all Aldrich, >99%)须在实验前干燥好以备使用，然后称量，和乙醇混合，并且把混合物在室温下研磨30分钟，使其充分混合。这种球形研磨可以缩短反应时间，降低烧结温度，从而使锂的损耗降到最低。然后将粉末在600-700℃的温度下干燥并烧结几个小时，重新研磨，再把粉末置于900-1000℃的温度下烧结1-2小时使其充分反应。最后吧样品再次球形研磨30分钟。
粉末种子衍射数据是用位于英国，牛津郡抵得渴特美国钢铁协会的阿普尔顿卢瑟福实验室中的高分辨率粉末衍射仪得到的。实验在一个充满标准钒的容器中进行，持续大概4小时(800 μA h)。飞行时间在2000- 19550 μs.范围内测定。用Rietveld方法的GSAS程序使结构精细、精纯。数据中大于11000μs的要冲最终的提纯数据中剔除，因为它们包含了最强的反射信息，会影响最终结果。
在电化学测试中，复合电极是在一个易变的溶剂中加入75 wt.%悬浮活性物质，10 wt.%活性极好的活性碳（提高电子电导），和15 wt.%聚乙烯（亚乙烯基氟化物）粘合剂[化学试剂公司]，然后把悬浮液置于铜箔上，在真空中75℃干燥一夜。脉冲横流测量方法是用两个电极单元和一个电脑辅助识别电池测试系统来进行测试的。锂被用来制相反电极，含1molal 锂 PF6[森田]的EC:DMC(1:2) [Grant Chemicals]做电解液。电势一般在0.00–2.00 V之间，电流密度是0.1mA •cm−2，大约C/10.。
Present day lithium-ion cells typically comprise a lithiated carbon anode, organic liquid electrolyte and a lithium transition metal oxide cathode. These cells offer obvious advantages over nickel-cadmium and lead-acid cells i.e. superior energy density and low toxicity; however, there are still safety concerns regarding the carbon anode. A possibly safer alternative is to use solid state inorganic oxides as both the negative and positive electrode materials.
In 1994, Thackeray et al. demonstrated Li4Ti5O12 spinel could be used as an electrode material in rechargeable lithium batteries . The capacity is very stable with cycling; however, the voltage is rather high for utilisation as the negative electrode at more than 1.5 V vs. Li/Li+. The aim of the present study was to investigate the effect on the electrochemistry of Li4Ti5O12 (≡Li1.33Ti1.67O4) of replacing some of the Ti4+ by other 3d transition metals according to the substitution mechanism: 3M3+ 2Ti4++Li+ where M3+=Fe3+, Ni3+, Cr3+. In particular, it was hoped that the working potential of 1.55 V vs. Li/Li+ could be lowered. Blasse first reported the crystal chemistry of the Fe solid solution ，although the electrochemical behaviour is not well known .Fe is preferable over other transition metals such as Co and V due to its abundance and low toxicity. One possible disadvantage is the tendency for Fe3+-containing spinels to be slightly inverse in character, i.e., some of the Fe3+ ions occupy the tetrahedral 8a sites as well as the 16d octahedral sites. It is generally believed that transition metals sharing the tetrahedral sites with the lithium ions in spinel oxides inhibit the lithium ions’ diffusion through the lattice during intercalation and cause a decline in performance with repeated charge–discharge cycling. Ni and Cr were chosen as alternative dopants for this work given their similar ionic radii to Ti4+ and preference for octahedral coordination. The spinel LiCrTiO4 is already known .
Results obtained from electrochemical and structural studies, including neutron diffraction, on all three solid solutions are presented.
The starting reagents, Li2CO3, Fe2O3, NiO, Cr2O3 and TiO2 (all Aldrich, >99%) were dried overnight prior to use then weighed, mixed with ethanol, and ball-milled at room temperature for 30 min to ensure intimate mixing. Ball milling was also found to shorten reaction times and reduce firing temperatures, thereby minimising the risk of lithia loss. The powders were then dried and heated at 600–700°C for several hours and reground before firing at 900–1000°C for 1–2 h to complete reaction. Finally, samples were ball-milled again for a further 30 min.
For phase identification, a Philips PW1830 X-ray diffractometer with Cu Kα radiation was used. Data were collected over the range 10°≤2θ≤70°. For lattice parameter determination, Si was used as an internal standard and data were collected up to 90° 2θ over several hours.
Powder neutron diffraction data were collected using the high resolution powder diffractometer (HRPD) situated at the ISIS neutron facility, Rutherford Appleton Laboratories, Didcot, Oxon, UK. Experiments were run at room temperature in a standard vanadium can for about 4h (800 μA h). Time-of-flight data were collected between 2000 and 19550 μs. Structural refinements were performed by the Rietveld method using the GSAS programme. Data greater than 11000 μs were excluded from the final refinement as they contained information from the most intense reflection and were found to falsely skew the result.
For electrochemical testing, composite electrodes were prepared by suspending 75 wt.% active material, 10 wt.% super S carbon (to enhance electronic conductivity) and 15 wt.% poly(vinylidene fluoride) binder [Aldrich] in a fugitive solvent, such as cyclopentanone. The slurry was cast onto Cu foil and vacuum dried overnight at 75°C. Galvanostatic measurements were made using two electrode cells and a Macpile computerised battery test system. Li metal was used as the counter electrode and 1 molal Li PF6 [Morita] in EC:DMC (1:2) [Grant Chemicals] as the electrolyte. The potential range was generally 0.00–2.00 V and the current density 0.1 mA cm−2, equivalent to about C/10.