Comparative Study of the Capacity and Rate Capability of LiNi y Mn y Co 1–2y O 2 (y 5 0.5, 0.45, 0.4, 0.33) Zheng Li, a, * Natasha A. Chernova, a Megan Roppolo, a Shailesh Upreti, a Cole Petersburg, b, * Faisal M. Alamgir, b, ** and M. Stanley Whittingham a, *** ,z a Institute for Materials Research, State University of New York at Binghamton, Binghamton, New York 13902-6000, USA b Department of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0245, USA An unresolved question for the layered oxides is: what is the optimum value of y in the formula LiNi y Mn y Co 1–2y O 2 for energy stor- age at moderate reaction rates? Here we report a systematic study of the specific capacity, rate capability and cycle life of Li x- Ni y Mn y Co 1–2y O 2 (y ¼ 0.5, 0.45, 0.4, and 0.333). The voltage of the Li/y ¼ 0.333 couple crosses over those of lower cobalt content for x < 0.55, as the Co redox begins to get involved. This early involvement of cobalt, rather than just Ni, leads to a slightly smaller specific capacity for y ¼ 0.333 than for LiNi y Mn y Co 1–2y O 2 with y > 0.333 when charging above 4 V. Overall the y ¼ 0.4 material has the optimum properties, having the highest theoretical capacity, less of the expensive cobalt and yet rate capabilities and capacity retention comparable to the y ¼ 0.333 material. V C 2011 The Electrochemical Society. [DOI: 10.1149/1.3562212] All rights reserved. Manuscript submitted January 10, 2011; revised manuscript received February 10, 2011. Published March 23, 2011. Since the pioneering work leading to the birth of the recharge- able lithium battery in the 1970s, 1,2 layered structured cathode mate- rials have been extensively studied. 3–5 The commercialization of the layered transitional metal oxide LiCoO 2 revolutionized the wireless communication and personal digital assistant. 6,7 Now the promising application of rechargeable lithium battery in plug-in hybrid electric vehicles (PHEV) requires cathode materials with higher volumetric energy density, higher power, lower cost, good structural stability, and thermodynamic stability of the delithiated phase MO 2 . LiFePO 4 has drawn much attention due to its high rate capability, abundance of its constituent elements and structural stability. 8–10 However, it has a low energy storage capability, particularly on a volumetric ba- sis, partly because of the need for conductive coatings. 11,12 Another promising cathode material the spinel LiMn 2 O 4 has a low storage capability because only 0.5 Li/Mn can be cycled. 13 Although LiCoO 2 has dominated the cathode market for lithium- ion cells, only in the last decade were the superior properties, better thermal stability, lower cost and higher capacity, of the mixed transition metal oxides LiNi y Mn y Co 1–2y O 2 highlighted. 14–20 LiNi 0.33 Mn 0.33 Co 0.33 O 2 is now found in many commercial cells. However, a still unresolved question is what is the optimum transition metal content. We have shown earlier that the LiNi y Mn y Co 1–2y O 2 with reduced cobalt also have equally good electrochemi- cal properties. 21,22 Therefore, we report here a systematic study on the effective capacity and rate capability of LiNi y Mn y Co 1–2y O 2 , for 0.33  y  0.50; we will use the following abbreviations, for y ¼ 0.5, 550; y ¼ 0.45, 992; y ¼ 0.4, 442; and y ¼ 0.33, 333. Experimental The series of layered compounds LiNi y Mn y Co 1–2y O 2 were syn- thesized by the co-precipitation method followed by solid-state reac- tion as described elsewhere. 22 A 5% excess of lithium was added to make sure that the molar ratio for Li/(Ni þ Mn þ Co) is around unity. The final sintering procedure was done at 800  C for 8 h and then the sample was quenched to room temperature by sandwiching it between two copper plates. The synthesis conditions for the four compositions were exactly the same. A temperature of 900  C was used to prepare pellets for density measurements. The x-ray diffraction (XRD) pattern was recorded using a Scin- tag XDS2000 h-h diffractometer equipped with a Cu-Ka radiation and a Ge(Li) solid state detector by step scanning (step size 0.02  / exposure time 10 s) in the 2h range of 10–90  . The Rietveld refine- ment was done by using GSAS/EXPGUI package. 23,24 The mor- phology of the samples was examined by transmission electron microscopy (TEM, Hitachi S570, Japan). The powder samples were dispersed on a copper grid with lacey carbon and studied under an accelerating voltage of 200 keV. Electrochemical properties were tested in 2325-type coin cells on VMP or VMP2 multichannel potentiostats (Biologic). The cath- ode paste was made by mixing 80 wt % active material with 10 wt % carbon black and 10 wt % PVDF (poly(vinylidene fluoride)) in NMP (1-methyl-2-pyrrolidinone), unless specified otherwise. The homogeneous paste was cast on aluminum foil then dried at 90  C overnight in a vacuum oven. The electrolyte used was 1 M LiPF 6 (lithium hexafluorophosphate) in a mixture of DMC (dimethyl car- bonate) and EC (ethylene carbonate) with 1:1 volume ratio (LP30 from EM Industries). The coin cells were assembled in a helium- filled glove box. The OCV (Open Circuit Voltage) profiles were obtained by charging or discharging the cell to the designated volt- age at 0.1 mA/cm 2 for 1 h. Following each pulse, the cell was left on open circuit for 1.5 h. The magnetic measurements were done using a Quantum Design SQUID magnetometer (MPMS XL-5). The temperature dependence of the dc magnetization was measured whilst cooling the samples from 350 to 2 K in a magnetic field of 1000 Oe. Two sets of partially delithiated Li x Ni 0.33 Mn 0.33 Co 0.33 O 2 samples were prepared electrochemically. One set is “powder” with x ¼ 1.00, 0.77, 0.70, 0.51, 0.29, the other set is the “electrodes” with x ¼ 1.00, 0.80, 0.70, 0.60, 0.50, 0.45, 0.40, 0.35, 0.30, i.e. the mixture of active material, carbon black and PVDF. The positive electro- des used in the cell for “powder” sample preparation consist of 94 wt % 333 material, 6% PVDF and no carbon. The as-assembled cells were charged continuously or intermittently to the targeted x-value for various Li x Ni 0.33 Mn 0.33 Co 0.33 O 2 , where the extracted amount of lithium was determined by coulometric titration. The charged cells were kept under open circuit for more than 12 h and were disassembled in the glove box. For the powders, the PVDF was washed away using NMP and the powders dried in the vac- uum oven. For the electrodes, the mixture was directly peeled off from the aluminum and washed with DMC, then dried in flowing nitrogen. All the samples were checked by XRD before further testing. The x-ray absorption data was acquired in the transmission mode at the beamline X23A2 at the National Synchrotron Light Source, Brookhaven National Laboratory. A Si(311) monochromator was used for an overall spectral resolution (DE/E) of 2  10 –4 . Energy calibrations were carried out by using the first inflection points in the spectra of Mn, Co or Ni metal foils (Mn K-edge ¼ 6539 eV, Co K-edge ¼ 7709 eV, Ni K-edge ¼ 8333 eV) as references. The XAS data analysis was carried out by standard background fitting and * Electrochemical Society Student Member. ** Electrochemical Society Active Member. *** Electrochemical Society Fellow. z E-mail: stanwhit@gmail.com Journal of The Electrochemical Society, 158 (5) A516-A522 (2011) 0013-4651/2011/158(5)/A516/7/$28.00 V C The Electrochemical Society A516 ecsdl.org/site/terms_use address. Redistribution subject to ECS license or copyright; see 128.61.52.138 Downloaded on 2013-08-13 to IP