Journal of Power Sources 159 (2006) 1340–1345 Low temperature performance of nanophase Li 4 Ti 5 O 12 J.L. Allen ∗ , T.R. Jow, J. Wolfenstine U.S. Army Research Laboratory, Adelphi, MD 20783-1197, USA Received 2 November 2005; received in revised form 7 December 2005; accepted 8 December 2005 Available online 25 January 2006 Abstract The low temperature electrochemical performances of 700 and 350 nm Li 4 Ti 5 O 12 were compared. At high rate, room temperature and at low rate and low temperature (0, -10, -20 and -30 ◦ C), the 350 nm Li 4 Ti 5 O 12 showed higher capacity than the 700 nm Li 4 Ti 5 O 12 . This difference is proposed to result from the shorter diffusion lengths and higher number of lithium insertion sites in the 350 nm Li 4 Ti 5 O 12 compared to the 700 nm Li 4 Ti 5 O 12 . However, at high rate and low temperature, a transition in performance was observed, that is, the 700nm material had higher capacity. At high rate and low temperature, it is proposed that interparticle contact resistance becomes rate limiting owing to the temperature dependence of this property and this accounts for the different behavior at low temperature and high rate. Published by Elsevier B.V. Keywords: Li 4 Ti 5 O 12 ; Anode; Li-ion battery; Low temperature; Nanophase 1. Introduction Li 4 Ti 5 O 12 is a potential anode material for Li-ion batteries with some unique and potentially useful characteristics [1–3]. For example, it is a zero-strain lithium insertion host suggest- ing virtually unlimited cycle life. It features a flat, operating voltage of about 1.5 V versus lithium, above the reduction poten- tial of common electrolyte solvents thus, it does not form a solid electrolyte interface based on solvent reduction which should be a favorable property for high rate and low tem- perature operation. This voltage also is sufficiently high such that the dangers of lithium plating that can occur at high rate and/or low temperature are removed. The use of nanophase Li 4 Ti 5 O 12 has been shown to yield improvements in rate capa- bility [4–6]. Furthermore, the use of nanostructured electrodes has been reported to enhance low temperature performance [7–9]. Up to this time, the low temperature performance of nanophase Li 4 Ti 5 O 12 has not been reported. Herein, we report on the low temperature performance as a function of particle size and rate. ∗ Corresponding author. Tel.: +1 301 394 0291; fax: +1 301 394 0273. E-mail address: jallen@arl.army.mil (J.L. Allen). 2. Experimental Two Li 4 Ti 5 O 12 samples were compared of differing particle size: (1) Nanomyte TM Li 4 Ti 5 O 12 , obtained from NEI corpora- tion, hereafter referred to as NEI-Li 4 Ti 5 O 12 [10] (2) a larger particle size Li 4 Ti 5 O 12 prepared at ARL using a solid-state method [11] from TiO 2 (rutile structure) and Li 2 CO 3 , here- after referred to as ARL-Li 4 Ti 5 O 12 . Three weight percent excess Li 2 CO 3 was used to compensate for lithia volatilization during the high temperature heating. The starting materials were ground with an alumina mortar and pestle with enough methanol to form a slurry. The dried and mixed reactant mixture was heated at 800 ◦ C for 12 h in air. The sample was reground, pelletized and heated for another 24 h at 800 ◦ C in air. The Li 4 Ti 5 O 12 samples were first characterized by X-ray diffraction with a Rigaku Ultima III diffractometer using Cu K radiation. Lattice constants were determined by obtaining diffraction data in a parallel beam diffraction geometry and fitting the data using Rietveld refinement [12] using RIQAS software (Materials Data Inc.). Crystal size was evaluated by col- lecting diffraction data in a Bragg–Brentano (focusing) geom- etry and correcting for instrumental broadening by using LaB 6 (NIST; 660A). Surface area measurements were done by the Brunauer– Emmett–Teller (BET) [13] method using N 2 as adsorbate gas. The average particle size diameter was calculated based on this 0378-7753/$ – see front matter. Published by Elsevier B.V. doi:10.1016/j.jpowsour.2005.12.039