AABC 2014 LLIBTA Symposium - Large Lithium Ion Battery Technology and Application - Track A: Cell Materials and Chemistry Session 1

This presentation will present recent work at Argonne National Laboratory focused on an atomic level understanding of degradation mechanisms in layered-layered composite cathode materials. A model of voltage fade and hysteresis will be presented and discussed in terms of:

• X-ray absorption spectroscopy
• Electrochemical cycling data
• Model, layered materials

A strong correlation between voltage fade and hysteresis will be established and certain aspects of the model tested. The presentation will conclude with summary remarks, including limitations of the model as well as plans for future research.

Transition metal phospho olivines LiMPO₄ (M= Fe, Mn, Co) are very promising alternative positive materials for lithium ion batteries. Cells using LiFePO₄ as cathode material have been introduced successfully to the market. However, the energy density of these cells is significantly lower compared to lithium-ion systems using layered materials like NCM or NCA. Therefore there is a great interest in the development of high voltage olivine materials like LiMnPO₄ or LiCoPO₄ as cathode materials for high energy applications.  Cycling stability and power capability of the pure materials are not sufficient and the thermal stability of the fully charged material is still under discussion.

Improved electrochemical performance  have been obtained by using binary and ternary mixed phosphates LiM(1)_xM(2)_1-xM(3)_1-x-yPO₄. The presentation will summarize recent developments on Lithium Mixed Metal Phosphates and will highlight challenges and options to use them in real systems. The presentation will start with a brief overview on the pure materials LiFePO₄, LiMnPO₄ and LiCoPO₄ and their properties. Strategies to improve these materials by partial substitution of Manganese and Cobalt by other metals will be presented and discussed including their implications for:

• Energy density,
• Rate capability,
• Cycling life, Safety,
• Electrode processing,
• Reactions with other components like electrolytes, separator, binder

Cycling tests and safety tests of complete cells will be presented and future options for optimization will be discussed.

Nanotechnology is a revolutionary area that has considerably influenced materials science and technology. Reduced dimensionalities have profoundly impacted the electrochemical properties of materials for myriad electrochemical applications. Since Fuji identified tin oxide nanocomposite anodes, there has been a resurgence of activities focused at identifying new anodes to replace graphite. Various strategies have been researched comprising new intermetallic anodes, generation of nanostructured matrixes containing electrochemically active species upon electrochemical Li insertion as well as creation of nanowires, nanotubes (NT) and ‘core-shell’ structures.

Over the years we have synthesized nanostructured intra type nanocomposites (ITN) comprising direct ex-situ generation of active and inactive phases by exploiting low-cost synthetic approaches. Recently, we have developed new approaches to nanoscale sacrificial templates for generating nanostructured hollow silicon nanotubes (SiNTs). These hollow nanotubular configurations of amorphous and nanocrystalline Si confine the colossal volume expansion stresses within the nanoscale and disordered domains minimizing the catastrophic effects observed with cycling. The electrochemically inactive species comprising transition metal non-oxides (TMN), carbon and carbon nanotubes (CNTs) were selected due to their thermodynamic and electrochemical stability towards Li. The ex-situ generated “active-inactive nanocomposites” are fabricated by cost effective mechanochemical and high energy mechanical milling (HEMM)-based approaches as well as low temperature liquid injection chemical vapor deposition techniques (CVD) and electrodeposition techniques, while the hollow SiNTs exhibiting impressive electrochemical response were generated by amorphous Si deposition and dissolution of the sacrificial template.

Initial results show that the first generation ITNs exhibit stable capacities (1000 mAh/g) while novel hybrid heterostructures of vertically aligned CNTs (VACNTs) containing nanoscale Si exhibit impressive capacities (~3000 mAh/g). Similarly, the hollow silicon nanostructures exhibit a high first discharge capacity (~2420 mAh/g) at 300 mA/g current density when cycled in the 0.01-1 V vs. Li⁺/Li voltage range. A relatively high first cycle irreversible loss (24%) was observed due to possible SEI formation owing to the large Si NT surface area. Additionally, at 2A/g, the Si NTs exhibit excellent capacities in the 1300-1700 mAh/g range with 88% capacity retention after 50 cycles corresponding to a 0.23% loss per cycle capacity fade. An intriguing aspect is the active material-matrix interface interactions and stabilization of nanoscale morphologies contributing to stable reversible capacities. Opportunities, status, and challenges related to synthesis and design of these nanostructured systems for next generation Li-ion batteries will be presented and discussed.

Layered Li-rich transition metal oxides have theoretical charge and energy densities exceeding 250 Ah-kg^-1 and 900 Wh-kg^-1 when these materials are charged beyond 4.5V vs. Li⁺/Li. However, significant capacity loss and impedance rise is observed when these materials are repeatedly cycled or held at high voltages against graphitic negative electrodes in electrochemical full-cells. In this presentation, we investigate one particular system, containing Li_1.2Ni_0.15Mn_0.55Co_0.1O₂-based positive electrodes, and show how capacity loss and impedance rise can be reduced by modification of the positive electrode. This modification is done in several ways that include the following: i) coating alumina onto composite electrodes via atomic layer deposition (ALD); ii) by blending commercially available Al₂O₃ powder with the other electrode constituents; (iii) by preparing electrodes from oxide particles coated with alumina by a sonochemical process. Full-cells containing the alumina-coated particles and positive electrodes, cycled between 2.2-4.6V, show better capacity retention and lower impedance rise than the baseline electrodes. Electrochemical and physicochemical data will be presented to discuss mechanisms that lead to the observed improvement in electrode performance and cell life.

Rechargeable lithium ion batteries have been applied to electric vehicle (EVs) and the number of EVs is foreseen to grow substantially in the near future. However, their application is hindered by performance and cost. The major component of battery costs is materials and associated processing. Thus, it is essentially critical to develop low cost materials and material processing to reduce battery cost. Recently, there is growing interested in switching manufacturing composite electrodes from conventional N-methyl-2-pyrrolidone (NMP)-based processing to water-based processing due to both economic and environmental advantages. Lots of efforts have been made in dispersing solid components in water suspension, improving wetting of electrode suspension on current collectors, and using various water soluble binders. However, most of the work has been limited to LiFePO₄. LiNi_xMn_yCo_1-x-yO₂ is another promising cathode material showing higher voltage and volumetric energy density than LiFePO₄.For example, LiNi_0.5Mn_0.3Co_0.2O₂ (NMC532) was selected as bench mark cathode for the Applied Battery Research (ABR) program of the Vehicle Technology Office (VTO). Thus, it would be interesting to evaluate the performance of NMC532 cathodes through aqueous processing. In addition, most of the electrode performance with aqueous processing in literature is from half coin cells only, which is insufficient for evaluation. In this work, full half cells with either or both electrodes fabricated through aqueous processing was assembled and tested. Additionally, large format pouch cells (>500 mAh) were also be evaluated.

NMC532 showed negligible dissolution in water within a short period. For instance, lithium solubility was ~12 μg/mL when NMC532 was suspended in water for 3 hours. This indicated if dispersion are mixed, coated and dried with 3 hours, exposure to water should not cause metal solubility issues. Carboxylmethyl cellulose was selected as a dispersant based on zeta potential result with NMC532. NMC532 suspension was mixed with three water soluble binders. Their rheological properties of aqueous suspensions were characterized. NMC532 cathodes were coated by a slot-die coater. Half coin cells, full coin cells and large format pouch cells were assembled for performance evaluation. Graphite A12 (ConocoPhillips) was used as the counter electrodes in full coin cells and pouch cells. Rate performance and cyclic performance were evaluated. Their performance was compared to that from NMC532 cathodes fabricated through NMP based processing.

Excellent rate performance and cyclic performance was observed from NMC532 cathode through aqueous processing which is comparable to that from NMP-based processing.