Transformational Thick Cathodes with Hierarchical Pore Structure for Enabling High-Energy Lithium Ion Cells at Lower Cost (submitted to DOE)


Active-material specific capacities for lithium ion batteries are relatively high, but the cells suffer lower specific capacity due to the inactive separator and current-collector foil masses, which limit practical cell energy density to about 150-175 Wh/kg. For example, the typical capacity of a graphite/LiNixMnyCo1-x-yO2 cell is about 150 mAh/g based on active material mass. When factoring in the inactive components, this value drops to about 60 mAh/g, (60% reduction). One way to address this limitation is to increase the cathode coating thickness, which raises unit-cell energy density while simultaneously lowering battery pack-cost. This processing modification has been considered by industry for quite some time; however, implementation has not occurred because thick cathodes are not compatible with current inactive material compositions and dispersion processing.

Practical limitations of the coating process such as breakdown of coating integrity (particle cohesion), delamination of coating from current collector foil (particle adhesion), and excessive flaking as the web traverses the coating line impose a maximum calendered cathode thickness of about 100 mm. If the cathode thickness could be doubled in a graphite/ LiNixMnyCo1-x-yO2 cell, then the energy density would increase by 17% from about 60 mAh/g to 70 mAh/g (total cell mass basis), reducing cell cost considerably. In addition, the amount of geometric surface area (current- collector, electrode, and interfacial contact area) that would need to be managed would be halved. Using the expertise of Oak Ridge National Laboratory (ORNL) researchers and our university and industry partners - the University of New Mexico (UNM), 3M, XG Sciences, and Cabot Corporation - we will address these thick cathode shortcomings though novel materials integration, cutting-edge colloidal science, and advanced coating and drying technology. Another important problem associated with these thick coatings is that capacity during high discharge rates is low, which limits power density, due to lithium-ion mass transport limitations. Our innovative approach will address this issue by implementing a hierarchical pore structure, from interparticle macropores within a graded electrode architecture down to intraparticle micropores, for optimizing lithium-ion mass-transport in the liquid phase.

This project will primarily focus on the VTO Applied Battery Research (ABR) baseline electrochemical couple, which consists of a ConocoPhillips (CP) A12 graphite anode and a TODA America LiNi0.5Mn0.3Co0.2O2 (NMC 532) layered, mixed-transition-metal-oxide cathode. The major risks associated with this project are successful integration of simultaneous new materials into a single electrode design, optimization of the resulting complex dispersion chemistry with dissimilar materials, finding the right balance of heat removal during drying, and technology transfer management to ensure successful scaling of the cathode active material. Increasing cathode binder content and optimizing mechanical properties are two keys to enabling thicker cathodes of 200 mm (calendered). However, increasing the conventional PVDF binder content to levels of 10-12 wt%, which is likely required for thick cathode processing, introduces new problems such as reduced ionic and electronic conductivities. Lithium conducting ionomeric binders to boost ionic conductivity and multilayer graphene conductive additives to boost electronic conductivity will be employed together with novel, higher-capacity cathode active materials to achieve thick, high energy structures. Commercial-scale, high-purity graphene conductive additives will be provided by XG Sciences, developmental graphene with tailored microstructures will be provided by UNM, and a variety of commercial-scale ionomer binders with different equivalent weights and mechanical properties will be provided by 3M. The binders are in a class of polymers known as perfluorosulfonic acid (PFSA), and will be converted to the "Li form"; for use as lithium-ion conductivity enhancement at the interface between cathode active material and electrolyte.

To offset inherent lithium-ion mass-transport limitations introduced by cathode thicknesses approaching 200 mm, the combination of engineered microstructures of the cathode active material particles (i.e., particles with a tailored network of pore sizes) and graded electrode architectures will be implemented to achieve a hierarchical pore structure from mm-sized macropores (interparticle porosity) to nm-sized micropores (intraparticle porosity). To achieve the desired macroporosity, the ORNL technique of dual slot-die coating with controlled settling (i.e., employing two dispersions in a simultaneous coating deposition process with different pump speeds, solids loadings, viscosities, and particle-size distributions) will be employed to achieve a graded electrode structure with a smaller, narrower pore-size distribution at the separator interface and a larger, broader pore-size distribution at the current-collector interface. UNM technology that provides a hierarchical network of meso and micropores (with controllable morphology - pore structure, pores modality, surface area) within NMC 532 primary particles will be scaled from tens of grams to the multi-kilogram level by Cabot. The morphology control is achieved by implementation of different pore formers/modifiers (for example, urea, melamine or hydrogen peroxide). Large-scale synthesis by Cabot is based on spray pyrolysis (semi-continuous) or the sacrificial support method (batch-based).The unique combination of approaches between ORNL and UNM/Cabot will result in a completely optimized system of porosity from micron-scale to the nanoscale, which will consist of macroporous "highways";, mesoporous "avenues";, and microporous "side streets";, for enhanced lithium-ion transport at high C rates.

Finally, to further contribute to cost reduction, water-based processing of the cathode dispersions will be employed using ORNL 's extensive expertise in aqueous electrode processing for lithium ion batteries. ORNL has acquired extensive data over the past 3+ years on water-based processing of lithium ion electrodes and is well poised to extend this colloidal chemistry toolkit to this project. The liquid-phase, lithium-ion mass-transport limitations of the baseline cathode and improved engineered cathodes will be experimentally studied and modeled on the macro-, meso-, and micropore scales. From this effort, we will determine the material distribution, pore structure and transport properties, and active material access as a function of thickness as well as the anticipated ionic and electronic conductivity improvements arising from the new 'inactive ' materials. This critical modeling and ex-situ experimentation will provide feedback into the design process, as well as into the coating formulations and methods to achieve targeted structures and properties in the electrode layers. This technology will significantly increase gravimetric and volumetric cell energy density while preserving power density by introducing a 150-200 mm (calendered) cathode with an appropriately balanced anode.

The project performance goals are based on the EV Everywhere PHEV40 and EV targets and are 225 Wh/kg and 400 Wh/L. We expect an 17% and a 14% increase in gravimetric and volumetric energy density, respectively, which is based on current state-of-the-art values for graphite/NMC cells of 130-190 Wh-kg and 250-350 Wh/L (Dow Kokam 2012 DOE AMR). We will also achieve a cell cost reduction (through the combination of reduced current collector foil and separator and aqueous electrode processing) of 15.

The key technical risks associated with this project are: 1) successful integration of three new materials (graphene conductive additive, ionomer binder, and active cathode) into a single electrode design; 2) optimization of the resulting complex dispersion chemistry with dissimilar materials; 3) finding the right balance of heat removal during drying to preserve the graded electrode architecture; 4) technology transfer management between UNM and Cabot to ensure successful scaling of the hierarchical pore-structure cathode active material. Even though the risks are significant, DOE funding would have a great impact on this project 's success. The size of this project is too large for battery makers to undertake, as they currently have dedicated production equipment and their budgets are stretched too thin to take on major formulation chemistry and coating process activities.

The major facility that will be used for completion of this project is the new $3M ORNL Battery Manufacturing Facility (BMF) and was specifically designed with this type of project in mind. It is a world-class open-access facility that was co-funded by VTO, AMO, and ORNL, and houses capabilities to produce electrode coatings based on any chemistry and subsequent assembly of full pouch cells up to 6 Ah. The BMF is highly flexible and can handle many diverse projects simultaneously, and it is set up to work seamlessly between parallel milestones of DOE and collaborative industry projects. It is unique because of the ability to produce scaled pouch cell sizes in pilot quantities, which can be used to demonstrate commercial readiness of a wide variety of LIB technologies.

The ORNL BMF will be available 60-75This multidisciplinary team will be led by ORNL PI Dr. David L. Wood, III who has 17 years of experience in the design of advanced electrodes and related subcomponents and materials for polymer electrolyte fuel cells and lithium ion batteries. Dr. Wood is an international leader in addressing relevant problems related to widespread commercialization of polymer electrolyte fuel cells and lithium ion batteries and manages a $7-8M annual budget at ORNL to this end. His diverse responsibilities entail management of substantial DOE EERE VTO and AMO project funds for lithium ion electrode processing and cell assembly cost reduction, management of significant industry partnerships related to increasing lithium ion cells energy and power density, and program management for the EERE FCTO ORNL hydrogen and fuel cell activities. ORNL brings several major additional elements to this R&D team: 1) formulation and colloid design expertise; 2) lithium-ion electrode fabrication understanding; 3) coating drying expertise; 4) advanced processing simulation; and 5) large-scale pouch cell assembly capability (100 mAh - 6 Ah at the Battery Manufacturing Facility). The advanced materials processing team members (David Wood, Beth Armstrong, Jianlin Li, Claus Daniel, and Adrian Sabau) will be responsible for colloidal science, dispersion formulations, electrode coatings, modeling of electrode drying, oven temperature profile mapping, and calendering, and the applied energy storage group (David Wood, Jianlin Li, Claus Daniel, and Tom Zawodzinski) will be responsible for electrode architecture design, cell (half-cell, coin, and 100 mAh to 1 Ah pouch) assembly, and testing management.

Using ORNL 's world-class materials characterization facilities, we will have a designated team member for novel materials and electrode structural characterization. Zawodzinski 's research group has access to the needed capabilities for studying transport and structure of the electrode layers and the modeling framework to describe the performance as a function of composition using measured transport and structural parameters as input, as well as fundamental electrochemical characterization. The University of New Mexico (UNM) collaborators (led by Plamen Atanassov) have 10+ years of experience in designing the engineered active material particles at the tens-of-grams scale and novel graphene conductive additives required for this project.

Statistical structure-to-property correlation of the NMC 532 active materials will be contributed, as well as electrochemical testing parameters for optimizing the design of active material. Through multivariate modeling of material design parameters obtained from spectroscopic, microscopic, electrochemical methods (i.e., chemical structure of electrode material, thickness, the binder content, mechanical properties, morphology and porosity quantified on the macro-, meso-, and micropore scales) against performance parameters (ionic and electronic conductivity and gravimetric and volumetric cell energy and power density), an optimal path towards accelerated design of the final design NMC 532 active materials will be obtained.

The industry partners 3M, XG Sciences, and Cabot Corporation bring the critical component of providing commercial-ready, novel electrode materials at the required scale to enable project success. Without their contribution, ORNL would not be able to demonstrate this technology at the 1-Ah pouch-cell level required for acceptance by lithium ion battery manufacturers. The industry partners will provide a combined 7.1% in-kind cost share.