Session A26: Focus Session: Physics of Energy Storage Materials - Materials, Stability and Transport

Nanostructured Materials for Portable and Stationary Energy Storage

Storing energy electrochemically involves electronic and ionic processes and chemical transformation inside and at the interface of materials. The ability to understand and design nanostructures and their interfaces afford the great opportunities for controlling these fundamental processes, which can ultimately lead to high performance energy storage devices. Here I will present several exciting examples on designing nanostructures and their interfaces to realize high performance energy storage devices. One example is on designing nanowires and heterostructured nanowires for ultrahigh capacity storage of lithium ions in silicon anodes and sulfur cathodes. The challenges associated with large volume expansion, electron and ion transport, and solid-electrolyte-interphase (SEI) have been addressed. Another example is to design open framework stucture of nanocrystals, which facility facile insertion of sodium and potassium ions. The high power, high energy efficiency and low-cost aqueous batteries can be enabled for grid scale stationary storage.   

Enhanced electrochemical performance in LiFePO4/graphene nanocomposite cathode material for lithium ion batteries

We synthesized LiFePO$_{4}$/graphene nano-composite employing a sol-gel method, where graphene oxide solution was added to the LiFePO$_{4}$ precursors during the synthesis. Electrical measurement reveals that the addition of 10{\%} graphene (by weight) to LiFePO$_{4}$ increases its conductivity by 5 orders of magnitude. SEM images of the composite show that the material consists of LiFePO$_{4}$ nanoparticles (with a mean particle size $\sim $ 50 nm) homogeneously mixed with graphene sheets; the latter provides a three-dimensional conducting network for Li+ ion and electron transport. A large specific capacity of 170 mAh/g was observed at a discharge rate of C/2. To further increase the conductivity and inhibit particle size growth of LiFePO$_{4}$ (thus to increase the rate capacity), we coated the nanoparticles with a thin carbon layer by adding 0.25M lauric acid as precursor in addition to graphene oxide during the synthesis. The respective roles of graphene and lauric-acid-induced carbon coating in the specific capacity and charge-discharge rate of the LiFePO$_{4}$ cathode material will be discussed.   

Simultaneous enhancement of electronic and Li+ ion conductivity in LiFePO4

The electronic conductivity is generally highly sensitive to the electronic properties of a material while the ionic conductivity is mainly sensitive to the structural properties. Thus, the simultaneous control of both is very challenging. Furthermore, many electrochemical systems for advanced energy technologies require materials in which both ionic and electronic conductivities are optimized. Here we explore the influence of biaxial strain on the electronic and Li$^{+}$ ion conductivities of the battery cathode material LiFePO$_{4}$ by performing first-principles calculations. We show that biaxial tensile strain leads to simultaneous increases in electronic and ionic conductivities of LiFePO$_{4}$. The results unveil a generic mechanism that should be present in other materials with polaronic transport and provide a promising new approach to enhance the performance of energy-related materials.   

Optimization of the surface stability of the LiMn$_{2}$O$_{4}$ spinel by employing DFT calculations

One of the most important materials for the lithium batteries electrodes is LiMn2O4. We used GGA+U method to calculate the bulk and surface properties of LiMn2O4. Our calculations show that the both correct AFM and electron localization (GGA+U) are necessary to obtain the semiconducting, Jahn-Teller distorted electronic ground state of LiMn2O4. Further we calculated energies of different surfaces such as (100), (110), and (111) to study their stability. Our calculations show that (111) surface has the lowest energy which makes it more stable than other surfaces and it also confirms the experimental results, whereas (101) and (001) similar energies. Absolute surface energies change with +U value, but the ratios between the energies are very similar. Based on these calculations we constructed the equilibrium (Wulff) shape of LiMn2O4 particle, which is similar to the cubo-octahedral shape with predominant {\{}1~1~1{\}} facets as it was found in experiments. Our density of states calculations show that the bulk and (100) are semiconducting, whereas (110) and (111) surfaces exhibit metallic behavior. We also calculated the LiMn2O4 bulk and surface potentials as a function of lithium concentration.   

Atomistic Simulation Studies of Nanostructured TiO$_{2}$

TiO$_{2}$ has been confirmed as a safe anode material in lithium ion batteries due to its higher Li-insertion potential, ($\sim $1.5V) in comparison with commercialised carbon anode materials. Recently in order to attain high rate capabilities of TiO$_{2}$ anode, for application in lithium ion batteries with both high power and high energy density, intensive attention has been paid to various TiO$_{2}$ nanostructures, such as nanoparticles, nanowires and mesoporous structures. In the current study, amorphisation recrystallization method is used to produce nano- porous, sheets and bulk structures for TiO$_{2}$ which have been extensively studied experimentally. Simulated X-ray diffraction patterns are produced from such structures and compared with the experimental XRDs. The simulated microstructures are analysed and compared with available high resolution transmission experimental results. Lithiation of TiO$_{2}$ nanostructures is considered and discussed in the context of current investigations and concentration profiles of different ions are shown in the structures. Semi-empirical models based on DFT tightbinding methods, are introduced for TiO$_{2}$ and lithiated its forms.   

First-principles study of electronic structures and phase transitions of lithiated molybdenum disulphide

By first-principles calculations, electronic structures of MoS2, intercalation-induced 2H to 1T phase transition and reversibility are investigated. It is revealed that change of interlayer stacking from 2H to 3R imposes negligible influence on the band structure and stability of MoS2. In contrast, the change of intralayer stacking from 2H to 1T changes the character of p-d repulsion, resulting in a semiconductor-to-metal transition. We demonstrate that the Kohn-Sham band energy, rather than the coulomb repulsion energy, plays dominant roles in both the phase stabilization and transition during Li intercalation. It is found that the evolution of 1T phase is crucially determined by chemical hardness, which underlies the cycle irreversibility. Due to the charge-density-wave (CDW) phase, Li extraction is impeded by the enhancement of Li-host binding. It is indicated that the cycle reversibility can be improved by electron-donor doping in MoS2, because the resultant pre-reduction of Mo and S eliminates the electron transfer from Li to host.   

Simulating Dendritic Formation on the Cathode of Lithium-ion Batteries

The Lattice-Boltzmann method was used to simulate the process of dendrite formation on the cathode of lithium-ion batteries. ``In-Situ'' maps of the electrodeposition process under constant charging current conditions were obtained. The results showed preferential dendrites formation on higher curvature spots of the cathode, because of the higher electrical field intensity. Different morphologies were obtained due to different initial roughness of the electrode. A mossy-like electrode can be observed after deposition on a rough electrode with randomly generated initial conditions. A tree-like dendrite can be observed when depositing on a single small dendrite. A smooth surface can be obtained when the initial electrode is ideally smooth. The influence of current density was also studied.   

Density functional and molecular dynamics studies of solid electrolyte Li$_7$La$_3$Zr$_2$O$_{12}$

Garnet-type structured Li$_7$La$_3$Zr$_2$O$_{12}$(LLZO) is considered as a promising candidate for Li-ion battery solid electrolytes because of its high ionic conductivity and electrochemical and chemical stability. We use first-principles density-functional theory calculations and molecular dynamics simulations to reveal the underlying mechanism that drives a tetragonal to cubic transition at elevated temperatures, and also to explain why the cubic phase can be stabilized with the incorporation of a certain amount of impurities such as Al. We show that the relationship between the observance of a cubic phase and the measurement of a substantially higher ionic conductivity is a secondary effect not directly attributable to the presence of Al in the crystal structure. Suggestions for enhancing the ionic conductivity in LLZO will also be discussed.   

First principles investigation of the superionic electrolyte Li$_7$P$_3$S$_{11}$

Li$_7$P$_3$S$_{11}$ has been shown to be a promising superionic conductor for solid state rechargeable batteries with a room temperature conductivity as high as $10^{-3}$ S/cm and a thermal activation energy as small as $E_A$=0.12 eV.\footnote {F. Mizuno et al., {\em{Solid State Ionics}} {\bf{177}}, 2721 (2006).} We have performed first principles modeling studies\footnote{N. A. W. Holzwarth, N. D. Lepley, Y. A. Du, {\em{J. Power Sources}} {\bf{196}}, 6870 (2011).} on this material in order to explain its stability and Li ion migration properties. Our investigation considers optimized crystal structures, migration involving both vacancy and interstitial mechanisms, as well as related materials. We find optimized crystal structures in reasonable agreement with experiment,\footnote{H. Yamane et al., {\em{Solid State Ionics}} {\bf{178}}, 1162 (2007); Y. Onodera et al., {\em{J. Phys. Soc. Jpn.}} {\bf{79}}, 87 (2010), suppl. A. } and the lowest calculated activation energy barrier was found to be $E_A$=0.15 eV in good agreement with the experimental value.   

Solid Electrolyte for Advanced Lithium Batteries

Lithium battery is a promising energy storage system due to its high energy density and high rate capability and its application ranges from micro to large scale megawatt batteries. The current technology is using liquid electrolyte that limits its application due to flammable nature of the electrolyte, particularly at high temperatures, and difficulty in fabrication and miniaturization of the device. We report a novel solid electrolyte with high lithium ion conductivity as a replacement for the current liquid electrolyte, particularly for electronic applications. The solid state lithium ion conductor is based on lithium germanium phosphorous sulfide compound. The compound is prepared by solid state reaction at 500 \r{ }C. The crystallinity and phase purity of the sample is checked by XRD. We also measured ionic conductivity of the sample using both 4-probe and impedance techniques. High lithium ion conductivity at room temperature is observed. In this study we have investigated the dynamics of ion conduction, XRD, and Raman spectra of the super ion conductor. Electrochemical performance of the solid electrolyte in a lithium cell and its stability against high voltage cathodes and lithium anode will also be presented.   

First principles computer simulations of Li$_{10}$GeP$_2$S$_{12}$ and related lithium superionic conductors

A recent paper by Kamaya {\em{et al.}}\footnote{ N. Kamaya {\em{et al.}}, {\em{Nature Materials}} {\bf{10}}, 682 (2011).} reported a new crystalline superionic conductor having a compact tetrahedral structure and a stoichiometry of Li$_{10}$GeP$_2$S$_{12}$. The room temperature conductivity was reported to be 0.01 S/cm, comparable to liquid electrolyte conductivies and five times higher the compositionally related thio-LISICON material Li$_{3.25}$Ge$_{0.25}$P$_{0.75}$S$_4$ developed earlier.\footnote{R. Kanno {\em{et al.}}, {\em{J. Electrochem. Soc.}} {\bf{148}}, A742 (2001).} This talk will present a progress report on our work to perform first principles computer simulations for these materials focussing on the structural, stability, and Li ion mobility properties of idealized crystalline models. From the perspective of our previous studies of Li ion conductivity in lithium thiophosphate electrolytes,\footnote{ N. A. W. Holzwarth {\em{et al.}}, {\em{J. Power Sources}} {\bf{196}}, 6870 (2011).} the effects of introducing Ge can be assessed.   

Exploring Li+ Potential Energy Surface in Poly(ethylene oxide)-based Sulfonate Ionomers

Ion-containing polymers are of interest as single-ion conductors for use as electrolytes in electrochemical devices, including lithium ion batteries. Current ion conductivities of the best ionomers are roughly 100X too small for practical applications and have a small fraction of their Li+ counterions participating in conduction. We are using ab initio methods to investigate the Li+ conduction mechanism, and specifically the role of transient positive triple ions (Li+A-Li+) in the conduction process. The positive triple ion has a lower energy separated state that allows for facile transport, if there is a pair within 1.4 nm. We will discuss the competition between cation solvation with ether oxygen atoms and cation-anion interaction. The importance of anion-anion separation in altering Li+ hopping barriers will be examined, as well as the variation in hopping rates with solvent identity. Ab initio calculations are used to evaluate the relative energy of ion states (contact and separated states), and this analysis is used to explain experimental phenomena of Li+ mobility in ionomers.   

Ab initio molecular dynamics simulations of organic electrolytes, electrodes, and lithium ion transport for Li-ion batteries

Optimizing the choice of electrolyte in lithium ion batteries and an understanding of the solid-electrolyte interphase (SEI) is required to optimize the balance between high-energy storage, high rate capability, and lifetime. We perform accurate ab initio molecular-dynamics simulations of common cyclic carbonates and LiPF6 to build solvation models which explain available Neutron and NMR spectroscopies. Our results corroborate why ethylene carbonate is a preferred choice for battery applications over propylene carbonate and how mixtures with dimethyl carbonate improve Li-ion diffusion. We study the role of functionalization of graphite-anode edges on the reducibility of the electrolyte and the ease of Li-ion intercalation at the initial stages of SEI formation. We find that oxygen terminated edges readily act as strong reductive sites, while hydrogen terminated edges are less reactive and allow faster Li diffusion. Orientational ordering of the solvent molecules precedes reduction at the interphase. Inorganic reductive components are seen to readily migrate to the anode edges, leading to increased surface passivation of the anode. We are currently quantifying Li-intercalation barriers across realistic SEI models, and progress along these lines will be presented.