Symposium EE4 : Electrode Materials and Electrolytes for Lithium and Sodium Ion Batteries

Na-ion battery as alternative high-efficiency and low-cost energy storage device to Li-ion battery has attracted wide interest for electrical grids, vehicles and portable electronics applications. However, demonstration of a full-cell battery with high energy (>300 Wh/kg-cathode) and long cycle life (>1000 cycles) remains a significant challenge. Here, we investigated the role of solid electrolyte interphase (SEI) formation on both cathode and anode materials and provided a potential pathway to achieve long term stability for Na-ion battery full-cells. Pre-cycle of Na_0.44MnO₂ cathode and hard carbon anode will lead to pre-formation of SEI on electrodes surface and hence mitigated electrolyte consumption in full-cells. The Na_0.44MnO₂-hard carbon full-cell with matched cathode-to-anode capacity and pre-formed SEI can deliver specific capacity of ~116 and 108 mAh/g based on Na_0.44MnO₂ at 1C and 2C rate (1C=120 mA/g). The corresponding specific energy is ~313 and ~286 Wh/kg, respectively. Superb cycling stability with capacity retention of ~77% and energy retention of ~74% over 2000 cycles at 2C rate was demonstrated.

High capacity anode material candidates for lithium-ion batteries, such as silicon, germanium, and tin, experience extreme and unavoidable expansion and contraction during the lithiation and delithiation processes. These volumetric changes lead to rapid morphology deterioration of the electrode materials (cracks, electrical isolation of particles, pulverization, etc), which severely limits the battery lifetime to a few charge-discharge cycles. While the use of nanoparticles or complex structures can avoid this volumetric deterioration, these strategies tend to be expensive and do not easily handle larger mass loadings. Instead of avoiding deterioration, lessons from nature suggest that utilizing self-healing functionality may provide the answer to stable battery operation. The idea of self-healing has recently been demonstrated to improve anodic cycling performance of low cost micron-sized silicon particles (SiMPs) via utilization of a Self-Healing Polymer (SHP), in lieu of traditionally used binders. This novel concept has shown to greatly increase the cycle life, even for thick electrodes, as compared to traditional polymer binders.
While the first iteration of the SHP-binder is a great step towards next-generation LIB, the function of self-healing within the battery anode under operation is not well understood. To this end, transmission x-ray microscopy (TXM) and small angle x-ray scattering (SAXS) were used to monitor operando structural changes of the SiMPs within the SHP-SiMPs anodes. Tracking of contrast and structural changes of individual SiMPs in images at various points along the first two charge/discharge cycles give key insights into the effects of the SHP binder. Furthermore, particle size changes in thick SHP-SiMPs anodes layers indicate maintained particle electrical connectivity during cycling. The data gathered from this study aids our mechanistic understanding of the self-healing process that occurs within the operating battery and provides certain design rules for enhanced performance in next-generation Si-based anode composites.

For portable and mobile electronic devices, improvements in current lithium ion batteries (LIBs) are urgently needed. We believe that in-operando observation at near-atomic spatial resolution, combined with millivolt electron energy-resolution spectroscopy (EELS) at the same spatial resolution, will prove the most powerful tool for understanding the atomic processes involved, and so suggest the means to improve performance. Therefore, we introduce a novel technique to study atom transport driven by local electric field in TEM/STEM and thus to simulate in situ atomic transport in a battery operation with sub-nm spatial resolution. The driving force for Li ion transport is the local electric field, which is induced by accumulated positive charges during exposure to the electron beam in the insulating electrolyte or anode materials [1]. These positive charges are produced by the ejection of secondary electrons (SEs) and Auger electrons due to excitations and ionizations by the incident electrons. The main advantage of this method is its inexpensive and easy setup for experiment. It is different from the method used in the conventional in situ TEM/STEM studies of battery materials, in which the driving forces are externally applied electric fields by the power supply, and thus a specialized TEM holder is required to set up a LIB cell.
The method has been applied to investigate the Li transport and lithiation of amorphous C [2] and Li ion migration in Li₄Ti₅O₁₂ anode under local electric field [3]. In the former, it was discovered that only a small portion of Li reacts with C forming Li-C compounds, while the majority of the Li is in form of “free” Li⁺ ion state. Spinel Li₄Ti₅O₁₂ has also been attractive for its usage as the anode in LIB due to its excellent cycling stability and accommodation of Li. However, under strong local electric fields, the original tetrahedral-coordinated Li ions migrate to the octahedral sites, resulting phase transformation from spinel to a random rocksalt structure. This transition is irreversible. In both cases, it demonstrated that the Li transport and migration can be achieved under electron irradiation alone, without an external power supply. Therefore, this work also raises reasonable doubts about operating a full LIB cell under electron beam illumination in TEM/STEM.
[1] N. Jiang, Report on Progress in Physics (2015), in press.
[2] N. Jiang, J. Mater. Res. 30, 424 (2015).
[3] D. Su, F. Wang, C. Ma, and Jiang, N., Nano Energy 2, 343 (2013).

The safety problems associated with conventional lithium ion batteries are largely related to electrolyte decomposition. One approach to address this problem is to replace the current liquid electrolyte with inorganic solids. A solid electrolyte must have high lithium ion conductivity, as well as chemical and electrochemical compatibility with the electrodes; however, decades of research have yet to discover a commercially viable replacement for organic liquid electrolytes. A promising technique is to use a combination of solid materials instead of a single material. Paired with a highly conductive separator, a bi- or tri-layer electrolyte offers a method to meet the intense demands on a solid electrolyte. Here we report our results on a new class of materials: doped Li₉S₃N as an anode protective layer. [1] We combine first principles calculations and experimental confirmation to support the use of this material in all solid state batteries. Further, we demonstrate a technique for identifying interphase layers when pairing the garnet electrolyte with several common cathode materials.[2]
[1] L. J. Miara, N. Suzuki, W. D. Richards, Y. Wang, J. C. Kim, and G. Ceder, “Li-ion conductivity in Li₉S₃N,” J. Mater. Chem. A, vol. 3, no. 40, pp. 20338–20344, Oct. 2015.
[2] L. J. Miara, W. D. Richards, Y. E. Wang, and G. Ceder, “First-Principles Studies on Cation Dopants and Electrolyte|Cathode Interphases for Lithium Garnets,” Chem. Mater., vol. 27, no. 11, pp. 4040–4047, Jun. 2015.

Currently, lithium-ion batteries incorporated in electric vehicles do not allow to store enough energy to ensure sufficient autonomy. Therefore, it’s necessary to develop new generations of batteries with higher energy density. Among battery technologies, the lithium-sulfur system (Li-S) is considered to be one of the most promising solutions due to a theoretical energy density of 2500 Wh/kg. However, its lifetime remains limited due to the polysulfide shuttle effect, induced by the use of liquid electrolyte. Indeed, many polysulfides (noted Li2Sx 1 ≤ x ≤ 8), formed during the sulfur reduction to Li2S, dissolve in liquid electrolyte and diffuse between both electrodes, leading to a rapid decrease of electrochemical performances.

All-solid-state Li-S batteries were studied in order to bypass the limitations of liquid electrolyte. The electrode is a composite material with sulfur (S8) as active material, argyrodite (Li6PS5Cl) as solid electrolyte for ionic conduction and Ketjen Black (KB) as additive conductor for electronic percolation. The liquid electrolyte between both electrodes is replaced by the solid electrolyte used in the composite electrode and metallic lithium is used as negative electrode.

The positive electrode was prepared by mixing and ball-milling at 370 rpm for 5h using a high-energy planetary ball mill, S8, Li6PS5Cl and KB with a weight ratio of 25:50:25. The battery displays a reversible capacity of 1565 mAh/g after 10 cycles at a current density of 134 μA/cm² (corresponding to a rate of C/10) with an initial discharge capacity of 1230 mAh/g and an extra capacity of around 335 mAh/g, observed for the first time and currently under studies.

At high current density, with lithium negative electrode, we observed strange potential decreases that can be related to short circuit phenomena. One explanation for short circuit phenomena can be the formation of lithium dendrites, although they are not supposed to grow in solid electrolyte. The presence of dendrites will be evidenced by solid state NMR. Using alternative Li-In electrode, this phenomenon was totally suppressed. In this presentation, we will then show our best optimized solid state batteries using a Li-Indium electrode with a good capacity (1660 mAh/g after 10 cycles at C/10), a good coulombic efficiency and a reasonable polarization for a solid-state battery and at room temperature. Perspectives in term of energy density and industrialization (upscale of the lithium ion conductors) will be presented.

The significant growth of both the number and the kinds of portable electronic systems, generally battery-powered has triggered the race for the development of high-performance microprocessors and systems on chips using low power consumption integrated circuits. As a consequence, the energy supply of such optimized components can also be ensured today by miniaturized power sources such as Li microbatteries. The latter have the advantage of being manufactured by vacuum deposition process also widely used in the microelectronics industry [1]. Nevertheless to be considered as an electronic components, the microbattery need to sustain temperature higher than 230°C for the reflow soldering to attach the component to the PCB . As Li metal anode melts at 181°C, Li microbatteries are not tolerant to such high temperature. For this reason we have manufactured Li-ion all-solid-state thin films batteries based on Si at the negative electrode and a new lithiated titanium oxysulfide as a cathode [2]. This new thin film material was prepared by sputtering from a home-made LiTiS₂ target. The resulting films are quite dense, amorphous with a composition close to Li_1.2Ti^IVO_0.5S^II_2.1. A full electrochemical delithiation of this electrode corresponding to a capacity of 64 µAh.cm^-2.µm-¹ was achieved in all-solid-state cells. As Ti⁴⁺ cannot be involved in the oxidation process, only sulfide species are oxidized during charge up to 2.9 V/Li⁺/Li forming S₂^2- disulfide anions. The subsequent lithiation and delithiation curves showing no capacity loss, demonstrate the perfect reversibility of the electrochemical processes. Actually, when cycled in the 3.2-1.5 V voltage window versus a lithium anode, extra lithium can be inserted in the material below 2V and titanium is partly reduced. The increase of the current density from 2 to 130 µA.cm^-2 (i.e from C/50 to 2C rates) induces only a slight capacity decrease, mainly due to an increase of the polarization near the end of charge. Besides, the capacity retention is found to be excellent during extended cycling with a capacity fading lower than -0.01 % per cycle and a mean coulombic efficiency close to 100%. Excellent electrochemical performances in terms of capacity value and cycling stability were also obtained in all-solid state Li-ion cells with Si as a negative electrode. After three successive solder-reflow processes (thermal treatment at 260°C), the capacity, the cycle life as well as the voltage profile remained unchanged. Finally, Li_1.2TiO_0.5S_2.1 is confirmed as a new interesting lithiated positive material for thin film lithium and lithium-ion microbatteries. Moreover, as its synthesis does not require any thermal treatment (contrary to the well-known LiCoO₂), is well-adapted to thermally sensitive substrates such as flexible polymers foils.

References:
[1] B. Pecquenard, F. Le Cras et al., J. Power Sources, 196 (2011) 10289
[2] F. Le Cras, B. Pecquenard et al., Adv. Energy Mater., 5 (2015) 1501061

The significant growth of portable electronics, generally battery-powered, has triggered the race for the development of high-performances microprocessors, Systems-on-a-Chip or DRAM, using low power consumption integrated circuits. As a consequence, the energy supply of such optimized components can be operated today also by miniaturized power sources such as microbatteries. Up to now, Li/LiPON/LiCoO₂, is the most used system, supplying a 4.2 V voltage. Nevertheless, their 4V operating voltage is not convenient for all applications, which may require often significantly lower ones. The replacement of Li/gamma-MnO₂ primary batteries by thin films batteries in selected devices has motivated the development of specific 3V rechargeable systems [1]. Among the possible positive electrode material able to reversibly insert/deinsert Li at 3V/Li⁺/Li or Na at 2.7 V/Na⁺/Na [2], iron molybdate Fe₂(MoO₄)₃ thin films appear as a suitable candidate as they combine the following features: ability to reversibly insert/deinsert Li or Na in bulk form due to a bidimensional open framework, low cost and environmental friendliness.
In the present study, radio-frequency sputtering, which is a versatile technique and a process widely used in the microelectronic industry was used to prepare iron molybdate thin films. After an optimization of the sputtering conditions especially the power, oxygen partial pressure and the total pressure, we have focused our thorough physico-chemical and electrochemical characterizations on stoichiometric films annealed at various temperatures between 200 and 500°C. The corresponding film with a monoclinic structure was then thoroughly studied (XRD, XPS, RBS, SEM, EPMA, Raman and Auger spectroscopies). Thin films electrodes were set in coin cells with liquid electrolytes for electrochemical characterization with metallic Li or Na as a negative electrode. The voltage curves displayed a plateau with a low hysteresis and a good cycle life (with a capacity corresponding to 96 % of the theoretical capacity). When changing from Li to Na, a modification of the shape of the voltage curve highlights a modification of the intercalation mechanism, from a biphasic to a monophasic process. XPS and Mössbauer analyses were performed on the inserted phases to investigate further the reaction mechanisms. Finally, all-solid-state cells were tested with Li as a negative electrode.

References:
[1] S. Cotte, B. Pecquenard, F. Le Cras, R. Grissa, H. Martinez, L. Bourgeois, Electrochimica Acta 180 (2015) 528-534
[2] A. Nadiri, C. Delmas, R. Salmon, P. Hagenmuller, Rev. Chim. Min, 21 (1984) 537-544

All-solid-state batteries with non-flammable solid-state electrolytes have significant advantages over Li-ion or Na-ion batteries with liquid electrolytes. In this talk we will present our recent advances in understanding the fundamental relationship between crystal structure and ionic transport in fast Li-ion conductors [1]. We find that the topology of the sites along the lithium hopping path, and the volume per anion, largely control the activation energy for Li motion. This topology is more favorable in compounds where the anions form a bcc lattice than in the typical close-packed fcc and hcp anion arrangements, explaining why most compounds are not fast-ion conductors. Furthermore, an accurate and efficient first-principles computational methodology has been developed to evaluate thermodynamic stability of the solid-state electrolyte against battery electrodes. These findings not only provide valuable insights towards the understanding of materials behaviors in discovered ionic conductors, but also serve as design principles for new Li and Na conducting materials and all-solid-state batteries. We will also report a few novel Li and Na ionic conductors that are predicted based on our design principles and experimentally confirmed by our collaborators.
[1] Y. Wang, et al., “Design principles for solid-state lithium superionic conductors,” Nat. Mater., 14, 1026-1031, (2015).

Lithium iron phosphate (LiFePO₄) has received much interest as a cathode material on account of its good reversibility, low toxicity, thermal stability, and high material abundance [1]. However, there are some key fundamental drawbacks in that it is a poor ionic and electronic conductor, which are apparent in the rate performance. Understanding the intercalation mechanism will provide insight on some of the core limitations of this material. Two-phase and solid solution mechanisms have been proposed, but details on the transformation at the single particle level remain elusive [2]. X-ray nanodiffraction measurements on single particles were carried out to produce direct observations of the electrochemical reaction at a fundamental level.

Argonne National Laboratory’s Advanced Photon Source boasts a hard x-ray nanoprobe beamline that has a very high spatial resolution of ~30nm [3]. Traditionally, this beamline has been used to study thin films. Here, its power was harnessed to observe single microcrystals in a powder sample. Two-dimensional area maps were taken of one particle each of LiFePO₄, FePO₄, and an intermediate average powder composition Li_0.5FePO₄, which contained both fluorescence and diffraction data. Combining the fluorescence and diffraction data, the phases involved in the LiFePO₄-FePO₄ transformation were identified within the particle from the distance between lattice planes, calculated with Bragg’s Law.

Identifying the grain boundaries between phases present in the particle highlights the effects the intercalation mechanism has on the stability of the material. In addition to phase identification, features such as strain and inhomogeneities were also identified with the utmost spatial resolution from the diffraction experiments. Visualizing these phenomena on the single particle level and how they correlate to each other sheds light upon the electrode operation and how battery failure ultimately occurs. This information may lead to advancements in this material as well as those with similar structures.


[1] Gibot P., Casas-Cabanas M., Laffont L., Levasseur S., Carlach P., Hamelet S., Tarascon J.-M., Masquelier C.,
Nat. Mater., 7 (9), 741-747, 2008.

[2] Meethong N., Shadow Huang H.-Y., Carter W. C., Chiang Y.-M., Electrochem. Solid- State Lett., 10,
A134-A138, 2007.

[3] Winarski, R.P., Holt, M.V., Rose, V., Fuesz, P., Carbaugh, D., Benson, C., Shu, D., Kline, D., Stephenson, G.B., McNulty, I., Maser, J., J. Syn. Rad., 19, 1056-1060, 2012.

Solid state electrolytes are critical enabling elements of emerging electrical energy storage technologies, such as sodium-based batteries. Realizing the promise of these developing technologies, however, depends on the development of chemically and structurally robust solid state electrolytes, stable against a variety of molten, organic, and even aqueous media. Here we explore the materials chemistry of a family of NaSICON (Sodium Super Ion CONductor) ceramics with the composition Na_1+xZr₂P_3-xSi_xO₁₂ (0≤x≤2), which show excellent low temperature sodium ion conductivity. Despite the promising stability of these ceramics against materials such as molten sodium and a range of organic electrolytes, there remain potentially significant materials stability challenges to using these materials in next generation technologies that might employ, for example, aggressive aqueous media. Here we specifically discuss potential “weak links” in the NaSICON phase chemistry and ceramic composition that make it potentially susceptible to degradation both in bulk and in novel thin film morphologies. Moreover, we describe our efforts to tune these characteristics during synthesis to produce improved NaSICON stability in aggressive electrolyte environments. Advancing the development of highly conductive, chemically stable solid state ion conductors stands to open the door to a range of innovative new sodium-based energy storage technologies needed to meet the ever-growing demand for safe, reliable, and effective electrical energy storage and delivery.

Sandia National Laboratories is a multi-program laboratory operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Company, for the US Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000.

Although Li-ion batteries (LIBs) are powering electric cars (GM’s Volt), problems still exist in terms of performance, mainly because current LIBs feature natural graphite (NG)-based anodes, which exhibit large irreversible capacity loss and short charging-discharging cycling life induced by chemical reactions between Li-ion inserted NG and organic electrolyte. This report shows that this problem is overcome by using a novel nitrogen-incorporated ultrananocrystalline diamond (N-UNCD) encapsulated NG/copper complex anodes. N-UNCD films (~5-10 nm grain size) exhibit electrically conductive grain boundaries induced by nitrogen incorporation into the boundaries, satisfying carbon sp² dangling bonds and providing electrons for conduction. In addition, N-UNCD films provide excellent chemical inertness, electronic/lithium ionic conductivity, and excellent lithium ions permeation between the electrolyte and the N-UNCD encapsulated NG matrix of NG/Cu anodes. N-UNCD enables robust solid electrolyte interphase (SEI) formed by anode/electrolyte chemical reactions and suppresses stress induced cracks of the SEI and NG particles and the subsequent loss of anode conductivity. X-ray diffraction and chemical analysis proved the integrity of N-UNCD/NG/Cu anodes and provided understanding of the new robust anode performance to enable next generation LIBs with potentially 10x longer life and superior performance than current LIBs.

With growing demands for high performance rechargeable batteries, metal alloy-based anodes are receiving great attention due to their high theoretical capacity. Among them, the most promising material is referred to silicon whose theoretical capacity is 3579 mAh/g. However, poor cycling stability caused by catastrophic volume change of 400 % during de-/lithiation is the major obstacle for the practical application of Si. To overcome the issue, nanostructuring or surface modification has been suggested by previous researches. However, they reported reduction in the capacity due to the electrochemical-inactive coating material, while the capacity retention was improved. Herein, a novel surface modification method to design high performance Si-based anodes was suggested. Based on modified RF sputtering system with vibration motors held by the holder, amorphous Si (a-Si) was uniformly deposited on the nanocrystalline Si (nc-Si) powders. The newly-designed a-Si/nc-Si anodes show distinct strong points. First, a-Si has higher robustness and resistance to failure than crystalline. Furthermore, a-Si contributed to the energy storage process unlike widely used coating material, e.g., oxides.
Nc-Si powder was used as the starting material for the P doped n-type amorphous Si. The average powder size of nc-Si was approximately 50 nm. The sputtering time and power were varied from 5 min to 3 h and from 50 W to 200 W to optimize the thickness of coating layer with the fixed Ar pressure and each RF power. The particle size and surface morphology were investigated by SEM and TEM. The effect of a-Si coating was analyzed by XRD and XPS. The electrochemical performances was investigated with coin-type half cell using 1 M LiPF₆ in an EC : DEC electrolyte.
We successfully controlled the thickness of a-Si layer with high uniformity according to the sputtering time and power. The existence of a-Si was verified by XRD peak compared to bare Si powder and the P atom was confirmed by XPS. Compared to bare Si, a-Si/nc-Si generally showed enhanced capacity and retention due to the modification effect. However, as the sputtering time and power exceed over the condition for critical thickness, the performance has been degraded. It can be explained by the crystallization of coated Si with increasing layer thickness. The best performance condition was obtained at 150 W, 5 min coated a-Si/nc-Si anode. It maintained superior capacity over 1500 mAh/g and C_50th/C_5th performance up to 75 %. A lower resistance in EIS and improved rate capability compared to bare Si have been obtained. The a-Si buffer layer dramatically managed crack formation and prevented capacity loss by adequate surface treatment. We introduced new coating method using modified RF sputter for making highly improved electrochemical performance by forming the doped layer in Si anode. It can open new synthetic route for mass production using slurry casting cell of stable Si anodes, which is simple and cost effective.

To develop high-capacity and low-cost lithium-sulfur (Li-S) batteries, the challenging problem related to the soluble lithium polysulfides must be probably addressed, since these soluble intermediates (Li₂S₈-Li₂S₆) can easily diffuse to the negative electrode, where they are reduced to Li₂S and deposit, resulting in low utilization of active materials and fast capacity fading. The strategies to solve this problem can be generally categorized as physical confinement in micropores or chemical immobilization through bonding of sulfur species in a carrier structure. Integrating physical and chemical immobilization together will be a better alternative approach without introducing extra inactive material. Herein, we report the design and characterization of a novel sulfur cathode with physical and chemical immobilization of sulfur species using microporous nitrogen-doped carbon fiber (MPNC). Cellulose aerogel was first activated to obtain mesoporous carbon nanofibers that are subsequently hydrothermally treated in aqueous ammonia for heavily nitrogen doping. The Li-S batteries based on MPNC-sulfur cathode showed high specific capacity (> 1000 mAh/g), excellent cycling and rate capability. The physical, chemical and electrochemical properties will be reported. It is concluded that the microporous architecture facilitates space confinement of sulfur species while the hetero-doped nitrogen species, with strong binding energy to polysufides, further assists their anchoring in the cathode.