Conduction below 100 °C in nominal Li6ZnNb4O14
- 675 Downloads
The increasing demand for a safe rechargeable battery with a high energy density per cell is driving a search for a novel solid electrolyte with a high Li+ or Na+ conductivity that is chemically stable in a working Li-ion or Na-ion battery. Li6ZnNb4O14 (LZNO) has been reported to exhibit a σ Li > 10−2 S cm−1 at 250 °C, but to disproportionate into multiple phases on cooling from 850 °C to room-temperature. An investigation of the room-temperature Li-ion conductivity in a porous pellet of a multiphase product of a nominal LZNO composition is shown to have bulk σ Li ≈ 3.3 × 10−5 S cm−1 at room-temperature that increases to 1.4 × 10−4 S cm−1 by 50 °C. 7Li MAS NMR spectra were fitted to two Lorentzian lines, one of which showed a dramatic increase with increasing temperature. A test for water stability indicates that Li+ may move to the particle and grain surfaces to react with adsorbed water as occurs in the garnet Li+ conductors.
KeywordsSolid Electrolyte LiNbO3 Scanning Transmission Electron Microscopy Solid Electrolyte Interphase Li3PO4
Since the introduction by the SONY Corp. in 1991 of a wireless cell phone powered by a rechargeable Li-ion battery (LIB), the LIB has been used extensively to power portable computers, numerous portable electronic devices from cell phones to digital cameras, and hand-held tools. In all of these applications, the LIB does not compete with the energy stored in a fossil fuel. With an increasing awareness of the societal costs of extraction of fossil fuels and of the gaseous emissions from their combustion, the storage of electrical energy generated by alternative energy sources such as wind or solar energy has become a global priority. The development of rechargeable batteries that can power electric vehicles competitively with the internal-combustion engine and can store and supply, competitive with coal or gas, power to the electric grid in a stationary battery has become a global priority. Li-ion rechargeable batteries are receiving world-wide attention because their cell-specific energy density can, theoretically, achieve over 400 Wh kg−1 . The Li–air and Li–sulfur batteries can, theoretically, reach specific energy densities per cell of 11400  and 2510 Wh kg−1 , respectively. However, LIBs face major hurdles of safety, cost, and cycle life. Conventional LIBs contain an organic–liquid electrolyte that is flammable and is reduced by an anode that maximizes the possible voltage of a cell. The voltage can be increased where a passivating solid electrolyte interphase (SEI) layer is formed on the anode surface by the addition of a suitable chemical to the liquid electrolyte; but the SEI layer introduces two fundamental problems: (1) it must be permeable to the Li+ ion, and the Li+ of the layer comes from the cathode on the initial charge of a cell fabricated in a discharged state to introduce an irreversible capacity loss of the cathode, and (2) the SEI layer adds to the resistance to charge transfer between the anode and the electrolyte. Too fast a charge may result in a plating of metallic lithium, Li0, on the SEI layer, and a Li0 anode forms dendrites on plating that grow during charge; if the dendrites grow across a thin electrolyte to the cathode, an internal short-circuit gives rise to thermal runaway with incendiary consequences . However, even where dendrites are blocked by a separator from reaching the cathode, they continuously create new anode surface area that needs to be pacified by new SEI formation, which leads to a capacity fade that limits the cycle life of the battery. Therefore, there is interest in the development of a solid Li+ electrolyte that can allow realization of a safe, low-cost LIB that approaches the theoretical limit of specific energy density . However, the catalysts for the reactions at an air cathode are more stable and active in an aqueous electrolyte, which is incompatible with a Li0 or other anode giving the high voltage needed for a Li–air or Li–S cell; and the soluble intermediates of the cathode reactions in a Li–sulfur cell need to be blocked from reaching the anode. Therefore, there is a strong motivation to find a solid Li+ electrolyte that is stable in water and can be interfaced with a Li0 anode .
Although the advantages of a solid electrolyte are widely acknowledged, their practical application is still limited by a low ionic conductivity, too small an energy gap, and/or poor chemical/electrochemical stability in a battery. Since the advent of the Na-sulfur battery in 1967 that operates with β-alumina as the solid electrolyte and molten electrodes at over 300 °C, many material systems have been investigated for good lithium-ion conductivity. This search has included crystalline, glassy, polymer gels, and polymer–gel/oxide composites. In the 1970s, Li3N was shown to have a room-temperature conductivity as high as 6 × 10−3 S cm−1 . However, this material has a small energy gap and, therefore, a poor electrochemical stability unless used only as an anode SEI layer. Many other Li+ conductors have since been discovered as is illustrated by the NASICON-structured LISICON Li1+x Al x Ti2−x (PO4)3 (LATP) , the perovskite-structured La3x Li(2/3)−x TiO3 (LLTO) , the thio-NASICON-structured LISICONS Li1.25Ge0.25P0.75S4  and Li10GeP2S12 . However, each of these materials has a fatal flaw that prevents their use as a solid electrolyte in a LIB. LATP and LLTO are unstable on contact with a Li0 anode and in an acidic aqueous electrolyte because the Ti4+/Ti3+ redox reaction is, respectively, at 1.8 and 2.5 V versus Li0 . The NASCICON-structured LISICON and thio-LISICON compounds are chemically unstable on contact with an aqueous and an organic–liquid Li+ electrolyte and they do not provide an all-solid-state battery of adequate cathode capacity. Lithium-phosphorous-oxinitride (LIPON) films can be prepared in situ by a sputtering technique with a Li3PO4 target in a controlled N2 atmosphere ; it is stable on contact with Li0, but its room-temperature Li+ conductivity is only 10−6 S cm−1. The garnet-structured Li7La3Zr2−x Ta x O12 (LLZO) can have a bulk Li+ conductivity as high as 10−3 S cm−1 and it is stable against Li0 [14, 15, 16]; but on exposure to air, Li+ moves to the surface of particles and grains to react with adsorbed water and CO2. The water sensitivity limits their application.
Previously, Kanovalova et al.  reported a Li6ZnNb4O14 (LZNO) phase with an interesting Li+ conductivity at 250 °C. However, this temperature is too high for a room-temperature LIB with solid electrodes and the compound is metastable. In this paper, we reinvestigate the Li+ conductivity of nominal LZNO to lower temperatures; the bulk conductivity is 3.28 × 10−5 S cm−1 at room-temperature and increases to 1.43 × 10−4 S cm−1 at 50 °C, typical of the good crystalline oxide Li+ conductors.
The composition of LZNO was prepared by a conventional solid-state reaction. Stoichiometric amounts of dried Li2CO3 (Alfa Aesar, 99 %), Nb2O5 (Alfa Aesar, 99.9 %), and ZnO (Alfa Aesar, 99.99 %) were mixed thoroughly in an Al2O3 mortar. 10 % excess Li and Zn were used to compensate for the loss of lithium and zinc during the high-temperature firing process. The mixture of starting materials was first fired at 600 °C for 3 h, and then pelletized and fired at 1080 °C for 20 h with a ramping rate of 5 °C/min, followed by grinding with a mortar and pestle, then pelletized and fired at 1080 °C for another 20 h. Phase purity and stability were characterized with a PANalytical Empyrean diffractometer equipped with Cu Kα radiation (λ = 1.5406 Å). The scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDX) measurements of the sintered samples were performed with a Hitachi S4800 SEM operated at 20 kV. The scanning transmission electron microscopy (STEM) was performed with a JEOL 2200FS TEM/STEM operated at 200 kV to examine the particle morphology of the ground powders.
For the impedance measurement, both parallel surfaces of the pellet were sputtered with a layer of gold contacts. The experiment was performed with a Novocontrol Alpha-A impedance analyzer in the frequency range of 0.1–107 Hz. Temperature was ramped from room-temperature (21.5) to 250 °C in increments of 50 °C; about 30 min was allowed for temperature equilibration after each temperature ramp. The temperature of the sample was controlled by a Novocontrol Quatro Cryosystem with flowing nitrogen gas.
In order to estimate roughly the scale of the Li+/H+ exchange rate, 1 g of LZNO powder was dispersed in 15 mL of de-ionized water at ambient conditions. The pH values were monitored by a VWR symphony pH meter (Model SB70P). The SEM images for the pellets after 50 h in water were taken with a JEOL JSM-6060 SEM microscope operating at 10 kV.
7Li MAS NMR measurements were carried out on ground powders with a 9.4T Bruker Avance NMR spectrometer. The spectrum was examined over the temperature range of –43 to 87 °C in a 3.2 mm MAS rotor spinning at 15 kHz. 7Li spectra were acquired with an echo sequence (90°–180°) having a 90° pulse length of 2 µs. 7Li chemical shifts are given with respect to 1 M LiCl (δ = 0 ppm) with solid LiCl as the secondary standard (−1.06 ppm).
Results and discussion
A platinum disk and lithium films of 1 µm thickness were sputtered on the as-synthesized LZNO pellet surfaces to make Li/LZNO/Pt and Li/LZNO/Li cells. The interfacial resistances were too large to allow testing of their chemical and electrochemical stability. However, excellent stability results were obtained with a study of 90 % β-Li3PS4 (LPS) and 10 % LZNO composite by Hood et al., which would suggest a possible good stability of LZNO. Cyclic voltammetry measurements show that the composite of LPS and LZNO is stable up to a potential of 5 V versus Li/Li+ and the symmetric Li/composite electrolyte/Li cell shows a long-term compatibility of the composite with metallic lithium .
A room-temperature fast Li-ion conducting solid electrolyte with a starting composition of LZNO was prepared by a solid-state reaction method. The bulk conductivity was measured to be 3.28 × 10−5 S cm−1 at room-temperature and increased to 1.43 × 10−4 S cm−1 at 50 °C. The nominal Li3ZnNb4O14 composition disproportionates by room-temperature into multiple crystalline phases on cooling from 1080 °C, but the Li+ conductivity remains comparable to that of other good crystalline oxide Li+ conductors.
The research was sponsored by the U.S. Department of Energy, Office Science, Office of Basic Energy Sciences, Materials Sciences and Engineering Division. The NMR research (L.W.G. and E.W.H.) was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Chemical Sciences, Geosciences and Biosciences Division. Scanning electron microscopy research was supported through a user project supported by ORNL’s Center for Nanophase Materials Sciences (CNMS), which is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy. Dr. John B. Goodenough was supported by the Materials Sciences and Engineering Division, Office of Basic Energy Sciences, Office of Science, U.S. Department of Energy grant number (DE-SC0005397).
This manuscript has been authored by a contractor of the U.S. Government under contract DE-AC05-00OR22725. Accordingly, the U.S. Government retains a paid-up, nonexclusive, irrevocable, worldwide license to publish or reproduce the published form of this contribution, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, or allow others to do so, for U.S. Government purposes.
- 28.Hood ZD, Wang H, Li Y, Paranthaman MP, Liang C (2015) The “filler effect”: a study of solid oxide fillers with β-Li3PS4 for lithium conducting electrolytes. Solid State Ionics (under review)Google Scholar