Solid-state batteries (SSBs) are a major focus for next-generation energy storage that is safe and has high energy density. In recent years, ion conductivity of solid-state electrolytes (SSEs) has developed to a level comparable to liquid electrolytes. However, great challenges are encountered at SSE/electrode interfaces when integrating the SSE into SSB cells. In addition to stability of the interface, which has been discussed extensively in the literature, a space-charge region around the electrode/SSE interface should exist and it can significantly affect Li-ion transport cross the interface [1]. Fermi-level alignment requires an ion charge transfer across the interface between the ion conductors that build an electric field around the interface. This space-charge region has been proposed theoretically,1 but direct measurement of the electric field or electrical potential has not been reported. On the other hand, the electric field in the semiconductor space-charge region has been studied extensively and directly measured and imaged; furthermore, the charge depletion and extension of the electric field into the semiconductor bulk have all been well documented. In fact, operations of many semiconductor micro-devices rely on the non-Ohmic current-voltage (I-V) characteristic across the space-charge region. And this makes the fundamental study and engineering of the semiconductor space charge among the farthestIn this contribution, we report on direct quantitative imaging of electrical potential on the space-charge region around the interface between Li6PS5Cl (LPSC) SSE and a NiMnCoO (NMC) cathode (Fig. 1a). We achieved this nm-scale, two-dimensional measurement by Kelvin probe force microscopy[2] (KPFM) imaging on the cross-section of a LPSC/NMC half-cell. KPFM is based on the non-contact mode of an atomic force microscope (AFM) setup in an Ar glove box. It maps the electrical potential of a sample surface in ~30-nm spatial resolution and ~10-mV voltage sensitivity. In our in-house KPFM setup, we used the second resonant oscillation of the cantilever to enhance the voltage sensitivity. To image the space charges, a well-defined SSE/electrode interface is needed. However, the sideview of the as-made LPSC/NMC half-cell by cold-pressing does not show a well-defined interface, and the surface of the sideview is rough for KPFM imaging. We polished the cross-section by ion-milling and via air-free sample transfer and polishing, and we obtained a well-defined interface exposed on the flat cross-sectional surface.The KPFM results show that the potential increased from the SSE side to the NMC side across the interface (Fig. 1b); this increase is due to the Li cation transferred from the SSE to NMC sides as well as the anion leveraged on the SSE side. This charge transfer is driven by the electrochemical potential difference of the Li ion in the two materials—in contrast to semiconductors, where the electronic charge transfer is driven by both the electronic band misalignment and charge-carrier concentrations at both sides of the interface. The potential is nonuniformly distributed, with 300‒500-mV variations in amplitude and 1‒2-μm variations in distance (Fig. 1c), indicating the nonuniformly distributed Li cations and anions. The nonuniform charge distribution can be caused by local chemical and structural defects and disorders such as grain boundaries, extended plane and line defects, and more. A simple estimate of the charge density is D=εε0V/d=2.3x10-9 C/cm2=1.5x108/cm2, with the assumptions that the potential is increased simply by an electrical double layer, the potential V = 400 mV, the dielectric constant ε=10, and the inter-distance of the double layer d=1.5 μm . This charge density corresponds to 3.3 charges per million (106) unit cells with a unit cell size of 1.5×1.5 nm2.This electrical potential of 300‒500 mV can significantly impede Li-ion transport from the SSE to the cathode during charging, and it should be mitigated. Dividing the potential difference between the two end-materials by engineering the interface (such as inserting a gradient layer) can be a promising approach, and it can also prevent interfacial chemical degradations. Interfacial engineering can effectively facilitate charge transport across the back contact of photovoltaic devices [3]. Further potential imaging on the engineered interfaces and degraded interfaces by cycling could lead to better understanding the interfacial electrical behavior. Our direct nm-scale potential imaging opens up a novel characterization technique for developing next-generation solid-state batteries. References Haruyama, K. Sodeyama, L. Han, K. Takeda, and Y. Tateyama, Chem. Mater. 26, 4248 (2014).C-S. Jiang et al., Nature Communications 6, 8397 (2015).Song, A. Moore, and J.R. Sites, IEEE J. Photovoltaics 8 , 293 (2018). Figure 1
Read full abstract