The all-solid-state-lithium battery (SSLB) is a key technology to use Li metal anode with a theoretical capacity of 3860 mAhg −1 while suppressing the growth of Li dendrites. We present the model of Li nucleation on a solid-state electrolyte with metal current collector (CC)/lithium phosphorous oxynitride (LiPON) interfaces as the nucleation sites. We also observe the initial stage of Li growth and following Li dissolution using an in-situ scanning electron microscope (SEM) technique. The Li nucleation overpotential increases with increasing the Young's modulus of the CC. Also, the achievable Li particle sizes drastically increase with the Young's modulus of the CC. Our calculations show agreements with the experimental results and reveal that tensile stresses in a CC generate 10 −1 −10 0 GPa pressures on Li nuclei. Those pressures are three orders of magnitude larger than the ultimate tensile strength of bulk Li.The ten times larger theoretical capacity of Li metal anode (3860 mAhg −1 ) compared to graphite anode (372 mAhg −1 ) has attracted a great deal of interest. However, despite significant prior efforts, Li dendrite growth and a resulting rapid degradation in Coulombic efficiency (CE) have been unsolved critical problems. 1 On the other hand, solid-state electrolytes can be a solution to mechanically suppress Li dendrite growth toward the cathode without the help of separators. Hence, the all-solid-state-lithium battery (SSLB) provides innovative changes for not only increasing the overall battery energy density, but also developing "rechargeable" Li metal anodes.The Oak Ridge National Laboratory (ORNL) pioneered thin-film SSLBs using lithium phosphorus oxynitride (LiPON) electrolyte. 2,3 The first charging process nucleates Li metal on the anode side. ORNL demonstrated that the solid-state electrolyte completely blocked Li growth toward the cathode. Instead, Li grew to penetrate thick Cu current collectors (CC). Although thin-film SSLBs can operate over thousands of charge/discharge cycles under specific conditions as the ORNL has demonstrated, it is still obscure what mechanism dominates Li electrodeposition and dissolution on solid-state electrolytes.Nucleation sites are located at solid/solid interfaces in a SSLB. 4,5 Hence, the Li nuclei must deform either electrode or electrolyte to create their own spaces and keep growing. This process must be associated with the generation of strain energies at nucleation sites. The nucleation work on a heterogeneous substrate in a liquid electrolyte is usually divided into the surface work and the chemical work. 6 We also consider the mechanical work for Li nucleation and growth in a solid electrolyte system. We study the initial stage of Li electrodeposition using an in-situ scanning electron microscope (SEM) technique. Finally, we verify our model with CCs of different metals. ExperimentalThe top and bottom surfaces of a Li 1+x+y Al x Ti 2−x Si y P 3−y O 12 (LATP) sheet (1.25 cm × 1.25 cm, Ohara Co.) were coated with 2.5-μm-thick LiPON layers by radio frequency ...

This paper studies the effect of current density on electrochemical Li deposition/dissolution at glassy solid electrolyte (LiPON) interfaces with a thin-film Cu current collector by in-situ scanning electron microscopy (SEM). The Li nucleation rate and the saturation density of Li nuclei increase with increasing current density. When the current density is smaller than 300 µA cm −2 , Li islands continue to separately grow under a Cu film to the critical sizes to produce small cracks in the Cu film resulting in isolated Li rod growth from the cracks. On the other hand, a current density of 1.0 mA cm −2 provokes the nucleation of Li islands at a number of sites. They rapidly coalesce under a Cu film in all lateral directions before cracking the Cu film, whereby Li growth is prevented.

This paper reports the results of in-situ scanning electron microscope (SEM) observations of Li plating and stripping with Pt current collector (CC) films on lithium phosphorus oxynitride (LiPON) glass electrolyte. Before the Li nucleation, Li-Pt alloying reactions occur while the voltage shows positive values. Li nucleates and starts to grow after the voltage decreases, which shows the lowest value in the negative voltage region. After Li-supersaturated-Pt surfaces are locally deformed into dome shapes where Li particles grow underneath, a hole opens in the center of each dome. This is a different CC fracture mechanism from that of non-alloying-CC films such as Cu. This difference in fracture mechanism is derived from a supersaturation process of Pt with Li, which plays a critical role in achieving smaller nucleation overpotentials and larger nucleation number densities compared to the case with Cu.

We have studied electrochemical Li deposition/dissolution processes at amorphous solid electrolyte (LiPON) interfaces with 30-nm-thick-Cu-current collectors at different current densities by in-situ scanning electron microscopy (SEM). When the current density is smaller than 300 μA cm−2, Li islands continue to grow under a Cu film without coalescing with their neighbors. Consequently, they produce small cracks in the Cu film leading to isolated Li rod growth from the cracks. On the other hand, a current density of 1.0 mA cm−2 provokes the nucleation of Li islands with a higher number density. They rapidly coalesce under a Cu film in all lateral directions before cracking the Cu film. High current density conditions therefore suppress Li rod growths.

This study examines an electro-chemo-mechanical theory that the nucleation overpotential (ηnc) for the Li nucleation at the Cu current collector (CC) film/lithium phosphorus oxynitride (LiPON) electrolyte interface is influenced by mechanical work to deform the CC film. The finite element method was used to simulate the mechanical pressure that the CC film exerted on the Li nuclei at the Cu/LiPON interface, and the results were in agreement with those in our previous study. By in-situ scanning electron microscopy observation for cycling of Li plating/stripping, it was found that Li was repeatedly nucleated and grew at positions where the CC film was locally fractured, and the ηnc decreased with repetition of Li plating/stripping. This was because the mechanical pressure to the Li nuclei was no longer applied at locations where the CC film was fractured. On the other hand, when thicker CC films that did not induce any cracks were used, the ηnc exhibited nearly consistent values even after repeating Li plating/stripping. Consequently, the experimental results obtained in this study supported our nucleation theory for a metal/solid-state-electrolyte interfacial system.