Research Article www.acsami.org
Nanoscale Chemical Characterization of Solid-State Microbattery Stacks by Means of Auger Spectroscopy and Ion-Milling Cross Section Preparation A. Uhart,† J. B. Ledeuil,*,† B. Pecquenard,‡ F. Le Cras,§,∥ M. Proust,⊥ and H. Martinez*,†
ACS Appl. Mater. Interfaces 2017.9:33238-33249. Downloaded from pubs.acs.org by IOWA STATE UNIV on 01/25/19. For personal use only.
†
IPREM ECP − UMR CNRS 5254, Université de Pau et des Pays de l’Adour, Hélioparc Pau-Pyrénées, 2 Avenue du Président Angot, Pau CEDEX 9 64053, France ‡ CNRS, ICMCB−UPR 9048 and Bordeaux INP, Université de Bordeaux, 87 Avenue du Dr. Schweitzer, F-33600 Pessac, France § CEA LETI, 17 rue des Martyrs, F-38054 Grenoble, France ∥ Université Grenoble Alpes, F-38000 Grenoble, France ⊥ ST Microelectronics, 10 rue Thalès de Milet CS 97155, 37071 Tours, France S Supporting Information *
ABSTRACT: The current sustained demand for “smart” and connected devices has created a need for more miniaturized power sources, hence for microbatteries. Lithium-ion or “lithium-free” allsolid-state thin-film batteries are adapted solutions to this issue. The capability to carry out spatially resolved chemical analysis is fundamental for the understanding of the operation in an all-solidstate microbattery. Classically cumbersome and not straightforward techniques as TEM/STEM/EELS and FIB preparation methods could be used to address this issue. The challenge in this work is to make the characterization of Li-based material possible by coupling ion-milling cross section preparation method and AES techniques to characterize the behavior of a LiCoO2 positive electrode in an all solid state microbattery. The surface chemistry of LiCoO2 has been studied before and after LiPON deposition. Modifications of the chemical environments characteristic of the positive electrode have been reported at different steps of the electrochemical process. An original qualitative and a semiquantitative analysis has been used in this work with the peak deconvolution method based on real, certified reference spectra to better understand the lithiation/delithiation process. This original coupling has demonstrated that a full study of the pristine, cycled, and post mortem positive electrode in a microbattery is also possible. The ion-milling preparation method allows access to a large area, and the resolution of Auger analysis is highly resolved in energy to separate the lithium and the cobalt signals in an accurate way. KEYWORDS: all-solid-state thin-film lithium microbatteries, LiCoO2 positive electrode, Auger electron spectroscopy, cross-sectional characterization, ion-milling
1. INTRODUCTION The sustained demand for “smart” and connected devices in the context of the development of the Internet-of-things and increased healthcare monitoring has induced a need for more miniaturized power sources and hence a need for microbatteries. A complete integration of the power source in miniaturized devices comprising generally CMOS (complementary metal oxide semiconductor) components and MEMS (microelectromechanical systems) sensors or actuators, although not always compulsory, is an asset to optimize their compactness and security. All-solid-state thin-film lithium batteries are adapted solutions to this issue since they can be manufactured with small footprints, by techniques commonly used in the microelectronics industry, i.e., physical or chemical vapor deposition of thin films and patterning methods. All© 2017 American Chemical Society
solid-state thin-film microbatteries are composed of thin layers of different materials. The latter obviously comprise not only the positive and negative electrode materials, the solid electrolyte, and the metal current collectors but also ionic and/or electronic insulating barriers deposited on the substrate and encapsulation layers preventing water and air to reach the active part of the cell.1The cells are manufactured through a bottom-up process, by using vacuum deposition techniques to place each material layer on the top of the previous one, to form at the end a monolithic component. Finally, at least 10 Received: May 23, 2017 Accepted: August 30, 2017 Published: August 30, 2017 33238
DOI: 10.1021/acsami.7b07270 ACS Appl. Mater. Interfaces 2017, 9, 33238−33249
Research Article
ACS Applied Materials & Interfaces different thin films are synthesized, hence forming at least as many solid−solid interfaces. When designing new batteries, the selection of the active materials (electrodes and electrolyte) is generally directly driven by their intrinsic properties (capacity, electronic/ionic conductivity, operation voltage, and electrochemical stability window) that are generally already known or that can be more or less easily forecast and by their processing ability and manufacturing costs. Unfortunately, the actual performance of the resulting cells in terms of cycle life, calendar life, voltage profile, and impedance is often less than expected due to unforeseen electrode/electrolyte reactions and evolution of the electrode/electrolyte interfaces. In all-solid-state batteries, such interactions can lead to both structural and chemical evolutions of the materials in the vicinity of the contact surface and possibly to the formation of solid interphases. Reactions between the solid electrolyte and electrode materials can occur either during the manufacturing process2,3 or the operation of the cell.4,5 As for microbatteries, their sequential manufacturing process by sputtering is particularly prone to induce chemical and structural modifications of the surface of the underlying material during the deposition of the following one that may not follow thermodynamics rules (formation of metastable interphases).6 In these all-solid-state thin-film stacks, it is generally preferable to limit intermixing of the components and to limit the formation of electrode/electrolyte interphase to preserve the intrinsic properties of each material. Nevertheless, in some particular cases, the formation of a proper interphase can be beneficial, for example, to decrease the charge transfer resistance,7 but in any case, the generation of adequate solid−solid interfaces has to be monitored and checked by characterization means. Actually, the study of such interfaces in all-solid-state batteries was only reported in few papers, despite the large variety of material combinations that are envisaged in electrochemical cells, due to the difficulty to probe these buried interfacial regions. Techniques used to characterize them include imaging by scanning electron microscopy (SEM) and high-resolution transmission electron microscopy (HR-TEM), chemical/structural analyses by electron energy loss spectroscopy (EELS)4 or energy-dispersive spectroscopy (EDS),5 elastic recoil detection (ERD), Rutherford backscattering spectroscopy (RBS),8 X-ray absorption near-edge structure (XANES),9 X-ray photoelectron spectroscopy (XPS),10 neutron depth profiling,11 and electrochemical methods (electrochemical impedance spectroscopy (EIS),12,13 cyclic voltametry).14 The specific manufacturing process of planar thin-film batteries by vacuum deposition techniques, which allow the deposition of very thin films with controlled thickness and chemical composition, could offer a unique opportunity to study the formation of their solid−solid interfaces by XPS spectroscopy. Indeed, despite the fact that XPS probes only a thickness up to 5 nm, it is possible to carry out depth profile analyses of thicker interfacial zones between two materials deposited successively, by using bilayered samples which top layer has an increasing thickness. Thus, the interfacial zone can be probed from the surface of the substrate material to the core of the second material. Nevertheless, it is very likely that the interface observed this way during the early stages of its formation (
