Search

All-solid-state silicon/LiPON/lithium model batteries are assembled via magnetron sputtering followed by a lithium evaporation process. The cells were characterized via cyclic voltammetry, linear sweep voltammetry and impedance spectroscopy. Cyclic voltammetry confirms a long-term stability and high coulombic efficiency of the solid cells. Also, the assembled batteries were tested at extraordinary high scan rates, showing remarkably high output currents and an exceptional fast-cycling capability. Further rate-dependent investigations via linear sweep voltammetry indicate clearly two different ion-transport mechanisms in the silicon electrode at low and high scan rates. While a diffusion-controlled mechanism is suggested at high scan rates, a homogenous concentration profile is expected at extremely low rates. As a decisive advantage of solid-state cells, electrochemical cycling can be performed at elevated temperatures. This is used to determine the temperature dependence of the diffusion coefficients of lithium in the silicon electrode from room temperature to 165 °C. The effective activation energy of the lithium-ion migration in silicon is obtained to 0.1 eV, which is remarkably lower than that of inside LiPON, i.e. 0.55 eV. Complementary impedance data reveal that the kinetics of transport in the solid-state setup at very high output currents and low temperatures is controlled by lithium-ion migration in the LiPON film.

The solid-state electrolyte LiPON is synthesized using reactive magnetron sputtering. The correlation between plasma conditions and chemical composition of the layers is studied by X-ray photo electron spectroscopy, accompanied by electrochemical measurements of the ionic conductivity of the films. It is found that the plasma conditions in the vicinity of the substrate affect the lithium and nitrogen contents in a reciprocal manner. The target sputter-age and an additional bias potential to the substrate are identified as important parameters to optimize the performance of the layers. Based on this information, bilayer samples are prepared, consisting of LiPON and a cathode of LiFePO4. Chronopotentiometry at these bilayers reveals the importance of a low interface resistance, and moreover, it identifies an insufficient lithium content at the LiPON/LiFePO4 interface as a reason for high interface resistance. Nitrogen is also essential to increase the conductivity and hence, a lack of nitrogen at the interface is suggested to have the same consequence. It is shown how the interface resistance can be controlled by the preparation conditions to achieve an optimized performance.

Thin film battery electrodes of the olivine structure LiFePO4 and the spinel phase LiMn2O4 are deposited through ion-beam sputtering. The intercalation kinetics is studied by cyclo-voltammetry using variation of the cycling rate over 4 to 5 orders of magnitude. The well-defined layer geometry allows a detailed quantitative analysis. It is shown that LiFePO4 clearly undergoes phase separation during intercalation, although the material is nano-confined and very high charging rates are applied. We present a modified Randles–Sevcik evaluation adapted to phase-separating systems. Both the charging current and the overpotential depend on the film thickness in a systematic way. The analysis yields evidence that the grain boundaries are important short circuit paths for fast transport. They increase the electrochemical active area with increasing layer thickness. Evidence is obtained that the grain boundaries in LiFePO4 have the character of an ion-conductor of vanishing electronic conductivity.

We investigate the electrochemical side reactions that occur during the cyclic lithiation/delithiation of sputter-deposited Li x V 2 O 5 films. For this purpose, the mass change of Li x V 2 O 5 films during lithiation/delithiation is measured by quartz crystal microgravimetry, while the electrical charge that is flowing during this reaction is measured by cyclic voltammetry. A time-resolved evaluation of the measurement data, in combination with an advanced type of data processing, finally allows us to calculate time-resolved quantitative mass spectra. These spectra provide information about the chemical species that take part during the electrochemical reactions. Based on this technique, we study the electrochemical side reactions between the Li x V 2 O 5 and the corresponding liquid electrolyte, i.e. we investigate the time-resolved formation of the solid electrolyte interface (SEI) layer during long term cycling. We are able to identify several chemical species that are formed during cycling and moreover, we identify three different stages of SEI formation.

A brief report is given on recent studies of the atomistic processes during charging of battery cathode material LixCoO2 by means of magnetometry and positron annihilation. A set-up for operando magnetometry is implemented which, for the first time, allowed to continuously monitor the distinct variation of the magnetic susceptibility χ of LixCoO2 which occurs during consecutive charging and discharging cycles. The variation of χ with Li+ content in the concentration range 1>x≥0.77 arises from a variation of the electronic density of states and from electronic correlation effects. The χ (x)-behaviour for xx)-variations in this concentration regime. First measurements on LixCoO2 thin-films performed at the positron beam line NEPOMUC of FRM II at the Heinz Maier-Leibnitz neutron source will be presented.

Thin film bilayers of silicon and LiPON are synthesized via rf magnetron sputtering, whereas the thickness of both layers is in the nanometer range, i.e. between 50 and 400 nm. Electron microscopy on the samples confirms a well-defined layer structure and a high interface quality of the layer stacks. Cyclic lithiation/delithiation of the bilayers via chronopotentiometry, moreover, reveals an astonishing capacity retention and high coulombic efficiency when compared to pure silicon films. This strong improvement in electrochemical performance is attributed to a stabilization of the silicon electrode by the nano-sized LiPON coating. Moreover, it is found that the rate of improvement depends on the thickness ratio between LiPON and silicon. Hence, samples with different thickness of LiPON and silicon, respectively, are investigated to get detailed insight into the electrochemical long term stability of the bilayer samples. Based on these investigations, an electro-mechanical model is suggested that describes in detail the microscopic degradation mechanism of LiPON coated silicon electrodes.

We report on the preparation of Li+ conducting lithium phosphorus oxynitride (LiPON) layers by ion beam sputtering, resulting in ultra-thin layers with a thickness down to only 12 nm. The morphology of the layers is studied by transmission electron microscopy, whereas homogeneity and excellent interface quality is observed. The ion conducting properties of the layers are studied as a function of temperature and layer thickness, using electrochemical impedance spectroscopy. In these measurements, a specific ionic conductivity at room temperature of around 2 · 10−7 S cm−1 is found that results in a maximum conductance of 0.2 S cm−2, thanks to their ultra-small thickness. To demonstrate the applicability of the thin films as functional solid state electrolyte, Ag/LiPON/Pt triple layers are studied, serving as models for all solid state batteries. The functionality of these model batteries is measured by cyclic voltammetry of up to 100 charge–discharge cycles, and their morphology after cycling is investigated by transmission electron microscopy. For the first time, these investigations give insight into the morphology of this kind of model cell after intensive cycling. In particular, they allow determining the specific lithium storage capacity of the interface reaction layer between Ag and LiPON, which is found to be around 12 μAh cm−2 μm−1.

The time-dependent lithium concentration of sputter-deposited lithium cobalt oxide (LCO) layers during delithiation is investigated. For this purpose, functional LCO thin films of well-defined thickness are prepared by ion-beam sputter deposition and delithiated electrochemically under potentiostatic conditions. The lithium concentration of the films during delithiation is determined optically, i.e. by measuring the relative transmission of the films. This electrochromic absorption technique is non-sensitive to electrochemical side reactions, and gives information about the ionic concentration of the LCO films. Therefore, it is used to study the lithium kinetics in the films as a function of their thickness. To evaluate the measurement data, different physical models are discussed. It is found that the experimental data can be accurately described by a thin-film diffusion model, taking into account a limited interface transport. Fitting this model to the measured concentration characteristics gives a lithium diffusion coefficient of D ¯ = 1.04 × 10 - 13 cm 2 s - 1 and an interface transport coefficient of κ ¯ = 1.59 × 10 - 7 cm s - 1 .

Thin films of lithium iron phosphate (LiFePO4, LFP), with a thickness between 80 nm and 320 nm are prepared by ion beam sputter deposition. X-ray diffraction, transmission electron microscopy, and electron energy loss spectroscopy show that the films exhibit the desired structure, morphology, and chemical composition. Chrono potentiometry and cyclic voltammetry are carried out to determine the electrochemical behavior of the LFP films. Within these electrochemical measurements, a well-defined lithium intercalation/deintercalation reaction at around 3.45 V vs. Li/Li+ and a specific capacity of 104 mAh g−1 are observed. Moreover, a quite high cycling stability is found: Measurement data at 1C rate indicate a capacity fading of only 20% after ≈1700 charge–discharge cycles. In addition, the diffusion of lithium within the LFP thin films is studied by cyclic voltammetry and galvanostatic intermittent transition technique. Within these measurements it is observed that the lithium kinetics in the sputter-deposited thin films is about two orders of magnitude faster, compared to the powder material and thin films deposited by pulsed laser deposition. This behavior may be explained by the strong texture of the sputtered films.

Li4Ti5O12 (LTO) thin films are deposited by dc-ion beam sputtering at different oxygen partial pressures and different substrate temperatures. In order to investigate, how these two parameters influence the atomic structure, the specimens are characterized by X-ray diffraction and transmission electron microscopy. Electrochemical characterization of the films is done by cyclic voltammetry and chrono-potentiometry. To determine an averaged chemical diffusion coefficient of lithium, a method is developed, evaluating c-rate tests. The results obtained by this method are compared to results obtained by the well established galvanostatic intermittent titration technique (GITT), which is used to determine a concentration dependent diffusion coefficient of lithium in LTO.

LiCoO 2 films, 200 nm in thickness, are produced by ion beam sputter-deposition. The structural, electrical and electrochemical properties of the films are studied by means of X-ray diffraction ( XRD ), electrical direct-current-measurements, and electron energy loss spectroscopy (EELS). The influence of preparation parameters like substrate temperature or oxygen partial pressure, are explored. While an oxygen deficiency was observed in films sputtered under pure argon atmosphere, the oxygen content of the thin films increases, significantly, in case of films sputtered under addition of O 2 . At substrate temperatures below 300 °C, XRD measurements reveal a lattice structure similar to the low-temperature-phase, while the formation of the high-temperature-phase is clearly observed at temperatures above 500 °C and an Argon/Oxygen ratio of 3/2. Furthermore, the EELS technique is demonstrated to be a sensitive tool to characterize the lithiation state of the sputter-deposited thin films.

Thin film ion-conductive materials may represent essential components of future all-solid-state batteries. Besides, they are also of interest from a fundamental point of view. In this article, the deposition of thin films of complex oxides and amorphous glasses is discussed. Methods of local chemical analysis by analytical electron microscopy and atom probe tomography are described and studies of atomic transport in sputtered network glasses are presented. As experimental examples, thin films of Li borate and silicate glasses, LiCoO2, V2O5, and Li4Ti5O12are addressed.

Complex impedance spectra of ion-conducting lithium borate network glasses are used to study the deformed shape of impedance semicircles which are usually described by a parallel circuit of an ohmic resistor and a phenomenological Constant Phase Element (CPE). Based on the Concept of Mismatch and Relaxation (CMR) of Funke et al. which provides a theoretical treatment of the ion dynamics in disordered materials, a quantitative description of the experimental impedance spectra is presented. It takes into account the contribution of the static glass network by introducing an additional capacitor, and provides both, a physical interpretation of the CPE and an easy-to-handle mathematical formula to calculate the ’real’ capacity of a CPE.

Thin films of lithium borate, sodium borate, rubidium borate, and lithium silicate glasses are prepared by ion beam sputter deposition in a thickness range between 70 and 1,400 nm. The chemical, structural, and electrical properties of the films are determined by transmission electron microscopy and impedance spectroscopy. A comparison between the thin glass layers and the corresponding bulk glasses, prepared from the melt, shows that both have similar chemical compositions and atomic short range orders. In contrast, the ionic conductivities of the borate glass films are found to be significantly increased compared to the corresponding bulk materials, while the increase depends on the chemical composition of the glasses and is not observed in the case of the silicate system. We propose a modified network structure of the sputter-deposited glass layers, namely, an increased nonbridging oxygen concentration, to be responsible for the elevated ionic conductivity.

Two distinct regions of the dc conductivity and its temperature dependence as a function of the mol fraction of alkali oxides, X, are observed in Na- and Li-borate glasses. At low alkali contents the dc conductivity σdc increases only moderately with X. However, at higher alkali contents, log σdc increases linearly and the activation enthalpy ΔH of σdc × T decreases linearly with log X, i.e. the dc conductivity reveals an effective power-law behavior. The transition between low alkali and high alkali behavior takes place at X ≈ 0.08 for Na-borate and at X ≈ 0.09 for Li-borate glasses. This behavior suggests that the diffusion mechanism changes at these alkali contents. The results are discussed in terms of ion separations and the transition from a single-ion jump to a collective diffusion mechanism. The vanishing of the mixed-alkali peak in Na–Rb borate and alumino-germanate glasses at sodium contents similar to that observed for the change in slope of σdc(X) in this work suggests that both phenomena share the same origin.

The mixed-alkali effect (MAE) of the ionic conductivity is traditionally defined as a negative deviation of the dc conductivity from that of an ‘ideal glass’ obtained according a linear mixing rule. An alternative definition is proposed in this paper which relates the conductivity of a mixed-alkali glass to the conductivity-sum of two single-alkali glasses with total alkali concentrations equal to the corresponding partial alkali concentrations of the mixed-alkali glass. A new reference glass denoted as independent component glass (ICG) is introduced. In contrast to the ‘ideal glass’ the conductivity of a real mixed-alkali glass is larger than the conductivity of the respective ICG resulting in a positive MAE. This fact manifests an enhancement of the conductivity when two single-alkali glasses are combined to a mixed-alkali glass. The conductivities and the tracer diffusivities of the mixed-alkali- and the corresponding single-alkali glasses are compared from this point of view. A subnetwork diffusion concept (SDC) is proposed to describe the diffusion dynamics in Na–Rb-borate mixed-alkali glasses.

The ionic conductivity of Na- and Na–Rb borate glasses is investigated as a function of hydrostatic pressure. Activation volumes of the dc conductivity are obtained from pressure-dependent ac conductivity measurements at a constant temperature of 180 °C. The activation volume of Na borate glasses decreases with increasing alkali content. It depends linearly on a separation parameter which is defined as the ratio of the average ionic distance and the average network atom distance. The activation volume of Na–Rb borate glasses increases from both end-members towards the mixed-alkali regime and passes through a maximum near the relative content Na/(Na + Rb) = 0.4. In addition to mobility anomalies, observed earlier in these borate glasses, the maximum in the activation volume is a further important fingerprint of the mixed-alkali effect.

Mass densities, molar volumes, glass-transition temperatures, and ionic conductivities are measured in series of YNa2O · (1 − Y)B2O3 glasses, with Y = 0.00, 0.04, 0.08, 0.12, 0.16, 0.20, 0.25, 0.30 and YRb2O · (1 − Y)B2O3 glasses, with Y = 0.00, 0.12, 0.16, 0.20, 0.25, 0.30. Measurements of the molar volumes indicate that the incorporation of rubidium ions leads to a considerable expansion of the network, which is not observed for sodium ions. The glass-transition temperature increases with increasing alkali content and reaches a maximum near Y = 0.25 for both glass systems. These trends are attributed to changes in the glass network. For each glass composition an Arrhenius-activated increase of the product of dc conductivity and temperature is observed. The activation enthalpy decreases with increasing number density of ions. A comparison between the binary sodium- and rubidium-borate glasses from this work, with the ternary sodium–rubidium borate glasses studied earlier in our laboratory, provides interesting insights in the influence of the glass structure on ionic transport processes and the mixed-alkali effect.

The mixed-alkali effect is studied in Y [X Na2O · (1−X) Rb2O] · (1−Y) B2O3 glasses as a function of the total alkali content Y. To this aim the ionic transport in glasses with Y = 0.3 was investigated by impedance spectroscopy and by the radiotracer diffusion technique. The results are compared with data obtained earlier by our group on a glass system with Y = 0.2. In both systems the relative compositions X = Na/(Na + Rb) are varied from 0.0 to 1.0 in steps of 0.2. The dc electrical conductivity is higher in Y = 0.3 glasses than in Y = 0.2 glasses and exhibits a minimum in each system near X = 0.4. Also near X = 0.4 a pertaining maximum in activation enthalpy occurs being more pronounced for Y = 0.3 as compared to Y = 0.2. The glass-transition temperature T g is almost identical for glasses with Y = 0.2 and 0.3 with a minimum near X = 0.2. The diffusivities of 22Na and 86Rb measured at a constant temperature (380 °C) show in both glass systems (Y = 0.3 and Y = 0.2) a similar dependence on the composition parameter X. The 86Rb diffusivity decreases exponentially with decreasing Rb content. The 22Na diffusivity decreases with decreasing Na content but reaches an almost constant plateau on the Rb-rich side. For each relative composition X the diffusivities of the two isotopes increase by nearly the same constant factor, when Y is increased from 0.2 to 0.3. In both systems 22Na- and 86Rb-diffusivities as functions of the composition parameter X intersect near X = 0.2 explaining the minimum in dc-conductivity. Composition dependent common Haven ratios were obtained by comparing the conductivity with the diffusion data.

Carbon-coated Li3V2(PO4)3 (LVP) displaying nanostructured morphology can be easily prepared by using ionic-liquid-assisted sol-gel synthesis. The selection of highly viscous and thermally stable ionic liquids might promote the formation of nanostructures during the sol-gel synthesis. The presence of these structures shortens the diffusion paths and enlarges the contact area between the active material and the electrolyte; this leads to a significant improvement in lithium-ion diffusion. At the same time, the use of ionic liquids has a positive influence on the coating of the LVP particles, which improves the electronic conductivity of this material; this leads to enhanced charge-transfer properties. At a high current density of 40 C, the LVP/N-butyl-N-methylpyrrolidinium bis(trifluoromethanesulfonyl)imide material delivered a reversible capacity of approximately 100 mA h g(-1), and approximately 99 % of the initial capacity value was retained even after 100 cycles at 50 C. The excellent high rate and cycling stability performance make Li3V2(PO4)3 prepared by ionic-liquid-assisted sol-gel synthesis a very promising cathode material for high-power electrochemical storage devices.

This study examines the influence of lithium intercalation on the optical properties of vanadium pentoxide films. The films with a thickness between 400 and 1000 nm were prepared by DC magnetron sputter deposition. Cyclic voltammetry and chronopotentiometry were used to set a well defined lithiation state of the LixV2O5 films between x = 0 and x = 1. The optical properties of these films were measured by optical reflectometry in the wavelength range between 500 and 1700 nm. From the reflectance data, the refractive index and the extinction coefficient of the films were finally calculated as a function of the wavelength using Cauchy's dispersion model. The results confirm that the optical behavior of LixV2O5 films varies significantly upon lithium insertion. It is demonstrated that the changes produced in the optical properties are completely reversible within the limits of permanent structure changes.

Amorphous lithium borate films are prepared by ion-beam sputtering using a $0.2\phantom{\rule{0.3em}{0ex}}{\mathrm{Li}}_{2}\mathrm{O}∙0.8\phantom{\rule{0.3em}{0ex}}{\mathrm{B}}_{2}{\mathrm{O}}_{3}$ glass as a target material. The films of a thickness between 7 and $700\phantom{\rule{0.3em}{0ex}}\mathrm{nm}$ are deposited onto silicon substrates between two metallic AlLi electrodes serving as electrical contacts. dc conductivities of the glass layers of different thicknesses are determined by temperature-dependent impedance measurements. An increase of the specific dc conductivity of about 3 orders of magnitude is observed, when the thickness of the layers decreases from 120 down to $7\phantom{\rule{0.3em}{0ex}}\mathrm{nm}$. This conductivity increase in the case of very thin films leads to an absolute conductance per unit area of $0.32\phantom{\rule{0.3em}{0ex}}{\ensuremath{\Omega}}^{\ensuremath{-}1}\phantom{\rule{0.3em}{0ex}}{\mathrm{cm}}^{\ensuremath{-}2}$ at $25\phantom{\rule{0.2em}{0ex}}\ifmmode^\circ\else\textdegree\fi{}\mathrm{C}$, which is sufficient for technical application of the glass layers in solid-state ionic devices. Three alternative explanations are suggested to explain this nontrivial conductivity increase: structural modifications at the interfaces, the formation of space-charge regions, and the existence of randomly distributed ion-conducting channels.

The electrochemical lithiation–delithiation of iron oxide is a rather complex process, which is still not fully understood. In this study we investigated the electrochemical lithiation–delithiation mechanism of hematite by means of X-ray diffraction (XRD), 57Fe Mossbauer spectroscopy, high-resolution transmission electron microscopy (HRTEM) and X-ray absorption spectroscopy (XAS). Since the delithiation process has been so far less investigated, particular attention was dedicated to the characterization of the chemical species that are formed during this process. The results of this investigation indicated that at the end of the delithiation process lithium iron oxide α-LiFeO2 is formed. The formation of this compound may be the explanation for the irreversible capacity loss in the first cycle, which is usually assigned to the formation of an organic gel-like layer. Based on these results a new charge–discharge mechanism of hematite in lithium-ion batteries (LIBs) is proposed and discussed.

Lithium borate network glasses are possible candidates for separator membranes in all-solid-state batteries. Thin films of a Li2O-borate glass were produced by argon beam sputtering and their specific ionic conductivities were measured by impedance spectroscopy. The conductivity of as-sputtered films is about two orders of magnitude higher compared to the conductivity of bulk glasses produced from the melt. Furthermore, thin films with a thickness of 7–125nm reveal a remarkable finite size effect after annealing: with decreasing thickness the specific dc conductivity increases about three orders of magnitude.