Volume 231, Issue 7-8 - Themed Issue: Mobility of Ions in Solids / Guest Editor: Paul Heitjans

Fundamental research on lithium ion dynamics in solids is important to develop functional materials for, e.g. sensors or energy storage systems. In many cases a comprehensive understanding is only possible if experimental data are compared with predictions from diffusion models. Nuclear magnetic resonance (NMR), besides other techniques such as mass tracer or conductivity measurements, is known as a versatile tool to investigate ion dynamics. Among the various time-domain NMR techniques, NMR relaxometry, in particular, serves not only to measure diffusion parameters, such as jump rates and activation energies, it is also useful to collect information on the dimensionality of the underlying diffusion process. The latter is possible if both the temperature and, even more important, the frequency dependence of the diffusion-induced relaxation rates of actually polycrystalline materials is analyzed. Here we present some recent systematic relaxometry case studies using model systems that exhibit spatially restricted Li ion diffusion. Whenever possible we compare our results with data from other techniques as well as current relaxation models developed for 2D and 1D diffusion. As an example, 2D ionic motion has been verified for the hexagonal form of LiBH 4 ; in the high-temperature limit the diffusion-induced 7 Li NMR spin-lattice relaxation rates follow a logarithmic frequency dependence as is expected from models introduced for 2D diffusion. A similar behavior has been found for Li x NbS 2 . In Li 12 Si 7 a quasi-1D diffusion process seems to be present that is characterized by a square root frequency dependence and a temperature behavior of the 7 Li NMR spin-lattice relaxation rates as predicted. Most likely, parts of the Li ions diffuse along the Si 5 rings that form chains in the Zintl phase.

Ti-based materials exhibit suitable properties for usage in secondary Li- and Na-ion batteries and were in the focus of several electrochemical and ion conductivity studies. A material of such interest is layer-structured, monoclinic Na 2 Ti 3 O 7 . Additionally, the sodium in Na 2 Ti 3 O 7 can be replaced completely with lithium to achieve monoclinic Li 2 Ti 3 O 7 , whose electrochemical properties were already investigated as well. Both materials exhibit interesting properties such as zero-strain behavior upon intercalation and high cycling stability. However, there is still a lack of fundamental understanding of the ion diffusivity of both Na and Li in the corresponding host structure. Solid-state nuclear magnetic resonance (NMR) spectroscopy is used here for the first time to reveal the cation dynamics in layered Na 2 Ti 3 O 7 and Li 2 Ti 3 O 7 . This includes activation energies for the ionic motion and jump rates on the microscopic scale from NMR spin-lattice relaxation (SLR), spin-alignment echo (SAE), and 2D NMR exchange techniques. Moreover, the dimensionality of the ionic motion is investigated by frequency-dependent NMR SLR. Structural details are studied using magic-angle spinning (MAS) NMR spectroscopy. Results for the electric field gradient at the Na and Li site, respectively, are compared with those from theoretical calculations performed within this study. The dynamics are similar for both cations, and the frequency-dependence of the 7 Li NMR SLR rate indicates Li motion confined to two dimensions. Thus, these two materials may be regarded a model system for low-dimensional diffusion of two different cations.

In many applications it has been found that the standard generalized gradient approximation (GGA) does not accurately describe weak chemical bond and electronic properties of solids containing transition metals. In this work, we have considered the intercalation material 1T-Li x TiS 2 (0≤ x ≤1) as a model system for the evaluation of the accuracy of GGA and corrected GGA with reference to the availabile experimental data. The influence of two different dispersion corrections (D3 and D-TS) and an on-site Coulomb repulsion term (GGA+U) on the calculated structural and electronic properties is tested. All calculations are based on the Perdew-Burke-Ernzerhof (PBE) functional. An effective U value of 3.5 eV is used for titanium. The deviation of the calculated lattice parameter c for TiS 2 from experiment is reduced from 14 % with standard PBE to −2 % with PBE+U and Grimme’s D3 dispersion correction. 1T-TiS 2 has a metallic ground state at PBE level whereas PBE+U predicts an indirect gap of 0.19 eV in agreement with experiment. The 7 Li chemical shift and quadrupole coupling constants are in reasonable agreement with the experimental data only for PBE+U-D3. An activation energy of 0.4 eV is calculated with PBE+U-D3 for lithium migration via a tetrahedral interstitial site. This result is closer to experimental values than the migration barriers previously obtained at LDA level. The proposed method PBE+U-D3 gives a reasonable description of structural and electronic properties of 1T-Li x TiS 2 in the whole range 0≤ x ≤1.

Geometric information about ion migration (diffusion pathways) and knowledge about the associated energy landscape (migration activation barriers) are essential cornerstones for a comprehensive understanding of lithium transport in solids. Although many lithium-ion conductors are discussed, developed, and already used as energy-storage materials, fundamental knowledge is often still lacking. In this microreview, we give an introduction to the experimental and computational methods used in our subproject within the research unit FOR 1277, “Mobility of Lithium Ions in Solids (molife)”. These comprise, amongst others, neutron diffraction, topological analyses (procrystal-void analysis and Voronoi–Dirichlet partitioning), examination of scattering-length density maps reconstructed via maximum-entropy methods (MEM), analysis of probability-density functions (PDFs) and one-particle potentials (OPPs), as well as climbing-image nudged-elastic-band (cNEB) computations at density-functional theory (DFT) level. The results of our studies using these approaches on ternary lithium oxides and sulfides with different conduction characteristics (fast/slow) and dimensionalities (one-/two-/three-dimensional) are summarized, focusing on the close orbit of the research unit. Not only did the investigations elucidate the lithium-diffusion pathways and migration activation energies in the studied compounds, but we also established a versatile set of methods for the evaluation of data of differing quality.

In order to improve our understanding of the Li-mobility in oxide glass networks with Li as the principle mobile particle, electrical conductivity and self-diffusivity of lithium was studied in two phosphate (0.2 Li 2 O·0.8 P 2 O 5 , 0.3 Li 2 O·0.2 MgO·0.5 P 2 O 5 ) and one borate (0.25 Li 2 O·0.75 B 2 O 3 ) glass compositions. Conductivity measurements provided information about ion dynamics while isotope-exchange experiments involving isotopically enriched Li diffusion glass couples provided information about long-range diffusivity of Li-isotopes through borate and phosphate networks. Due to the limitations of individual glass stabilities, the temperature range for selected experiments was very small, e.g. as in the case of Li–phosphate composition between 373 and 520 K. The activation energy for Li-migration derived from conductivity measurements was similar for Li–Mg–phosphate and Li–borate, 90.4 and 85.2 kJ/mol, while for pure Li–phosphate the value was 74.7 kJ/mol. In the case of self-diffusion, the activation energies were comparable with Li–Mg–phosphate having the highest value of 76.9 kJ/mol while Li–phosphate and Li–borate had almost the same value of 72.9 and 72.2 kJ/mol, respectively. In these glass compositions with similar Li-cation concentration, the differences in the mobility predominantly depend on structural arrangement of building units and the spatial distribution of negative potentials, as reflected in the value of H R / f , i.e. the Haven ratio divided by the correlation factor, as a mean to better understand the diffusion mechanism in glass structures, where vacancy vs. interstitial diffusion cannot be clearly defined. For Li–phosphate almost unconstrained Li-migration was indicated with the H R / f value of 0.98, while Li–Mg–phosphate had the most structural constraint on mobilized Li-cations, with the H R / f value of 0.30. Findings are compared with silicate (Li 2 O·3 SiO 2 ) and aluminosilicate (Li 2 O·Al 2 O 3· 4 SiO 2 ) glasses from our previous studies in order to elaborate the effect of network topology.

Physical properties of solid materials can be strongly modified by pressure treatment at elevated temperatures. This study focuses on the compaction-induced behavior of powdered amorphous solids using Li 2 Si 3 O 7 -glass as an example. Experiments were carried out on distinct fractions with particle sizes from <25 μm to 224–250 μm. Measurements of electrical conductivity using impedance spectroscopy were carried out in situ at pressures up to 930 MPa and at temperatures from 373 K to 667 K. Simultaneous monitoring of volume changes allows correlating conductivity and porosity of samples. To study the effect of adsorbed water on surfaces, the material was pretreated by flushing with water-bearing nitrogen before the experiment. Continuous increase of electrical conductivity upon pressurization was observed for all particle size fractions both in the brittle and in the plastic deformation regimes. The pressure derivative of DC conductivity strongly increases with grain size at low T (373 K). At high T (608–665 K) the effect is less pronounced due to the onset of welding of particles forming continuous pathways for charge transport without grain boundaries as barriers. Welding of particles occurs already at temperatures significantly below the glass transition temperature, induced by strong local forces at grain-grain contacts. No effect by pretreatment of glass powder with water vapor was observed at low temperature, while at high temperature surface modification by adsorbed water resulted in enhancement of electrical conductivity, probably caused by lowering of viscosity at grain surfaces, which facilitates welding of particles.

Abstract: Reaction with n-butyl lithium (BuLi) is commonly used to study the kinetics of intercalation of lithium in electrode materials for batteries. We performed lithium isotope exchange experiments on TiS 2 single crystals as model system to determine the irreversible step in the intercalation process. Single crystals of TiS 2 were prepared by chemical vapor transport and intercalated by lithium with natural isotopic signature (7% 6 Li 93% 7 Li) using a 2.5 molar solution of BuLi in hexane. Crystals were homogenized at 313 K in a dry argon atmosphere for 1–2 months and then exposed for several days to ~1 molar BuLi solution enriched in the light lithium isotope (95% 6 Li, 5% 7 Li). After the isotopic exchange experiments concentration profiles of the isotopes were measured parallel to the a/b-plane of the crystal using Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry (LA-ICP-MS). Profiles show that already intercalated lithium isotopes can be completely exchanged without significant increase in total lithium content. This observation implies that, after adsorption of BuLi at the crystal surface, the butyl radical can freely jump from one lithium atom to a neighbouring one. The jump sequence is either finished by desorption of a BuLi molecule or by combination of two adjacent butyl radicals forming an octane molecule or other kinds of deactivation of butyl radicals. The latter step is irreversible. Self diffusion coefficients determined by fitting of the lithium isotope profiles are in good agreement with chemical diffusivities determined by lithium intercalation experiments.

Ceramics with nm-sized dimensions are widely used in various applications such as batteries, fuel cells or sensors. Their oftentimes superior electrochemical properties as well as their capabilities to easily conduct ions are, however, not completely understood. Depending on the method chosen to prepare the materials, nanostructured ceramics may be equipped with a large area fraction of interfacial regions that exhibit structural disorder. Elucidating the relationship between microscopic disorder and ion dynamics as well as electrochemical performance is necessary to develop new functionalized materials. Here, we highlight some of the very recent studies on ion transport and electrochemical properties of nanostructured ceramics. Emphasis is put on TiO 2 in the form of nanorods, nanotubes or being present as mesoporous material. Further examples deal with nanocrystalline peroxides such as Li 2 O 2 or nanostructured oxides (Li 2 TiO 3 , LiAlO 2 , LiTaO 3 , Li 2 CO 3 and Li 2 B 4 O 7 ). These materials served as model systems to explore the influence of ball-milling on overall ionic transport.

Titanium dioxide nanotubes were obtained by a simple alkaline hydrothermal route forming titanate nanotubes in a first step and subsequent ion exchange with hydrochloric acid. The obtained TiO 2 nanotubes were structurally characterized by powder X-ray diffraction, Raman spectroscopy, transmission electron microscopy. Electrochemical analysis by cyclic voltammetry shows the presence oif anatase and TiO 2 -B phase in the tubes. From Randles-Ševcik plots a diffusion-controlled anodic reaction is found for the anatase phase, the TiO 2 -B phase, however, exhibits a pseudocapacitive behavior. A quite high amount of TiO 2 -B phase present in our TiO 2 nanotubes is responsible for high charge and discharge capacities exceeding those found by other research groups for comparable nanotube systems. Furthermore, galvanostatic long-time cycling proves a high stability of the TiO 2 nanotubes.

This article reports on Li self-diffusion in lithium containing metal oxide compounds. Case studies on LiNbO 3 , Li 3 NbO 4 , LiTaO 3 , LiAlO 2 , and LiGaO 2 are presented. The focus is on slow diffusion processes on the nanometer scale investigated by macroscopic tracer methods (secondary ion mass spectrometry, neutron reflectometry) and microscopic methods (nuclear magnetic resonance spectroscopy, conductivity spectroscopy) in comparison. Special focus is on the influence of structural disorder on diffusion.

Solid-state nuclear magnetic resonance spectroscopy is an efficient technique to characterize dynamics and structure of materials. It has been widely used to elucidate ion dynamics in lithium ion conductors. Fast moving lithium ions are needed in energy storage devices, whereas slow ion motion is exploited in some materials used, for example, as blankets in fusion reactors. β -lithium gallium oxide (LiGaO 2 ) is a slow Li + ionic conductor similar to γ -lithium aluminum oxide (LiAlO 2 ). In an ion conductor, in addition to the main diffusion process, localized motions (to-and-fro jumps) may be present. In the present work, with the help of solid-state NMR experiments, we report on the localized movements of Li + ionic species in β -LiGaO 2 in the temperature range between 300 K and 450 K. In this work, we have mainly extracted the peculiarities of ion dynamics from 7 Li spin-alignment echo NMR measurements and the observation of the motional narrowing of the central transition signal of 7 Li.

We have studied the Li diffusion in the complex hydride Li 3 (NH 2 ) 2 I which appears to exhibit fast Li ion conduction. To get a detailed insight into the Li motion, we have applied 7 Li nuclear magnetic resonance spectroscopy methods, such as spin-lattice relaxation in the laboratory and rotating frames of reference, as well as spin-alignment echo. This combined approach allows us to probe Li jump rates over the wide dynamic range (~10 2 –10 9 s −1 ). The spin-lattice relaxation data in the range 210–410 K can be interpreted in terms of a thermally-activated Li jump process with a certain distribution of activation energies. However, the low-temperature spin-alignment echo decays at T ≤200 K suggest the presence of another Li jump process with the very low effective activation energy.