Wouter J. H. Borghols

Size Effects in the Li4+xTi5O12 Spinel

The nanosized Li(4+x)Ti(5)O(12) spinel is investigated by electrochemical (dis)charging and neutron diffraction. The near-surface environment of the nanosized particles allows higher Li ion occupancies, leading to a larger storage capacity. However, too high surface lithium storage leads to irreversible capacity loss, most likely due to surface reconstruction or mechanical failure. A mechanism where the large near-surface capacity ultimately leads to surface reconstruction rationalizes the existence of an optimal particle size. Recent literature attributes the curved voltage profiles, leading to a reduced length of the voltage plateau, of nanosized electrode particles to strain and interface energy from the coexisting end members. However, the unique zero-strain property of the Li(4+x)Ti(5)O(12) spinel implies a different origin of the curved voltage profiles observed for its nanosized crystallites. It is proposed to be the consequence of different structural environments in the near-surface region, depending on the distance from the surface and surface orientation, leading to a distribution of redox potentials in the near-surface area. This phenomenon may be expected to play a significant role in all nanoinsertion materials displaying the typical curved voltage curves with reduced length of the constant-voltage plateau.

Lithium Storage in Amorphous TiO2 Nanoparticles

Amorphous titanium oxide nanoparticles were prepared from titanium isopropoxide. In situ measurements reveal an extraordinary high capacity of 810 mAh/g on the first discharge. Upon cycling at a charge/discharge rate of 33.5 mA/g, this capacity gradually decreases to 200 mAh/g after 50 cycles. The origin of this fading was investigated using X-ray absorption spectroscopy and solid-state nuclear magnetic resonance. These measurements reveal that a large fraction of the total amount of the consumed Li atoms is due to the reaction of H2O/OH species adsorbed at the surface to Li2O, explaining the irreversible capacity loss. The reversible capacity of the bulk, leading to the Li0.5TiO2 composition, does not explain the relatively large reversible capacity, implying that part of Li2O at the TiO2 surface may be reversible. The high reversible capacity, also at large (dis)charge rates up to 3.35 A/g (10C), makes this amorphous titanium oxide material suitable as a low cost electrode material in a high power battery.

Dynamic solubility limits in nanosized olivine LiFePO4.

Because of its stability, nanosized olivine LiFePO(4) opens the door toward high-power Li-ion battery technology for large-scale applications as required for plug-in hybrid vehicles. Here, we reveal that the thermodynamics of first-order phase transitions in nanoinsertion materials is distinctly different from bulk materials as demonstrated by the decreasing miscibility gap that appears to be strongly dependent on the overall composition in LiFePO(4). In contrast to our common thermodynamic knowledge, that dictates solubility limits to be independent of the overall composition, combined neutron and X-ray diffraction reveals strongly varying solubility limits below particle sizes of 35 nm. A rationale is found based on modeling of the diffuse interface. Size confinement of the lithium concentration gradient, which exists at the phase boundary, competes with the in bulk energetically favorable compositions. Consequently, temperature and size diagrams of nanomaterials require complete reconsideration, being strongly dependent on the overall composition. This is vital knowledge for the future nanoarchitecturing of superior energy storage devices as the performance will heavily depend on the disclosed nanoionic properties.

The electronic structure and ionic diffusion of nanoscale LiTiO2 anatase

Upon lithium insertion in the pristine TiO2 anatase phase the theoretical maximum of LiTiO2 can be reached in crystallite sizes less than approximately 10 nm, whereas bulk compositions appear limited to Li(x) approximately 0.6TiO2 at room temperature. Both X-ray absorption spectroscopy (XAS) and ab initio calculations have been applied to probe the electronic structure of the newly formed LiTiO2 phase. These results indicate that a large majority of the Li-2s electrons reside at the Ti-3d(t2g)/4s hybridized site. About 10% of these electrons are transferred to non-localized states which makes this compound a good electronic conductor. Ionic conductivity is probed by nuclear magnetic resonance (NMR) relaxation experiments indicating relatively small hopping rates between the Li-ion sites in LiTiO2. Formation of the poor ionic-conducting LiTiO2 at the surface of the particles explains why micro-anatase Li(x)TiO2 is not able to reach the theoretical maximum capacity at room temperature, and why this theoretical maximum capacity reached in nano-sized materials cannot be (dis)charged at high rates.

Direct synthesis of nanocrystalline Li0.90FePO4: observation of phase segregation of anti-site defects on delithiation

Solid solutions of LixFePO4 are of tremendous interest because of a proposed increase in ion transport properties, but the formation of these solutions at high temperatures is difficult if not impossible and direct synthesis is difficult and rarely reported. Here we report modified polyol syntheses which produce nanocrystalline Li1−yFePO4 directly, where the maximum Li substoichiometry on the M1 site sustained at synthesis temperatures of 320 °C is about 10%. High target lithium vacancy concentrations promote the increase in anti-site disorder of Li+ and Fe2+, as this process is driven by vacancy stabilization. Combined neutron and X-ray diffraction on partial delithiated substoichiometric olivines reveals segregated defect-free (where Li is extracted) and defect-ridden (where Li remains) regions. This proves (1) that the anti-site defects obstruct Li+ diffusion explaining the detrimental electrochemistry and (2) that the anti-site defects form clusters. Finally, preferential anisotropic strain broadening in the bc-plane indicates the existence of a coherent interface between the Li-poor and Li-rich phases. Along with the size broadening upon delithiation this proves that in nano-sized LixFePO4 the two phases coexist within a single particle, which is not expected based on thermodynamics arguments due to the energy penalty associated with the coherent interface. Thereby, these results give important and unique insight and understanding in the properties of nano sized LiFePO4.

Competing ferri- and antiferromagnetic phases in geometrically frustrated LuFe2O4.

We present a detailed study of magnetism in LuFe(2)O(4), combining magnetization measurements with neutron and soft x-ray diffraction. The magnetic phase diagram in the vicinity of T(N) involves a metamagnetic transition separating an antiferro- and a ferrimagnetic phase. For both phases the spin structure is refined by neutron diffraction. Observed diffuse magnetic scattering far above T(N) is explained in terms of near degeneracy of the magnetic phases.