Fast ion conducting garnet materials have been identified as promising electrolytes for all solid-state batteries. However, reliable synthetic routes to materials with fully elucidated cation site occupancies where an enhancement in lithium conductivity is observed remains a challenge. Ca-Incorporation is developed here as a promising approach to enhance the ionic conductivity of garnet-type Li7−xLa3Zr2−xTaxO12 phases. Here we present a new sol–gel synthetic strategy as a facile route to the preparation of materials of a desired stoichiometry optimized for Li+ conductivity. We have found that the ionic conductivity of Li6.4La3Zr1.4Ta0.6O12 is increased by a factor of four by the addition of 0.2 mol of Ca per formula unit. Ca is incorporated in the garnet lattice where it has no effect on the sinterability of the material and is predominately located at the La sites. We anticipate that the ease of our synthetic route and the phases presented here represents a starting point for the further realization of solid state electrolyte compositions with similarly high Li+ conductivities using this methodology.

The series of cation ordered double perovskites Ba2Nd1−xFexMoO6 undergo a compositionally-driven transition from localised to delocalised electronic behaviour, as exhibited in the end members Ba2NdMoO6 and Ba2FeMoO6 respectively. Rietveld structural analyses against neutron diffraction data indicate that all compounds are stoichiometric in oxygen and show replacement of Nd3+ with Fe3+ on the larger of the two octahedral sites in the cation-ordered structure. A tetragonal distortion persists up to x = 0.25 and Ba2Nd0.9Fe0.1MoO6 shows freezing of magnetic moments at 5 K. Neutron scattering indicates an absence of long range magnetic ordering suggesting the formation of a spin glass phase below this temperature. Ba2Nd0.75Fe0.25MoO6 shows high electrical resistivity with a temperature dependence indicative of fully localised electronic behaviour. Despite the Fe3+ occupation (0.25) being above the percolation limit (0.195) for the face centred cubic lattice, this compound shows no magnetic ordering at 2 K. Compositions in the range 0.30 ≤ x ≤ 0.85 give a mixture of two perovskite phases with lattice parameters of ca. 8.4 and 8.1 Å. The single phase compositions Ba2Nd0.10Fe0.90MoO6 and Ba2Nd0.05Fe0.95MoO6 form face centred cubic structures with long range magnetic ordering of the Fe3+ moments below ferrimagnetic ordering transitions of 270 and 285 K respectively. Neutron diffraction shows almost complete parallel alignment of the Fe3+ moments and, combined with conductivity measurements showing delocalised electronic behaviour in Ba2Nd0.10Fe0.90MoO6, indicate ferrimagnetic ordering of Fe3+ and delocalised Mo5+.

The fast ionic, high temperature (HT) phase of LiBH4 can be stabilized by Br¯ substitution. Lithium borohydride bromide compounds, Li(BH4)1–xBrx, have been synthesized mechanochemically, with and without thermal treatment and the resulting phase behavior determined as a function of composition. Single phase materials exist for 0.29 ≤ x ≤ 0.50 with conductivity 2 orders of magnitude higher than LiBH4 at 313 K. Powder neutron diffraction has been used to resolve the details of the crystal structure of one such compound. These demonstrate that 7Li(11BD4)0.667Br0.333 retains the HT structure (hexagonal space group P63mc, a ≈ 4.2 Å, c ≈ 6.7 Å) from 293 to 573 K. The borohydride bromide exhibits considerable static and dynamic disorder, the latter invoking complex rotational motion of the (BH4)¯ anions.

Grinding together the solid acid HLaTiO4 with stoichiometric quantities of lithium hydroxide monohydrate gives the solid solution H1–xLixLaTiO4. The structures of these crystalline phases have been refined against neutron powder diffraction data to show that all of these compounds crystallize in the centrosymmetric space group P4/nmm. The protons and lithium cations occupy sites between the perovskite layers; the former in hydroxide groups that hydrogen-bond to adjacent layers while Li+ is in four-coordinate sites that bridge the perovskite slabs with a geometry intermediate between square-planar and tetrahedral. The reaction proceeds rapidly, but the unit cell size continues to evolve over the course of days with a gradual compression along the interlayer direction that can be modeled using a power law dependence reminiscent of an Ostwald ripening process. On heating, these materials undergo a mass loss because of dehydration but retain the layered Ruddlesden–Popper structure up to 480 °C before a substantial loss of crystallinity on further heating to 600 °C. Impedance spectroscopy studies of the dehydrated materials shows that Li+ mobility in these materials is lower than the LiLaTiO4 end member, possibly because of microstructural effects causing large intergrain resistance through the defective phases.

The structures of new phases Li6CaLa2Sb2O12 and Li6.4Ca1.4La2Sb2O12 have been characterised using neutron powder diffraction. Rietveld analyses show that both compounds crystallise in the space group la3̄d and contain the lithium cations in a complex arrangement with occupational disorder across oxide tetrahedra and distorted oxide octahedra, with considerable positional disorder in the latter. Variable temperature neutron diffraction experiments on Li6.4Ca1.4La2Sb2O12 show the structure is largely invariant with only a small variation in the lithium distribution as a function of temperature. Impedance spectroscopy measurements show that the total conductivity of Li6CaLa2Sb2O12 is several orders of magnitude smaller than related lithium-stuffed garnets with σ=10−7 S cm−1 at 95 °C and an activation energy of 0.82(3) eV. The transport properties of the conventional garnets Li3Gd3Te2O12, Li3Tb3Te2O12, Li3Er3Te2O12 and Li3Lu3Te2O12 have been evaluated and consistently show much lower values of conductivity, σ≤4.4×10−6 S cm−1 at 285 °C and activation energies in the range 0.77(4)≤Ea/eV≤1.21(3).The lithium-stuffed garnets Li6CaLa2Sb2O12 and Li6.4Ca1.4La1.6Sb2O12 accommodate lithium in a complex distribution across oxide tetrahedra and octahedra. The total conductivity of Li6CaLa2Sb2O12 is considerably lower than reported for related fast-ion conducting garnets due to a much larger intra-grain contribution to the resistivity than is commonly found for this family of compounds.Research highlights► Lithium stuffed garnets have a complex distribution of Li+ across multiple sites. ► They show fast lithium ion conductivity with an activation energy of 0.82 eV. ► This conductivity is much lower than related garnets due to grain boundary effects. ► The stoichiometric garnets show minimal ion mobility.

Garnets are capable of accommodating an excess of lithium cations beyond that normally found in this prototypical structure. This excess lithium is found in a mixture of coordination environments with considerable positional and occupational disorder and leads to ionic conductivity of up to 4 × 10−4 S cm−1 at room temperature. This high value for total conductivity, combined with excellent thermal and (electro)chemical resistance makes these candidate materials for operation in all solid-state batteries. This review looks at garnets with a wide range of stoichiometries and lithium concentrations and the impact of complex lithium distributions and crystallographic order/disorder transitions on the transport properties of these materials.

The cation-ordered perovskites Ba2NdMoO6 and Ba2Nd1−xYxMoO6 have been structurally characterised by a combination of neutron and X-ray powder diffraction. Ba2NdMoO6 retains the tetragonal room temperature structure on cooling to 150 K [I4/m; a = 5.98555(5) Å, c = 8.59510(10) Å] although the MoO6 octahedra distort with an elongation of two trans Mo–O bonds. Neutron diffraction data collected at T ≤ 130 K show that this compound has undergone a structural distortion to a triclinic space group, although the MoO6 octahedra do not distort any further on cooling below this temperature [at 130 K: I; 5.97625(14) Å, 5.9804(2) Å, 8.59650(13) Å, 89.876(2)°, 89.921(3)°, 89.994(2)°]. The room temperature tetragonal space group symmetry of Ba2NdMoO6 is preserved in the series Ba2Nd1−xYxMoO6 up to composition 0.35 ≤ x < 0.5. The lattice parameters converge as the value of x increases until cubic symmetry is reached for the composition of Ba2Nd0.5Y0.5MoO6 [Fmm; a = 8.4529(3) Å]. Magnetic susceptibility measurements show that all of these compounds display the Curie–Weiss behaviour associated with a fully localised electronic system. The paramagnetic moments show good agreement with those anticipated to arise from the spin-only contribution from Mo5+ (S = 1/2, µso = 1.73 µB) and the moment of 3.62 µB associated with the spin–orbit coupling of the 4I9/2 ground state of Nd3+. For x ≤ 0.125 this series shows a magnetic transition in the range 10 to 15 K indicative of a distortion of the MoO6 octahedra in these compounds that is similar to Ba2NdMoO6.

The layered perovskite HLaTiO4 reacts stoichiometrically with LiOH·H2O at room temperature to give targeted compositions in the series HxLi1−xLaTiO4. Remarkably, the Li+ and H+ ions are quantitatively exchanged in the solid state and this allows stoichiometric control of ion exchange for the first time in this important series of compounds.

Lithium molybdate has been prepared by grinding LiOH·H2O with MoO3 in air at room temperature. X-Ray powder diffraction data show that the formation of highly crystalline Li2MoO4 is largely complete after 10 min. The phenacite structure of this material is the same as that derived from an X-ray diffraction study of a single crystal obtained from aqueous solution [R; a = 14.3178(14) Å, c = 9.5757(9) Å]. Anhydrous lithium hydroxide fails to give the same reaction indicating that the water of crystallisation of LiOH·H2O is a vital component in this rapid synthesis. Differential scanning calorimetry measurements show that this reaction can proceed spontaneously between the two stable solid reagents at sub-ambient temperatures and is driven by the liberation of water from the crystalline lattice. Lithium molybdate prepared in this manner has significantly smaller and more regularly shaped particles than samples prepared by other synthetic methods.

Polycrystalline samples of the garnets Li3+xNd3Te2−xSbxO12 have been prepared by high temperature solid state synthesis. X-ray and neutron powder diffraction data show that all compounds crystallize in the space group Ia3̅d with lattice parameters in the range 12.55576(12) Å for x = 0.05 to 12.6253(2) Å for x = 1.5. The lithium is distributed over a mixture of oxide tetrahedra and heavily distorted octahedra. Increasing the lithium content in these compounds leads to the introduction of vacancies onto the tetrahedral position and an increasing concentration of lithium found in the octahedra. The latter exhibit considerable positional disorder with two lithium cations positions within each octahedron. Impedance measurements show fast ion conduction with an activation energy of ca. 0.59(6) eV that is largely invariant with composition. Solid-state Li NMR measurements indicate that there is no exchange of lithium between the different coordination environments. These results indicate that lithium conduction in the garnet structure occurs exclusively via a network of edge-linked distorted oxide octahedra and that the tetrahedrally coordination lithium plays no part in the transport properties.

The garnet system Li5+xBaxLa3−xTa2O12 shows an unprecedented Li+ content (x ≤ 1.6) and short Li–Li distances of ca 2.44 Å between majority occupied sites suggesting that the high Li+ mobility requires a complex cooperative mechanism.

The garnets Li3Nd3W2O12 and Li5La3Sb2O12 have been prepared by heating the component oxides and hydroxides in air at temperatures up to 950 °C. Neutron powder diffraction has been used to examine the lithium distribution in these phases. Both compounds crystallise in the space group Ia3¯d with lattice parameters a=12.46869(9) Å (Li3Nd3W2O12) and a=12.8518(3) Å (Li5La3Sb2O12). Li3Nd3W2O12 contains lithium on a filled, tetrahedrally coordinated 24d site that is occupied in the conventional garnet structure. Li5La3Sb2O12 contains partial occupation of lithium over two crystallographic sites. The conventional tetrahedrally coordinated 24d site is 79.3(8)% occupied. The remaining lithium is found in oxide octahedra which are linked via a shared face to the tetrahedron. This lithium shows positional disorder and is split over two positions within the octahedron and occupies 43.6(4)% of the octahedra. Comparison of these compounds with related d0 and d10 phases shows that replacement of a d0 cation with d10 cation of the same charge leads to an increase in the lattice parameter due to polarisation effects.Li3Nd3W2O12 contains lithium on a filled tetrahedrally coordinated site. The Li-rich phase Li5La3Sb2O12 accommodates lithium on a mixture of tetrahedrally and octahedrally coordinated sites linked by a shared oxide face. The d0 garnets show substantially contracted unit cells compared to d10 analogues.

The garnet system Li5+xBaxLa3−xTa2O12 shows an unprecedented Li+ content (x ≤ 1.6) and short Li–Li distances of ca 2.44 Å between majority occupied sites suggesting that the high Li+ mobility requires a complex cooperative mechanism.

The layered perovskite HLaTiO4 reacts stoichiometrically with LiOH·H2O at room temperature to give targeted compositions in the series HxLi1−xLaTiO4. Remarkably, the Li+ and H+ ions are quantitatively exchanged in the solid state and this allows stoichiometric control of ion exchange for the first time in this important series of compounds.

Lithium molybdate has been prepared by grinding LiOH·H2O with MoO3 in air at room temperature. X-Ray powder diffraction data show that the formation of highly crystalline Li2MoO4 is largely complete after 10 min. The phenacite structure of this material is the same as that derived from an X-ray diffraction study of a single crystal obtained from aqueous solution [R; a = 14.3178(14) Å, c = 9.5757(9) Å]. Anhydrous lithium hydroxide fails to give the same reaction indicating that the water of crystallisation of LiOH·H2O is a vital component in this rapid synthesis. Differential scanning calorimetry measurements show that this reaction can proceed spontaneously between the two stable solid reagents at sub-ambient temperatures and is driven by the liberation of water from the crystalline lattice. Lithium molybdate prepared in this manner has significantly smaller and more regularly shaped particles than samples prepared by other synthetic methods.

Fast ion conducting garnet materials have been identified as promising electrolytes for all solid-state batteries. However, reliable synthetic routes to materials with fully elucidated cation site occupancies where an enhancement in lithium conductivity is observed remains a challenge. Ca-Incorporation is developed here as a promising approach to enhance the ionic conductivity of garnet-type Li7−xLa3Zr2−xTaxO12 phases. Here we present a new sol–gel synthetic strategy as a facile route to the preparation of materials of a desired stoichiometry optimized for Li+ conductivity. We have found that the ionic conductivity of Li6.4La3Zr1.4Ta0.6O12 is increased by a factor of four by the addition of 0.2 mol of Ca per formula unit. Ca is incorporated in the garnet lattice where it has no effect on the sinterability of the material and is predominately located at the La sites. We anticipate that the ease of our synthetic route and the phases presented here represents a starting point for the further realization of solid state electrolyte compositions with similarly high Li+ conductivities using this methodology.

The cation-ordered perovskites Ba2NdMoO6 and Ba2Nd1−xYxMoO6 have been structurally characterised by a combination of neutron and X-ray powder diffraction. Ba2NdMoO6 retains the tetragonal room temperature structure on cooling to 150 K [I4/m; a = 5.98555(5) Å, c = 8.59510(10) Å] although the MoO6 octahedra distort with an elongation of two trans Mo–O bonds. Neutron diffraction data collected at T ≤ 130 K show that this compound has undergone a structural distortion to a triclinic space group, although the MoO6 octahedra do not distort any further on cooling below this temperature [at 130 K: I; 5.97625(14) Å, 5.9804(2) Å, 8.59650(13) Å, 89.876(2)°, 89.921(3)°, 89.994(2)°]. The room temperature tetragonal space group symmetry of Ba2NdMoO6 is preserved in the series Ba2Nd1−xYxMoO6 up to composition 0.35 ≤ x < 0.5. The lattice parameters converge as the value of x increases until cubic symmetry is reached for the composition of Ba2Nd0.5Y0.5MoO6 [Fmm; a = 8.4529(3) Å]. Magnetic susceptibility measurements show that all of these compounds display the Curie–Weiss behaviour associated with a fully localised electronic system. The paramagnetic moments show good agreement with those anticipated to arise from the spin-only contribution from Mo5+ (S = 1/2, µso = 1.73 µB) and the moment of 3.62 µB associated with the spin–orbit coupling of the 4I9/2 ground state of Nd3+. For x ≤ 0.125 this series shows a magnetic transition in the range 10 to 15 K indicative of a distortion of the MoO6 octahedra in these compounds that is similar to Ba2NdMoO6.

Garnets are capable of accommodating an excess of lithium cations beyond that normally found in this prototypical structure. This excess lithium is found in a mixture of coordination environments with considerable positional and occupational disorder and leads to ionic conductivity of up to 4 × 10−4 S cm−1 at room temperature. This high value for total conductivity, combined with excellent thermal and (electro)chemical resistance makes these candidate materials for operation in all solid-state batteries. This review looks at garnets with a wide range of stoichiometries and lithium concentrations and the impact of complex lithium distributions and crystallographic order/disorder transitions on the transport properties of these materials.

The series of cation ordered double perovskites Ba2Nd1−xFexMoO6 undergo a compositionally-driven transition from localised to delocalised electronic behaviour, as exhibited in the end members Ba2NdMoO6 and Ba2FeMoO6 respectively. Rietveld structural analyses against neutron diffraction data indicate that all compounds are stoichiometric in oxygen and show replacement of Nd3+ with Fe3+ on the larger of the two octahedral sites in the cation-ordered structure. A tetragonal distortion persists up to x = 0.25 and Ba2Nd0.9Fe0.1MoO6 shows freezing of magnetic moments at 5 K. Neutron scattering indicates an absence of long range magnetic ordering suggesting the formation of a spin glass phase below this temperature. Ba2Nd0.75Fe0.25MoO6 shows high electrical resistivity with a temperature dependence indicative of fully localised electronic behaviour. Despite the Fe3+ occupation (0.25) being above the percolation limit (0.195) for the face centred cubic lattice, this compound shows no magnetic ordering at 2 K. Compositions in the range 0.30 ≤ x ≤ 0.85 give a mixture of two perovskite phases with lattice parameters of ca. 8.4 and 8.1 Å. The single phase compositions Ba2Nd0.10Fe0.90MoO6 and Ba2Nd0.05Fe0.95MoO6 form face centred cubic structures with long range magnetic ordering of the Fe3+ moments below ferrimagnetic ordering transitions of 270 and 285 K respectively. Neutron diffraction shows almost complete parallel alignment of the Fe3+ moments and, combined with conductivity measurements showing delocalised electronic behaviour in Ba2Nd0.10Fe0.90MoO6, indicate ferrimagnetic ordering of Fe3+ and delocalised Mo5+.