We present a review of the structural, physical, and chemical properties of both the bulk and the surface layer of lithium iron phosphate (LiFePO4) (LFP) as a positive electrode for Li-ion batteries. Depending on the mode of preparation, different impurities can poison this material. Several impurities are identified, and a quantitative estimate of their concentration is deduced from the combination of X-ray diffraction analysis, Fourier transform infrared spectroscopy, Raman spectroscopy, and magnetic measurements. The surface layer also plays an important role in case the size of the particles is reduced to few tenths of nanometers, and it is characterized by the same means. These properties are correlated with the electrochemical performance of the positive electrode materials. An optimized preparation provides samples with carbon-coated particles free of any impurity phase, insuring structural stability and electrochemical performance that justify the use of this material as a cathode element in new generation of lithium secondary batteries operating at temperatures from 23 up to 60 °C. The electrochemical performances cycle life of commercial batteries 26,650 and ultrarare cell shows over 10,000 cycles at 100% DOD and 100% SOC at 23 °C and 7700 cycles at 60 °C.

In this work, we studied the local environment of a series of LiNiyMnyCo1-2yO2 materials synthesized by wet chemistry via oxalate route. Measurements included the temperature dependence of the magnetic susceptibility chi(m)(T), the magnetization M(H), and the electron spin resonance. The influence of the synthesis conditions on the magnetic and electronic properties is presented and discussed. Magnetic properties and electron spin resonance spectroscopy provides further information on Ni2+, Mn4+ ion distribution in LiNiyMnyCo1-2yO2. The magnetic data allows the determination of the cationic disorder, i.e. the concentration of Ni2+ ions in the 3a lithium sites, as a function of the composition, in quantitative agreement with the results of XRD experiments. We find that the lithium extraction proceeds by different steps. In a first step, Ni2+ in converted first in Ni3+, then in Ni4+ as x decreases from x = 1, while Co remains in the Co3+ valence state. The ESR spectra show a resonance due to Mn4+, which is analyzed self-consistently with magnetic experiments. (C) 2011 Elsevier B. V. All rights reserved.

We report electrical properties of the lithium manganospinel LiMn2 O 4 and its delithiated/lithiated forms, Li1−x+∂Mn2−∂O4 with 0.0≤x≤0.4 and 0.00≤∂0.18. The electrical conductivity has been determined from d.c. measurements as a function of temperature and lithium content in the host lattice. LiMn2O4 exhibits a phase transition in the vicinity of 280 K, which disappears in the lithium-rich samples. Electrical data are analysed using the model of small-polaron transport. ESR spectroscopy has been applied to identify the singular structural features. Investigations as a function of temperature show a reliable determination of the modifications in the cationic sublattice and of the lithium overstoichiometry. The hopping conductivity mechanism between the Mn3+ and Mn4+ sites gives a coherent explanation for the observed ESR signal of cyclotron resonance.

We show that the investigation of magnetic properties is the best tool to identify and quantify the impurities and defects that limit the ability of the lamellar intercalation compounds for use as cathode for Li-ion batteries. The results are illustrated for LiNiO2, LiNi1-yCoyO2, LiCoO2, LiNi0.5Mn0.5O2, and LiNi1/3Mn1/3Co1/3O2 (LNMCO). Despite the extensive studies of these ionic compounds in the past, not only for practical use, but also for themselves, the present work reveals that the magnetic properties of these lamellar compounds have been largely misunderstood, and that, at contrast with the common belief, they do not belong to the family of two dimensional frustrated antiferromagnets. The misunderstanding comes from confusion between extrinsic and intrinsic effects. This distinction allows for an overall understanding of the intrinsic properties of these materials, and opens the route to their optimization, in particular for LNMCO that is the most promising element of this family.

A series of LiNi1/3Mn1/3Co1/3O2 samples with α-NaFeO2 structure were synthesized using tartaric acid as a chelating agent by wet chemical method. Different acid to metal-ion ratios R have been used to investigate the effect of this parameter on the physical and electrochemical properties. The cationic mixing (Ni on the 3a Li sites) have been evaluated by Rietveld refinement and magnetic measurements We determined that the LiNi1/3Mn1/3Co1/3O2 sintered at 900 {degree sign}C for 15 h with an acid to metal ion ratio R=2 was the optimum condition for this synthesis. For this optimized sample, only 1.26% of nickel-ions occupied the 3a Wyckoff site of the lithium-ion sublattice. This material delivered an initial discharge capacity of 172 mAh/g and displayed Coulomb efficiency of 94%.

Li1+x(Ni1/3Mn1/3Co1/3)1−xO2 layered materials were synthesized by the co-precipitation method with different Li/M molar ratios (M=Ni+Mn+Co). Elemental titration evaluated by inductively coupled plasma spectrometry (ICP), structural properties studied by X-ray diffraction (XRD), Rietveld analysis of XRD data, scanning electron microscopy (SEM) and magnetic measurements carried out by superconducting quantum interference devices (SQUID) showed the well-defined α-NaFeO2 structure with cationic distribution close to the nominal formula. The Li/Ni cation mixing on the 3b Wyckoff site of the interlayer space was consistent with the structural model [Li1−yNiy]3b[Lix+yNi(1−x)/3−yMn(1−x)/3Co(1−x)/3]3aO2 (x=0.02, 0.04) and was very small. Both Rietveld refinements and magnetic measurements revealed a concentration of Ni2+-3b ions lower than 2%; moreover, for the optimized sample synthesized at Li/M=1.10, only 1.43% of nickel ions were located into the Li sublattice. Electrochemical properties were investigated by galvanostatic charge–discharge cycling. Data obtained with Li1+x(Ni1/3Mn1/3Co1/3)1−xO2 reflected the high degree of sample optimization. An initial discharge capacity of 150mAhg−1 was delivered at 1C-rate in the cut-off voltage of 3.0–4.3V. More than 95% of its initial capacity was retained after 30 cycles at 1C-rate. Finally, it is demonstrated that a cation mixing below 2% is considered as the threshold for which the electrochemical performance does not change for Li1+x(Ni1/3Mn1/3Co1/3)1−xO2.

Amorphous analogues of lithium-iron phosphates (LFP), which are promising cathode materials for Li ion batteries, were prepared by the standard press-quenching method and their thermal stability, as well as structural and electrical properties, were studied for the first time. The glass transition temperature, T-g, determined by the differential thermal analysis (DTA) is composition-dependent and lies in the 492-523 degrees C range. The local structure, studied by the FTIR absorption spectroscopy, and the thermal stability are found to be almost insensitive to the lithium content. Studies on the electrical properties, carried out by impedance spectroscopy, have shown that the total electrical (predominantly polaronic) conductivity at 450 degrees C approaches 10(-2) S.cm(-1). The room temperature conductivity of samples after their nanocrystallization (induced by annealing at the temperature of the beginning of crystallization) was higher by a factor of 4-10 (depending on composition) than that of the as-received glass. Therefore, nanocrystallization seems to be a promising way to enhance the electrical conductivity of amorphous lithium-iron phosphates.

The structure and magnetic properties of lithium iron pyrophosphate, i.e., Li2Fe3(P2O7)2 or LiFe1.5P2O7, synthesized using a facile metal acetate approach for application in lithium-ion batteries, are investigated in detail. The high-resolution transmission electron microscopy, selected area electron diffraction, and x-ray diffraction measurements indicate that Li2Fe3(P2O7)2 is crystallized in the monoclinic structure, without any indication of crystallographic defects such as dislocations or misfits, and exhibit smooth surface morphology. The evaluated lattice parameters are a=0.698 76 nm, b=0.812 36 nm, c=0.964 22 nm, and β=111.83° (P21/c space group). Infrared spectroscopic measurements indicate the presence of P2O7 groups, which are formed by the two PO4 tetrahedral groups connected together. The magnetic measurements indicate that Li2Fe3(P2O7)2 is a weak antiferromagnetic material with TN=20 K exhibiting a Curie constant Cp=3.38 emu K/mol per Fe ion and a negative value of the Weiss temperature (Θp=−15 K). The absence of higher valence state Fe impurities and antiferromagnetic interactions due to the greater distance between two equivalent magnetic ions, which vanishes the Fe–O–Fe superexchange interactions, is confirmed.

The structural properties of LiFePO4 prepared by the hydrothermal route and chemically delithiated have been studied using analytical electron microscopy and Raman spectroscopy. High-resolution transmission electron microscopy and selected area electron diffraction measurements indicate that the partially delithiated particles include LiFePO4 regions with cross-sections of finite size along the ac-plane, as a result of tilt grain boundary in the bc-plane, and dislocations in other directions. Only the boundary along the bc-plane is accompanied by a disorder over about 2nm on each side of the boundary. The Raman spectrum shows the existence of both LiFePO4 and FePO4 phases in the shell of the particles at a delithiation degree of 50%, which invalidates the core–shell model. This result also invalidates the recent model according to which each particle would be single-domain, i.e. either a LiFePO4 particle or a FePO4 particle. On the other hand, our results, like prior ones, can be understood within the framework of a model similar to the spinodal decomposition of a two-phase system, which is discussed within the framework of morphogenesis of patterns in systems at equilibrium. Both end-members, however, are well crystallized, suggesting a recovery similar to that observed in superplastic alloys, with dynamics that are due to the motion of nucleation fronts and dislocations, and not due to a diffusion phenomenon associated with a concentration gradient.

A series of LiMnyFe1−yPO4 samples have been prepared in the whole range 0≤y≤1. Chemical delithiation could be achieved to obtain MnyFe1−yPO4 in the range 0≤y≤0.8, keeping the same crystal phase (olivine structure, space group Pnma). The composition y=0.8 is the limit where the delithiated phase is still crystallized, but abruptly suffers strains at the molecular scale evidenced by both optical spectroscopy and X-ray diffraction. The analysis of the magnetic properties shows that in all the samples the concentration of impurities is negligible. The concentration of polarons, either holes associated to Li vacancies in LiMnyFe1−yPO4 or electrons associated to the existence of Li left in the matrix of MnyFe1−yPO4, is found to be small (≤1%) in all the samples. For y≤0.6, all the Mn3+ ions MnyFe1−yPO4 are in the high-spin state (S=2). At larger manganese concentration, however, the Mn3+ ions in excess of the critical concentration yc=0.6 undergo a transition to the low-spin state (S=1). As a consequence, and in contrast with prior works, we find that Mn0.8Fe0.2PO4 has magnetic interactions that are much smaller, and no antiferromagnetic ordering in this compound is detected, at least above 20K. Antiferromagnetic ordering that had been reported so far for MnyFe1−yPO4 at large y-composition might come from incomplete delithiation. The spin-transition of Mn3+ in concentration (y–yc) to the low-spin state is at the origin of the strain fields at the molecular scale that increase with y for y>0.6, and ultimately prevents the full delithiation for y>0.8. This result sheds light on the reason for the degradation of cathode properties in Mn-rich compounds of the heterosite–purpurite series, while the electrochemical properties are good in the range y≤0.6 but only at slow rates, due to the very small hopping mobility of the small polaron.

Deviation from ideal stoichiometry of LiFePO4 has been investigated. Any attempt to increase the Li concentration of samples prepared either by the precursor precipitation route or by the continuous aqueous precursor synthesis route results in the formation of lithium phosphate impurity, in addition to stoichiometric LiFePO4 free of any Li vacancy. On the other hand, Li-deficient homogeneous solid solutions of composition Li1-2xFexFePO4 could be obtained. For x >= 0.06, however, a sarcopside impurity phase is formed. Investigations of structural properties allow us to define the defect responsible for the solid solution as Fe-Li(center dot) + V-Li' in the Kroger-Vink notation. Because the chemical formula of the sarcopside is obtained by writing x = 1/2 in the chemical formula of the solid solution, this impurity phase can be viewed as a condensation of the Fe-Li(center dot) + V-Li' defects. Magnetic measurements show that isolated lithium vacancies V-Li' are also diluted in the Li1-2xFexFePO4 matrix. The negative charge of the isolated V-Li'is compensated by the valence change Fe2+ -> Fe3+ of an iron ion in its vicinity, forming a small magnetic polaron that is detected by magnetic measurements. The concentration of such polarons, however, remains very small as it saturates to a concentration of 0.2-0.3 mol%, much smaller than the concentration x in V'(Li) bound to Fe-Li(center dot). The electrochemical features are significantly damaged by the Fe-Li(center dot) defects that block the diffusion of lithium along the corresponding channel, while the Li3PO4 only acts as an inert mass.

The spinet of nominal composition Li4Mn5O12 has been prepared by the wet chemistry technique and characterized by structural analysis (X-ray diffractometry, Raman spectroscopy), magnetic properties including electron spin resonance (ESR), and thermal properties (specific heat). Both structural and magnetic analyses reveal that the physical and electrochemical properties are importantly influenced by the presence of Li2MnO3 impurity phase. The presence of this additional phase reduces the average oxidation state of manganese according to a disproportionation reaction so that the final composition of the spinet phase is Li3.8Mn5.2O12 (or Li1.27Mn1.73O4). Because of the important geometric frustration of the magnetic interactions and dilution of the antiferromagnetic interactions, no magnetic ordering is observed in the temperature range investigated. The anomalous magnetic properties, including the Dysonian profile of the ESR line, show that the material is metallic. The Sommerfeld constant is 308 mJ/(K-2) per mole of Li1.27Mn1.73O4, which shows that this material belongs to the class of heavy-fermion systems like LiV2O4 or LiTi2O4. These heavy fermions are the minority-spin t(g)down arrow electrons of Mn3+ ions that have an reduced effective masse m/m(0) = 467. The electrochemical properties show that the specific capacity of Li3.8Mn5.2O12 at charge rate 1C is 163 mA h/g, a large value that is possibility due to the to insert Li up to the composition Li6.8Mn5.2O12. The origin of the disproportionation, and the flat voltage in the lithiation process, are discussed in the framework of the stability of the Mott insulator phase with respect to the metallic phase.

A novel lithium iron phosphate has been synthesized by a solution route at moderate temperature. The structure was determined from powder by XRD, HRTEM and SAED experiments. LiFe2P3O10 tripolyphosphate crystallizes in the monoclinic system, space group P21/m, with lattice constants a=4.597(7)Å, b=8.566(4)Å, c=9.051(4)Å, β=97.47° and Z=2. Internal and external vibrational modes (Raman and FTIR) show that the dominant spectral features come from the (P3O10)5− oxo-anions displaying internal and external modes along with the P–O–P bridging modes. Magnetic measurements are consistent with the high-spin configuration of Fe2+ cation with an effective magnetic moment 5.51μB. A weak antiferromagnetic ordering is observed below the Néel temperature at TN=19K. Electron paramagnetic resonance spectroscopy confirms this electronic configuration and provides evidence of the presence of a carbonaceous layer onto the particle surface. Electrochemical measurements were carried out in lithium cells with LiPF6–EC–DEC electrolyte at 25°C. The material delivered a capacity 70mAh/g in the voltage range 2.7–3.9V, close to the theoretical value (72mAh/g). The resulting cyclic voltammogram indicates a stable structure with a good reversibility with the redox peaks at 3.26 and 3.13V vs. Li0/Li+.

Amorphous analogs of lithium-iron-phosphates (LFP) with olivine-like local order were prepared by standard press-quenching method and their thermal stabilities as well as structural and electric properties were studied for the first time. Glass transition temperature T, determined by DTA method is equal to 492-523 degrees C depending on the glass composition. Fourier transform infrared (FTIR) absorption spectroscopy was used to study local order. Local structure and thermal properties are almost independent on the lithium content. Electrical properties were studied by impedance spectroscopy. Impedance measurements indicated that the studied lithium-iron-phosphate glasses with olivine-like local order exhibit predominantly electronic (polaronic) conductivity. Total electrical conductivity at 450 degrees C for a glass of nominal composition LiFePO4 approaches 10(-2) S/cm. Electrical conductivity at 25 degrees C of Li0.4FePO4 glass after nanocrystallization (caused by heating up from room temperature to the beginning of crystallization temperature) increased by a factor close to 4. Nanocrystallization seems to be the way for electrical conductivity improvement of olivine-like cathode materials. (c) 2008 Elsevier B.V. All rights reserved.

Currently, LiFePO4 is considered a good candidate as a positive electrode in rechargeable lithium-ion batteries for hybrid-electric vehicle (HEV) applications. A major difficulty with the preparation of the olivine structure (LiFePO4) comes from the existence of two oxidation states of iron, namely Fe(H) and Fe (111). Careful control of the synthesis procedure is needed to avoid any impurity which can poison the electrochemistry of LiFePO4 electrodes in Li-ion batteries. Analysis of the structure and morphology of the phospho-olivine LiFCPO4 is presented as a function of the synthetic conditions. The combination of analytical methods, i.e. magnetization, M(H), susceptibility, X.(T), and Raman scattering spectroscopy, appears to be a powerful tool for the detection of small amount of impurities. The electrochemical performance of optimized LiFePO4 is evaluated in Li4Ti5O12/LiPF6-EC-DEC/C-LiFePO4 cells operating at high temperature (60 degrees C). (c) 2007 Elsevier B.V. All rights reserved.

Li1+xMn2-xO4 spinels have been prepared by wet route technique in the whole concentration range 0⩽x⩽13. Structural analysis has been performed by X-ray diffraction and scanning electron microscopy and Raman spectroscopy. The investigation of impurity phases by magnetic experiments complete the sample characterization, allowing for a comparison with samples obtained by other chemical routes. The analysis of the intrinsic magnetic properties show that the magnetic interactions are Mn3+–O–Mn3+ interactions in a geometrically frustrated lattice, while the Mn4+–Mn4+ interactions are negligible, with the consequence that the paramagnetic Curie temperature vanishes in the limit x≃13 where only Mn4+ ions remain. A consistent description of both the extrinsic and the intrinsic properties including former investigations of electron paramagnetic resonance has been achieved in relation to the electrochemical properties.

The small polarons in LiFePO4 are associated with the presence of Fe3+ ions introduced by the native defects in relative concentration \({{\left[ {{\text{Fe}}^{3 + } } \right]} \mathord{\left/ {\vphantom {{\left[ {{\text{Fe}}^{3 + } } \right]} {\left[ {{\text{Fe}}^{2 + } + {\text{Fe}}^{3 + } } \right]}}} \right. \kern-\nulldelimiterspace} {\left[ {{\text{Fe}}^{2 + } + {\text{Fe}}^{3 + } } \right]}} = 3 \times 10^{ - 3} \) in the samples known to be optimized with respect to their electrochemical properties. The nearest iron neighbours around the central polaron site are spin-polarized by the indirect exchange mediated by the electronic charge in excess. These small magnetic polarons are responsible for the interplay between electronic and magnetic properties that are quantitatively and self-consistently analysed. Comparison is made with other magnetic polaron effects in other members of the family of magnetic semiconductors to which this material belongs.

A series of LiNi x Mn y Co z O2 (x = y, z = 1 − 2y) oxides have been synthesized by “chimie douce” and investigated as positive electrodes in rechargeable lithium batteries. Layered LiNi y Mn y Co1 − 2y O2 materials with high homogeneity and crystallinity were synthesized using the wet-chemical method assisted by carboxylic acid as the polymeric agent. The long range and local structural properties are investigated with experiments including X-ray diffraction, Fourier transform infrared spectroscopy, and electron paramagnetic resonance spectroscopy. The evolution of the structure is discussed as a function of the cobalt content that confers layer-like behavior on the framework. Electrochemical performance of LiNi y Mn y Co1 − 2y O2 oxides is tested in cells using nonaqueous 1 M LiPF6 dissolved in ethylene carbonate–diethyl carbonate. Charge–discharge profiles are investigated as a function of the rate capability and the voltage window. A relation is found between the gravimetric capacity and the cation disorder of the positive electrode as indicated by structural analysis. Fast lithium extraction attributed to the larger interslab space has been observed in the cobalt-rich oxides.