An in situ Raman study of the intercalation of supercapacitor-type electrolyte into microcrystalline graphite

An initial Raman study on the effects of intercalation for aprotic electrolyte-based electrochemical double-layer capacitors (EDLCs) is reported. In situ Raman microscopy is employed in the study of the electrochemical intercalation of tetraethylammonium (Et₄N⁺) and tetrafluoroborate (BF₄⁻) into and out of microcrystalline graphite. During cyclic voltammetry experiments, the insertion of Et₄N⁺ into graphite for the negative electrode occurs at an onset potential of +1.0 V versus Li/Li⁺. For the positive electrode, BF₄⁻ was shown to intercalate above +4.3 V versus Li/Li⁺. The characteristic G-band doublet peak (E_2g2(i) (1578 cm⁻¹) and E_2g2(b) (1600 cm⁻¹)) showed that various staged compounds were formed in both cases and the return of the single G-band (1578 cm⁻¹) demonstrates that intercalation was fully reversible. The disappearance of the D-band (1329 cm⁻¹) in intercalated graphite is also noted and when the intercalant is removed a more intense D-band reappears, indicating possible lattice damage. For cation intercalation, such irreversible changes of the graphite structure are confirmed by scanning electron microscopy (SEM).     

Preparation and characterization of ramsdellite Li2Ti3O7 as an anode material for asymmetric supercapacitors

Ramsdellite Li₂Ti₃O₇ was first synthesized via sol-gel process with good crystallity of an average particle size of 0.175 μm. The product was thoroughly investigated as a lithium intercalation compound, and as an active anode material in asymmetric supercapacitors coupling with activated carbon as cathode. Lithium intercalation reactions were found occurring at 1.32 and 1.62 V versus Li/Li⁺, respectively. A reversible specific capacity of 150 mA h g⁻¹ at 1C was obtained on Li₂Ti₃O₇ electrode in a nonaqueous electrolyte. The charge current was found to strongly influence the anodic discharge capacity in the asymmetric cell. The capacity retention at 10C charge-discharge rate was found to be 75.9% in comparison with that at 1C.     

Electrochemical lithium insertion into anatase-type TiO2: An in situ Raman microscopy investigation

Anatase type TiO₂ has been previously largely reported as a candidate negative electrode material for lithium-ion batteries. We report here for the first time the complete in situ Raman study of lithium insertion and de-insertion into three variously nano-sized TiO₂ anatase powders (Prolabo, ca. 80 nm, AK1, ca. 15 nm and MTi5 ca. 8 nm), of which AK1 and MTi5 show superior capacity and cyclability. From these measurements realized in a galvanostatic mode between 3 and 1 V versus Li/Li⁺, the phase transition from a tetragonal to an orthorhombic structure was clearly observed to take place at different quantities of x in Li_xTiO₂. These results confirm the extension of the solid solution domain as particle size is reduced. For the smaller TiO₂ nano-sized materials (AK1 and MTi5), a more pronounced decrease in band intensity when x > 0.3 for Li_xTiO₂, was observed and may be related to the decrease in the optical skin depth linked to the conductivity increase as lithiation proceeds.     

Preparation of micro-dot electrodes of LiCoO2 and Li4Ti5O12 for lithium micro-batteries

Fabrications of micro-dot electrodes of LiCoO₂ and Li₄Ti₅O₁₂ on Au substrates were demonstrated using a sol-gel process combined with a micro-injection technology. A typical size of prepared dots was about 100 μm in diameter, and the dot population on the substrate was 2400 dots cm⁻². The prepared LiCoO₂ and Li₄Ti₅O₁₂ micro-dot electrodes were characterized with scanning electron microscopy, X-ray diffraction, micro-Raman spectroscopy, and cyclic voltammetry. The prepared LiCoO₂ and Li₄Ti₅O₁₂ micro-dot electrodes were evaluated in an organic electrolyte as cathode and anode for lithium micro-battery, respectively. The LiCoO₂ micro-dot electrode exhibited reversible electrochemical behavior in a potential range from 3.8 to 4.2 V versus Li/Li⁺, and the Li₄Ti₅O₁₂ micro-dot electrode showed sharp redox peaks at 1.5 V.     

Electrochemical and ex situ XRD investigations on (1−x)LiNiO2·xLi2TiO3 (0.05≤x≤0.5)

An attempt to understand the unusual electrochemical behaviors in (1−x)LiNiO₂·xLi₂TiO₃ (0.05≤x≤0.5), an excess initial charge capacity exceeding the oxidation of transitional metal to +4 accompanying the appearance of an irreversible initial charge plateau when x reached 0.075, was performed. The decreased charge-discharge polarization after charging to 4.6 and 4.8 V and increased columbic reversibility after charging to 4.6 V typically for x=0.1 and 0.2, in contrast to charging to 4.4 V, suggested that the excess initial charge capacity possibly did not come mainly from electrolyte decomposition; while ex situ XRD results in the sample with x=0.2 confirmed that Li⁺ were really extracted at the stage of the charge plateau, ruling out the possibility that electrolyte decomposition mainly accounted for the unusual electrochemical behaviors. It was inferred that the species responsible for charge compensation for the excess charge capacity must be oxygen ions in these materials, considering that Ni⁴⁺ and Ti⁴⁺ are generally impossible to be oxidized to a higher valence. Various electrochemical cycling experiments demonstrated that the sample for x=0.05 with high resistant ability to high voltage and temperature is very promising cathode material in view of observed capacity and cycleability from a viewpoint of application.     

Influence of manganese(II), cobalt(II), and nickel(II) additives in electrolyte on performance of graphite anode for lithium-ion batteries

Manganese dissolution into an electrolyte from the spinel LiMn₂O₄ in the lithium-ion cell has been recently investigated. In order to study the influence of the dissolved manganese species on the lithium intercalation/deintercalation into a natural graphite electrode, the electrochemical behavior of graphite was investigated in 1 mol dm⁻³ LiClO₄ electrolyte solution containing a small amount of Mn(II) by the addition of manganese(II) perchlorate. During the charging process, Mn(II) ions were firstly electroreduced on the electrode around 1.0 V versus Li/Li⁺ followed by irreversible decomposition of the electrolyte and lithium intercalation into the graphite. By microscopic observation of the graphite surface, manganese deposition was confirmed after the charge/discharge test. Due to the manganese deposition, the reversible capacity of the graphite electrode was drastically decreased. Furthermore, the cyclability of the anode was degraded with the amount of the manganese additive increasing. We compared these results with those of the cobalt(II) and nickel(II) additives by dissolving the corresponding perchlorates. Furthermore, we discussed the influence in practical cells based on the consideration of electrochemistry of the deposited metals.     

Electrochemical and thermal studies of Li[NixLi(1/3−2x/3)Mn(2/3−x/3)]O2 (x = 1/12, 1/4, 5/12, and 1/2)

Samples of the layered cathode materials, Li[Ni_xLi_(1/3−2x/3)Mn_(2/3−x/3)]O₂ (x = 1/12, 1/4, 5/12, and 1/2), were synthesized at 900 °C. Electrodes of these samples were charged in Li-ion coin cells to remove lithium. The charged electrode materials were rinsed to remove the electrolyte salt and then added, along with EC/DEC solvent or 1 M LiPF₆ EC/DEC, to stainless steel accelerating rate calorimetry (ARC) sample holders that were then welded closed. The reactivity of the samples with electrolyte was probed at two states of charge. First, for samples charged to near 4.45 V and second, for samples charged to 4.8 V, corresponding to removal of all mobile lithium from the samples and also concomitant release of oxygen in a plateau near 4.5 V. Li[Ni_xLi_(1/3−2x/3)Mn_(2/3−x/3)]O₂ samples with x = 1/4, 5/12 and 1/2 charged to 4.45 V do not react appreciably till 190 °C in EC/DEC. Li[Ni_xLi_(1/3−2x/3)Mn_(2/3−x/3)]O₂ samples charged to 4.8 V versus Li, across the oxygen release plateau, start to significantly react with EC/DEC at about 130 °C. However, their high reactivity is similar to that of Li_0.5CoO₂ (4.2 V) with 1 μm particle size. Therefore, Li[Ni_xLi_(1/3−2x/3)Mn_(2/3−x/3)]O₂ samples showing specific capacity of up to 225 mAh/g may be acceptable for replacing LiCoO₂ (145 mAh/g to 4.2 V) from a safety point of view, if their particle size is increased.     

Cycling-induced stress in lithium ion negative electrodes: LiAl/LiFePO4 and Li4Ti5O12/LiFePO4 cells

In this work, we examined the electrochemical behaviour of lithium ion batteries containing lithium iron phosphate as the positive electrode and systems based on Li-Al or Li-Ti-O as the negative electrode. These two systems differ in their potential versus the redox couple Li⁺/Li and in their morphological changes upon lithium insertion/deinsertion. Under relatively slow charge/discharge regimes, the lithium-aluminium alloys were found to deliver energies as high as 438 Wh kg⁻¹ but could withstand only a few cycles before crumbling, which precludes their use as negative electrodes. Negative electrodes consisting solely of aluminium performed even worse. However, an electrode made from a material with zero-strain associated to lithium introduction/removal such as a lithium titanate spinel exhibited good performance that was slightly dependent on the current rate used. The Li₄Ti₅O₁₂/LiFePO₄ cell provided capacities as high as 150 mAh g⁻¹ under C-rate in the 100th cycle.     

An investigation of intercalation-induced stresses generated during lithium transport through sol-gel derived LixMn2O4 film electrode using a laser beam deflection method

The stress changes Δσ generated during lithium transport through the sol-gel derived Li_xMn₂O₄ film electrodes annealed at 773 and 873 K were quantitatively determined as a function of the lithium stoichiometry x using a laser beam deflection method (LBDM). Δσ generated during a real potential step between an initial electrode potential and a final applied potential was uniquely specified by the Δσ versus x curve. The Li_xMn₂O₄ film annealed at 773 K for 24 h (low temperature (LT)-Li_xMn₂O₄) showed larger capacity than the Li_xMn₂O₄ film annealed at 873 K for 6 h (high temperature (HT)-Li_xMn₂O₄) and this result is ascribed to the fact that the smaller the grain size is, the more increases the electrochemically active area of the film electrode. From the analysis of the normalised Δσ transient measured simultaneously along with the cyclic voltammogram in the potential range of 2.5-3.4 VLi/Li⁺, it is found that normalised Δσ generated in the LT-Li_xMn₂O₄ was smaller than that in the HT-Li_xMn₂O₄ during the lithium intercalation/de-intercalation around 3.0 VLi/Li⁺ region. This result gives an experimental evidence for the fact that the Jahn-Teller distortion is suppressed by the increase in the average oxidation state of manganese with decreasing in annealing temperature.     

In situ X-ray diffraction study of different graphites in a propylene carbonate based electrolyte at very positive potentials

Synchrotron based in situ X-ray powder diffraction was applied for the investigation of the PF₆⁻ intercalation behaviour into graphite of different particle size and various crystallinity at very positive potentials above 5 V vs. Li/Li⁺. In a propylene carbonate based electrolyte, small particles with low crystallinity show almost no anion intercalation and the particle structure remains intact. In graphite particles with increased particle size and/or higher crystallinity PF₆⁻ intercalation up to a stage 4 phase occurs but the reversibility is poor and exfoliation of graphene layers takes place as it was proven by post-mortem scanning electron microscopy. Thus, graphites with small particles and less crystallinity are preferred as conductive additive in positive “high voltage” cathodes in terms of structural integrity.     

Electrochemical behavior of LiCoO2 in a saturated aqueous Li2SO4 solution

The electrochemical behavior of a commercial LiCoO₂ with spherical shape in a saturated Li₂SO₄ aqueous solution was investigated with cyclic voltammetry and electrochemical impedance spectroscopy. Three redox couples at E_SCE = 0.87/0.71, 0.95/0.90 and 1.06/1.01 V corresponding to those found at ELi/Li⁺=4.08/3.83, 4.13/4.03 and 4.21/4.14 V in organic electrolyte solutions were observed. The diffusion coefficient of lithium ions is 1.649 × 10⁻¹⁰ cm² s⁻¹, close to the value in organic electrolyte solutions. The results indicate that the intercalation and deintercalation behavior of lithium ions in the Li₂SO₄ solution is similar to that in the organic electrolyte solutions. However, due to the higher ionic conductivity of the aqueous solution, current response and reversibility of redox behavior in the aqueous solution are better than in the organic electrolyte solutions, suggesting that the aqueous solution is favorable for high rate capability. The charge transfer resistance, the exchange current and the capacitance of the double layer vary with the charge voltage during the deintercalation process. At the peak of the oxidation (0.87 V), the charge transfer resistance is the lowest. These fundamental results provide a good base for exploring new safe power sources for large scale energy storage.     

Li diffusion in LiNi0.5Mn0.5O2 thin film electrodes prepared by pulsed laser deposition

Kinetic and transport parameters of Li ion during its extraction/insertion into thin film LiNi_0.5Mn_0.5O₂ free of binder and conductive additive were provided in this work. LiNi_0.5Mn_0.5O₂ thin film electrodes were grown on Au substrates by pulsed laser deposition (PLD) and post-annealed. The annealed films exhibit a pure layered phase with a high degree of crystallinity. Surface morphology and thin film thickness were investigated by field emission scanning electron microscopy (FESEM). The charge/discharge behavior and rate capability of the thin film electrodes were investigated on Li/LiNi_0.5Mn_0.5O₂ cells at different current densities. The kinetics of Li diffusion in these thin film electrodes were investigated by cyclic voltammetry (CV) and galvanostatic intermittent titration technique (GITT). CV was measured between 2.5 and 4.5 V at different scan rates from 0.1 to 2 mV/s. The apparent chemical diffusion coefficients of Li in the thin film electrode were calculated to be 3.13 × 10⁻¹³ cm²/s for Li intercalation and 7.44 × 10⁻¹⁴ cm²/s for Li deintercalation. The chemical diffusion coefficients of Li in the thin film electrode were determined to be in the range of 10⁻¹²-10⁻¹⁶ cm²/s at different cell potentials by GITT. It is found that the Li diffusivity is highly dependent on the cell potential.     

Insignificant impact of designed oxygen release from high capacity Li[(Ni1/2Mn1/2)xCoy(Li1/3Mn2/3)1/3]O2 (x + y = 2/3) positive electrodes during the cycling of Li-ion cells

The properties of graphite/Li[(Ni_0.5Mn_0.5)_xCo_y(Li_1/3Mn_2/3)_1/3]O₂ (x + y = 2/3, y = 1/12 and 1/6) Li-ion cells are reported. There is an extended plateau near 4.5 V during the first charging of the cells that corresponds to the simultaneous removal of Li and oxygen from the Li[(Ni_0.5Mn_0.5)_xCo_y(Li_1/3Mn_2/3)_1/3]O₂ (x + y = 2/3, y = 1/12 and 1/6) electrodes. The release of this oxygen directly within a Li-ion cell has been a cause for concern. However, it was found that subsequent to O₂ release, Li-ion cells delivered a high reversible positive electrode specific capacity near 250 mAh/g at C/30 between 2.5 and 4.8 V, the cells did not display increased irreversible capacity relative to counterparts having Li metal negative electrodes and the cells retained 85% of their initial capacity after 70 cycles at C/6 between 2.5 and 4.6 V. Therefore, the O₂ released during the first charge does not significantly impact the electrochemical properties of graphite/Li[(Ni_0.5Mn_0.5)_xCo_y(Li_1/3Mn_2/3)_1/3]O₂ (x + y = 2/3) lithium-ion cells.     

Impedance study of protecting film formation on lithium and lithiated graphite induced by bis(fluorosulfonyl) imide anion

The process of Li⁺ reduction from room temperature ionic liquids consisting of N-methyl-N-propylpyrrolidinium cation (MPPyr⁺) and bis(fluorosulfonyl) imide (FSI⁻) or bis(trifluoromethanesulfonyl) imide (TFSI⁻) anions was studied with the use of impedance spectroscopy. Reduction was carried out on both metallic lithium (Li) and graphite (G) electrodes. It has been found that the FSI anion in high amounts is able to form a protective film on both graphite and metallic lithium. The Li⁺/Li couple should rather be represented by a Li⁺/SEI/Li system. The SEI structure depends on the manner of its formation (chemical or electrochemical) and is not stable with time. The rate constant for the Li⁺ + e⁻ → Li process at the Li/SEI/Li⁺ (in MPPyrFSI) interface is k_o = 4.2 × 10⁻⁵ cm/s. In the case of carbon electrodes (G/SEI/Li⁺ interface), lithium diffusion in solid graphite is the rate determining step, reducing current by ca. two orders of magnitude, from ca. 10⁻⁴ A/cm², characteristic of the Li/SEI/Li⁺ electrode, to ca. 10⁻⁶ A/cm².     

Enhanced inactivation of E. coli bacteria using immobilized porous TiO2 photoelectrocatalysis

Immobilized TiO₂ nanotube electrodes with high surface areas were grown via electrochemical anodization in aqueous solution containing fluoride ions for photocatalysis applications. The photoelectrochemical properties of the grown immobilized TiO₂ film were studied by potentiodynamic measurements (linear sweep voltammetry), in addition to the calculation of the photocurrent response. The nanotube electrode properties were compared to mesoporous TiO₂ electrodes grown by anodization in sulfuric acid at high potentials (above the microsparking potential) and to 1 g/l P-25 TiO₂ powder. Photocatalyst films were evaluated by high resolution SEM and XRD for surface and crystallographic characterization. Finally, photoelectrocatalytic application of TiO₂ was studied via inactivation of E. coli. The use of the high surface area TiO₂ nanotubes resulted in a high photocurrent and an extremely rapid E. coli inactivation rate of ∼10⁶ CFU/ml bacteria within 10 min. The immobilized nanotube system is proven to be the most potent electrode for water purification.     

A novel lithium intercalation compound based on the layered structure of lithium nitridonickelates Li3−2xNixN

New lithium nickel nitrides Li_3−2xNi_xN (0.20 ≤ x ≤ 0.60) have been prepared and investigated as negative electrode in the 0.85/0.02 V potential window. These materials are prepared from a Ni/Li₃N mixture at 700 °C under a nitrogen flow. Their structural characteristics as well as their electrochemical behaviour are investigated as a function of the nickel content. For the first time are reported here the electrochemical properties of a lithium intercalation compound based on a layered nitride structure. The Li_3−2xNi_xN compounds can be reversibly reduced and oxidized around 0.5 V versus Li/Li⁺ leading to specific capacities in the range 120-160 mAh/g depending on the nickel content and the C rate. Due to a large number of lithium vacancies, the structural stability provides an excellent capacity retention of the specific capacity upon cycling.     

Electrochemical characterization of amorphous LiFe(WO4)2 thin films as positive electrodes for rechargeable lithium batteries

Amorphous LiFe(WO₄)₂ thin films have been fabricated by radio-frequency (R.F.) sputtering deposition at room temperature. The as-deposited and electrochemically cycled thin films are, respectively, characterized by X-ray diffraction, scanning electron microscopy, transmission electron microscopy, selected area electron diffraction, and X-ray photoelectron spectra techniques. An initial discharge capacity of 198 mAh/g in Li/LiFe(WO₄)₂ cells is obtained, and the electrochemical behavior is mostly preserved in the following cycling. These results identified the electrochemical reactivity of two redox couples, Fe³⁺/Fe²⁺ and W⁶⁺/W^x+ (x = 4 or 5). The kinetic parameters and chemical diffusion coefficients of Li intercalation/deintercalation are estimated by cyclic voltammetry and alternate-current (AC) impedance measurements. All-solid-state thin film lithium batteries with Li/LiPON/LiFe(WO₄)₂ layers are fabricated and show high capacity of 104 μAh/cm² μm in the first discharge. As-deposited LiFe(WO₄)₂ thin film is expected to be a promising positive electrode material for future rechargeable thin film batteries due to its large volumetric rate capacity, low-temperature fabrication and good electrode/electrolyte interface.     

Ionic liquids in electrochromic devices

The ionic liquid (PYR₁₄TFSI) has proved to be the key material to make a Li-ion conducting element of a complete electrochromic device, when interposed between transparent film electrodes like WO₃ and Li-charged V₂O₅. The key features of this ionic liquid and its mixtures with LiTFSI are the excellent transparency in the visible and NIR optical regions, the good ionic conductivity and the electrochemical compatibility with inorganic Li-intercalation oxide thin film electrodes used in electrochromic devices. The higher optical contrast found during WO₃ colouration with PYR₁₄TFSI-LiTFSI, compared to that in a conventional non-aqueous electrolyte like PC-LiTFSI, was attributed to the larger inertness of the former one (no decomposition reaction at the lowest electrode potential). This highly conductive ionic liquid has been incorporated into a polymer matrix (P(EO)₁₀LiTFSI), in order to obtain a transparent solid electrolyte with high Li ion conductivity and good mechanical stability. Finally this solid PYR₁₄TFSI-P(EO)₁₀LiTFSI transparent ion conductor was interposed between the same electrodes as above in order to yield a fully solid-state, Li-ion electrochromic device. This new solid electrolyte was able to transfer reversibly a Li ionic charge between 5 mC cm⁻² and 10 mC cm⁻² from the lithium storage electrode Li_xV₂O₅ to the WO₃ electrochromic electrode in less than 100 s at room T, darkening the device from an initial 80% to a final 30% transmittance (at 650 nm). Such a device has been tested first under various constant current conditions, and later under potentiostatic control using ±2 V steps. The latter method allows not only for a faster response of the electrochromic system, but provides also an easier life stability test of the device, which withstood 2000 cycles with little changes in its optical contrast.     

Structural and electrochemical behaviors of metastable Li2/3[Ni1/3Mn2/3]O2 modified by metal element substitution

Layered metastable lithium manganese oxides, Li_2/3[Ni_1/3−xMn_2/3−yM_x+y]O₂ (x = y = 1/36 for M = Al, Co, and Fe and x = 2/36, y = 0 for M = Mg) were prepared by the ion exchange of Li for Na in P2-Na_2/3[Ni_1/3−xMn_2/3−yM_x+y]O₂ precursors. The Al and Co doping produced the T^#2 structure with the space group Cmca. On the other hand, the Fe and Mg doped samples had the O6 structure with space group R-3m. Electron diffraction revealed the 1:2 type ordering within the Ni_1/3−xMn_2/3−yM_x+yO₂ slab. It was found that the stacking sequence and electrochemical performance of the Li cells containing T^#2-Li_2/3[Ni_1/3Mn_2/3]O₂ were affected by the doping with small amounts of Al, Co, Fe, and Mg. The discharge capacity of the Al doped sample was around 200 mAh g⁻¹ in the voltage range between 2.0 and 4.7 V at the current density of 14.4 mA g⁻¹ along with a good capacity retention. Moreover, for the Al and Co doped and undoped oxides, the irreversible phase transition of the T^#2 into the O2 structure was observed during the initial lithium deintercalation.     

Synthesis of LiNi0.9Co0.1O2 from Li2CO3, NiO or NiCO3, and CoCO3 or Co3O4 and their electrochemical properties

Cathode active materials with a composition of LiNi_0.9Co_0.1O₂ were synthesized by a solid-state reaction method at 850 °C using Li₂CO₃, NiO or NiCO₃, and CoCO₃ or Co₃O₄, as the sources of Li, Ni, and Co, respectively. Electrochemical properties, structure, and microstructure of the synthesized LiNi_0.9Co_0.1O₂ samples were analyzed. The curves of voltage vs. x in Li_xNi_0.9Co_0.1O₂ for the first charge–discharge and the intercalated and deintercalated Li quantity Δx were studied. The destruction of unstable 3b sites and phase transitions were discussed from the first and second charge–discharge curves of voltage vs. x in Li_xNi_0.9Co_0.1O₂. The LiNi_0.9Co_0.1O₂ sample synthesized from Li₂CO₃, NiO, and Co₃O₄ had the largest first discharge capacity (151 mA h/g), with a discharge capacity deterioration rate of −0.8 mA h/g/cycle (that is, a discharge capacity increasing 0.8 mA h/g per cycle).