It is an old concept to fabricate lithium batteries in the discharged state with only an appropriate current collector as the negative electrode (anode). On the initial charge, metallic Li is electroplated at this anode current collector, and so, during electrochemical cycling, the battery operates as a Li battery which contains only the amount of lithium that is supplied by the positive electrode (cathode).Despite some improvements, 1 the manufacture of a practical, rechargeable Li battery that operates exclusively with an in situ plated Li anode has been prevented by the formation of mossy, dendritic, and granular metallic lithium deposits on metal foils. 2 In this paper, we report on the feasibility and electrochemical properties of a Cu/solid lithium electrolyte/LiCoO 2 thin-film battery, where Cu represents an anode current collector that does not form intermetallic compounds with lithium. Prior to the initial charge, all of the battery components are stable in air for several hours, which facilitates handling and processing of this "Li-free" lithium battery.During operation, this Li-free battery shows the maximum potential and high rate capability inherent in a Li battery but avoids the major drawbacks of a battery fabricated with a Li metal anode. The vapor deposition of a metallic lithium film is more complicated than the deposition of other metal films that are less air sensitive, and due to the low melting point of lithium (181ЊC) a Li battery does not survive the 250ЊC solder reflow process commonly used to assemble electronic circuit boards. 3 In this paper, we demonstrate that the Li free battery with an in situ plated Li anode shows no signs of degradation after being heated at 250ЊC in air for 10 min in the Li-free state. Furthermore, complete stripping of the electroplated Li anode is reversible, and since no excess Li is present, the Li-free battery cannot be destroyed by overdischarging the LiCoO 2 cathode, in contrast to a conventional Li-LiCoO 2 battery.The Li-free battery prior to operation resembles a Li-ion battery, which is also assembled in its fully discharged state. However, the Li-free battery avoids the capacity problem inherent to tin oxide-and oxynitride-based insertion materials 3-7 which have recently been proposed as anodes for Li-ion cells. These anodes are known to consume, irreversibly, between 40 and 60% of the lithium inserted during their electrochemical activation in the initial cycle due to the formation of an amorphous matrix containing Li 2 O or Li 3 N. [3][4][5]8 Despite the technically high capacity of these anodes, between 600 and 700 mAh/g, the unsatisfactory ratio of reversible to irreversible capacity severely limits the utilization of the cathode, which serves as the Li-ion cell's initial lithium source. Recently, we showed that the cathode utilization could be improved by fabricating cathodeheavy cells in which a good deal of the total reversible cell capacity was obtained by electrochemical plating and stripping of metallic lithium from the overlithiated anode ...

Polycrystalline films of LiCoO 2 deposited by radio frequency magnetron sputtering exhibited a strong preferred orientation or texturing after annealing at 700ЊC. For films thicker than about 1 m, more than 90% of the grains were oriented with their (101) and (104) planes parallel to the substrate and less than 10% with their (003) planes parallel to the substrate. As the film thickness decreased below 1 m, the percentage of (003)-oriented grains increased until at a thickness of about 0.05 m, 100% of the grains were (003) oriented. These extremes in texturing were caused by the tendency to minimize volume strain energy for the thicker films or the surface energy for the very thin films. Films were deposited using different process gas mixtures and pressures, deposition rates, substrate temperatures, and substrate bias. Of these variables, only changes in substrate temperature could cause large changes in texturing of thick films from predominately ( 101)-( 104) to (003). Although lithium ion diffusion should be much faster through cathodes with a high percentage of (101)-and (104)-oriented grains than through cathodes with predominately (003)-oriented grains, it was not possible to verify this expectation because the resistance of most cells was dominated by the electrolyte and electrolyte-cathode interface. Nonetheless, cells with cathodes thicker than about 2 m could deliver more than 50% of their maximum energies at discharge rates of 5 mA/cm 2 or higher.

Mass transport and thermodynamic properties of Li x CoO 2 were studied by the potentiostatic intermittent titration technique ͑PITT͒. We determined the chemical diffusion coefficient (D Li ) and the thermodynamic factor ͑⌰͒ of Li in the region 0.45 Ͻ x Ͻ 0.7, where Li x CoO 2 exists as a single phase having either a rhombohedral or a monoclinic structure. Solid-state thin-film batteries were used in order to ensure a well-defined diffusion geometry. Both D Li and ⌰ have minima at the phase boundaries of the Li vacancy ordered phase Li 0.5 CoO 2 . The self-diffusion coefficient of Li (D Li ) has a minimum at x ϭ 0.5 associated with the Li vacancy ordering. As the degree of ordering increases, nonmonotonic variations of D Li , ⌰, and D Li become more pronounced when approaching x ϭ 0.5 in Li x CoO 2 .High power applications of rechargeable Li batteries require fast Li ion mobility within intercalation compounds. The ␣-NaFeO 2 structure of LiCoO 2 ensures a high chemical diffusion coefficient of Li (D Li ) and in turn, a high rate capability. 1 The structure is based on a nearly cubic close-packed arrangement of O 2Ϫ with the Li ϩ and Co 3ϩ alternatively occupying the octahedral sites between adjacent oxygen ion layers ͑space group R3 m). 2 According to previous theoretical studies, Li diffusion in the Li layer occurs via tetrahedral sites by a divacancy mechanism. 3-5 The values of D Li in the literature vary from 10 Ϫ13 to 10 Ϫ7 cm 2 /s. 6-20 This large difference is attributed to different assumptions for the geometrical factors ͑diffu-sion length and cross-sectional surface area͒ used in the calculation of D Li . 17,18 In the case of conventional composite powder electrodes, for example, when the diffusion length is erroneously identified as the thickness of the electrode, the value of D Li will be overestimated particularly for thick electrodes. Single-crystal electrodes would provide a well-defined geometry, but they are not easily available. 19 In terms of both geometry and feasibility, thin-film electrodes combined with solid electrolytes are ideal for diffusion studies. Thin-film electrodes with liquid electrolytes were employed in previous diffusion studies; 10,14,18 however, because permeation of liquid electrolytes into cracks in thin-film cathodes cannot be ruled out, use of a solid electrolyte is preferrable.Extraction of Li from LiCoO 2 occurs via several phase transitions. In Li x CoO 2 , two rhombohedral phases coexist for 0.75 р x р 0.93, 21 which has been attributed to the insulator metal transition upon Li extraction. 22,23 D Li is not defined as a single value in the two-phase region. At compositions near Li 0.5 CoO 2 , Li ions are ordered in rows separated by rows of vacancies within the Li layers, 21 which is accompanied by a lattice distortion to a monoclinic symmetry ͑space group P2/m). 4,21,24 The degree of ordering is sensitive to the sample purity. 25 Recently, Van der Ven and Ceder predicted that D Li and the self-diffusion coefficient of Li (D Li ) have a minimum at x ϭ 0.5 due to Li or...

No abstract

No abstract