M. N. Obrovac

Structural Changes in Silicon Anodes during Lithium Insertion/Extraction

The structural changes in silicon electrochemically lithiated and delithiated at room temperature were studied by X-ray powder diffraction. Crystalline silicon becomes amorphous during lithium insertion, confirming previous studies. Highly lithiated amorphous silicon suddenly crystallizes at 50 mV to form a new lithium-silicon phase, identified as Li 1 5 Si 4 . This phase is the fully lithiated phase for silicon at room temperature, not Li 2 1 Si 5 as is widely believed. Delithiation of the Li 1 5 Si 4 phase results in the formation of amorphous silicon. Cycling silicon anodes above 50 mV avoids the formation of crystallized phases completely and results in better cycling performance.

Reversible Cycling of Crystalline Silicon Powder

A method is described in which crystalline silicon can be used as a practical anode material for lithium-ion batteries. Commercial lithium-ion cells are typically charged at a constant current to a fixed voltage and then are held by the charger at constant voltage until the current decreases to a certain value (also known as constant current/constant voltage or CCCV charging). It is first shown that CCCV charging can be used to reversibly cycle crystalline silicon and limit its capacity. A cycling method is then demonstrated in which crystalline silicon is first partially converted to amorphous silicon, in situ, during conditioning cycles. After the conditioning cycles the silicon can be cycled normally, using CCCV cycling limits, with good coulombic efficiency and little overlithiation during the first cycle.

Alloy Design for Lithium-Ion Battery Anodes

A set of guidelines is proposed for designing high-energy-density alloy anode materials. It is first shown that the molar volume of lithium is about 9 mL/mol in a wide variety of lithium alloys and is independent of lithium content. Using this property of lithium alloys, simple relationships between the volumetric energy density and the volumetric expansion of an alloy are derived. These relationships are extremely powerful for designing alloys with the maximum possible energy density for a given electrode-coating performance.

The Electrochemical Displacement Reaction of Lithium with Metal Oxides

The electrochemical reaction of lithium with a-LiFeO2, b-Li5FeO4 , and CoO is studied by in situ X-ray diffraction and in situ Mossbauer measurements. The results of the measurements show that these metal oxides are immediately decomposed during discharge to form lithia and the reduced metal. This reaction proceeds through a single intermediate or surface phase. The reaction products are nanometer-sized, but are not amorphous as was suggested previously. During charge the metal displaces the lithium in lithium oxide to form a metal oxide and lithium. In the case of CoO, the original lithium oxide oxygen lattice is preserved and the reaction resembles an ion exchange process. This also appears to be the case for the iron oxides. Upon discharge, the reverse occurs and the lithium replaces the metal in the metal oxide, once again forming lithium oxide and reduced metal on the bottom

A Mossbauer effect investigation of the Li-Sn system

Abstract A thorough Mossbauer effect investigation of all the phases in the Li–Sn binary phase diagram has been made. The compounds studied were verified to have the same crystal structures as those reported in the literature. The center shifts (relative to CaSnO 3 ), quadrupole splittings, number of Sn sites and their abundance have been determined for each of the phases by fitting model spectra to the observed data. The observed spectra show mean center shifts near 2.4 mm/s for the Sn-rich phases (Sn, Li 2 Sn 5 and LiSn) and mean center shifts between 2.1 and 1.8 mm/s for the Li-rich phases (Li 7 Sn 3 , Li 5 Sn 2 , Li 13 Sn 5 , Li 7 Sn 2 and Li 22 Sn 5 ). These results are important for the interpretation of Mossbauer effect measurements on high capacity Sn-based electrode materials for Li-ion batteries.

Structure and electrochemistry of LiMO2 (M=Ti, Mn, Fe, Co, Ni) prepared by mechanochemical synthesis

Abstract LiMO 2 (M=Ti, Mn and Fe) were prepared by ball milling mixtures of lithium transition metals and transition metal oxides. The resulting oxides were found to have the rocksalt structure with the transition metal and lithium ions randomly ordered in the cation sites, except for LiMnO 2 which had some short range cation order. LiCoO 2 and LiNiO 2 samples prepared at high temperature were also observed to transform to the disordered rocksalt structure after subsequent milling. However, during milling these samples lost lithia and oxygen to form Li x M 1− x O (0.25×0.5). The electrochemical performance of the milled samples as cathodes in lithium batteries was poor, presumably due to the disorder of the cation lattice. However, mechanochemical synthesis may prove to be a useful synthesis step in the preparation of novel active cathode materials.

Effect of Heat Treatment on Si Electrodes Using Polyvinylidene Fluoride Binder

The electrochemical performance of Si electrodes using polyvinylidene fluoride binder heated at different temperatures ranging from 150 to 350°C was investigated. Compared to unheated electrodes that have no capacity after the first formation cycle, the heat-treated electrodes show an increasingly improved cycling performance as the heating temperature increases from 150 to 300°C. In particular, Si electrodes heated at 300°C retain a specific capacity of ∼600 mAh/g for 50 cycles with a lower cutoff potential of 0.170 V vs Li/Li + . The reasons for such improvements are considered based on results from thermal analysis, optical microscopy, and adhesion tests. It is suggested that heat-treatment improves the binder distribution, the adhesion to the Si particles and to the substrate, thereby leading to a better cycling performance.

Electrochemically Active Lithia/Metal and Lithium Sulfide/Metal Composites

This paper demonstrates Li 2 O/M and Li 2 S/M (M = Co, Fe) composites that contain electrochemically active lithium. The composites were made by ballmilling Li 2 O or Li 2 S with metal powder. The resulting materials had average grain sizes of ∼100 A and enormous reversible capacities of up to 600 mAh/g. When cycled within appropriate voltage limits, the electrodes showed no capacity fading. These composites are compatible with lithium-ion technology and could provide a safe, high-energy density alternative to conventional intercalation electrodes.

A Comparison of the Reactions of the SiSn, SiAg, and SiZn Binary Systems with L3i

Alloys of Si with elements such as Sn, Ag, or Zn may exhibit large specific and volumetric capacities when cycled electrochemically vs lithium. Combinatorial films of Si 1 - x M x in the range 0 < x < ∼ 0.60 have been prepared by magnetron sputtering. where M is Sn, Ag, or Zn. These films have been analyzed for structure, composition, and electrochemical performance. The electrochemical performance of the film is found to be dependent on the element M as well as the amount, x, of the element in the film. The three Si-based binary systems are compared in terms of phases formed, amorphous vs crystalline structure, and capacity retention. It is found that electrodes which remain as a single amorphous phase during insertion and removal of Li exhibit superior capacity retention as compared to electrodes that form multiple or crystalline phases.

Al-M ( M = Cr , Fe , Mn , Ni ) Thin-Film Negative Electrode Materials

Comprehensive investigations of thin-film Al-M (M = Cr, Fe, Mn, Ni) negative electrode material libraries were performed using combinatorial and high-throughput techniques. Mossbauer effect spectroscopy and X-ray diffraction were used to characterize library structure. The electrochemical performance of over 200 compositions of Al x M 1-x (0.75 < x < 1) was determined at both room and elevated temperature. All Al x M 1-x alloys are completely inactive above 15 atom % M. Thermodynamic calculations based on the macroscopic atom model of de Boer et al. predict compositions with at least 33-65 atom % M to be inactive. A phenomenological model to describe the observed capacity as a function of transition metal content and cycling temperature is presented.