Lisa C. Klein

CoO[sub 2], The End Member of the Li[sub x]CoO[sub 2] Solid Solution

While has been widely studied in the past 15 years as a promising positive electrode material in lithium‐ion batteries, suprisingly, many questions are still unanswered concerning the electrochemical characteristics of the lithium intercalation material. Among these is the existence of an end member phase on complete lithium deintercalation. The use of dry plastic lithium‐ion battery technology has allowed the construction of an in situ x‐ray diffraction cell which allows structural characterization of at x values at and close to 0 for the first time. Instead of the expected destruction of the core structure of by a drastic increase in structural disorder, an increase in crystallographic quality occurred as x approached 0. For the first time, the end member phase was isolated. This phase is a hexagonal single‐layered phase (O1) believed to be isostructural with and has lattice parameters of a = 2.822 A and c = 4.29 A. The phase converted immediately back to a three‐layer (O3) delithiated type phase on lithium reinsertion. Electrochemical studies show that 95% of lithium can be reinserted back into the structure on complete delithiation and reversible cycling properties are maintained when cycled back to 4.2 V.

Cobalt dissolution in LiCoO2-based non-aqueous rechargeable batteries

Abstract Rechargeable lithium batteries using a variety of differently prepared LiCoO 2 samples were cycled to 4.5 V versus Li and the negative electrode was investigated for Co deposits after the cell was cycled over 25 times. Qualitative analysis revealed evidence of cobalt dissolution for samples showing high capacity loss with cycle number. A direct correlation between the percentage capacity loss and percentage cobalt loss was established. Cycling LiCoO 2 -based cells to voltages ranging from 4.1 to 4.5 was found to result in an increase in the capacity fade for voltages greater than 4.2 V, consistent with the traditional limit of reversibility of 4.2 V for LiCoO 2 . For voltages greater than 4.2 V, the capacity fading scaled with the measured amount of cobalt dissolution which was found to correlate with structural changes observed by in-situ X-ray diffraction.

The Electrochemistry of Zn3 N 2 and LiZnN A Lithium Reaction Mechanism for Metal Nitride Electrodes

LiZnN has been isolated by way of an electrochemical conversion reaction of Zn 3 N 2 with Li. We show that Zn 3 N 2 reversibly reacts with lithium electrochemically, exhibiting a large reduction capacity of 1325 mAh/g corresponding to the insertion of 3.7 Li per Zn. Of this initial capacity, 555 mAh/g were found to be reversible. Through the use of extensive in situ and ex situ X-ray diffraction, the reaction mechanism with lithium was identified as a conversion reaction of Zn 3 N 2 into LiZn and a matrix of βLi 3 N, the high pressure form of Li 3 N. Upon oxidation, LiZn transformed into metallic Zn, while βLi 3 N contributed to the transformation into LiZnN. This is the first identification of a reversible Li 3 N conversion mechanism. The formation of LiZnN as the new end member of the electrochemical reaction with lithium was identified as the cause of the irreversible loss observed during the first cycle. The βLi 3 N and LiZn oxidation into LiZnN was found to be reversible upon subsequent cycles. Poor cycle life was mainly attributed to the electromechanical grinding of the Li-Zn alloying reaction. Cu 3 N is also introduced as a material utilizing a similar conversion reaction but exhibiting improved cycle life.

Electrochemistry of Cu3N with lithium: A complex system with parallel processes

Cu 3 N was examined as a candidate negative electrode material for rechargeable Li-ion batteries. Cu 3 N electrodes exhibited good cycle life and excellent rate capabilities. The investigation of the materials electrochemical reaction mechanism revealed that the electrochemistry of Cu 3 N is rich with several parallel processes. In addition to a reversible lithium/copper nitride conversion process and the formation/decomposition of an organic layer at the surface of the nanocomposite, large cycle number and elevated temperature were found to promote a reversible lithium/copper oxide conversion process. Although the lithium/metal nitride conversion process was found to exhibit poor cycling stability, it constituted a fundamental step in the electrode chemistry as it generated highly active Cu nanoparticles which may have activated the formation of an organic layer and the formation of copper oxide. The oxidation of Cu metal into Cu 2+ to form CuO and its reduction contributed to the increase in capacity with cycle number. However, the maximum capacity obtained at high rate and elevated temperature far exceeded the theoretical capacity associated to the reduction of pure Cu 2+ into Cu metal These results suggest that the formation/decomposition of an electrolyte interface layer, which may become more substantial with cycling, and other reaction processes, such as a Li-Cu alloying reaction, may provide the additional capacity during cycling.

Transport properties of Nafion™ composite membranes for proton-exchange membranes fuel cells

Abstract The dielectric and electrical behaviors of a Nafion™ membrane, Nafion™/7SiO 2 –2P 2 O 5 –ZrO 2 (SPZ), Nafion™/SiO 2 +H 3 PO 4 (iHPO), Nafion™/SiO 2 +H 3 PW 12 O 40 (iHPW) and Nafion™/ZrP composites were investigated using water desorption measurements and ac impedance spectroscopy as a function of time, relative humidity and temperature. The composite membranes except Nafion™/ZrP exhibited only marginally higher conductivity than Nafion™ at high temperature. Unlike the other membranes, the Nafion™/ZrP membrane showed reduced proton mobility with a higher activation energy. The results suggest that improvements in fuel cell performance, previously observed in Nafion™/SPZ and Nafion™/ZrP composite membranes, are not due simply to increased hydrophilic properties, but that there is an interplay between structure and proton mobility.

Effect of precursors on the structure of phosphosilicate gels 29Si and 31P MAS-NMR study

Gels were prepared using phosphoric acid, triethyl phosphate and trimethyl phosphite as the precursors of phosphorus. The upper limit for P 2 O 5 for gel formation is 90 mol% for gels using phosphoric acid and trimethyl phosphite, and 70 mol% for gels using triethyl phosphate. Gels prepared from phosphoric acid with more than 10 mol% of P 2 O 5 partially crystallize to Si 5 O(PO 4 ) 6 after heat treatment at 200°C for 10 h. Under the same heat treatment condition, all the gels prepared from trimethyl phosphite crystallize. Gels prepared from triethyl phosphate are amorphous even after heat treatment at 800°C for 10 h. However, chemical analysis indicates that most of the phosphorus in the triethyl-phosphate-prepared gels is driven off during the heating process; no more than 8 mol% of the initial amount of P 2 O 5 is retained in the gels. The effects of phosphorus precursors on the phosphosilicate gels were studied by X-ray diffraction, and 29 Si and 31 P magic angle spinning (MAS)-nuclear magnetic resonance (NMR). 29 Si and 31 P spectra show that Si O P and P O P linkages do not form in the xerogels (i.e. gels heat treated to ∼ 60°C). The gels heat treated at 200°C or above (those that did not crystallize) show evidence of Si O P and P O P network formation.

Reversible Conversion Reactions with Lithium in Bismuth Oxyfluoride Nanocomposites

For the first time, the effect of the oxygen anion substitution on the reversible electrochemical activity of metal fluorides was investigated through the use of bismuth oxyfluorides. We demonstrate that a reversible conversion occurs in BiO x F 3 - 2 - x ( x = 0 , 0 . 5 , 1 , 1 . 5 ) /C nanocomposites with the formation of Bi°, Li 2 O, and/or LiF during lithiation and the reformation of BiO x F 3 - 2 x during subsequent delithiation as opposed to a two-phase mixture of Bi 2 O 3 + BiF 3 . We also show evidence that in BiOF and BiO 0 . 5 F 2 the fluoride component reacts first at higher voltage during lithiation to form Bi°, LiF, and Bi 2 O 3 . The newly formed Bi 2 O 3 is then subsequently reduced into Li 2 O and Bi° on a second plateau at lower voltage. During delithiation the process is reversed, Li 2 O reacting first at low voltage followed by LiF. Although cycling is presently poor, our results show that a relatively low oxygen content can significantly improve the electrochemical activity of the metal fluoride, suggesting that optimized metal oxyfluorides could be an attractive alternative to metal fluorides for applications as positive electrode materials in lithium batteries.

Synthesis of electrochemically active liCoO2 and liNiO2 at 100°c

Abstract Presently, fabrication of LiCoO2 requires calcination and annealing at temperatures ranging from 850 °C to 1000 °C from one to two days. Herein we show that LiCoO2 powders can be prepared at temperatures as low as 100 °C through the use of a two days ion exchange reaction between CoOOH and an excess of LiOH · H2O. The resultant LiCoO2 powders were studied through the use of X-ray diffraction, thermal gravimetry, and infrared analysis. Low-temperature fabricated LiCoO2 powders were well crystallized with lattice parameters agreeing with 850 °C prepared samples. The samples had an infrared spectrum similar to that of LiCoO2 with, however, weak extra lines indicating the presence of both carbonate and hydroxyl species. The presence of these species was found to severely affect both the capacity and reversibility of the electrochemical intercalation reactions. Heating these powders at moderate temperatures, around 250 °C, was effective in removing the species and drastically improving their electrochemical cycling properties. The same ion exchange reaction was successfully applied to the synthesis of LiNiO2.

The Electrochemistry of Germanium Nitride with Lithium

Ge 3 N 4 was investigated for its electrochemical activity with lithium as a possible negative electrode material for Li-ion batteries. Ge 3 N 4 was found to reversibly react with Li, exhibiting high capacity, 500 mAh/g, and maintaining good cycling stability. The reaction mechanism of Ge 3 N 4 with lithium was investigated in detail using in situ and ex situ X-ray diffraction (XRD) in reflection, in situ XRD in transmission, ex situ transmission electron microscopy, and selected-area electron diffraction (SAED). The two phases, α- and β-Ge 3 N 4 , of the electrode material mostly maintained their respective crystalline microstructure during cycling. A substantial integrated intensity decrease in the XRD Bragg reflections observed during the first lithiation and the concurrent emergence of diffuse rings in SAED suggest the reaction of Ge 3 N 4 with lithium may be limited thereby converting only the outermost shell of the Ge 3 N 4 crystal. The identification of α-Li 3 N and Ge at the end of the first delithiation using SAED supports a lithium/metal nitride conversion reaction process. The formation of the Li 3 N matrix was found to be consistent with a 50% irreversible capacity loss in the first cycle.

Densification of monolithic silica gels below 1000°C

Abstract Changes in the microstructure of an acid-catalyzed silica gel have been studied through the gel to glass conversion temperature. The gel composition selected was known to produce a dense silica glass at 1000°C that was not distorted or cracked. Surface area, porosity and pore size distribution were measured. Shrinkage during constant rate heating was observed. Effects of heating rate and atmosphere on densification were noted. An optimum heating rate of 2.5°C/min, which is a balance between the temperature dependence of the viscosity and the temperature dependence of the dehydroxylated surface tension, was found to give the greatest shrinkage below 1000°C.