Birgisson, Steinar9; Jensen, Kirsten Marie Ørnsbjerg9; Christiansen, Troels Lindahl10; Bøjesen, Espen Drath9; Christensen, Mogens9; Tyrsted, Christoffer7; von Bülow, Jon Fold8; Iversen, Bo Brummerstedt9
1 Department of Chemistry, Science and Technology, Aarhus University2 Department of Engineering - Membrane and Sensor Technology, Department of Engineering, Science and Technology, Aarhus University3 Department of Chemistry - Centre for Materials Crystallography (CMC), Department of Chemistry, Science and Technology, Aarhus University4 Interdisciplinary Nanoscience Center - INANO-Kemi, Langelandsgade, Interdisciplinary Nanoscience Center, Science and Technology, Aarhus University5 Interdisciplinary Nanoscience Center, Science and Technology, Aarhus University6 Department of Chemistry - Centre for Energy Materials (CEM), Department of Chemistry, Science and Technology, Aarhus University7 Interdisciplinary Nanoscience Center, Faculty of Science, Aarhus University, Aarhus University8 Haldor Topsøe A/S9 Department of Chemistry, Science and Technology, Aarhus University10 Department of Chemistry - Centre for Materials Crystallography (CMC), Department of Chemistry, Science and Technology, Aarhus University
Spinel LiMn2O4 is a well-known cathode material for Li-ion batteries. It is considered to be a safer, more environmentally friendly and cheaper alternative to the widely used LiCoO2 cathode material. During charging/discharging of a Li-ion battery it is necessary for the Li-ions to be extracted/inserted into the cathode material. The specific capacity of the cathode material is determined by how many Li-ions can diffuse in and out of the cathode material structure per mass unit. Furthermore the rate of the electrochemical reaction (and therefore the power output of the battery) can be limited by how fast the Li-ions can diffuse in and out of the structure. The spinel structure of LiMn2O4 allows for a three dimensional Li-ion diffusion via tetrahedral and octahedral holes in the cubic close packed oxide structure. By using LiMn2O4 nanoparticles as a cathode material for Li-ion batteries instead of micronsized particles the diffusion lengths inside the cathode material structure are significantly shorter. Therefore higher specific capacity and more specific power can be achieved by using nanoparticles instead of micro particles [S.H. Ye et al, Electrochim. Acta 49 (2004) 1623]. We have shown that it is possible to produce spinel LiMn2O4 nanoparticles by a hydrothermal reaction route. In the reaction KMnO4 is reduced in a redox reaction using an organic reduction agent (here ethanol) with Li-ions present in the solution. It is seen that the primary product of this reaction is either spinel or layered structure LiMn2O4, depending on the initial concentration if Li-ions. An impurity phase, identified as Mn3O4, is also detected in different concentrations depending on reaction time and temperature. We have developed an experimental technique for in-situ measurements of solvothermal reactions under sub- and supercritical conditions [Becker et al, J. Appl. Crystallogr. (2010) 43]. The technique uses synchrotron X-ray radiation to measure time resolved powder x-ray diffraction patterns while the reaction is happening thereby giving real time information on crystalline phase formation, particle sizes and other structural properties for the reaction being studied. Normally the reactions are started by heating and a constant temperature is kept throughout the experiment. In this study the hydrothermal reaction previously shown to produce spinel LiMn2O4 nanoparticles is studied in-situ to learn more about the phase formation, particle growth and hopefully use the information to optimize the reaction for production of spinel LiMn2O4 nanoparticles in a flow reactor. The information can possibly be used to control the particle size and phase purity of the flow reactor reaction. Also due to the time resolved data acquired in the in-situ measurements it gives a unique opportunity to study reaction kinetics and thermodynamic quantities of the reactions. A temperature study of the reaction has been conducted to see how the formation rate and particle growth is affected by temperature while the precursor concentration is kept constant. The precursor solution is an aqueous solution with Li:Mn:EtOH molar ratio of approximately 1:2:7 and the reactions conditions are constant temperature at 220°C, 260°C, 300°C, 350°C and 400°C at 250 bar. First results show that the initial product of the hydrothermal reaction is spinel LiMn2O4 nanoparticles but it is seen that this phase is unstable towards the formation of Mn3O4 nanoparticles under the reaction conditions studied. In other words after the initial formation of the spinel LiMn2O4 nanoparticles a decomposition to Mn2O3 is observed. From the result a probable reaction route is suggested and presented in equations (1) and (2). (1) 4LiOH + 8KMnO4 + 7CH3CH2OH --> 4LiMn2O4 + 8KOH + 7CH3COOH + 5H2O (2) 12LiMn2O4 + 5CH3CH2OH + H2O --> 8Mn3O4 + 12LiOH + 5CH3COOH Looking at the weight fraction of each of the phases as a function of reaction time for the different temperatures it is evident that the transformation from LiMn2O4 to Mn3O4 happens faster at higher temperatures and is complete within 30 minutes reaction time at 350°C and 400°C. Further data analysis is ongoing. For example it is possible to fit the formation curves to Johnson–Mehl–Avrami kinetic model and extract rate constants for the transformation reaction from LiMn2O4 to Mn3O4 at different temperatures. Using this information it is possible to calculate activation energy of the reaction using the Arrhenius equation. Further in-situ measurements using total scattering technique are in progress. The hope is to extract information about the initial transformation from the amorphous (or poorly crystalline) precursor solution to the crystalline LiMn2O4 nanoparticles.