In this project, we developed a new concept for the production of ultra-thin Li-ion batteries on flexible substrates, by printing the electrode materials directly on top of the substrate. As negative electrode materials, we will use nanoparticles of metal oxides, such as ZnO, in combination with amorphous carbon. Films of these metal oxides have been prepared by printing techniques and have a high specific charge capacity [1-3]. Furthermore, for the next step we will use nanocrystalline LiFePO₄ und Li₄Ti₅O₁₂, that have not been prepared by printing processes yet, as positive electrodes. These materials are studied extensively with respect to applications in conventional Li-ion batteries and are already used commercially [4, 5].  

In future applications, the printable battery production process has the potential to be integrated into the production of the complete active devices. The advantage of printing batteries is obvious: the electrodes can be structured in a rapid and simple way, enabling the combination of different electrode materials for hybrid battery systems that include components with high energy density as well as components with high power density. For applications in mass markets the processes are very cost effective. Furthermore, the combination of printable batteries with solar cells, i.e. the concept of storing electric energy directly at the site of the energy conversion, gives an alternative to feeding the renewable energy into the electric grid.

The first part of this project belongs to the preparation and stabilization of suitable dispersions of the metal oxides for the printing process. The nanomaterials and the morphology of the printed films were investigated by using X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and atomic-force microscopy (AFM). Subsequently, the dispersions were processed on flexible paper by ink-jet-printing process or spin coating. The battery performance of the samples was tested in Swagelok type batteries. From the galvanostatic cycling we got the information about the specific charge capacity, energy density, power density and cycle lifetime.

Dispersion and stabilization of nanoparticulate ZnO

For a solution process like ink-jet printing and spin coating the material has to be dispersed in a solvent. However, particles in a solution tend to agglomerate resulting in instable dispersions. To reduce the agglomeration of the particles in the solution the particles have to be stabilized. The stabilization of particles depends on many parameters, e.g. the size, the surface conditions, the stabilizer molecules, and the solvent. In all experiments the solvent and the stabilizer molecule was fixed. We used 2-Methoxyethanol as solvent and a commercial co-polymeric (Tego 751) as stabilizer. The dispersion was processed by a homemade dispersing process. After dispersing the size of the particles within the dispersion was measured by dynamic light scattering (DLS). The particle sizes in the solution are much bigger compared to the values of the dry powder. This is due to further agglomeration of the particles in the solution. The stabilization against agglomeration is not sufficiently effective for all particles. The results show that the smallest particles create the biggest agglomerates, although no clear correlation between powder particle size and the agglomeration in the solution can be observed.

In addition to the size of the particles the surface modification and the surface condition are important for stabilization. This is due to the interaction between the surface defects in ZnO and the stabilizer molecules. Higher concentration of surface defects will enhance the stabilization. Surface defects in ZnO are well studied by photoluminescence.

Battery performance

The different ZnO powders were tested in two-electrode coin cells against Li metal counter electrodes. Galvanostatic cycling was performed in the voltage window 0.02 V - 3 V. We used a rate of C/100, i.e. complete discharge/charge within 100 h, and 1M LiPF₆ in EC/DMC as electrolyte. The results correspond to an insertion of about 3 Li ions per formula unit ZnO and thus is consistent with the mechanism described by the subsequent reactions:

In the following charging/discharging steps strong decreases of the specific capacity are observed. This has to be attributed to the formation of new phases coming along with big volume changes (especially during the LiZn alloy formation), separation of the discharging products and thus very limited reversibility of the above described reactions. We could show that the capacity retention is best for the electrodes with the smallest particle sizes.

 References

List of Publications 2006-2011 as PDF

Subproject Report 2006-2010 as PDF