(Invited) Use of Secondary Materials in NMC Co-Precipitation: Effect of Impurities on the Cell Performance

Abstract

The growing demand for raw materials in lithium-ion batteries increasingly requires the increased use of secondary raw materials in the co-precipitation of NMC battery chemicals. Traditionally, pure raw materials as metal solutions are required for co-precipitation, which is a challenge for battery producers. Removal of impurities is also expensive and challenging step. Aim of this research is to study the impact of impurities present in metal sulfate solutions on the battery cell performance. These metal sulfate solutions were obtained from metal industry’s side-stream and from recycled spent batteries.Metal sulfate solutions with known amount of impurities were used in (Ni0.8Co0.1Mn0.1)OH2 (NMC811) precursor co-precipitation. Commercial pure manganese and nickel sulfates were used as reference samples. Key impurities in the solutions were zinc, iron, calcium, potassium, lithium, and magnesium. Composition of samples was determined using ICP-OES. We also demonstrate actions to control precipitation of impurities during the NCM precursor co-precipitation. Solubilities of impurities were calculated using MEDUSA and HYDRA.Dried hydroxide precursor (sieved <25µm) was mixed with LiOH in molar ratio of Li:(Ni0.8Co0.1Mn0.1:Al) 1.04:0.996. LiOH excess was used to compensate lithium loss during high temperature calcination and ensure homogenous lithiation. The mixture calcined with 2.5°C/min heating ramp and 6 hours holding time at 800 °C under O2 flow. Material was milled and sieved <40µm in the dry room conditions. Residual lithium washed from surface with certain amount of de-ionized water and dried in vacuum oven.Cathode slurry was mixed with Thinky ARE 250 mixer. Slurry composition was 4% of PVDF, 4% carbon and 92% active material and NMP as solvent. Cathode slurry was spread on aluminum foil with 100 µm applicator, then dried at 50 °C for few hours and finally at vacuum oven at 120 °C overnight. Cathode foil was calendered three times before coin cell assembly. The active material loading on the foil was about 12 mg/cm2. Theoretical capacity used to calculate the C-rate was 200 mAh/g. Electrochemical performance was carried out using one electrode pair pouch cell (50 mAh) was prepared with graphite anode, and electrolyte was 1.15M LIPF6 in EC:DMC:EMC (2:4:4) and 1 % of vinylane carbonate. All foils and coin cells were prepared in the dry room. Cells tested between 4.2V-2.8V. 1100 cycles at 1C/1C currents including capacity check every 100 cycles at 0.2C/0.2C currents. Before the capacity check, cell was discharged at 0.2C.The microstructure shown by field emission scanning electron microscope (FESEM) images was obtained using a Zeiss Sigma FESEM operating at 5kV at the Centre for Material Analysis of the University of Oulu. X-ray diffraction (XRD) analysis was done for the calcined NCM811 cathode materials by using Rigaku SmartLab 9 kW with Co-Kα radiation at 40 kV and 135 mA. Diffractograms were collected in the 2θ range (5–120° at 0.01° intervals). Diffraction peaks and crystallite parameters were identified using International Centre for Diffraction Data (PDF-4+ 2020). The specific surface areas, pore sizes and pore volumes of samples were determined from nitrogen adsorption desorption isotherms at the temperature of liquid nitrogen (−196 ◦C) using a Micromeritics ASAP 2020.The tapped density of powders was measured based on the ISO EN 787/11 standard. Residual lithium was measured with an automatic titrator. Particle size distribution (PSD) during the co-precipitation was measured with a Malvern Mastersizer 3000.Based on the results, co-precipitation of impurities to the NMC particles followed the thermodynamic calculations. As expected, iron, zinc, magnesium and calcium co-precipitated in NMC811 to a large extent, but potassium and lithium remain in the mother liquor. The presence of impurities was also confirmed by several characterizations. Impurities did not affect the particle morphology or tap density of NMC811. Surprisingly, the effect of impurities was also minor to the cell performance. During the cycling, these cells provided good cyclability and remain high retention (>80%) after 1100 cycles compared to reference samples.