Machine learning-guided inkjet printing of tin oxide nanoparticle inks on laser-textured copper foils for high-performance sodium-ion battery anodes

• Tin (Sn) nanoparticle inks (96 nm, 1.77 mPa·s) were synthesized via laser ablation for sodium-ion battery anodes. • Laser texturing increased surface roughness by 367 %, enabling Sn inkjet printng on Cu foils. • Sn anodes exhibited 400 mAh/g at 0.1C in SIBs. • Machine learning can predict surface roughness of printing substrate based on images. • Integration of real-time laser ablation monitoring and AI-driven process control enabled reproducible Sn ink fabrication for scalable anode production. Inkjet printing tin oxide (SnO 2 ) nanoparticle inks on copper foils for sodium-ion battery (SIB) anodes presents challenges such as particle size control (large particles can clog nozzles and cause uneven prints), viscosity (values above 3 mPa·s are unprintable on this system), concentration, and surface adhesion. Since copper foils are hydrophobic, surface modification is necessary to improve adhesion. This study investigates SnO 2 ink formulation, copper substrate preprocessing, and inkjet printing using design of experiments and characterization. SnO₂ inks with 96 nm particles and a viscosity of 1.77 mPa·s at 15 °C were synthesized via laser ablation with real-time monitoring. Laser texturing (50–70 µm track spacing at 10–20 kHz) using a Nd:YAG laser (1064 nm) increased copper surface roughness by 367 % (up to 2.1 µm) and created micro-cavities, enabling successful ink deposition. The repetition rate was identified as the most influential parameter during texturing. Printing was carried out at 50 °C with a 1-minute interlayer delay to prevent runoff. Machine learning models (Random Forest, Linear Regression, Extreme Gradient Boosting, Support Vector Machine, and K-Nearest Neighbour) were used to predict surface roughness based on laser parameters. Printed SnO 2 nanoparticles on copper foils demonstrated an initial discharge capacity exceeding 400 mAh/g at 0.1 C.