2024 Impact factor 1.7
Condensed Matter and Complex Systems

EPJ ST Highlight - Data-Driven Insights into Inkjet Droplet Formation

Visualising morphologies of inkjet-printed droplets

High-speed image analysis shows how the control parameters of inkjet printers are linked to the shapes of the ink droplets they produce – helping researchers to optimise the printing process.

Inkjet printing has become a cornerstone of high-tech microfabrication, underpinning applications ranging from microchip production and drug delivery to DNA sequencing and tissue engineering. In these fields, precision is paramount - the ability to reliably place picolitre-sized droplets with exact morphology determines the success of both medical treatments and microelectronic device fabrication.

Despite advances in computational fluid dynamics (CFD), simulating and controlling droplet formation in real-world conditions remains a challenge due to the complexity of two-phase flows and the vast number of operational parameters involved. To address this, researchers are turning to data-driven approaches as a complementary or alternative strategy. These methods can reduce reliance on time-consuming simulations and enable real-time analysis and decision-making in manufacturing environments.

In a new study published in EPJ Special Topics (EPJ ST), researchers at CIMNE/UPC present a comprehensive data-driven investigation of droplet morphology in inkjet printing. The team, led by Pavel Ryzhakov, began by performing extensive controlled droplet-generation experiments using a piezoelectric inkjet dispenser. Each droplet was captured via high-speed imaging, yielding a rich dataset of raw images and extracted geometrical features.

The study’s image processing pipeline, openly published on GitHub, automatically identified and measured key droplet features such as shape, size, and the presence of satellite droplets. After rigorous data cleaning, the researchers conducted correlation analyses to examine how droplet formation is influenced by voltage pulse parameters including amplitude, width, and delay.

The results reveal strong, quantifiable relationships between signal characteristics and droplet morphology, offering a practical roadmap for tuning dispenser settings to consistently produce stable, single droplets - minimising waste and optimising performance in advanced manufacturing.

This work marks a significant advance at the intersection of experimental microfluidics and AI-driven optimisation, laying the foundation for more responsive, precise, and sustainable digital manufacturing systems.

Editors-in-Chief:
Reinhold Egger and Philipp Hövel
I am naturally indebted to you and the referees who contributed to this success with your time and constructive advice.

Hamid Assadi

ISSN (Print Edition): 1434-6028
ISSN (Electronic Edition): 1434-6036

© EDP Sciences, Società Italiana di Fisica and Springer-Verlag