Tracking UV Degradation of Inkjet Prints Using ToF-SIMS and Machine Learning

Abstract

Inkjet printing is widely deployed in a variety of commercial and consumer contexts and is an excellent reference system for exploring complex and subtle changes that occur in response to ageing and environmental exposure. This study describes a robust and unbiased assessment of changes in inkjet ink chemistry in response to real-world environmental exposure (sunlight and ambient heat). Over a period of 21 weeks, commercial inks are exposed to solar UV light and monitored for chemical changes via time-of-flight secondary ion mass spectrometry (ToF-SIMS). Self-organizing map with relational perspective mapping (SOM-RPM), hierarchical cluster analysis (HCA), and partial least squares (PLS) regression are used to analyze the ToF-SIMS data and elucidate chemical changes both spectrally and spatially over time. Ultraviolet (UV) light is found to cause significant changes to ink chemistry. Each of the inks (black, cyan, magenta and yellow) react in a unique manner, with the black and yellow inks being most affected. This study exemplifies how machine learning can pinpoint subtle changes in hyperspectral data and provide insight into the long-term thermal and UV degradation of inkjet inks.