Colloidal Crystals Based Time-Temperature Integrators

In the realm of advanced materials, the autonomous and manipulation-free recording of temperature states over extended periods is gaining critical importance for, e.g., battery safety assessments. Our research introduces a novel concept for time–temperature integrators (TTIs) utilizing colloidal crystals, which provide a versatile and effective optical readout for low-tech visual inspections.<br/>The innovative approach leverages two key features of colloidal crystals. First, the film-formation kinetics of these crystals can be precisely controlled by blending particles with distinct glass transition temperatures (Tg). Second, by creating a linear gradient of these particle mixtures within the colloidal crystal, we enable a localized and gradual readout mechanism. Specifically, tailor-made latex particles, uniform in size but with varying <i>T</i>_g, create a homogeneous photonic stopband. The disappearance of this opalescence directly correlates with the local particle ratio and the specific time-temperature exposure, offering a straightforward and visual indication of thermal history.¹ This material is capable of autonomously recording isothermal heating events.<br/>Increasing the complexity of the colloidal particle mixtures introduces a redundancy of film formation kinetics. We use these in the form of an array of colloidal crystal spots to train an artificial neural network to learn the individual time and temperature dry sintering kinetics. This approach allows for a machine learning-enabled precise and independent measurement of time and temperature parameters, using only a standard smartphone camera for readout.² Our findings highlight the significant potential of integrating machine learning in materials science to enable novel forms of functional devices.<br/><br/>(1) Schöttle, M.; Tran, T.; Feller, T.; Retsch, M. Time-Temperature Integrating Optical Sensors Based on Gradient Colloidal Crystals. <i>Adv. Mater. </i><b>2021</b>, <i>33</i> (40), e2101948. DOI: 10.1002/adma.202101948<br/>(2) Schöttle, M.; Tran, T.; Oberhofer, H.; Retsch, M. Machine Learning Enabled Image Analysis of Time-Temperature Sensing Colloidal Arrays. <i>Advanced Science </i><b>2023</b>, <i>10</i> (8), 2205512. DOI: 10.1002/advs.202205512