From Bug to Feature: Harnessing Cross‐Sensitivity for Multiparametric Luminescence Sensing

ABSTRACT Luminescence sensing is a powerful technique for remote monitoring of chemical and physical parameters. Yet, measurement accuracy is often compromised by cross‐sensitivity—confounding modulation of the optical readout feature by nontarget parameters. Conventionally treated as a “bug”, cross‐sensitivity is generally mitigated through strategies aimed at minimizing or compensating for its effects. In this work, we reconceptualize cross‐sensitivity as a “feature” and unveil its latent potential to enable multiparametric luminescence sensing using a single sensor. We achieve this by employing ruby (Al 2 O 3 :Cr 3 + ) microspheres as a model system—a popular luminescent pressure standard with marked cross‐sensitivity to temperature. First, we explore the cross‐sensitivity of various conventional optical readout features in ruby and propose a framework for quantifying accuracy loss as a function of the interfering parameter, enabling a fair evaluation of cross‐sensitivity across features. We then harness cross‐sensitivity using Linear Discriminant Analysis (LDA)—a supervised machine learning algorithm trained to isolate the effects of pressure and temperature, enabling simultaneous measurement of both parameters without the need for correction. This work lays the foundation for multiparametric luminescence sensing using a single sensor, expanding the capabilities of this technique and broadening its application potential.