🔬 materials science

Multimodal Visual Sensing of Temperature and Pressure: From Spectroscopic Readout to Multiple Linear Regression-Enhanced RGB Analysis

This study presents a concentration-tunable KGaGeO4:Bi3+,Eu3+ platform that enables selective temperature or pressure sensing through multimodal spectroscopic readouts, achieving significantly enhanced sensitivity by integrating RGB imaging with multiple linear regression analysis.

Original authors: Maja Szymczak, Yufan Meng, Guanjun Xiao, Miguel A. Hernández-Rodríguez, Iga Sawaryn, Bo Zou, Lukasz Marciniak

Published 2026-09-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Maja Szymczak, Yufan Meng, Guanjun Xiao, Miguel A. Hernández-Rodríguez, Iga Sawaryn, Bo Zou, Lukasz Marciniak

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the modern world, machines and processes are becoming increasingly complex, often operating inside sealed containers, buried under layers of material, or shrinking down to sizes where traditional sensors simply cannot fit. To monitor these hidden environments, scientists have turned to light. Instead of inserting a physical probe that might break or interfere with the system, they use materials that glow when struck by light. The color and brightness of this glow change depending on the conditions around it, such as heat or crushing force. This allows researchers to "see" temperature and pressure from a distance, using a camera rather than a wire. While this method works well for measuring one thing at a time, a major challenge remains: making a single material that can accurately report two different conditions simultaneously without confusing them. If a material changes color when it gets hot, it becomes difficult to tell if a color change is due to heat or pressure.

A team of researchers has developed a new approach to solve this problem using a specific type of glowing powder. They created a material made of potassium, gallium, germanium, and oxygen, which they then doped with two different glowing elements: bismuth and europium. Think of the material as a canvas where the researchers could paint with light. By carefully adjusting the amount of europium mixed into the powder, they could tune the material to act as either a thermometer or a pressure gauge. When the mixture contained more europium, the material became highly sensitive to temperature changes. When the mixture contained very little europium, it became a precise pressure sensor that was largely unaffected by temperature fluctuations. This flexibility comes from the unique way the two glowing elements behave. The bismuth ions produce a broad, blue-green glow that reacts strongly to both heat and pressure, while the europium ions emit a sharp, red light that is much more stable. By balancing these two, the researchers created a system where the competition between the blue-green and red light shifts in predictable ways depending on the environment.

The researchers tested these materials in two different ways. First, they heated the samples to see how the light changed. They found that as the temperature rose, the red light from the europium faded while the blue-green light from the bismuth grew brighter. This shift caused the overall color of the glow to move from red to white and finally to blue. By capturing images of this glowing powder with a standard digital camera, they could measure the intensity of the red, green, and blue channels separately. Instead of just looking at the ratio of red to blue, they used a statistical method called multiple linear regression. This technique allowed them to combine the information from all the color channels at once, rather than looking at them in isolation. This simple step dramatically improved the accuracy of the temperature reading. The method increased the sensitivity of the temperature measurement from about one percent per degree to nearly nine percent per degree, making it possible to detect very small changes in heat with a simple camera.

Next, the team tested the material's ability to sense pressure. They placed a sample containing very little europium inside a diamond anvil cell, a device that uses two tiny diamonds to squeeze a sample with immense force. As they increased the pressure from zero up to nearly fifteen gigapascals, the blue-green light from the bismuth dimmed rapidly, while the red light from the europium remained relatively steady. This caused the glowing powder to change color from turquoise to white and finally to deep red. Just as with the temperature tests, they took pictures of the glowing sample at different pressure levels. Using the same statistical method to analyze the red, green, and blue channels together, they found that the pressure sensitivity jumped from about one hundred and sixty-nine percent per gigapascal to over seven hundred percent per gigapascal. This means the material can detect extremely small changes in pressure with high precision. Crucially, they also checked how much the temperature affected these pressure readings. They found that the pressure signal remained stable even as the temperature changed, meaning the sensor could work reliably in environments where the heat fluctuates.

The significance of this work lies in its dual nature and its simplicity. The researchers demonstrated that a single material can be engineered to perform two distinct sensing jobs simply by changing its chemical recipe. Furthermore, they showed that high-tech, expensive spectrometers are not always necessary to get precise data. By combining standard digital photography with advanced data analysis, they created a low-cost, portable way to map both temperature and pressure across a surface. This approach opens the door to new types of sensors that can be embedded in materials or placed in hard-to-reach areas, providing detailed visual maps of physical conditions that were previously impossible to measure without complex equipment. The study confirms that by carefully controlling the balance of glowing elements and using smart data processing, scientists can create versatile tools for understanding the physical world.

Technical Summary: Multimodal Visual Sensing of Temperature and Pressure via MLR-Enhanced RGB Analysis

Problem Statement
The development of optical sensors capable of reliable, sensitive, and straightforward readout of physical parameters is critical for automated and digitalized technological processes. While luminescence-based sensing offers high sensitivity and remote readout capabilities, distinguishing between temperature and pressure often requires complex spectroscopic setups or suffers from cross-sensitivity. Furthermore, while ratiometric luminescence thermometry using RGB imaging is established, its application to luminescence manometry (pressure sensing) remains underexplored. Existing color-image-based approaches for pressure sensing typically rely on simple intensity ratios between single color channels, potentially underutilizing the information contained in the full spectral response and limiting sensitivity. There is a need for a multifunctional platform that can be tuned for specific sensing modalities while integrating simple digital imaging with advanced data analysis to enhance sensitivity and spatial resolution.

Methodology
The authors synthesized a concentration-tunable luminescent platform, KGaGeO4:Bi³⁺,Eu³⁺, using a high-temperature solid-state reaction method. The study leveraged the distinct spectroscopic responses of two Bi³⁺ centers (Bi³⁺(1) and Bi³⁺(2)) and Eu³⁺ emission to different external stimuli:

Composition Control: By varying the Eu³⁺ concentration (0.1% to 5%), the authors tailored the material for specific sensing regimes. A high-Eu³⁺ composition (5%) was selected for thermometry due to pronounced temperature-dependent spectral shifts, while a low-Eu³⁺ composition (0.1%) was chosen for manometry to minimize temperature cross-sensitivity.
Spectroscopic Characterization: The materials were characterized using XRD, Raman spectroscopy, SEM, and EDS. Temperature-dependent (93–673 K) and pressure-dependent (0–16 GPa) luminescence measurements were performed using spectrofluorometers and a diamond anvil cell (DAC).
Imaging and Data Analysis: Digital luminescence images were acquired using a standard Canon EOS 400D camera. The images were decomposed into Red (R), Green (G), and Blue (B) channels.
- Traditional Readout: Sensitivity was evaluated using Luminescence Intensity Ratios (LIR), CIE1931 chromaticity coordinates, and spectral shifts.
- Advanced Readout: Multiple Linear Regression (MLR) was applied to combine information from multiple color-channel ratios (specifically R/G and R/B) simultaneously. This multivariate approach was used to construct calibration models for both temperature and pressure, aiming to extract more information than single-ratio methods.

Key Contributions and Results

Material Design and Mechanism: The study identified that Bi³⁺ ions occupy distinct, distorted six-coordinated K⁺ sites, while Eu³⁺ occupies higher-coordination sites. The two Bi³⁺ centers exhibit different excitation selectivities and temperature dependencies. Crucially, Bi³⁺ emission is highly sensitive to both temperature and pressure, whereas Eu³⁺ emission is relatively stable under pressure but sensitive to temperature. This differential response enables the separation of stimuli through composition tuning.
Thermometry (High-Eu³⁺ Composition):
- The KGaGeO4:1%Bi³⁺,5%Eu³⁺ sample exhibited a strong temperature-induced redistribution of emission from red (Eu³⁺) to blue-green (Bi³⁺).
- Conventional ratiometric thermometry (LIR) achieved a maximum relative thermal sensitivity ( $S_R$ ) of 1.55% K⁻¹.
- RGB-based sensing: Using standard RGB ratios, the R/G ratio yielded a sensitivity of 0.75% K⁻¹, while the R/B ratio yielded 1.05% K⁻¹.
- MLR Enhancement: By applying MLR to the R/G and R/B ratios simultaneously, the maximum relative thermal sensitivity was significantly enhanced to 8.82% K⁻¹, representing an approximately ninefold improvement over individual ratios. This allowed for quantitative, spatially resolved temperature imaging.
Manometry (Low-Eu³⁺ Composition):
- The KGaGeO4:1%Bi³⁺,0.1%Eu³⁺ sample showed a strong pressure response (Bi³⁺ quenching) with minimal temperature interference.
- Pressure-induced color changes were observed from turquoise (ambient) to red (16 GPa).
- Spectroscopic Readout: The LIR-based manometer achieved a maximum relative pressure sensitivity of 54.47% GPa⁻¹. The Thermal Invariability Manometric Factor (TIMF) was calculated to be 114.9 K GPa⁻¹, indicating low temperature cross-sensitivity.
- RGB-based sensing: Standard RGB ratios yielded a maximum sensitivity of 169% GPa⁻¹ (for the R/G ratio).
- MLR Enhancement: The application of MLR to the RGB data increased the maximum relative pressure sensitivity to 721% GPa⁻¹, a 4.3-fold improvement. This represents the first demonstration of MLR-assisted RGB analysis for quantitative luminescence pressure sensing.

Significance and Claims
The paper claims that this work establishes a new route for multifunctional optical sensing by integrating material engineering with multivariate data analysis. The primary significance lies in the demonstration that:

A single host-dopant system (KGaGeO4:Bi³⁺,Eu³⁺) can be compositionally tuned to function as either a highly sensitive thermometer or a manometer, minimizing cross-sensitivity.
Multiple Linear Regression (MLR) can effectively exploit the complementary information contained in multiple RGB color channels, drastically improving sensing sensitivity compared to conventional single-ratio approaches for both temperature and pressure.
The proposed strategy enables quantitative, spatially resolved sensing using low-cost, commercially available digital cameras rather than complex spectrometers.
This is the first application of MLR-enhanced RGB imaging specifically for luminescence pressure sensing, extending the utility of multivariate analysis from thermometry to manometry.

The authors conclude that this synergy between composition-controlled functionality, multimodal spectroscopic sensing, and data-assisted imaging offers a powerful strategy for designing next-generation, highly sensitive, and practical optical sensors.

More like this