*Important notice: This news reports on an unedited version of an accepted paper and is awaiting final editing. Therefore, the paper should not be regarded as conclusive or treated as established information.
A compact, low-power, proof-of-concept optical system, termed SMILE (Sensing of Microplastics in Liquid Environments), has been introduced to detect and classify microplastics in flowing water. This innovative system combines flow-based particle transport, dual-channel ratiometric fluorescence sensing, and a solvatofluorochromic dye to distinguish polymer types based on variations in their chemical environment. The findings were published in Scientific Reports.
Study: A dual-channel optical sensor for real-time microplastic classification via ratiometric fluorescence analysis. Image Credit: Floren Horcajo/Shutterstock.com
Microplastic Pollution Drives Demand for Faster Water Testing
Microplastic pollution is a growing concern in aquatic environments, with particles detected in up to 83% of global tap water samples. Conventional detection methods, including Fourier-transform infrared (FTIR) spectroscopy and Raman microspectroscopy, provide detailed information about particle composition but require expensive equipment and time-consuming sample preparation, hindering their use in continuous monitoring.
Existing flow-based detection systems and commercial flow cytometers offer alternatives for analyzing particles in liquid samples. However, many rely on complex instrumentation or high-power excitation sources, increasing costs and limiting flow rates. Fluorescent dyes like Nile Red can count microplastic particles but do not reliably distinguish their polymer compositions.
To address these challenges, the solvatofluorochromic sensing employed in SMILE detects changes in fluorescence emission as the surrounding environment's chemical polarity changes, offering a potential solution for continuous microplastic detection.
SMILE Uses Dual-Channel Fluorescence to Identify Plastic Types
Researchers constructed the SMILE system by merging fluorescence confocal microscopy with a flow-based detection system, assembled on a portable 45 × 30 cm aluminum breadboard.
An analog pump transports aqueous samples containing suspended particles through a glass capillary with a 200 μm inner-diameter constriction at a flow rate of 430 mL/h, with adjustments possible up to 600 mL/h. An inline metal filter with a 49 μm mesh is installed upstream to remove larger particles and prevent blockage.
Before measurement, microplastic samples are treated with 4-dimethylamino-4'-nitrostilbene (DANS), a solvatofluorochromic dye whose fluorescence emission wavelength changes with the polarity of the surrounding medium. A 405 nm laser operating at 4.5 mW provides optical excitation. A dichroic mirror then directs the laser beam through a lens onto the capillary constriction, producing an elliptical focal spot (200 × 50 μm).
The same lens collects fluorescence emitted by particles passing through the detection region. A secondary dichroic beam splitter separates the emission into two spectral channels: channel one (CH1) detects wavelengths above 550 nm (yellow-to-red emissions from polar materials), while channel two (CH2) collects wavelengths below 550 nm (blue-to-green emissions from non-polar polymers).
Each channel directs the collected light through optical filters to a dedicated photomultiplier tube, converting the fluorescence signals into electrical pulses.
The resulting signals are digitized using a 16-bit data acquisition board. Custom MATLAB algorithms perform baseline correction, identify coincident events across both channels, and remove unwanted signals using pulse-duration thresholds of 10–120 μs. The software then calculates integrated pulse areas from both channels for ratiometric analysis.
Optical Sensor Achieves 77.8% Microplastic Classification Accuracy
Calibration tests demonstrated that the ratio of integrated pulse areas from the two detection channels can effectively distinguish different polymer types. DANS exhibited a redshift in fluorescence emission as the surrounding polymer environment became more polar.
Non-polar polypropylene (PP) and polyethylene (PE) produced emission peaks near 480 nm, with most of their signals detected in CH2. Polypropylene showed an almost vertical signal distribution, corresponding to a reference angle of 90.0°, while polyethylene had a fitted slope of 18.2 and an angle of 86.9°. Polystyrene (PS) produced an intermediate emission peak near 530 nm, with a fitted slope of 5.77 and an angle of 80.2°.
Polymethyl methacrylate (PMMA), polyvinyl chloride (PVC), and polyethylene terephthalate (PET) produced emission peaks between 560 and 580 nm, resulting in stronger signals in CH1. Their fitted slopes were 2.79, 2.40, and 1.84, respectively. These differences in fluorescence responses provided a basis for distinguishing the tested polymers.
Download a free copy of this page here!
Fivefold cross-validation using 816 detected particle events produced a classification accuracy of 77.8%. Recall was highest for PS (98.6%), PE (95.4%), and PP (93.2%).
In contrast, classification was less accurate for polar polymers, with recall rates of 41.1% for PMMA, 24.5% for PVC, and 59.6% for PET, partly due to overlapping fluorescence responses. Independent micro-Raman spectroscopy confirmed the polymers' identities.
However, the authors note the approach’s limited particle-count recovery numbers, ranging from 40% for PVC to 65% for PET, relative to the nominal concentrations of the prepared suspensions. The study therefore highlights the need to improve counting performance before the system can be reliably applied for environmental monitoring.
Low-Cost Sensor Shows Promise for Water Quality Monitoring
The SMILE sensor's compact, low-power design has significant implications for screening in municipal drinking water facilities, filtration plants, and industrial wastewater systems. By analyzing particles directly in flowing water at rates of hundreds of milliliters per hour, the platform reduces the need for multi-stage sample preparation and manual substrate transfers, which can lead to particle loss or contamination.
Its inline operation could support automated monitoring across water intake facilities and distribution networks, providing early warnings of microplastic concentrations and acute contamination events.
Researchers Target Improved Accuracy and Real-World Water Testing
The SMILE system demonstrates the potential of solvatofluorochromic fluorescence combined with dual-band optical detection for high-throughput microplastic screening in flowing liquids. The sensor uses fluorescence pulse-area ratios to identify polymer types while minimizing the influence of common optical noise sources.
Future work should test weathered, degraded, and biofouled polymers and expand the optical setup with additional channels to improve classification accuracy. Researchers also plan to develop machine learning (ML) models to enhance polymer discrimination and extract particle size, supporting continuous environmental monitoring.
Journal Reference
Nicolai, E., et al. (2026). A dual-channel optical sensor for real-time microplastic classification via ratiometric fluorescence analysis. Scientific Reports. DOI: 10.1038/s41598-026-78675-z, https://www.nature.com/articles/s41598-026-78675-z.
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.