# Optical void sensor distinguishes model nanoplastics by shape and material

> Nanoscale cavities in gallium arsenide trap model plastic particles and report changes in reflected colour. The study combines shape sorting with polymer-sensitive optical signals; testing real environmental and biological samples is proposed as future work.

URL: https://foodradar.org/optical-nanoplastic-sensor-model-particles-and-validation-limits
Language: en
Author: FoodRadar Editorial
Published: 2026-10-19T06:00:10.000Z
Updated: 2026-10-25T01:29:13.706Z
Source publication date: 2026-08-07
Category: Food safety
Türkçe: https://foodradar.org/tr/optical-nanoplastic-sensor-model-particles-and-validation-limits
Español: https://foodradar.org/es/optical-nanoplastic-sensor-model-particles-and-validation-limits
Deutsch: https://foodradar.org/de/optical-nanoplastic-sensor-model-particles-and-validation-limits
Italiano: https://foodradar.org/it/optical-nanoplastic-sensor-model-particles-and-validation-limits
Русский: https://foodradar.org/ru/optical-nanoplastic-sensor-model-particles-and-validation-limits

## Nanoscale cavities provide both sorting and colour readout

Researchers at the universities of Stuttgart and Bayreuth have demonstrated an optical sensing platform that distinguishes model nanoplastic particles using their shape and material-dependent colour response. Reported in Nature Communications on 7 August 2026, the study uses nanoscale cavities, called Mie voids, in gallium arsenide. A trapped particle changes the optical response of the cavity, producing a change in reflected colour.

Circular and elliptical voids serve as both trapping sites and optical reporters. The experiments use optical microscopy and camera images, with colour analysis and, for selected comparisons, controlled illumination polarisation. Scanning electron microscopy provides corresponding views of the particles and their positions in the voids. Single-polymer reference measurements support the assignment of the measured colours.

## Reference measurements distinguish the model polymers

The first material comparison used polystyrene, or PS, and poly(methyl methacrylate), or PMMA, particles with a nominal diameter of 300 nm. Dynamic light scattering gave mean diameters of 363 nm for PS and 374 nm for PMMA, averaged over three measurements. Voids containing these particles produced distinct colour groups alongside the empty voids. In a separate example, a dirt particle produced an outlying colour response.

The later mixture experiment combined spherical PS, PMMA and PET particles with elongated PS particles. The PET particles were made with PS seeds and contained at least 75% PET; their exact composition was unknown and was not explicitly represented in the simulations. The measured mean diameter of this PET particle population was 303 nm. Separate reference measurements were used for the final mixture, and the colours of circular and elliptical voids were analysed separately because an empty elliptical void could overlap in colour with a PMMA-filled circular void.

## Shape and polarisation affect the optical signal

The researchers compared spherical particles with elongated PS particles designed with aspect ratios of 2:1, 3:1 and 4:1. Matching void geometry and subsequent cleaning contributed to selective trapping. In one experiment, 11 of 7,500 circular voids retained an elongated particle tip-first, corresponding to 0.15%. Reference measurements showed a colour difference between these cases and circular voids containing spherical particles.

Colour representation and illumination also mattered. Four empty or PS-filled circular and elliptical void states formed distinct groups in RGB space, while the CIE representation did not distinguish all four groups. For elliptical voids, changing the polarisation altered the colour separation between empty and filled states. The supplementary controls showed negligible changes for empty round voids and for ideally filled round voids under polarisation variation. Experimental images were white-balanced against bare gallium arsenide before each image, whereas the simulations used an ideal white reference.

## Real-sample measurements remain a proposed next step

The demonstrated mixtures used prepared model particles. In their discussion, the authors propose further optimising void geometry to match particles and enhance the optical resonances, and exploring surface functionalisation for more complex environments. They identify measurements with real environmental and biological samples, including blood specimens, as a future way to validate system performance.

The authors envisage test strips and mobile detection devices for large-scale nanoplastic screening. This is their proposed direction for development. When discussing earlier Mie-void approaches in the introduction, they state that low-cost field operation by non-specialists requires further research.

## Sources and rights

- [All-optical Mie void sensor for complex nanoplastic mixtures](https://doi.org/10.1038/s41467-026-76199-8?utm_source=foodradar.org&utm_medium=referral&utm_campaign=editorial&utm_content=optical-nanoplastic-sensor-model-particles-and-validation-limits-en). Nature Communications. Dominik Ludescher, Julian Schwab, Serkan Arslan, Evelyn Kubacki, Monika Ubl, Markus Retsch, Harald Giessen and Mario Hentschel (2026). All-optical Mie void sensor for complex nanoplastic mixtures. Nature Communications 17, 8010. DOI: 10.1038/s41467-026-76199-8. © The Author(s) 2026. CC BY 4.0. Adapted, edited and translated by FoodRadar. [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/?utm_source=foodradar.org&utm_medium=referral&utm_campaign=editorial&utm_content=optical-nanoplastic-sensor-model-particles-and-validation-limits-en-license).

Adapted by FoodRadar.

[Editorial standards](https://foodradar.org/editorial-policy)