Label-Free Cell Sorting with SERS
High-resolution SERS mapping and deep learning enable label-free classification and microfluidic sorting of cancer and non-cancer cells.
A microfluidic cell sorter built around high-resolution SERS mapping and deep learning has separated cancer and non-cancer cells without fluorescent labels or Raman tags.
The researchers fabricated periodic ring-shaped gold nanostructures using laser near-field reduction, creating a SERS substrate with uniformly distributed plasmonic hotspots. Raman mapping showed a spatial resolution of around 185 nm and highly consistent enhancement across the substrate, allowing spectra to be collected from different regions within individual cells.
Because SERS spectra varied across different regions of the same cell, those high-resolution maps preserved local chemical information that would otherwise be averaged together. Signals from HeLa, A549, and NIH 3T3 cells reflected differences associated with proteins, lipids, and nucleic acids and were used to train a convolutional neural network for cell classification.
“This technique allows the creation of plasmonic ring-shaped nanostructure arrays by introducing cetyltrimethylammonium bromide into the precursor solution,” said Koji Sugioka in a press release.
Using the high-resolution substrate, the classifier reached 96.2 percent accuracy for label-free cell identification. With a more randomly distributed gold-nanoparticle substrate, accuracy fell to 75 percent, supporting the researchers’ argument that spatially resolved, reproducible SERS information was critical to classification performance.
As a proof of concept, the SERS substrate was integrated into a Y-shaped microfluidic chip. Cells identified from their Raman signal were directed into separate outlets using a magnetically actuated gate, enabling separation of HeLa cancer cells from NIH 3T3 fibroblasts without fluorescent biomarkers or Raman tags.
“The preliminary results indicate that the developed sorter achieves precise cell sorting and separation without the need for biomarker or Raman-tag labeling,” said Sugioka.
A Dual View of Fruit-Surface Contamination
A compact system aligns Raman and LIBS measurements to identify organic and inorganic contaminants on heterogeneous fruit surfaces.
Raman and laser-induced breakdown (LIBS) spectroscopy have been combined in a compact system that can distinguish pesticide residues from metal contamination on fruit surfaces.
The hybrid system (Hy-R-LIBS) aligns Raman and LIBS through a common optical path so both measurements probe the same region of the fruit surface. Raman is collected first to preserve the surface for molecular fingerprinting, before the higher-energy LIBS pulse ablates the spot to reveal elemental composition. Measurements were repeated across ten positions on each sample, reducing spatial mismatch between the two techniques on the heterogeneous fruit surface.
The researchers tested the approach on orange peel contaminated with the pesticides thiabendazole and thiram, alongside copper and zinc standards. Silver nanoparticles were deposited onto the peel to improve sensitivity in both channels. For Raman, the nanoparticles enhanced pesticide bands associated with characteristic molecular vibrations, while LIBS produced stronger atomic emission from the metal contaminants.
Raman distinguished thiabendazole from thiram through their vibrational fingerprints, while LIBS resolved characteristic copper and zinc emission lines. The two techniques also provided cross-checks: LIBS detected sulfur associated with thiabendazole, while Raman identified the pesticide through its C=N stretching bands. Detection limits reached around 0.05 ppm for the strongest features.
In a mixed sample containing both thiabendazole and zinc, the system recovered the expected molecular and elemental signatures within the same measurement sequence. The authors present Hy-R-LIBS as a proof of principle, with performance beyond the four tested analytes still to be established.
Recycling Starts with the Container
Terahertz spectroscopy links container geometry to polymer-classification errors and guides designs that are easier to sort for recycling.
A terahertz (THz) spectroscopy study has linked packaging geometry to errors in polymer identification, then used those findings to guide the design of containers that are easier to sort for recycling.
The researchers focused on polyethylene terephthalate (PET) and polystyrene (PS), two visually similar plastics that can be difficult to separate by appearance alone. In a community collection experiment, 383 used containers were analyzed by continuous-wave terahertz spectroscopy at 75, 95, and 100 GHz, with a pre-trained model assigning each sample to PET or PS. ATR-FTIR was used to confirm the material identity.
The field test showed that geometry contributed to classification errors. Ribs, slopes, curvature, and deformation altered THz transmittance, with the effect strongest under S-polarized measurements at 45° incidence.
“We wanted to rethink packaging design so that recyclability becomes part of the design process while everyday usability and practical requirements are retained,” said lead author Juniya Yoshihara in a recent press release.
The team then tested prototype container bottoms with different geometries and found that a flat central sensing region gave more stable THz readings, while surrounding grooves could still preserve practical functions such as draining water away from soba noodles. The selected design maintained stable transmittance while remaining stackable and easy to handle.
“By integrating sensing considerations into product geometry, we can explore packaging that is easier to identify after disposal without giving up its familiar functions,” said Yoshihara.
The work suggests that more reliable THz sorting may depend not only on better sensing and classification, but also on designing packaging that presents a more consistent measurement surface.
What’s in the Toy Box?
XRF screening of 333 toys sold along the US–Mexico border finds potentially toxic elements in 78 percent of samples.
A binational X-ray fluorescence screening of 333 children’s toys sold in El Paso and Ciudad Juárez found potentially toxic elements in nearly four out of five samples, with jewelry emerging as the highest-risk product category.
The researchers collected toys from major retail centers on both sides of the US–Mexico border over 11 months, using sampling proportional to the local population of children under 14. X-ray fluorescence (XRF) spectroscopy provided a rapid, non-destructive way to screen the products for lead, cadmium, arsenic, chromium, nickel, zinc, and titanium.
“Children are not small adults. Their bodies take in heavy metals more readily than ours do,” said senior author Jorge Gardea-Torresdey in a recent press release. “And because their organs and detoxification systems are still developing, the same exposure that an adult might tolerate can cause far greater harm.”
Overall, 78 percent of the toys contained at least one element of concern, while 12 percent contained three or more. Forty samples also contained concentrations above limits referenced in US or Mexican toy-safety standards, with products purchased in Ciudad Juárez around three times more likely to exceed those thresholds than those bought in El Paso. Lead was most notable in inexpensive jewelry, while arsenic and cadmium were detected only in Ciudad Juárez samples.
“A bracelet or a pair of earrings can cross an international line in an afternoon,” said Gardea-Torresdey. “Our results make the case for harmonized safety standards, better labeling and continued monitoring to protect the roughly 565,000 children who call this binational community home.”
Because XRF measures total elemental content rather than metal release during use, the findings identify products for follow-up rather than confirming exposure or regulatory violation. Planned leaching studies will test which toys actually release metals under conditions intended to mimic contact with saliva or sweat.
The Spectacular and Strange
An Ancient Recipe for Iridescence
More than a millennium ago, Abbasid potters were engineering metallic surfaces before nanotechnology had a name. A new study shows how silver and copper nanoparticles helped give 9th-century Islamic ceramics their shimmering greens, ochres, browns, and reds.
The researchers studied six Abbasid luster shards using UV-Vis spectroscopy, micro X-ray absorption spectroscopy, micro X-ray diffraction, and electron microscopy. The goal was to connect each color with the chemistry and structure of the thin luster layer beneath the glaze surface.
“Today we still find it very difficult to reproduce the effects they did in the 9th century, so we wanted to find out what chemical transformation the painting applied on the ceramics went through to create such effects,” said Trinitat Pradell, professor at the Universitat Politècnica de Barcelona and co-corresponding author of the study, in a press release.
Luster decoration began with a paint rich in silver and copper compounds applied to an already glazed ceramic. During a second firing, metal ions diffused into the glaze and were reduced, forming metallic nanoparticles embedded just below the surface rather than simply deposited on top. The balance between silver and copper proved crucial: silver-rich layers produced green lusters, copper-rich layers produced red and red-brown colors, and intermediate compositions gave ochre and brown tones.
The study also points to the delicacy of the firing process. The luster had to be fired between the glass transition and softening temperatures, allowing ions to move through the glaze without causing the paint to stick. Future mock-up samples could help isolate the effects of individual manufacturing parameters, including the roles of iron and tin.
“This study was carried out by analyzing tiny fragments from historical ceramics,” said Marine Cotte, scientist at the ESRF and co-corresponding author. “The next step will be to produce and analyze mock-up samples to distinguish the effects of each manufacturing parameters.”
Newsletters
Receive the latest analytical science news, personalities, education, and career development – weekly to your inbox.
