The study, published in Food Science of Animal Resources, focused on acesulfame potassium, sodium saccharin, sucralose, and aspartame, four high-intensity sweeteners approved for use in Korea. Each is hundreds of times sweeter than table sugar, which means only tiny amounts are needed to deliver sweetness with little or no caloric load. But that very potency creates a regulatory challenge: because the effective doses are so small, food safety agencies need analytical tools sensitive enough to detect them at vanishingly low concentrations before they can verify that manufacturers are staying within legal limits.

The researchers turned to liquid chromatography coupled with tandem mass spectrometry, or LC-MS/MS, a technique that has become the gold standard for quantifying multiple compounds in complex food matrices. The method works by first separating the chemical components of a sample as they travel through a chromatography column, then fragmenting the molecules of interest and detecting them based on their characteristic mass-to-charge ratios. In this study, the team used a Thermo Scientific Vanquish UHPLC system coupled to a TSQ Quantis mass spectrometer, with chromatographic separation achieved on a C18 column using a mobile phase of ammonium acetate and acetic acid in water and methanol. Each analytical run took just 11 minutes, a significant improvement in throughput over earlier methods that sacrificed no accuracy in exchange for the speed.

One of the trickiest aspects of analyzing dairy products is that milk proteins and fats can interfere with the detection of target compounds, a phenomenon known as the matrix effect. To combat this, the researchers optimized a deproteinization step using two classic reagents: potassium ferrocyanide trihydrate and zinc sulfate heptahydrate, which together precipitate proteins out of solution. Through a systematic experiment involving ten different reagent volume combinations applied to a protein-rich thick fermented milk matrix spiked with known amounts of each sweetener, they identified the sweet spot: 0.6 milliliters of each reagent. Statistical analysis using one-way ANOVA followed by Duncan’s multiple range test confirmed that this combination maximized recovery rates for all four sweeteners simultaneously.

The validation results were striking. Calibration curves showed linearity with coefficients of determination of 0.9998 or better for all four sweeteners, far exceeding the US Food and Drug Administration’s minimum criterion of 0.995. Limits of detection ranged from just 0.002 milligrams per liter for acesulfame potassium to 0.103 milligrams per liter for sucralose, while limits of quantification spanned 0.007 to 0.312 milligrams per liter. Accuracy, measured through recovery experiments at low, medium, and high spike concentrations across all four dairy matrices, fell between 84.40 and 103.97 percent, and precision remained within a relative standard deviation of 6.67 percent. All of these figures comfortably satisfied the validation criteria set by both AOAC International and the International Council for Harmonisation.

Matrix effects, the perennial bogeyman of LC-based food analysis, proved remarkably benign in this method. Following European Commission SANTE guidelines, the team compared analyte responses in matrix-matched standards against solvent-only standards across fermented milk, processed milk, thick fermented milk, and lactic acid bacteria beverages. All four sweeteners showed matrix effects within plus or minus 10 percent, meaning the dairy matrix barely disturbed the signal at all. This allowed the researchers to use simple solvent-based calibration curves rather than more laborious matrix-matched calibration, streamlining routine analysis without compromising reliability.

To demonstrate that the method would hold up outside a single laboratory, the team conducted an inter-laboratory validation between two independent facilities. Thick fermented milk and lactic acid bacteria beverage samples were spiked with all four sweeteners at three concentration levels and analyzed in triplicate at both sites. Recovery rates ranged from 90.89 to 102.74 percent, and relative standard deviations stayed below 7.78 percent, both well within AOAC acceptance criteria for analytes in the 0.1 to 10 milligram per kilogram range. The researchers also calculated measurement uncertainty, accounting for contributions from stock solutions, sample preparation, calibration, and repeatability, and found expanded uncertainties between 2.74 and 10.97 percent across all matrices, comfortably below the Codex Alimentarius threshold of 22 percent.

With the method fully validated, the researchers applied it to 40 dairy products purchased from Korean retail stores, supermarkets, convenience stores, and online platforms between June and September 2024. The results painted a nuanced picture of sweetener use across the industry. Acesulfame potassium appeared exclusively in processed milk, detected in five of ten samples at concentrations between 120.82 and 161.76 milligrams per kilogram, well below the Korean Food Code limit of 0.50 grams per kilogram. Sucralose was the most ubiquitous, found in 88 percent of all samples across every category, with concentrations ranging from 14.53 to 172.96 milligrams per kilogram, all safely under the 0.40 grams per kilogram ceiling. Sodium saccharin, which is entirely prohibited in these products, was not detected in any sample, affirming regulatory compliance.

Aspartame told a more interesting story. It was found only in fermented milk, where five of ten samples tested positive at concentrations between 65.08 and 269.42 milligrams per kilogram. Because Korean regulations impose no specific numeric limit on aspartame, requiring only that it be used at the minimum level necessary to achieve its intended technical effect, these findings underscore the importance of continued surveillance. The researchers note that recent studies have raised questions about potential health implications of artificial sweetener consumption, including metabolic disturbances, alterations to gut microbiota, and possible links to cardiovascular disease, making reliable monitoring tools all the more essential for both regulatory compliance and health risk assessment.

The broader significance of this work lies in its completeness. Many analytical studies validate a method on paper but stop short of demonstrating real-world applicability or cross-laboratory reproducibility. By combining optimized sample preparation, thorough single-laboratory validation, formal uncertainty analysis, inter-laboratory confirmation, and a market survey of 40 commercial products, this study delivers a turnkey solution for food safety laboratories. As demand for reduced-sugar dairy products continues to grow globally, regulators and manufacturers alike now have a faster, more sensitive, and more robust tool for keeping the sweeteners in our yogurt and milk exactly where the law says they should be.

Subject of Research: Validation of an LC-MS/MS method for quantifying four artificial sweeteners in dairy-based foods and beverages

Article Title: Quantitative analysis, validation, and application of an LC-MS/MS method for four artificial sweeteners in dairy-based foods and beverages

Article References: Kim, K.-W., Kim, D., Kang, J., Yun, C.-I., & Kim, Y.-J. (2026). Quantitative analysis, validation, and application of an LC-MS/MS method for four artificial sweeteners in dairy-based foods and beverages. Food Science of Animal Resources, 46(1), Article 48. https://doi.org/10.1007/s44463-025-00010-3

Image Credits: AI Generated

DOI: 10.1007/s44463-025-00010-3

Keywords: artificial sweeteners, LC-MS/MS, dairy products, food safety, acesulfame potassium, sucralose, aspartame, sodium saccharin, method validation, matrix effects, Korean Food Code, mass spectrometry

A team of Vietnamese researchers led by Le Dinh Hung of the Institute of Public Health in Ho Chi Minh City, together with Phan Long Ho and colleagues at the University of Science, Ho Chi Minh City, Vietnam National University, and Ho Chi Minh City University of Education, has developed an analytical framework that explicitly accounts for the chemical composition of evaporative water residues when calculating gross alpha counting efficiency. The work, published in Environmental Geochemistry and Health, introduces the concept of an effective alpha-particle mass range, a quantity that describes how deeply alpha particles can penetrate a specific residue matrix before being absorbed, and integrates that quantity directly into closed-form efficiency equations that can replace the matrix-specific empirical calibration curves conventionally used in gas-proportional counting.

The core physical problem is self-absorption. Alpha particles are heavy, doubly charged helium nuclei that lose energy rapidly as they traverse matter, traveling only tens of micrometers in typical solids. When radionuclides are distributed throughout a dried residue layer on a planchet, particles emitted deep within the layer are stopped before they can escape toward the detector window. The fraction that escapes depends on the residue’s mass thickness, expressed in milligrams per square centimeter, and on its stopping power, which in turn depends on which elements compose the residue. A residue dominated by light elements such as calcium, carbon, and oxygen absorbs alpha particles differently than one dominated by sodium and chloride, yet most laboratories calibrate their counters with a single reference material, typically calcium sulfate dihydrate, and apply that calibration to samples of entirely different chemistry.

The new framework attacks this bias at its physical root. Building on classical descriptions of alpha-particle geometry and on the stopping and range calculations embodied in the SRIM code developed by Ziegler and colleagues, the researchers model the probability that an alpha particle emitted at a given depth within the residue can escape either directly toward the detector or after backscattering from the underlying planchet. They derive an analytical expression for the backscattering coefficient as a continuous function of alpha-particle energy and the mass number of the planchet material, homogenized from Monte Carlo-based correlations reported by Fernández Timón and Jurado Vargas. The escape probabilities for direct emission and backscattering are then integrated over the residue thickness, producing closed-form equations for counting efficiency in three distinct thickness regimes: an extremely thin region where both escape mechanisms operate across the whole layer, a transition region where a scattering dead zone emerges near the surface, and a thick region where the residue exceeds the full penetration range and efficiency falls off inversely with mass thickness.

The pivotal innovation is the effective mass range of the alpha particle in a compound matrix. Rather than treating the residue as a generic substance, the framework reconstructs its elemental composition from the water’s measured physicochemical properties, including total dissolved solids and major ion concentrations, and computes a weighted effective range that reflects the actual stopping power of the mixture. This means that a sodium chloride dominated residue from a saline coastal aquifer and a calcium carbonate dominated residue from a hard-water well are treated as physically distinct counting sources, each with its own efficiency curve, without requiring the laboratory to prepare new matrix-matched calibration standards for every sample type.

The validation was unusually thorough. The team compared the analytical model against independent experimental calibration datasets prepared with two very different matrices, calcium sulfate dihydrate and calcium carbonate, spanning the thin-source regime in which residue mass remains below 100 milligrams, corresponding to a mass thickness of up to about 5.2 milligrams per square centimeter in this study. The model predictions closely tracked the measured efficiency trends for both materials. The framework was then exercised against proficiency testing samples distributed by the International Atomic Energy Agency between 2021 and 2025, and every testing outcome satisfied the acceptance criterion of an absolute Z-score below 1.5, indicating that the calculated activities were statistically consistent with the reference values.

The most striking demonstration came when the method was applied to real evaporative residues from coastal groundwater samples. Because the researchers could reconstruct the elemental makeup of each residue, they discovered that the residues were composed primarily of sodium and chloride, chemically far removed from the calcium sulfate calibration standard that would normally be used. When gross alpha activities calculated with the analytical framework were compared against those derived from the conventional single-matrix calcium sulfate calibration, the two approaches diverged systematically, with discrepancies ranging from 12.05 to 23.46 percent. The largest difference appeared in the sample with the highest residue mass thickness, 3.99 milligrams per square centimeter, exactly where self-absorption effects are most pronounced. In other words, a laboratory relying on a standard calibration could underreport or overreport gross alpha activity in saline groundwater by more than a fifth, purely as an artifact of matrix mismatch.

That magnitude of bias matters for regulatory decisions. The World Health Organization’s drinking water guidelines, the European Council Directive 2013/51/Euratom, the United States Environmental Protection Agency Method 900.0, and Vietnam’s national technical regulation QCVN 01-1:2024/BYT all set screening thresholds for gross alpha activity in water intended for human consumption. A systematic error approaching 25 percent could push a compliant water source over a regulatory limit or, conversely, mask a genuine exceedance, with direct consequences for public health protection in regions with elevated natural radioactivity. Coastal aquifers, where seawater intrusion enriches groundwater in sodium and chloride and where dissolved solids can be high, are precisely the environments where the mismatch between calibration matrix and sample residue is most severe.

To make the method practical, the team has released both the complete source code of the computational tool on GitHub and an interactive web application built on the Streamlit platform, allowing any laboratory to compute matrix-corrected counting efficiencies from routinely measured water chemistry data. The framework also includes a full uncertainty propagation treatment, with closed-form expressions for the combined standard uncertainty of the counting efficiency in each thickness regime, accounting for uncertainties in the alpha-particle range, the physical absorber thickness including the air gap, detector window, and discriminator threshold, the residue mass thickness, and the effective backscattering coefficient itself.

The researchers emphasize that the framework is not a replacement for careful sample preparation but a physically grounded alternative to empirical calibration curves, one that treats residue composition as an input rather than an uncontrolled variable. Because the model requires only routine data such as total dissolved solids and major ion concentrations that most water quality laboratories already collect, adoption could be straightforward. The authors suggest that incorporating residue composition into efficiency calculations enhances the reliability of gross alpha activity determination and may facilitate radiological screening and water quality assessments in regions with elevated natural radioactivity. For the millions of people who depend on groundwater in coastal and granitic terrains worldwide, the study offers a quiet but consequential improvement: a more honest number at the foundation of every radiological safety decision.

Subject of Research: Matrix-dependent self-absorption effects on gross alpha activity determination in environmental water residues

Article Title: Matrix effects on gross alpha determination in environmental water residues: an analytical framework based on effective alpha-particle mass range

Article References: Hung, L. D., Ho, P. L., Minh, V. T., Van Anh, L. T., Vuong, L. Q., Minh, L. H., Loan, V. T. T., Linh, B. N. T., Thanh, T. T., & Van Tao, C. (2026). Matrix effects on gross alpha determination in environmental water residues: an analytical framework based on effective alpha-particle mass range. Environmental Geochemistry and Health, 48(15), Article 596. https://doi.org/10.1007/s10653-026-03492-2

Image Credits: AI Generated

DOI: 10.1007/s10653-026-03492-2

Keywords: gross alpha, self-absorption, groundwater, environmental radioactivity, gas proportional counting, counting efficiency, drinking water, matrix effects, alpha particles, water quality, IAEA proficiency testing, Vietnam