The research, published in the Journal of Agriculture and Food Research, focused on Camellia sinensis, the species behind green, black, oolong, and white teas. While previous studies have largely examined aroma in processed tea, the team led by Xinxin Lv and colleagues at Chinese tea research institutions turned the spotlight on fresh shoots, plucked at a uniform one-bud-one-leaf standard from the National Germplasm Hangzhou Tea Repository. Because all 20 cultivars grew under identical field conditions and were harvested on the same two days in March 2024, any chemical differences could be attributed to genetics rather than environment or handling.

The analytical centerpiece was headspace solid-phase microextraction coupled with gas chromatography and tandem mass spectrometry, a technique that captures the airborne molecules released by a sample and separates them with high sensitivity. From just half a gram of freeze-dried leaf powder, the method detected 65 volatile metabolites spanning nine chemical classes. Alcohols and esters were the most numerous, each accounting for about a quarter of identified compounds, followed by terpenes and aldehydes. When the researchers averaged relative abundance across all cultivars, terpenes dominated the volatile pool at 61.4 percent, with esters contributing 24.4 percent and alcohols 9.8 percent.

That average, however, concealed striking variety-specific patterns. Cultivars such as JMD, BYTZ, and EC7H showed terpene proportions exceeding 60 percent and in some cases 70 percent, while GH3H was distinguished by an unusually high ester fraction of 44.13 percent. FDDB stood out for its alcohol content above 16 percent. To disentangle these differences statistically, the team applied orthogonal partial least squares-discriminant analysis, a multivariate method that separates samples along axes of maximum group distinction. The model performed well, explaining 86.5 percent of the variance in the volatile data and achieving a predictive measure of 0.82, and a 200-iteration permutation test confirmed it was not overfitted.

The analysis identified 16 discriminant volatile metabolites with variable importance scores above 1, meaning they contributed disproportionately to separating the cultivars in the statistical space. Twelve of these had authentic reference standards and published odor thresholds available, allowing the researchers to move from relative to absolute quantification. They then calculated odor activity values, or OAVs, by dividing each compound’s measured concentration by its odor threshold in water. A compound with an OAV above 1 is present at a level that could, in principle, be perceived by the human nose.

All 12 compounds cleared that bar in every cultivar, but their magnitudes varied enormously. Linalool, a floral monoterpene with an exceptionally low odor threshold of 0.22 micrograms per liter, posted OAVs between roughly 285,000 and 1.09 million, the highest of any compound measured, peaking in cultivar QM809. Geraniol, another rose-scented terpene, reached an OAV above 151,000 in JMD, which also led for beta-myrcene, a citrus-and-herbal compound. SFX recorded the highest OAV for (Z)-hex-3-enyl acetate, a green-fruity ester, while JGY topped the list for 1-octen-3-ol, a herbal compound. Each cultivar, in other words, carried a distinct fingerprint of potentially odor-active molecules.

Grouping the 12 compounds by their reported odor descriptors revealed four aromatic themes. Geraniol, phenethyl alcohol, linalool, indole, nerol, and trans-nerolidol clustered around floral notes; 1-octen-3-ol and methyl salicylate defined herbal territory; (E)-hex-3-en-1-ol and (Z)-hex-3-enyl acetate represented green and fresh scents; and beta-myrcene and benzyl alcohol fell into the fruity sector. Within the floral group, accumulation patterns diverged sharply: linalool was abundant everywhere, exceeding 250 micrograms per gram in all cultivars and approaching 958 micrograms per gram in QM809 and EC7H, whereas geraniol concentrated in JMD at 604 micrograms per gram, indole peaked above 30 micrograms per gram in GH3H, and trans-nerolidol was highest in LJCY.

To make sense of these patterns, the authors mapped each compound onto established plant biosynthetic routes. Five of the differential volatiles were terpenoid-derived: linalool, geraniol, nerol, and beta-myrcene trace back to the plastidial MEP pathway, while the sesquiterpene trans-nerolidol arises from the cytosolic mevalonate pathway, with terpene synthase enzymes assembling the final structures. Four compounds, including phenethyl alcohol, benzyl alcohol, methyl salicylate, and indole, derive from phenylalanine or tryptophan metabolism, and three, the green-leaf volatiles, originate from the lipoxygenase-mediated oxidation of linoleic and linolenic acids. The lack of synchronized variation within each class suggested that cultivars differ not simply in how much of a pathway they run, but in which specific branches dominate.

As an exploratory structural complement, the team performed molecular docking, computationally fitting each volatile into homology-modeled structures of biosynthesis-related proteins including terpene synthase, lipoxygenase, and salicylic acid methyltransferase. Predicted binding scores ranged from minus 3.9 to minus 6.1 kilocalories per mole, with nerol paired to terpene synthase showing the strongest predicted interaction, followed by trans-nerolidol and geraniol with the same enzyme class. The authors were careful to frame these results as hypothesis-generating: docking scores reflect a computational scoring function rather than measured binding energies, and no enzyme-activity or substrate-conversion experiments were conducted. The models identify plausible ligand-protein contacts, mostly hydrophobic with some hydrogen bonding, but functional relevance awaits biochemical validation.

The practical implications reach from breeding programs to tea factories. Because volatile profiles shift substantially during withering, fixation, and drying, the fresh-leaf OAV patterns documented here represent the raw aromatic potential that processing then transforms. Knowing which cultivars are naturally rich in linalool versus geraniol, or in green esters versus herbal phenolics, gives breeders a chemical basis for matching germplasm to green, black, or oolong manufacturing, and gives processors a predictive handle on what a given batch of fresh leaves can deliver. The study also fills a genuine gap: systematic comparisons of fresh-leaf volatiles across diverse Chinese cultivars, combining broad profiling, targeted quantification, and odor-relevance screening, have been rare. As the authors note, the dataset provides a foundation for subsequent sensory, processing, and mechanistic work, and a reminder that the character of a great tea begins in the chemistry of the untouched leaf.

Subject of Research: Cultivar-associated volatile metabolite profiles and aroma chemistry of fresh Camellia sinensis leaves

Article Title: Unraveling the cultivar-associated aroma profiles of Camellia sinensis via untargeted and targeted volatile metabolomics coupled with molecular docking

Article References: Lv, X., Liu, P., Zhang, F., Chen, M., Yang, L., Zhu, X., Wang, J., Zhang, G., Yuan, H., Yang, Y., & Hua, J. (2026). Unraveling the cultivar-associated aroma profiles of Camellia sinensis via untargeted and targeted volatile metabolomics coupled with molecular docking. Journal of Agriculture and Food Research, Article 103368. https://doi.org/10.1016/j.jafr.2026.103368

Image Credits: AI Generated

DOI: 10.1016/j.jafr.2026.103368

Keywords: tea, Camellia sinensis, volatile metabolomics, aroma compounds, linalool, geraniol, odor activity value, terpene synthase, molecular docking, tea cultivars, HS-SPME-GC-MS, food chemistry

The premise is audacious. Flavor, after all, is a chemical phenomenon, governed by hundreds of volatile compounds that reach the nose and non-volatile molecules that land on the tongue. Nutritional content, meanwhile, is measured through wet chemistry: protein by Kjeldahl or combustion methods, fat by solvent extraction, carbohydrate by difference or enzymatic assay. None of these quantities would seem to be encoded in the pixels of a photograph. Yet the researchers reasoned that a meal’s appearance carries indirect but systematic information about its composition, from the browning of seared meat to the sheen of rendered fat to the visible proportion of sauce, vegetables, and grain in a tray.

To ground the model in chemical reality, the team first performed a rigorous flavor chemistry characterization of 32 ready-to-eat meal variants spanning pork-, chicken-, and beef-based products. Using headspace solid-phase microextraction coupled with gas chromatography-mass spectrometry, or HS-SPME/GC-MS, they identified 77 volatile flavor compounds across the product set. From these, they screened 22 key odor-active compounds using the odor activity value, a metric that divides a compound’s concentration by its sensory detection threshold; compounds with a value of one or higher are considered genuinely perceptible to human noses. Aldehydes and spice-derived terpenes emerged as the major contributors separating the products from one another, a chemically sensible finding since aldehydes typically arise from lipid oxidation and the Maillard reactions of cooking meat, while terpenes trace back to seasonings such as pepper, star anise, and citrus-derived spices.

This chemical ground truth matters because it anchors what the neural network ultimately learns. Rather than asking the model to guess abstract labels, the researchers tied the image data to measured flavor profiles and physicochemical nutrient indices, allowing the network to internalize the correlations between what a dish looks like and what it contains. The architecture at the heart of the system is the Transformer, the same attention-based design that revolutionized natural language processing. Transformers excel at weighing which parts of an input matter most in relation to every other part, and in a multimodal setting they can fuse heterogeneous streams of information, visual features drawn from images alongside structured chemical and physicochemical descriptors, into a single coherent representation.

The headline result is a classification accuracy of 91.5 percent in distinguishing the 32 meal variants from images alone. For a task that requires separating closely related dishes of the same protein base, that figure suggests the model is picking up on subtle visual cues rather than trivial differences. More demanding still were the regression targets: predicting actual macronutrient content. After an optimization step using a weighted-Huber loss, a robust objective that limits the influence of outlier samples and weights the training signal across classes, the model achieved determination coefficients of 0.86 for protein, 0.80 for fat, and 0.79 for carbohydrate. In practical terms, an R-squared near 0.8 means the image-based predictions explain roughly four-fifths of the variance observed in conventional physicochemical measurements, a level of agreement the authors describe as high consistency with laboratory reference values.

The consumer dimension of the study adds a commercial edge. In a panel of 100 participants, flavor attributes dominated purchase decisions, confirming the industry intuition that taste, not nutrition labels, drives what shoppers actually put in their baskets. Critically, the trained model was able to identify the core flavor descriptors of the products, effectively translating pixels into sensory language. That capability points toward a workflow in which a quality-control engineer photographs a finished dish and receives, within seconds, an estimate of its macronutrient profile and its expected flavor character, without opening a package or sending a sample to a lab.

The implications for centralized kitchens and ready-to-eat meal enterprises are considerable. Modern meal production operates at enormous scale, with single facilities turning out tens of thousands of portions daily. Batch-to-batch variation in raw materials, cooking time, and sauce distribution can shift both flavor and nutrition in ways that periodic lab sampling catches only after the fact. An image-based screening tool could function as a continuous, non-destructive checkpoint on the production line, flagging trays whose appearance deviates from the expected profile before they are sealed, shipped, and reviewed by consumers. Because the method requires only a camera and a trained model, its marginal cost per inspection is close to zero, in contrast to GC-MS runs that require expensive instrumentation, skilled operators, and consumable reagents.

The authors also sketch a consumer-facing trajectory for the framework. A smartphone application that estimates macronutrients and calories from a photo of one’s plate is not a new idea, but existing systems typically rely on generic food recognition and portion heuristics. A model trained on systematic flavor chemistry could go further, offering personalized meal recommendations that account for both nutritional needs and flavor preferences, the two axes the consumer panel showed jointly govern acceptance. For individuals managing diabetes, athletic training diets, or simply the daily question of what to eat, an AI that understands both what a dish contains and what it will taste like represents a meaningful step beyond calorie counting.

Still, the study’s scope invites caution. The model was trained and validated on 32 variants of pork-, chicken-, and beef-based ready-to-eat meals, a deliberately bounded product space. Generalizing to the sprawling diversity of global cuisine, with its vast range of plating styles, lighting conditions, and ingredient combinations, will require far larger and more varied datasets. Image-based prediction is also inherently indirect: the model infers chemistry from appearance, so it can be fooled by plating choices, sauces that conceal the food beneath, or lighting that alters perceived browning. The authors position the framework as an AI-assisted quality-control tool rather than a replacement for definitive laboratory analysis, a framing that acknowledges these limits while underscoring the practical value of rapid screening.

What makes the work conceptually significant is its demonstration that flavor, often treated as the least quantifiable of food attributes, can be folded into a multimodal learning pipeline alongside nutrition. By pairing rigorous HS-SPME/GC-MS characterization and odor activity analysis with modern deep learning, the researchers built a bridge between sensory chemistry and computer vision, two fields that rarely share a vocabulary. If subsequent studies extend the approach across broader food categories and real-world imaging conditions, the humble photograph could become a standard instrument of food science, one that judges a meal the way a hungry, discerning customer would: by how it looks, and by what that appearance promises about taste and nourishment.

Subject of Research: Transformer-based multimodal deep learning for image-based synchronous evaluation of flavor and nutritional quality in ready-to-eat meals

Article Title: Transformer-based multimodal deep learning and its application in flavor-nutrition synchronous evaluation

Article References: Qi, W., Ding, M., Zhang, Y., Wang, B., Shen, C., & Liu, D. (2026). Transformer-based multimodal deep learning and its application in flavor-nutrition synchronous evaluation. npj Science of Food. https://doi.org/10.1038/s41538-026-01186-8

Image Credits: AI Generated

DOI: 10.1038/s41538-026-01186-8

Keywords: deep learning, Transformer, multimodal AI, flavor chemistry, food nutrition, ready-to-eat meals, GC-MS, volatile compounds, odor activity value, food image recognition, quality control, macronutrient prediction

Keemun black tea, produced from the Camellia sinensis cultivar Qimen in Anhui province, is prized for its pronounced floral, fruity and sweet notes, a profile that has earned it a place among the most famous black teas on Earth. Its manufacture follows a strict sequence of withering, rolling, fermentation and drying. While withering and fermentation have attracted considerable research attention, the drying step, despite being the final thermal process that locks in color, taste and aroma, has remained surprisingly underexplored. Previous studies largely focused on drying temperature or variable temperature profiles, leaving the role of drying time itself largely unquantified.

To close that gap, the research team, led by Erdenechimeg Otgonbat and colleagues, sampled tea directly from a commercial-scale hot-air dryer operating at 110 degrees Celsius. Subsamples of roughly 100 grams were pulled from three separate grids of the drying bed before drying and after 10, 15, 20, 25 and 30 minutes, providing biological replicates that eliminated micro-environmental gradients inside the machine. Crucially, every sample was immediately plunged into liquid nitrogen and freeze-dried. This lyophilization step served two purposes: it standardized all samples to a uniform dry weight, because headspace partitioning of volatiles is exquisitely sensitive to moisture, and it halted residual enzymatic activity and thermal reactions, preserving a genuine chemical snapshot of each moment.

The volatiles were then captured using headspace solid-phase microextraction, a solvent-free technique, followed by gas chromatography-mass spectrometry. In total, the team identified 104 volatile compounds spanning alcohols, aldehydes, esters, alkenes, ketones, alkanes and heterocyclics. Alcohols dominated the profile, accounting for between 66 and 71 percent of total volatiles, with aldehydes close behind. The total volatile content followed a strikingly non-linear trajectory: it fell from 18,148 micrograms per kilogram before drying to a low of 14,623 at 15 minutes, then surged to 16,856 at 20 minutes before settling near 15,679 at the end of the process.

That transient spike at 20 minutes is one of the study’s most intriguing observations. The researchers hypothesize that it reflects a dynamic equilibrium between physical evaporation, which strips volatiles away in the hot-air stream, and thermal chemical generation, which creates new aroma molecules. As heat accumulates, non-enzymatic cascades such as the Maillard reaction and Strecker degradation of amino acids churn out fresh aldehydes and heterocycles, while thermal hydrolysis cleaves non-volatile glycosidically bound precursors, liberating a wave of free linalool and geraniol derivatives. For a brief window, the rate of generation outpaces the rate of loss. Beyond 20 minutes, precursors become depleted and the balance tips back toward evaporation. The authors caution that this mechanism remains a working hypothesis requiring isotope-labeling studies to confirm.

Sensory evaluation, conducted under China’s national standard for tea assessment with a panel of seven trained assessors, corroborated the chemistry. As drying progressed, the infusion color deepened from bright orange-yellow to darker orange-red, while green and astringent notes faded completely between 20 and 30 minutes. By the end of drying, the tea reached its peak sensory quality, displaying the classic high sweet and floral Keemun aroma alongside a bright red liquor. The phenotypic trajectory of the panel aligned neatly with the molecular data, giving the chemical findings a human anchor.

To identify which compounds actually mattered, the team deployed partial least squares discriminant analysis, validated with a 999-permutation test to rule out overfitting. This chemometric screen flagged 20 differential volatile markers that reliably distinguished the drying intervals. A subsequent odor activity value analysis, which compares each compound’s concentration against its odor threshold in water, narrowed the field to 39 aroma-active compounds, of which six emerged as the primary drivers of Keemun’s signature scent: trans-beta-ionone, linalool, geraniol, benzeneacetaldehyde, nonanal and theaspirane.

The dynamics of these key odorants reveal two competing chemical stories. Trans-beta-ionone and related C13-norisoprenoids surged during the 20-to-30-minute window, direct evidence of thermal oxidative cleavage of carotenoid pigments such as beta-carotene and neoxanthin, which fragment into sweet-floral and sweet-woody molecules once thermal energy overcomes the activation barrier under low-moisture conditions. Meanwhile, the monoterpene alcohols linalool and geraniol fluctuated as residual glycoside-hydrolyzing enzymes released them early in drying, only for desiccation to inactivate those enzymes past 20 minutes, letting volatilization win. Most tellingly, benzeneacetaldehyde, the honey-like note, accumulated late in drying through Strecker degradation of the amino acid phenylalanine within the Maillard reaction cascade, confirming that drying acts not merely as dehydration but as a thermal-chemical reactor.

The practical implications extend beyond academic curiosity. Because the most critical accumulation of Keemun’s key odorants occurs in the middle-to-late drying stages, manufacturers now have a quantitative basis for tuning drying time as a precision parameter alongside temperature, potentially enabling smarter, aroma-targeted control in tea production. The authors acknowledge limitations, noting that absolute quantification could be cross-validated with solvent-assisted flavor evaporation and that future multi-omics work should trace precursor degradation kinetics. Still, the study delivers something rare in food science: a minute-by-minute molecular biography of a flavor in the making, showing exactly when a leaf stops smelling green and starts smelling like Keemun.

Subject of Research: Time-dependent changes in volatile and aroma compounds during the hot-air drying of Keemun black tea

Article Title: Time-dependent changes of aroma and volatile compounds during drying of Keemun black tea

Article References: Otgonbat, E., Cui, J., Luo, Z., Song, Y., Gao, T., & Song, C. (2026). Time-dependent changes of aroma and volatile compounds during drying of Keemun black tea. Food Chemistry: X, Article 104531. https://doi.org/10.1016/j.fochx.2026.104531

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104531

Keywords: Keemun black tea, tea drying, aroma chemistry, volatile compounds, Maillard reaction, Strecker degradation, HS-SPME-GC/MS, linalool, trans-beta-ionone, odor activity value, food chemistry, sensory evaluation

Superheated steam is not a new idea. It is produced by heating ordinary steam above its saturation point, so that instead of condensing into droplets when it touches a cooler surface, it remains a dry, high-energy gas. That distinction matters enormously in food processing. Saturated steam delivers moisture along with heat, which can leave surfaces wet and hospitable to microbes. Superheated steam, by contrast, transfers intense heat while simultaneously dehydrating the surface, and it does so in an environment nearly devoid of oxygen. For oxygen-sensitive foods, that combination is potentially ideal: rapid microbial kill, suppressed oxidation, and no added water. Until now, however, most applications have focused on drying and on decontaminating low-moisture foods such as spices, nuts, and food-processing equipment, leaving a significant gap in understanding how the technology treats delicate, high-moisture, ready-to-eat produce.

The research team, led by Jingyu Gao and colleagues including Boran Yang and Xin Wen, sourced Fuji apples from Yantai in Shandong Province and cut them into uniform pieces roughly one centimeter thick. The pieces were briefly dipped in a dilute sodium chloride solution to limit browning during preparation, then subjected to one of five steam treatments: conventional saturated steam at 100 degrees Celsius, or superheated steam at 150, 200, 250, or 300 degrees Celsius, each for exactly ten seconds. The treated samples were sealed in sterile bags and stored at 4 degrees Celsius, with detailed measurements taken on days one, three, five, seven, and nine. The parameters tracked ranged from the routine, such as microbial counts, color, and firmness, to the sophisticated, including enzyme activities, oxidative stress markers, total phenolics, and a full gas chromatography-mass spectrometry analysis of volatile aroma compounds.

The microbial results were unambiguous. Bacteria, molds, and yeasts all multiplied over time in every group, but the hotter the superheated steam, the slower the growth. After nine days, samples treated at 300 degrees Celsius carried the lowest total bacterial count of any group, at 4.61 log colony-forming units per gram, and the counts in the 200, 250, and 300 degree groups were reduced by roughly one log unit compared with the saturated steam control. The effect on molds and yeasts was even more dramatic: reductions of up to 2.85 log units relative to the control. The authors attribute this to the greater sensible heat carried by higher-temperature steam, which damages microbial proteins and membranes within the brief exposure window, combined with the dry, non-condensing environment that denies microbes the surface moisture they need to thrive.

Enzyme activity told a more nuanced story. Polyphenol oxidase and peroxidase are the two principal enzymes behind enzymatic browning in cut apples, and controlling them is central to preserving visual appeal. Peroxidase responded predictably: treatments at 200 degrees Celsius and above kept its activity significantly lower than the control throughout storage, with the hottest treatment producing the strongest and most sustained inhibition. Polyphenol oxidase, however, behaved in a strikingly non-linear way. At 250 degrees Celsius, its activity actually rose above that of the control, apparently because intermediate temperatures partially disrupt the plastid membranes that sequester the enzyme from its substrates, releasing it into an active state. At 300 degrees Celsius, the extreme heat denatured the released enzyme outright, restoring strong inhibition. This dual effect, the authors note, underscores the importance of carefully optimizing superheated steam conditions rather than assuming that more heat is always better.

Curiously, these enzyme shifts did not translate into visible color differences. Lightness values, total color difference, and the browning index all evolved gradually during storage, and while the steam-treated groups consistently showed numerically lower browning indices than the control, the differences never reached statistical significance. The authors point out that browning development in fresh-cut produce is not always tightly correlated with enzyme activity, and that the enzyme changes induced by a ten-second treatment may simply have been too modest to alter the overall color trajectory. Weight loss was similarly unaffected, and firmness remained stable across all groups during storage, although after nine days the steam-treated samples were slightly softer than the control, hinting that even brief high-temperature exposure can nudge cell wall structures toward degradation.

Where the study delivers its most commercially relevant surprise is in aroma chemistry. Using headspace solid-phase microextraction coupled with gas chromatography-mass spectrometry, the team identified 41 volatile compounds, dominated by esters, the chemicals responsible for the fruity, sweet character of apples. Esters, alcohols, and aldehydes together accounted for more than 99 percent of the total volatile content. After nine days, every superheated steam group retained more total volatiles than the saturated steam control, and the 150 degree treatment consistently preserved the highest levels throughout storage. The relative proportion of esters in the treated samples exceeded 49.5 percent by day nine, compared with just 40 percent in the control, whose volatile profile had shifted toward aldehydes associated with green, grassy notes. In other words, the conventional steam treatment left the apples smelling less like ripe apples and more like cut grass.

To translate chemistry into perception, the researchers calculated odor activity values, which weigh each compound’s concentration against its sensory threshold. Ten aroma-active compounds emerged, led by ethyl 2-methylbutyrate, 2-methylbutyl acetate, and hexanal. Every superheated steam group maintained higher total odor activity values than the control across storage, and the 150 degree group peaked at 1217.35 on day three, nearly double the control’s value. The 300 degree group, despite its superior antimicrobial performance, retained fewer aroma compounds than its milder counterparts, suggesting that excessive heat accelerates the thermal degradation of volatile molecules. The authors caution that these conclusions rest on instrumental analysis alone and that formal sensory panels will be needed to confirm what consumers would actually perceive.

Weighing all the evidence, the team identified the 250 degree Celsius, ten-second treatment as the best overall compromise. It delivered effective microbial control and sustained peroxidase inhibition, kept superoxide anion formation rates relatively low, and avoided the excessive aroma losses seen at 300 degrees, even though polyphenol oxidase activity remained elevated. The authors acknowledge the limitations of their single fixed exposure time and call for future work on temperature-time combinations, sensory validation, and pilot-scale trials addressing treatment uniformity, throughput, and energy consumption. Still, the study makes a compelling case that superheated steam deserves a place in the fresh-cut conversation. In an industry long reliant on antibrowning dips and modified atmosphere packaging, a chemical-free, oxygen-free, ten-second heat pulse that simultaneously tames microbes and protects the fruit’s signature fragrance is a proposition worth taking seriously.

Subject of Research: Effects of short-time superheated steam treatment on quality and aroma-active volatiles in fresh-cut Fuji apples during refrigerated storage

Article Title: Beyond microbial control: Effects of short-time superheated steam on storage quality and aroma-active volatiles in fresh-cut Fuji apples

Article References: Gao, J., Wei, S., Li, M., Ni, Y., Yang, B., & Wen, X. (2026). Beyond microbial control: Effects of short-time superheated steam on storage quality and aroma-active volatiles in fresh-cut Fuji apples. Food Chemistry: X, 39, Article 104527. https://doi.org/10.1016/j.fochx.2026.104527

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104527

Keywords: superheated steam, fresh-cut apples, Fuji apples, food preservation, aroma volatiles, polyphenol oxidase, peroxidase, microbial inactivation, enzymatic browning, odor activity value, gas chromatography-mass spectrometry, refrigerated storage

Kiwifruit is what botanists call a climacteric fruit, meaning it continues to ripen after harvest by producing its own ethylene, a gaseous plant hormone that orchestrates softening, sweetening, and aroma development. Global production of the fruit has surpassed four million tons annually since 2017, reaching roughly 4.4 million tons by 2024, with the green-fleshed Hayward cultivar dominating commercial markets thanks to its characteristic flavor. Because kiwifruit is typically picked while still firm and physiologically mature but far from ripe, its flavor, sweetness, and aroma remain underdeveloped at harvest and must be coaxed along with ethylene treatment before it reaches consumers.

The research team, led by Inhwan Kim and Eunyoung Park of Chung-Ang University and Seoul National University along with colleagues at the Korea Food Research Institute, set out to answer a question that has lingered in postharvest science: how exactly do ethylene and 1-methylcyclopropene alter not just the quantity of aroma compounds in kiwifruit, but their actual sensory impact? Previous studies had catalogued volatile changes during ripening, but raw concentrations can be misleading, because odor thresholds vary enormously between compounds. A chemical present in large amounts may barely register to the human nose, while a trace compound can dominate the aroma experience.

To capture the full picture, the researchers subjected Hayward kiwifruit imported from New Zealand to three treatments in sealed chambers at 20 degrees Celsius: an ethylene exposure of 1000 microliters per liter for 22 hours, a 1-methylcyclopropene treatment at 1.4 microliters per liter, and an untreated control. Calcium hydroxide was included in each chamber to absorb carbon dioxide and keep conditions consistent. Fruit from each group was then analyzed at days zero, three, five, seven, and ten of storage, yielding a time-resolved portrait of aroma chemistry under each regime.

The analytical workhorse of the study was headspace solid-phase microextraction coupled with gas chromatography-mass spectrometry, a technique that allows volatile compounds released from the fruit pulp to be captured on an absorbent fiber and then separated and identified with high precision. Using isotopically labeled internal standards, the team quantified 27 distinct volatile compounds: nine aldehydes, seven esters, four alcohols, five terpenes, one ketone, and one furanone. Ten of these, including hexanal, hex-2-enal, nonanal, benzaldehyde, methyl hexanoate, and the pinenes, were confirmed against authentic reference standards, while the rest were identified through mass spectral libraries and retention indices.

The headline numbers are striking. Over ten days of storage, total volatile content in untreated control fruit climbed from 761 to 6335 micrograms per kilogram, an 8.32-fold increase, while ethylene-treated fruit surged even further, from 927 to 8771 micrograms per kilogram, a 9.46-fold rise. In stark contrast, fruit treated with 1-methylcyclopropene, which jams the ethylene receptors on cell membranes and prevents the ripening signal from being received, saw its volatile content fall to just 71.5 percent of its starting level by day ten. The fruit had essentially been frozen in an immature aromatic state.

But raw abundance tells only half the story, so the researchers calculated relative odor activity values, or rOAVs, which compare each compound’s concentration to its known odor threshold and express it relative to the most potent odorant in the sample. Compounds with rOAV values of one or higher are considered critical to the characteristic aroma, those between 0.1 and one contribute to overall flavor, and those below 0.1 are merely potential aroma players. This sensory-weighted lens revealed a clear narrative arc: early in storage, the aroma of all three groups was dominated by aldehydes, particularly hexanal and hex-2-enal, the C6 compounds responsible for the fresh, green, grassy notes characteristic of freshly cut kiwifruit.

As storage progressed, a dramatic shift occurred in the control and ethylene-treated fruit. Esters, the class of compounds responsible for fruity and sweet aromas, began to accumulate explosively. Methyl isobutyrate in stored control fruit reached 1067 micrograms per kilogram, a staggering 97-fold increase over fresh fruit, and in ethylene-treated samples it climbed 370.9-fold. Ethyl benzoate, undetectable in fresh ethylene-treated fruit, reached 2401 micrograms per kilogram after storage and ultimately displayed the highest odor activity of any compound identified by day ten in the ethylene group. Methyl butyrate, methyl isobutyrate, and ethyl benzoate all crossed the critical rOAV threshold of one by day seven in controls and by day five in ethylene-treated fruit, marking the point at which the kiwifruit’s aroma pivots from green to genuinely fruity.

Interestingly, the study punctured a long-standing assumption. Ethyl acetate is routinely monitored as the signature ester of fruit ripening, and its concentration did indeed rise during storage. But its odor threshold is exceptionally high at 6200 micrograms per kilogram, so its rOAV remained below 0.01 throughout the experiment, meaning it contributes almost nothing to the aroma humans actually perceive. The real aromatic powerhouses, the authors argue, are the short-chain esters like methyl isobutyrate and the benzoates like ethyl benzoate, which should be considered the true indicators of fruity aroma development in ripe kiwifruit. The biochemistry behind this shift involves the lipoxygenase pathway, which generates aldehyde precursors, and the enzyme alcohol acyltransferase, which esterifies acyl-CoA molecules with alcohols to form esters; prior work has shown that genes encoding these enzymes are positively correlated with ester accumulation and are suppressed by 1-methylcyclopropene treatment.

The practical implications are considerable. Principal component analysis showed that 1-methylcyclopropene-treated fruit retained an early-stage aroma profile throughout the entire storage period, clustering with fresh samples rather than progressing toward the ripe, ester-rich profile of the other groups. This confirms that blocking ethylene perception preserves shelf life and firmness but comes at the cost of aroma development, a trade-off that distributors must weigh depending on their timeline. Conversely, ethylene treatment accelerates the arrival of full fruity aroma, making it a tool for preparing ready-to-eat fruit on demand. By combining quantitative volatile profiling with odor activity analysis, the Korean team has provided the industry with a sensory-meaningful framework for deciding exactly when and how to intervene, transforming what was once an art of guesswork into a science of precision flavor control during the long journey from vine to table.

Subject of Research: Effects of ethylene and 1-methylcyclopropene on aroma volatile profiles of Hayward kiwifruit during postharvest storage

Article Title: Ethylene and 1-methylcyclopropene modulate aroma volatile profiles in kiwifruit (Actinidia spp.) during postharvest storage

Article References: Kim, I., Park, E., Lee, H., Ahn, D., Choi, J. H., Park, K.-J., Lim, J.-H., & Lee, J. (2026). Ethylene and 1-methylcyclopropene modulate aroma volatile profiles in kiwifruit (Actinidia spp.) during postharvest storage. Food Science and Biotechnology. https://doi.org/10.1007/s10068-026-02248-z

Image Credits: AI Generated

DOI: 10.1007/s10068-026-02248-z

Keywords: kiwifruit, ethylene, 1-methylcyclopropene, aroma volatiles, postharvest storage, esters, aldehydes, odor activity value, climacteric fruit, HS-SPME-GC-MS, food science, fruit ripening