The research team, led by Yong Liang, Wei Wang, Kai Huang, Zhengxing Chen, and Tao Wang, began from a well-known obstacle. Rice proteins are nutritionally attractive because they offer a balanced profile of essential amino acids and low allergenicity, but as foam stabilizers they are notoriously difficult. They are largely insoluble in water and prone to severe aggregation, which prevents them from diffusing quickly to the air-water interface and forming a coherent film. Foams stabilized by plant protein particles alone often collapse within seconds or minutes, because these proteins tend to sit more comfortably in the air phase than in the liquid phase, leaving the foam vulnerable to rapid water drainage. Enhancing the water retention capacity of the interfacial film therefore became the central design goal of the work.

To solve the solubility problem, the researchers co-assembled rice proteins with soy protein isolates using a pathway of alkali solvation followed by acid precipitation. Both proteins were first unfolded at high pH, allowed to interact, and then refolded together into uniform, water-dispersible nanoparticles. Transmission electron microscopy showed that while rice proteins alone formed aggregates of roughly thirty-nanometer elementary particles, the co-assembled rice-soy protein nanoparticles, prepared at a one-to-one mass ratio, emerged as monodisperse particles of about forty nanometers. Zeta potential measurements revealed something unexpected: the complexes carried stronger surface charges than either protein alone, while their surface hydrophobicity was lower than that of either parent protein. This indicated that the hydrophobic groups of both proteins had been buried inside the co-assembled structures while the charged groups were exposed outward, a rearrangement driven by hydrophobic interactions, electrostatic attraction, and hydrogen bonding.

These nanoparticles already improved foaming. Soy protein isolates alone could generate foams filling seventy-five percent of the initial dispersion volume, but only five percent survived thirty minutes. The co-assembled rice-soy nanoparticles at optimal ratios reached one hundred percent foaming ability and pushed thirty-minute stability to roughly fifteen percent at higher concentrations, plateauing at fifty percent survival once the interface became saturated. That was progress, but more than eighty-five percent of the foam still vanished within half an hour. The team then turned to starch nanocrystals, rigid platelets produced by hydrolyzing waxy corn starch in sulfuric acid for seven days. These crystals carry abundant hydroxyl groups on their surfaces, making them strongly hydrophilic and mechanically robust, but they also aggregate heavily through hydrogen bonding between their faces.

When the protein nanoparticles and starch nanocrystals were simply mixed together, the two components associated into hybrid complexes the researchers call RSP@SNCs. Atomic force microscopy revealed that the aggregated starch crystals disaggregated and reappeared as sheet-like hybrid nanoparticles of one hundred to two hundred nanometers, a morphology distinct from either component. Intrinsic fluorescence spectroscopy provided quantitative evidence of close association: as starch nanocrystal concentration increased, the fluorescence of the protein’s aromatic amino acid residues was progressively quenched, following an apparent Stern-Volmer-type relationship with a mass-based quenching coefficient of 1.896 milliliters per milligram and a correlation coefficient of 0.9677. Fourier transform infrared spectroscopy showed no new absorption bands or significant shifts, confirming that the binding was non-covalent.

The molecular glue holding the complexes together turned out to be hydrogen bonding. In a clever chemical blocking experiment, the team repeated the complexation in the presence of sodium chloride, sodium dodecyl sulfate, or thiourea, which selectively disrupt ionic forces, hydrophobic attractions, and hydrogen bonding respectively. Thiourea dramatically weakened the protein-starch interactions and restored fluorescence, while the salt and surfactant had only minor effects. Because the starch nanocrystals self-aggregate primarily through hydrogen bonds between their hydroxyl groups, competition from protein-starch hydrogen bonding appears to reduce the surface energy of the crystals, suppressing their self-aggregation and yielding dispersed hybrids with balanced amphiphilic character, exactly what is needed to anchor at an air-water interface.

The foaming results were striking. Adding starch nanocrystals doubled the initial foam formation of the protein nanoparticles at the highest tested ratio of 1.2, while optical microscopy showed that bubbles became smaller and more homogeneous as the starch proportion rose. Confocal laser scanning microscopy, with proteins and starch labeled in different fluorescent colors, revealed that both components were jointly trapped at the air-water interface, forming a thick physical barrier around each bubble. In foams stabilized by protein alone, bubbles approached a millimeter in size with thin, patchy coverage; with the hybrids at a ratio of 0.8, bubble sizes dropped to hundreds of micrometers and the liquid-filled spaces between adjacent bubbles widened, reducing direct bubble contact and hindering coalescence.

The stabilization mechanism operates on three coupled fronts. First, the interfacial film itself: contact angle measurements showed starch nanocrystals are highly hydrophilic at forty degrees while rice-soy protein nanoparticles are more hydrophobic at seventy-nine degrees, and the hybrids strike a balance that wets both phases, lowers interfacial tension, and satisfies the Gibbs stability criterion against coarsening. Second, the bulk viscosity: shear rheology demonstrated that viscosity of the stabilizer dispersions increased with starch content, and cryo-scanning electron microscopy performed at minus 145 degrees Celsius revealed ordered hybrid structures arranged throughout the continuous phase, not just at the interface. These ordered microstructures elevated system viscosity, slowing bubble disproportionation and liquid drainage. Notably, the dispersions showed shear-thinning behavior, meaning that during the violence of whipping the mixture temporarily thins, allowing particles to race to the interface, after which viscosity rebuilds to lock the structure in place. Third, water retention: the hydrophilic starch crystals trap water in the interstitial spaces between bubbles, deterring the gravity-driven drainage that kills most foams.

The practical implications extend well beyond the laboratory. The authors point to applications in milky tea, lattes, and other foamed beverages where stability over hours rather than minutes would transform product quality, particularly for delivery and takeaway markets. Because every ingredient is edible and plant-derived, the system sidesteps the limitation of surfactant crystals, which can produce extraordinarily stable foams but are not food-grade. The work also demonstrates a broader design principle: rather than searching for a single miracle stabilizer, combining a diffusion-competent protein nanoparticle with a mechanically strong, hydrophilic nanocrystal produces synergy that neither component achieves alone. As demand grows for plant-based, low-calorie foods with indulgent textures, engineering the air-water interface with hybrid biocolloids may become a standard tool in the food engineer’s repertoire, turning one of the most fragile structures in cuisine into one of the most durable.

Subject of Research: Stabilization of aqueous food foams using rice and soy protein nanoparticles reinforced with starch nanocrystals

Article Title: Starch nanocrystals reinforce rice/soy protein-structured air-water interface to stabilize aqueous foams

Article References: Liang, Y., Wang, W., Huang, K., Chen, Z., & Wang, T. (2026). Starch nanocrystals reinforce rice/soy protein-structured air-water interface to stabilize aqueous foams. Journal of Agriculture and Food Research, Article 103352. https://doi.org/10.1016/j.jafr.2026.103352

Image Credits: AI Generated

DOI: 10.1016/j.jafr.2026.103352

Keywords: aqueous foams, starch nanocrystals, rice protein, soy protein isolate, food colloids, air-water interface, hydrogen bonding, foam stability, plant-based foods, Pickering stabilization, interfacial tension, food engineering

Orally disintegrating films, or ODFs, are thin sheets that melt rapidly against the oral mucosa, releasing their payload for absorption without the need to swallow. Conventional ODFs rely almost exclusively on hydrophilic polysaccharides such as pullulan, hydroxypropyl methylcellulose, and pregelatinized starch, which disintegrate quickly; published examples include hydroxypropyl cellulose films that vanish in water in about 12.5 seconds and hydroxypropyl methylcellulose formulations that break down in roughly 43 seconds. But these matrices carry little nutritional value, and protein hydrolysate alternatives are prohibitively expensive. Soy protein isolate, a widely available and inexpensive plant protein with excellent film-forming ability and a rich amino acid profile, seemed like an obvious candidate, except that its dense, flexible molecular packing causes films to disintegrate sluggishly in water, sometimes lagging far behind their polysaccharide competitors.

The team’s answer was β-cyclodextrin, a cyclic oligosaccharide with a hydrophobic interior cavity and a hydrophilic, hydroxyl-covered exterior. Cyclodextrins are workhorses in food and pharmaceutical science, used to emulsify, solubilize, mask unpleasant flavors, and protect sensitive bioactive compounds. Previous work by some of the same authors had shown that β-cyclodextrin could accelerate the disintegration of soy protein films by disrupting the tight packing of protein chains. The new study asked whether the molecule could do double duty: remodel the protein network for rapid release while also protecting ergothioneine, a sulfur-containing antioxidant first isolated in 1909 from the fungus Claviceps purpurea and produced by edible fungi and cyanobacteria.

Ergothioneine was a demanding guest. Its unusually low redox potential of −60 millivolts makes it resistant to autoxidation, and studies have shown it outperforms glutathione at scavenging hydroxyl radicals, peroxyl radicals, and singlet oxygen. Yet antioxidant activity in ergothioneine-containing systems is known to decline during thermal processing and storage, likely through oxidative degradation. The researchers cast films by dissolving 2.0 grams of soy protein isolate in water at 75 degrees Celsius, adding 10 milligrams of ergothioneine, 0.7 grams of D-sorbitol as a plasticizer, and β-cyclodextrin at 0, 5, 10, and 15 percent of the dry protein weight, then drying the solutions in Petri dishes at 40 degrees Celsius for eight hours.

Microscopy revealed how dramatically the additive reshaped the material. The pristine soy protein film was rough and inhomogeneous, riddled with micropores and aggregated particulate domains, while adding ergothioneine alone smoothed the surface, apparently by promoting the unfolding of protein polypeptide chains. The real transformation came at 10 percent β-cyclodextrin, where the films displayed an extremely smooth, dense, defect-free surface organized into a continuous reticular network. The authors attribute this to hydrogen bonding between the cyclodextrin’s abundant hydroxyl groups and the protein network, which suppresses pore formation during drying. At 15 percent, however, the strategy backfired: bright, angular crystalline aggregates appeared across the surface, evidence that excess cyclodextrin had exceeded its solubility limit and phase-separated into β-CD-rich crystalline domains, shattering the structural continuity of the protein matrix.

Those structural changes translated directly into performance. The water contact angle, a measure of surface wettability, fell from 63.77 degrees in the control film to 36.28 degrees at 10 percent cyclodextrin, reflecting a far more hydrophilic surface that welcomes water penetration. Tensile strength climbed from 10.37 megapascals in the control to 12.15 megapascals at the optimal loading, and elongation at break rose from 3.462 to 4.376 percent, meaning the films became both stronger and more flexible. Then came the headline result: in vitro disintegration time dropped from 31.67 seconds in the control to 15.46 seconds at 10 percent cyclodextrin, a 45 to 52 percent improvement across the cyclodextrin-containing formulations, comfortably meeting the sub-60-second benchmark for fast-disintegrating films. Intriguingly, the 15 percent formulation showed no significant further gain, because its added thickness lengthened the path water had to travel, canceling out its extra hydrophilicity.

The loading data told a similarly encouraging story. High-performance liquid chromatography showed that ergothioneine loading efficiency rose from 66.83 percent in films without cyclodextrin to 76.33 percent at the 10 percent level, with loading capacity peaking at 2.16 micrograms per milligram of film. Relative standard deviations across replicate films stayed below 4 percent, confirming that the solvent casting method produces reproducible, uniformly dosed films. Each standard 15-by-15-millimeter dosing unit carried about 16.21 micrograms of ergothioneine in the optimal formulation. Surface pH values for all films ranged from 5.55 to 6.48, safely within the range tolerated by oral tissue, and mucoadhesive forces were essentially unchanged at moderate additive levels, reaching 0.69 newtons only in the phase-separated 15 percent formulation, where crystalline protrusions increased contact area.

The protective half of the dual function emerged under stress. The team sealed films in ordinary plastic bags and stored them for 28 days at 40 degrees Celsius, 75 percent relative humidity, under continuous fluorescent light to simulate accelerated oxidative aging. Films containing ergothioneine alone watched their DPPH radical scavenging activity collapse from 83.47 percent on day zero to 40.86 percent by day 28, with ABTS activity falling in parallel from 85.21 to 41.74 percent. Films with 10 percent cyclodextrin retained 60.57 percent DPPH activity and 60.68 percent ABTS activity over the same period, significantly better than every other formulation. Spectroscopic characterization supported the mechanism: X-ray diffraction showed the amorphous protein film acquiring the characteristic crystalline peaks of β-cyclodextrin, while Fourier-transform infrared spectroscopy revealed strengthened hydrogen bonding bands at 3288, 1641, and 1537 reciprocal centimeters, along with new peaks marking the cyclodextrin skeleton. Thermogravimetric analysis showed the maximum decomposition temperature rising steadily with cyclodextrin loading, from 300.28 degrees Celsius in the control to a peak of 314.63 degrees, confirming a more thermally robust composite.

The authors are careful about how far the interpretation can go. Because the antioxidant assays measure total radical scavenging of the entire film matrix rather than residual ergothioneine concentration directly, and because no inclusion complex between cyclodextrin and ergothioneine was demonstrated, the improved retention may reflect reduced oxidative exposure through intermolecular interactions and a denser matrix rather than the specific stabilization of individual ergothioneine molecules. The disintegration tests also used distilled water rather than simulated saliva, and all evaluations were in vitro. Future work employing HPLC or LC-MS quantification of ergothioneine, phase-solubility analysis, differential scanning calorimetry, and two-dimensional NMR is needed to pin down the molecular fate of the antioxidant during storage.

Even with those caveats, the implications are considerable. The study breaks the disintegration barrier that has kept soy protein out of serious consideration for orally disintegrating films, and it does so with a cheap, food-grade cyclodextrin that simultaneously improves mechanical strength, thermal stability, and antioxidant retention. For the food and pharmaceutical industries, the platform suggests a practical route to delivering sensitive bioactive compounds, from ergothioneine to other oxidation-prone nutraceuticals, in a fast-dissolving, plant-protein-based format that adds nutritional value instead of diluting it. If longer-term storage studies and in vivo safety testing bear out the accelerated results, the humble soybean may find itself at the leading edge of oral thin-film technology, one 15-second melt on the tongue at a time.

Subject of Research: Development of soy protein isolate-based orally disintegrating films using β-cyclodextrin to enhance disintegration and protect ergothioneine

Article Title: Development of soy protein isolate-based orally disintegrating films: The dual role of β-Cyclodextrin in enhancing disintegration performance and ergothioneine stability

Article References: Zhou, Y., Du, X., Lv, C., Tian, Y., Guo, S., Guo, Z., & Wang, Z. (2026). Development of soy protein isolate-based orally disintegrating films: The dual role of β-Cyclodextrin in enhancing disintegration performance and ergothioneine stability. Food Chemistry: X, 39, Article 104451. https://doi.org/10.1016/j.fochx.2026.104451

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104451

Keywords: orally disintegrating films, soy protein isolate, β-cyclodextrin, ergothioneine, oral drug delivery, antioxidant stability, film disintegration, hydrogen bonding, plant protein, Food Chemistry: X, bioactive compound protection, solvent casting