The study, published in the journal Advanced Composites and Hybrid Materials, was led by Ki-Hyun Ryu and Sungju Lee of the Korea Institute of Science and Technology, together with colleagues across several KIST research centers. Their central insight is deceptively simple: instead of trying to push ions into a densely packed nanotube fiber after the fact, why not build molecular-scale breathing room into the material from the very beginning? The answer they arrived at relies on a thermodynamically driven co-assembly process borrowed from the world of liquid crystals, a field more often associated with display screens than with supercapacitors.

To understand why the problem matters, it helps to picture what happens inside a conventional carbon nanotube fiber. Individual nanotubes, each essentially a rolled-up sheet of graphene a few nanometers wide, are drawn together into bundles, and those bundles are packed into a fiber with inter-bundle spacing measured at the molecular scale. This ultra-dense arrangement maximizes the contact between tubes, which is exactly what you want for load transfer and for electrons hopping from one tube to the next. But for electrochemical applications, the picture inverts. Ions from an electrolyte can only access the outermost surfaces of the fiber; the deep internal channels between bundles remain inaccessible, so the enormous theoretical surface area of the nanotubes goes largely unused. Materials scientists have long described this as a packing-accessibility trade-off, and it has been a persistent bottleneck in translating the extraordinary properties of individual nanotubes into practical devices.

The Korean team’s strategy attacks the problem at the molecular level using soft poly(vinyl alcohol), or PVA, chains. PVA is a hydrophilic, water-soluble polymer, and the researchers exploited its strong non-covalent adsorption onto carbon nanotube surfaces to intercalate the polymer chains directly into the nanotube bundles during fiber formation. Because the process occurs during a lyotropic liquid crystalline co-assembly stage, the rigid nanotubes retain the high axial alignment that is essential for efficient electron transport along the fiber, while the flexible polymer chains wedge themselves between the bundles and hold the channels open. The researchers describe the resulting architecture as unlocked, in the sense that the interstitial channels that would normally collapse into dense packing are instead molecularly secured at a controlled spacing.

The thermodynamics of this intercalation are worth appreciating. When long, flexible polymer chains are confined between rigid nanotubes, they lose some conformational entropy, but the strong adsorption energy between the polymer and the carbon surface compensates for that loss, driving the hybrid structure to self-assemble spontaneously. Once in place, the entropic relaxation of the soft chains, their tendency to coil and spread, effectively props open the interstitial channels without any external scaffolding, template, or post-processing step. This bottom-up approach stands in contrast to more common strategies that graft bulky functional groups onto nanotube walls or hybridize the fibers with additional active materials, both of which can disrupt the intrinsic structure of the nanotubes or add complexity and cost to manufacturing.

The benefits of the unlocked architecture extend beyond simple spacing. Carbon nanotube surfaces are inherently hydrophobic, which impedes wetting by aqueous electrolytes and further limits ion access. The intercalated PVA chains, being hydrophilic, impart wettability to the composite fiber, so electrolyte ions are not only given physical pathways into the interior but are also chemically encouraged to penetrate. In effect, the polymer performs double duty: it acts as a molecular spacer that maintains the channels, and it converts the fiber’s internal environment from one that repels electrolytes to one that welcomes them. The dormant internal volume of the fiber, previously inaccessible, becomes an active participant in electrochemical processes.

The quantitative evidence for this internal activation is striking. When the researchers measured the electrochemical performance of their composite fibers, they found an 11.2-fold enhancement in specific capacitance compared with pristine carbon nanotube fibers. Specific capacitance is a direct measure of how much charge a material can store per unit mass, and an improvement of more than an order of magnitude indicates that ions are genuinely reaching the deep core of the fiber rather than merely decorating its surface. The composite fibers also showed superior rate performance, meaning they retained much of their charge-storage capability even when charged and discharged rapidly. Rate capability is a sensitive probe of ion diffusion resistance: if ions had to fight their way through dense packing, performance would collapse at high rates. The fact that it did not confirms that the engineered channels substantially mitigate diffusion resistance throughout the fiber’s cross-section.

What makes this work particularly significant is its generality. The authors frame their methodology as a universal design paradigm for next-generation composite materials that require dynamic ion transport. Dense assemblies of one-dimensional nanomaterials are not limited to carbon nanotubes; nanowires, nanofibers, and other elongated building blocks face the same packing-accessibility dilemma whenever they are assembled into macroscopic forms. A spacing strategy that works during co-assembly, requires no complex active-material hybrids, and preserves the intrinsic structure of the constituent nanomaterials could therefore be applied far beyond the supercapacitor context in which it was demonstrated here. The approach also avoids covalent modification of the nanotubes, which is important because covalent chemistry on carbon surfaces introduces defects that degrade the electrical conductivity that makes nanotubes valuable in the first place.

The research was carried out at the Advanced Conductive Materials Research Center, the Carbon Composite Materials Research Center, and the Functional Composite Materials Research Center of the Korea Institute of Science and Technology in Wanju, Republic of Korea. It was supported by the Nano and Material Technology Development Programs through the National Research Foundation of Korea, funded by the Ministry of Science and ICT, as well as by a KIST Institutional Program. The work was published as an open-access article on 10 September 2026, making the full technical details freely available to researchers worldwide.

For a field that has spent decades chasing ways to make carbon nanotube assemblies live up to their nanoscale promise, the message of this study is refreshingly constructive. Rather than treating dense packing as an unavoidable compromise, the researchers showed that molecular-level design, using nothing more exotic than a soft polymer and the thermodynamics of liquid crystalline assembly, can unlock the hidden interior of a nanotube fiber while keeping its celebrated strength and conductivity intact. If the strategy proves as transferable as the authors suggest, the dense-but-dead interiors of one-dimensional nanomaterial assemblies may soon become some of the most valuable real estate in materials science.

Subject of Research: Liquid crystalline co-assembly of carbon nanotube and poly(vinyl alcohol) hybrid fibers to open interstitial channels for enhanced interfacial ion transport

Article Title: Bottom-up co-assembly of carbon nanotube and soft polymer hybrids: unlocking interstitial channels for enhanced interfacial ion kinetics

Article References: Ryu, K.-H., Lee, S., Lim, S.-I., Kim, S. G., Kim, S. M., Kim, N. D., Hwang, J. Y., Moon, S. Y., Yu, H., Jeong, H. S., & Kim, D.-Y. (2026). Bottom-up co-assembly of carbon nanotube and soft polymer hybrids: unlocking interstitial channels for enhanced interfacial ion kinetics. Advanced Composites and Hybrid Materials. https://doi.org/10.1007/s42114-026-02076-z

Image Credits: AI Generated

DOI: 10.1007/s42114-026-02076-z

Keywords: carbon nanotubes, poly(vinyl alcohol), liquid crystals, co-assembly, nanocomposites, composite fibers, interstitial channels, ion diffusion, specific capacitance, supercapacitors, soft matter, Korea Institute of Science and Technology

The research team, led by Ying Han and Xin Yang, screened a wide range of sugars, from monosaccharides like arabinose and fructose to polysaccharides and sugar derivatives, before settling on chitosan oligosaccharides, or COS, with a degree of polymerization of just 2 to 6 sugar units. Unlike fructose, which has previously been shown to induce curcumin self-assembly but merely stabilizes environmental charge without participating in the assembly itself, COS are the only naturally charged amino saccharides and carry meaningful biological activity of their own. That distinction matters: if the sugar is not just a passive inducer but an active structural partner, then the biological activities of both the carrier and the cargo can be exploited simultaneously. The resulting co-assembled nanoparticles achieved a drug loading content of 41.56 percent and an entrapment efficiency of 76.73 percent, figures that stand well above the roughly 10 to 15 percent loading typical of many conventional curcumin delivery systems.

Getting the particles to the right size proved to be the study’s central technical challenge. Initial co-assemblies formed regular spheres but aggregated into micron-scale clusters of roughly 1 to 1.2 micrometers, far too large for the enhanced permeability and retention effect that allows nanoparticles to accumulate passively in tumor tissue, which generally requires particles below about 200 nanometers. Intriguingly, the researchers found that almost none of the obvious processing variables mattered. Solvent polarity, tested across methanol, ethanol, dimethyl sulfoxide, acetone, and ethyl acetate, had no effect on particle size. Neither did carrier concentration, stirring speed, stirring time, or stirring method. Because the COS molecules themselves self-assemble into structures below 50 nanometers, the team concluded the micron-sized objects were agglomerates of small nanoparticles, and that the answer lay in controlling aggregation rather than formation.

The fix came from an unexpected quarter: an inorganic electrolyte. After testing various salts and surfactants, the team found that adding sodium carbonate during assembly kept the particles below the critical size threshold. Dynamic light scattering confirmed the adjusted particles measured around 200 nanometers, with scanning electron microscopy revealing irregular, sphere-like accumulations and transmission electron microscopy showing that each particle was itself composed of several smaller nanoparticles with a sharp light-dark contrast between center and edge. Contact angle measurements showed the composite surface had become substantially more hydrophilic than free curcumin, consistent with a structure in which hydrophobic curcumin is buried inside while hydrophilic COS chains face outward toward water. The particles carried a slightly negative zeta potential, a surface charge profile associated with improved blood compatibility, reduced clearance by the reticuloendothelial system, and longer serum half-life.

Spectroscopy and molecular dynamics simulations together revealed the forces holding the assembly together. Ultraviolet-visible spectra showed a slight red shift of curcumin’s characteristic 435-nanometer absorption peak upon incorporation, indicating hydrogen bonding between the sugar and the polyphenol, a shift that weakened as sodium carbonate levels rose, presumably because the salt competed with hydrogen ions and disrupted the bonding. When the nanoparticles were exposed to 0.2 percent sodium dodecyl sulfate, the absorption peak shifted blue, signaling disassembly and confirming that hydrophobic interactions also contribute to the structure. Fourier transform infrared spectroscopy showed characteristic shifts in the hydroxyl, carbonyl, and carbon-oxygen stretching regions of the composite relative to a simple physical mixture of the two components. Simulations performed with the COMPASS force field predicted strong hydrogen bonds between chitotetraose, the dominant oligomer in the mixture, and curcumin, with bond lengths of roughly 2.02 and 2.56 angstroms, and showed the molecules spontaneously tending toward spherical arrangements.

The functional payoff was dramatic. Free curcumin exhibits an apparent aqueous solubility of just 0.0923 percent, but once incorporated into the COS nanoparticles, that figure rose to 99.87 percent, an enhancement of approximately 1,100-fold. Under simulated gastrointestinal digestion, moving from gastric fluid at pH 2.0 containing pepsin to intestinal fluid at pH 7.0 containing pancreatin, the area under the release curve for the nanoparticles was more than 15 times higher than for free curcumin, indicating far greater gastrointestinal availability. The authors are careful to note that this demonstrates enhanced gastrointestinal release performance rather than directly establishing systemic bioavailability, but the implication for functional foods and nutraceuticals is clear: a curcumin that actually disperses in aqueous food matrices and survives to be absorbed.

Release behavior under other conditions reinforced the picture of a stable but responsive carrier. At physiological temperature, the nanoparticles released about 31 percent of their curcumin payload at pH 7.4, simulating blood circulation, about 33 percent at pH 6.7, simulating the tumor microenvironment, and about 45 percent at pH 5.6, simulating the endosomal and lysosomal compartments inside cells. The absence of an initial burst release suggests the hydrogen bonding and hydrophobic interactions that build the assembly also restrain premature curcumin leakage during transport. The particles showed no obvious change in particle size over 30 days of storage, and hemolysis assays on fresh mouse blood found rupture rates below the 5 percent safety threshold at all tested concentrations, supporting the system’s biocompatibility.

Biological testing showed the co-assembly did more than improve solubility. In 4T1 mouse breast cancer cells, the IC50 of the nanoparticles dropped to 10.54 micrograms per milliliter compared with 20.64 for free curcumin, a roughly 49 percent reduction corresponding to nearly a doubling of antiproliferative potency, with a combination index of 0.81 suggesting synergy between the sugar and the drug. In MCF-7 human breast cancer cells the effect was even more pronounced: the IC50 fell from 16.96 to 1.33 micrograms per milliliter, a 12.75-fold increase in apparent potency with a combination index of 0.78. Flow cytometry showed the nanoparticles pushed cells out of the G0/G1 phase and accumulated them in G2/M, with 46.33 percent of treated cells in that phase versus 3.75 percent of controls. Fluorescence microscopy confirmed time-dependent accumulation of the particles in the cytoplasm of 4T1 cells.

In tumor-bearing BALB/c mice, fluorescently labeled nanoparticles injected via the tail vein rapidly distributed to liver, kidney, and tumor tissue; signals in liver and kidney faded over time while the tumor retained a strong, persistent fluorescence, consistent with passive EPR-mediated targeting. Mice treated with the nanoparticles showed significantly lower tumor volumes than those given free curcumin or COS alone, with the relative tumor inhibition rate rising from 58 percent for free curcumin to 73 percent for the co-assembled formulation over 12 days of treatment. Body weight, liver and kidney function markers including ALT, AST, ALP, bilirubin, creatinine, uric acid, and blood urea nitrogen, and histopathological staining of major organs showed no significant toxicity. The work was funded by the National Natural Science Foundation of China and approved by the Experimental Animal Ethics Committee of Ningbo University.

What makes the study notable beyond its headline numbers is its simplicity. Lipid carriers can solubilize hydrophobic compounds but suffer from physical and oxidative instability; synthetic polymer nanoparticles offer design flexibility but involve materials and processes less suited to food applications. The COS-curcumin system requires nothing more exotic than mixing a water solution of short sugar chains with a curcumin stock in dimethyl sulfoxide under stirring, plus a pinch of sodium carbonate to tame aggregation. By establishing a direct structure-assembly-function relationship, in which molecular complementarity between hydrophilic oligosaccharides and the hydrophobic polyphenol drives both nanoparticle formation and performance, the researchers point toward carbohydrate-based nanomaterials as sustainable, food-compatible platforms for delivering not just curcumin but a broader class of poorly water-soluble phytochemicals into functional foods and nutritional products.

Subject of Research: Chitosan oligosaccharide-curcumin co-assembled nanoparticles for enhancing curcumin solubility and bioavailability

Article Title: Co-assembly of Highly Active Chitosan Oligosaccharide with Curcumin to Improve Its Bioavailability

Article References: Han, Y., Fu, S., Que, C., Zhang, Z., Song, A., Cao, J., Wang, J., Zhang, H., & Yang, X. (2026). Co-assembly of Highly Active Chitosan Oligosaccharide with Curcumin to Improve Its Bioavailability. Current Research in Food Science, Article 101582. https://doi.org/10.1016/j.crfs.2026.101582

Image Credits: AI Generated

DOI: 10.1016/j.crfs.2026.101582

Keywords: curcumin, chitosan oligosaccharides, nanoparticles, co-assembly, bioavailability, drug delivery, functional foods, nutraceuticals, hydrogen bonding, antitumor activity, water solubility, food science