There is a question that sits quietly underneath much of chiral optics: how much can you add to a quantum state before it stops remembering the shape that made it chiral? An exciton — one electron bound to one hole — inherits its optical handedness from the structure that hosts it. Add a third particle, an extra electron or hole, and the resulting trion is a different beast: a Coulomb-bound three-body state with its own energies and selection rules, and, until now, no demonstrated structural chirality of its own. Whether the trion would keep the parent exciton’s handedness, scramble it, or wash it out entirely was an open question.
A paper published in Nature Communications on 30 September 2026 answers it, for one well-controlled system, with unusual clarity. Hiroyuki Fujinami, Satoshi Kusaba, Satoru Konabe, Weilu Gao, Jana Zaumseil and Kazuhiro Yanagi report the circular dichroism of trions in enantiopure single-wall carbon nanotubes: the trion’s circular-dichroism sign is the same as that of the parent E11 exciton, and its dissymmetry factor is almost the same too. The extra charge binds to the exciton without erasing its chiroptical signature.
A note on timing before the physics. This is a September paper that has only now surfaced in trade coverage, and this article treats it accordingly: as a result to examine, not as breaking news. The paper itself is open access, and every number below comes from it, not from the aggregator summary that brought it to attention.
The physics first: what makes a nanotube chiral
A single-wall carbon nanotube is a graphene sheet rolled into a cylinder, and the roll matters. The roll is specified by a pair of integers, (n, m), that describe the direction in which the sheet is wrapped. Most (n, m) combinations produce a tube whose atomic lattice winds helically around its axis, like the thread of a screw. Such a tube is chiral: it cannot be superimposed on its mirror image. The mirror image is a different tube, the enantiomer, with its own (n, m) indices. The pair studied here — (6, 5) and (11, −5) — are enantiomers of one another, related by the general rule that the enantiomer of (n, m) is (n + m, −m).
Chirality in the lattice becomes chirality in the optics through circular dichroism, or CD. A chiral material absorbs left- and right-circularly polarised light by different amounts, and the difference, ΔA = A_L − A_R, is the CD signal. For an exciton in a nanotube, the sign and size of that signal depend on which subband transition is involved: the first subband transition, E11, is the lowest-energy excitonic absorption, with E22 and E33 higher. A given enantiomer shows a characteristic pattern of positive and negative CD across these transitions, and the mirror-image enantiomer shows the mirror-image pattern.
The standard way to compare that fingerprint across transitions and samples is the absorption dissymmetry factor, g_abs = ΔA / A: the CD signal divided by total absorbance. It measures how chiral a transition is relative to how strong it is. It is the right quantity here, because a trion transition is weaker than its parent exciton transition in absolute terms, and comparing raw CD signals would conflate strength with handedness.
That distinction is what makes the trion question genuinely uncertain. There are two ways a trion could acquire CD, and they were not equivalent. One is spin-based: circularly polarised light selectively excites particular spin or valley states through selection rules, and trion CD of that kind had already been demonstrated in quantum dots, quantum wires, perovskites and monolayer semiconductors. The other is configurational: CD from the chiral structure itself, the helical lattice, independent of spin selection. Configurational trion CD had never been observed, and there was no guarantee that adding a charge would leave the structure-derived response intact.
The trion, and why nanotubes are the place to test it
A trion is an exciton plus one additional carrier: two electrons and one hole for a negative trion, one electron and two holes for a positive trion. In most bulk semiconductors, trions are fragile, bound by a milli-electron-volt or less, visible only at cryogenic temperatures. Carbon nanotubes are an exception, and the reason is dimensionality. Confined to a quasi-one-dimensional tube, carriers interact far more strongly than in a three-dimensional crystal, and the paper cites trion binding energies exceeding 100 meV — more than four times room-temperature thermal energy. That is why nanotube trions remain stable at room temperature, and why the experiment could be done on a benchtop spectrometer rather than in a cryostat.
The experiment also needed something that has only recently become routine: enantiopure nanotubes. Raw nanotube material is a mixture of chiralities and both handednesses, and a mixed sample’s CD cancels out. Fujinami and colleagues separated (6, 5) and (11, −5) tubes by gel chromatography from a commercial starting material, CoMoCAT SG65i, and formed thin films of each by vacuum filtration onto quartz substrates. The films were deliberately made from randomly oriented tubes, with linear dichroism below 0.01, so the CD measured was real chirality, not an artefact of alignment. Enantiomeric purity was around 90 percent, confirmed by the strong, mirror-image CD spectra of the two films at E11, E22 and E33, against a racemic control film that showed nearly zero CD.
To make trions, the team used electrolyte gating: a droplet of the ionic liquid EMIM-TFSI spread over the film, with a gold working electrode beneath and a large lateral gate electrode. Applying a gate voltage injects carriers into the film — holes at negative voltage, electrons at positive voltage — and the injected carriers bind to photoexcited excitons to form trions. Spectra were recorded at each gate voltage step in a vacuum chamber below 10^-4 Pa. The same knob that creates the trion also lets the experimenter watch the exciton disappear and the trion appear, in both absorption and CD, on the same sample, in the same run.
What the gating showed
At zero gate voltage, the undoped films show the familiar excitonic spectrum, with the E11 peak around 1.2 eV in the (11, −5) film. As the gate voltage is swept — 0 to −2.5 V for hole injection, 0 to +2.5 V for electron injection — the E11 peak bleaches in both absorption and CD, and a new feature grows in on its low-energy side, at around 1.06 eV. That is the trion: lower in energy than the parent exciton by the trion binding energy, appearing only when carriers are present, and fading again at high carrier density when the first subband fills completely. The behaviour is the same for hole and electron injection, for both enantiomers. The new part is in the CD channel: the trion feature has its own CD signal, at the same 1.06 eV position, and its sign matches the E11 exciton’s sign in every case — positive for (11, −5), negative for (6, 5).
The control experiment seals the structural interpretation. In the racemic film, where the two enantiomers cancel, no trion CD is observed at all. If the trion CD came from spin selection rules or from the gating geometry, it would survive in a racemic mixture; it does not. It comes from the chiral structure, and only from the chiral structure. That is the sense in which this is the first configurational CD of a trion: the three-particle state’s optical handedness is inherited from the lattice, not imposed by the light’s spin.
The quantitative result is the dissymmetry factor. Rather than taking a single CD-to-absorption ratio at one gate voltage, which is sensitive to baseline offsets, the authors extracted g_abs from the slope of the CD-versus-absorption correlation as the gate voltage was swept from 0 to ±3.0 V. For the E11 exciton, the slope was 65 ± 1 mdeg per optical-density unit, corresponding to g_abs = (1.97 ± 0.03) × 10^-3. For the trion, the slope was 64 ± 2 mdeg/OD, corresponding to g_abs = (1.94 ± 0.06) × 10^-3. The two values agree within their uncertainties: adding a charge changes the absolute transition strength, but changes the left- and right-circularly polarised strengths by nearly the same relative factor, so the handedness survives essentially untouched.
That near-equality is more informative than it first appears, because it is not what happens to every exciton-derived state in nanotubes. The paper notes the contrast with quantum defects in the same material: when an exciton is localised at a defect, its g_abs is reduced relative to the E11 exciton. A defect rewrites the optical response; a trion, apparently, does not. The authors are careful about what this does and does not prove: a trion is a genuine three-particle correlated state, not reducible to a simple exciton-plus-carrier picture, and a fully microscopic theory of trion CD in chiral nanotubes remains a subject for future work. What the data support is a phenomenological reading — the optically active trion inherits the helicity-dependent response of the parent E11 manifold — and that should not be inflated into a general theorem about all chiral three-body states.
Why it matters, and what it does not yet show
The immediate significance is fundamental. Chirality is a property of a structure, and many-body states are properties of interactions. It was not obvious that the two would compose cleanly — that a property imprinted by a helical lattice would survive the addition of a strongly interacting third particle. This result says it does, at least for the brightest trion channel in this system, and it gives the field a method — gated CD on enantiopure films — for asking whether four- and five-body states behave the same way.
The device implications are real but second-order. Trions are charged, so unlike neutral excitons they can be created and tuned electrically, and this experiment is itself a demonstration of that: the chiral optical response of the film is modulated by a gate voltage, at room temperature, in the near-infrared, around 1.06 to 1.2 eV. A material whose circular-polarisation response can be switched electrically is a candidate ingredient for circular-polarisation-resolved photodetectors, modulators, and chiral light sources. But no device is demonstrated here — no photodetector, no modulator, no measured extinction ratio in a device geometry. The films are randomly oriented networks, 100–110 nm thick, measured in a spectrometer; the path to a working component runs through alignment, contacts, and integration. None of that is in this paper, and none of it should be implied by coverage of it.
The limits of the result deserve equal weight. First, it is demonstrated for one enantiomeric pair, (6, 5) and (11, −5), at around 90 percent purity. The paper’s reasoning suggests generality across chiralities, but generality is not yet shown, and other chiralities have different CD magnitudes and sign patterns. Second, the measurement is absorption CD, not emission; trion emission CD, which a chiral light source would need, is a separate measurement this paper does not report. Third, the trion CD exists only in a window of doping, appearing with carrier injection and disappearing again at high density. Fourth, the theory is phenomenological; a fully microscopic theory is still to come, so the mechanism is described, not yet derived from first principles.
None of these limits diminishes the result; they locate it. The paper’s contribution is a clean, quantitative, well-controlled fact: in enantiopure carbon nanotubes, the trion’s dissymmetry factor is (1.94 ± 0.06) × 10^-3 against the parent exciton’s (1.97 ± 0.03) × 10^-3, with matching signs in both enantiomers and no signal in the racemic control.
What to watch
Three follow-ups would each change the assessment in a different direction. The first is chirality generality: the same gated-CD measurement on other enantiomeric pairs, to test whether sign-matching and g_abs preservation are universal. The second is emission: whether trion photoluminescence shows the same CD, which would connect this absorption result to chiral emitters. The third is theory: a microscopic calculation of trion CD in a chiral tube that predicts the near-equality of g_abs from the three-body wavefunction, rather than rationalising it after the fact.
There is also a quieter implication for anyone who works with nanotube optics. Trion formation is often treated as a nuisance in excitonic devices: doping bleaches the exciton, shifts the spectrum, and complicates the lineshape. This result reframes that nuisance, slightly. The trion is not a featureless charge-induced absorption that obscures the exciton’s chirality; it carries the chirality forward, at nearly the same relative strength. In a doped chiral nanotube film, the chiroptical signal does not vanish when the exciton bleaches — it moves, to lower energy, into the trion.
That robustness is the paper’s real theme: chiral trions exist in enantiopure single-wall carbon nanotubes. The third particle does not erase the handedness. It inherits it, quantitatively, to within experimental uncertainty.
Sources and further reading
Fujinami, H., Kusaba, S., Konabe, S., Gao, W., Zaumseil, J. & Yanagi, K. “Chiral trions in enantiopure single-wall carbon nanotubes.” Nature Communications 17, 11235 (2026). Published 30 September 2026; received 8 May 2026, accepted 14 September 2026. DOI: 10.1038/s41467-026-78073-z. Open access. Affiliations: Tokyo Metropolitan University (Fujinami, Kusaba, Yanagi); Hosei University (Konabe); University of Utah (Gao); Heidelberg University (Zaumseil). Data deposited on Figshare; source data provided with the paper. The authors declare no competing interests.
For background on nanotube exciton CD and enantiomer separation, see Wei et al., Nature Communications 7, 12899 (2016). For the contrast case of quantum-defect CD in nanotubes, see Sebastian et al., Nature Communications 16, 5107 (2025). This publication’s earlier primer, “Trions, Biexcitons, and the Exciton Zoo,” covers trion binding, gate dependence and the exciton-polaron picture in broader context.