The numbers, stated plainly: a red quantum-dot-electrolyte LED with 24.2% external quantum efficiency and a T95 lifetime of 13,100 hours at 1,000 cd/m²; a green device at 20.3% and 14,400 hours at the same brightness; and a blue device at 13.5% with a T95 of 244.5 hours at 1,000 cd/m². The paper, “Efficient and stable inverted RGB quantum-dot-electrolyte LEDs for active-matrix displays,” by Yunfei Ren, Baiquan Liu, Chuan Liu, and colleagues at Sun Yat-sen University and the South China University of Technology, was published in Nature Communications on 2 November 2026 (DOI: 10.1038/s41467-026-79228-8). Its claim for the blue device is carefully scoped: a record stability for solution-processed blue LEDs without light-outcoupling technologies. But the number that matters most is that all three colors now exist in the same device family, built the same way, in the same paper.
One caveat belongs up front. The figures here come from the paper’s abstract; the full text could not be verified at the time of writing. The ionic liquid’s identity, the insulating layer’s material, the emission wavelengths, peak luminance, the blue device’s roll-off, and the display pixel data are therefore not stated here. Where the 2025 precursor paper names a material, this article attributes it to that paper.
Why QE-LEDs stopped at red
The quantum-dot-electrolyte LED is barely twenty months old. The same Sun Yat-sen group introduced it in Advanced Materials in March 2025, in a paper that reported only red devices. The idea was a deliberate hybrid: a conventional quantum-dot LED is a sandwich of charge-transport layers around a quantum-dot emitting layer, and its weak point has always been charge balance — electrons flood in easily through the oxide transport layer while holes struggle in from the other side. A light-emitting electrochemical cell uses mobile ions to build up charge at the interfaces, which helps injection, but it is slow and hard to control. The QE-LED takes that ion trick and puts it inside a QD-LED stack: dope an ionic liquid into the quantum dots, and under bias the mobile ions migrate to the interfaces and form electrical double layers that sharpen the local field and pull carriers into the dots.
The 2025 red device showed the trick worked. With an insulating polymer, PMMA, mixed into the emitting layer to choke off excess electron flow, it reached 20.5% external quantum efficiency and a projected T95 lifetime above 374,000 hours at 100 cd/m². The group also built a red active-matrix display on a commercial thin-film-transistor backplane, with uniform emission that held up after 400 hours of continuous operation at high brightness. It was a strikingly complete first demonstration — and it was red only. The reason was not the emitters; red, green, and blue quantum dots are all mature materials. The blockage was process: every new layer in a solution-processed stack is deposited from a solvent that must not dissolve the layer beneath it, and the ionic liquid that makes the QE-LED work is itself a polar, mobile species. In the conventional 2025 stack, that constraint could be managed for one color; extending it to green and blue — whose dots carry different surface ligands and sensitivities — meant the stack itself had to change.
The flip: an inverted stack removes the solvent conflict
The new paper’s first move is architectural. In an inverted LED stack, the layer order is reversed: the electron-transport layer goes down first on the cathode, the emitting layer sits on top of it, and the hole-transport layer and anode are deposited last. Here the abstract assigns the inversion a specific job: it “eliminates solvent compatibility issues,” and that elimination is what enables green and blue QE-LEDs, which had not been reported before. The abstract does not spell out which solvent conflict the inversion resolves, and the full text was not accessible to check. What can be said is that the constraint it removed was the binding one: the same ionic-liquid doping that made the red device work could not previously be carried into green and blue stacks, and with the stack flipped, it can — producing a single architecture for all three colors rather than three separate recipes, the precondition for a display, where red, green, and blue pixels must be fabricated side by side in one process flow.
The second change is more surgical. The paper places an insulating layer between the emitting layer and the electron-transport layer, specifically to suppress excessive electron leakage — a different placement from the 2025 fix, where the insulating polymer PMMA was mixed into the emitting layer itself. The problem is the same, electrons arriving faster than holes and leaking through the dots without recombining, but the solution has moved from inside the emitter to the interface where the leak happens. An insulating layer at that boundary throttles the electron current while leaving the emitting layer’s chemistry untouched. The abstract does not name the insulating material, so this article does not either.
The ions: a double layer that does the injecting
The core of the device remains the ionic liquid doped into the quantum dots. The abstract describes the resulting electrolyte emitting layer as enhancing charge injection and transport, and the 2025 paper gives the mechanism: under bias, the positive and negative ions migrate to opposite electrodes and accumulate at the interfaces, forming electrical double layers that concentrate the applied voltage into a nanometer-scale region — a local field far stronger than the average across the device, strong enough to pull holes into the dots.
The 2025 paper named its ionic liquid: tetrahexylammonium tetrafluoroborate, THA-BF4. The new paper’s abstract does not name the ionic liquid it uses, and it would be wrong to assume it is the same one — the green and blue devices may need a different ion pair, a different loading, or both. What carries over is the design rule: mobile ions inside the emitting layer, forming double layers under bias, doing work that a static stack cannot. That rule is now demonstrated across three colors rather than one. And that generalization was not guaranteed: blue quantum dots are the hardest emitters in any LED family, with deeper energy levels, larger hole-injection barriers, and more energetic excitons that find more ways to decay without emitting light. That the same ionic-liquid approach lifts blue to 13.5% efficiency is evidence that the double-layer field enhancement is doing real work at the interface, not merely polishing an already easy red device.
Three colors, three scorecards
The device figures, all from the paper’s abstract, deserve to be read as a set. Red: 24.2% EQE and a T95 of 13,100 hours at 1,000 cd/m², which the abstract says surpasses previous best red QE-LEDs. Green: 20.3% EQE and a T95 of 14,400 hours at 1,000 cd/m². Blue: 13.5% EQE and a T95 of 244.5 hours at 1,000 cd/m².
Two things stand out. The first is the test condition. T95 — the time for luminance to fall to 95% of its initial value — is a stricter metric than the T50 half-life commonly quoted for LEDs, and 1,000 cd/m² is a demanding initial luminance, rather than the 100 cd/m² at which the 2025 red device’s lifetime was quoted. That matters when comparing the two red results: the 2025 paper’s projected T95 of more than 374,000 hours was measured at 100 cd/m², while the new red figure of 13,100 hours is at ten times that luminance. The two numbers are not on the same scale; the new figure is the more demanding measurement, not a regression. The second is the shape of the set. Red and green are now in the same league — efficiencies above 20% and T95 lifetimes above ten thousand hours at display brightness, with green edging red, 14,400 hours against 13,100 — while blue is the outlier on both axes, lower in efficiency and shorter-lived by a factor of fifty. That gap is the honest shape of the field: blue is the color that has gated every display technology, from OLED to perovskite to conventional QD-LED, and a blue device merely worse than its siblings by a moderate factor, rather than by orders of magnitude, is itself the news.
Blue: the number to watch, and how to read it
A 13.5% external quantum efficiency for a solution-processed blue LED is solid but not by itself extraordinary — blue QD-LEDs and blue perovskite LEDs have both reported higher peak efficiencies. What is extraordinary is the stability figure attached to it: a T95 of 244.5 hours at 1,000 cd/m², claimed as a record stability for solution-processed blue LEDs without light-outcoupling technologies. The scope of that claim needs to be kept intact: “solution-processed” excludes vacuum-deposited OLEDs, the commercial blue standard; “without light-outcoupling technologies” excludes devices whose headline numbers depend on lenses, scattering layers, or nanostructures added to extract more light; and T95 is the strict metric, the device holding 95% of its brightness rather than 80% or 50%.
For scale: a 2025 Advanced Materials pure-blue perovskite LED reached 20.3% EQE at 469 nm — higher efficiency than the QE-LED blue — but its reported lifetime was a T80 of 4.8 hours at 1,070 cd/m², a looser metric at similar brightness. The QE-LED blue gives up efficiency and gains roughly fifty times the operating time under a stricter definition of failure — the right direction for a display, where a pixel must hold its brightness for years rather than hours. The caveats are real, though. A T95 of 244.5 hours is still far short of a commercial specification, measured in tens of thousands of hours, and the abstract gives no emission wavelength, so there is no way to check the device against the Rec. 2020 blue primary window of roughly 460 to 475 nanometers. Efficiency roll-off — whether the 13.5% holds at display brightness or only at low current density — is likewise not stated in the accessible text. The record claim is a stability claim at a stated luminance under a stated metric, and it should be quoted that way, not inflated into a claim that blue is solved.
The displays: a set, not a single color
The paper’s final claim is that the group built RGB active-matrix displays from these devices, “exhibiting excellent stability and uniformity” — the difference between three good laboratory devices and a display technology. An active-matrix display means the LEDs are integrated with a thin-film-transistor backplane that addresses each pixel individually, the architecture the 2025 paper demonstrated in red. Doing it in RGB means the inverted stack and the ionic-liquid process work for all three colors in a single fabrication flow, the minimum requirement for any display product. What the abstract does not provide is any pixel data that would let a reader judge the demonstration: no display size, no resolution or pixel pitch, no driving voltage or uniformity figure, no operating lifetime for the panel as a whole. “Excellent stability and uniformity” is the authors’ summary, not a number.
What this is not — yet
First, it is not a commercial display: no cost analysis, no yield data, and no large-area fabrication result in the accessible text. Second, it is not a solved blue problem: the blue device’s 244.5-hour T95 is a record in its stated scope and also roughly sixty times shorter than the green device’s lifetime in the same paper.
Why completing the set is the story
Strip away the device numbers and what remains is a design rule other groups can test: dope mobile ions into the emitting layer so that double layers form under bias and do the injecting; throttle the electron current with an insulating layer at the transport interface rather than inside the emitter; and flip the stack so the solvents stop fighting the layers. A red-only device, however efficient and long-lived, is a materials result — it demonstrates a trick. An RGB set in one architecture, with an active-matrix display behind it, demonstrates a process: the same stack, the same ionic-liquid step, the same insulating layer, producing three colors that a display needs side by side. The blue device is the proof that the process extends to the hard color, and its stability figure is the first blue number in this device family that can be discussed in the same breath as a display requirement rather than as a laboratory curiosity.
Sources and further reading
Ren, Y., et al. “Efficient and stable inverted RGB quantum-dot-electrolyte LEDs for active-matrix displays.” Nature Communications (2026). Published 2 November 2026. DOI: 10.1038/s41467-026-79228-8. All figures for the new devices — red 24.2% EQE and 13,100-hour T95, green 20.3% and 14,400-hour T95, blue 13.5% and 244.5-hour T95, all at 1,000 cd/m² — are from this paper’s abstract; the full text was not accessible at the time of writing, so the ionic liquid identity, insulating-layer material, emission wavelengths, peak luminance, roll-off, and display pixel data are not stated here.
Ren, Y., et al. “Quantum-Dot-Electrolyte Light-Emitting Diodes for Displays.” Advanced Materials 37, e2417330 (2025). DOI: 10.1002/adma.202417330. Cited for the 2025 red-only QE-LED: the THA-BF4 ionic liquid, the double-layer mechanism, the PMMA insulating polymer, the 20.5% EQE, the projected T95 above 374,000 hours at 100 cd/m², and the red active-matrix display, stable after 400 hours.
Yao, Z., et al. “Decoupling the Exciton-Carrier Interaction for Highly Efficient Pure Blue Perovskite Light-Emitting Diodes Exceeding 20%.” Advanced Materials 37, e20131 (2025). DOI: 10.1002/adma.202420131. Cited for the pure-blue perovskite comparison: 20.3% EQE at 469 nm, T80 of 4.8 hours at 1,070 cd/m².
Park, J., et al. “Phase Regulation via Dual Additives for Pure-blue Emitting Quasi-2D Perovskite Light Emitting Diodes.” Faraday Discussions (2026). DOI: 10.1039/D6FD00036C. Cited for the Rec. 2020 pure-blue window of roughly 460–475 nanometers.