Pentagon-heptagon carbon nanoribbon conducts like a metal, in short pieces

Lifted off a silver crystal by a microscope tip, ribbons of carbon pentagons and heptagons conducted about 0.5 to 0.9 of the maximum one quantum channel allows, though the pieces were short and made in low yield.

Peer reviewed · Laboratory measurement on a metal surface · One research group

Researchers at the University of Science and Technology of China (USTC) in Hefei have built a carbon nanoribbon from five- and seven-membered rings, with no hexagons, and report in Nature that it behaves as a metal. Very narrow graphene ribbons are usually semiconductors, so a metallic all-carbon strip could one day serve as nanoscale wiring. This “pentaheptite” carbon was predicted to be metallic in 1996. The catch: the ribbons came in low yield, as short pieces on silver, and one peer reviewer was not convinced they are intrinsically metallic.

Evidence at a glance

Is this a big deal?

For chemists, a solid step: a carbon structure predicted 30 years ago now exists in ribbon form, made by a new rearrangement route. For electronics it is an early hint. Nobody has wired one into a circuit, and the authors say the synthesis needs more work first.

What scientists found

Graphene is a sheet of carbon hexagons. Cut it into a strip a few atoms wide and a band gap, an energy range electrons cannot occupy, usually opens, making the strip a semiconductor.

The USTC team, led by Bing Wang and Qitang Fan, changed the rings instead. Each ribbon has a zigzag chain of heptagons down the middle, flanked by two chains of pentalene, a pair of fused pentagons. In CPHN 1 the two pentalene chains point the same way. In CPHN 2 they point opposite ways, and that alone flips the ribbon from metal to semiconductor.

The paper calls CPHN 1 a covalent metal. Its atoms share electrons in bonds, as in graphene or diamond and lonsdaleite, rather than sitting in the sea of loose electrons that binds copper. Yet one of its electron bands is only partly filled, so electrons at the conduction energy, the Fermi level, move freely.

How they found it

Chemists at Nankai University in Tianjin made the starting molecule, an eight-carbon ring carrying two bromine atoms. Deposited on silver at room temperature (300 K), the molecules linked into chains. Heating to 623 K (about 350 °C) rearranged each eight-membered ring into a pentalene unit, and two hours at 773 K (500 °C) fused neighbouring chains side by side into ribbons.

To read the electronics, the team slid bromine atoms under the ribbons to loosen the silver’s grip, then mapped electron standing waves along them. A short ribbon traps electrons the way a guitar string traps a vibration: only certain wavelengths fit. Matching wavelength to energy traces the band. For CPHN 1 it dips below the Fermi level, the signature of a metal. The analogy has a limit: a string can be any length, but these ribbons were about 10 nm, so their levels are discrete and “metal” strictly describes the long-ribbon limit.

Next, the STM tip grabbed one end of a ribbon and peeled it 1 to 5 nm off the surface while current flowed. One CPHN 1 ribbon carried about 0.9 G₀, where G₀, the conductance quantum, is the most one perfect quantum channel can pass. That is a resistance of about 14 kΩ. Others gave about 0.5 G₀, or jumped between 0.3 and 0.8 G₀ as the tip’s grip changed, and conductance barely fell over the first 3 nm of lifting.

Why it matters to you

This won’t change your phone or laptop. Chip wiring is copper and other metals; the abstract names electrodes and interconnects as possible future uses for metallic carbon ribbons. That would need long, defect-free ribbons made reliably, which this work does not deliver.

What the evidence does not show

There is no device. Every measurement was made in vacuum on silver, or with a ribbon dangling between the silver and an STM tip.

Yield is low. By the Supplementary Information’s count, 17.4% of pentalene units ended up in ribbons, and 13.5% in segments at least six repeat units long. The best segments ran past 20 units, about 10 nm: 10,000 end to end would span just 0.1 mm.

A strong signal at zero voltage in the lifting tests is also unexplained; the authors tentatively blame the tip contact.

How this fits what we already knew

Vincent Crespi, Steven Louie and colleagues predicted in 1996, in Physical Review B, that a flat sheet of pentagons and heptagons would be a metal. Fan and co-workers made ribbons containing 5–7 ring pairs by fusing polyazulene chains in 2019, and a Berkeley team led by Daniel Rizzo made graphene ribbons metallic in 2020 by engineering their edges. Carbon nanotubes were shown in 1998 to conduct in quantized steps of G₀. The new ribbons reach metallicity a different way: through ring shape alone.

Independent perspective

SciQuest found no on-record comment from an independent researcher as of 29 October 2026, 08:19 UTC, and has not yet interviewed one. The published peer-review file shows a split among the anonymous referees. Through three rounds, Referee 2 questioned the yield, the ribbon quality and the electronic interpretation, concluding that “the conclusion of intrinsic metallicity is not convincingly demonstrated.” Referee 3 agreed that quality and yield could improve but found the band crossing the Fermi level convincing, and recommended publication. Referee 1 supported publication after revisions.

How other headlines framed it

The only other coverage found is Nature‘s own Research Briefing. Its freely readable summary describes metal-like transport and a route to highly conducting ribbons, without the low yield or short segments; the rest is paywalled. The referees’ disagreement is public only in the peer-review file.

What happens next

The team tried six other copper, silver and gold surfaces without beating silver’s (111) face, and heating to 823 K broke the rings. The authors say the synthesis must improve before devices. Longer ribbons would let other groups test the metallic band directly, and the raw data are public on Zenodo.

In one sentence

Rearranging carbon’s rings into pentagons and heptagons produced a narrow ribbon that conducts like a metal, though so far only in short pieces on silver.

Read next

• Single-atom NMR with an STM tip: the same kind of microscope, used to probe one atom’s nucleus.
• Is lonsdaleite stronger than diamond? Another case of carbon’s properties changing with how its atoms are arranged.
• Scientists reconstruct an exciton’s first 400 femtoseconds: more on how electrons move in nanoscale materials.

Sources and disclosure

1. Huang, J., Fan, Q., Meng, J. et al. “Narrow carbon pentaheptite nanoribbon as a covalent metal.” Nature 658, 1176–1182, published online 28 October 2026. doi:10.1038/s41586-026-11202-2. Read: abstract, figure and Extended Data captions, Supplementary Information and the CC BY 4.0 peer-review file (source of the referee quote).
2. Nature Research Briefing. “Ribbon made of carbon pentagons and heptagons has metallic properties.” 28 October 2026. doi:10.1038/d41586-026-03298-3.
3. Crespi, V. H., Benedict, L. X., Cohen, M. L. & Louie, S. G. “Prediction of a pure-carbon planar covalent metal.” Physical Review B 53, R13303 (1996). doi:10.1103/PhysRevB.53.R13303.
4. Rizzo, D. J. et al. “Inducing metallicity in graphene nanoribbons via zero-mode superlattices.” Science 369, 1597–1603 (2020). doi:10.1126/science.aay3588.
5. Fan, Q. et al. “Nanoribbons with nonalternant topology from fusion of polyazulene.” Journal of the American Chemical Society 141, 17713–17720 (2019). doi:10.1021/jacs.9b08060.
6. Frank, S., Poncharal, P., Wang, Z. L. & de Heer, W. A. “Carbon nanotube quantum resistors.” Science 280, 1744–1746 (1998). doi:10.1126/science.280.5370.1744.

Disclosure: This article was drafted with AI assistance and checked against the sources above; the paper’s main text is paywalled, so details come from its abstract, captions, Supplementary Information and peer-review file. The research was funded by Chinese national and Chinese Academy of Sciences programmes and the New Cornerstone Science Foundation; the authors declare no competing interests. Calculations by SciQuest: G₀ = 2e²/h ≈ 77.5 µS, so 1/(0.9 × 77.5 µS) ≈ 14.3 kΩ; 623 K ≈ 350 °C, 773 K ≈ 500 °C; 10,000 × 10 nm = 0.1 mm. The graphic is a SciQuest original drawn from the paper’s values. Spotted an error? See our corrections policy.