A Chinese team has broken through part of the 'impossible triangle' of one-dimensional metal wires—vacuum-compatible, stable, and long. Published online August 20 in *Science* (Vol. 393, Issue 6813, pp. 826–830, DOI: 10.1126/science.aeg0028), a team led by Prof. Li Kuo at the Center for High Pressure Science and Technology Advanced Research (HPSTAR) has, for the first time, pushed the length of single-atom-diameter metal wires to the micrometer scale: individual chains exceed 1 μm with over 4,000 continuously arranged copper atoms—more than two orders of magnitude longer than the longest single-metal atomic chains (SMACs) previously reported internationally.
How it was done: high-pressure solid-state topochemical polymerization
For decades, SMACs have been trapped in a trilemma: vacuum-compatible, stable, and long—pick only two. Solution-synthesized SMACs rarely exceeded 10 atoms; a few Ni chains topped out at 28 before falling apart. Li's team's solution in one phrase: squeeze first, then lock in.
The precursor is β-copper phthalocyanine (β-CuPc), the main crystal form of phthalocyanine blue pigment—central copper ions are periodically arranged along the stacking direction, surrounded by 18π-electron unsaturated aromatic rings. When single crystals are compressed to ~0.9 GPa in a diamond anvil cell, β-CuPc undergoes a structural phase transition: the intermolecular slip angle drops from 45.5° to 28.6°, tightening the stacking. At 21.5 GPa the peripheral aromatic rings polymerize; at 25.0 GPa / 533 K held for 12 hours, an sp³-hybridized carbon skeleton forms a dense protective sheath around the copper atoms, freezing the pressure-aligned copper chains for ambient conditions.
To overcome the trace-sample limits of diamond anvil cells, the team used a Paris-Edinburgh press at ~40 GPa for bulk synthesis, yielding single crystals up to ~940 × 250 × 50 μm³. X-ray diffraction, selected-area electron diffraction, and HAADF-STEM jointly confirmed a Cu–Cu periodicity of 2.57 Å along highly ordered chains.
An acid-assisted ultrasonic exfoliation strategy pulled individual chains out of the lattice in 3% trifluoroacetic acid: exfoliated single chains exceed 1 μm, corresponding to 4,000+ continuous copper atoms. Even after hours of sonication and prolonged soaking in TFA, the chains remain intact—thanks to the sp³ carbon sheath.
'Sandwich structure' and anomalous electrical properties
The surprise is not the chain length but the physics. DFT calculations and measurements paint this picture: adjacent copper ions compressed to 2.57 Å still do not form significant Cu–Cu chemical bonds; instead the nanowire core forms a unique one-dimensional antiferromagnetic chain—Curie-Weiss temperature θ_CW = −34.2 K, confirmed by SQUID magnetization measurements.
The electrical behavior is even more interesting. Conductivity along the chain direction measures 5.5×10⁻³ S/m (limited by grain boundaries), while DFT predicts 2.5×10⁷ S/m for an ideal defect-free structure; perpendicular to the chains it is only 1.2×10⁴ S/m—anisotropy exceeding three orders of magnitude. Crucially, the conductive channel is not Cu-d electrons but p electrons of the underlying carbon skeleton—strong densification makes the p orbitals overlap along the chain, forming a continuous 1D charge-transport channel. Theorists summarize this as an 'insulating sheath + conductive interlayer + antiferromagnetic core' sandwich structure.
Generality: not just copper
The team tested CoPc, NiPc, ZnPc, and metal-free H₂Pc—after high-pressure treatment, all precursors produced structurally similar carbon-coated 1D polymeric structures. This means the strategy extends to cobalt, nickel, zinc and other metals, and could even construct mixed-metal atomic chains and 1D heterostructures, opening an experimental platform for molecular wires, spintronic devices, and single-electron transistors.
Why this matters
First, it solves the long-standing bottleneck of 1D condensed matter physics: the lack of 'long enough, stable enough, large enough' samples. Classic theories—Peierls transitions, Tomonaga-Luttinger liquids, 1D magnetism—have long lacked suitable experimental samples. Now they have them.
Second, it is a genuinely 'bottom-up' candidate for nanoscale circuits. As silicon-based processes approach physical limits, the road ahead leads to the 1D limit—single-atom-diameter metal wires are the material foundation of that route.
Third, academician Mao Ho-kwang's commentary highlights the outlook: 'Atom-scale ultrathin metal wires are expected to offer entirely new material options for next-generation nanocircuits, flexible wearable electronics, and more.'
The research was completed by HPSTAR jointly with Nankai University, Peking University, and South China University of Technology; PhD student Zhang Jie is first author and Li Kuo is corresponding author. The work was funded by the National Key R&D Program and the National Natural Science Foundation of China, with support from the Facility for Extreme Condition User Facility, Shanghai Synchrotron Radiation Facility beamlines BL17UM/BL14W1, and Japan's SPring-8 BL10XU beamline.
One-dimensional metal wires are no longer 'figures in papers'—they are real objects that can be exfoliated, measured, and arrayed.
References: *Science* online 8/20 (DOI: 10.1126/science.aeg0028), Xinhua 8/21, CCTV News, HPSTAR full-text release, China.com.cn, commentary by academician Mao Ho-kwang, *Science* Vol. 393, Issue 6813, pp. 826–830.