Researchers achieve room-temperature ferromagnetism in wafer-scale porous graphene

A team of researchers spanning eight institutions across Asia has grown a sheet of porous graphene that behaves like a magnet at room temperature—a property that carbon materials have long resisted, and one that opens a path toward carbon-based electronics that exploit magnetic spin. The work, conducted by scientists at Shenzhen University, the Japan Atomic Energy Agency, the National Institutes for Quantum Science and Technology, the University of Tokyo, the University of Science and Technology of China, Xi'an Jiaotong University, Shenzhen Technology University, and The Hong Kong Polytechnic University, represents the first time this kind of intrinsic magnetism has been confirmed in a bulk, wafer-scale carbon system at room temperature.

The material itself is unusual in structure. Rather than lying flat like conventional graphene films, the layers of this hyper-porous graphene network stand nearly perpendicular to the silicon substrate beneath them, weaving together into a three-dimensional lattice riddled with tiny pores—each smaller than about 2 nanometers across. The researchers grew it across an entire 2-inch silicon wafer using a plasma-assisted vapor deposition process combined with low-energy electron irradiation. What emerged was a network whose magnetic strength reached roughly 1,000 times that of defected graphite, a conventional reference material. At room temperature, the porous network achieved a saturation magnetization of 4.7 emu/g, translating to about 0.06 microbohrs of magnetic moment per carbon atom.

The critical question in any claim of magnetic carbon is contamination. Trace amounts of iron, cobalt, nickel, or other magnetic metals can masquerade as intrinsic carbon magnetism. The team addressed this head-on using X-ray magnetic circular dichroism tuned to the carbon absorption edge—a technique sensitive enough to detect the magnetic signature of carbon itself. The signal reversed as expected when the magnetic field or X-ray polarization flipped, while both conventional graphite and an amorphous carbon control sample produced no comparable response. Scans at the characteristic energies of iron, cobalt, and nickel found none of their signatures. Mass spectrometry analysis showed that trace magnetic impurities in the sample could account for at most 0.23 percent of the measured magnetization, ruling out contamination as the source.

The researchers also built in a structural control. Using the same deposition equipment under different conditions, they produced an amorphous carbon film with a far more disordered atomic arrangement and no comparable network of exposed edges. This material, made in the identical growth environment, showed no magnetic signal—a direct comparison that strengthened the case that the porous structure itself, not some artifact of the process, was responsible for the effect.

When the team annealed the porous network, heating it to collapse the three-dimensional mesh into a denser structure, the magnetic signal vanished and magnetization dropped to the level of the amorphous control. This confirmed that the pore geometry was essential. Atomic-resolution transmission electron microscopy had already revealed the source: the pores exposed abundant zigzag edges of graphene, many of them separated by more than roughly 1.4 nanometers. Density functional theory calculations explained why this spacing mattered. When opposing zigzag edges sit far enough apart—around 1.4 nanometers or more—their unpaired electron spins favor parallel alignment and reinforce each other magnetically. Closer spacing flips the preferred alignment and causes the magnetic moments to cancel out. The geometry of the pores, not simply the abundance of edges or defects, determines whether the material's magnetism adds up or cancels.

This finding extends a mechanism that had been demonstrated only in individual graphene nanoribbons and small nanographene molecules to a scalable, wafer-scale film. The authors describe it as the first element-resolved confirmation of intrinsic carbon magnetism at room temperature in a bulk system. Because pore geometry appears to be the controlling variable, the researchers suggest that if that geometry can be reliably tuned during growth, the resulting magnetic state could become tunable as well. That possibility points toward a new frontier: carbon-based spintronics and magnetic devices that operate at room temperature and could eventually reach commercial scale.