A “happy accident” has yielded a new, stable form of pure carbon made from cheap feedstocks, researchers say. Like diamond and graphene, two other guises of carbon, the material seems to have extraordinary physical properties. It is harder than stainless steel, about as conductive, and as reflective as a polished aluminum mirror. Perhaps most surprising, the substance appears to be ferromagnetic, behaving like a permanent magnet at temperatures up to 125°C. The discovery, announced by physicist Joel Therrien of the University of Massachusetts in Lowell on 4 November here at the International Symposium on Clusters and Nanomaterials, could lead to lightweight coatings, medical products, and novel electronic devices.
Therrien’s talk elicited both excitement and caution among the dozens of researchers attending the meeting. “Once it is published and the work has been replicated by others, it will generate a lot of interest for sure,” says Qian Wang, an applied physicist at Peking University in Beijing. She notes that carbon is much lighter than other ferromagnetic elements such as manganese, nickel, and iron. Moreover, carbon is nontoxic in the body, she says. “If it can be magnetic, it could be very useful for making biosensors or drug-delivery carriers” that could be magnetically interrogated or directed to diseased tissues.
Robert Whetten, a materials scientist at Northern Arizona University in Tucson, says he’s “actually persuaded they have something.” He recalls that in the mid-1980s, when scientists first created “buckyball” spheres made of 60 carbon atoms, “there was the same degree of skepticism, despite all the evidence.” But he also points out that past claims of magnetism in pure carbon, such as one published in Nature in 2001, have fallen apart when contamination was later found in the samples.
The researchers have only made thin films of the material, which they have studied with electron microscopes and x-ray spectrometers. “Much more characterization needs to be done,” says Pulickel Ajayan of Rice University in Houston, Texas. “We are trying to be extremely careful.” He and the other researchers say they have so far seen no signs of impurities that could account for the properties of the material. Based on both theoretical models and the analytical data, they believe it consists of corrugated layers of linked carbon atoms stacked like washboards, with additional bonds between the layers.
Sumio Iijima, a nanomaterials expert at Meijo University in Nagoya, Japan, renowned for his 1991 discovery of carbon nanotubes, another carbon “allotrope,” says the limited data presented are “not good enough” to convince him the group has found a new allotrope. He wants the team to perform x-ray crystallography, a gold standard technique to determine structure, on larger samples.
Therrien says the discovery came in a failed attempt to synthesize pentagraphene, a sheet of carbon atoms bound in pentagonal rings that has been predicted but never created. His idea was to exploit a technique known as “geometrical frustration.” He placed a catalyst, a sheet of copper foil, on a pedestal in the center of a chemical vapor deposition (CVD) oven and heated it to about 800°C. But rather than pumping in a feed gas of the usual small hydrocarbons, such as methane, he injected a more complex precursor: 2,2 dimethylbutane, a cheap petrochemical that is available by the ton. As a branched hexane, the chemical has six carbon atoms arranged along a bent backbone.
The high temperatures of CVD normally break down complex hydrocarbons at the catalyst surface into atoms or simple molecules, which then self-organize into graphene sheets or crystals such as diamond. But previous experiments led Therrien to believe the hexane’s crooked backbone would persist, creating a geometric constraint that would force the “frustrated” carbon groups to link up into the pentagonal cells of pentagraphene.
They did not—but the atoms settled into a different structure. Therrien says that on 13 November 2017, after teaching an evening class, he returned to his lab to check on his oven and noticed the smell of tar. The inside of the furnace was caked in black pitch. But the copper foil was covered in something that looked like polished silver. “I just stopped dead,” he says. “I literally spent half an hour just staring at this, trying to figure out what on Earth had happened.”
After 2 years of further experimentation with other branched hexanes, the team was able to reliably create the substance again and again, in films up to 1 micron thick and several centimeters across. Puru Jena, a collaborator at Virginia Commonwealth University in Richmond, says models of molecular dynamics predict that the carbon atoms form corrugated layers tiled with six- or 12-atom rings and linked together by covalent bonds. Graphite, in contrast, consists of flat carbon layers that freely slide across each other.
Ajayan says the x-ray and electron-microscope data support the predicted structure. He says those data have also confirmed another prediction of Jena’s: that the bonds in the material are formed by shared electrons in a mix of particular orbital shells—what chemists call sp2 and sp3 bonds. The sp2 bonds leave electrons “not involved in any bond and kind of dangling,” which makes it easier for their spins to align, Jena says. “That gives you the ferromagnetism.”
At the meeting, Therrien dramatized the magnetism by showing a video of millimeter-size flakes of the substance suspended in a water droplet. As he slowly waved a small bar magnet over the water, the flakes followed it back and forth. More tantalizing still is his claim that the material remains magnetic even at the elevated temperatures at which motors and computers typically operate.
The magnetism adds to a suite of properties never before seen together in a form of pure carbon. They include tremendous hardness that presumably results from the bonds joining adjacent layers: “We’ve tried scratching it with steel wool, and it comes off clean,” Therrien says. “The only thing we can say verifiably scratches it is a diamond scribe.” Though the group has yet to measure the tensile strength of the material, the fact that vanishingly thin flakes hold together at millimeter size suggest it may be as strong as some metals, he says.
Then there is the mirrorlike appearance, seen in photos Therrien showed at the meeting. The team’s measurements indicate that the film, even when just 50 nanometers thick, reflects more than 90% of incoming light at wavelengths ranging from the far-ultraviolet to the midinfrared. That attribute could make it a useful reflective coating, more durable than the standard aluminum, for mirrors in cameras and telescopes.
Its electrical conductivity turned out to be just shy of that of stainless steel. But it can also display other electronic properties. Annealing the material by slowly heating it to 1000°C dims its shine and turns it into a semiconductor with a band gap—the energy required to liberate an electron—similar to that of amorphous silicon, which can turn light into electricity. That makes it a candidate material for photovoltaic cells, Therrien suggests.
Therrien is bullish about the long-term potential of geometrical frustration to synthesize novel allotropes of carbon and other elements. “Even if it works only for carbon, the very fact that there are probably hundreds of different allotropes that you might be able to make using this approach is going to really open things wide up.”
The group has not yet settled on a name for the mystery material. Jena calls it U-carbon—U for unusual. But Therrien, inspired by medieval alchemists who sought in vain for “adamant,” an unbreakable lodestone, is calling it adamantia.
*Correction, 15 November, 11 a.m.: An earlier version of this story incorrectly suggested U-carbon was the first carbon allotrope to display ferromagnetism at relatively high temperatures, when in fact other forms of pure carbon have displayed that behavior.