First UAV test flight with Cella solid-state hydrogen storage and fuel-cell power system
New SwRI consortium to target advances in gasoline and diesel engine aluminum cylinder head designs

Berkeley Lab scientists find that polycrystalline graphene is not very resistant to fracture

Scientists at the US Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab) have developed the first known statistical theory for the toughness of polycrystalline graphene, which is made with chemical vapor deposition, and found that it is indeed strong (albeit not quite as strong as pristine monocrystalline graphene), but more importantly, its resistance to fracture is quite low. Their study, was published recently as an open-access paper in Nature Communications.

Graphene, a material consisting of a single layer of carbon atoms, has been touted as the strongest material known to exist, 200 times stronger than steel, lighter than paper, and with extraordinary mechanical and electrical properties.

This material certainly has very high strength, but it has particularly low toughness—lower than diamond and a little higher than pure graphite. Its extremely high strength is very impressive, but we can’t necessarily utilize that strength unless it has resistance to fracture.

—Berkeley Lab scientist Robert Ritchie

Ritchie-graphene
Polycrystalline graphene contains inherent nanoscale line and point defects that lead to significant statistical fluctuations in toughness and strength. (a) A one-atom-thick polycrystalline graphene sheet composed of carbon atoms arranged in regular hexagonal rings, except at the grain boundaries (GBs) and triple junctions (TJs). The defected atoms at the GBs and TJs that are part of non-hexagon rings are drawn in red for clear identification.

(b) A zoom-in of the area marked in a. The red arrows indicate the orientation of the grains on either side of the GB; the GB itself is composed of rings of five (colored pink) and seven (colored blue) carbon atoms. These are the dislocation cores with the shortest Burgers vectors in graphene, and thus represent low-energy configurations of GBs. A TJ formed at the intersection of three GBs is highlighted.

(c) A perspective view of the graphene sheet showing its 3D structure.

(d) The principal residual stresses in the graphene sheet are in units of GPa. The grain interiors are defect free, while the GBs and TJs have significant residual stresses, and thus are the weak points where fracture nucleates.

(Credit: Berkeley Lab) Click to enlarge.

Ritchie, a senior scientist in the Materials Sciences Division of Berkeley Lab and a leading expert on why materials fail, was co-author of the study along with Ashivni Shekhawat, a Miller Research Fellow in his group. Together they developed a statistical model for the toughness of polycrystalline graphene to better understand and predict failure in the material.

The mathematical model takes into account the nanostructure of the material. The team found that the strength varies with the grain size up to a certain extent; most importantly, however, the model defines graphene’s fracture resistance, Ritchie said.

Toughness, a material’s resistance to fracture, and strength, a material’s resistance to deformation, are often mutually incompatible properties.

A structural material has to have toughness. We simply don’t use strong materials in critical structures—we try to use tough materials. When you look at such a structure, like a nuclear reactor pressure vessel, it’s made of a relatively low-strength steel, not an ultrahigh-strength steel. The hardest steels are used to make tools like a hammer head, but you’d never use them to manufacture a critical structure because of the fear of catastrophic fracture.

—Robert Ritchie

As the authors note in their paper, many of the leading-edge applications for which graphene has been suggested—such as flexible electronic displays, corrosion-resistant coatings, and biological devices—implicitly depend on its mechanical properties for structural reliability.

Although pure monocrystalline graphene may have fewer defects, the authors studied polycrystalline graphene as it is more inexpensively and commonly synthesized with chemical vapor deposition. Ritchie is aware of only one experimental measurement of the material’s toughness.

Our numbers were consistent with that one experimental number. In practical terms these results mean that a soccer ball can be placed on a single sheet of monocrystalline graphene without breaking it. What object can be supported by a corresponding sheet of polycrystalline graphene? It turns out that a soccer ball is much too heavy, and polycrystalline graphene can support only a ping pong ball. Still remarkable for a one-atom thick material, but not quite as breathtaking anymore.

—Robert Ritchie

Next, Shekhawat and Ritchie are studying the effects of adding hydrogen to the material.

The research was funded by the DOE Office of Science.

Resources

  • Ashivni Shekhawat & Robert O. Ritchie (2016) “Toughness and strength of nanocrystalline graphene” Nature Communications 7, Article number: 10546 doi: 10.1038/ncomms10546

Comments

Herman

Well...

Since almost every advancement in EV and Autonomous technology is by default led by Tesla (according to Henrik and others) I'm certain Tesla knew this. (Of course I've already seen plenty of claims that Tesla was planning on using grapheme so I guess they were only saying that to fool the oil companies or something).

Brotherkenny4

Fracture toughness can be measured by mechanical means. I fail to understand why a model, even if validated by experimental results, is superior to actual testing. I guess testing is mundane and anyone can do it, but if your a national lab you need to attached some equations to your papers.

Engineer-Poet

If you have an accurate model you know why something is, not just that it is.  It allows you to predict the consequences of changing things.

Perhaps there's some way to re-align polycrystalline graphene to make it monocrystalline.  That is something that would probably be tested out in models and only then tested to see how the model reflects reality.

The comments to this entry are closed.