THE STORY IN BRIEF
  • The team examined 1,183 possible sites in a model of monolayer amorphous carbon.
  • Local ring geometry changes the calculated strength of hydrogen binding.
  • A model of promising sites does not establish the performance or cost of a working electrolyser.
MAKE TIME FOR A GOOD READ
Inside this story 2 perspectives

A perfect drawing of graphene looks like a page of hexagonal graph paper. A sheet of amorphous carbon has no such repeating pattern. To a scientist looking for a surface that will briefly hold hydrogen and then let it go, the untidiness may be an advantage. On 16 September India's Press Information Bureau described work by researchers at the Indian Institute of Technology Gandhinagar who investigated 1,183 possible reaction sites in a computer model of an irregular, one-atom-thick carbon sheet.

Their paper, published in npj 2D Materials and Applications, addresses one part of the problem of producing hydrogen by splitting water. A catalyst helps the chemical reaction proceed, and efficient systems often rely on costly metals. Carbon is an attractive material to investigate for this purpose. Knowing that a surface is made of carbon, however, says little about how strongly any one part of it will hold hydrogen during the reaction.

Why some imperfections matter

Imagine mapping an identical point in each hexagon across a perfectly repeating sheet. The local surroundings would be similar from one point to the next. The researchers' disordered material contains rings of five, six and seven carbon atoms in different arrangements. That changes the local geometry, and the geometry in turn changes how hydrogen interacts with a site.

There is a balance to find. If hydrogen binds very weakly, the site offers little help to the reaction. If it binds too strongly, releasing it becomes difficult. Researchers can calculate a quantity known as adsorption free energy to compare sites. Values near zero are considered promising for this particular step. A single average for the entire sheet can obscure a small number of sites with more favourable behaviour.

The group combined detailed calculations with a machine-learning approach to cover more sites than would be practical using the most expensive calculation on every one. The result is a map of local environments worth investigating. It gives experimental researchers a more specific problem than whether carbon in general makes a good catalyst: which patterns are useful, how common are they, and can a real sample retain them in use?

Distance to a working device

A promising value on a computer screen cannot establish how fast a complete electrolyser will produce hydrogen, how long a physical surface will last or how much the resulting fuel will cost. The model mainly helps explain why different sites made of the same element may behave differently. That is valuable information before a laboratory commits time to making and testing a material.

The study also offers a modest change in how to think about making materials well. In an application where repeating geometry gives too few useful places for a reaction, perfect regularity may be the wrong target. The test now is whether experimentalists can turn some of the calculated advantages of an irregular sheet into a property they can measure repeatedly.

BB
ABOUT THE AUTHOR

Beyond Business Editors

Editorial desk

Beyond Business connects global developments with the decisions, ambitions and everyday lives of its readers. Our editorial desk explores the stories shaping our world.

More from this author
Keep the perspective.