Author: Lisa Ercolano
Template clusters used for the data driven method to expedite the filling of the database.

A first-of-a-kind database of quantum nanoclusters created by Johns Hopkins materials scientists has the potential to facilitate the development of new technologies ranging from highly efficient catalysts and improved sensors to data storage and photodynamic therapy. The nanoclusters range in size from a few- to several hundred atoms, and have unique electrical, magnetic, optical, and reactivity properties.

The project, funded by a Multidisciplinary University Research Initiatives (MURI) grant from the Office of Naval Research and led by  Tim Mueller,  an assistant research professor in the Whiting School of Engineering’s Department of Materials Science and Engineering, is reportedly the most extensive collection of computed cluster structures ever assembled.

A nanocluster is a tiny group of atoms that is larger than a single atom but smaller than a full-fledged nanoparticle. Their unique properties also make them a promising option for the development of cost-effective catalysts that minimize the use of expensive elements. Mueller’s team’s database provides detailed information about these physical properties, which should aid researchers in the development of innovative nanocluster-based technologies.

“We’re very excited for people to make use of the tool and to see the sorts of new discoveries it facilitates,” Mueller said. The team’s research was published in Nature Scientific Data.

In the process of amassing the database, Mueller’s team discovered several new low-energy clusters and was able to define the energies, relaxed structures, and physical properties of more than 50,000 clusters across 55 elements. This number included 593 structures observed to have energies lower than any previously on record, as well as 1,320 entirely new cluster structures.

“Nanoclusters can be significantly altered, even when their sizes fluctuate by merely a few atoms. This characteristic creates tons of possibilities for fine-tuning them to derive diverse properties. Although our database primarily comprises elemental clusters, those can serve as a foundation for more complex nanoclusters,” said study lead author Yunzhe Wang, a doctoral student in materials science and engineering.

Wang offers as an example the creation of small groups of metal atoms called ligated metal clusters that have useful applications like making light-emitting devices or for medical imaging.

“You can do this by attaching special molecules to the metal core. By changing the number and type of these molecules, we can change the properties of the clusters. We can also make metal clusters with different types of metals by swapping out certain atoms with others,” Wang said.

The small size of the clusters Wang and Mueller identified in the database gives them distinct properties that can’t be found in larger materials, and the variety of structures provides researchers with a wide range of new possibilities to explore.

“Our data set can be used to guide the experimental synthesis of predicted nanoclusters, help identify low-energy clusters in different chemical environments, computationally screen for clusters suitable for a variety of applications, or train machine learning models,” Mueller said.