Luminescent Nanoparticles Reveal Surprising Clusters

Professor Remco van der Hofstad

By coating nanoparticles with specific molecules they can be made to function in new and exciting ways. New research shows that these molecules typically group together in distinct clusters.

Nanoparticles increasingly find their way into practical applications. By coating nanoparticles with specific molecules they can be made to function in new and exciting ways, for instance in medicine. The distribution of molecules across the surface of nanoparticles has long been thought to be evenly spread. Now research by chemists and mathematicians at the University of Technology in Eindhoven led by professor Remco van der Hofstad shows that these molecules typically group together in clusters. This surprising insight, which is published in the leading journal Nature Communications, can open the way to a more efficient production of these complex nanomaterials, and at the same time improve their effectiveness.

Establishing the randomness of molecule distribution on nanoparticles is far from a trivial matter. Even with the best microscopes it is hard to pinpoint the exact position of the molecules on a sphere. And then, what exactly is randomness? Take for instance the three images below. Which do you think represents a completely random distribution?

Random and non-random distribution

Many people assume that only the third image reveals a fully random spread of points, while the patterns in the other images are more or less clustered. Think again! In reality, the first two pictures are generated in a fully arbitrary way, whereas the points in the right image show an order that is way too neat and regular to be fully random.


So to find out what was really going on, the researchers not only needed knowledge of the chemistry and a high-quality microscope (in this case the high-resolution dSTORM-microscope), but also the expertise of mathematicians. They helped to interpret the experimental measurements of molecules and to compare them to statistical models.  

The experiment started by labeling the molecules with luminescent dyes. These make the molecules show up as patterns of light under the microscope. Because of disturbances in the light, these patterns are quite variable. Using state-of-the-art statistical techniques these disturbances were filtered out, enabling the researchers to establish the actual distribution on the molecules.


They then compared this distribution with a fully random pattern from a computer simulation. The comparison clearly showed up a more clustered pattern of molecules on the nanoparticles than was to be expected had the distribution been fully random. The common assumption that the binding of molecules to nano particles is fully arbitrary, is therefore not correct.

Why exactly the molecules cluster, is still not known. For this, the researchers say, more research is needed. Still they are quite happy with the results. “The better we understand the behaviour of molecules on nanoparticles, the more efficient the production of these kinds of complex nanomaterials, and the better we will be able to develop new practical applications of nano technology”, says Remco van der Hofstad, professor in statistics at the department of Mathematics and Computer Science at the TU/e.

He is also very pleased with the unique collaboration between chemists and mathematicians in this research. “This certainly is a good basis for new work in the future”.