Birth of a crystal nucleus

Laura Filion is one of the three 2019 Vidi grantees at the UU Faculty of Science. It is her wish to tackle an old problem in materials science. “When a fluid freezes, how does it choose what crystal structure to form?” Using computer simulations of colloidal systems, she hopes to provide a clear answer to this question.

The main topic of Laura Filion’s research is the self-assembly of colloids. “Colloids are microscopic particles that float around in fluids and gases. They exist in various shapes and are all over the place”, says Filion, computational statistical physicist at the Debye Institute for Nanomaterials Science. “Think of fat drops in milk, oil drops in latex paint, or even the ink in e-readers. In the lab, colloids can be made from a variety of materials.” Although colloids are invisible to the human eye, they are very large in relation to atoms. The largest atom is smaller than 0.3 nanometer, while colloids have sizes between 1 and a few thousand nanometers.

This size difference has a strong effect on materials built up out of colloids. “Materials made out of colloids have an internal structure of the same size as the particles. This influences the properties of the material, such as the softness, and the way it interacts with light. Specific arrangements of colloids can be used to reflect visible light in different ways. This would be very useful in, for example, paint.”

Random motion leads to ordered structures

Filion is receiving a Vidi grant of 800.000 euros from the Dutch Research Council NWO, to study the way colloidal particles arrange themselves into ordered structures. “On a microscopic scale, colloids are constantly bouncing through the suspension, as they are being hit by the atoms in the surrounding gas or fluid. However, when you put enough of them into a small volume, this random motion can actually lead to beautifully ordered structures, where the particles organize themselves into a crystal. We call this phenomenon ‘crystal nucleation’, as it typically starts with a tiny crystal nucleus, which then grows out to a larger and larger crystal.”

Surprisingly, nucleation can sometimes lead to unpredictable results. Filion: “In systems where multiple crystal structures can appear, we do not have a good theory for predicting what crystal will emerge from the fluid. So when a fluid freezes, how does it choose what crystal structure to form? Using computer simulations, I want to find out how this is selected, and how we can control this process.”

On a microscopic scale, colloids are constantly bouncing through the suspension.

Structured liquids

Until recently, it was unclear where to look for the relevant factors that drive the selection of crystal structures, but the last few years, physicists are asking new questions, says Filion. “It is now thought that not just the bulk characteristics of the crystal and liquid phases matter, but also the structure of the liquid is a key element to crystallization. Liquids may seem unstructured, but they actually contain a large number of complex local structures.

I suspect that some of those characteristics are good for the nucleation of a given crystal, while others are not. So with my Vidi, I want to identify and control the features that enhance or inhibit nucleation into different crystal structures.” During the next five years, Filion will work on this important question with two PhD candidates.

Lab mice of atom research

Her research on colloid nucleation is relevant not only to colloidal materials. “Atoms and colloids often have the same phases: they can be liquids, gases, or crystals. So colloids are small enough to share their phase behaviour with atoms, but large enough to be visible under the microscope. Consequently, colloidal particles are great for studying processes that also apply to atoms.”

So colloids can be considered the lab mice of atom research? “They are certainly a good model system, where simulations and theories can be directly tested in the lab. As a fundamental physicist, I don’t have a specific application in mind. But for example in pharmaceutical sciences, it is essential that medicine always crystallizes into the same structure, with the same effects on the body. A better understanding of how we can predict and control crystal nucleation could be extremely helpful in ensuring this.”