Sustainable solvents — do they exist?
Text: Santina Russo
Many of the solvents used in industry are toxic and require elaborate disposal procedures. Researchers from Basel, Bern and Zurich are laying the groundwork for replacing these harmful liquids with safe new substances.
Whether it’s for dying clothes, producing plastics or paints, or removing hazardous waste from contaminated soils, many branches of chemistry rely on solvents for their processes. Although some of these solvents are harmless, many are toxic and carcinogenic and must therefore be processed and disposed of using elaborate procedures that are both cost- and energy-intensive. For several years, researchers have therefore been studying a specific class of solvents known as “eutectic liquids,” which could replace harmful substances in the future.
Their complicated name actually has a fairly simple meaning — it refers to a mixture of two substances that has a lower melting point than the individual components. We take advantage of a similar effect when we apply salt to roads in winter: Since salt water has a lower melting point than pure water, it doesn’t turn to ice until it reaches a lower temperature.
Solvents with soft skills
Like conventional solvents, eutectic liquids are capable of dissolving molecules and can therefore, in principle, be used for the same tasks. Yet, when it comes to practical applications, they offer some key advantages. First of all, they are easy to manufacture — they just need to be mixed in the correct ratio. Second, there is a specific group of these liquids that is completely harmless to health and the environment and therefore also significantly easier to dispose of than conventional solvents.
“This would make eutectic solvents more sustainable and more cost-effective at the same time,” says Markus Meuwly, Professor of Physical Chemistry at the University of Basel. As the solvents have only been studied for around 20 years, however, knowledge about them remains limited.
One sticking point in the past was that little was known about the molecular structure of eutectic mixtures, and there were also no methods of obtaining information about these structures. “This meant it was impossible to determine how their functionality came about and how their properties could be influenced with a view to practical applications,” says Meuwly. “One important concept in molecular sciences is that the structure of substances determines their function,” he explains. “Conversely, the functionality can be adapted by changing the structure.
To do that, however, we need a way of determining the arrangement of components in such liquids.” Meuwly’s team has now developed one such method in collaboration with researchers from the Universities of Bern and Zurich. In doing so, the team has laid the groundwork for studying structure–function relationships in eutectic liquids.
Experiments and computational modeling
The researchers developed and validated the new technique using mixtures of potassium thiocyanate and acetamide, a sort of model substance among eutectic liquids. On the one hand, they used specialized methods based on infrared spectroscopy to study specific interactions between the molecules and ions in the liquids. This allowed them to draw conclusions about the distances between the particles and their relative arrangement.
For example, the thiocyanate (SCN-) ions were usually far apart at a low water content and correspondingly high acetamide content, and they adopted wide-ranging orientations relative to one another. With increasing water content, the ions adopted more-rigid orientations and moved closer together — despite the strong repulsive forces between two negatively charged SCN- ions. At the same time, the ions were surrounded by ever larger clusters of water molecules. Thanks to these observations, the research team now has a more molecularly resolved understanding of how properties of eutectic mixtures emerge.
The key finding for the researchers, however, was that the results of their computer simulations fundamentally agreed with those of the spectroscopic measurements. “We’ve therefore validated our computer model,” says Töpfer.
The researcher has since used the results to further improve his model with the help of techniques including machine learning so that it now provides an even better representation of the interactions taking place within the mixtures. “Now, we can go a step further and start to predict the structures and properties of liquids.”