Scotch and water: an imperfect blend.

Our knowledge about the geometry and electronic structure of molecules in the liquid phase is very limited. Not only do the molecular arrangements change rapidly—on the scale of picoseconds to femtoseconds—but the properties of the individual molecules are also constantly changing, and interactions between the molecules cannot be neglected. Thus it is not surprising that there is still much to learn about common and simple liquids.

The Perfect Balance

Imbibers have long been combining alcohol and water in myriad beverage concoctions. For example, Homer in his account of the Trojan War reported that the Greeks mixed their wine with water to avoid insanity. Scientists, for entirely different reasons, have spent the past four decades studying how alcohol and water mix at the molecular level. Alcohol in water is one of the fundamental liquid-liquid solutions endemic to countless chemical and biological processes. Until now, however, experimental data on the molecular properties that emerge from such solutions came primarily from neutron diffraction analysis and sent mixed messages. The study by Guo et al. of methanol, a nondrinkable form of alcohol that can provide insights into the behavior of other alcohols, sheds light on a 40-year controversy over what kinds of molecular structures are formed in pure liquid methanol. It also suggests an explanation for the well-known puzzle of why alcohol and water do not mix completely: the system must balance nature's tendency toward greater disorder (entropy) with the molecules' tendency to form hydrogen bonds.

One big mystery has involved the fact that, when alcohol and water mix, the disorder or entropy in the resulting system does not increase as expected for ideal solutions. This anomaly has traditionally been explained in terms of hydrophobic interactions involving alcohol molecules that induce a static, ice-like structure in the surrounding water. However, despite a great deal of effort spanning four decades, a convincing description of the details of the incomplete mixing is lacking, and no consensus on the correct explanation has been reached.

To shed light on this puzzle, researchers turned to ALS Beamline 7.0.1, where they studied the absorption and emission of x-rays by liquid methanol in and out of solution with water. The spectra obtained reflect the local electronic structure; in particular, the oxygen line shape is sensitive to the hydrogen bonding configurations. Information about the molecular arrangements can therefore be obtained by comparison to theoretical predictions.

Experimental oxygen emission spectra recorded at three excitation energies (A, B, and C) compared with theoretical emission spectra of chain and ring structures in methanol.

The results show that the structure of liquid methanol at room temperature is a combination of rings and chains, each made up of either 6 or 8 methanol molecules. When water is added, the methanol chains interact with varying numbers of water molecules. These "bridging" water molecules bend the chains into open-ring structures that are stable because their glue-like hydrogen bonds are saturated. This means that the mixing of alcohol and water on the microscopic level is incomplete no matter how long you wait.

At the molecular level, very little mixing of alcohol and water occurs in solution. Instead, chains of methanol molecules react with clusters of water molecules to form stable open-ring structures, which lower the solution's entropy.

The high degree of order in these clusters reduces the overall entropy of the liquid. Yet entropy must either stay the same or increase in the liquid. So to preserve the second law of thermodynamics, nature discourages the formation of too many such clusters in the liquid. Indeed, the measurements indicate that only a portion of the chains are being bridged. While the formation of clusters prevents full mixing, the second law of thermodynamics limits the degree of order in the system, suggesting a competition between increasing entropy and hydrogen bonding of clusters.

This study establishes a valuable tool for probing the molecular properties of liquids and solutions, something that until now has been difficult to do. The results have substantially refined both our knowledge of structure and order in methanol and methanol-water solutions and our understanding of the unusual thermodynamic properties of this common liquid mixture.

Research conducted by J.-H. Guo and D.K. Shuh (Berkeley Lab); Y. Luo, S. Kashtanov, and H. Agren (Royal Institute of Technology, Sweden); A. Augustsson, J.-E. Rubensson, and J. Nordgren (Uppsala University).

Research funding: U.S. Department of Energy, Office of Basic Energy Sciences (BES); the Swedish National Science Foundation; the Swedish Natural Science Research Council; and the National Supercomputer Center of Sweden. Operation of the ALS is supported by BES.

Publication about this research: J.-H. Guo, Y. Luo, A. Augustsson, S. Kashtanov, J.-E. Rubensson, D.K. Shuh, H. Agren, and J. Nordgren, "The Molecular Structure of Alcohol-Water Mixtures," Phys. Rev. Lett. 91, 157401 (2003).