X-ray study of the liquid potassium surface:structure and capillary wave excitations

From Soft-Matter
Revision as of 07:13, 25 September 2009 by Yang (Talk | contribs) (Added content)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Soft Matter Keywords

Capillary Wave, Liquid Metal, Liquid-vapor interfaces, Surface tensions, Surface-Induced Layering


Illustration of layering at the surface of a liquid metal. Figure taken from Prof. O.G. Shpyrko's website http://liquids.seas.harvard.edu/oleg/Research_Oleg.htm
Experimental setup. Figure taken from this paper.

The surface of a liquid is usually disordered, but theories and computational models have long predicted different behavior for liquid metals. Recently, scientists have been able to prove experimentally that the surfaces of certain liquid metals (including gallium, indium, and mercury) are indeed ordered. These liquids have layering of molecules close to the surface, which eventually transitions into the randomly packed molecules in the bulk of the liquid. Even after this discovery, it was unclear whether the surface layering was due to intrinsic properties of liquid metals or due to high surface tension, because all the liquid metals tested had high surface tension. This paper presents an experiment on liquid potassium, a liquid metal with a comparably low surface tension (5 to 7 times less than that of the other liquid metals mentioned). They find that liquid potassium also has a layered surface, meaning that the root cause of this phenomenon lies in the metallic properties of the liquid, not in its high surface tension.

Why couldn't the layering effect of liquid metals be observed earlier on? Firstly, alkali metals have both high evaporation rates and high reactivity to oxygen and water, so people thought it would be impossible to maintain a pure surface on the liquid metal. This was proven to be a false notion, probably because the high solubility of the alkali oxides in the bulk of the liquid kept the surface unadulterated. Secondly, low surface tension in the liquid allows capillary wave excitations that perturb the surface enough to make it difficult to obtain precise x-ray diffraction readings to measure the extent of layering (in a way similar to Bragg diffraction). The authors were able to construct a capillary wave model and recover the x-ray diffraction data and in turn the layering structure on the liquid surface.

Relevance to Soft Matter

Liquid-vapor interfaces are very important in soft matter, and the dominating forces at an interface are different than those in a bulk liquid. Here we see how differently the atoms in the liquid metal behave when they encounter a vapor fluid instead of being surrounded by other liquid state atoms. Moreover, the effect penetrates further than just the first layer of atoms, meaning that there is some kind of characteristic length scale at work. Interestingly, the surface of the liquid is ordered and crystal-like, while the bulk of the liquid is disordered and liquid-like. Why is it that the atoms closest to transitioning to the vapor phase are the least disordered? Furthermore, why does this phenomenon only occur in liquid metals and not other liquids? Surface tension ends up not being the culprit in this case. It is important in soft matter physics to be able to identify the dominant forces in a situation. Here, we have interesting metallic properties leading to a perhaps counterintuitive phenomenon.

Further Developments

The paper goes on to propose an explanation for why layering only occurs in liquid metals. The authors view the electrons in the liquid as a Fermi sea and the ions as a classical gas. The electric interactions between the two somehow suppress surface fluctuations, which normally overwhelm the layering effect.

More recently, in "Atomic-Scale Surface Demixing in a Eutectic Liquid BiSn Alloy" (see references), it was shown that when there is a mixture of two liquid metals, they will actually form alternating single-layers of atoms of each element. This is very intriguing!

Illustration of alternating bismuth and tin atom monolayers in a mixture of those two liquid metals. Taken from Argonne Natl Labs, http://www.anl.gov/Media_Center/News/2005/CNM051125.html


O.G. Shpyrko, P. Huber, P.S. Pershan, B.M. Ocko, H. Tostmann, A. Grigoriev, and M. Deutsch, Phys. Rev. B 67, 115405 (2003).

Prof. Oleg Shpyrko's website: <http://liquids.seas.harvard.edu/oleg/Research_Oleg.htm>

O.G. Shpyrko, A.Y. Grigoriev, R. Streitel, D. Pontoni, P.S. Pershan, M. Deutsch, B. Ocko, M. Meron, and B. Lin. "Atomic-Scale Surface Demixing in a Eutectic Liquid BiSn Alloy". Phys. Rev. Lett. 95, 106103 (2005).

Argonne National Laboratory. "Mixed metals not so mixed up at the nano-level". 25 November 2005. <http://www.anl.gov/Media_Center/News/2005/CNM051125.html>