A New State Of Matter Turns A Solid World Into A Melting One

An experiment described in last week's Science (Sept. 12) paints an exotic portrait of certain substances seemingly confounding nature by existing as a liquid, instead of a solid, at room temperature.

George Bertsch, a theoretical physicist at the University of Washington, describes how the experiment with clusters of sodium atoms found that the atoms did not follow sodium's normal pattern, melting at 97.8 degrees Centigrade (208 degrees Fahrenheit). Instead, the small clusters of atoms melted at minus 6 degrees Centigrade (21 degrees Fahrenheit), well below room temperature.

The discovery was the work of Hellmut Haberland at the University of Freiburg in Germany. Bertsch, who has been following the field of cluster research for the past decade, writes that as a result of the experiment, scientists are now challenged "to understand what happens to the liquid and solid phases in small particles."

European researchers are working on a number of practical applications for the cluster phenomenon. Attempts are being made to produce thin films of silicon clusters that would process signals carried by light. Others are researching the use of clusters to improve the magnetic recording of data. And Haberland has been reported to have produced clusters of the element molybdenum that will even stick to Teflon.

Clusters have been called a new type of matter, says Bertsch, because they appear to be a bridge between atoms and the world of normal size, and have strange magnetic, electrical and optical properties. What is particularly curious, he says, is that the properties of the clusters depend on the number of atoms they contain.

Bertsch notes that most clusters are very unstable collections -- "they touch a wall and they are gone." But a decade ago it was discovered that certain clusters contain "magic numbers" of atoms that make them particularly stable. These numbers begin with just two atoms, and continue through eight, 20 and 40 and into the hundreds of atoms.

The German researcher used magic-number clusters of 139 sodium atoms. The melting point was observed by forming condensation 'droplets', rather like hot steam hitting a cold window, and passing the condensate through a mass spectrometer and finally an electric field.

Bertsch concedes that the research is controversial, and there are physicists who insist there can only be one melting temperature for each of the 92 natural elements. But, he says, "as scientists we have to look at the evidence." What's more, he believes there is evidence that the same phenomenon that the German researcher demonstrated with sodium, also exists with atomic clusters of both tin and lead.

The UW scientist is hesitant to attempt an explanation of what is causing the lower melting point of these elements. However, he notes, it has been suggested that there is a relation to a theory known as surface melting: when a substance reaches melting temperature, only a small surface layer melts immediately. "As solid sodium reaches melting temperature, a small layer of liquid might form on top of the solid," he says. "In a cluster, all you would have is this outer, liquid layer."

It could be said, says Bertsch, that this new type of matter is "practically all surface." If there is already something strange happening at the surface of certain elements, "then you accentuate that behavior when you create clusters."