An apt description of crawling home after a long night of bar hopping might be, Wandering Home in Many Dimensions: Making Sense of Chaos. But in this case, it is the title of a lecture about the mysterious journey of sodium chloride particles into precise crystalline formations, bypassing the trillions (literally) of other options these particles have of becoming ordinary, molecularly-messy, glass. It’s as though atoms want to arrive at a state of beauty. But how does a scientist even begin to formulate the right questions, let alone answer the problem of how these atoms overcome entropy to become orderly structures?

Thursday, at 8:00 p.m., at the Elementary School, in a talk open to the public, Dr. R. Stephen Berry, distinguished professor of chemistry at the University of Chicago, a McArthur Fellow, and one of five officers of the National Academy of Sciences, will be illustrating the particular way in which scientists attack a problem.

“People – even my students at the University of Chicago – rarely get a sense of why scientific theories are different from others,” says Berry, who has spent decades exploring the mysteries of atomic and molecular structures and still maintains a profound sense of wonder and amazement that the human mind can invent predictive concepts like energy and entropy. “These abstract concepts put together completely diverse kinds of experiences into these precise, quantifiable concepts with predictive powers. To me this is overwhelmingly amazing. The unique aspect of science is its quantifiable, predictive power.”

But the steps needed to move from a diffuse problem to a predictive concept are far from apparent in the lab. “The key steps, and I think the most exciting steps, come in trying to figure out what are the right, precise, addressable questions,” says Berry. Setting priorities is critically important in science, he explains, otherwise in this computation age the danger is being overwhelmed with data.

Berry faces new scientific problems by asking himself, What knowledge is worth having in this particular context? “The first time I heard of anyone asking that question --What knowledge is worth having? -- was when Wayne Booth in the English department of the University of Chicago wrote a book with that title. He was thinking of it in more philosophical and social terms, but then as my colleagues and I began thinking about complex problems in science and the whole issue of translating a diffuse idea into a precise idea, I realized that Wayne had asked exactly the right question.”

The scientific process is all about asking the right questions. For example, let’s think about trying to solve the crystal formation problem noted above. We might ask, “Why do some things form glass and other things form crystals?” Simple enough. Yet, according to Berry, this is not an addressable scientific problem. It is too diffuse. It is not specific enough. “You need to go conceptually several steps further. You have to figure out what you can do at a much more precise, quantifiable level.”

In describing his research, Berry takes us through his thinking: “When studying atomic clusters of small particles with 10 to 100 atoms – basically the small end of nano scale materials – that were either solid or liquid, we could see that some of them would form glasses while some of them, like clusters of salt molecules, would find their way ‘home’ to ordered, regular crystalline structures. Even though we could show that they had hundreds of billions more ways of becoming amorphous, and glassy, they still found their way to crystalline structures. Our reaction was, ‘Wow! Why? What is it about the forces between them that makes them different from the things that got stuck as glass?”

Atoms and molecules that make up crystals are laid out in a very regular, orderly way. In a crystal, each “lattice” is a discreet unit absolutely identical to the next. Each crystal is made up of repeating lattice units. In glasses, however, atoms and molecules are every which way. They’re “messy.” There aren’t any discreet units.

Berry and his colleagues looked at a salt structure of 32 molecules with 64 ions, half of them sodium, half chlorine. “In its most stable form, it is a little rock salt cube, that is 4 ions by 4 ions by 4 ions The most perfect little cube all arranged orderly. Or they might form little rectangular blocks or strips. There are only a few hundred of these shapes compared with the trillion or 100 trillion amorphous glassy structures they could form.”

Still in the pre-experimental stage, they have found that if they model a liquid structure of these 32 molecules, that it is almost impossible to prevent the cluster from finding its way to a solid, crystalline, salt structure, despite all the other glassy possibilities. “It finds its way to very very low entropy. The driving forces have to be very very strong,” Berry says. “And the question is: What is it that makes those driving forces so strong? Why does it find its way to an ordered structure and not give way to all that high entropy. That’s the question. But how can you make an inference about that when you are faced with an impossibly large amount of information in this many dimensional world? What is it in that big body of information that is worth knowing? What will help you answer that question?” Somehow he felt that the forces between the particles were responsible for the establishment of this order.

So he set out to map the movement of a few random atoms within a whole cluster, and discovered that some follow a saw-tooth pattern, while others a steep gradient that resembles a craggy staircase. It turned out that a cluster with a saw-tooth pattern makes glass called “glass formers.” One with a staircase pattern makes crystals called “structure seekers.” Berry notes that when it is cooling and solidifying, it is the forces between the particles that determine where the sodium and chloride ions will go.

Further, they discovered that if the forces between the particles are short range (imagine the closeness of Velcro strips) then the potential surface is going to be saw-tooth-like and will be a glass former. If the forces are long range, then the motion of the particles are very collective (they all influence each other, even the far away things). “Long range forces: staircase potential, structure seekers. Short ranges forces: saw-tooth potential, glass formers,” Berry summarizes.

On a hunch, Berry collaborated with other scientists to discover that complex protein molecules, which fold in precise patterns, have staircase topographies as well. But given the statistical odds, Berry likes to say, “These little salt clusters are a greater miracle than protein molecules!”

“Structure seekers” defy all odds to reach home, wandering in many dimensions, not unlike some Telluride locals or tourists in the dark after one too many.