By Saren Johnston
In the furthest stretches of their imaginations, physicists never believed they would have molecular structures in which they could embed small geometrical arrangements of interacting magnetic ions. But now that chemists have become remarkably adept at preparing such structures, physicists have access to the kinds of uncomplicated systems they need for investigating magnetism at the fundamental level.
And, although they may be a long time coming, the potential applications for molecular magnetic clusters, especially in the field of computing technology, are staggering. Each molecule can be viewed, in certain cases, as a microscopic magnet with magnetization that can be oriented up or down, thus carrying a bit of binary information.
Imagine data-storage capacities thousands of times greater than what are available in today's computer systems. Magnetic molecular clusters may one day make it possible to stockpile hundreds of gigabytes of data in an area equivalent to the head of a pin, an astounding feat when you consider that just 1 gigabyte can store enough information to fill 62,500 typed, double-spaced pages.
Beyond the incredible data storage possibilities, there also exists the prospect that these clusters could be manipulated at the atomic level to form logic components in a quantum computer. Quantum computers would operate at speeds hundreds or thousands of times as fast as the supercomputers that amaze us today.
Another fascinating, potential spinoff of the study of magnetic clusters is the connection with magnetic molecules that are present in biological systems and whose functions are still largely unknown. But before we look too longingly at what might eventually develop from the study of molecular magnetic clusters, a great deal of fundamental research must still be done.
In the molecules of interest, there are as few as 12, 10, six or even four atoms grouped together, with each atom having a net magnetic moment. Because these magnetic moments arise from the intrinsic angular momentum, or spin, of the atomic electrons, physicists frequently refer to the magnetic interactions among "spins."
Driven by an enthusiastic curiosity about these systems, Ferdinando Borsa, an Ames Laboratory senior physicist, is carrying out nuclear magnetic resonance (NMR) experiments with certain molecules to better understand their basic magnetic properties. NMR uses the spectroscopic properties of the energy levels of the magnetic and quadrapole moments of the nuclei in strong magnetic fields to obtain microscopic information about the atomic magnetic moments, or spins, in the molecules. This information may ultimately provide scientistsgreater knowledge about the magnetic properties of more complex materials and make it possible to design materials with specific characteristics.
The structures Borsa is investigating are unique because, although the small molecular structures chemists have designed contain many different atoms, the magnetic properties of those molecules are determined by only a handful of interacting magnetic ions arranged in very ordered ways.
"The beautiful thing about these molecules is that for the first time it's possible to control the number of magnetic moments," says Borsa. "You can also control where they are located in the molecule, their distance apart and what kind of interactions take place among these moments. So these very simple systems are ideal model systems for doing theoretical calculations that are almost exact and can be compared directly to the experimental data." Developing analytical theories based on these model systems should allow scientists to accurately predict the magnetic properties of the molecules as a function of parameters such as temperature and applied magnetic field.
Marshall Luban, also an Ames Laboratory senior physicist and a collaborator with Borsa on the research, is involved in the theoretical aspects of small magnetic systems. He joyfully notes that the structures chemists have created in the laboratory are a theorist's dream. "They're a kind of utopia -- simple systems with just a few interacting ions as opposed to a number such as 1023 power, which you might normally worry about," he says. "Those kinds of problems are so difficult that it's a pleasure to be able to work on something far simpler, the other end of the spectrum in range of difficulty. What we're talking about, in a sense, are realizable toy systems that are ideal for theoretical calculations."
Borsa adds, "In more complicated systems, there is a big gap between the theories and the experiments. The experiments allow you to get some information on magnetic properties, but when you want to reproduce that with a theory, you have to use approximate theories or use computer simulations on a small number of particles and extrapolate the large numbers. With these small systems, you don't need to do all that because you only have six, 10, 20 spins -- you can solve the problem exactly."
Zeehoon Jang, an Ames Lab graduate student, is doing experimental work on small molecules with Borsa, but is also intrigued by the theoretical possibilities that are Luban's specialty. He has done many NMR measurements on small molecules and worked on some of the theoretical calculations associated with those molecules. "Zeehoon is our line of connection between the experiments and the theory," says Borsa.
"I think physics is beautiful when it can be made simple," says Borsa. "The logical and mathematical tools are complicated, but you are aiming at simplicity in physics to get to the root of the problem."
The simplicity Borsa requires to get to the source of magnetic molecule puzzles is supplied primarily by Dante Gatteschi, a professor of chemistry at the University of Florence in Italy. Gatteschi and members of his research group, among them the brilliant, young researcher Alessandro Lascialfari, produce and characterize the new materials that provide Borsa the molecular systems he requires for his NMR investigations.
Although simple in terms of the number of interacting magnetic ions, these molecules are at the same time complicated, containing lots of organic materials, such as oxygen, carbon and hydrogen, which shield the magnetic moments from outside interference. "These materials serve as scaffolding," says Luban. "They keep the ions carrying the magnetic moments positioned as they are. If you ripped off this stuff, the ions would just wander away."
Borsa elaborates, comparing a molecule to a big ball. "The magnetic moments are inside the ball; they don't see each other, so the molecules act independently of one another. Although you have a number of these molecules in a crystal, when you do an experiment, you're essentially probing the behavior of one molecule. You have a repetition of all units -- identical -- so you have the number (n) of molecules times the same result that you have for one. But the advantage of having n times is that you have big NMR signals.
"With NMR, you use the nuclei as probes or spies that are sitting in the molecules," Borsa continues. "And what the nuclei see is the fluctuation of the spins. So you are studying the motion of the spins by looking at the effect on the NMR."
Some of the molecules that Borsa, Luban and Jang are investigating include: manganese 12 (Mn12), iron 6 (Fe6), iron 10 (Fe10), copper 8 (Cu8), copper 6 (Cu6) and, the latest addition, chromium 4 (Cr4). That translates to molecules with 12 or fewer spins, and Luban reminds us, "That makes for an enormous simplification in any effort or attempt to describe the magnetic properties of those molecules."
As an example, he invites us to look at a diagram of an Fe10 molecule, sometimes called the "ferric wheel," because ferric means made of, containing or derived from iron; and because the Fe10 structure diagram closely resembles a carnival Ferris wheel.
NMR studies of Fe10 probe the interactions of the 10 iron ions under various conditions to gain more information on the magnetic properties of the molecule. "Suppose one of the iron guys at a given instant of time does something, say it rocks, rotates in space or jumps -- it screams," says Luban, anthropomorphizing the iron ions. "Then we ask the question: 'If this guy screams now, at this place, what impact does it have elsewhere at these other iron locations at a later time?' This is called the equilibrium/time correlation function."
Because of the simplicity of the Fe10 system, there's a good chance that Luban may be able to calculate the equilibrium time correlation function from first principles (i.e., without approximations or modeling). He can then directly compare his predictions about the magnetic properties of Fe10 with Borsa's experimental data.
"Hopefully, they will be in pretty good agreement," Luban says. "Then we will be able to say, 'Yes, we can in fact predict all of the properties of this system.' The applications, I think, will be forthcoming. The more we understand from these simple systems, the more we can predict. Then we can start to get greedy and design systems that will exhibit the properties we would like."
Borsa and Jang hope the NMR measurements they do on these various systems will give them a better idea of how the fundamental physical law for magnetism determines the magnetic behavior of interacting spins. "If we can understand that, we may be able to design a material that will work for better computer magnetic memory," says Jang.
Borsa admits that application is down the road. Right now, he's happy to indulge his curiosity by using these simple systems to observe how particles behave when they are coupled together. "One of the fundamental problems in condensed matter physics is what makes many particles behave in a collective way," says Borsa. "Why would a bunch of particles, when they're together, have some new behavior that is not a direct consequence of the addition of the behavior of each individual particle? The hope is to find some basic law that determines the collective behavior of particles, and then you can use it in any kind of field."
For more information:
Ferdinando Borsa, (515) 294-9901
borsa@ameslab.gov
Current research funded by:
DOE Basic Energy Sciences Office
Last revision: 12/17/99 sd
Home | Comments
| Search | Disclaimer