Issue Date: December 13, 2004
SINGLE-MOLECULE MAGNETS EVOLVE
One of the most pressing technology needs today is to find more efficient ways to store and process digital information. There are a couple of possibilities: One is to squeeze more data onto storage devices by making currently used magnetic nanoparticles even smaller. Another is to develop fundamentally different ways to process information, such as quantum-based computing. For any of these potential applications, the use of chemistry to develop new materials will be critical. That's where single-molecule magnets (SMMs) may fit in.
Conventional magnets rely on the collective behavior of the unpaired electron spins of hundreds of thousands or millions of individual metal centers in a particle or bulk material. SMMs, on the other hand, are transition-metal clusters that individually exhibit the classical properties of a magnet below a critical temperature called the blocking temperature, which is currrently limited to about 4 K. Because of their small size, SMMs also have been shown to exhibit quantum tunneling of magnetization and quantum phase interference--key properties needed for materials to function as quantum bits (qubits).
The past couple of years have seen a burst of activity in SMM research, as fine details of the properties of the tiny magnets have become better understood and the synthesis of more diverse compounds has evolved.
"We have many types of magnets today made from iron, cobalt, and nickel, and metal alloys and metal oxides," notes University of Florida chemistry professor George Christou, one of the pioneers of the SMM field. "Magnetic materials make up a multi-billion-dollar-per-year industry. So we're not trying to replace the magnets we already have--they are already fantastic. But single molecules are much smaller than even the smallest magnetic particles, and if a molecule can function as a magnet, then you can use a molecule to store information--one bit of information per molecule."
The smallest metal nanoparticles used in research are about 3 nm in diameter, which is in the neighborhood of 1,000 atoms, Christou says. "The most information that can be stored on hard drives and other devices currently is 3 billion bits, or 3 gigabits, in 1 cm2 area of a cobalt-based magnetic material," he notes. "The much smaller size of our molecules means we could get 30 trillion of them into 1 cm2, and thus a storage density of 30 trillion bits, or 30 terabits, is feasible. This is 10,000 times greater than the current best by computer manufacturers. One of the research challenges now is to find better SMMs that function at higher temperatures."
The most studied SMMs so far have been clusters with manganese oxide cores surrounded by a sheath of organic carboxylate ligands. The oxygen atoms bridge the manganese atoms to allow interactions between the large number of unpaired electrons, leading to the magnetic properties, Christou explains.
BESIDES THEIR size, SMMs have a number of other advantages over traditional magnets, Christou says. They are soluble in organic solvents because of the peripheral ligands, while magnetic particles in general are insoluble. The alterable shell of organic ligands that helps solubilize SMMs also makes them highly crystalline and prevents the magnetic cores of the molecules from making contact with neighboring molecules. Unlike magnetic particles made by breaking down larger bulk materials, SMMs have a uniform size. They also are air-stable for long-term storage. "There is a lot of operator control, or synthetic manipulation, we can do with these molecules that is not really possible with traditional magnetic particles," he adds.
Christou and his group have prepared about three-fourths of the SMMs reported thus far. One of his primary collaborators in studying SMM properties during the past 15 years has been University of California, San Diego, chemistry professor David N. Hendrickson. Christou and Hendrickson have been working since 2001 with physicists Stephen O. Hill of the University of Florida and Andrew D. Kent of New York University, as well as physical chemist Naresh S. Dalal of Florida State University, under a National Science Foundation Nanoscale Interdisciplinary Research Team grant.
The remainder of the synthesis and investigation of the properties of SMMs has been carried out by several groups around the world, and at one time or another nearly all the primary researchers in the field have worked together at some level. "The SMM field is truly an interdisciplinary area," Christou says.
THE EARLY TYPES of molecule-based magnets, first developed in the late 1960s, consist of three-dimensional arrays of inorganic or organic molecules, particularly organometallic complexes. In these compounds, some of which are magnets at room temperature, the magnetism arises from the collective unpaired electrons of the bulk material, similar to iron or cobalt metal particles. A lot of progress has been made on developing these 3-D molecule-based magnets since the late 1980s.
The first identified single-molecule magnet was a manganese oxide cluster with acetate ligands, Mn12O12(O2CCH3)16(H2O)4. It has eight Mn(III) and four Mn(IV) metal centers in a Mn12O12 core that is surrounded by 16 acetate groups. The molecule, first reported in a paper in 1980, has 20 unpaired electrons.
Not much happened with this compound for a decade, Christou recalls. His group, which had initially been interested in studying manganese-based biomolecules, ended up synthesizing the Mn12 benzoate derivative in the late 1980s. Christou sent some of the benzoate to Dante Gatteschi of the Laboratory for Molecular Magnetism at the University of Florence, in Italy. Gatteschi's group was already working in the area of 3-D molecule-based magnets.
Studies of the magnetic properties of the Mn12 compounds led to a paper by Roberta Sessoli, then a graduate student in Gatteschi's group, along with Gatteschi, Christou, Hendrickson, and others. The paper reported the low-temperature magnetic susceptibility of the clusters, establishing them as the first known SMMs [J. Am. Chem. Soc., 115, 1804 (1993)]. Sessoli, Gatteschi, and coworkers also published a paper describing the magnetic hysteresis of the compounds [Nature, 365, 141 (1993)]. Hysteresis is the ability of a material to remain magnetized after an applied magnetic field is removed. It's required to keep the spin magnetic moment from randomly flipping and losing stored information--that is, the ones and zeros of computer binary code.
Those two papers set off a gold rush for physicists to study the magnetic properties of the Mn12 compounds, Christou says. "It became a physics playground. These compounds were crystalline, soluble, and monodisperse. And because they are so small, they show a lot of the quantum properties that theoretical physicists predicted should be exhibited by small magnetic particles but that no one had seen before.
"The physicists wanted to understand quantum tunneling, the factors that control it, and the quantum mechanical mathematical expressions to describe the effects," Christou continues. "There are hundreds of papers on Mn12 acetate now, and the physicists probably know more about it than the chemists."
A KEY THEORIST working on SMMs has been Philip Stamp, director of the Pacific Institute of Theoretical Physics at the University of British Columbia, Vancouver. His work was among the first to focus on magnetic and quantum properties of nanoparticles. Another key figure--and one many SMM researchers have worked with--is Wolfgang Wernsdorfer of the National Center for Scientific Research's Louis Neél Laboratory in Grenoble, France. He is an expert on the low-temperature study of magnetic properties of SMMs.
SMMs that contain metals other than manganese, including iron, vanadium, cobalt, and nickel, have been reported. Christou and his collaborators lately have been exploring SMMs that have various numbers of manganese atoms. They now have synthesized compounds ranging from Mn4 to Mn84, including an Mn25 cluster with a record 51 unpaired electrons. They also are looking for new methods to synthesize SMMs, including soluble Mn(III) starting materials and combining preformed manganese clusters of different sizes to make clusters not available from the usual manganese starting materials.
While many of the structures of SMMs are impressive to view, perhaps the most striking is an Mn84 cluster shaped like a holiday wreath that was reported earlier this year by Christou, Wernsdorfer, postdoc Anastasios J. Tasiopoulos, and coworkers [Angew. Chem. Int. Ed., 43, 2117 (2004)]. This cluster, the largest SMM made so far, has a Mn84O72 core that was built up by the reaction of Mn12 acetate with a permanganate salt.
How does Mn84's size compare to other nanomagnets? The giant ring of 1,032 atoms has an outside diameter of 4.2 nm and is 1.2 nm thick, making it a little larger than the smallest known cobalt magnetic particles, Christou says. The Mn84 cluster crystallizes in a hexagonal-close-packed arrangement resembling graphite sheets, with multiple layers aligning to form nanotubes. This architecture offers a variety of possibilities for new materials, such as inserting guest molecules into the tubes. "There is something gorgeous about a large molecular cluster like Mn84, the way it forms these beautiful nanotubular stacks," Christou observes. "I think the structural aesthetics of SMMs are not to be underestimated."
Getting to a point where the electronic properties of clusters can be fine-tuned to make SMMs with desired properties is a direction Christou and others would like to move toward. One system developed by Christou's group is a manganese SMM dimer prepared from Mn4O3Cl4(O2CCH2CH3)3(pyridine)3. The dimer is held together by six hydrogen bonds between chlorine and hydrogen atoms [Nature, 416, 406 (2002)].
Single-crystal studies carried out by Wernsdorfer using superconducting quantum interference devices reveal that the hydrogen bonding leads to coupling of the magnetic moments of the two Mn4 units and different quantum behavior than for the individual SMMs. The data show that quantum tunneling still occurs in the usual steplike manner, but it's the collective behavior of the dimer, rather than a single molecule. The results show that even weak electronic exchange interactions can have a large influence on the quantum properties of SMMs, Christou says, and suggest that this approach could be a chemical way to tune the quantum tunneling.
SEVERAL RESEARCH groups are starting to branch out in other directions to create different families of SMMs and study their properties. For example, groups led by chemistry professors Kim R. Dunbar at Texas A&M University and Jeffrey R. Long at the University of California, Berkeley, have been independently synthesizing SMMs with cyanide bridging groups, rather than the standard oxide bridging groups.
"Christou and Hendrickson's research in the area of manganese carboxylate SMMs, as well as that of other groups such as Gatteschi and Sessoli, has had a tremendous impact on the field of molecular magnetism," Dunbar says. "Our work on cyanide-based SMMs complements the research on metal oxide-based analogs and demonstrates that the SMM phenomenon is more general." Cyanide chemistry offers the advantages of being easier to control than cluster self-assembly reactions based on oxides, and it's more predictable in terms of the nature of the spin coupling between metal centers, she adds.
One of Dunbar's compounds, recently reported by graduate student Eric J. Schelter and postdoc Andrey V. Prosvirin, is a Mn4Re4 cyanide cluster with chloride ligands on the Mn(II) ions and tripodal phosphorus ligands on the Re(II) ions [J. Am. Chem. Soc., 126, 15004 (2004)]. The complex is important, as it is the first example of a cube-shaped SMM cluster and the first SMM to contain a 5d element, Dunbar notes. The effect of the magnetically anisotropic low-spin Re(II) ions on the high-symmetry environment induces an unexpected type of magnetic behavior, she says. Magnetic anisotropy is the nonuniform distribution of magnetic properties in a molecule or particle.
The findings on Mn4Re4, and related clusters made by graduate student Curtis P. Berlinguette, "are important to the field, as they lend insight into how first-order single-ion anisotropy of a metal ion can lead to SMM behavior of a different origin than that observed for Mn12 acetate," Dunbar says. These single-ion effects were first observed for SMMs last year by Naoto Ishikawa and coworkers of Tokyo Institute of Technology, in Japan, for a set of lanthanide phthalocyanine double-decker complexes [J. Am. Chem. Soc., 125, 8694 (2003)].
Dunbar's group is planning to prepare and study a series of molecular cubes, squares, and trigonal bipyramids, and use them as building blocks for larger assemblies. The chloride ligands on the Mn4Re4 complex, for example, provide ideal sites for substitution, Dunbar says.
Continuing the exploration of the lower part of the periodic table, a few researchers are beginning to make mixed transition-metal and lanthanide SMMs. The first reported examples are Cu2Tb2 and Cu2Dy2 clusters synthesized by chemistry professor Naohide Matsumoto of Kumamoto University in Kumamoto, Japan, and coworkers [J. Am. Chem. Soc., 126, 420 (2004)]. The complexes were shown to have SMM behavior, although magnetic hysteresis was not observed.
Graduate student Abhudaya Mishra in Christou's group more recently synthesized an SMM with a Mn11Ln4O8 core, where Ln = Nd, Gd, Dy, Ho, or Eu [J. Am. Chem. Soc., 126, 15648 (2004)]. The dysprosium cluster, described in detail in the paper, is the first mixed 3d-4f SMM to exhibit hysteresis and quantum tunneling, Christou notes.
A DIFFERENT approach to making lanthanide SMMs that uses macrocyclic chemistry is being pursued by graduate student Curtis M. Zaleski and chemistry professors Vincent L. Pecoraro of the University of Michigan and Martin L. Kirk of the University of New Mexico and their coworkers. The team has focused on developing SMMs that, like the 3d-5d and other 3d-4f complexes, provide more anisotropy.
The researchers have concentrated on making metallacrowns, the inorganic analogs of crown ethers that "are particularly attractive for SMMs because they allow for a high density of metal ions," Zaleski says. Their collaborator, Dimitris P. Kessissoglou and his group at Aristotle University of Thessaloniki, in Greece, had earlier made the first metallacrown-based SMM, a Mn26 cluster that has a Mn16O12 core surrounded by strands of manganese atoms and ligands to give a metallacryptate structure--a 3-D version of a metallacrown [Angew. Chem. Int. Ed., 42, 3763 (2003)].
In a different approach, Zaleski, Pecoraro, and their coworkers turned to combinations of lanthanide ions with transition metals to make a class of Ln6Mn6 clusters [Angew. Chem. Int. Ed., 43, 3912 (2004)]. The complexes have a hexagonal Ln6 core with two Mn3 "wings." The dysprosium analog shows SMM behavior, but gadolinium and terbium analogs don't, they note. The team is continuing to synthesize mixed-metal clusters with metallacrown networks and has yet to complete full characterization of the magnetic properties.
Another new type of molecular magnet includes metal complexes that form extended linear structures, like a chain of beads. The magnetic properties of these single-chain magnets (SCMs), as they are called, come from coupling interactions between one link in the chain and the next. The one-dimensional chains are a fundamentally different type of magnet because they aren't discrete molecules like SMMs and they don't have a 3-D network structure like traditional iron or cobalt metal magnets or organometallic-based molecular magnets. There are now a few examples of SCMs, the first one being a cobalt-based molecule synthesized in 2001 by Andrea Caneschi, Sessoli, and Gatteschi at the University of Florence and their coworkers.
One type of SCM is a family of manganese(III)-nickel(II) complexes synthesized by chemists Hitoshi Miyasaka of Tokyo Metropolitan University and Masahiro Yamashita of Tohoku University, Sendai, both in Japan, and Rodolphe Clérac of Paul Pascal Research Center, Pessac, France. The chains are made up of a Mn2 complex with salen-type bidentate ligands linked to a nickel complex containing pyridine ligands.
The repeat unit of the chain is Mn–ON–Ni–NO–Mn–O2, and the chain lengths are approximately 110 repeat units long, or about 140 nm, Miyasaka notes. The chains are magnetically isolated from each other in the crystal structure by void spaces occupied by inorganic counterions. The team recently reported a Ni–Mn2–Ni tetramer version [Inorg. Chem., 43, 5486 (2004)].
Clérac and Miyasaka, in collaboration with Wernsdorfer and others, studied the magnetic properties of their original manganese-nickel SCM, observing that the manganese-nickel repeat unit displays the properties of an SMM [Phys. Rev. B., 69, 132408 (2004)]. "Our point in this work is not only to reveal the correlation between SMM and SCM systems, but also to search for the variation in quantum effects from the atomic quantum regime to the classical bulk regime," Miyasaka says. The type of interaction between SMM units shown to affect quantum tunneling in Christou's Mn4 dimer might also be a key interaction affecting the quantum behavior of SCMs, he adds.
A few companies have been interested in developing SMMs into commercial applications, Christou says, but no one in industry has jumped in yet because the operating temperature is still too low. In fact, he and his collaborators have yet to patent any of their work.
"It's still kind of early," Christou observes. The first thing that has to be done is to get blocking temperatures above 5 K, he says. But the temperature shouldn't be an economic barrier to their application, as the use of superconducting magnet coils for nuclear magnetic resonance applications has shown, Christou points out.
As far as applications go, some academics are working to deposit Mn12 clusters on surfaces, but that too is not very advanced, Christou says. "We have been avoiding putting Mn12 on surfaces in our lab because two dimensions might not be the future of information storage," he notes. "A lot of us believe the future of SMMs and information storage is going to be three-dimensional. And Mn12 is probably not going to be the future of SMMs either. It's the best at the moment, but we need better compounds."
One of the things Christou likes to point out as giving the SMM field high potential is that it's an area of materials research where scientists have all the advantages of molecular solution chemistry--solubility, room-temperature synthesis, and easy control of molecular structure. "The great thing about SMMs is that the molecule is not the precursor to the material, the molecule is the interesting material," he concludes.
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society