Features:

BUILDING NANOCRYSTALS
By Katherine A. Caponi, NCSA Science Writer

We normally think of building something as putting parts together to make a whole. Think of a carpenter building a piece of furniture, like a chair. She gathers her plans, tools, wood and other hardware. Legs, seat, nails and all, she uses those parts to make a whole, useable chair that she can then sit on. That's even how Webster's defines building--to make by putting together materials, parts, etc. By contrast, we typically think of taking something apart as demolition, destroying the purpose of the original object. But what if what we could build something in such a way that we actually had to start with the "whole" and take parts away to get a useful object? Think of it like whittling or carving a delicate ice sculpture. That's what some scientists are doing with nanomaterials.

In a lab at the University of Illinois at Urbana-Champaign (UIUC), for example, scientists are building nanomaterials for future lasers, which will be one billionth of a meter long! To do this, the scientists first put a silicon wafer through a process called electrolysis reproduction. They gradually immerse a silicon wafer through an electrolytic process. They gradually immerse the wafer into an etching bath of hydrofluoric acid and hydrogen peroxide. At the same time, the scientists apply an electrical current to the bath. Munir Nayfeh, the lead researcher and member of the Illinois experimental group, explains, "This process erodes the surface layer of the material, leaving behind a delicate network of weakly interconnected nanostructures." They then remove the wafer from the etching bath, immerse it in a liquid solvent bath, and subject it to ultrasound. The ultrasound bath causes the delicate nanostructures to crumble into particles. The particles, silicon nanocrystals, are easily sorted into different sized groups that will emit corresponding colors of light when hit with infrared or ultraviolet light.

Because the nanocrystals emit visible light and are made out of biologically benign silicon, they are useful for a variety of purposes such as sensors, semiconductor lasers, single electron transistors, and even for tagging cancer cells for study. In particular, they could provide a viable alternative to dye markers like the barium drinks used in upper gastrointestinal X-rays and some MRIs that have been used for decades. The nanpocrystals are small enough to fit through the pores in a cell, and according to Nayfeh, they are also photostable--a quality that is quite valuable when researchers need to take repeated measurements under intense radiation.

But how do scientists find out which kinds of molecular structures have such special properties and how do they determine what uses can be made of them? The answer is in the electronic structure of molecules. If scientists can perform accurate calculations detailing molecules' electronic structures, they can predict how the molecules will behave under certain conditions. Lubos Mitas, assistant professor of physics at North Carolina State University and former NCSA research scientist; researchers from UIUC's physics department; and teams of scientists from other institutions have been working together to use supercomputers to calculate the molecular electronic structure of silicon nanocrystals and several other materials.

Exciting discoveries in nanocrystals In the example of the tiny lasers, Nayfeh and his team are investigating the special properties of silicon nanocrystals. They emit light in the visible range from blue to red, they appear to be quite stable, and the electrolysis reproduction technology for homogenous crystals is inexpensive. "Moreover," Mitas states, "the observed nanoparticles tend to appear in 'magic' sizes, such as 'blue', 'yellow', 'orange', etc. This clearly points to the existence of "sweet spots" in the cluster sizes that have enhanced structural and chemical stability.

Mitas and his team have joined Nayfeh to carry out an extensive study on the silicon nanocrystals, charged with the task of finding out what causes the enhanced chemical, structural, and optical properties. Because the traditional theories used to predict a material's properties based on electronic structure (which requires solving highly complicated differential equations in many dimensions) tend to be less reliable when applied to new materials with unexpected physical and chemical properties, there is a clear opportunity for new theoretical alternatives.

One of the traditional methods, density functional theory (DFT), describes an interacting system of subatomic particles in terms of its density instead of its wave function and energy levels. The DFT method works well for determining geometries of molecules, but the Quantum Monte Carlo (QMC) method provides a significantly more accurate approach of calculating the energy differences such as the optically active excitations.

Using QMC on NCSA supercomputers, Mitas and his team of researchers came to several interesting conclusions about the silicon nanocrystals. They identified the structures of those "magic" cluster sizes with strong photoluminescence in the blue, yellow, and orange ranges of the visible spectrum that were repeatedly obtained in experiments.

They also carried out calculations of the Stokes shift, or the difference between the energy absorbed by the molecule and the energy that the molecule emits afterward, for the excitation at the absorption edge of the molecule. Some of these calculations have been done in collaboration with a research group and supercomputers at Lawrence Livermore National Laboratory (LLNL) led by Giulia Galli. One of the most interesting questions was whether the excited state was localized within the molecule and how much the excitation could affect the atomic geometries. Because QMC was capable of describing the excitonic effects very well, the team found that the exited electrons were delocalized across the whole structure. They also found that the excitation causes very small changes to the structure of the molecule, and the Stokes shift calculation was in excellent agreement with the experimental data showing the actual difference in frequencies between absorbed and emitted energy.

Answering these two basic questions means that researchers now have a way to predict the color of the emitted light from silicon nanocrystals of varying sizes. It also tells them how any change of the electronic structure of the molecule would manifest itself in change of the optical properties. In particular, they were able to predict changes in the color of the emitted light upon attaching additional molecules such as methyl and ammonia groups to nanocrystals. The results of their QMC simulations of chemical doping, which have been addressed in articles in the Physical Review Letters (http://prl.aps.org/) of the American Physical Society, were in excellent agreement with physical experiments.

One more fact that the researchers revealed is that hydrogen peroxide is absolutely crucial to the production of silicon nanocrystals in order for them to be stable. If hydrogen peroxide is not present in the etching bath during electrolysis, the molecules will not appear in the specific structural configurations necessary to make them a good alternative to the currently used dye markers.

The method to their magnets The extensive calculations on silicon nanocrystals are just an example of the useful applications of calculating the electronic structures of molecules. Mitas and his team have logged close to 500,000 hours on the NCSA SGI Origin2000 and the NCSA Titan Linux cluster to help other researchers in a variety of projects.

For example, they applied QMC method to the properties of molecular magnets. Molecular magnets are of great importance to the microelectronic industry, opening a window of opportunity for prototyping increasingly smaller devices. Eventually the study of electronic structures of molecular magnets will provide the potential to tremendously increase the storage capacity to just a few electrons/spins per bit of information. Additional projects involve applying QMC to ferroelectric and magnetic solid materials used in making transducers, capacitors and sensors as well as biomolecules, which form complex systems that carry out very vital tasks in our bodies.

These projects are very diverse in nature yet they have something in common- the physics at work is incredibly complex. However, the development of new methods enables the scientists to reveal the electronic structure of materials on the quantum level and opens breathtaking opportunities rapidly advance medical, microelectronic, and optical solutions in the era of supercomputing.

Funding This research is supported by the Office of Naval Research, University of Illinois at Urbana-Champaign, Defense Advanced Research Projects Agency, the National Science Foundation, and NCSA.

Access Online URL: http://access.ncsa.uiuc.edu/CoverStories/electronicstructures/.

For further information: http://altair.physics.ncsu.edu/.