Login   |    ITaP Home > ITaP Newsroom

 
This article has been moved to the new ITaP Newsroom website. It will remain available here until Dec. 31, 2013. To find it in its new location, visit itap.purdue.edu/newsroom/. Send questions about this change to itap-comm@purdue.edu.

Purdue researchers propose method to image electron wavefunctions

Researchers at Purdue University and the University of Melbourne, Australia, are proposing a novel way to accomplish something normally impossible to do — 3-D mapping of electron wavefunction, which is the probability that an electron will be in a given position around an atom in a solid.

The technique, which the researchers used ITaP high-performance computing systems to develop, could add to understanding of how things work at the atomic level — a field called quantum mechanics. A map of electron wavefunctions and their response to an electric field should aid in designing, engineering and manufacturing nanoscale devices a few atoms in size that act quantum-mechanically, as well as next-generation microchips and other electronics with nanoscale features.

It also could be useful in advancing quantum computing. Computers taking advantage of quantum properties may have potential for exponential leaps in speed and processing power, for example, by being able theoretically to solve all of the permutations of a problem at once.

Moreover, the technique may have wide applicability. In principle, it could be used to map wavefunctions in single-electron silicon quantum dots, quantum wells and other nanostructures of interest from basic-science, nanoscale-device and quantum-computing perspectives.

The idea was conceived by University of Melbourne researcher Professor Lloyd Hollenberg together with Purdue Electrical and Computer Engineering Professor Gerhard Klimeck, former Klimeck student Rajib Rahman, now at Sandia National Laboratories, and Seung Park, a student in Klimeck’s group. The method employs silicon isotopes in a silicon crystal and electric and magnetic field control to establish a statistical map or fingerprint of an electron wavefunction, a key to understanding electron behavior, which is vital for enabling nanoscale engineering and quantum computing.

The researchers worked for two years developing the technique, currently detailed online in the journal Physical Review Letters, where they are proposing it for testing by experimentalists.

Park already has tested its feasibility in computer models involving more than a million atoms, said Klimeck, director of the Network for Computational Nanotechnology (NCN) based at Purdue. The modeling was done with NEMO 3-D — software developed by the Klimeck research group to realistically simulate structures one atom at a time — on ITaP’s Steele supercomputing cluster. Steele is operated by the Rosen Center for Advanced Computing. The Rosen Center is the research-computing arm of ITaP, which is Purdue's central information technology organization.

Additionally, in experiments in the 1970s, Purdue researchers Edward Hale and Robert Mieher similarly probed bulk, rather than atomic-scale, silicon samples.

Klimeck and Rahman said the new technique employs silicon with impurities that make it easier to bind additional electrons to silicon atoms and create, in effect, hydrogen atoms in crystal complete with hydrogen-like wavefunctions. Such a system also forms the building block of a popular quantum-computer proposal.

Silicon atoms around a wavefunction are then used to probe it in different locations over different devices with different frequencies of electromagnetic fields. In particular, the idea is to monitor electromagnetic excitation at readily discernable Silicon 29 isotope sites, which in turn reveals properties of the electron wavefunction at that location. Repeating enough of those probing actions yields the data for a statistical map of the wavefunction.

As described by Hollenberg, electrons normally can only be thought of as existing in a kind of probability cloud around an atom where they might be found if they could actually be observed. The method the researchers are proposing essentially pins the atom in a crystal lattice, magnifies it and allows an observer to “peer inside the cloud” by capturing signals from the nuclear spin of the Silicon 29 isotopes as the wavefunction is deformed by an electromagnetic field.

The work was supported by the Australian Research Council and the U.S. National Security Agency and Army Research Office. In modeling the technique, Park also used NCN and nanoHUB resources built on the HUBzero technology developed at ITaP.

Image caption: This visualization from computer simulations shows an electron wavefunction — a key to understanding electron behavior, which is vital to nanotechnology and the idea of super-powerful quantum computing. Purdue and Australian researchers are proposing a 3-D method for mapping wavefunctions, which normally cannot be observed. In the image, an electron wavefunction (purple) is attracted by an impurity potential and an electrical field against an interface. The confinement potential is shown in shaded gold. The wavefunction is hybridized between the impurity and interface confinement. Credits: Gerhard Klimeck, Rajib Rahman, Insoo Woo, David Ebert, Purdue University

Writer: Greg Kline, science and technology writer, Information Technology at Purdue (ITaP), (765) 494-8167, gkline@purdue.edu

Sources: Gerhard Klimeck, (765) 494-9212, gekco@purdue.edu

                Rajib Rahman, (505) 844-0151, rrahman@sandia.gov

                Lloyd Hollenberg, +61 3 8344 4210, lloydch@unimelb.edu.au

Last updated: Sept. 25, 2009