This article originally appeared in the February 2000 issue of the ORNL review as part of the Brave New Nanoworld feature.

Quantum Computing by Connecting the Dots

The world's first practical computation on a fabricated nanoscale device may be achieved by a team led by Jacob Barhen of ORNL's Center for Engineering Science Advanced Research (CESAR) and the Computer Science and Mathematics Division (CSMD). The team has received LDRD funding to develop a quantum dot array to perform this task. For this project, quantum dots are clusters of atoms, a few nanometers in diameter, surrounded by an insulating layer. The relatively small size of the metallic dots planned for the ORNL array will enable operation at room temperature (because no cooling will be needed).

Using colloid chemistry, Leon Maya fabricated these 2-NM platinum quantum dots, which appear as round yellow spots in this atomic force microscope image.
Leon Maya, a team member from CESAR and CASD, has already fabricated platinum particles about 2 NM in diameter. For this project, gold particles approximately 1.5 NM in diameter will be produced by trapping them in tiny cavities of polymer molecules.

Because of his knowledge of the structure of DNA, which he has studied using an atomic force microscope, Thomas Thundat will use a DNA molecule as an architectural template for positioning quantum dots on a surface between two electrodes to program the device. "By attaching complementary four-base-DNA sequences to the quantum dots," Thundat says, "we should be able to place them at predetermined spatial locations on the DNA backbone with sub-nanometer precision."

Gold electrodes that provide access to the quantum dots from the macroscopic environment will be fabricated. The two electrodes, which will anchor each end of the predesigned DNA sequence, will be laid on a flat substrate and separated by a narrow gap a few nanometers in length. A novel technique will be used to decompose the DNA template without altering the geometrical position of the electrodes and quantum dots.

The practical goal of the ORNL team is to build a device that emulates a neural network. Instead of the neurons and connecting synapses found in the brain, the nanoscale computer will depend on electrically charged quantum dots connected by electrons that tunnel between them at different rates.

(See " Nanosensor Probes Single Living Cells")

Using colloid chemistry, Maya will fabricate and coat the gold quantum dots. Thundat will assemble them using the DNA templates and an atomic force microscope. Jack Wells of CESAR/CSMD and David Dean and Michael Strayer, both of the Physics Division, will develop a first-principles simulation of the device on ORNL's IBM-SP3 parallel supercomputer. "The simulation," Barhen says, "should provide immediate and invaluable feedback to the experimental design and help us accurately specify the parameters needed to use the device as a computer.

Barhen, Yehuda Braiman, Vladimir Protopopescu, and Nageswara Rao, all from CESAR/CSMD, will develop the methodology and algorithms needed to implement neuromorphic computations on the quantum-dot array. This implementation involves innovative techniques that modify the electron tunneling rate between dots. For demonstration purposes, the researchers intend to solve a pattern recognition problem. An example of such a problem would be the analysis of sensor data to identify seismic patterns indicating the presence of a porous sandstone layer that might contain oil.

"This project," Barhen says, "is motivated by the information processing needs of future generations of intelligent systems. On one hand, there is a need to meet the tremendous constraints on power consumption, size, and temperature. On the other hand, novel sensors, such as hyperspectral cameras used for imaging landscapes, require ever more powerful, dedicated processors. Thus, we are targeting applications that can exploit the emergent collective computational properties of an ensemble of nanostructures."

Current PR | PR archive
ornl | csm | research | people | sitemap | search