CSC Home PageSkip to Main Content
About Us | Services | Client Results | Insights | Contact Us | Careers
Corporate Governance
Investor Relations
Newsroom
Cycling Sponsorship
Locations
Blogs, Podcasts, Videos & RSS
E-mail Story Story Feedback Print Version
Features
Home Page Home Arrow Features 2003

CSC Scientists Discover Ways to Build Chips From the Atom Up

Nanotube Y-Junction
Model of a Carbon Nanotube Switch
Transistor manufacturers are beginning to reach the physical limits of how small silicon and copper chips can become. CSC scientists have helped find a way to create transistors far smaller than current chips — not by reducing the size of existing chips but by building them up from individual molecules.

The research of four CSC scientists working at NASA’s Ames Research Center in Moffet Field, Calif., has helped pave the way to building transistors from carbon nanotubes, tiny structures created from a single layer of carbon atoms whose size is measured in nanometers (a nanometer is one billionth of a meter). Scientists had theorized that transistors could be made from nanotubes. CSC researchers Manjeri Anantram, Cun-Zheng Ning, Deepak Srivastava and Toshishige Yamada, joined by researchers at NASA and academic institutions, took a closer look at how to actually create such transistors. Their discoveries include algorithms to model such applications, new techniques of information transmission, carbon structures that operate like transistor switch terminals and methods for harnessing nanotube chains for electronic systems.

The carbon alternative

One of the best-known principles of information technology is Moore’s Law, created by Intel cofounder Gordon Moore, which states that the number of transistors per square inch on integrated circuits will double every 18 months. While many expect the law to hold for the foreseeable future, it won’t hold forever. Chip fabricators may soon reach the conventional limits of chips.

The current development technique for semiconductors calls for reducing the size of silicon transistors or simply driving existing transmission systems faster. This top-down approach to miniaturizing transistors and diode lasers — the fundamental building blocks for computing and communication systems — is unlikely to be able to meet the ever-increasing demand for higher-speed information processing and transmission. Even if chips could be made small enough, such densely packed circuits eventually would emit too much heat to be cooled effectively.

If chips were to continue to decrease in size, a new way of building them was needed. Since nanotubes were discovered in 1991, they have emerged as a candidate for the next advance in miniaturization. What intrigued physicists about these tubes were their electronic properties: They can function either as metals or semiconductors. As metals, they conduct very high currents without the deterioration and heating that still are problems with copper wires. As semiconductors, they can be used for high-performance nanotransistors.

In exploring alternatives to the traditional top-down method of reducing the size of silicon transistors, CSC’s researchers realized that miniature nanotube-based devices could be built from the bottom up — with atomic precision. The resulting devices, which have since been created by other scientists, are a new breed of transistors. A nanotube-based transistor is as much as 60,000 times smaller than a conventional transistor. "This alters the kind of devices that you can build," says Srivastava, who focused on making nanotubes act like switches. "You can pack more and more transistors in a smaller amount of space."

Normally, Srivistava says, increasing the density of transistors also increases the power density, which would throw off so much heat that the device would burn itself up. But the carbon structure requires less power so the transistor can run with lower heat and energy thresholds. The size of the carbon nanotubes would also make it easier to integrate processing and sensing in the same device, which could give nanotube-based chips the functions of bioelectric or neural systems. "It’s feasible that they would interact with the environment and have learning characteristics," Srivastava says.

CSC’s contributions to nanotube research

The CSC scientists and their collaborators studied possible nanotechnology materials, theoretically or through computer simulation. They discovered advantages and disadvantages of constructing nanoscale switches and transistors using carbon nanotube hetero-junctions, atomic chains made of individual atoms or even DNA molecules. But their primary contribution to nanotechnology research was in their focus on nanodevice construction. They reasoned that if developers are to build nanoscale devices from the bottom up, they will need entirely new approaches to development.

Following are some of the findings of each of the four researchers:

  • One thing developers needed was a way to model nanoscale devices. Classical methods simply cannot describe how electrical current flows though a nanodevice, but fully quantum mechanical theory is too computationally complex for modeling. Anantram’s team created innovative algorithms and approximations that allow developers to include quantum effects as appropriate, while keeping the overall computations manageable for practical simulations. Anantram’s team then used the simulation software they developed to study how the current-carrying capacity or connectivity of a carbon nanotube depends on its structure and mechanical strain.
  • Ning’s research is primarily concerned with information transmission. He discovered that a transmission system can be based on heating electrons in a semiconductor nanowell or nanowire rather than by turning electrical current on and off. The team’s approach can be as much as three orders of magnitude faster than the one currently being used. NASA has filed a patent for the ultrafast switching based on coupled tiny lasers.
  • After systematically studying the formation, stability, structure and electronic response behavior of carbon nanotube hetero-junctions, Srivastava invented a series of entirely carbon-based structures that can perform all the functions of the three-terminal devices necessary for computing circuitry. The key innovation of this work is that a complete functioning device can be built entirely by carbon nanotubes.
  • Yamada has invented a method to make semiconducting atomic chains for electronics applications. This is a bottom-up approach to build electronics with atomic-scale elements. He developed nanodevice models, which are critical in future atomic-scale electronics. Based on his work, NASA has filed a patent for atomic-scale devices and methods of fabrication.

Ongoing benefits to NASA and CSC

The NASA scientists were honored for their nanodevice research with CSC’s Award for Technical Excellence, which is regarded as the most prestigious form of recognition bestowed upon CSC employees. Each year, CSC’s Leading Edge Forum recognizes the company’s leading technologists by presenting them with the award, and in 2003 six teams were selected from more than 70 team nominees.

The team was honored in part because of their discoveries’ potential impact on CSC and NASA. Srivastava says that his group’s research adds to CSC’s already considerable expertise in large-scale computer modeling and shows that CSC has the intellectual depth to handle large, research-driven computing projects. The researchers say that computers, networks, chips and switches based on molecular-scale devices will also be important for other CSC government clients, such as the U.S. Department of Defense and the Department of Energy.

The benefits the advances could bring to NASA are even clearer. Nanoscale computers will make it possible to build dramatically smaller navigation and communications devices for spacecraft, allowing the spacecrafts themselves to become smaller, lighter and less expensive. Carbon nanotubes could benefit NASA at a macro level, too. Because the carbon structures are heat-resistant and 100 times stronger than steel for weight, carbon nanotubes could be used as lightweight building material for the craft themselves.

About the Leading Edge Forum and the Award for Technical Excellence

CSC’s Leading Edge Forum serves as "The Technology Voice of CSC," providing thought leadership and a technology point of view for both the marketplace and CSC. Through its portfolio of programs, the LEF synthesizes its learnings into a holistic view of the technology marketplace: where it is today and where it is headed.

Established in 1989, the Award for Technical Excellence encompasses the entire spectrum of technology services CSC offers to clients. Achievements can range from custom software development to complex integration of off-the-shelf software. More than 200 CSC field technologists representing every CSC organization around the globe review the nominations.

Related Information:

Contact Us and Let Our Experience Help You Produce Results.

Go to the LEF Awards for Technical Excellence page.

 

© Copyright 2008 Computer Sciences Corporation | Privacy Policy | RSS