Duke Chemists Describe Progress At Making "Buckytubes" Suitable For Nanoelectronic Devices

A team led by Duke assistant professor of chemistry Jie Liu is producing nanotubes in larger numbers by altering their recipes for making the molecules. They also are growing the molecules on silicon surfaces to guarantee their purity and favorable electronic properties.

The scientists have further adapted the nanotubes for use as ultra tiny electrical sensors by dressing them in "skirts" and attaching polymers that conduct electricity. The researchers fabricate the polymers in a billionths-of-a-meter scale "chemical reactor" at the tip of an atomic force microscope (AFM), using sensitive electronic guidance to guide the microscope's ultrasharp tip. The group then deposits materials with nanoscale precision ("nano" meaning "billionths) through well controlled chemical reactions.

In one of four presentations at the American Chemical Society's Aug. 18 22 national meeting in Boston, Liu and some of his graduate students described a new and improved alternative method for growing nanotubes directly on a surface of semiconducting silicon dioxide -- the workhorse material of microelectronics.

Nanotubes were discovered in the early 1990s. One of the leading groups in the field is at the laboratory of Nobel laureate Richard Smalley at Rice University, where Liu did postdoctoral research before coming to Duke. Such nanotubes have a range of favorable physical and other properties for engineering the diminutive structures of "nanoscience" a current national focus of basic research.

The closed cylinders of carbon atoms, measuring just billionths of a meter, are exceptionally strong and lightweight. Their electronic properties can vary with the particular arrangement of their carbon atoms, making them act like metals, semiconductors or substances that are almost insulators, Liu said.

Liu and his colleagues seek to produce quantities of nanotubes for use in devices such as molecular sized sensors that can work with individual cells to detect biological molecules. Nanotubes can now be made relatively easily in "powder" form by using an electric arc, a laser, or vapor synthesis. Modifying that powder for nanoelectronic use, however, requires purification with heat and acid, a technique that Liu said is "known to create defects."

In the experiments reported at the American Chemical Society meeting, Liu described how his research team has been able to build up nanotube structures by sprouting them from particulate seeds of iron and molybdenum catalysts. The team carried out the process within a chemical vapor deposition chamber, where heat and the catalyst decompose hydrocarbons. Grown by this method, the nanotubes are electronically correct for sensor use, Liu said.

"Previously people grew these nanotubes on a surface using methane as a feeding gas, but less than 5 percent of the catalyst particles sprout nanotubes," he said. "By switching the feeding gas from methane to a carbon monoxide and hydrogen mixture we can significantly improve the amount of nanotubes we can grow on a surface."

The researchers' new cooking method currently allows 15 to 20 percent of the seeding catalyst nanoparticles to sprout nanotubes. That's "still not satisfactory," Liu acknowledged. "We want each nanoparticle to sprout a nanotube."

Growing nanotubes on silicon isn't enough to make a nanotube sensor, Liu cautioned. To attach the necessary additional molecules without destroying the protodevice's electronic properties requires avoiding the "covalent" bonds bonds with equally shared electrons that atoms like carbon, oxygen and hydrogen naturally form.

"If you covalently link molecules to a nanotube, it becomes insulator like," he said. "You destroy the path for the electron to go through the nanotube." A separate challenge is the inherent chemical stability of carbon nanotubes, which makes bonding with other molecules difficult.

The solution for avoiding covalence and ensuring reactivity is to "put clothes on the nanotubes," Liu said. To do that, his group has learned how to add a "primer" layer of 3 aminopropyltriethanoxysilane (APTES) to a nanotube's surface. Over the APTES they then grow a coating of silicon dioxide, a non covalent semiconducting material that can form chemical bonds with conductors.

They need to use APTES because that promoter serves to confine the silicon dioxide coating to the nanotube's surface, Liu explained. "Without this, the silicon dioxide may grow everywhere."

Another direction for Liu's group is to fabricate nanostructures that incorporate chain like molecules called polymers. The polymers can serve as "molecular wires" for sensors applications, according to Liu.

To attach the polymers, the scientists use a device normally employed to map out nanoscale features on surfaces, the AFM. In the late 1990s Chad Mirkin at Northwestern University invented a method of using an AFM to pick up "organic" molecules and deliver them to a surface. The organic molecules, all containing carbon, travel suspended inside the tiny "nanomeniscus" of water that condenses on an AFM's tip from the surrounding air.

Liu's team's has further adapted Mirkin's approach by making the nanomeniscus into what he calls "a chemical reactor."

His team starts out by coating the AFM tip with the building blocks for synthesizing conducting organic polymers. After a nanomeniscus forms, its watery environment dissolves the building blocks. When the scientists then bring the nanomeniscus into contact with a surface such as a nanotube, the building blocks are combined electrochemically into complete polymers and deposited there.

To create a nanomeniscus "you want it pretty humid," Liu said. "The humidity we work with is between 35 and 60 percent." At the chemical society's Boston meeting, one of Liu's graduate students is "going to describe ways and conditions to fabricate different polymers on a surface," he added.