Scientists have characterized nanotubes as the Superman of the molecular world. Composite materials made with nanotubes can be 100 times stronger than steel at one-sixth the weight. They conduct heat as well as diamond and electricity as well as copper. They might even solve the world's energy problems through hydrogen storage. Now, researchers have triggered IR fluorescence from carbon nanotubes.
Nobel laureate and fullerene pioneer Richard Smalley of Rice University (Houston, TX) says he was surprised to detect fluorescence in semiconductor, single-walled nanotubes. "I really didn't think it was likely to see fluorescence in the tubes because we knew the bandgaps, and you typically find molecular mechanisms that quench the fluorescence before you get it out," Smalley says.
His initial concerns were keeping nanotubes from aggregating into ropes or bundles and separating the various kinds of nanotubes by size and type. "Our motive was entirely to develop the science and technology necessary to manipulate individual nanotubes to get metals sorted from semiconductors, and semiconductors by bandgap," Smalley says.
Typically formed by a single graphite sheet seamlessly wrapped into a tube, nanotubes can be classed as single-walled or multiwalled and metallic or semiconductor. Multiwalled nanotubes are concentric circles of nanotubes. The orientation of the graphite sheet used to make the nanotube determines whether a tube is semiconductor or metallic; if semiconductor, the diameter of the tube is the major factor in determining the tube's bandgap.
Smalley's group first pelted a suspension of single-walled nanotubes with high-frequency sound waves to separate the ropes into single tubes. After centrifugation to isolate the single tubes from the remaining bundles of tubes, they obtained samples measuring an average of 0.7-nm diameter and 130-nm long (the tubes typically start longer, but the group believes that sonication cuts the average tube length to around 130 nm). Sodium dodecylsulfate (SDS) molecules encased the tubes to keep them from regrouping and centrifugation removed the remaining bundles from the sample.
Analysis of the absorption and emission spectra as it relates to the pH of the sample showed that as the pH decreased to below three, fluorescence diminished. "We believe that the mechanism is a hydronium ion that migrates through the SDS monolayer coating and gets to the side of the tube. If it's acidic enough, [the ion] will compete with the side of the tube for a proton and literally covalently bond to the side of the tube," Smalley says.
This bonding results in tying up one of the electrons in the nanotube that was otherwise part of the highest occupied valance band of the semiconductor. If enough ions bond to the side of the tube, it reduces the number of excitons created when the tube is peppered with 8 ns pulses from a 532-nm laser. When the pH is above seven, the excitons are free to move along the long nanotube and emit photons in several distinct peaks between 800 nm and 1600 nm, with each peak approximately 200 to 300 cm-1. As the group intensified optical stimulation during the experiments, though, the number of excitons increased and the electron/electron-hole pairs cancelled each other. Smalley says that optimal excitation levels give quantum yields of 10-3. In other words, for every 1000 excitons, the yield is about three photons.
Although Smalley doesn't think visible excitation is possible with the nanotubes at any levels worth pursuing, the sharp emission peaks in the near-IR (NIR) spectrum where tissue is most transmissive bode well for using nanotubes as in-vivo fluorescent dyes or tags. "NIR is a region that you can detect things quite well with a standard CCD," he says. "This is very practical for medical diagnosis and treatment. The fluorescence wouldn't be as practical if it were at 2 µm or more because of water absorption."