Surface solution for writing nanodata

Materials World magazine
1 Mar 2009
Carbon nanotube attached to the tip of an atomic force microscope. Image courtesy of Purdue University, USA

A carbon nanotube (CNT) probe tip, with a silicon oxide outer layer for wear resistance, could enable ultrahigh density data storage.

The ‘nanopencil’ can write bit sizes as small as 6.8nm onto ferroelectric films. Silicon oxide makes it less prone to bending or buckling.

This improves on previous research, which found that CNTs and other nanowires used for data writing are prone to wear, explains Professor Konstantinos Giapis of the California Institute
of Technology (Caltech) in Pasadena, USA. ‘When they are made longer to increase their useful life, they become easier to bend and are less precise,’ he adds.

The silicon oxide solution was born when international semiconductor firm Intel approached scientists at Caltech for help.

Dr Yuegang Zhang of the Lawrence Berkeley National Laboratory, USA, who previously worked for Intel on this project, explains, ‘The buckling force of a naked nanotube probe is of a few nano-Newtons, which is comparable with the force needed for the device to operate. The silicon oxide significantly increases the mechanical stiffness of the probe, [creating a] buckling force that is calculated as high as 14.5 micro-Newtons, which is more than three orders of magnitude higher than the operation force’.

The trick is to maintain a conformal layer of silicon oxide around the CNT using plasma coating techniques, as well as sanding it down precisely enough to expose the nanotube tip. A diamond
surface was used for the latter, which proved to be time-consuming. ‘It is similar to sanding a macroscopic surface using Crocus paper,’ says Giapis. ‘It removes extremely little material at each pass. However, with better understanding of the process, the time required can be reduced

Another difficulty in up-scaling the nanoprobe’s production will be improving how the nanotube is attached to the scanning probe microscope, says Zhang. Growing the CNTs on the probe is not always reliable. However, if this can be solved, ‘probe-based data storage could one day replace all hard disks and some flash memory storage devices,’ he insists.

While the original Intel-Caltech project has ended, the group at Caltech is continuing to develop the technology. Researchers are creating CNTs filled with liquid metals to improve conductivity. They are also trying to ‘functionalise’ the exposed CNT tips with organic and biological molecules for biosensors.