In big-data sciences like genomics so much data is produced that researchers are forced to resort to manually shipping hard disks around between centers. Currently, a single desktop computer hard drive can have a maximum storage density of 600 gigabits per square inch, equal to a capacity of about three or more terabytes. The industry goal is to increase this density by more than 15 times, providing the ability to store thousands of high-resolution images and music files on an area no larger than a human fingernail. Now, researchers have found that material roughness plays a key role in denser storage.
Scientists from the Institute of Materials Research and Engineering (IMRE) - part of the Agency for Science, Technology and Research (A*STAR), based in Singapore - and the National University of Singapore published their work in Nature's open-access journal, Scientific Reports.
It explains how making the surface of a substrate material smoother, allows the self-assembly of its nanostructures - polystyrene-polydimethylsiloxane block copolymer - to become more efficient. Self assembly is a cheap, high-volume, high-density information-patterning process at sub-billionths-of-a-meter scales.
The researchers found that a height close to 10 atoms (10 angstroms) breaks the self-assembly technique, because this is the limit of surface roughness which allows successful self assembly of dots - or information - used to eventually make high-density storage.
Last year, IMRE researchers demonstrated a technique of using salt to create super-fine, nano-sized structures, increasing storage density from 0.5 terabits per square inch to 3.3 terabits per square inch. Now, their latest finding brings the goal of 10 terabits per square inch, or more storage capacity on smaller devices, closer. This density goal is even beyond current solid-state memory abilities, due to reach its limit by 2017, according to the researchers who produced the paper.
"If self-assembly is to be employed for bit-patterned media, then the key issue of surface roughness highlighted in the paper needs to be addressed and to be sorted out," says Mohammad Saifullah, one of the research scientists who made the discovery. "As for the discovery being a 'game changer', it is difficult to say at the moment as more research needs to be done to reach the goal of 10 terabits per square inch."
But, Thomas Russell from the department of Polymer Science and Engineering at the University of Massachusetts Amherst, US, who was not involved in the research says: "I really do not comprehend why the authors are seeing what they are seeing. I feel there is another factor entering in that they are not realizing. Angstrom-scale roughness changes should not affect the ordering or orientation of block copolymers where the relevant length scales are much larger."
In response, Saifullah says that the fact remains their observations are experimentally supported after removing all possible factors, so he and his colleagues are confident of the results.
Tighter control
Roughness data was extracted using Nanoscope Analysis software, supplied by Veeco Instruments, which came along with the Digital Instruments Nanoscope IV atomic force microscope used to analyze the surface topography of the substrate.
"If we want large scale, large area nano-patterning using very affordable self-assembly, the surface needs to be extremely smooth so that we can achieve efficient, successful self-assembly and with lower incidences of defects," says Saifullah.
Saifullah and his colleagues conclude in their research paper that to achieve one terabit per square inch and further, bit-sizes on storage material would have to shrink to 12 nano-meters, which may require tighter control over the surface roughness of magnetic media.
