Behind the buzz and beyond the hype:
Our Nanowerk-exclusive feature articles
Posted: Oct 23, 2008
True colors - a novel way of information storage using color lithography
(Nanowerk Spotlight) The method that has been traditionally used in binary information storage is by making a distinction between storage (designated as 1) and non-storage (designated as 0). In reality, each imprint (or non-imprint) can store either 1 or 0. Thus the sequence and the numbers of 1 and 0 define everything with respect to the amount of information that can be stored and retrieved at the hardware level, no matter how sophisticated the overlaying software routines are. Ever since computers were developed, information storage has adhered to the eight-bit system.
No matter how sophisticated information storage technologies have become – exploiting magnetoresistance, developing optical storage media such as CDs, DVDs and blue-ray discs, or the development of holographic storage media – a bit is always represented by manipulating a single feature, i.e., a transition or non-transition.
"This technology is fundamentally limited by the fact that a transition cannot represent two bits or more than that" Dr. Arun Chattopadhyay explains to Nanowerk. "Now, in contrast, consider the following: There are four colors, each of which could at least represent two or more bits; whereas in conventional methods only a single bit is available. In terms of color, this is somewhat similar to a black and white system that can support at most two kinds of transitions – 0 to 1 and 1 to 0. On the other hand, in four-color coded systems there can be 16 such unique transitions."
The primary aim in this earlier work was to demonstrate that lithography in color could be an alternative to high-resolution black and white storage. In a new paper just published in Microsystem Technologies, Chattopadhyay and his team report a new scheme of information storage in color and demonstrate its superiority over conventional binary schemes ("A new way of information storage using red, green, blue and black color imprints").
"There has been no known example of presentation of a scheme of information storage using the color coding similar to ours or anywhere near that" says Chattopadhyay. "On the other hand, there are several methods available for printing different colors next to each other at reasonably high resolution.
Thus there are two challenges: The imprinting of different (but chosen) colors at chosen spots in an organized manner with the resolution being the highest is the first of the problems. The second being an information technology way of using those imprinted colors for high density information storage with appropriate development of schemes."
The recent work of the IIT scientists addresses both points. Interestingly enough, although this method of information storage may appear to be an obvious choice, no one has actually done it yet. The practical challenge lies in developing methods of storage of information in multiple colors and their retrieval.
According to Chattopadhyay, conventional photolithographic principles could be utilized in storing information. "On the other hand, retrieval of the stored information is equally significant as capability of that would also determine the effective and useful information that could be stored and read" he says. "An innovative and efficient way to read a color pattern would be using a CCD camera, considering the rapid progress being made in the technology related to
this. A set of lenses with appropriate illumination should, in principle, be able to capture a spot of length 400–700 nm. Hence, it is primarily a microscope with a CCD camera fitted on the eye piece that could be used for reading the stored data."
Given this set-up, the appropriate choice of colors would be the primary colors red, green, and blue, and black. One important issue here is the fact that color spot would have different sizes, depending on the color chosen, and this could potentially limit the efficiency of data storage. For instance, if one assigns 11 = red (wavelength 650 nm) the requirement of storage space would be much more in comparison to the scheme where 11 = blue (wavelength 470 nm).
"Since the minimum spot size is related to the color, the best way could be to assign the most frequently occurring bit pair the color with the shortest spot size, i.e., blue" says Chattopadhyay. "However since black is the background color (the length of which could be kept shorter than that of blue), the most frequently occurring bit pair could be assigned black color."
Following this argument, an example of who the colors are allotted to the bit pairs in an example are shown below. Since there are four bit pairs and four different colors, there can be 24 different mappings.
Schematic representation of the assignment of color to bit pairs based on their frequency of occurrence. (Reprinted with permission from Springer)
The researchers point out that another important improvement over conventional modes of storage is that in addition to the colors representing bits, a transition from one color to another can also represent bits. For example, if 01 is blue and 10 is green then blue to green transition can also represents bits.
Taking into account the higher density of storage using the color coding but could also taking additional advantage of the 16 different color transition – in contrast to only two transitions (pit to land and vice versa) available in conventional CD's and DVD's – the researchers calculated an up to 75% higher data storage capacity for their system.
Chattopadhyay mentions that, since every single transition will not be meaningful, one has to use a method to differentiate between the meaningful and non meaningful transitions. "This can easily be done by extending the previous color pattern by 100 nm compared to its minimum length. Hence whenever reading data, a spot which is 100 nm more than its assigned size represents a meaningful transition and using the data stored in the lead in area one can then interpret that transition."
The IIT team actually demonstrated the feasibility of their color lithographic scheme by capturing part of a video on photographic film and then successfully decoded the imprinted bit pattern.
Information storage in red, green, blue and white, a demonstration of the capability of this method. The image represents part of a video file stored in color imprints. Each color block is approximately 45 micron x 45 micron in dimensions. (Image: Dr. Chattopadhyay/IIT Guwahati)
"Although, the dimensions of the spots mentioned here are large, in practice they can be significantly reduced using modern optical setups especially with the aid of confocal laser scanning microscopy – both to write and read data" Chattopadhyay explains. "This would significantly improve the density of storage as well as speed of storage and retrieval which is vital to commercial success."
In terms of ultimate miniaturization, one could even think of introducing quantum dots as the medium of color information storage. These nanocrystals can be small – less than 10 nm in diameters – and their size, shape and level of doping with other ions determine their fluorescence emission characteristics. These materials could be used in future storage of color information with the highest resolution, accompanied by an appropriate development in optics.
Density of information storage and the speed of writing and reading the stored information limit the usage of any memory device. In that respects there are several points that need to be addressed for realistic applications of any color lithographic storage medium.
Firstly, the materials on which color codes would be imprinted need to be developed further. What the IIT scientists have done is a sort of one-time, permanent write-only process. There are at least two steps (imprinting and then developing the films) which need to be reduced to one and ideally it should be similar to writing on a CD or DVD, but in colors.
Secondly, one has to find a suitable reading/writing device like a CCD camera that needs to be incorporated in the whole construct of the machine. But Chattopadhyay and his team believe that this would not be very difficult.
Finally, software code to be used with this kind of hardware and fully exploits its vastly improved data storage capability needs to be developed.