Slashdot Mirror
New Polymer Ideal For Secure Data Storage
aphexbrett writes "Clever geometry is the basis of a new material that is said to be ideal for secure data encryption and dense optical information storage. The material consists of a lattice of onionlike spheres in which the particle core and its layers each contain a different dye. The material can hold four or more pieces of information in one spot--not just two as in binary optical data storage. And it opens a door to high-density three-dimensional optical data storage. Read a summary of the research over at C&EN News."
A binary bit holds one piece of information, it has two states but is still only one bit (piece) of information.
This process is not very useful for the proprosed applications of data storage. The main hurdle in that case is dynamic, accurate access to setting flags one way or another and then subsequently reading them. This is nothing more than a way to trap molecules in concentric shells of layers of polymer, a far cry from high performance data storage. Don't hold your breath yet.
That's weird, because DVD's can be dual layered, in other words, more than one piece of data in one "spot" (2-dimensional spot that is). And so is the upcoming bluray discs. Of course, when DVDs were first developed, 10-layers was "planned". And there's FMD (prototypes only) that uses multiple layers but of fluorence not reflective optics. And there's the holographic storage technologies (which is truly 3D, unlike layering).
But what would happen if you mix this multi-dye technique with the existing layering technique....and blue lasers....man....just think of the p0rn possibilities! Each dye can store a different angle (or an "alternate ending").
the article doesn't seem to mention how robust such a material would be - will the dyes last for a long period of time, and if not, will some dyes fade before others?
Given the correct photo-stabilisers, the dye layers could be made to last for "extended" periods of time. Maybe up to years? The problem lies with the light-fastness of dyes; when a dye molecule undergoes the electronic interaction with light that produces colour there is a % chance that the molecule will be damaged by that change. The higher the energy of the electronic interaction, the higher the % chance of damage.
Blue (visible) dyes are generally amongst the most intrinsically stable as their interactions are with the red (low energy) portion of the visible spectrum. On the other hand, UV reactive dyes (such as Optical Brighteners/Flourescent Whiteners) are degraded very quickly by their high energy interactions. Put a sheet of copier paper out in the sun for a couple of days, and then hold it next to a new sheet - you'll see how quickly the OBA's have been destroyed!
Now, photo-stabilisers can be added to the dye mix to counteract these degredation processes but in a system where you are wanting several dyes to be active at differing wavelengths it will be difficult in the extreme to arrange the system so that one of these "onion layers" doesn't absorb the wavelength required by another layer!
Finally, making a reader for the material is one thing
And what a thing it would be! The nice thing about silicon chips is that the access time is constant (IIRC each bit is activated in parallel?) across the storage unit. You can read bits 1, 2 and then 3 with the same latency as bits 1, 1583945856 and then 393758273589235892253. With a "three dimensional matrix" of discrete units, you first have to find your bit before it can be read! Imagine with current mass-use technology... a read head housing 4 lasers (as in the 4 dye example in the article) trying to access a bit at the "start" of the data, then one that's physically 1.5cm away, and then again, and again, and again.... the latency would be huge!! Maybe I don't know enough about
So yes, security tagging would be OK - relying on the macro-structure of the matrix under different lighting and the good old Human Eyeball Mk1 - but data storage? I think it'll be a while before this gets used...
That's right. For example, CD was invented in 1979 (CMIIW). Started to be introduced in 1983. Beginning of adoption is around early nineties. So, it takes 10-15 years before it's truly popular.
But sometimes, inventions wither before they see the daylight or poorly marketed. I just hope that it will soon hits the market with the right price.
--
Error 500: Internal sig error
One possibility (which I encountered in another context) is that you create a piece of this material, then record its "signature" under various wavelengths shining from different directions. Then you hand out the material as the ID card and publish the signatures.
Creating a duplicate is infeasible because you'd have to more-or-less exactly duplicate the position of #bignumber of nano-scale particles inside the containing matrix...
The SCO lawsuit makes me wish my company were in Utah. We need a new building.
... with a few more details: Nanoparticle dyes boost storage
Eric Smalley
If they can arrange for 8 different wavelengths, each sphere becomes a byte. Then, if each sphere is 1000nm in diameter (which is a pretty large item in relation to the rest of nanotech, and therefore seems reasonable to manufacture), they could fit 100K X 100K = 10GB into a wafer 10cm by 10cm by 1um in size. Or, if you add another 10cm to the height you get a 1 petabyte cube that you can hold in your hand, like the one Arthur C. Clarke mentions in 3001. Even a little memory stick type of thing 3cm by 1cm by 1mm in size would hold 300GB, while something the size of a current HDD would be around 100TB. Not too shabby.
And what a thing it would be! The nice thing about silicon chips is that the access time is constant (IIRC each bit is activated in parallel?) across the storage unit. You can read bits 1, 2 and then 3 with the same latency as bits 1, 1583945856 and then 393758273589235892253. With a "three dimensional matrix" of discrete units, you first have to find your bit before it can be read! Imagine with current mass-use technology... a read head housing 4 lasers (as in the 4 dye example in the article) trying to access a bit at the "start" of the data, then one that's physically 1.5cm away, and then again, and again, and again.... the latency would be huge!! Maybe I don't know enough about...
:-(
You probably don't. Silicon chips aren't truly random access anymore, at least for large RAMs. , They do have a seek time, due to the northbridge and to the way that DRAM works in general. These days it's on the order of 50-100 nanoseconds (for a CPU, that's 200 clock cyles). While the bandwidth is impressive (gigs per second), that latency to RAM from the CPU is large, and it's what hyperthreading is all about: when you miss cache and have to go to RAM, let the other thread run.
The same is true for hard drives, but on a larger scale: seeks take miliseconds, but bandwidth is large. So while this new polymer won't enable solid-state storage, it might be able to compete with hard drive eventually.
That said, I think it's vaporware too
I hereby place the above post in the public domain.