Researchers Explore Quantum Dot Based NVRAM

I Don't Believe in Imaginary Property brings us an article describing the possibility of a new type of non-volatile storage based on quantum dot technology. So far, researchers in Germany have achieved 10ns access times and 0.7Hz refresh rates. Their calculations predict that the access time could be maintained for up to a million years. We have discussed other technologies based on quantum dots, such as solar panels and information teleportation. From the Ars Technica article: "Quantum dots can do this because there is more design freedom in setting them up. Normal flash memory relies on the huge potential barrier created by a silicon oxide layer. However, to get electrons across that barrier when writing data to a flash cell requires a lot of energy, energy that destroys the silicon oxide layer. Quantum dots, in contrast, have tunable properties, so the barrier can be kept low."

Comments

Strange suimmary

[niceone]

10ns access times and 0.7Hz refresh rates. Their calculations predict that the access time could be maintained for up to a million years.

What TFA says is that at the moment the memory needs refreshing at 0.7Hz, but they calculate that eventually they'll be able to make one that only needs refreshing once every million years.

have they solved QD blinking yet?

[davidknippers]

Given that quantum dot 'blinking' seems to be stochastic, I don't see how quantum dots could be utilized for reliable RAM.

Re:have they solved QD blinking yet?

[davidknippers]

I actually can't figure out where all the hype is coming from with quantum dots. They have interesting biomedical applications as fluorophores in cellular imaging and molecular detection. They have advantages to organic dyes traditionally used in imaging applications, as they're tunable, have highly specific emission ranges, very high quantum yields and are resistant to photobleaching. Precise measurements of changes in biomolecules can be measured with quantum dots, such as detection of fluorescence intensity changes of Förster resonance energy transfer processes. However, the use of QDs isn't revolutionary, and they're not exactly cheap to make (a quick trip to the Invitrogen website, searching for their 'QDot' line of products illustrates this). The technical difficulties that still have to be ironed out with quantum dots, such as the 'blinking' problem (likely due to some sort of twist on Auger photoionization) makes them useless in single-molecule excitation situations (which I'm sure would be necessary to control in any sort of storage environment). At this point, numerous problems stand in our way of making them the wonderful solution described in these articles. Their toxicity, steps required to ensure proper surface passivation, limited solubility in aqueous medium without extensive modification to their surface, etc...