Talk:Three month report

From Wise Nano

"Squishy" Silicon  by Brett Bellmore 14:38, 13 Nov 2004 (CST)

The bulk modulus (Stiffness) of silicon is somewhere between 200 GPa and 60 GPa, depending on crystal orientation and the degree of deflection. (The surface is evidently "squishier" than the bulk material.)

For diamond these values are in the neighborhood of 440 GPa. Cubic boron nitride comes in at about 400 GPa.

So, yes, silicon is quite a bit "squishier" than diamond. When you factor in the higher atomic mass of silicon, 28 vs 12, it's clear that the resonant frequency of silicon structures would be much lower than comparable carbon structures. You might have trouble with them at room temperature.

Slight mistatement  by Brett Bellmore 23:01, 13 Nov 2004 (CST)

The relevant properties for determining resonant frequencies would be modulus and density, not modulus and atomic weight. Which still leaves comparable structures in silicon with resonant frequencies several times lower than if they were made of diamond.

What's the focus?  by Tom Craver 10:05, 15 Nov 2004 (CST)

These notes seemed a bit all over the place. I guess part of the problem may be that the project is a broad survey of applicability, but needs to defend the case that nanoblocks can function usefully in a space environment.

Some suggestions:

• Minimize attention paid to potential construction of nanoblocks - at some point you've got to move beyond convincing people it can be done and just point them to appropriate reference works defending the feasibility. Just detail the confidence level of various properties - e.g. 99% confident of 1nm precision in shape, 80% confident in some error level at construction. I agree that a focus on error accumulation due to radiation is a key topic to address, so setting the baseline properties and potential for degradation is important.
• Define levels of confidence regarding capabilities - 99% confidence that nanoblocks could be used for "weak to moderate" structural construction, 90% confidence they'll be strong enough for applications requiring high strength, 80% confidence they'll handle applications involving repeated and varying stress. (Just made-up examples, ignore details.) The higher the probability, the more solidly you need to make your case. Your worst critics may downgrade probabilities by a level - making your case for you by setting a lower bound that still shows high capability. Including lower confidence levels allows you to leave capabilities for future evaluation (and papers! :-), without discarding them. Later reports could upgrade confidence - showing progress.
• Radiation being good for a human is not very relevant - early nanoblocks will not have self-repair. Perhaps it's a useful point to bring up, as a contrast - but if so, the point would be that errors will accumulate, and you need to put some bounds on how fast that will occur, and the impact for certain types of functionality and under certain conditions. (You could point to plans for future work to consider repair via test and replace at the nanoblock level - but I think that must be left out of this paper's scope, and probably isn't worth doing anyhow.) Probably all functional failure levels derive from atomic position and bond error levels. For each key functionality you focus on for high confidence, you need to specify a failure metric, and establish (with a decent confidence level) some sort of acceptable error accumulation level, and show the time period to exceed that level under assumptions like "Mars Surface", "Lunar Surface", "Free space", and probably with "Solar storm duration" - ideally allowing combination of the last with the other three, perhaps giving an example based on solar storm frequency and a Mars cycler ship with a hull and sensors made of nanoblocks (close to worse case - long term exposure with little planetary protection).