Sapphire: A Liquid That Won't Get Things Wet 843
eaglebtc writes "Tuesday on Good Morning America, a representative from Tyco Fire & Security demonstrated an amazing new substance called Sapphire: a water-like fluid that does not get things wet. He filled a small fish tank with Sapphire and submerged a book, a laptop, and a flat panel TV. Both electronics were turned on when submerged; all three items came out completely unharmed. Click here for a slideshow of the demonstration. The official name for Sapphire is actually Novec 1230. Read about it here (PDF). Tyco sees practical applications of Sapphire in fire extinguisher systems for museums and libraries. By the same token of practicality, regular readers of Slashdot probably have something else in mind: total-immersion watercooling. Just think of the possibilities!"
Fluorocarbons (Score:5, Interesting)
Of course we had to have an entire floor below us dedicated to refrigeration, but hey. Governments can afford this kind of stuff.
Pricey (Score:5, Interesting)
After all, 3M's not stupid: they price things correctly. These are the guys behind the Post-It Note.
Re:Uhm (Score:5, Interesting)
Yep, you are no expert. In many fires by far the most damage is not caused by the fire itself, but by the massive amounts of water used to put it out.
Re:"Water"-cooling (Score:1, Interesting)
Potentially very useful depending on the properties of that gas.
If it doesn't get things wet... (Score:5, Interesting)
-j
Re:"Water"-cooling (Score:5, Interesting)
The actual usefulness of the fluid in any state depends on the specific heat capacity, which I can't see 'coz the site is /.'d ...
Since the phase change itself is be a major energy-absorber, that could be very helpful indeed as long as fresh condensed fluid comes in after a radiator of some sort.
Re:Distilled Water? (Score:2, Interesting)
Re:Distilled Water? (Score:3, Interesting)
Re:"Water"-cooling (Score:3, Interesting)
I'm sure that's the boiling point at standard atmospheric pressure at sea level, although the page is slashdotted, so I can't verify that.
It's not difficult to raise a liquid's boiling point by pressurizing it. Cooks do it all the time: it's called a pressure cooker.
But really, there's no reason to bother with that. Cooling a processor isn't about dunking a computer in a liquid and letting the heat evenly distribute. You're gonna want to chill it, no? It's probably more relevant to talk about the liquid's freezing point than its boiling point.
Re:"Water"-cooling (Score:5, Interesting)
If you put it deep enough in the fluid, the bubbles will condense on the way up. If the heat of vaporization is anything reasonable, this should work quite well to remove heat from your chip--the fluid changing to a gas absorbs a bunch of heat, and then swims away with it, while more cool fluid rushes in to take its place. Probably look cool, too.
Fluorinert (Score:5, Interesting)
I seem to remember hearing that the fluorinert they cooled the processors with was perfectly safe unless turned into a gas, in which case it was roughly as toxic as mustard gas. So, if there was ever an electrical fault in one of the machines that caused the coolant to boil off, there was a distinct possibility that you'd end up with a few dead operators.
Can anyone confirm/deny this? Actually, don't deny -- this is one of my best geek stories.
Re:Distilled Water? (Score:5, Interesting)
In a lab that I worked in we had water that had been passed through several kinds of filters and ion exchangers. You were good to go when the machine said that it had 10+ megaohm of resistance. We stored the water in clean glass bottles, but after about a week had to dump it because the ions leeched from the glass.
Re:Safe? (Score:3, Interesting)
Re:"Water"-cooling (Score:2, Interesting)
Pure H2O (Score:3, Interesting)
Am I right?
Re:Ted Kennedy (Score:1, Interesting)
How does it put out a fire? (Score:3, Interesting)
Would be great for machine rooms.
Distilled? How about ultrapure? (Score:2, Interesting)
So, if you could exclude air, coat the components, and then recirculate the ultrapure water through a resin bed (which is how it's made ultrapure in the first place), it *could* be done- but it would be a lot of work. Plus, very pure water is surprisingly corrosive, so the inerting layer would have to be pretty specific, like the polyethylene that coats the inside of soda cans (cheap but effective).
Water has the benefit of having a large specific heat (4.18 kJ/kg), which is about as good as it gets. While Novec 1230 is good stuff, it has low specific heat (1.103 kJ/kg). It's a trade-off, though, since you can't get pure water below 0 C without the risk of it freezing, unlike Novec, which gets down to -108 C.
Read the datasheet! (Score:4, Interesting)
Heat conductivity & some math (Score:5, Interesting)
A liquid conducts heat EXTREMELY well. You're thinking in terms of a solid, where atoms are fixed and have to transfer energy to each other. However, in a liquid, if one portion of the liquid is heated, this creates a stream of molecules in the liquid to disperse the heat. The heated molecules will actively move away from the heat source, giving room to cooler liquid molecules, which is a hell of a lot more efficient than normal solid-state heat conductivity.
Additionally, it has an heat capacitivity of about 1.1 kJ/kg/degree C, which compares to 4.2 for water. This means that 1.1 kJ (1.1 kW for one second) will heat one kilogram of the stuff one degree Celsius.
One can use this number for some interesting math. A normal box draws maybe 250W, all of which becomes heat. The density of the stuff is 160% of water's. I guesstimate that my tower will hold about twelve liters of water, or about 20 kg of this stuff.
(Note the scientifically correct notation "this stuff".)
Anyway, 20 kg exposed to 250W means that this stuff will heat by 0.75 degrees C every minute if the heat is not dissipated. Assuming a room temperature of 25 deg C, and an electronics-critical point of 45 deg C (the upper bound of operating temperature for some things I've seen; hell, some even have 40 tops), we have a span of 20 degrees, or about 30 minutes of operation until components are out of spec in their operating environment.
Again, this assumes that no heat is dissipated. A miditower probably has about 0.5 to 0.75 square meters of dissipating surface, with good heat transfer from this stuff inside.
Anybody knows if hard drives are built to operate immersed in liquid?
convenient timing... (Score:2, Interesting)
Re:"Water"-cooling (Score:5, Interesting)
Mustard gas [beowulf.org]
Corporate site for Fluorinert [3m.com]
Total immersion (Score:2, Interesting)
Re:"Water"-cooling (Score:5, Interesting)
With a high enough recirculation flow rate, any boiling the would take place would be at the chip. Small bubbles would form and be swept away by the fluid flow.
This process is called "Nucleate Boiling" and is the best heat transfer method there is. The latent heat of vaporization is absorbed by the liquid in it's phase change to a gas. Then the tiny gas bubbles are swept away by the fluid flow and the gas bubbles collapse, giving their latent heat to the surrounding fluid. This heat is later removed by the cooling radiator at the other end.
As long as the bulk temperature of the fluid stays well below the boiling point and the fluid flow is sufficent to strip the small bubbles that form on the heat source surface, this is really the best setup imaginable!
I live the greatest adventure anyone could want -- Tosk the Hunted.
Re:Pure H2O (Score:3, Interesting)
And then, of course, *zap*
Get off my planet (Score:3, Interesting)
What planet are you from? Here on earth we only care about what happens to their stock price!
Get off my planet, you're not ruining it for the rest of us. :-)
Re:Fluorinert (Score:2, Interesting)
As far as Flourinert being used for deep diving, I think the poster is just completely incorrect. I have seen no valid evidence of liquid breathing. Most deep divers use Trimix (a mixture of oxygen, nitrogen, and helium.) One of the major challenges of breathing liquids, other than your alveoli require a gas exchange not likely to happen with a liquid, is how do you circulate the liquid from inside the lungs to the outside environment. Your diaphram and accessory muscles don't have enough strenght to move mass quantities of liquid.
not an ozone killer (Score:2, Interesting)
As further explanation, the presence of fluorine atoms alone doesn't make a ozone depleting compound. CFCs are a problem due to the chlorines getting into the upper atomosphere, the clorine radical is a catalyst for the decomposition of O3. The fluorines aren't an issue since F radicals are too reactive to exist for a long period of time.
Also, the 3M compound is too heavy to make it up to the ozone layer, CFCs were a combination of inertness to the troposphereic environment and being light enough to rise to the ozone layer.
Re:"Water"-cooling (Score:2, Interesting)
The CRAY 2 was completely flouronics cooled. The cooling fluid, made by 3M, allowed the whole computer to be immersed in the electrically insulated fluid, and yet conduct the heat away by conduction and ebullient vaporization. It looked much like a fish tank.
Re:"Water"-cooling (Score:3, Interesting)
But remember, this is at ONE atmosphere or "standard" pressure. It is quite possible that in a relatively small pressure vessel with leads etched through it, that the boiling point could be raised. How much, that is something that has to figured out, but it's not completely unreasonable.
Also, processors burn hotter than 120F because the cooling sucks. If a liquid were to be well circulated, it might be possible to keep the temperature of the outside of the heat sink at a reasonable temp (~100F). But that will depend mainly on the thermal diffusivity of the liquid as well as the circulation rate of the liquid.
So yeah, a boiling point of 120F is pretty shitty, but there are reasonable ways around it, if the properties of this liquid have cooling properties that are worth the effort.
Re:"Water"-cooling (Score:5, Interesting)
total immersion isn't anything new... (Score:2, Interesting)
Re:Specs Data (Score:1, Interesting)
Actually, 3M's FAQ (the PDF linked in the post) claims that it has "zero ozone depletion potential". But it also seems to indicate that it degrades when it gets into contact with UV radiation (ie. keep it out of direct sunlinght). This leaves open the question: is it only ozone-safe as long as it doesn't decompose?
Still the fluorine content of the chemical raises concerns beyond ozone depletion. This stuff may very well be even nastier to produce than, say, teflon. (With similar ecological and safety concerns)
Additional info:
http://www.tldp.com/issue/202/Notes_Fluori
http://www.fluoridealert.org/fluorosis-ind
Re:Fluorinert (Score:3, Interesting)
And besides that, if you watch the scene in question, there's just no way they could've built such a lifelike robo-rat. At least, not cheaply enough to make it worth the cost, compared to just faking the effect.
Environmentally Safe? Cl vs F (Score:3, Interesting)
How do you clean it up? (Score:4, Interesting)
Re:If it doesn't get things wet... (Score:5, Interesting)
The point is that Mercury WILL stick to zinc (amalgam, anyone?) and copper, but zinc is a bit less toxic than copper to have around.
Similarly, I work with 3M's fluorinert liquids quite a bit, although not for cooling. They're useful for some of their other properties (which I'm not revealing right now, because my research could get scooped -- that's science in action!), but the BEST property is that they're STABLE, and they're awfully nonreactive with organics (humans). There've been studies where fish were immersed in fluorinert liquids for long times. Just bubble some oxygen through the stuff so the fish can breathe, and choose the right density, and the fish do fine. Choose the wrong densities, though, and the fish'll be bobbing about on the surface, wondering how the hell to deflate their swim bladders.
Re:If it doesn't get things wet... (Score:2, Interesting)
Re:"Water"-cooling (Score:5, Interesting)
Re:"Water"-cooling (Score:4, Interesting)
I suspect a dunking wouldn't allow any or enough water inside the hard drive, but continued use in such an environment would obviously allow the liquid in.
And obviously it would make CDRoms very hard to use.
But, it does sound like a good idea for a full submersion motherboard. But, you'd probably have to mount the motherboard so all the connectors (keyboard, mouse, etc) were on top, or it would leak.
I'd be just a bit nervous of about having a power supply submerged.
And what happens if there's any condensation, or the liquid gets contaminated by any sort of conductive material? The liquid may be non-conductive, but contaminants would be.
It does sound like a better solution for current design liquid cooling systems. If they leak, it won't fry anything. That is assuming it conducts heat, rather than insulate from it.
Total sub (Score:4, Interesting)
Carbon Tetrachloride worked well too (Score:5, Interesting)
Re:Immersion won't work..... (Score:3, Interesting)
Re:Mod Parent Up (Score:3, Interesting)
Also I believe the gpp is comparing apples with oranges. When you boil your kettle, these bubbles occur all the time but still the temperature of the heating element stays same until amount of cavity increases incredibly and the element heats up, the circuit gets open again. Kettles only have natural convection which is not enough to sustain the steady heat transfer between the fluid and the heating element.
Re:Fluorinert (Score:2, Interesting)
I saw a news report about this, and it doesn't even require an open flame or "ruining" a pan - according to the reporter, all it requires is the same level of heat you'd use to fry bacon.
Re:"Water"-cooling (Score:4, Interesting)
There have been accidents in steel mills where someone allowed water to becomed trapped unter liquid steel, the problem is two fold first is like putting water on a grease fire the instant evaporation causes the hot grease (or steel) above it to explode and 4000 degree steel can cause the water to crack into H2 and O2. Very bad stuff.
LK
Re:"Water"-cooling (Score:5, Interesting)
Because gas makes it soo easy!
As the hard-drive spins, it pulls in some air along with it at the surface. The read/write head literally floats above this fast moving airstream. This allows the head to be made cheeply - they don't have to have inteligent or actuators to keep the head right above the disk itself. They just float.
If you read the spec sheets for most hard drives - they ususall have a limit of 10,000 ft. Above that, the air is not dense enough.
One could argue, why don't the just presurise the drives and seal them off: I've heard that the drives need to out gas for a few months after manufacture.
Re:"Water"-cooling (Score:5, Interesting)
Halon's not a good example of the point you're making.
Halon works at low concentrations, without displacing oxygen, by chemically jamming the chain of reactions in the combustion process. You could think of it as the opposite of a catalyst. The bromine atoms do it, by binding to free radicals that normally keep the fire going. Sorry, I've never found a good online description of exactly how this works.
WAY back when, the manufacturer ran a TV commercial showing a guy in a chair surrounded by a ring of fire. They put out the fire with Halon to show that it didn't interfere with the guy's breathing.
If it were just a matter of displacing oxygen, don't you think they would have used something cheaper?
Re:"Water"-cooling (Score:3, Interesting)
While that makes sense, didn't you read the post?
He filled a small fish tank with Sapphire and submerged a book, a laptop, and a flat panel TV. Both electronics were turned on when submerged; all three items came out completely unharmed.
He submerged a laptop, which obviously has ventilation holes which likely lead to the hard drive. Laptop was completely unharmed, or so they say: just because it was working for the 5 seconds it was on TV doesn't mean it'll continue to function.
Tried it, ... (Score:2, Interesting)
An old P200 tower under the desk was running happily; immersed up to the bottom of the RAM chips! It never missed a beat.
The UPS beside it [there to power the important things] fried beyond repair. The battery ionised stuff maybe, cause it corroded dramatically over the next few days.
The P200 still ticks away in the corner.