Followup to: Cryonics wants to be big
We've all wondered about the wisdom of paying money to be cryopreserved, when the current social attitude to cryopreservation is relatively hostile (though improving, it seems). In particular, the probability that either or both of Alcor and CI go bankrupt in the next 100 years is nontrivial (perhaps 50% for "either"?). If this happened, cryopreserved patients may be left to die at room temperature. There is also the possibility that the organizations are closed down by hostile legal action.A
The ideal solution to this problem is a way of keeping bodies cold (colder than -170C, probably) in a grave. Our society already has strong inhibitions against disturbing the dead, which means that a cryonic grave that required no human intervention would be much less vulnerable. Furthermore, such graves could be put in unmarked locations in northern Canada, Scandinavia, Siberia and even Antarctica, where it is highly unlikely people will go, thereby providing further protection.
In the comments to "Cryonics wants to be big", it was suggested that a large enough volume of liquid nitrogen would simply take > 100 years to boil off. Therefore, a cryogenic grave of sufficient size would just be a big tank of LN2 (or some other cryogen) with massive amonuts of insulation.
So, I'll present what I think is the best possible engineering case, and invite LW commenters to correct my mistakes and add suggestions and improvements of their own.
If you have a spherical tank of radius r with insulation of thermal conductivity k and thickness r (so a total radius for insulation and tank of 2r) and a temperature difference of ΔT, the power getting from the outside to the inside is approximately
25 × k × r × ΔT
If the insulation is made much thicker, we get into sharply diminishing returns (asymptotically, we can achieve only another factor of 2). The volume of cryogen that can be stored is approximately 4.2 × r3, and the total amount of heat required to evaporate and heat all of that cryogen is
4.2 × r3 × (volumetric heat of vaporization + gas enthalpy)
The quantity is brackets for Nitrogen and a ΔT of 220 °C is approximately 346,000,000 J m-3. Dividing energy by power gives a boiloff time of
1/12,000 × r2 × k-1 centuries
Setting this equal to 1 century, we get:
r2/k = 12,000.
Now the question is, can we satisfy this constraint without an exorbitant price tag? Can we do better and get 2 or 3 centuries?
"Cryogel" insulation with a k-value of 0.012 is commercially available Meaning that r would have to be at least 12 meters. A full 12-meter radius tank would weigh 6000 tons (!) meaning that some fairly serious mechanical engineering would be needed to support it. I'd like to hear what people think this would cost, and how the cost scales with r.
The best feasible k seems to be fine granules or powder in a vacuum. When the mean free path of a gas increases significantly beyond the characteristic dimension of the space that encloses it, the thermal conductivity drops linearly with pressure. This company quotes 0.0007 W/m-K, though this is at high vacuum. Fine granules of aerogel would probably outperform this in terms of the vacuum required to get down to < 0.001 W/m-K.
Supposing that it is feasible to maintain a good enough vacuum to get to 0.0007 W/m-K, perhaps with aerogel or some other material. Then r is a mere 2.9 meters, and we're looking at a structure that's the size of a large room rather than the size of tower block, and a cryogen weight of a mere 80 tons. Or you could double the radius and have a system that would survive for 400 years, with a size and weight that was still not in the "silly" range.
The option that works without the need for a vacuum is inviting because there's one less thing to go wrong, but I am no expert on how hard it would be to make a system hold a rough vacuum for 100 years, so it is not clear how useful that is.
As a final comment, I disagree that storing all patients in one system is a good idea. Too many eggs in one basket is never good when you're trying to maximize the probability that each patient will survive. That's why I'm keen on finding a system that would be small enough that it would be economical to build one for a few dozen patients, say (cost < 30 million).
So, I invite Less Wrong to comment: is this feasible, and if so how much would it cost, and can you improve on my ideas?
In particular, any commenters with experience in cryogenic engineering would delight me with either refinement or critique of my cryogenic ideas, and delight me even more with cost estimates of these systems. Its also fairly critical to know whether you can hold a 99% vacuum for a century or two.
A: In addition to this, many scenarios where cryonics is useful to the average LW reader are scenarios where technological progress is slow but "eventually" gets to the required level of technology to reanimate you, because if progress is fast you simply won't have time to get old and die before we hit longevity escape velocity. Slow progress in turn correlates with the world experiencing a significant "dip" in the next 50 or so years, such as a very severe recession or a disaster of some kind. These are precisely the scenarios where a combination of economic hardship and hostile public opinion might kill cryonics organizations.
With the development of commercial space flight, at some point launching cryonauts into space might become cost-effective.
Right now, with a human head weighting about 5 kg, launching it would cost about $150,000 (not counting the cryopreservation equipment, which is probably significant, and has to withstand the launching stress). Comparing this with a price tag of Alcor full-body preservation, which is also $150,000, it's not totally bonkers to suppose that in a few decades it could become competitive, even without the fancy space elevators.
If it's possible to use the low temperature in space, despite the solar radiation, to keep the temperature down, or somehow keep the package in the shadow, maybe of a specifically crafted accompanying object (I'm not sure about that -- while it's a critical question), it could be a no-maintenance solution, where one would have to perform a deliberate and rather costly procedure to disturb it.
Is the radiation going to cause significant information-theoretic damage? In how long?
Shopping is hard, let's do math!
First, we need a conceptual framework. The whole point of cryonics is to stop chemistry, so if you're cryopreserved and then exposed to ionizing radiation over any period of time, you'll experience the same amount of damage as if you were alive and exposed to that much radiation all at once. (Being alive and exposed to radiation over a period of time is different; you experience less damage because your cells have time to repair themselves.)
Wikipedia says "Estimates are that humans unshielded in interplanetary space would receive annually roughly 400 to 900 milli-Sieverts (mSv) (compared to 2.4 mSv on Earth)". Wikipedia also says that an acute exposure of 4500 to 5000 mSv is "LD50 in humans (from radiation poisoning), with medical treatment". Now, LD50 isn't LD100, but we can agree that it's a Very Bad Dose.
Generously, assuming that the Very Bad Dose is 5000 mSv, and Outer Space's Death Rays are 400 mSv/yr, being Cryopreserved In Space will give you a Very Bad Dose in 12.5 years. This is compared to roughly 2000 years on Earth.
That answers one half of Eliezer's question. My answer to the other half (is this significant in informat... (read more)
Good idea! A few refinements:
You probably don't want a literally spherical tank; it might roll away and hit something or bother someone. Trading a few % of efficiency for a flattened, ridged bottom might be a good idea.
If you're going to rely on social taboos against disturbing graves, you probably have to keep bodies/tank down to 30, if not an even lower number. A group of family and friends who are buried together in the same crypt are eccentric; a community of essentially unrelated people who are buried together in the same crypt are a cult, and
I think that the idea is good, and the engineering is fine for back-of-the-envelope, but can we please call it a "vault" or something instead of a grave? Cryonics already has an image problem, and we don't want to suggest the people in the grave are permanently dead.
Why? Several baskets certainly make sense if you're trying to maximize the probability that at least a few patients survive, and might make sense if you assign significantly negative utility to higher variance in your probability distribution about survival percentage. If you just care about the mean, why would more baskets be better?
(a) Because a single, large target is very inviting for people to try and break deliberately, but if there are N small targets spread out all over the place, the effort required to inflict a lot of damage is extreme, especially if each grave is under 5 meters of concrete and in a remote location.
Also, the idea of killing "the elitists" by rupturing their collective cryograve seems more righteous than going to Jack Smith's grave and specifically killing Jack. It seems more like murder that way. So the optimal solution is one person per grave.
(b) because graves can use slightly different technology, and in the time between when you set the scheme up and when civilization or just cryonics companies collapse, you can see which designs actually fail, and rescue the patients inside them. A large population of say, 100 graves, with 10 examples of each design type will yield information about what works best as the worst ones break. Then over time you should become more confident in the remaining designs that have zero instances of failure.
I think you're Rokomorphizing an awful lot. You just need to be in a state of mind where smashing a cyro container seems cool, something that can score points with your friends, and where you think you can get away with it.
With liquid nitrogen at -196°C and the average temp in the places you suggest well below freezing (A few minutes of googling suggests it wouldn't be hard to find an average annual temp of -20°C.), I think you could use a more-optimistic ΔT of 175°.
If you care about cryonics and its sustainability during an economic collapse or worse, chemical fixation might be a good alternative. http://en.wikipedia.org/wiki/Chemical_brain_preservation
The main advantage is that it requires no cooling and is cheap. People might be normally buried after the procedure, so it would seem less weird.
However, a good perfusion of the brain with the fixative is hard to achieve.
Chemical fixation could also be combined with those low maintenance cryonic graves just in case the nitrogen boils off.
Australia claims 42% of Antarctica. That should be plenty of room.
Why limit yourself to no maintenance at all in your feasibility speculations? Tending graves is common across cultures. As long as you're spinning a tank of liquid nitrogen as a "grave", why not spin a nitrogen topoff as equivalent to keeping the grass trimmed or bringing fresh flowers?
A couple thoughts on places to look for ideas, places where people have probably been thinking about similar challenges:
My comments in this sub-thread brought out more challenges and queries than I expected. I thought that by now everyone would expect me to periodically say a few things out of line regarding identity, consciousness, and so on, and that only the people I was addressing might respond. I want to reply in a way which provides some context for the answers I'm going to give, but which covers old territory as little as possible. So I would first direct interested parties to my articles here, for the big picture according to me. Those articles are flawed in various... (read more)
You seem to jump to the conclusion that, in the favorable case, (that consciousness only exists in quantum computers AND quantum coherence is the fundamental basis of persistent identity), the coherence timescale would obviously be your whole lifetime, even if hypothermia, anesthetics, etc happen, but as soon as you are cryopreserved, it decoheres, so that the physical basis of persistent identity corresponds perfectly to the culturally accepted notion.
But that would be awfully convenient! Why not assign most of your probability to the proposition that evolution accidentally designed a quantum computer with a decoherence timescale of one second? ten seconds? 100 seconds? 1000 seconds? 10,000 seconds? Why not postulate that unconsciousness or sleep destroys the coherence? After all, we know that classical computation is perfectly adequate for evolutionarily adaptive tasks (because we can do them on a classical computer).
Mass cryonic suspension does not seem likely to be affordable anytime soon: "As of 2010, only around 200 people have undergone the procedure since it was first proposed in 1962" - http://en.wikipedia.org/wiki/Cryonics