Could a "Santa Claus Machine" ever be built?

[szputnyik]

http://en.wikipedia.org/wiki/Santa_Claus_machine

The Santa Claus Machine is the hypothetical ultimate matter replicator/factory/3D printer.

It could essentially take in matter, disassamble it on the subatomic level and from the acquired cache of particles, reassemble any other matter.

For example, it could suck in air surrounding it, and output a bucket of fried chicken, a light bulb, and Barack Obama at the age of 10 with all the knowledge and experiences he had up to that point.

As long as a perfect, coded description of an item is available for it, it could disassemble inserted matter and reconstruct it indistiguishably from the original item.

Does physics allow such a machine to work, provided we have the technology to construct things on a subatomic level and the computer power to process such complex code, which describes an item particle-by-particle?

Edited February 18, 2014 by szputnyik

[NGTOne]

You're overlooking the fact that it would need an equivalent cache of the appropriate materials - for instance, unless you fed it a 1-ton lump of iron, you wouldn't get a car out of it.

[Streetwind]

He didn't overlook it. He stated "disassemble on the subatomic level". That literally means it can build iron atoms out of neutrons, protons and electrons.

[szputnyik]

But if it could construct atoms from cached protons, neutrons and electrons, or even those from cached quarks, it could also make a car from 1 ton of air or anything else.

[Rhidian]

Converting water into gold (or as you are saying, air into chicken) would require reassembling the subatomic particles themselves to create the necessary materials, which is a matter of nuclear activities (and would likely require a massive input of energy).

If you have the required elements already, that might be more feasible, but that would still require huge amounts of energy. Just turning Graphite into Diamonds, for example, isn't something we have the capability of doing yet (that I am aware of).

[Simon Ross]

In a word...No

While it would be conceivably possible to replicate simple 'non living' items in the way described, you would still need the raw materials (chemicals, proteins etc..) that make up the original item to replicate it.

As for a living organism, forget it. Basic uncertainty principle blocks that straight away

[ZedNova]

In a word...No

While it would be conceivably possible to replicate simple 'non living' items in the way described, you would still need the raw materials (chemicals, proteins etc..) that make up the original item to replicate it.

As for a living organism, forget it. Basic uncertainty principle blocks that straight away

We just need a Heisenberg compensator. Not sure how it'll work, but we'll figure that out later.

[DaveofDefeat]

Tea, Earl Grey, Hot

[TechnicalK3rbal]

No idea if it could work, but I lve your list of example output materials

[Camacha]

While it would be conceivably possible to replicate simple 'non living' items in the way described, you would still need the raw materials (chemicals, proteins etc..) that make up the original item to replicate it.

As for a living organism, forget it. Basic uncertainty principle blocks that straight away

I see no scientific reason why living things could not be created when inanimate objects are not a problem. If you replicate either accurately enough, it will work. The faults probably just become more obvious.

And no, you would not need proteins and all that fuzz, just protons, electrons and neutrons should suffice. At worst you end up with a massive energy difference that needs to be leveled out, but hey, when technology like this comes around we will probably also be a lot closer to easy energy > matter > energy conversions.

So, yeah, unless humanity wipes itself out, it might very well happen, if you just keep on developing technology and keep doing research.

[Seret]

Does physics allow such a machine to work, provided we have the technology to construct things on a subatomic level and the computer power to process such complex code, which describes an item particle-by-particle?

Subatomic assembly is sci-fi, but constructing with atomic precision is one of the goals of nanotechnology. It can already be done on a very small scale, tools like the atomic force microscope can be used to build things atom by atom. The problem is that it's very slow. There are two alternatives to moving atoms around piece by piece, you can either go "top down", where you add or remove bulk sections of material (eg: photolithography, x-ray lithography), or you can go "bottom up", where you use self-assembling structures.

However, the idea of creating macroscale structures with these techniques is not really on the table. We can't even really create fully functioning nanomachines yet.

So my vote goes for "maybe, but not for a very, very long time".

[Scotius]

One hundred years ago we couldn't even imagine a computer Nuclear physics was still a newborn, quantum physics still circled Bohr's and Heisenberg's heads. And look where we are now. Who knows what our grandchildren will have?

[Simon Ross]

Sorry guys, but for this to work on anything above the most simple objects you would pretty much have to re-write every known law of Quantum Mechanics !

Uncertainty Principle states that you can make an accurate observation of a particle's position and momentum... However you cannot do both at the same time !

This actually works very much the same with events in the universe we live in day to day....

Imagine a bullet in flight, you want to log it's exact position and velocity in a single instant of it's trajectory. Well, lets freeze frame it at that instant !

From here we can give an extremely accurate measurement of the bullet's position in 3D space, however we cannot gather any data on the velocity of the bullet, it's no longer moving.

So lets set the bullet moving again. Now we can take an extremely accurate measurement of the bullet's velocity, however, the positional measurement we took of the bullet is no longer valid as the bullet has moved on from that position.

Given this verified law, it is impossible to accurately scan and measure a single particle, let alone a complex object.

For a very simple (non living object) there are ways around this in terms of multiple sampling and determing average positions and energy states. So say we scan a house brick for instance... After billions of individual scans we have collected enough data to enable us to replicate a house brick, however even an object as simple as a replicated house brick will still not be an EXACT copy of the house brick you originally scanned, there will always be accumulated errors in the data.

Now go try this with something as complex as a human being !

Trust me, your going to end up with a puddle of organic mush !

Edited February 18, 2014 by Simon Ross

[fenderzilla]

yes, you probably could, but you need the same amount of matter to be put in as you take out (obviously.) if you're going to make fried chicken or a 10 year old boy from something with the density of our atmosphere, you'd have to suck in a LOT.

[Seret]

Uncertainty Principle states that you can make an accurate observation of a particle's position and momentum... However you cannot do both at the same time !

I agree, that's a definite no to scanning and duplicating anything with atomic precision. Constructing things from a pre-existing blueprint is potentially feasible at some point down the track though. I've always thought that (since cows eat grass) it'd be great to have a machine into which you could dump your lawn clippings and it would make you an artificial steak. In reality though I suspect it would take 6 months to herd all those atoms into the right places!

[Simon Ross]

I agree, that's a definite no to scanning and duplicating anything with atomic precision. Constructing things from a pre-existing blueprint is potentially feasible at some point down the track though. I've always thought that (since cows eat grass) it'd be great to have a machine into which you could dump your lawn clippings and it would make you an artificial steak. In reality though I suspect it would take 6 months to herd all those atoms into the right places!

In truth, I tend to agree with you, it's why I don't completely rule out the possibility of actual replicater type technology in the future. If all you are trying to replicate is a generic blueprint of a non living object be it a steak or a house brick then I think the level of complexity drops a LONG way !

What the steak tastes like however :-)

[lajoswinkler]

No, it's impossible. Heisenberg principle is the fundamental scientific obstacle, and technological obstacles are just... unimaginable.

Anyone comparing this with "we never knew cellphones will be possible" lacks basic understanding of physics.

[LordOfTheOrbits]

I actually think we could do this if we just split it into two parts:

1. Take the input atoms one by one and use nuclear fusion/fission to transform it into the desired atom.

2. Use existing atom level construction to build the desired thing.

The only problem is that it would take several months or years to even build a tiny steak, so when your steak is ready, it's already rotten.

[MBobrik]

Flat out impossible at our or any foreseeable level of technology. Not because of the Heisenberg principle, though. After all, you don't want to force the particles into predefined positions, you want to force them into predefined quantum states and so the Heisenberg principle does not even come into play.

But from there it is all uphill. or more like up a vertical cliff.

First, no known physics allows us to manipulate nucleons with that kind of precision. If we could, even on a extremely crude level, all the mighty fusion machinery like ITER would be rendered obsolete by a rudimentary put_this_proton_over_there contraption used not to build stuff, but to catalyze H+He3 fusion. But no one even attempts that, no known physics can be used that way.

Second, even if it could rearrange atomic nuclei, the nuclear binding energy of the input feed would have to be extremely closely matched with the desired product.

If you were to produce water from a ton of scrap metal, your energy bill would be truly astronomical. And in the other direction, if you tried to make a new car out of water, your machine would have extremely hard time dissipating the excess several megatons of energy in a less destructive way than several megatons of energy, suddenly produced within a small volume, usually tends to dissipate...

Third. Even if the nuclear binding energy would not be an issue, mixing atoms from distinguishable arrangements into an indistinguishable mess produces entropy. And in the other direction, sorting out an indiscriminate mass of atoms into your desired precise arrangement got HUGE entropy cost, that would have to be paid elsewhere.

Fourth. We have means to control positions of individual atoms, ( not precise enough yet ) but we don't have any means of enforcing arbitrarily molecular bond formation. So we could only build stuff that got the molecular bonds that form when atoms are forcibly put together, which prevents us from building directly anything much more complicated at molecular level than bulk mass. We could of course do the building by first building simple chemical precursors, then laboriously guiding them together to let the chemistry do the part we can't directly, then place the end product, but the the process would then get enormously more complicated, and correspondingly slow. Perhaps slower than producing the desired molecules the old way in a chemistry lab.

Edited February 18, 2014 by MBobrik

[Aghanim]

Converting water into gold (or as you are saying, air into chicken) would require reassembling the subatomic particles themselves to create the necessary materials, which is a matter of nuclear activities (and would likely require a massive input of energy).

If you have the required elements already, that might be more feasible, but that would still require huge amounts of energy. Just turning Graphite into Diamonds, for example, isn't something we have the capability of doing yet (that I am aware of).

That is how HPHT diamond works, it converts graphite into diamond

[Camacha]

Given this verified law, it is impossible to accurately scan and measure a single particle, let alone a complex object.

Those are the laws of physics, 'tis true. However, there are many ways to work around or otherwise turn laws of physics that work against you to your favor. You do not need to be spot on to the Planklength. With some trickery sufficient accuracy could possibly be achieved to get a good enough result to work.

Also, you do not have to build the whole thing at once. You could go with all kinds and types of substructures, to make life a little easier. I have a feeling such a machine will be a hybrid of a lot of technologies and concepts, not just a really accurate bricklayer. That is a very straightforward and traditional approach, but that will not get us there - you are absolutely right there.

Mind you, I am not claiming that this is something that is easily done, far, very far from it. But it should not be utterly impossible.

Also, OP never mentioned an object being scanned to the proton, just one object being broken down and another being built, so that discussion is pretty much moot.

[RuBisCO]

Yeah there is never going to be such a machine unless they discovery a way to break all of physics and make 2+2=fish.

A more realistic option would be a machine that can take in energy and raw elements and push out any product made from said energy and elements, so just remove the whole making new atoms from scratch part. How much energy would it require, probably not that much. Do you know how much energy it takes to mine and melt down and make a steel hammer? About 10-12.5 MJ, or about as much energy as a single human being consumes in a day. So depending on how common the elements of your product is the amount of energy needed to make it could vary by several orders of magnitude. If it's a lump of food all you need is a little bit of carbon, hydrogen, oxygen, a dash of phosphorous and sulfur and your good to go at a few MJ per kg of chick nugget goodness, if you want a solid Ruthenium paperweight on the other hand we are talking thousands of giga-joules, at which point atomic synthesis might start to look like a good option.

As for making perfect replicates: there is perfect and than there is 'good enough'. Has anyone ever watched the classic, "Citizen Kane", supposedly one of the greatest movies ever... no you haven't! For the original celluloid tape of Citizen Kane has long since fallen to pieces and oblivion. The original Citizen Kane does not exist, what you have watched is a copy, of a copy, of a copy, usually 'digitally enhance' and reformatted even, this is true of just about every movie you have seen... I don't hear anyone complaining.

[Camacha]

As for making perfect replicates: there is perfect and than there is 'good enough'.

What do you think DNA does? It is actually a requirement to copy that less than perfect, otherwise a lot of evolution would be a much slower process. The error rate is a fairly constant number, just the how and what is sort of random.

[jwenting]

If you have the required elements already, that might be more feasible, but that would still require huge amounts of energy. Just turning Graphite into Diamonds, for example, isn't something we have the capability of doing yet (that I am aware of).

we've actually been doing that for decades, typically using a microscopic bit of diamond dust as a seed on which to grow the stone.

[problemecium]

Impossible? No.

Likely? Also no. I have little confidence our species will both survive long enough and maintain a net positive rate of technological advancement for enough of its lifetime.

And by the time this would be invented, stuff like ITER would be ancient, primitive technology anyway. Fusion is one thing - a Santa Claus machine on the other hand is faaar-future stuff. Like Dyson spheres and, well, magic.