# What could possibly power an Alcubierre drive?

I understand that in order to compress and expand spacetime, an Alcubierre drive would require some sort of "exotic matter" with a negative energy density.

Is that sort of exotic matter purely hypothetical? If not, what would it look like? How would it behave? Is it related in any way to antimatter?

Lastly, would that the exotic matter either fill or form the rings that surround the ship? (as seen in various conceptual designs from NASA) And am I right in assuming that it would be wrong to treat the exotic matter as some sort of fuel that's depleted, rather than as an integral part of the ship itself?

Thanks in advance for any pointers. This is my first question, so I apologize if it's too broad or if it overlaps with any others I may have overlooked.

It is important to note that you do not need negative mass for an Alcubierre drive to work, but negative mass density. The difference is subtle, but important for actually being able to build a warp drive.

Essentially what the equations say is that to bend space in the appropriate manner for a warp drive (i.e. to expand space behind your spacecraft and compress it in front), you need to be able to sum up all of the matter in a given portion of space and get something along the lines of "-10 kg / m3."

Space-time curvature is caused not simply by the mass of matter, but also by how compact it is in a given volume, and what we're trying to do with a warp drive is to curve space in a very specific way, so it doesn't matter what we're doing with the matter-energy in a given volume, so long as we can make its mass density appear to be negative.

The easiest way to do this is to just take a lump of matter that weighs -10 kg, and put it in your space. This however, doesn't make sense, since how can something "weigh" negative 10 kilos? That's like saying I have negative 3 apples. This is why it is sometimes said we need "exotic matter" to make warp drives work, however, there are a few examples of ways to cheat this.

I'll start with an analog in silicon doping. In order to create integrated circuits, bulk silicon is doped to either be positive or negative charged. The way you do this is by adding impurities to pure silicon to change the number of valence electrons in its structure. To make positively charged silicon, boron is added. Because boron has one fewer valence electron vs. silicon, this results in a deficiency in electrons. Similarly, to make negatively charged silicon, phosphorus is added. Because phosphorus has an extra valence electron vs. silicon, there is a surplus of electrons, leading to a negative charge. Because these electrons are surplus, they're free to move around the silicon lattice, and controlling this flow is how circuits work.

The more interesting thing here, however, is what happens when you consider not the electrons as a particle, but the absence of electrons. In circuit design, we call this a hole, and we treat it as a positively charged particle, one that doesn't actually exist.

In this way, you can actually view this as a negative electron density. Essentially, we have created a positively charged particle using only negatively charged matter! This is very similar to what we want for our warp drive if you think of mass not as an electrical charge, but instead a sort of gravitational charge.

There are many equations that treat mass like a gravitational charge, and it's telling that gravity is unique among the fundamental forces in that it does not, to our knowledge, have opposite charges. This means gravity forms monopoles, and everything we've seen about monopoles suggests that nature abhors them. Really it makes no sense, when viewing mass as a kind of charge, for there to not be the concept of an opposite charge.

This probably indicates our models are incomplete with respect to how we view mass and gravity, and indeed, we are unable to combine gravity with quantum mechanics.

So can we find a little more concrete example of negative mass density? Yes, actually, we can find a few.

The first, as others have mentioned, is the Casimir effect. In a nutshell, what happens in the Casimir effect is that two plates close together are pulled together with an extra, anomalous force (i.e. not due to charge or gravity).

The reason for this comes when we look at empty space from a quantum point of view. Quantum physics says that vacuum isn't really empty, but is a seething mass of "virtual particles" constantly popping into existence in pairs then combining with each other and annihilating. Because they recombine, the energy density of the vacuum balances back out to zero (in fact, that it exactly cancels out to zero is considered one of the great unsolved mysteries of physics).

So what does this have to do with the Casimir effect? Well, if space is empty, the vacuum energy can easily cancel back out to zero as virtual particles pop into existence, do their thing, and pop out of existence. However, things change once we add mass (i.e. our two plates).

Virtual particles are treated as a spectrum. As far as the math is concerned, we treat it as an infinite sum of particles that all exactly cancel out to zero. Because the virtual particle spectrum has infinite energies, per De Broglie it also has infinite wavelengths. This is where things start to get interesting once we add our plates.

For a particle to fit between those plates, it must have a wavelength smaller than the distance between them. Now, because our plates are close together, they block some of these virtual particles from popping into existence, namely those with a larger wavelength than the gap between them. Now, more particles of the larger wavelength are forming outside the plates than in between them. What this ultimately means is we've altered the density of virtual particles in space. Because we can say these virtual particles have mass (even if it is short-lived), we have now changed the mass density of the space in between and around the plates. So if normal, empty space has a mass density of zero, what does that mean for the mass density between the plates? Because it has less of a density than the space outside the plates, and that space has a density of essentially zero, it must, for all intents and purposes have negative mass density. We've just created a form of negative mass! The Casimir force however, is incredibly weak, as a particle's energy is directly related to its wavelength. In order to increase the Casimir force, you must decrease the separation of the plates (and thus block shorter wavelength, higher energy virtual particles), and so you can only create very small amounts of negative mass density with it.

Just like our silicon doping example, we've created a negatively charged attribute just from manipulating the distribution of a positively charged one. We have, in effect, "doped the vacuum." Unfortunately, we can't really use this for creating our warp drive, but it does at least show us that negative mass density is possible.

As a sidenote, this "pairing" of positive and negative attributes cancelling out to zero was first envisioned in a vacuum model known as the Dirac Sea. This model basically states that the vacuum is an infinite sea of negative energy particles (hey, that's exactly what we want), and although it turned out to be not true, it did make valuable scientific predictions that were later confirmed, namely the existence of antimatter. Even though the Dirac Sea view is incomplete with regards to the vacuum, it is directly analogous to our silicon doping example.

So are there more useful examples of negative mass density in nature we can use? Yes, it turns out there are. Two of them, in fact.

I will start with the more scientifically accepted version first, one that we can see all around us: The accelerating expansion of the Universe.

It is generally agreed, based on astronomical observations, that the Universe's expansion is accelerating. This creates a bit of a problem: something must power this acceleration, and that means energy. So to make our equations match our observation, we add some energy to the vacuum, making it expand faster.

We treat this as an intrinsic property of space, so as space expands, its density doesn't decrease. That means adding more space is also adding more energy! This creates a new problem: energy means mass, and mass means gravity. If we're adding more and more energy to the Universe as it expands, we're adding more and more mass! The Universe shouldn't be expanding at all then, the energy making it expand faster and faster should itself generate gravity that slows the expansion!! We call this strange energy that seems to come from nowhere "dark energy" because we don't understand what it is at all.

So what do we do about this problem? Mass and energy curve space, so adding more energy should curve space more, and indeed it does (the Universe accelerates faster), but the space is curving in the wrong direction since the Universe is speeding up, not slowing down! Clearly whatever "dark energy" is, it's not "normal" energy in any way.

Looking at things from a general relativity standpoint, the accelerating expansion means that empty space has negative curvature. This may not sound weird, but what it's basically saying is that the negatively-curving space has a negative mass-energy density! All of our data is telling us at every point in the entire Universe, we have negative-mass! We have exactly what we need to build our warp-drive, everywhere, but we don't have the slightest clue what it is! It's like the Universe itself is taunting us!

But once again, it shows us that negative mass is, in fact, a real, natural thing. The only question is how do we create this negative mass in the configuration we need to go really fast?

Our last, best hope lies in a very controversial conjecture of physics: The Woodward effect.

Essentially, the Woodward effect says that if you accelerate any object that is absorbing or discharging energy, that you can create a transient change in the mass of that object. In fact, it is theorized that this transient change can even be negative!

So what kinds of systems can absorb energy and release it that we can also easily accelerate? Well, one of the simplest ways to model such a system is using capacitors and inductors. These are objects that are simple, cheap, and can absorb and discharge energy rapidly without relying on mechanical components.

This leads us to our modern day research into warp fields. Nearly every warp field experiment is using capacitors in some kind of configuration designed to make use of the Woodward effect. Often these capacitor banks take the form of a ring, this is because you can maximize the bending of space inside the ring, thus making it easier to detect.

It is perhaps also telling that fringe theories such as the EmDrive rely on microwave resonant cavities, which themselves behave as a capacitor and inductor tied together. In other words, an EmDrive can be modeled as the exact system predicted to create negative mass by the Woodward effect!

In the end, if it turns out the Woodward effect is real (big if), then we have a way of generating negative mass density, and interestingly enough, you can probably do it with a microwave resonant cavity (which is essentially what an EmDrive is). I find it interesting how all of these seemingly unrelated fringe ideas seem to be converging at the same point.

## So why do NASA's artistic depictions of Alcubierre drive spaceships use rings?

This is a very different question, but one still related to negative mass density. When the Alcubierre drive was first proposed, many scientists did some calculations and decided that even if it were possible to construct negative mass, that you'd need planet-sized amounts of this negative mass to create a warp field (on the order of the mass of Jupiter).

This is clearly unrealistic, and seemed to mean bad news for our plans to go fast, but some other scientists did more calculations and determined that no, you don't need planet-sized amounts of negative mass, you can just do it with less than a metric ton.

So why the huge discrepancy? It all boils down to the topology of your warp field. The scientists who came up with the planetary mass estimate created a warp field that was essentially spherical in shape, whereas the scientists who came up with the much smaller estimate devised a warp field that is toroidal shaped.

But it gets more interesting than that: the speed at which you can travel faster than light, known in warp research fields as the boost velocity, is directly proportional to how thick your warp bubble is. It becomes very difficult to shrink the thickness of your warp bubble since it requires increasing the mass density along the edge of your warp bubble. But in doing so, you minimize the amount of exotic mass you need while maximizing the amount of boost you can get. It turns out the easiest way to create a thin, yet dense edge to your warp bubble like this is to make your bubble the shape of a torus.

Since the warp field is a function of how you arrange your negative mass density, that's where your rings come from.

So do the rings contain the exotic matter? Well, yes and no. As we've already established, you don't need actual matter weighing negative 10 kilos, you need negative mass density, and as we've shown before, it should be possible to create such a mass density using normal matter.

It is therefore more appropriate to say that the rings contain the mechanism for creating negative mass density, and most likely that negative mass density exists as a field around the rings.

To finish my answer, I will leave you with a world-building scenario that allows you to create a warp-drive with minimum hand-waving:

Let's say your warp drive works by taking energy and using it to boost the appropriate type of virtual particle into existence. The how it does this is the only hand-waving part we have.

Virtual particles are everywhere in empty space. They are represented as an infinite spectrum of properties, and always appear in pairs and annihilate, returning the average energy density of the vaccuum to zero. So long as the particles "pay back" the energy for their existence by recombining, they can literally have any characteristics, as long as the pair allows them to combine back to zero. In general, we might say the pair have opposite charges, but the equations work just as well to say the particles have "opposite masses," so long as everything sums up to zero energy in the end.

What our warp drive does then is use energy to force virtual particles that have negative mass into existence. It does this by "paying" for the energy that would have otherwise been paid by the virtual particles annihilating. This is analogous to Hawking radiation, where a black hole "pays" the Universe for a particle in a virtual particle pair, causing one of the virtual particles to be boosted into reality. This particle then escapes the black hole, carrying energy away from it.

So by specifically tuning how we move energy around our spaceship (hand-wave part), like a black hole, we too boost virtual particles into existence, but only those with negative-mass, and use them to form our warp bubble. This warp bubble causes space in front of our ship to compress, and space behind it to expand in a manner similar to how the Universe expands. Because there is no limit to how fast space can expand, there is essentially no limit to how fast we can go. Inside our warp bubble, everything is stationary, but to an outside observer, we seem to be moving faster than light. Because we aren't actually moving, time dilation is only a problem at the infinitesimally small edge of our warp bubble.

This process takes constant energy input, as these negative mass particles want to decay to more stable forms (and in the process presumably acquire positive mass), but they exist on timescales long enough for us to create a warp bubble. Without that energy input, our warp field collapses, and we're stuck at sublight speeds.

• "This may not sound weird, but what it's basically saying is that the negatively-curving space has a negative mass-energy density!" As I said in my comment on Starfish Prime's answer, this is a misconception--dark energy is sometimes said to have negative pressure, but in the main cosmological models that use it, especially the lambda-CDM model which is the current favorite, dark energy has positive energy density. Oct 8 '19 at 0:42
• Of note: "Holes" in electrical components are, in fact, positively charged particles, also known as protons. The distinction is that protons are usually tightly bound in the nucleus and don't actually move, instead forcing the electron to come to the hole. (Ions, such as minerals dissolved in water, often have moving charges that are both positive and negative, moving in opposite directions.) Oct 8 '19 at 2:59
• The Casimir force is a quantum correction to the electromagnetic force that has been widely misunderstood as a vacuum effect because it doesn't go to infinity in the α→∞ limit. See Jaffe (2005). Gravity is different from the other forces because it is spin-2 and the others are spin-1. Spin-1 forces have a charge conjugation symmetry while spin-2 forces don't. Monopole doesn't mean what you seem to think. This answer is a mess of misconceptions and fringe science; please don't upvote it just because it suggests that we'll have a Star Trek future. Oct 8 '19 at 7:14
• As Hypnosifl and benrg noted, this answer mixes true basic statements with errors and misconceptions in way which crafts convincing sounding lie (I always say that best lies have foundation of truth, especially when based on widely known or easily verifiable facts, and introduce fabrications in parts which due to complexity mark is likely to just trust the con artist after accepting and verifying the foundation). Now, since question is not marked with "hard science" or even "science based" tag, this answer fits general needs of soft sci-fi author's worldbuilding. Oct 8 '19 at 12:18
• @MichaelS : Electron holes are mobile quasiparticles and protons are not. In a semiconductor valence band, electron states with energies near the top of the band have negative effective masses -- meaning they move as if they were positively charged -- and these states contribute to current in the material. In semiconductors, p-doped regions have negatively charged particles and excess positively charged quasiparticles moving in opposite directions. Both contribute to conductivity. Oct 8 '19 at 15:01

An Alcubierre drive can't exist at all, if by "drive" you mean something analogous to what you find in Star Trek/Wars: you have a region of your ship filled with some unspecified phlebotinum, you send a go signal to it, it powers up and you go zipping off in whatever direction you want. This is impossible because the exterior part of the Alcubierre warp bubble is a spacelike surface: no part of it lies in the future light cone of any other part. So nothing you do at the start point of your journey can cause the correct spacetime geometry to exist at later points of the journey. Either you have to happen upon existing infrastructure, or you have to start preparing your own infrastructure at a point whose future light cone includes the entire trip. (So, for example, you have to start preparing at least 1000 years in advance if you want to travel 1000 light years.) The Wikipedia article briefly discusses this in the section Alcubierre drive#Placement of matter.

That aside, the main problem is that the "exotic matter" in the solution doesn't follow any physical laws. The field equation of general relativity has the basic form $$\textit{complicated function of second-order derivatives of the metric} = \textit{mass/energy distribution}$$

Finding exact solutions to the field equation for realistic mass/energy distributions is extremely difficult. But finding a solution with no constraint on the mass/energy distribution is trivial: any twice-differentiable spacetime metric is a solution. Alcubierre's warp drive is in the trivial no-constraints category. And, in fact, the mass/energy distribution that you get from it makes no sense. I don't just mean that the energy density is negative; I mean that there's no cause and effect. The exotic matter just appears from nowhere when it's needed and disappears into nowhere when it's not.

In principle one could invent a hypothetical form of matter, invent some plausible physical laws for it to follow, and then find a warpy solution to GR in which the stress-energy tensor actually reflected your exotic matter behaving according to the laws you invented. That would be an interesting theoretical accomplishment. As far as I know no one has succeeded in doing that. So at the moment there is no warp drive physics. There is just warp drive magic, with a false veneer of respectability from the use of the GR field equation.

• Doesn't your first statement only apply to faster-than-light travel with a warp drive? Honestly, I'd be happy with a sublight reactionless drive, if someone could arrange that. Oct 8 '19 at 12:10
• Is your objection just to the requirement of locally superluminal motion of negative energy (in which case I believe the Krasnikov tube avoids this problem), or are you saying that even if there is no specific issue like that, the practice of choosing a spacetime and then finding the matter/energy field to match should be dismissed as "magic", whereas the procedure where you "invent a hypothetical form of matter, invent some plausible physical laws for it to follow" and use that to come up with curved spacetimes is not "magic", even if the matter you imagine is wholly unknown to science? Oct 8 '19 at 17:22
• @Hypnosifl What I meant by "magic" is that you can write down a metric where you accio some object to yourself and it moves for no local causal reason, or a gravitating body spontaneously appears in what was formally a vacuum, or whatever, and they're solutions to GR in the same sense as the Alcubierre metric. Invented but plausible laws would be more like hard-ish sci-fi where there are at least some rules governing what can and can't happen. Also it's really hard to find "lawful" solutions to GR so it would be impressive in a way that Alcubierre's exercise isn't. Oct 8 '19 at 20:53
• @Hypnosifl I read Krasnikov's paper and it appears that all he's saying is that you can prepare the return trip at the same time as the outgoing trip, so the round trip can be arbitrarily short. That's true. What I was saying in that paragraph is that you can't just build a spaceship with warp nacelles in a roughly Minkowski background and travel FTL on a whim (i.e. just after you decide where to go). The trip can only include spacetime points in the Minkowski future light cone of the spacetime point where you plan it. The round-trip time can be as short as you like, though (or negative). Oct 8 '19 at 21:07
• "you can write down a metric where you accio some object to yourself and it moves for no local causal reason" Well, the Einstein field equations allow for non-geodesic motions of the matter/energy field due to non-gravitational forces--in a case like the Krasnikov tube where the negative energy does not have to move locally FTL, if one posits neg. energy is due to the Casimir effect, it seems like it'd be possible in principle to use semiclassical gravity (where matter fields obey quantum field theories like QED) to supply a causal backstory for the motion of the negative energy regions. Oct 9 '19 at 3:28

Is that sort of exotic matter purely hypothetical?

Mostly. There appear to be things that might be suitable... Alcubierre suggested the Casimir effect, but that's pretty weedy. Dark energy might also fit the bill, but as to how you'd actually make it do what you want it to do, instead of appearing out of nowhere and blowing up your universe... your guess is as good as mine. Negative mass cosmic strings may also fit the bill.

If not, what would it look like? How would it behave? Is it related in any way to antimatter?

It almost certainly wouldn't behave or look like matter. Calling it matter is probably something of a misnomer. I expect you wouldn't be able to see it at all as it would be unlikely to interact electromagnetically (though you might get some gravitational lensing effects), and (aside from the associated space-time warping and resulting mangling of you and your ship) you'd probably be able to fly right through a cloud of the stuff without crashing into it as it. As to the rest of its behaviour, who knows?

For specific examples... we don't really have a handle on dark energy, but you probably can't see it or feel it , other than by inferring its presence from accelerating inflation.

Cosmic strings aren't made of "stuff" at all, but are interestingly warped regions of space that can't simply unkink. You can't see it directly, but it will strongly bend light that passes close to it, similar to gravitational lensing.

The casimir effect is an artefact of the quantum vacuum, and so isn't really made of "stuff" either.

Lastly, would that the exotic matter either fill or form the rings that surround the ship? (as seen in various conceptual designs from NASA)

In those designs, probably yes.

However.

Many of the problems with existing warp metrics boil down to needing unbelievably colossal amounts of exotic handwavium to make a warp that could encompass an entire spacecraft. One solution to this is to make a region of warped space that's small on the outside and large on the inside (not really a tardis, because you have to expand and collapse the whole thing to get out). There are a few interesting papers on this (with a few more at the bottom of this Orion's Arm page on reactionless drives) which may or may not interest you.

To avoid crunching and uncrunching space every time you want to get out (which may be risky), and to handle the fact that superluminal travel doesn't appear to be possible with an alcubierre-type drive anyway, the Orion's Arm creators construct a warp drive, crunch it up into a warp bubble and then use it to tow or push a regular spacecraft. The big lumpy things and rings on the outside of their ships are the connecting mechanisms, with all the interesting stuff folded up in a picometre-sized bubble.

And am I right in assuming that it would be wrong to treat the exotic matter as some sort of fuel that's depleted, rather than as an integral part of the ship itself?

As no-one really knows what the stuff would actually be, or how you'd get it, or manipulate it, there's no good answer to this question. Maybe the stuff evaporates when you unclench your warp fields. Maybe the stuff isn't stable. Maybe keeping the stuff in one place is so hard that you just can't help but have it leak out over time, requiring new stuff to be pumped back in to stop your warps going flat. The hands are yours to wave; it seems unlikely that anyone will prove your fictional statements on this matter wrong in your lifetime.

And if they do, it'll probably be for awesome real-life-warp-drive reasons anyway, so its kind of a win-win.

• Most cosmological models involving dark energy say it has negative pressure, but not negative energy--see here on the "fixed positive energy density in all space". A physicist blogger gives a conceptually different (but physically equivalent) explanation here--dark energy has a constant energy density even as space expands, whereas other forms of mass/energy get diluted (on large scales) as space expands. Oct 7 '19 at 22:04
• @Hypnosifl isn't negative pressure more useful than negative energy in this context anyway? Maybe I'm getting my warp drives and wormholes mixed up. Oct 8 '19 at 12:06
• Page 8 of Alcubierre's paper says "one can show that these observers will see an energy density given by [equation] ... The fact that this expression is everywhere negative implies that the weak and dominant energy conditions are violated." The weak energy condition says the matter/energy density is non-negative everywhere. Traversable wormholes violate it too, see p. 1 - 2 here. So, both require negative energy. Oct 8 '19 at 16:34
• @Hypnosifl thanks for clearing that one up! Oct 8 '19 at 19:41

As far as I'm aware (I'm neither an astrophysicist, nor do I play one on television, and I didn't sleep at a particular hotel chain last night), that's been one of the perceived sticking points for construction of an actual Alcubierre drive -- the inability to even define what exotic matter is, never mind say for certain how to obtain or create it.

In order to have negative energy, you'd need to have negative mass, and to a layman, it's not at all clear what that actually means -- since both normal and antimatter have positive mass (i.e. are gravitationally attracted by other masses). It might require a local reversal of the Higgs Field, if that even makes sense (I'm not a particle physicist, either). Or it might require some unfamiliar arrangement of quarks (though I believe I've read that all the possible combinations of two or three of the known quarks and antiquarks -- up, down, and strange and their counterparts -- still produce positive mass).

In the end, for worldbuilding purposes, you can treat "exotic matter" as just another form of handwavium -- something your characters or their culture have discovered how to obtain or make, in order for their warp drives to work, without the need for the reader to know the actual details of what it is, where it's found, or how it's made.

As far as the rings in illustrations of Alcubierre warp ships, I believe you have it exactly right -- the exotic matter is in the rings, and is (in current iterations of the theory) vibrated radially to produce the warp effect, then the warp field somehow offset so that space ahead of the ship shrinks, while that behind expands, carrying the ship (protected by the bubble of the warp field) forward at potentially superluminal speeds.

• "since both normal and antimatter have positive mass (i.e. are gravitationally attracted by other masses)" That's only probably true: en.wikipedia.org/wiki/Gravitational_interaction_of_antimatter Oct 7 '19 at 20:17
• @Harabeck I am certain I've read that there has been experimental confirmation that antihydrogen has weight -- the same weight as the same quantity of hydrogen, to several decimals -- in the Earths' gravity field. Oct 8 '19 at 11:09
• @ZeissIkon I was just looking at that too, but the stuff I found with a few minutes of lazy searching suggested that the error bars are still a bit too big to be really sure. Oct 8 '19 at 12:08
• Photons are affected by gravity, and they are their own antiparticle. Whatever it is that makes mass appears to work on an absolute scale. There is no indication that negative mass is even possible. Oct 8 '19 at 23:56

The answer offered by Zeiss is well-put-together, and so I avoid repeating those thoughts and instead offer another potential solution.

The man behind the Alcubierre drive, Miguel Alcubierre, realized that his demand for a matter with negative energy may not ever be feasible, so he proposed an alternative: a Casimir vacuum. Just as the other answerer, I am nowhere near a proper physicist, much less a quantum physicist -- but in essence, it has to do with the way vacuums interact.

The idea at the core of the Alcubierre drive is to contract the space in front of it, and expand the space behind it, in order to travel faster than light without breaking the laws of relativity. With space mostly being a vacuum, Alcubierre saw potential in the Casimir effect to achieve these ends. How, exactly, is either up to quantum physicists, or in the case of your world, to you.

• The Casimir effect is often taken as a demonstration that what we think of as empty space is not actually at a "ground state" of energy content, but rather is full of particle-antiparticle pair production/annihilation events -- and the two conductive plates used to demonstrate Casimir effect affect that process in a way that appears to reduce the ground state energy between them. Oct 7 '19 at 16:28
• The casimir vacuum between parallel plates isn't exactly an alternative to negative energy, its energy really is negative compared to the energy density of the vacuum in the absence of parallel plates in quantum field theory, though there is uncertainty about how the energy density of the vacuum should be tabulated in the context of the theory of general relativity which describes how mass/energy curves spacetime, this can probably only be resolved by a complete theory of quantum gravity, which we don't have yet. Oct 7 '19 at 17:19
• Also, there are some quantum inequalities that have been derived theoretically which might pose problems for using the Casimir effect for an Alcubierre drive or traversable wormhole, though this paper by Krasnikov (who showed the theoretical possibility of another FTL solution called the Krasnikov tube) argues these inequalities present no fundamental obstacle. Oct 7 '19 at 17:22
• One other point: if vacuum energy is treated as a field with positive energy density in the context of general relativity, and if we calculate that value based on assuming it's the dark energy needed to explain accelerating expansion in GR (method 1 in my 'there is uncertainty' link), then the energy in the Casimir effect can be below vacuum energy by an amount greater than the positive energy density of the vacuum, see post 51 here. So the Casimir vacuum is a good candidate for negative energy in GR. Oct 8 '19 at 17:37

I came here because the title says "what could power an Alcubierre drive", although you actually want to know what drives it. Still, I'll quickly give you my thoughts on the power question.

The power for an Alcubierre drive would have to come from antimatter or black holes because no other known source of energy would be plentiful enough. With black holes there are basically two ways of producing energy: Evaporating small black holes, or throwing matter into black holes and using the resulting radiation.

Throwing matter into black hole has a terrible efficiency: Most of the matter ends up in the black hole which gets heavier in the process.

By contrast, evaporation, like matter-antimatter annihilation, has 100% efficiency — all the mass is turned into radiation. Because only microscopic black holes radiate at all significantly, one would need a huge number of them. The evaporation can be slowed down by throwing mass into the holes (which, as a side effect, also produces some radiation) because larger holes do not radiate much. But there is no way to re-start a large black hole, which is a big disadvantage compared to antimatter which can readily be annihilated. Ideally on would always feed fast-burning black holes at a steady state just below "criticality" and somehow dispose of the energy produced.

Overall it seems that antimatter is easier to handle and use. (I didn't think I'd ever write that sentence.)