Can the EmDrive actually work for space travel?

The "impossible" EmDrive engine, which purportedly generates thrust by bouncing microwaves around inside a cone-shaped chamber.
The "impossible" EmDrive engine, which purportedly generates thrust by bouncing microwaves around inside a cone-shaped chamber. (Image credit: SPR Ltd./

The "EmDrive" claims to make the impossible possible: a method of pushing spacecraft around without the need for — well, pushing. No propulsion. No exhaust. Just plug it in, fire it up and you can cruise to the destination of your dreams. 

But the EmDrive doesn't just violate our fundamental understanding of the universe; the experiments that claim to measure an effect haven't been replicated. When it comes to the EmDrive, keep dreaming. 

Related: Superfast spacecraft propulsion concepts (images)

Microwaves of the future

It goes by various names — the EmDrive, the Q-Drive, the RF Resonant Cavity, the Impossible Drive — but all the incarnations of the device claim to do the same thing: bounce some radiation around inside a closed chamber, and presto-chango you can get propulsion.

This is a big deal, because all forms of rocketry (and indeed, all forms of motion across the entire universe) require conservation of momentum. In order to set yourself in motion, you have to push off of something. Your feet push off of the ground, airplanes push themselves off of the air, and rockets push parts of themselves (e.g., an exhaust gas) out the back end to make them go forward.

But the EmDrive doesn't. It's just a box with microwaves inside it, bouncing around. And supposedly it is able to move itself.

Explanations for how the EmDrive could possibly work go past the boundaries of known physics. Perhaps it's somehow interacting with the quantum vacuum energy of space-time (even though the quantum vacuum energy of space-time doesn't allow anything to push off of it). Perhaps our understanding of momentum is broken (even though there are no other examples in our entire history of experiment). Perhaps it's some brand-new physics, heralded by the EmDrive experiments.

Don't play with momentum

Let's talk about the momentum part. Conservation of momentum is pretty straightforward: in a closed system, you can add up the momenta of all the objects in that system. Then they interact. Then you add up the momenta of all the objects again. The total momentum at the beginning must equal the total momentum at the end: momentum is conserved.

The idea of the conservation of momentum has been with us for centuries (it's even implied by Newton's famous second law), but in the early 1900s it gained a new status. The brilliant mathematician Emmy Noether proved that conservation of momentum (along with other conservation laws, like conservation of energy) are a reflection of the fact that our universe has certain symmetries.

For example, you can choose a suitable location to perform a physics experiment. You can then pick up your physics experiment, transport it to anywhere in the universe and repeat it. As long as you account for environmental differences (say, different air pressures or gravitational fields), your results will be identical.

This is a symmetry of nature: physics doesn't care about where experiments take place. Noether realized that this symmetry of space directly leads to conservation of momentum. You can't have one without the other.

So, if the EmDrive demonstrates a violation of momentum conservation (which it claims to do), then this fundamental symmetry of nature must be broken.

But almost every single physical theory, from Newton's laws to quantum field theory, expresses space symmetry (and momentum conservation) in their base equations. Indeed, most modern theories of physics are simply complicated restatements of momentum conservation. To find a breaking in this symmetry wouldn't just be an extension of known physics — it would completely upend centuries of understanding of how the universe works.

The universe: Big Bang to now in 10 easy steps

The reality of experiment

That's certainly not impossible (scientific revolutions have happened before), but it's going to take a lot of convincing to make that happen.

And the experiments so far have not been all that satisfying.

Ever since the introduction of the EmDrive concept in 2001, every few years a group claims to have measured a net force coming from its device. But these researchers are measuring an incredibly tiny effect: a force so small it couldn't even budge a piece of paper. This leads to significant statistical uncertainty and measurement error. 

Indeed, of all the published results, none have produced a measurement beyond "barely qualifying for publication," let alone anything significant.

Still, other groups have developed their own EmDrives, attempting to replicate the results, like good scientists should. Those replication attempts either fail to measure anything at all, or found some confounding variable that can easily explain the measured meager results, like the interaction of the cabling in the device with the Earth's magnetic field.

So that's what we have, nearly 20 years after the initial EmDrive proposal: a bunch of experiments that haven't really delivered, and no explanation (besides "let's just go ahead and break all of physics, violating every other experiment of the past 100 years") of how they could work. 

Groundbreaking, physics-defying revolution in space travel or a pipe dream? It's pretty clear which side Nature is on.

Paul M. Sutter is an astrophysicist at SUNY Stony Brook and the Flatiron Institute, host of Ask a Spaceman and Space Radio, and author of How to Die in Space. He contributed this article to's Expert Voices: Opinions and Insights.

Learn more by listening to the episode "Could the "EmDrive" really work? on the Ask A Spaceman podcast, available on iTunes and on the Web at Thanks to Mitchell L. for the questions that led to this piece! Ask your own question on Twitter using #AskASpaceman or by following Paul @PaulMattSutter and

Follow us on Twitter @Spacedotcom or Facebook. 

Paul Sutter

Paul M. Sutter is a research professor in astrophysics at  SUNY Stony Brook University and the Flatiron Institute in New York City. He regularly appears on TV and podcasts, including  "Ask a Spaceman." He is the author of two books, "Your Place in the Universe" and "How to Die in Space," and is a regular contributor to, Live Science, and more. Paul received his PhD in Physics from the University of Illinois at Urbana-Champaign in 2011, and spent three years at the Paris Institute of Astrophysics, followed by a research fellowship in Trieste, Italy. 

  • FB36
    There is actually no need for exotic physics for creating a working EmDrive!

    ALL photons (including MW photons) have/carry momentum!
    So, for example, any laser device in space would act like an EmDrive and produce (tiny) thrust!

    The problem w/ current EmDrive prototype is that it does NOT have a hole to let generated MW photons to escape and carryaway momentum (& so produce thrust)!

    (One may ask, if so then, what exactly would be best EmDrive for space?
    Any photon (light/laser, RF, MW) source would act as an EmDrive in space (since ALL photons carryaway momentum)!
    Then it depends on what kind of device can produce highest energy/momentum photons most efficiently!
    (It maybe a UV laser (array)?))
  • Chem721
    FB36 said:
    (It maybe a UV laser (array)?

    This reminds me a bit of a fusion reactor generating propulsion. People have suggested that such a source of energy might provide for enormous thrust from a low mass fuel source. You have an onboard reactor that is working at 15 miilion K or so, and feeding tiny amounts of hydrogen keeps it cranking out a lot of energy. But how do you get propulsion from that energy?

    Can you convert the energy into some form of massive "photon thrust" in order to obtain propulsion? Can any one see how to get a "UV laser (array)", or even higher, like X-Ray or Gamma-Ray drive? You would still need to exhaust a lot of photons, and how would you generate them from the energy of fusion?

    Clearly a static device producing a lot of energy is not going anywhere. But could there be a future for fusion-based EmDrive?!

    See my thread on asking how a fusion-powered drive could work at:
    As expected, no one came up with even a reasonable idea of how to get this to work. But fusion and EmDrive, that is a new concept, at least for me!! :)
  • funny guy
    its may be thermodynamic . use a big elastic with the outside pressure then its a wave back and forth motion and the gas pressurising . a cylinder could also help. It would be interesting if it ads a vector to a 7 or 8 mass system. a spining bearing moving up and down , a spin mass with elongating wobble, a surface spin like with a heavy mass object for astounding . You could maybe even create a geometrical horizon and perhaps a mini mass of the 5th dimension. also more vectorial spin of a pentagonal maybe even octogonal pattern .Like when they say things falling in a blackhole do.
  • Inspector
    Why don't we just use laser ablation?
  • Chem721
    Inspector said:
    Why don't we just use laser ablation?

    Are you suggesting that one or more large lasers inside of "thrust chambers" hammers some kind of fuel that provides exhaust gases as super-high temperatures? That might work, using containment fields like in the fusion reactor.

    It is the best idea I have heard yet regarding fusion-powered spacecraft, although it does not involve EmDrive, or does it? Are you proposing it for an EmDrive mechanism, or fusion-powered drive?