Does the EM drive (electromagnetic drive) work despite the apparent lack of propellant?

Question

In recent headlines, a supposed EM drive, or RF resonant cavity thruster, has been reported to produce thrust despite the apparent lack of any propellant. The drive is purported to work by pushing microwaves through a cone, thereby creating more force on the back end, pushing the object along. Naturally, the scientific community is skeptical that such a device can even exist. Popular Mechanics is blaming populism for the hype (is that irony?). Forbes insists "Physics Falls Apart If The EMdrive Works". The conservation of momentum seems violated by such a theory; there needs to be something of mass moving in the opposite direction to produce thrust. Yet, they, Eagleworks, claim to have something that produces thrust without expelling mass in the opposite direction. Are they lying? Does it work?

Answer 1

From the EagleWorks paper:

The test campaign included a null thrust test effort to identify any mundane sources of impulsive thrust; however, none were identified.

They are not lying, they did detect thrust, but the source of that thrust remains a mystery. The fact that the EM drive would violate Newton's third law, one of the most established principles of physics makes this highly suspicious. Notice that they do not claim that the thrust could not be caused by experimental error (proving a negative is impossible) but that they simply were not able to detect one.

The paper does suggest an alternative theory (pilot-wave), but that in itself is not fully accepted by the scientific community (which doesn't mean it's wrong!).

In conclusion, thrust was measured, but it's not clear if the thrust is caused by experimental error or by currently ill understood quantum effects. The burden of proof lies with the proponents of the EM drive.

Answer 2

Currently, the answer actually seem to be yes!

It was previously reported that radio-frequency (RF) resonant cavities generated anomalous thrust on a low-thrust torsion pendulum [1,2] in spite of the apparent lack of a propellant or other medium with which to exchange momentum. It is shown here that a dielectrically loaded, tapered RF test article excited in the transverse magnetic 212 (TM212) mode (see Fig. 1) at 1937 MHz is capable of consistently generating force at a thrust-to-power level of 1.2±0.1  mN/kW1.2±0.1  mN/kW with the force directed to the narrow end under vacuum conditions.

• Aerospace Research Central of the American Institute of Aeronautics and Astronautics - "Measurement of Impulsive Thrust from a Closed Radio-Frequency Cavity in Vacuum" - November 17, 2016

In a simple explanation, a model of the technology was suspended on a pendulum in a vacuum and was able to produce thrust. In comparison of how much thrust, the Hall-effect thruster produces about 50 times more, but the EM drive seems to produce thrust nonetheless and future models may be able to produce something more comparable. Source

It is shocking and certainly does seem to be a violation of physics as we currently understand it, however, there are a few theoretical explanations. For example:

The team [Eagleworks] also offers a hypothesis about how the drive actually works without contradicting the laws of physics: "[The] supporting physics model used to derive a force based on operating conditions in the test article can be categorised as a nonlocal hidden-variable theory, or pilot-wave theory for short." ...

"If a medium is capable of supporting acoustic oscillations, this means that the internal constituents were capable of interacting and exchanging momentum," the team writes.

"If the vacuum is indeed mutable and degradable as was explored, then it might be possible to do/extract work on/from the vacuum, and thereby be possible to push off of the quantum vacuum and preserve the laws of conservation of energy and conservation of momentum."
Source

I am not a physicist, but that explanation sounds like nonsense, and the scientific community remains skeptical. A very common criticism centers around uncertain instrumentation, and therefore bad measurements. For example:

The more important point is that the individual uncertainties in their instrumentation don't account for the variation in the thrust that they measure, which is a very strong hint that there is an uncontrolled experimental parameter playing havoc with their measurements.
Source

Some also criticize that there are so few data points, which is an indication of lazy experimentation. For example:

Indeed, although the researchers have numerous variables at their hands to change between experiments, they only play with one. In previous papers, they played with two, but still this limited exploration and limited data is really disheartening.
ibid

Conversely, some will likely and eagerly try to reproduce the experiment's results. The above paper from the AIAA, if the findings are legitimate, marks what may be the biggest discovery of our century; the promise is too good to pass up for some.

Answer 3

The answer is no.

The paper does not claim that the thrust comes without "propellant", and it includes a possible physical interpretation of the results not violating Newton's third law (emphasis mine):

If the vacuum is indeed mutable and degradable as was explored, then it might be possible to do/extract work on/from the vacuum, and thereby be possible to push off of the quantum vacuum and preserve the laws of conservation of energy and conservation of momentum. It is proposed that the tapered RF test article pushes off of quantum vacuum fluctuations, and the thruster generates a volumetric body force and moves in one direction while a wake is established in the quantum vacuum that moves in the other direction.

This explanation relies on the Bohmian interpretation of quantum mechanics. In practice, the authors are (partially) trying to substantiate that interpretation with this experiment. Note that this is mostly an interpretation of accepted physics, and not a novel theory.

Regarding whether the thrust is effectively measured, the experiment does take a number of precautions to make sure they eliminate some bias, but it still shows a huge error margin of about 8%:

Thrust data from forward, reverse, and null suggested that the system was consistently performing with a thrust-to-power ratio of 1.2 ± 0.1 mN/kW.

We need to see if the results can be consistently replicated in the future before drawing conclusions.