A new article in the New Scientist begins with a statement of Newton’s third law that is blissfully ignorant of the irony. The article’s headline is:
The magazine is notorious for its use of sensationalist headlines and seems to have done it again. Jon Cartwright, the author of the article, has done a decent job of explaining the ‘helical drive’ proposed by a manager at NASA named David Burns, and hasn’t himself suggested that the drive violates any laws of physics. It seems more like someone else was responsible for the headline and decided to give it the signature New Scientist twist.
The featured image is a disaster, showing concept art of Robert Shawyer’s infamous em-drive. Shawyer had claimed the device could in fact violate the laws of physics by converting the momentum of microwaves confined in a chamber into thrust. Various experts have debunked the em-drive as fantasy, but their caution against suggesting the laws of physics could be broken so easily appears to have missed the New Scientist.
Update, 7.06 am, October 16, 2019: In a new article, Chris Lee at Ars Technica has explained why the helical drive won’t work, and comes down harshly on Burns for publicising his idea before getting it checked with his peers at NASA, which would’ve spared him the embarrassment that Lee dished out. That said, Lee is also a professional physicist, and perhaps Cartwright isn’t entirely in the clear if the answer to why the helical drive won’t work is as straightforward as Lee makes it to be.
With the helical drive, Burns proposes to use an object that moves back and forth inside a box, bouncing off either end. Each bounce imparts momentum to the box but the net momentum after two bounces is zero because they’re in equal and opposite directions. But if the object could become heavier just before it strikes one end and lighter before it strikes the other, the box will receive a ‘kick’ at one end and start moving that direction.
Burns then says if we could replace the object with a particle and the box with a particle accelerator, it should be possible to accelerate the particle in one direction, let it bounce off, then decelerate it in the other direction and recover most of the energy imparted to it, and repeat. This way, the whole setup can be made to constantly accelerate in one direction.
The flip side is that the mass-energy equivalence is central to Burns’s idea, but according to the theory of special relativity that it’s embedded in, it’s actually the mass-energy-momentum equivalence. As Lee put it, special relativity conserves energy and momentum together, which means a heavier particle bouncing off one end of the setup won’t keep accelerating the setup in its direction. Instead, when the particle becomes heavier and acquires more momentum, it does so by absorbing virtual photons from an omnipresent energy field. When the particle slows down, it emits these photons into the field around it.
According to special relativity and Newton’s third law, the release process will accelerate the setup, and the absorption process will decelerate the setup. The particle knocking on either ends is just incidental.