Towards traversable wormholes from force-free plasmas

1) Very well written paper
2) Exceptionally clear
2) Very novel idea
3) Clear description of calculations

1) The mechanism they study unfortunately doesn't lead to large traversable wormholes

This paper attempts to construct large traversable wormholes using standard physics arising from charged particles interacting with a strong magnetic field, in regimes where a particular low energy effective theory is useful. The problem of constructing large wormholes without invoking physics beyond the standard model is of great interest, and the approach taken here is a novel and interesting mechanism, certainly worthy of exploration.

This is a very well written paper, that is exceptionally clear and well organized -- and overall very enjoyable to read. I therefore recommend the paper for publication.

Constructing traversable wormholes in the low energy theory of our universe is an important and exciting problem. This paper proposes an interesting new idea to achieve this by using collective light modes arising from the force-free electrodynamics description (which has proven useful in astrophysical setups). At the end, the simple setup described in the paper doesn't lead to a consistent wormhole solution. Nonetheless, the idea is interesting and deserves further study. The nearly-AdS2 perspective on these collective modes is also of some interest. The analysis is clear and convincing. Therefore, I recommend this paper for publication in SciPost Physics.

That said, I would like to ask the authors some minor clarifications/questions:

1) Is it be possible to clarify a little bit the physical setup, to give some intuition to a reader who doesn't know much about FFE? If I understand correctly, the Alfven modes are supposed to describe coherent excitations of the plasma and EM field, so does this mean that the hypothetical wormhole must be filled with plasma? Does this has any physical significance?

2) The Alfven wave modes are supposed to follow the magnetic field lines in closed loops threading the wormhole. A question that I haven't seen addressed is the following: is it clear that the FFE approximation is also valid in the outside region, far from the black holes? Shouldn't this put a constraint on the outside distance d?

3) In [1], a large magnetic field was crucial to have a large number of Casimir modes. This doesn't seem to be the case here where it's only needed for the FFE to be valid. So could we repeat the analysis in a more realistic astrophysical setup? For example by replacing the RN black hole with near-extreme Kerr in a strong magnetic field?

The paper studies the possibility of making travesable wormholes in 4d Einstein gravity by using Casimir energy of Alfeven waves of force-free electrodynamics(FFE). Although the idea of using Casimir energy is not new, and ultimately authors encountered serious obstacles to realizing this idea, FFE near black holes is an interesting topic which recently drew some attention in the literature. Also taking into account that the paper is clearly written and authors discuss various physical effects, I recommend the paper
for publication with major revisions.
Here is the list of questions I would like the authors to address:
1) FFE seems to be an emergent phenomena when standard electrodynamics interacts with matter. The authors considered the correction to the (total) stress-energy tensor from Alfven waves. Does this take into account the stress-energy tensor of underlying matter?
2) It seems that eventually authors impose simple periodic boundary conditions on matter. Is Section 4 necessary then?
3) The conclusion of the paper seems that it is hard to sustain a wormhole with FFE in 4d. In higher dimensions the black hole story is well studied(AdS2 near horizon, 2d Casimir energy, etc), _except_ the dispersion relation for Alfven waves. It would be nice if the author could comment on this.
4) The lengthy calculation in Section 5.2 results in the condition $m^2 < 1/(R l)$. The authors explain this result as: "...AdS kinematics imply that the relevant mass scale is the geometric mean...". I think it would be very helpful for readers to see this kinematic estimate explicitly. Naively, in eq. (41) mass is responsible for $1/(\cos^2 \sigma)$ potential, which is cutoff at $|\sigma-\pi/2| \sim R/l $. Since $\omega \gtrsim 1$, the effect of the mass is negligible for $m l \lesssim 1$.