atyy said:
Is this paper right?
http://arxiv.org/abs/1108.0320
Unruh effect without trans-horizon entanglement
Carlo Rovelli, Matteo Smerlak
Since in this setup the state of the field in the Rindler wedge is pure, we argue furthermore that the relevant entropy in the Unruh effect cannot be the von Neumann entanglement entropy. We suggest, in alternative, that it is the Shannon entropy associated with Heisenberg uncertainty.
I think it probably is right, if you read carefully what it says. It is talking about the Unruh effect as described in Unruh's original paper, on an accelerating detector. As they say in the paper
if you mean something else by "Unruh effect" then it can be correct to attribute it to entanglement entropy as is often done. But if you focus on the thermality of the detector clicks then they argue there is more to the story.Is there a connection with some of the other discussion, and papers mentioned in this thread?
For clarity, I'll quote the full abstract, since in the fragment of it you quoted it is not clear what "this setup" is and what Unruh effect they are talking about.
==quote==
We estimate the transition rates of a uniformly accelerated Unruh-DeWitt detector coupled to a quantum field with reflecting conditions on a boundary plane (a “mirror”). We find that these are essentially indistinguishable from the usual Unruh rates, viz. that the Unruh effect persists in the presence of the mirror. This shows that the Unruh effect (thermality of detector rates) is
not merely a consequence of the entanglement between left and right Rindler quanta in the Minkowski vacuum. Since in this setup the state of the field in the Rindler wedge is pure, we argue furthermore that the relevant entropy in the Unruh effect cannot be the von Neumann entanglement entropy. We suggest, in alternative, that it is the Shannon entropy associated with Heisenberg uncertainty.
==endquote==
Something to notice about 1108.0320 is that it was submitted to Physical Review D on 4 August 2011 and the final version [v3] was published in PRD on 25 June 2012, almost 11 months later.
http://prd.aps.org/abstract/PRD/v85/i12/e124055
There was no mathematical or logical change that I can see. But in April 2012 a paragraph (highlighted) was added in the introduction which simply repeats and emphasizes a crucial distinction which had already been remarked briefly, in passing, in the first paragraph.
It's possible that a reader of the first version might have MISSED that crucial point.
==excerpt page 1==
An accelerated particle detector clicks even in the vacuum. This is not surprising per se: the detector receives energy from whichever device is accelerating it, and there is no reason why this energy should not be exchanged with the field. What is surprising, however, is the thermal character of these transitions in the case of uniform acceleration, discovered by Unruh [1]: thermal states are states of maximal entropy—whence the entropy of “acceleration radiation”?…
...In this light, the entropy of the Unruh radiation appears to be related to the von Neumann entropy of the improper mixture of right Rindler quanta [6, 7].
If by “Unruh effect” one means the thermal character of the vacuum field fluctuations observable within a Rindler wedge, this is clearly correct. But if we restrict the attention just to the detector’s transition rates, and by “Unruh effect” one means—as in Unruh’s original work and as we do here—the thermal character of the
detector’s transition rates, then, we argue here, the story is subtler and there is more to learn.
A difficulty with the entanglement interpretation of the Unruh effect (in the sense specified) has been pointed out repeatedly, e.g. in [9–11]: it violates causality. The Rindler horizon of an accelerated observer depends on its entire worldline, with proper time ranging from minus to plus infinity. But a physical effect cannot depend on the future history of the observer. This motivated Schlicht to study the Unruh effect in causal terms [10]; he concluded…
==endquote==