Quantum physics paper pulled for “serious theoretical errors,” notice accidentally paywalled

A paper on photonic quantum walks has been retracted over a theoretical disagreement.

The notice is also paywalled, which the editorial director has assured us is a mistake that is being corrected.

We sent the COPE guidelines on retraction to the American Physical Society, which publishes Physical Review Letters. Editorial director Dan Kulp told us the paywall was the unintentional consequence of a web redesign, and that they are in the process of restoring public access to “all Errata-types, including Retractions.”

Here’s the rest of his statement:

Thank you for pointing out our inconsistency with the COPE guidelines as far as Retractions are concerned.  This was not intentional and actually an oversight.  I believe that the “paywall” appeared when we recently redesigned our webpages and this inconsistency was missed.

…
I have the IT group working the paywall issue.  Our business logic should have all Errata-types, including Retractions, outside the paywall.  This will, hopefully, be corrected soon.

As for the Retraction, the error was discovered by readers (those thanked in the Retraction) soon after the publication of the Letter.  The group questioned the theoretical existence of the localized bound states (as far as I understand the issue).  Over the intervening months, the two group corresponded and reached the mutual conclusion that the Letter was indeed flawed.  The authors then submitted a Retraction and that was published roughly three months after the publication of the Letter.

Here’s the notice for “Observation of Topological Structures in Photonic Quantum Walks” (still paywalled as of publication):

Because of serious theoretical errors that have consequently led to misinterpretations of experimental data, the main result of our recent Letter concerning the observation of topological structures and localization in photonics quantum walks is incorrect and has to be withdrawn.We thank Peng Xue, Hao Qin, Rong Zhang, Zhihao Bian, and Xiang Zhan as well as János Asbóth for pointing out these errors to us.

The paper has been cited twice, according to Thomson Scientific’s Web of Knowledge.
Hat tip: Murray