Maybe it’s an occupational hazard of dealing with quantum physics — uncertainty and all that — but a group of Swiss researchers has retracted their paper on quantum dots after discovering “major errors” that undermined their conclusions.
The article, published in 2010 as a research letter in Nature Photonics, was titled “Polarization-entangled photons produced with high-symmetry site-controlled quantum dots,” by Eli Kapon and colleagues.
According to the retraction notice:
In this Letter, we claimed the generation of entangled photons from highly symmetric site-controlled pyramidal quantum dots. Apart from the observation of isotropic emission and the measurement of a vanishing fine-structure splitting of 0 μeV ± 2 μeV, a fidelity of >0.5 was obtained by using the density matrix derived from quantum state tomography of 16 different polarization-resolved correlation measurements. After further investigation of the second-order correlation functions, g(2)(τ), major errors were found and the best fidelity found is now 0.45 for a gate width of 0.225 ns. As a result, we can no longer claim emission of polarization-entangled photons. We therefore wish to retract this Letter and sincerely apologize for any adverse consequence that may have resulted from the paper’s publication.
The paper has been cited 69 times, according to Thomson Scientific’s Web of Knowledge. We note that the researchers, who announced their Nature Photonics paper in a brief press release on their website, had a very similar paper in IEEE Xplore, which appears to be conference submission. That work does not appear to have been retracted.
Kapon, a prominent scientist, is director of the Laboratory of Physics of Nanostructures at the Federal Institutes of Technology in Lausanne (one of two such institutes there). He also is listed as being the founder of a company called BeamExpress,
a privately held company developing in cooperation with the Swiss Federal Institute of Technology (EPFL) advanced long wavelength Vertical Cavity Surface Emitting Lasers (VCSELs) with wavelengths ranging from 1200nm to 1650nm, for application in high-speed optical communications.
We’re not sure what “adverse consequences” might have come from the publication of this letter, but we agree with move to retract rather than issue a “mega correction” to try to patch up the tattered analysis.
Does anyone reading this blog know enough applied advanced high energy infrared photonics to comment on the problems with this paper? After all, it has been cited sixty nine times, which is a lot, so it must have had some influence at the time it was published. Comments from an informed source?
Quite a few of those citations were most likely for fashion. It is fashionable to cite papers from high-profile journals in order to make your manuscript look important; the citation then does not endorse all the claims made in a cited paper.
Actually, there is the huge field of semiconductor quantum dots (self-assembled and other like the on in the paper) that do this kind of optics experiments. Normally, for single photon source in quantum computing etc.
But the community seem to know for some time that the results were not as they should be – a colleague just told me last month that they have not cited that paper because the group would go to retract it. The manuscript in question was written half a year ago as far as I know.
The “adverse consequences” bit is no more than a boilerplate. Unless satellites will start falling from the sky because the photons are not really polarization-entangled.
That’s if you’re discounting the (hypothetical) hapless physics PhD students who have tried to replicate theses findings for the last two years and will have to re-boot their work…
I thought all physics PhD students were hypothetical.
For clarification: you should probably state that when you said “Federal Institute of Technology in Lausanne (one of two such institutes there)” the word “there” refers to Switzerland, not Lausanne. The other EPF is in Zurich.
Perhaps there is only one EPF, but in a quantum superposition of states (or, in this case, cantons).