That’ll do it: Physics paper retracted for a “pattern that is unphysical”

j phys dLast December, we brought you the story of a math paper that was retracted because it made “no sense mathematically.” Today, we have that retraction’s cousin: A physics paper retracted because some of the data are “unphysical.”

Here’s the notice for “Room temperature ferromagnetism in pure and Co- and Fe-doped CeO2 dilute magnetic oxide: effect of oxygen vacancies and cation valence,” which was published in April 2011 in the Journal of Physics D: Applied Physics:

It has come to the attention of IOP Publishing that some of the data in this paper appears to have been fabricated. The Editorial Board has investigated this and found that the XPS spectra shown in figure 3 all exhibit an identical noise pattern that is unphysical and is inconsistent with the different samples being investigated. Consequently, this paper has been retracted by IOP Publishing.

The following authors have agreed to the publication of this retraction:

Y T Xing

S Kumar

S N Dolia

E Baggio Saitovitch

T Shripathi

U P Deshpande

The journal was unable to contact the following authors:

P Kumari

M Alzamora

The first author of the paper, R. K. Singhal, appears on neither of those lists. He declined to sign the notice, the IOP tells us. We asked him for comment, and will update with anything we learn.

The paper has been cited seven times, according to Thomson Scientific’s Web of Knowledge.

Update, 10/1/13: The Brazilian authors of this paper give some details on what happened (in Portuguese).

8 thoughts on “That’ll do it: Physics paper retracted for a “pattern that is unphysical””

  1. Slowly, but surely, spectroscopy comes under the same spotlight as western blots and some electron microscopy. As we get better at identifying these issues, the number of retractions will rise, though until journals take a proper stance (as IOP has here), we will continue to get ‘corrections” when it should be a retraction.
    Good to see IOp use the correct work, too: ‘fabrication” rather than groping for some laughable euphemism.

    1. So it looks like the x axes of the graphs in one figure were was mislabelled as mg/ml, not pg/ml, since multiple mentions in the text and tables make it clear that it is 60 ng/ml
      Sloppy, not obviously dishonest.

      1. Professor Gallin, thanks for reading this post and that paper. I have been working in the field of endocrinology for 15 years. 60ng/ml (0.22 uM) of E2 in a volume of 20ul medium by a single ovarian follicle is extremely high (I would say it is impossible). Based on my experience, a 90% or more discount of that value seems more reasonable (I am trying to find if other labs have published results of similar range; so far, I failed). Many sloppy things can be found in that lab. Please check out the following papers
        1) PMID:10079409
        2) PMID: 9988412;
        The figure 1 in Papers (1) and (2) are identical.

        3) PMID: 11334907; Figure 1.
        4) PMID: 11821100; Figure 2.
        5) PMID: 16758350: Figure 1.
        How many photographs have been re-used?

        What I presented here should mean something.

        Scientific data go GREEN: Recycling.

  2. I fear the reaction of the Brazilian authors did not seem right to me as they claim no participation on the paper when it got exposed and retracted but they claimed full responsibility for it when the time was to stamp it on their CVs. So one is a real author only when things seem OK?
    I would bet the same authors tried to prevent the retraction and public exposure before the retraction. Why have not issue an authorship correction during investigation then. It is remarkable how their public statement (not a retraction) begins by boasting how grand their image in the field (supposedly) is.

  3. I guess “a pattern that is unphysical” is just one way off saying (some of) the authors fabricated data. It seems that their method of fabrication (take a curve, duplicate it, give it a different color, and say it’s a spectrum from a different compound) is rather un-sophisticated. It probably should have been caught by the reviewers of the paper or the editor. I doubt that this will remain the only retraction. The Applied Physics Letter (Ref 4 of the retracted paper) by some of the same authors, also contains suspect XPS spectra, while the M-H curves in that paper are identical (including noise) as those in Fig 2a of the retracted paper. Only the y-scale has been changed and the ferromagnetic behavior is ascribed to the vacuum annealed CeO2 instead of the hydrogenated CeO2 in the retracted paper.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.