In the notice, Chang Ming Li from the Institute for Clean Energy & Advanced Materials (ICEAM) at Southwest University in China, states that there is “insufficient evidence to conclusively” identify the composition of the nanowire array described in the article, which “severely undermines the validity of the reported conclusions.”
The 2015 paper has been considered “highly cited” by Clarivate Analytics’ Web of Science, formerly part of Thomson Reuters, meaning it has received a disproportionate amount of cites given its field and publication year.
Li also said that the paper — which appeared in Physical Chemistry, Chemical Physics — was “submitted and published without my knowledge or permission.” He has not responded to our request to explain how that could have happened, given that he was the corresponding author.
According to the notice, first author Bai Sun based his work with the nanowire array on a 2007 paper, which prepared the same nanowire array using a different method. When Sun identified three peaks in the raw data that the 2007 authors had not reported, he attributed them to the substance in which the nanowire array had been grown. Sun, therefore, believed the peaks to be background and removed them from the figure. The notice states that Li “was not made aware of these alterations.”
More notable, perhaps, is that the authors of the 2007 paper issued a correction to their paper that year, in which they explained that they had mischaracterized the composition of the nanowire array and now believed the material was BiOCl instead of BiMnO3. Li said that this correction, of which he had only recently been made aware, “cast doubt on our own interpretation and findings.”
Li and his research group decided to synthesize the material again, using the same method reported in the original paper’s supplementary information, and reanalyzed the composition of the nanowire array. This time they included the three peaks that Sun had removed from the original assessment. In the new analysis, Li concluded that the nanowire array was not, in fact, composed of BiOCl — and may not even be BiMnO3. Given this uncertainty, Li decided to retract the paper, “Light-controlled resistive switching memory of multiferroic BiMnO3 nanowire arrays.”
Here’s the detailed retraction notice:
I, the corresponding author, wholly retract this paper, which was submitted and published without my knowledge or permission.
In this paper, a multiferroic BiMnO3 nanowire array prepared by a hydrothermal method for light-controlled resistive switching memory is reported. The material phase identification was analyzed by the first author, Bai Sun, based on a previous report (ref. 21).1 However, the measured raw data were not exactly in agreement with ref. 21, the raw data displayed three additional small peaks. During analysis these peaks were incorrectly interpreted as the pattern of Ti substrate, and therefore were treated as part of the background and removed. The corresponding author was not made aware of these alterations.
I have recently been made aware that the authors of ref. 21 have corrected their report to attribute their XRD patterns to BiOCl instead of BiMnO3. This cast doubt on our own interpretation and findings. Therefore my research group resynthesized our material according to our reported method and repeated the XRD analysis.
The newly obtained XRD pattern is shown in Fig.1a (black line) and is consistent with the raw data collected in our original study (prior to peak removal). The asterisk labeled peak is attributed to the Ti substrate. The standard XRD patterns of BiMnO3 (JCPDS 89-4544) and BiOCl (JCPDS 82-0485) are also shown in blue and red, respectively. With the exception of the peaks at 30.35, 42.40 and 57.001, the measured peak positions and intensities in Fig. 1a agree well with those in PDF card 82-0485, and not with those in PDF-89-4544.
We further characterized the composition by using energy dispersive spectroscopy (EDS) and X-ray photoelectron spectroscopy (XPS). In Fig. 1b and c, C, Ti, Bi, Mn and O elements are identified. The Ti element is from the Ti substrate. The XPS spectrum of Cl2p in Fig. 1d shows a minor detected amount of Cl element, which we believe is attributed to the use of the chlorinated raw materials rather than the presence of BiOCl. The single crystalline feature shown in Fig. 2d of this PCCP paper suggests that the sample is not a composite of bismuth oxide and manganese oxide. However the elemental compositions tabulated in the inset of Fig. 1b in this notice are not consistent with non-stoichiometric BiMnO3.
Although the evidence presented from the XPS and EDS analysis suggests that the material is not BiOCl, due to the indeterminate identity of 3 peaks in the XRD pattern we have found that there is insufficient evidence to conclusively state that the reported material is BiMnO3. Therefore, due to the uncertain characterization of the material used to form the reported nanowire array, which severely undermines the validity of the reported conclusions, I am retracting the article.
Signed: Chang Ming Li, 7th March 2017.
<<Bai Sun, co-author to the original paper, could not be contacted.>>
Retraction endorsed by Anna Simpson, Executive Editor for PCCP.
The 2015 paper has been cited 40 times, according to Clarivate Analytics’ Web of Science, formerly part of Thomson Reuters.
The notice does not reveal how the paper was submitted and published without Li’s knowledge. We reached out to Li but have not heard back. We could not find contact information for Sun.
Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.