For an IEEE retracted paper citation, check the publisher's current record, read the linked notice, and decide why the source belongs in your manuscript. Use the notice to document the publication event. Check the underlying evidence before retaining a technical claim.
Open the publisher status
Start from the DOI in your reference manager and open the publisher's article page. Look near the title for a correction, retraction, expression of concern, or update history. Then check the passage you intend to use. An older PDF may still contain the wording or equation that prompted a later notice.
Crossmark offers a useful shortcut where a publisher supplies it. Its button opens status information and links to reported updates. Crossref also cautions that the presence of Crossmark is no guarantee. Treat it as another route to the editorial record. If the button is absent, inspect the publisher's update history and linked notices directly.
IEEE's post-publication policy explains why a retracted paper may remain readable. It normally stays in the archive with its status clearly indicated. Continued access preserves the scholarly record; it does not restore confidence in the findings. Check the status even when the DOI resolves and the PDF downloads normally.
This site's generator formats references and retrieves bibliographic details; it has no retraction scanner. Use the broader reference-audit guide for author, DOI, and numbering errors. Here we checked two Nature-family publications and their formal notices to show the additional work needed for an IEEE reference list.
Follow the correction or retraction notice
The first record is the 2023 lutetium-hydride report. Its original DOI is 10.1038/s41586-023-05742-0. The publisher marks the article as retracted and links a notice dated November 7, 2023, with DOI 10.1038/s41586-023-06774-2. The notice identifies the original article, allowing us to verify that the warning concerns this publication.
The notice describes concerns about the materials' provenance, measurements, and data processing, together with unresolved concerns about resistance-data reliability. That is enough to reject using the paper as confirmation of the claimed superconductivity. Marking the bibliography entry as retracted would still leave the scientific claim unsupported.
The second record is Bryant, Pozzati, and Elofsson's 2022 AlphaFold2 interaction-prediction paper, DOI 10.1038/s41467-022-28865-w. Its page links an Author Correction published March 24, 2022, DOI 10.1038/s41467-022-29480-5. The notice changes four pDockQ parameters in Methods and states that both HTML and PDF were corrected.
The DOI log below records each chain from the original to its notice. Keep both identifiers. Crossref's update guidance treats a correction or retraction notice as a separate document with its own DOI and metadata. Replacing the original DOI with the notice DOI while retaining the original title would identify the wrong publication.
| Original DOI | Publisher status and notice date | Notice DOI |
|---|---|---|
10.1038/s41586-023-05742-0 | Retracted November 7, 2023 | 10.1038/s41586-023-06774-2 |
10.1038/s41467-022-28865-w | Author Correction March 24, 2022 | 10.1038/s41467-022-29480-5 |
Decide which claim remains usable
For the correction case, the old parameter set was L=0.958, x0=160.11, k=1, and b=0.001. The corrected set is L=0.724, x0=152.611, k=0.052, and b=0.018. If your methods section copied the earlier coefficients, update that section and check any calculations that used them. Adding a correction reference without checking the implementation leaves the practical problem unresolved.
The correction changes the Methods parameters. Read the updated passage and check how that change affects your use of the method. Its scope gives you no basis to label the whole paper retracted or to assume that every result has been independently validated.
To discuss the lutetium-hydride retraction as a publication event, cite the retraction notice and state the status in the sentence. For a technical argument that previously relied on the retracted result, remove that reliance or find independently supported evidence. Do not convert the withdrawn result into a general statement that the broader scientific question has been settled.
For a manuscript using the corrected pDockQ method, cite the research paper for the method and the correction when explaining the changed coefficients. The three sample references below keep the retraction notice, the original method paper, and its correction separate. Their numbers are illustrative; assign your own numbers by first appearance in the manuscript.
Record the check and citation purpose
The sample entries use ordinary IEEE journal-reference fields. Preserve the notice's published title, including Author Correction or Retraction Note, so readers can identify what you cited. Check the destination's instructions for any additional retraction label on the original article. Never transfer the notice's date, page, or article number onto the original paper.
How to cite a correction or erratum in IEEE
These are separate journal records. The Nature notice entry uses its December issue date; the log records its November 7 online publication date.
[1] N. Dasenbrock-Gammon et al., “Retraction Note: Evidence of near-ambient superconductivity in a N-doped lutetium hydride,” Nature, vol. 624, no. 7991, p. 460, Dec. 2023, doi: 10.1038/s41586-023-06774-2.
[2] P. Bryant, G. Pozzati, and A. Elofsson, “Improved prediction of protein-protein interactions using AlphaFold2,” Nat. Commun., vol. 13, no. 1, Mar. 2022, Art. no. 1265, doi: 10.1038/s41467-022-28865-w.
[3] P. Bryant, G. Pozzati, and A. Elofsson, “Author Correction: Improved prediction of protein-protein interactions using AlphaFold2,” Nat. Commun., vol. 13, no. 1, Mar. 2022, Art. no. 1694, doi: 10.1038/s41467-022-29480-5.We checked the two DOI chains on November 1, 2026. The downloadable log records the original DOI, observed status, notice date, notice DOI, publisher links, and intended use. Download it as CSV or JSON and adapt the columns for your own bibliography. Its coverage is limited to the two records shown.
Download the checked DOI log as CSV · Download JSON
Keep the notice's publication date separate from your checking date. Revisit the publisher page before submission, especially if a source previously carried an expression of concern. That label records an unresolved editorial concern and calls for reading the accompanying notice; it should not be silently treated as either a correction or a final retraction.
A move from Early Access to an assigned issue is a different update. Our Early Access guide explains which bibliographic fields change in that situation. A retraction or correction requires returning to the sentence supported by the source, in addition to updating the reference manager.
After editing, refresh the linked bibliography and compare the affected in-text numbers with their records. Check a static list manually. If a notice or corrected passage remains inaccessible, leave that specific item unresolved so your coauthor can address it before signing off the reference list.