What Is a DOI?
A DOI (Digital Object Identifier) is a permanent ID assigned to a scholarly work — a journal article, dataset, book chapter, or preprint. It looks like 10.1152/physrev.00032.2012, and unlike a URL it never breaks: https://doi.org/10.1152/physrev.00032.2012 resolves to the article even if the journal redesigns its entire website.
How a DOI works
The string has two parts: the prefix (10.1152) identifies the registrant — usually the publisher — and the suffix identifies the specific work. The publisher registers the DOI with the work's full metadata (authors, title, journal, volume, pages) in a central registry (CrossRef for most articles), and commits to keeping the link pointed at the current home of the work, forever.
That registry is why a DOI is citation gold: paste it into the generator and the citation is built from the publisher's own record — the exact author list, spelling and all.
Where to find a DOI
- Article landing page — near the title or in the "Cite this" panel.
- The PDF's first page — header or footer, often as a full https://doi.org link.
- Database records — PubMed, JSTOR, Scopus, and Google Scholar display it.
- Search CrossRef — search.crossref.org finds the DOI from a title when all else fails.
The same DOI, cited in two styles
APA 7 and MLA 9 want the DOI as a full https://doi.org/ link at the end. Chicago appends it after the page range. Harvard uses it where a URL would go. IEEE writes doi: 10.xxxx without the link. The generator applies each convention automatically.
Rasch, B., & Born, J. (2013). About sleep's role in memory. Physiological Reviews, 93(2), 681-766. https://doi.org/10.1152/physrev.00032.2012
B. Rasch and J. Born, "About sleep's role in memory," Physiological Reviews, vol. 93, no. 2, pp. 681-766, 2013. doi: 10.1152/physrev.00032.2012.
DOI vs. URL
Journal URLs rot — platforms migrate, sessions expire, proxies wrap links in university domains nobody else can open. The DOI is permanent, public, and proxy-free. The rule in every modern style: if the work has a DOI, cite the DOI, not the URL. Reserve URLs for works that genuinely have no DOI — most web pages, older articles, magazine pieces.
Keep every citation you make.
A free account saves your citation history and organizes sources into projects with notes and tags.
Create a free account →Find the sources you should be citing.
Premium searches 250 million scholarly works by topic, recommends citations for your claims, and flags statements in your writing that need support.
Go Premium — $5/monthFrequently asked questions
- Do all articles have DOIs?
- Most published since ~2000, plus many older ones registered retroactively. Books, chapters, datasets, and preprints increasingly have them too. No DOI? Cite with the journal URL or print form.
- Is a DOI the same as a PMID or arXiv ID?
- No — PMID is PubMed's internal ID and arXiv IDs are repository-specific. Both usually map to a DOI, which is the one citation styles want. (Modern arXiv papers have DOIs of the form 10.48550/arXiv.xxxx.)
- How do I cite a DOI in APA?
- End the reference with the full link: https://doi.org/10.xxxx — no period after it, no "Retrieved from". The generator formats this automatically.
- My DOI link asks for payment — is it still correct to cite?
- Yes. The DOI points to the version of record; whether the reader has access is separate. You may also legally find the accepted manuscript free via the author's page or their institution's repository.