How to Find a Paper When the Journal Website Has Disappeared

Almost every week, someone arrives at a library help desk holding a citation that leads nowhere. The reference looks complete — authors, title, volume, issue, maybe even a link — but the link returns an error, the journal’s domain now sells something else, or the “archive” page is empty. This happens more often than most students expect, and it is especially common with small open-access journals founded in the 2000s and 2010s, many of which hosted their PDFs on a single self-managed website.

The good news: a paper that was published once usually left traces. The method below is the one we use at the desk. Work through it in order; most searches end by step 4.

Why you may be reading this page. This domain once hosted issue files for third-party journals. Those files are no longer here, and this library is not the original publisher. If you followed an old link looking for a specific article, the steps below are the fastest honest way to find it. See our archive notice for what we know about the old addresses.

Before you start: write down what you actually have

Librarians call this “verifying the citation”. Take two minutes and note, separately:

Exact title of the article (copy it; don’t retype from memory).
All author names, including initials.
Journal name — full name and any abbreviation you have seen.
Volume, issue, year, month, page range.
Any identifier: DOI, ISSN, article number, or the old URL itself.

Small differences matter. Journals with near-identical names are common (“International Journal of Advanced…”, “International Journal of Advance…”), and a single missing word can send you to the wrong title. When in doubt, the ISSN is the most reliable key, because it identifies one serial publication regardless of how its name is written.

Step 1 — Search the exact title in quotation marks

Start broad, but precise:

Google Scholar — paste the full title inside quotation marks. Then look at the small “All N versions” link under a result: it often reveals copies on university repositories, author pages or preprint servers that the main link does not show.
A general web search with the title in quotes plus filetype:pdf.
Semantic Scholar and OpenAlex — both index large numbers of papers from smaller journals and show linked versions.

If you find the paper here, check that it is the published version (look for the journal name, volume and pages on the PDF) before you cite it. A preprint or a thesis chapter with the same title is a different document.

Step 2 — Look for a DOI, then resolve it

A DOI (Digital Object Identifier) is designed to outlive websites: the publisher can move the article and update where the DOI points.

Search the title at Crossref Search. If the journal registered DOIs, the record will appear with the DOI and publisher metadata.
Resolve any DOI you find at https://doi.org/ followed by the DOI.

A caution from experience: many small journals of the early 2010s did not register DOIs with Crossref, or printed “DOI-like” numbers that were never registered. If doi.org says the DOI cannot be found, the number was probably not a real registered DOI. That is useful information in itself — note it and move on.

Step 3 — Try the Internet Archive, properly

Most people try the Wayback Machine once, see a blank result, and give up. Use it more carefully:

Go to web.archive.org and paste the exact old URL, including the file name (for example, a path ending in .pdf).
If there’s no capture, paste the folder above it (the URL without the file name) and add an asterisk: example.com/path/*. The Wayback Machine will list every archived address under that folder — sometimes the PDF was captured under a slightly different URL.
Try the journal’s old issue or archive page as well. Even if the PDF was never captured, the issue page may list the paper’s full citation, authors and sometimes the author email.
Search Internet Archive Scholar, a separate index focused on research papers, including many from journals that have gone offline.

Step 4 — Search open-access aggregators and repositories

Authors often deposit their own copies. Search the exact title in:

CORE — aggregates full texts from repositories and journals worldwide.
BASE (Bielefeld Academic Search Engine) — a large index of institutional repositories.
Zenodo — many small journals and individual authors now deposit papers and whole issues here.
Your own and the author’s institutional repository, if the author’s affiliation is known.
The Unpaywall browser extension, which flags legal free copies when you open a paper’s landing page.

Step 5 — Go to the source: the journal, then the author

The journal. Journals sometimes change domain rather than close. Look up the ISSN at the ISSN Portal: the record may list the current URL, the publisher, and whether the title has ceased or changed name. Treat what you find on a new website with care — confirm that the new site is the same publication (same ISSN, same editorial history) and not an unrelated journal with a similar name.

Preservation services. The Keepers Registry shows whether a journal’s content was deposited with a digital preservation service. For small independent journals the answer is often “no”, but when it is “yes”, it tells you where the content survives.

The author. This is the most underrated step and it very often works. Find the author through their ORCID record, their current university staff page, or academic profile sites, and write a short, polite email:

Dear Dr. …, I am a [student/researcher] at [institution] working on [topic]. I am trying to read your article “[title]”, published in [journal], [year]. The journal’s website is no longer available. Would you be willing to share a copy? Thank you for your time.

Authors usually hold the right to share their own work for research purposes, and most are glad someone wants to read it.

Step 6 — Ask a library

If none of the above works, ask your institution’s library for an interlibrary loan / document delivery request. Librarians have access to union catalogues and networks that no public search engine covers. Bring the verified citation from the first step — it makes their search faster.

When you still cannot find it

Sometimes a paper really is lost to the open web. In that case:

Do not cite what you have not read. Citing a paper you only know from someone else’s reference list is a recognised source of propagated errors. If you must refer to it, say so, e.g. “as cited in …”, following your style guide’s rules for secondary citation.
Look for the same work elsewhere. Authors often presented the same study at a conference or developed it into a thesis. A search on the authors’ names plus two or three key terms from the title often finds the later, better-preserved version.
Consider whether you need it. If a key claim depends on a paper nobody can access, that is worth mentioning to your supervisor — it may be a sign to find stronger, verifiable evidence.

A note on why this happens

Journals are only as permanent as the arrangements behind them. Large publishers deposit content with preservation services and register DOIs; a small journal run from one website may do neither. When the website lapses, the only surviving copies are the ones someone else made. This is why librarians encourage authors to deposit their work in an institutional repository or a service such as Zenodo, and why, when you choose where to publish, long-term preservation is a question worth asking — see our guide on how to evaluate a journal.

Quick checklist

Citation verified (title, authors, journal, ISSN, year)
Exact-title search: Google Scholar (“All versions”), Semantic Scholar, OpenAlex
DOI checked at Crossref and doi.org
Old URL, folder (/*) and issue page checked in the Wayback Machine; Internet Archive Scholar searched
CORE, BASE, Zenodo, repositories, Unpaywall
ISSN Portal and Keepers Registry checked
Author contacted
Library request made

Found an error or a tool that should be here? See our corrections policy.