Crossref Scholarly Metadata — Papers, DOIs & Citations

Search 150M+ scholarly works from Crossref — the authoritative DOI registry behind academic publishing — and get clean, structured JSON. Filter by topic, year, and work type; pull DOIs, titles, authors, journals, publishers, and citation counts. Built for research, bibliometrics, publishing analytics, and AI/RAG over scientific literature. No API key.

Keywords: Crossref API, DOI lookup, scholarly metadata, citation data, bibliometrics, academic papers API, research data, publishing analytics.

---

Why this actor

Crossref is the registry that mints and resolves DOIs for most of the world's scholarly literature — journals, conference papers, books, datasets, preprints. Its REST API is excellent but returns deeply nested records and uses deep cursor pagination. This actor flattens it into one tidy record per work and handles paging for you.

It pairs with our OpenAlex actor: Crossref is the publisher-registry view (DOIs, journals, citation counts); OpenAlex adds institutions and concepts. Use both to own research-data workflows.

• Full-text-ish search across titles, authors, and metadata.
• Filters — publication-year range and work type (journal-article, book-chapter, preprint…).
• Clean output — DOI, title, year, authors, journal, publisher, citation count, ISSN, URL.
• Deep pagination — pull a handful or tens of thousands.
• No API key — official Crossref API; a contact email joins the faster "polite pool."

---

What you can build

• Literature reviews — assemble a filtered corpus on a topic with DOIs and citation counts.
• Bibliometrics — quantify output by year, journal, or publisher.
• Reference enrichment — resolve titles/DOIs to clean metadata for a citation manager or knowledge base.
• Publishing analytics — track a journal's or publisher's output and impact.
• AI / RAG — build a scholarly metadata layer for a research assistant.

---

Input

Field	Type	Default	Description
query	string	large language models	Search across titles/authors/metadata.
fromYear / toYear	integer	—	Publication-year bounds.
workType	enum	—	journal-article, book-chapter, proceedings-article, posted-content, etc.
maxResults	integer	200	Max works.
mailto	string	—	Contact email for the polite pool (recommended).

Example

{ "query": "perovskite solar cells", "fromYear": 2022, "workType": "journal-article", "maxResults": 1000, "mailto": "you@org.com" }

---

Output

{
  "doi": "10.1162/coli_a_00558",
  "title": "Large Language Models Are Biased Because They Are Large Language Models",
  "type": "journal-article",
  "year": 2025,
  "authors": ["Jane Doe", "John Smith"],
  "journal": "Computational Linguistics",
  "publisher": "MIT Press",
  "cited_by_count": 19,
  "issn": ["0891-2017"],
  "url": "https://doi.org/10.1162/coli_a_00558",
  "scraped_at": "2026-06-12T00:00:00.000Z"
}

A final {"_type":"summary"} record reports total matches and how many were returned.

---

Pricing

First 25 works free per account, then $0.002 per work ($2 per 1,000). Zero charge on empty searches. No monthly rental, no API key.

---

FAQ

Do I need a key? No. Crossref is open; a contact email just gets you the faster polite pool.

How many works? Over 150 million across journals, books, conferences, datasets, and preprints.

Difference from OpenAlex? Crossref is the DOI/publisher registry (citation counts, journals); OpenAlex adds institutions, concepts, and open-access status. They complement each other.

Can I filter by year and type? Yes — fromYear/toYear and workType.