In February 2026, Chemical & Engineering News (C&EN) reported a new twist on an old integrity problem: citation cartels that invent fake author names rather than recruiting real ones. Research-integrity investigator Anna Abalkina, a social scientist at the Free University of Berlin, shared a spreadsheet of roughly 20 chemistry papers she believes were published for no purpose other than to sell citations to paying clients. The papers carried fabricated authorship, and several were placed in journals from Taylor & Francis, Springer Nature, and Elsevier, each of which told C&EN it was investigating.
What the investigation found
According to C&EN’s reporting, the operation charged researchers a per-reference fee (reported at roughly $5-10 per citation) to have their published work cited in these manufactured papers. Rather than recruiting a real co-author, the operators simply invented one, along with a fabricated institutional affiliation, sidestepping several of the usual failure points: no real person has to consent to authorship, no real person is exposed if the paper is later retracted, and no genuine researcher’s name accumulates a track record of retracted work that editors or database curators could flag.
How this differs from a classic citation cartel
CASRAI’s citation cartel entry describes the traditional pattern: real authors, editors, or journals that agree, explicitly or tacitly, to cite one another’s work more than the underlying scholarship justifies. That pattern is detectable through bibliometric network analysis, because it eventually shows up as anomalously dense citation clustering among identifiable, real accounts. A cartel built on invented authors is structurally different. There is no real account to cluster around, no ORCID profile to check, no institutional email to verify, and no career incentive tying a specific person to the fraud. The fake-author variant is a citation cartel in effect — the goal is still artificially inflating a paying client’s citation count — but it uses fabricated identity rather than a real, ongoing citation relationship as the delivery mechanism, which is precisely what makes it harder for existing bibliometric-anomaly detection to catch.
Exploiting article-processing-charge waivers
Per C&EN, several of the flagged papers listed fabricated authors as based in countries including Senegal, Kenya, Somalia, Namibia, Saudi Arabia, Iraq, and Egypt — jurisdictions where legitimate researchers commonly qualify for article-processing-charge (APC) waivers at open-access journals. Claiming one of these affiliations for a fake author accomplishes two things at once: it supplies a name and institution that isn’t easily cross-checked against a real, easily contactable person, and it removes the cost barrier that would otherwise apply to publishing junk content purely to carry paid-for citations. The tactic effectively launders both the authorship problem and the cost problem through a legitimate piece of publishing-equity infrastructure that was built to lower barriers for researchers in under-resourced systems.
Why a later retraction does not undo the benefit
One structural weakness C&EN’s reporting surfaces is that citations from a paper still typically count toward metrics like the h-index even after that paper is retracted, unless a citing database specifically reprocesses and discounts them. That means a client who has already paid for citations retains most of the bibliometric benefit even if the manufactured paper carrying those citations is eventually pulled — the harm to metrics integrity happens at the moment of citation, not at the moment of retraction, and by the time a paper mill-style operation is caught, the citation counts it inflated have often already done their work in tenure files, grant applications, and journal rankings.
Publisher and platform response
Representatives of Taylor & Francis, Springer Nature, and Elsevier all confirmed to C&EN that they were investigating the flagged papers at the time of reporting. As with prior large-scale paper-mill discoveries (see CASRAI’s overview of the paper mill and tortured-phrases landscape), the practical response cycle tends to run: independent sleuth identifies a pattern and shares evidence → publisher confirms an internal investigation → a subset of flagged papers are eventually corrected, retracted, or the journal is reviewed for indexing status. That cycle takes time, and fabricated-author schemes are specifically designed to survive longer within it, since there is no real individual’s reputation motivating a fast correction.
What this means for research-integrity practice
For journal editors, research-integrity officers, and anyone relying on citation counts as an evaluation signal, the practical implication is that author verification needs to extend past checking whether a name and affiliation exist on paper. Useful checks include confirming a persistent identifier such as an ORCID iD with a plausible publication history predating the paper in question, checking whether an institutional affiliation resolves to a real, contactable department, and treating an unusually narrow citation pattern — a paper that cites one client’s work disproportionately relative to its actual topic — as a red flag regardless of whether the authors listed appear individually legitimate. See CASRAI’s research integrity officer (RIO) entry and paper mill entry for related detection concepts, and coercive citation for a related but distinct pressure-based citation-inflation tactic where the demand for extra citations comes openly from an editor rather than being laundered through a fake author.
Frequently asked questions
Is this the same thing as a paper mill?
It overlaps but isn’t identical. A paper mill sells authorship slots on fabricated manuscripts, typically to real researchers who need a publication credit. The scheme C&EN reported is narrower: the manuscripts exist mainly to carry citations for sale, and the authorship itself is fabricated rather than sold to a real client.
Why chemistry journals specifically?
C&EN’s reporting focused on chemistry titles because that is the beat the investigation originated on and where Abalkina’s sample was drawn from; the underlying tactic (fabricated authorship plus waiver misuse) is not chemistry-specific and could in principle recur in any discipline with open-access APC waivers.
Can journals detect fabricated authors before publication?
Not reliably with current standard workflows. Verifying an ORCID iD’s publication history, confirming an institutional affiliation independently rather than trusting the manuscript’s own contact details, and flagging papers with narrow, client-concentrated citation patterns are the checks integrity researchers currently point to, but none of them are yet mandatory or automated at most journals.
