Hundreds of academic journals that launched as free to publish in and free to read have ended up switching to a fee-charging model, according to a study that examined the fate of titles using the diamond open-access (OA) model.

The analysis, published as a preprint earlier this year¹, identified 440 titles that initially used the diamond OA model — in which publication costs are covered by sources such as institutions, non-profit organizations and libraries, rather than readers and authors — but changed to other publishing models between 2009 and 2026.

Of the 440 titles, 369 switched to gold OA, under which authors pay an article-processing charge (APC); 52 changed to hybrid OA, for which subscription-based titles make individual articles free to read for an extra fee; and 19 went entirely behind a paywall.

The APCs that the journals began implementing ranged from US$8 to $5,300 and were mostly dependent on the type of publisher: what the authors refer to as ‘professional publishers’ typically started charging higher APCs, whereas research organizations and society publishers mostly started charging lower amounts.

Over the past few years, journals have increasingly moved away from diamond OA, the analysis revealed.

The authors, led by Lisa Matthias, an information scientist at the Humboldt University of Berlin, found at least one ‘diamond fracture’ — the term they used to describe journals that moved away from diamond OA — every year since 2009. “But they do seem to be accelerating, with the majority occurring from 2020 onwards and 2025 marking the highest annual count in the data set,” the researchers report.

So, why are some journals moving away from a model that has been popular with both authors and readers?

Tracking diamond fractures

The diamond OA model has been used by academic journals for decades, but its popularity took off in the late 2010s, stemming partly from frustration that the high APCs of various journals were pricing out some researchers, especially those in lower-income regions. A 2021 analysis estimated that, at the time, there were as many as 29,000 active diamond OA journals worldwide².

There is no single source that records all diamond OA journals globally, so Matthias and her colleagues drew on several data sets to identify journals that had switched models.

This means that the study does not provide an exhaustive list of all the diamond OA journals that moved away from the model over the study period, nor could it calculate the proportion of journals that stuck with the diamond OA model. It also didn’t examine the reasons why individual journals moved away from the model.

But the findings do provide insights that could help to explain the increase in diamond fractures over the period.

For example, of the flipped journals, 88% stayed with the same publisher, which means that the trend can’t be explained by the journals simply being sold or acquired by other publishers, says Matthias.

She adds that the journals that moved away from the diamond model were “disproportionately older, more established and larger titles, and not struggling young ones”, which suggests that the decisions weren’t necessarily driven by financial pressures. More than half of journals in their data set (57%) ‘fractured’ after at least nine years of operation, the study found.

“It doesn’t look like a survival story,” says Matthias. “It seems like these journals and their reputation had become commercially worth converting to a different model.”

The study also found that although commercial publishers accounted for the majority (78%) of the diamond fractures, the journals in the data set as a whole did not markedly increase their output after changing model, “as might be expected from a pure profit motive”, the authors write.

Saray Córdoba, a retired information scientist at the University of Costa Rica in San José, suspects that some journals might have shifted away from the model “not because diamond open access is financially unsustainable”, but because they think that the more commercial models “will bring them more prestige”.

Uncertain fate

When looking at regional differences in the journals that moved away from diamond OA, the study authors noted a “relative scarcity” in Latin America — despite the dominance of diamond OA in the region. “To me, that hints that the diamond model can work and it can be protected,” says Matthias.

The authors note that support for diamond OA from institutions in Latin America, such as through the OA publishing platforms Redalyc and SciELO, as well as from local universities, could be enough to “stave off fractures” for existing journals. But they add that it’s unclear whether these organizations can support a fresh crop of diamond OA journals in the future.

According to Córdoba, diamond OA titles are a “source of pride” for Latin America’s researchers and public, which could also explain the low rate of diamond fractures in the region.