Disappearing Polymorphs: The Crystals That Vanish — and Whether They’re Real

For two years, Abbott Laboratories made its HIV drug Norvir without incident. Then, in 1998, the crystal at its core stopped forming — replaced by a denser twin that no lab it had touched could seem to shake. The mistake cost an estimated quarter of a billion dollars, and it wasn’t a fluke of one factory.

Bottom Line Up Front

What the evidence supports: Disappearing polymorphs are a real, peer-reviewed phenomenon. A molecule’s metastable crystal form can be displaced by a more stable one that, once it nucleates, out-competes the original in every subsequent batch — occasionally spreading across a whole manufacturing site through trace seed crystals.

What’s overstated or unsupported: The idea that a lost form is gone forever, or that the effect is mystical or “contagious” in any spooky sense. The driver is ordinary nucleation thermodynamics plus inadvertent seeding — and forms once declared lost have since been recovered in the lab.

⚕️ LyfeiQ Score: 8/10 — a genuine, consequential phenomenon that the popular retelling routinely exaggerates.

What Actually Happens When a Crystal “Disappears”?

A disappearing polymorph is not a molecule that vanishes — it’s a crystal packing arrangement that becomes suddenly, stubbornly unreproducible. Polymorphs share an identical chemical formula but pack their atoms into different lattices. Diamond and graphite are both pure carbon; they behave nothing alike. The same molecule can crystallize into forms that differ in solubility, melting point, and stability. When a lower-energy form appears for the first time, the higher-energy one it displaces can become nearly impossible to grow again under the old recipe.

The term entered the vocabulary through a 1995 paper by Jack Dunitz and Joel Bernstein in Accounts of Chemical Research, built partly on cases from their own labs and the ranitidine hydrochloride patent fights. Their point was not that chemistry had broken. It was that a metastable form lives on borrowed time, and once a more stable competitor nucleates, thermodynamics quietly picks the winner.

Ritonavir is the case everyone cites. Abbott launched the protease inhibitor as Norvir in 1996 knowing of only one crystal form. Two years later, a second, denser form — roughly half as soluble — began precipitating inside the capsules, which then failed dissolution testing. Follow-up work showed the two forms differ in molecular conformation and hydrogen bonding, not just packing. Once Form II existed it seemed to seed itself everywhere; chemists who had handled it were suspected of carrying microscopic crystals into clean rooms. Abbott pulled the capsule, reformulated, and lost an estimated $250 million.

The story kept going. A third form, Form III, was reported in 2022 — 24 years after the crisis — as a reminder that “all the forms” is a claim no chemist should make with confidence.

What Do Chemists and Drug Makers Actually Do About It?

The practical response has been to stop being surprised. Polymorph screening is now routine before a drug is filed: high-throughput crystallization across dozens of solvents, temperatures, cooling rates, and deliberate seeds, hunting for hidden forms while there’s still time to react. Regulators expect this work, and it is standard practice in modern development in a way it simply wasn’t in the mid-1990s.

Seed hygiene matters more than it sounds. A trace of a stable form can tip an entire batch, so labs that need a metastable form guard against contamination the way a bakery guards a sourdough starter — in reverse. And when a stubborn form refuses to appear, chemists now reach for mechanochemistry (ball milling) and melt or supercooled-liquid crystallization to force the issue. Those same tools, as it turns out, can bring “lost” forms back. If you work with crystalline solids, the honest operating assumption is that more forms exist than you’ve found — so screen like it.

Three Ways to Read the Phenomenon

The Solid-State Chemistry View

To crystallographers, nothing here breaks physics. Bernstein and Dunitz framed disappearing polymorphs as metastability giving way to a thermodynamically preferred form, with nucleation acting as a probabilistic bottleneck — rare until it isn’t, then dominant. More recent work from groups led by Aurora Cruz-Cabeza at Manchester and collaborators at Cambridge has dissected the crystal size, shape, and conformation factors that decide which form wins. The field’s consensus: puzzling in the specifics, not mysterious in principle.

The Pharmaceutical Industry and Regulatory View

For drug makers, a polymorph is a solubility problem with a legal shadow. Solubility governs how much of a pill actually reaches the bloodstream, so a denser, less soluble form can render a medicine inert without changing a single atom of its chemistry. That is why screening is now baked into development and why patent law leans on it: the paroxetine and ranitidine disputes both turned on whether a “new” crystal form was truly novel or an inevitable byproduct of the existing process. Crystal chemistry and courtroom argument have been tangled together for decades.

The Public and Science-Communication View

In the popular retelling, disappearing polymorphs edge toward horror. The go-to analogy is Kurt Vonnegut’s Ice-nine from Cat’s Cradle — a seed crystal that freezes the world on contact. Outlets and blogs have run the ritonavir story as a “contagious crystal” cautionary tale, and the facts mostly hold up. The “poisoned the whole planet” flourish does not. A useful dissent comes from inside chemistry itself: the dramatic “seeded across labs by human carriers” claims were plausible but rarely proven at the time, and some are better explained by coincidence plus a strong prior expectation of finding the new form.

Where Does the Science End and the Folklore Begin?

The honest answer to “are they real?” is yes — with an asterisk. The phenomenon is documented and expensive; the ritonavir episode alone settles the question. What’s overstated is permanence and spookiness. The 2015 review Disappearing Polymorphs Revisited argued that much of the mystique reflects a misunderstanding of nucleation and a lot of undocumented seeding rather than anything supernatural. And in 2024, a team led by Cruz-Cabeza did what the folklore calls impossible: they recovered ritonavir’s disappeared Form I in a ball mill, showing that relative stability can be flipped by crystal size and shape. “Gone forever” turned out to mean “gone until someone changed the conditions.”

What’s Next for the Vanishing Crystal?

Prediction is starting to outrun disaster. Crystal structure prediction software is approaching the point where it can forecast which hidden forms a molecule might adopt before one ambushes a factory. Mechanochemistry offers a way to access and interconvert forms on demand rather than waiting for nature to cooperate. And machine-learning models of nucleation may eventually predict which polymorph wins under a given set of conditions. Put together, they point toward a future where a disappearing polymorph is something you anticipate on a screen, not something you discover when the capsules stop working.

The Verdict: Disappearing Polymorphs’ LyfeiQ

Disappearing polymorphs are real, well-evidenced, and far less magical than their reputation.

Credibility Rating: 8/10

Evidentiary Basis: 9/10 — decades of peer-reviewed cases across multiple compounds

Mechanistic Understanding: 7/10 — nucleation and seeding explain most cases; some specifics remain open

Real-World Impact: 8/10 — genuine drug recalls, reformulations, and patent battles

Risk-Benefit Ratio: Neutral — a real manufacturing hazard, but increasingly detectable and even reversible

Scientific Consensus: widely accepted as real; the “gone forever / contagious” framing is considered overstated

👉 Who should care: anyone in pharma, materials science, or crystallography — plus curious readers who like a true story stranger than the myth built on top of it.

👉 Who can skip the worry: consumers. Modern polymorph screening makes a ritonavir-style surprise far less likely in the medicines on shelves today.

⚕️ LyfeiQ Score: 8/10 — a real phenomenon worth understanding. Treat the “vanished forever” version with a healthy dose of skepticism.

Citations

1. Dunitz JD, Bernstein J. Disappearing Polymorphs. Accounts of Chemical Research, 1995, 28(4), 193–200. pubs.acs.org
2. Chemburkar SR, Bauer J, Deming K, et al. Dealing with the Impact of Ritonavir Polymorphs on the Late Stages of Bulk Drug Process Development. Organic Process Research & Development, 2000, 4(5), 413–417. pubs.acs.org
3. Bauer J, Spanton S, Henry R, et al. Ritonavir: An Extraordinary Example of Conformational Polymorphism. Pharmaceutical Research, 2001, 18(6), 859–866. cmbe.engr.uga.edu
4. Bučar D-K, Lancaster RW, Bernstein J. Disappearing Polymorphs Revisited. Angewandte Chemie International Edition, 2015, 54(24), 6972–6993. doi.org
5. Ritonavir Form III: A New Polymorph After 24 Years. Journal of Pharmaceutical Sciences, 2022. jpharmsci.org
6. Sacchi P, Wright SE, Cruz-Cabeza AJ, et al. Crystal size, shape, and conformational changes drive both the disappearance and reappearance of ritonavir polymorphs in the mill. PNAS, 2024. doi.org
7. The curious case of disappearing polymorphs. Chemistry World, 2026. chemistryworld.com

Disclaimer: This content includes personal opinions and interpretations based on available sources and should not replace medical advice. This content includes interpretation of available research and should not replace medical advice. Although the data found in this blog and infographic has been produced and processed from sources believed to be reliable, no warranty expressed or implied can be made regarding the accuracy, completeness, legality or reliability of any such information. This disclaimer applies to any uses of the information whether isolated or aggregate uses thereof.