Same molecule. A different crystal.

Ritonavir’s Form II packed the same molecule into a more stable, less soluble crystal. Explore the difference between a chance first appearance and growth from an existing seed.

Teaching model
Illustrative polymorph lattice beside sourced solubility comparisonOne molecule, two packing arrangementsSchematic cells, not a molecular structureteal: Form I orange: Form IIHigh push starts more Form I cells (toy).Press Run a new draw to begin.Sourced solubility comparisonEthanol · 25°CForm IForm II38.5% of Form IForm I / Form II ≈ 2.6Toy nucleation & growth at left. Solvent-specific literature ratios at right.

Enlarge for bigger labels, then swipe sideways →

Chance of a new nucleus (toy)2.5 % per draw
Existing seedAbsent
Form I / Form II solubility2.6 times
Form II relative solubility38.5 % of Form I

Press Run a new draw. Repeat it to see chance first appearance, then try a seed. The unseeded chance is 2.5% in invented teaching units. The solubility comparison is 2.6 times in ethanol at 25°C for Form I relative to Form II. It does not predict a capsule’s dissolved fraction.

What this model assumes

This is a stochastic teaching toy, not a ritonavir process model. P(at least one new nucleus) = 1 − exp(−rate × exposure), using an invented Form II rate of 0.025 and invented exposure units. High push starts more schematic Form I cells; it does not change the Form II probability or fit a real nucleation rate. No units represent years, batch size, or an actual plant probability. Run a new draw samples a uniform random number; a supplied seed bypasses first nucleation. The growth slider deterministically replaces schematic Form I cells with Form II after its appearance, showing a possible solvent-mediated conversion, not measured growth rates or phase fractions. The initial cell population is a qualitative packing schematic, not measured nucleation kinetics. Solvent and temperature influence real nucleation, growth and transformation, but here the solvent selector changes only the separate literature solubility comparison. Wang 2021 reports Form I/Form II ≈2.6 in ethanol at 25°C and 3.9–5.7 in ethanol/water at 5°C, a composition-dependent range. No absolute dose, dissolution, bioavailability, clinical advice or capsule concentration is calculated. Degradation products, solvent composition, temperature cycling and real kinetic competition are omitted.

A form that is slow to appear can spread quickly once its seeds are present.

Ritonavir launched in 1996. In 1998, a previously unseen, less soluble crystal form caused manufacturing and formulation problems. Form I and Form II contain the same molecule in different conformations and crystal arrangements.

Form I can nucleate preferentially at high supersaturation. Form II has a more stable crystal structure but can have a long induction time before it first appears. Once a seed exists, nucleation no longer has to start from nothing.

The lattice below is a teaching schematic, not a molecular structure or a crystallization prediction. A random draw illustrates the first-appearance idea. Its rate, exposure units and growth rules are invented. Increasing exposure increases the chance in this toy model; it does not forecast years of real plant operation.

Wang and colleagues report Form I about 2.6 times as soluble as Form II in ethanol at 25°C. In ethanol/water at 5°C the reported ratio is 3.9–5.7, depending on composition. These are solvent-specific solubility comparisons, not percentages of a capsule absorbed or dissolved in the body.

The original facilities struggled to return to Form I after Form II appeared. Seeds can persist and promote further growth. This does not mean every batch in every solvent must always become Form II; temperature, solvent, supersaturation, impurities and processing history all matter.

After a Form II speck appears, why might returning to the old Form I process fail?

Taking a real process
beyond the lab?

Jon and I are building RARE Labs toward an autonomous pilot plant builder using AI and robotics. Join the waitlist for updates, and tell us what you’re trying to scale.

Join the RARE Labs waitlist ↗

Go a little deeper.