The same molecule, packed two ways: why chocolate turns white and a drug once vanished from pharmacy shelves

Take a bar of chocolate that has sat too long in a warm cupboard and you will find a dull white film on it. Nothing has gone off, nothing has been contaminated, and the chemical composition is exactly what it was when it left the factory. The molecules have simply rearranged themselves into a different crystal packing. That is polymorphism, and the same phenomenon once forced a major pharmaceutical company to pull a life-saving HIV drug off the market with no chemical change to its formula at all.

Composition is not the whole story

A crystal is molecules arranged in a repeating three-dimensional pattern. The chemical formula tells you which molecules; it says nothing about how they are stacked, and for a solid, the stacking is where most of the useful properties live.

Change the arrangement and you change melting point, hardness, colour, density, electrical behaviour, how fast the solid dissolves, and how stable it is on a shelf — all without adding or removing a single atom. The textbook illustration is carbon: diamond and graphite are the same element in two arrangements, one the hardest natural material and the other soft enough to write with. For elements this is called allotropy; for compounds the same idea is polymorphism, and calcium carbonate is the everyday case — calcite and aragonite are the same CaCO₃, but shells and coral build with one and limestone with the other.

Some forms are stable, most are only comfortable

At a given temperature and pressure, exactly one polymorph is thermodynamically stable. Every other form is metastable: perfectly capable of existing for years, but sitting in an energy well it will eventually leave for the deeper one.

That "eventually" is the whole engineering problem. Crystallisation is governed by kinetics as much as thermodynamics, and Ostwald's rule of stages captures what usually happens — a solution tends to crystallise first into whichever form is easiest to nucleate, not the most stable one. So a manufacturing process can reliably produce a metastable form for years, and then, one day, produce something else.

Getting a chosen form on purpose is a genuine craft. Solvent, cooling rate, supersaturation, stirring, impurities and deliberate seeding all steer which form nucleates, which is why the same compound made by two routes in two plants can arrive as two different solids.

When the wrong form is a medical problem

For a drug taken by mouth, dissolution is usually the bottleneck. A tablet must dissolve in the gut before anything can be absorbed, and a more stable crystal form is by definition harder to break apart — lower energy means lower solubility and a slower dissolution rate. Two tablets with an identical amount of an identical molecule can therefore deliver very different amounts into the blood. The polymorph is not a chemical detail; it is a bioavailability decision.

The case every formulation scientist is taught is ritonavir, an HIV protease inhibitor launched in 1996. Two years later, batches began failing dissolution testing. A previously unknown, more stable and far less soluble crystal form — Form II — had appeared, and once it existed in a plant, the original form became extremely difficult to make again: trace seeds of the stable form kept converting new batches. The capsules were withdrawn and the product had to be reformulated, at enormous cost.

That pattern has a name — the disappearing polymorph. It is not mysticism about contaminated laboratories; it is seeding. Microscopic crystals of a more stable form spread through equipment, air and people, and thereafter every crystallisation has a template pointing at the form you did not want.

The opposite lever exists too. Where a stable form dissolves too slowly to work, formulators deliberately use a metastable or amorphous solid — one with no long-range order at all — to raise solubility, then spend the rest of the development programme stopping it from crystallising in the packet.

A chemical formula describes what a substance is made of. Its crystal form describes how it will behave. Manufacturing is largely the business of getting the second one right, repeatedly.

Chocolate is the same physics, in public

Cocoa butter crystallises in six forms, and only one of them makes good chocolate. Form V melts just below body temperature, which is why good chocolate is glossy and hard at room temperature and yields cleanly in the mouth, with the snap that confectioners test for.

Tempering — the careful heating, cooling and reheating cycle chocolatiers perform — exists solely to seed Form V and suppress the others. Untempered chocolate sets into softer, duller forms with no snap. And the white film on old chocolate is bloom: Form V slowly converting towards the more stable Form VI, with fat migrating to the surface and recrystallising there. It is entirely safe to eat, and it is the identical thermodynamics that took ritonavir off the shelves — a metastable form doing what metastable forms eventually do.

How you tell one form from another

Because composition is unchanged, ordinary chemical analysis cannot see the difference. The characterisation is structural.

Powder X-ray diffraction is the standard fingerprint: each packing arrangement scatters X-rays into its own distinct pattern of peaks, so a diffractogram identifies the form and can quantify a mixture of two. Differential scanning calorimetry finds it thermally, since each form melts at its own temperature and conversions show up as extra thermal events. Raman and infrared spectroscopy pick up the subtler shifts in molecular environment, and solid-state NMR resolves cases where two forms differ in ways diffraction handles poorly. Regulators expect this evidence: a drug filing must specify which form is being made, prove it is the form that was tested, and show it does not convert over the product's shelf life.

Why India writes different patent law about this

A new crystal form of a known drug is, in most jurisdictions, patentable subject matter — and since forms are frequently discovered late, polymorph patents became a standard way of extending commercial protection after the original molecule's patent expires.

India is deliberately different. Section 3(d) of the Patents Act bars the patenting of a new form of a known substance unless it differs in efficacy, and the Supreme Court applied it in 2013 to reject a patent on a specific crystalline form of imatinib mesylate. The reasoning was not that the crystal form is trivial — the judgment accepted the form had improved properties — but that improved processability or storage is not enhanced therapeutic efficacy. Whatever one's view of the outcome, it is a rare instance of national law turning on a point of solid-state chemistry, and it is why polymorph screening in India has a legal dimension that formulation textbooks written elsewhere do not cover.

Why it matters for students and researchers

Polymorphism is where crystallography stops being a structural curiosity and starts deciding whether a product works. It is also unusually well suited to research at modest scale: a screening study across solvents and cooling rates, careful diffraction work, and a stability study are within reach of a well-equipped university laboratory, and negative results — a form that will not form, a metastable phase that survives longer than expected — are genuinely publishable.

The field extends well past pharmaceuticals. Pigment polymorphs decide colour and lightfastness; energetic materials change sensitivity between forms; organic semiconductors and pharmaceutical co-crystals depend on packing for charge transport and solubility respectively; and computational crystal-structure prediction is getting good enough to warn a chemist about a form nobody has yet made.

That whole span — synthesis, characterisation, structure, properties and performance of crystalline materials — is the remit of the International Journal of Crystalline Materials (ISSN 3107-877X), a peer-reviewed journal launched in 2024. For chemistry, pharmacy and materials students, following that literature is how you learn to ask the question that saves a process: not "what is this compound", but "which form of it did we just make".

Frequently asked questions

What is polymorphism in chemistry?

Polymorphism is the ability of a single chemical compound to crystallise in more than one arrangement of its molecules. The composition stays identical while properties such as melting point, solubility, density and stability change.

Why does the crystal form of a drug matter?

Because the form controls how quickly the solid dissolves, and a drug must dissolve before it can be absorbed. Two tablets containing the same amount of the same molecule in different crystal forms can deliver very different amounts into the bloodstream.

What happened with ritonavir?

Two years after its 1996 launch, a previously unseen and far less soluble crystal form appeared and batches began failing dissolution tests. Because trace crystals of the more stable form seeded every later attempt, the original form became very hard to produce, and the capsules had to be withdrawn and reformulated.

Why does old chocolate turn white?

That is fat bloom. Cocoa butter set in the desirable Form V slowly converts towards the more stable Form VI, with fat migrating to the surface and recrystallising as a pale film. It is harmless to eat and is the same metastable-to-stable transition seen in pharmaceutical solids.

How are polymorphs identified?

Mainly by powder X-ray diffraction, since each packing produces a distinct peak pattern. Differential scanning calorimetry, Raman and infrared spectroscopy, and solid-state NMR are used alongside it to confirm the form and detect conversion.