Two purchase orders can both say “terbinafine” and still describe two different materials. One is the free base, the other the hydrochloride. They carry different CAS numbers, different molecular weights, different solubility, and, for anyone building a formulation, different behaviour. On an API specification the word in brackets after the drug name is not a suffix. It is the identity of what is in the drum. Below is where that word changes the purchase, using substances from our own pharmaceutical list.

The word after the name is the product

Few small-molecule drugs are traded as the neutral molecule. Most are weak acids or weak bases, and the neutral form is often an oil, a gum or a barely soluble solid that no tabletting line can handle. Terbinafine hydrochloride (CAS 78628-80-5) is the form that actually moves: a white powder melting at 204–208 °C with good solubility in methanol. The base is a synthetic intermediate; the salt is the saleable product.

Which counter-ion is chosen is a deliberate decision, and each one buys something different. The hydrochloride is the cheapest and the most common. Ropivacaine hydrochloride (CAS 132112-35-7, melting point 267–269 °C, specific rotation −7.28° in water) shows the standard pattern: the salt is a crystalline, water-soluble solid, and the rotation figure is there to confirm which enantiomer was supplied rather than the racemate.

The methanesulfonate route appears where a base will not crystallise usefully as the hydrochloride. Osimertinib mesylate (CAS 1421373-66-1, decomposing above 232 °C) is a case in point: methanesulfonic acid is strong, non-nucleophilic and gives a salt that crystallises well and dissolves far more readily than the free base, which is what lets an oral tablet dose reliably.

Other counter-ions appear where the standard ones fail. Cisatracurium besylate (CAS 96946-42-8) uses benzenesulfonic acid, a large and well-tolerated organic acid, to stabilise a molecule that would otherwise degrade by Hofmann elimination. Tiamulin fumarate (CAS 55297-96-6) pairs a veterinary diterpene base with fumaric acid, a diacid that can form more than one stoichiometry, which is why the molar ratio belongs on the specification and not in a footnote. Apramycin sulfate (CAS 41194-16-5) takes the aminoglycoside route, where the sulfate of a polycationic sugar is the water-soluble grade and the free base is not the form anybody injects.

Two further patterns are worth naming. A counter-ion can be the active part of the medicine: strontium ranelate (CAS 135459-87-9) works because the strontium cation is delivered, and bacitracin zinc (CAS 1405-89-6) salts a peptide with zinc largely for stability in a topical base. A counter-ion can also be chosen for tolerance rather than crystallinity: docusate sodium (CAS 577-11-7, C₂₀H₃₇NaO₇S, MW 444.56) is a sodium sulfosuccinate whose wetting behaviour comes precisely from being the dissolved sodium salt. And a salt can simply be a difficult solid: etamsylate (CAS 2624-44-4) is the diethylammonium salt of a dihydroxybenzenesulfonic acid, very soluble in water and openly listed as hygroscopic.

The parenthetical can carry an entire structure. Gadolinium DOTA meglumine (CAS 92943-93-6) is the clearest example: the active species is a gadolinium ion locked inside a macrocyclic cage, and meglumine is the counter-ion that turns the resulting anionic complex into a water-soluble injectable.

Hydrates: water is part of the molecule

A hydrate is not a damp powder. The water sits in the crystal lattice at fixed stoichiometry, and it changes molecular weight, density and dissolution rate. The prefix states the ratio, and that ratio is a specification rather than a description.

The comparison that makes the point sits on our own list. Dasatinib monohydrate (CAS 863127-77-9) and dasatinib (CAS 302962-49-8, melting point 275–286 °C, pKa 10.94) are separate trade items and not interchangeable. The monohydrate is the less hygroscopic, more stable form and the one a solid dosage form is built around; the anhydrous material takes up water from the air and converts. Buying the cheaper CAS and discovering that loss on drying has moved between batches is how blend uniformity fails.

Multiply the water where the stoichiometry says so. Pemetrexed disodium hemipentahydrate (CAS 357166-30-4) carries two sodium ions and two and a half molecules of water per molecule of drug; “hemipentahydrate” is 2.5, not a rounding. The same discipline runs across the list: methylene blue trihydrate (CAS 7220-79-3) is three waters, sodium thiosulfate pentahydrate (CAS 10102-17-7) is five, and calcium ascorbate dihydrate (CAS 5743-28-2, C₁₂H₁₄CaO₁₂·2H₂O, MW 426.35) is two, a figure its own sheet corroborates with a loss on drying capped at 11% and a calcium content of 9–10%.

Hydrated and anhydrous grades are also distinct commodities for simple acids. Citric acid monohydrate (CAS 5949-29-1) and the anhydrous grade are two CAS numbers with different solubilities and different caking behaviour, and substituting one for the other shifts the water balance of the whole batch.

Not every bracketed name is a salt

Platinum cytotoxics sit outside the salt framework. Carboplatin (CAS 41575-94-4) and oxaliplatin (CAS 61825-94-3) are neutral coordination compounds: a platinum(II) centre with two amine ligands and one bidentate leaving group. In carboplatin that leaving group is cyclobutane-1,1-dicarboxylate, which hydrolyses more slowly than the chlorides of cisplatin, and it is that slower hydrolysis which drives both the lower nephrotoxicity and the higher water solubility. Oxaliplatin swaps the ammines for a (1R,2R)-cyclohexane-1,2-diamine carrier, and its specification controls the stereochemistry with an optical rotation of +74.5 to +78.0°. Dicycloplatin (CAS 287402-09-9) belongs to the same family. For these materials there is no counter-ion to name, because there is no ionic bond — which is exactly why a request for “the hydrochloride” is meaningless here.

Same salt, different crystal

Once the salt and the hydration state are fixed, the solid can still exist in more than one crystal form, and form is a separate row on the specification for good reason. Bortezomib (CAS 179324-69-7, pKa 9.66) illustrates why: the substance is a boronic acid, not a salt, and it is hygroscopic and moisture-sensitive, which is why the first marketed presentation was a lyophilised, mannitol-buffered powder rather than a simple vial of the active. The general rule holds for any polymorphic material: the name on the label identifies the molecule, not the lattice, and only the specification can settle that.

The numbers to put on an enquiry

• The salt form and its counter-ion, with the CAS of that exact salt rather than of the parent base.
• The hydration state, and the loss on drying that goes with it. For a dihydrate or a pentahydrate the water is stoichiometric, so the assay and the LOD have to agree.
• The basis of the assay — as is, dried or anhydrous — and, where dose depends on it, the free-base equivalent.
• The crystal form, with the method used to establish it. XRPD is the usual answer; a melting point alone will not separate two polymorphs.
• Water content and residual solvents separately, since a true hydrate and a wet powder can report the same total volatiles and behave nothing alike.

A specification sheet is a commercial document as much as a technical one. The parenthetical is the first thing worth reading and the easiest to skip.