InChI means International Chemical Identifier and describes a chemical structure in standardised machine-readable layers.
This topic is closely related to What is a SMILES structure? and What is PubChem and what is a PubChem CID?, which provide additional context for the interpretation and characterization of chemical materials.
Layers can encode formula, connectivity, hydrogen, charge, stereochemistry and isotopes, showing where structures agree or differ.
InChIKey is a compact fixed-length string derived from an InChI and is convenient for searching and indexing.
The complete structure cannot be reconstructed directly from the InChIKey alone; the key needs to link back to a structural record.
InChIKey is useful for comparing database records but says nothing about purity or batch quality of physical material.
Chemical identity involves more than a name. Formula, molar mass, salt or free form, stereochemistry and, where relevant, hydration or solvation should all refer to the same chemical entity. A mismatch between these fields is an important reason to recheck source data.
Databases describe chemical entities and collect reference information. A Certificate of Analysis, by contrast, generally reports measurements for a specific batch. Correct database identity therefore does not automatically prove the composition or purity of physical material.
A robust check compares name, CAS number, PubChem record, formula, mass and structural identifiers. When several independent fields consistently describe the same form, documentation becomes more reliable. Contradictions deserve additional attention.
A measurement gains meaning from the method, sample and conditions. HPLC, LC-MS, NMR, FTIR and solid-state techniques provide different kinds of information and should not be treated as interchangeable evidence.
A frequent error is combining data for closely related but non-identical forms, such as a free base with the molar mass of a hydrochloride or an unspecified structure with a stereospecific identifier. Cross-checking is particularly useful for finding such errors.
SDS, TDS and CoA serve different functions. An SDS primarily addresses safety, a TDS technical characteristics and a CoA batch-specific test results. The same type of value can therefore have a different context in different documents.
No database, identifier or analytical technique automatically describes every aspect of a material. Molecular structure, chemical purity, water content and solid-state form are different information levels. Strong assessment uses information appropriate to the actual question.
Most errors are avoided by not relying on one prominent number or one name. Check whether the different data are mutually consistent, distinguish structural identity from experimental batch data, and use complementary analysis when the question requires it.