Substance identity under REACH

REACH regulates substances, not products or processes, and this is what makes it awkward for bio-based chemistry. If a fermentation route and a petrochemical route deliver the same substance, the newcomer joins the existing joint submission and relies on data already generated — an enormous saving, since the testing packages required above ten and one hundred tonnes per year include long-term ecotoxicity and repeated-dose toxicity studies. If it is a different substance, it is an unregistered substance and cannot be placed on the EU market at all until it has its own dossier. Nothing in the regulation cares that the molecule was made from sugar. Everything depends on whether identity can be demonstrated.

What identity actually consists of

Under Annex VI a substance is identified by more than its name and CAS number. The dossier must state the composition: the main constituent, each impurity and each intentionally added stabiliser, with concentration ranges and analytical evidence — spectra, chromatography — supporting them. ECHA’s guidance on identifying and naming substances sets the practical arithmetic. A well-defined mono-constituent substance is one whose main constituent is present at roughly eighty per cent or more; below that and above ten per cent the additional constituents make it a multi-constituent substance, described by all of them. Anything not deliberately present and below the threshold is an impurity, and impurities are part of the identity rather than a footnote to it.

That is the whole difficulty in one line. Bio-succinic acid and petrochemical succinic acid are the same molecule and are not automatically the same registered substance, because the two routes leave behind different things. A fermentation stream carries residual sugars and other organic acids, salts from neutralisation, protein and cell-derived residues; a chemical route carries catalyst metals, solvents and reaction by-products the bio-route never sees. Sameness is an argument about those tails, made on analytical data, and it is won or lost on purification rather than on chemistry.

When the impurity profile stops being cosmetic

Two consequences follow, and both are hazard-relevant rather than administrative. First, classification under CLP takes account of impurities: an impurity that is itself classified, present above the relevant generic concentration limit, can pull the classification of the substance as placed on the market, so two materials with identical main constituents can carry different labels. Second, an ecotoxicity or toxicity study is performed on a specific test material, and its validity for another manufacturer’s material depends on the compositions being close enough that the reported properties transfer. A registrant relying on shared data is asserting exactly that.

Where the molecule is not well defined at all

Many bio-based streams are not single molecules. Plant extracts, tall oil derivatives, fermentation broths used as such and hydrolysates fall into the UVCB category — substances of unknown or variable composition, complex reaction products or biological materials — which are identified by their source and their process precisely because their composition cannot be listed exhaustively. This has a hard consequence: for a UVCB, the process is part of the identity, so a different feedstock or a changed process step can mean a different substance rather than a variation of one.

Some naturally occurring materials are exempted from registration under Annex V, but the exemption is lost when the material has been chemically modified — which most useful conversion steps do.