D.4.1. What is a Substance dataset?

The Substance dataset is the central core of information in IUCLID. It contains all data related to a chemical substance like the chemical identity including the substance composition, information on manufacture, use and exposure, information on the classification and labelling, and all required and available endpoint study summaries. A substance to be handled in an IUCLID Substance dataset can be a chemical element, a molecule, or a combination of molecules derived from any manufacturing process or from a biological source, including any additive and any impurity deriving from the process used, but excluding any solvent which may be separated without affecting the stability of the substance or changing its composition (see also chapter B.4.2 Substance-related information).

A Substance dataset is the repository of data, which is used as a basis to create a registration or submission Dossier. When a Substance dataset is created for a given chemical substance, it is assigned to a Reference substance, which in turn is based on the EC Inventory or, if not listed, newly defined. The Reference substance is nothing else than a link to the identity of the chemical. On the user interface, it appears as a data entry field, which however is edited in a separate inventory of Reference substances (see also chapter D.11 Reference substance (Create and update Reference substance related information)).

The difference between the (submission) substance after which a Substance dataset is named and the assigned Reference substance is briefly explained based on the following examples of (i) a mono-constituent substance and (ii) a multi-constituent substance:

The second example implies that a Substance dataset is also used as repository for so-called unintentional mixtures. These include:

These kinds of mixtures are handled like any discrete substances. That is, a Reference substance is defined for the mixture and assigned to the Substance dataset created for that mixture.

Important

The IUCLID feature Mixture is designed for handling preparations, which according to the definition given in REACH are intentional mixtures of substances gained by blending of two or more constituents. The components retain their own chemical identity and properties and do not react. The composition of a preparation can be fully characterised. See chapter D.7 Mixture (create and update mixture related information).

A Substance dataset is characterised by the following data structure:

The data structure of a Substance dataset is illustrated by the following figures:

The relations between the database objects shown in the figures above are as follows: