Prefixes — or combining forms?

With “endings” out of the way, shall we move on to “prefixes”?

In a number of IUPAC publications, the entities that are referred to as “prefixes” include

• Numerical prefixes [1], aka multiplicative prefixes [2] ‘di’, ‘tri’, ‘tetra’, etc. and ‘bis’, ‘tris’, ‘tetrakis’, etc.;
• Prefixes indicating atoms or groups, either substituents, e.g. ‘hydro’, ‘chloro’, ‘cyano’, or ligands, e.g. ‘hydrido’, ‘chlorido’, ‘cyanido’ [2];
• Prefixes ‘de’ and ‘an’ in subtractive nomenclature as well as their combinations with the names of atoms or groups, e.g. ‘dehydro’, ‘anhydro’, ‘demethyl’, ‘deoxy’, etc.;
• The ‘a’ prefixes for skeletal replacement and Hantzsch-Widman names, e.g. ‘aza’, ‘oxa’, ‘thia’, as well as their combinations with multiplicative prefixes, as in ‘dioxa’ [3];
• Geometrical and structural prefixes such as catena-, arachno-, quadro-, etc. [3];
• Configurational prefixes of inositols such as allo-, chiro-, cis-, epi-, muco-, myo-, neo- and scyllo- [4];
• Prefixes retro- and ‘apo’ in nomenclature of carotenoids [5];
• Configurational prefixes in nomenclature of carbohydrates [6];
• Prefix sn- (for stereospecifically numbered) in nomenclature of glycerol derivatives [7];
• Prefixes ‘abeo’, ‘cyclo’, ‘homo’, ‘nor’ and ‘seco’ in nomenclature of natural products [8];
• Prefix ‘poly’ and qualifiers such as branch-, net-, or star- in polymer names [9].

I like “qualifiers”. I also don’t mind saying “multiplicative prefix” or “configurational prefix” as long as we understand that they actally might be not prefixes, just like vegetarian sausages are not sausages and white chocolate is not chocolate.

Are there any genuine prefixes in this list? Yes! You don’t have to be a chemist or a linguist to recognise them. Prefix ‘de’ meaning “reversal, undoing, removing” is found in many English words of Latin origin, e.g. deactivate (cf. activate), decrease (cf. increase), descend (cf. ascend), destruction (cf. construction) and so on. While the prefix ‘an’, from the Ancient Greek ἀν, meaning “not, lacking, want of”, is found in words of Greek origin such as analgesia, anarchy or anorexia^*.

What about “multiplicative prefixes”? I hope you won’t mind me calling them “multipliers” for short. In my opinion, they are overqualified for jobs of true prefixes. Why? Because they do have independent meaning, viz. that of cardinal numbers. Since there is an infinite number of cardinal numbers, we can say that, in theory, there is an infinite number of multipliers, which means that they cannot be closed-class. In practice, chemical nomenclature employs only a small set of multipliers but this should not prevent us from seeing the bigger picture. Which is: multipliers are either content morphemes or combinations thereof. More specifically, they are combining forms, i.e. linguistic forms that usually are not used as separate words but appear as part of other words. Most of multipliers used in science are based on Greek or Latin roots.

Let’s have a look now at “prefixes” indicating atoms or groups. For example, the molecule CH₃Cl could be named substitutively chloromethane. According to IUPAC, this name contains the prefix ‘chloro’. Alternatively, we can give it a radicofunctional name methyl chloride. This binary name is grammatically a noun phrase, consisting of two nouns, that is, two content words. Each content word should have a root, so we have two roots: ‘meth’ and ‘chlor’. But the same two roots are found in the substitutive name chloromethane!

See what I mean? ‘Chloro’ is not a prefix because it contains a content morpheme. Since substituents can, in turn, contain other substituents, e.g. ‘4-chlorophenyl’ in 4-amino-3-(4-chlorophenyl)butanoic acid, there is an unlimited number of possible substituent names. Thus, substituent names are open-class. A substituent name could be used either as a word part (in substitutive names) or as a separate word (in radicofunctional names): cf. ‘methyl’ in 5,7-dichloro-2-methylquinolin-8-ol and “methyl chloride”. Plus, of course, it could be a name of a radical molecular entity: e.g. methyl on its own means CH₃•.

Similar story with ligand names. Red Book refers to them as “prefixes” while Bünzli-Trepp, for reasons unknown, refers to anionic ligand names ‘oxido’, ‘sulfido’, ‘selenido’ and ‘telludrido’ as “pseudoprefixes” [10]. Well they are not prefixes. You can see that with the names of neutral and cationic ligands. For example, ethylenediamine is a name of both a separate entity (a) and a ligand in tris(ethylenediamine)iron(3+) (b). So, ligand names are also open-class; each ligand name could be either a standalone word or a word part.

(a)	(b)

ethylenediamine (trivial) ethane-1,2-diamine (substitutive) tris(ethylenediamine)iron(3+) (trivial + additive)

If both multipliers and group/ligand names are not prefixes, even less are their combinations: ‘dihydro’, ‘tetramethyl’, ‘hexaaqua’.

Skeletal replacement “prefixes” such as ‘aza’, ‘bora’, or ‘phospha’, are combining forms composed of roots ‘az’, ‘bor’, ‘phosph’, respectively, and the functional morpheme ‘a’. Geometrical and structural “prefixes” are also combining forms; some of them have more than one root, for instance triangulo-, tetrahedro- and octahedro-. The “prefix” ‘cyclo’, used in slightly different ways in different context, e.g. in organic and inorganic nomenclature, is nonetheless unmistakeably a combining form.

In short: if something called “prefix” has an identifiable root but usually is not found as an independent word, it is not a prefix but rather a combining form.

*	The prefix ‘an’ is an alternative form of the prefix ‘a’, from the Ancient Greek ἀ. This latter prefix is found in words such as agnostic, apathy, aphonia and so on. The form ‘an’ is used when followed by ‘h’ or vowels.

References

1. Leigh, G.J., Favre, H.A. and Metanomski, W.V. Principles of Chemical Nomenclature: A Guide to IUPAC Recommendations. Blackwell Science, 1998, p. 71.
2. Connelly, N.G., Hartshorn R.M., Damhus, T. and Hutton, A.T. Nomenclature of Inorganic Chemistry: IUPAC Recommendations 2005. Royal Society of Chemistry, Cambridge, 2005.
3. Hellwich, K.-H., Hartshorn, R.M., Yerin, A., Damhus, T. and Hutton, A.T. (2020) Brief guide to the nomenclature of organic chemistry (IUPAC technical report). Pure and Applied Chemistry 92, 527—539.
4. IUPAC and IUPAC-IUB (1974) Nomenclature of cyclitols (1973 recommendations). Pure and Applied Chemistry 37, 283—297.
5. IUPAC and IUPAC-IUB (1975) Nomenclature of carotenoids (rules approved 1974). Pure and Applied Chemistry 41, 405—431.
6. McNaught, A.D. (1996) Nomenclature of carbohydrates (IUPAC recommendations 1996). Pure and Applied Chemistry 68, 1919—2008. 2-Carb-4. Configurational symbols and prefixes.
7. IUPAC-IUB Commission on Biochemical Nomenclature (CBN) (1978) Nomenclature of lipids. Recommendations, 1976. Biochem. J. 171, 21—35. Lip-1.13. Stereospecific Numbering.
8. Giles, Jr., P.M. (1999) Revised Section F: Natural products and related compounds (IUPAC recommendations 1999). Pure and Applied Chemistry 71, 587—643.
9. Hiorns, R.C., Boucher, R.J., Duhlev, R., Hellwich, K.-H., Hodge, P., Jenkins, A.D., Jones, R.G., Kahovec, J., Moad, G., Ober, C.K., Smith, D.W., Stepto, R.F.T., Vairon, J.-P. and Vohlídal, J. (2012) A brief guide to polymer nomenclature (IUPAC technical report). Pure and Applied Chemistry 84, 2167—2169.
10. Bünzli-Trepp, U. Systematic Nomenclature of Organic, Organometallic and Coordination Chemistry. EPFL Press, 2007, pp. 36—37.