Let’s name a simple inorganic chain (a):
 |
| (a) |
|---|
|
The shortest systematic name I can think about is 1,2-dinitrosodioxidane, based on the parent hydride dioxidane (aka hydrogen peroxide). Alternatively, we can emphasise the structure’s symmetry by naming it as a dinuclear entity, bis(nitrosyloxygen)(O—O).
Or we can have a go at it employng yet another type of nomenclature developed for inorganic chains and rings (ICR): 2,5-diazy-1,3,4,6-tetraoxy-[6]catena [1, 2 IR-7.4]. What’s going on here?
The ‘catena’ bit means that the structure in question is a chain; ‘[6]catena’ simply says that the chain consists of six atoms:
| X1–X2–X3–X4–X5–X6 |
All atoms in the chain are named with ‘y’ terms [1, Table 1], in our case ‘azy’ for nitrogen and ‘oxy’ for oxygen, and listed alphabetically. They are preceded by corresponding Greek-derived multipliers, thus ‘diazy’ and ‘tetraoxy’. The locants indicate the positions of each type of atom in the chain, so ‘2,5-diazy’ and ‘1,3,4,6-tetraoxy’.
| O1–N2–O3–O4–N5–O6 |
The resulting name, 2,5-diazy-1,3,4,6-tetraoxy-[6]catena, is as straightforward as it is boring. To interpret it, you don’t need to know the names of parent hydrides (such as dioxidane), groups (like nitroso) or ligands (nitrosyl). Still, one might wonder if it’s worth the bother to create such a long name for ONOONO.
We can name ring structures in a similar fashion, except for using ‘cycle’ instead of ‘catena’.
 |
| (b) |
|---|
|
For instance, the structure (b) will be 2,4-diazy-1,3-disulfy-[4]cycle. This name requires minimal a priori knowledge (just that of ‘y’ terms) while describing exactly what (b) is. In contrast, the name like 1λ4,3,2,4-dithiadiazete calls for familiarity with Hantzsch-Widman nomenclature as well as λ-convention [3], while the binary-type name, disulfur dinitride, does not even tell us that (b) is a ring.
So far so good. To quote the Red Book [2, p. 118],
the advantage of this nomenclature system lies in the simplicity with which complicated structures can be derived from the name and vice versa.
What if chains or rings have groups/ligands attached? Then they can be named as, well, groups or ligands [1, p. 1681]:
In principle all atoms in the molecule can be treated as a part of the nodal framework. The resulting names will, however, become too cumbersome to be practical. Therefore it is preferable to name some atoms or groups of atoms as ligands to the nodal skeleton.
Consider triphosphoric acid (c):
 |
| (c) |
|---|
|
Its “practical” ICR name will be 1,7-dihydrido-2,4,6-trihydroxido-2,4,6-trioxido-1,3,5,7-tetraoxy-2,4,6-triphosphy-[7]catena [2, p. 134]. As you can see, it is a combination of a chain named in a “pure” ICR fashion with standard additive nomenclature for ligands*.
 |
| (d) |
|---|
|
Likewise, the ICR name for (d) will be 2,4,6-trihydroxido-2,4,6-trioxido-1,3,5-trioxy-2,4,6-triphosphy-[6]cycle [2, p. 133].
Now I don’t know about you but I find the whole ligand-attaching business disappointing; this is how the hope of simplicity begins to fade. But wait, you ain’t seen nothing yet.
Examples (a)—(d) show simple chains and rings. Imagine that you need to name the structures that contain branched chains, or more than one ring, or combinations thereof. The overall topology of the structure is specified by combining ‘catena’ and/or ‘cycle’ with Greek multipliers ‘di’, ‘tri’, etc., as in ‘dicatena’, ‘tricycle’, ‘catenadicycle’ and so on. The connectivity of chains or rings is indicated by a nodal descriptor, which looks like a bridge descriptor in von Baeyer nomenclature. Except it isn’t. Observe the structure (e):
 |
| (e) |
|---|
|
It could be thought of as a branched chain consisting of two inorganic chains, thus ‘dicatena’. The main chain Si1–O2–P3–O4–Si5 contains five nodes while the branch –O6–Si7 has two nodes and is attached at the position 3; this is reflected in the nodal descriptor ‘[5.23]’. Assigning the atom identities to the nodes, we get ‘2,4,6-trioxy-3-phosphy-1,5,7-trisily-[5.23]dicatena’. Finally, attaching the ligands ‘methyl’† and ‘oxido’ results in 1,1,1,5,5,5,7,7,7-nonamethyl-3-oxido-2,4,6-trioxy-3-phosphy-1,5,7-trisily-[5.23]dicatena.
 |
 |
| (f) | (g) |
|---|
|
Similarly, nodal descriptors for polycyclic structures consist of two parts divided by the full stop. The first numeral corresponds to the number of nodes in the main ring, that is, the ring with the largest number of nodes. To indicate that the descriptor is for a cyclic structure, this numeral is preceded by a zero. Thus, ‘04’ for the structure (f) and ‘08’ for (g). The numerals after the full stop specify the bridges. In (f), the bridges contain no nodes and so are indicated by zeros with pairs of superscript locants, viz. ‘01,3’ and ‘02,4’, so the complete descriptor is ‘[04.01,302,4]’‡ and the ICR name is tetraphosphy-[04.01,302,4]tricycle. In (g), the bridges contain one node each and therefore are indicated as ‘11,5’ and ‘13,7’ which gives us the complete descriptor ‘[08.11,513,7]’‡ and the name 1,3,5,7-tetrahydrido-2,4,6,8,9,10-hexaoxy-1,3,5,7-tetrasily-[08.11,513,7]tricycle.
ICR nomenclature uses the combining forms ‘catenium’ and ‘cyclium’ for cations and ‘catenate’ and ‘cyclate’ for anions.
 |
 |
| (h) | (i) |
|---|
|
So (h), the monocationic form of white phosphorus (f), is named tetraphosphy-[04.01,302,4]tricyclium(1+). The structure (i), the fully deprotonated form of catena-triphosphoric acid (c), is named 2,2,4,4,6,6-hexaoxido-1,3,5,7-tetraoxy-2,4,6-triphosphy-[7]catenate(5−).
The recommendations [1] are a quarter century old already and I think it’s fair to say that this nomenclature didn’t exactly catch on. Why? For one thing, the names generated according to it are just too long to be “practical”. Given that the optimal line length in most printed texts is between 60 and 70 characters per line, the ICR names for even simple molecules won’t fit a single line; e.g., the name 1,7-dihydrido-2,4,6-trihydroxido-2,4,6-trioxido-1,3,5,7-tetraoxy-2,4,6-triphosphy-[7]catena (c) contains 91 characters. For another thing, the superficial similarity of nodal descriptors to well-established von Baeyer descriptors can be is quite confusing.
| * | Note those terminal ‘hydrido’ ligands attached to the terminal oxygen atoms: in this way, the name is based on the longest chain in the molecule. Otherwise, (c) could be called 1,1,3,5,5-pentahydroxido-1,3,5-trioxido-2,4-dioxy-1,3,5-triphosphy-[5]catena. This name is shorter than 1,7-dihydrido-2,4,6-trihydroxido-2,4,6-trioxido-1,3,5,7-tetraoxy-2,4,6-triphosphy-[7]catena but, you may have noticed already, ICR nomenclature is not about giving shortest names. |
| † | In ICR nomenclature, organic substituents such as ‘methyl’ are treated as ligands and do not form part of chain. |
| ‡ | Note that here, in contrast to von Baeyer descriptors, the numerals indicating the bridges are not separated by full stops. |
References
- Fluck, E.O. and Laitinen, R.S. (1997) Nomenclature of inorganic chains and ring compounds (IUPAC Recommendations 1997). Pure and Applied Chemistry 69, 1659—1692.
- 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.
- Powell, W.H. (1984) Treatment of variable valence in organic nomenclature (lambda convention) (Recommendations 1983). Pure and Applied Chemistry 56, 769—778.
No comments:
Post a Comment