# Naming ionic compounds of group B

When naming an ionic compound which includes a transition metal, the oxidation state is written between parantheses, yet some metals show an exception. Which elements represent this exception, and what is the reason?

• What is ‘group B’? – Jan Jan 11 '17 at 19:55

The relevant document including the rules of formal IUPAC nomenclature is the Nomenclature of Inorganic Chemistry — red book from 2005. Section IR-5.4.2.2 (Use of charge and oxidation numbers) is relevant for the cases you discuss. It reads:

[…]

[…]

The oxidation number (see Sections IR-4.6.1 and IR-9.1.2.8) of an element is indicated by a Roman numeral placed in parentheses immediately following the name (modified by the ending ‘ate’ if necessary) of the element to which it refers. The oxidation number may be positive, negative or zero (represented by the numeral $$0$$). An oxidation number is always non-negative unless the minus sign is explicitly used (the positive sign is never used). Non-integral oxidation numbers are not used for nomenclature purposes.

Several conventions are observed for inferring oxidation numbers, the use of which is particularly common in the names of compounds of transition elements. Hydrogen is considered positive (oxidation number $$\mathrm I$$) in combination with non-metallic elements and negative (oxidation number $$\mathrm{-I}$$) in combination with metallic elements. Organic groups combined with metal atoms are treated sometimes as anions (for example, a methyl ligand is usually considered to be a methanide ion, $$\ce{CH3}$$), sometimes as neutral (e.g. carbon monooxide). Bonds between atoms of the same species make no contribution to oxidation number.

Note that oxidation numbers are no longer recommended when naming homopolyatomic ions. This is to avoid ambiguity. Oxidation numbers refer to the individual atoms of the element in question, even if they are appended to a name containing a multiplicative prefix, cf. Example 12 above. To conform to this practice, dimercury($$2+$$) (see Section IR-5.3.2.3) would have to be named dimercury($$\mathrm{I}$$); dioxide($$2-$$) (see Section IR-5.3.3.3) would be dioxide($$\mathrm{-I}$$); and ions such as pentabismuth($$4+$$) (see Section IR-5.3.2.3) and dioxide($$1-$$) (see Section IR-5.3.3.3), with fractional formal oxidation numbers, could not be named at all.

While the section in itself does not mention the possibility of omitting oxidation numbers, the examples obviously do:

1. $$\ce{Na2[Fe(CO)4]}$$   sodium tetracarbonylferrate($$\mathrm{-II}$$), or sodium tetracarbonylferrate($$2-$$)

The generally accepted rule is that oxidation numbers can be left out if the element in question is commonly known to exist in only one oxidation state. That, for example, includes groups 1 and 2 but also elements such as zinc of which only one state is common.

Do you mean which metals or which transition metals will be an exception. Base metal ionic compounds (groups 1 and 2) do not show a Roman numeral because they have no D electrons and so will lose 1 or 2 electrons corresponding to their group number. If you mean which other metals, if I remember correctly Silver (+1) Zinc (+2) as well as Aluminum (+3 but in the Boron group) are the three non base metals that do not require a roman number to be shown in the compounds that they form.

So among ALL metals, Base metals (alkali and alkaline earth), silver, zinc and aluminum I believe do not require a roman numeral to be shown in compounds that they form unless maybe they deviate from their usual oxidation number.

Mostly elements with numerous oxidation states have their oxidation state written to specify which compound is reacting. For example:

• iron(III) chloride is a brown yellow compound, whereas iron(II) chloride is a green compound
• chromium(III) chloride is violet compound, chromium(IV) chloride is red and chromium(II) chloride is green

The general exception is when we are sure that mostly this element exists at one oxidation state. For example with sodium, we usually do not write oxidation state as it is known (group I, II etc).

• So, perhaps i should write orbitals (1s2 2s2...) to decide ? – alpy Jan 11 '17 at 18:59