Normal metals like sodium or Calcium have a positive charge as $\ce{Na}^+$ or $\ce{Ca}^{2+}$. Transition metals have a loot of variable oxidation states.

Yesterday I read about Iridium in Wikipedia and it stated that

a +9 oxidation state has been found in $\ce{IrO_4^+}$

which is astounding for me...

So i was thinking can metals have a net negative charged ion also?

According to Google :

Metals USUALLY have positive charges

Since it is written that metals USUALLY have positive charges which means there may be conditions or compounds when they have negative charges too. Is it possible?

Also can Noble gasses form ions too?

  • 2
    $\begingroup$ Addressing your noble gas ion question, they have been known for over 50 years to make compounds with formal positive charges on the noble gas atoms, though discrete noble gas cations (i.e. $\ce{Ng^{x+}}$) are not known in compounds. $\endgroup$ – Nicolau Saker Neto Mar 16 '15 at 14:46
  • 2
    $\begingroup$ More interestingly, regarding the possibility of noble gas anions, all noble gases up to radon have an endothermic electron affinity, and could not even in principle stabilize a negative charge on them. However, due to relativistic effects, the superheavy noble gas ununoctium may be able to form an isolated $\ce{Uuo^-}$ anion, though it would presumably be quite reactive. Whether a compound can be made with a $\ce{Uuo^-}$ anion is not known. $\endgroup$ – Nicolau Saker Neto Mar 16 '15 at 14:49
  • 1
    $\begingroup$ @Nocolau there is no longer "ununoctium". It's now oganesson, Og. $\endgroup$ – Oscar Lanzi Jan 21 '18 at 11:16
  • 1
    $\begingroup$ Search “Alkalide” for information on things like $\ce{Na-}$ and other first column negative states $\endgroup$ – H. Khan Jun 18 '18 at 21:06

These species usually do not exist in nature, but they can be synthesized.

  1. Silver has been reduced in liquid ammonia to give $\ce{Ag-}$.
  2. A lot of anionic metal carbonyl complexes $\ce{M(CO)_{n}^{m-}}$ have been synthesized:


  • $\ce{[V(CO)6]-}$, $\ce{[Nb(CO)6]-}$, $\ce{[Ta(CO)6]-}$, $\ce{[Mn(CO)5]-}$, $\ce{[Ir(CO)4]-}$, $\ce{[Co(CO)4]-}$, $\ce{[Rh(CO)4]-}$


  • $\ce{[Ti(CO)6]^{2-}}$, $\ce{[Zr(CO)6]^{2-}}$, $\ce{[Hf(CO)6]^{2-}}$, $\ce{[Fe(CO)4]^{2-}}$, $\ce{[Ru(CO)4]^{2-}}$, $\ce{[Os(CO)4]^{2-}}$


  • $\ce{[V(CO)5]^{3-}}$, $\ce{[Nb(CO)5]^{3-}}$, $\ce{[Ta(CO)5]^{3-}}$, $\ce{[Mn(CO)4]^{3-}}$, $\ce{[Re(CO)4]^{3-}}$


  • $\ce{[Cr(CO)4]^{4-}}$, $\ce{[Mo(CO)4]^{4-}}$

There's a review by John E. Ellis: Highly Reduced Metal Carbonyl Anions: Synthesis, Characteri's zation, and Chemical Properties on these compounds.


While Christian Aichinger's comment concerning $\pi$-backbonding in metal carbonyls is correct, it is important to realize that $\ce{CO}$ is a neutral ligand!

With other words: The Mond process: $\ce{Ni(s) + 4CO -> Ni(CO)4}$, using nickel powder and carbon monoxide, is not a redox reaction and the oxidation state of the metal still is $0$ in the metal carbonyl.

| improve this answer | |
  • 2
    $\begingroup$ The metal carbonyls have pretty much the same old positive oxidation state if we remove those carbonyls... But the Ag- is good .... $\endgroup$ – NeilRoy Mar 16 '15 at 10:43
  • 7
    $\begingroup$ Another noteworthy example would be CsAu which contains gold in the -1 oxidation state. $\endgroup$ – J. LS Mar 16 '15 at 11:53
  • 4
    $\begingroup$ I don't fully agree with @NeilRoy, but even though CO is neutral, it can pull electron density away from the metal via π backbonding ($ \ce{ M^{-}-C=O <-> M=C-O^{-} } $). I think it's helpful to remember that oxidation states are a helpful formalism to keep track of electrons, but they don't always describe the real electron distribution very well. $\endgroup$ – Christian Aichinger Mar 16 '15 at 12:40
  • 6
    $\begingroup$ Very interesting examples of metal anions are the alkalides $\ce{Na^-}$, $\ce{K^-}$, $\ce{Rb^-}$, and $\ce{Cs^-}$. Yes that's right, monoanions of the familiar alkali metals! Who said old dogs can't learn new tricks? $\endgroup$ – Nicolau Saker Neto Mar 16 '15 at 14:34
  • 1
    $\begingroup$ @NicolauSakerNeto Change your comment into an answer $\endgroup$ – NeilRoy Mar 16 '15 at 16:58

Not the answer you're looking for? Browse other questions tagged or ask your own question.