Why do non-metals not have delocalised electrons, whilst metals do have delocalised electrons? I understand that delocalised electrons is defined as “electrons that are not bound in place to a single atom or a single bond between two atoms”, and I think that delocalised electrons are necessary for the bonding of two elements (please correct me if I am wrong). If I am correct about my latter statement, shouldn’t both metals and non-metals have delocalised electrons (because they form ionic bonds)?

Clarification about my understanding of delocalised electrons would be much appreciated.

Edit: I would appreciate it if someone could comment on how I could improve my post please since it is being downvoted. That would be grateful since I am new to Chemistry Stackexhange, thank you.

  • $\begingroup$ See that "or a single bond" clause? This is about electrons delocalised between two atoms, and that's what typically happens in non-metals. $\endgroup$ Commented Feb 18, 2023 at 9:47
  • 1
    $\begingroup$ Are you aware that you are making an incorrect generalization? Or perhaps you should be clear whether by "metals" you are referring to elements or materials with particular properties? $\endgroup$
    – Buck Thorn
    Commented Feb 24, 2023 at 20:04
  • $\begingroup$ Well, instead of simply asking what delocalisation means, you asked... the thing you did and apparently wasted 50 rep. $\endgroup$
    – Mithoron
    Commented Feb 24, 2023 at 21:19
  • $\begingroup$ @BuckThorn No, I am not aware I am making an incorrect generalisation. That is why I was hoping someone could post an answer to correct my incorrect statements and help me clear my understanding. $\endgroup$
    – qwerty
    Commented Feb 25, 2023 at 11:33
  • 3
    $\begingroup$ You should probably also be aware that the explanation of particular observables (such as conductivity/delocalization) are not always unique. The mechanisms that explain electron delocalization in metals, aromatic compounds and conducting organic polymers may differ. $\endgroup$
    – Buck Thorn
    Commented Feb 26, 2023 at 16:51

1 Answer 1


Delocalization of electrons is used as an explanation for different properties. The OP asks why the electrons in non-metals are less delocalized than in metals, although some delocalization is required to bond the atoms of metals and non-metals alike. To answer the question, we need to measure a property that is tightly related to delocalization of electrons.

Thermal conductivity measures how well heat (thermal energy in motion) can pass through a material under a temperature differential. In metals, both electric current (flow of charge) and heat transfer (flow of thermal energy) are primarily carried by electrons. (Ref 1) We could expect, then, that materials with extensive delocalization of electrons would have high values of thermal and electric conductivity, and conversely, that materials with small values of thermal and electrical conductivity would have minimal delocalization of electrons, and these materials would be categorized as non-metals.

The data have been categorized in different ways for the whole periodic table (Ref 2). Several graphs illustrate the thermal

enter image description here

and electrical conductivities for the whole periodic table.

enter image description here

Both graphs use the same axes, and the blue vertical line marks manganese, at. wt. = 25 . The non-metals are bunched at the lower right corner of both graphs, but surprisingly, there are many metals that are not much better conductors than the non-metals. Use of a log scale for the vertical axis gives an even sharper separation and makes the class of metals look more uniform.

enter image description here

Nevertheless, there are differences in thermal and electrical conductivities that cannot be ascribed simply to a sea of a certain number of electrons. Other factors must include bond energy levels, including empty levels. The existence of other classifications, like metalloids and semiconductors (and doping) might suggest other mechanisms that might affect electron mobility and delocalization. A look at the plot by atomic number (below) could point to other factors.

enter image description here

The lanthanides stand out as a group, all of which have relatively low, but not zero, conductivity.

Some properties of the elements that might be related to electron mobility, like electronegativity, show no unique regions. While there are sharp distinctions between some non-metals and some metals, it seems that there are always a few borderline cases.

enter image description here

Different ways of graphing the data can exaggerate (and simplify) the differences, whereas linear plots might just blur distinctions. But the conclusion must be that delocalization of electrons is a variable that ranges from near zero to near complete and depends on factors in addition to a classification into metal and non-metal.

Edit: A new theory of metal bonding suggests that instead of visualizing cations of metal floating in an anionic sea of electrons, the spaces between metal cations may be considered voids which contain orbitals which localize some or all of the electrons in the sea. These orbitals are compressible and/or rearrangeable under pressure (see image and black box of explanation):

enter image description here

enter image description here

The images are from Chemistry World (Ref 3), which refers to the article at Ref 4, which is behind a paywall (no images in the abstract). In any case, this new model, which is based on extensive computations, gives a new view to the cation-in-a-sea-of-electrons picture, and could help us understand the mechanism of electron mobility in pure metals, similar to how mobility is understood in doped semiconductors.

Ref 1. https://materion.com/-/media/files/alloy/newsletters/technical-tidbits/issueno-104-electrical--thermal-conductivity.pdf

Ref 2. https://periodictable.com/Elements/029/data.html

Ref 3. https://www.chemistryworld.com/news/new-theory-provides-answers-to-why-metals-have-the-structures-that-they-do/4017056.article?utm_source=cw_monthly&utm_medium=email&utm_campaign=cw_newsletters

Ref 4. Y Sun et al, Proc. Natl. Acad. Sci. USA, 2022, DOI: 10.1073/pnas.2218405120


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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