Talk:Nonmetal

Source-checking tag

@Ldm1954: I noticed your comment of 02:51 April 21st 2025 that, following subsequent clean-up, most of your earlier concerns about problematic cites had been addressed, and that you had considered removing the RS tag.

On that basis I’m inclined to remove the tag, while of course leaving any specific citation issues to be dealt with individually as they arise. Sandbh (talk) 12:08, 7 July 2026 (UTC)

@Sandbh, while @YBG, @Johnjbarton and I have done a fair amount of cleanup, it is not clear that any concensus has been reached that this page is good enough for the tag to be removed. In a quick spot check I see that there are still errors. For instance the table Some general chemistry-based differences between metals and nonmetals claims that non-metals do not form alloys with metals. This is wrong, for instance carbon in ferrous steels is a famous solid solution. Many non-metals are critical at low levels in metal alloys, this is standard metallurgy. Ldm1954 (talk) 13:16, 7 July 2026 (UTC)

@Ldm1954: Thanks for the prompt reply. I agree that carbon in steel, and other nonmetals present at low levels in metal alloys, are real and important cases. However, I don’t think the table claims that nonmetals do not form alloys with metals. Rather, it is headed Some general chemistry-based differences between metals and nonmetals and is intended as a broad general-chemistry comparison, not an exhaustive account of every possible metal–nonmetal interaction.

The entry in question was thus intended to indicate the general-chemistry contrast: metals combine with metals as alloys.

Separately, I note that the tag concerns citations that may not verify the text, rather than article errors or possible refinements more generally. So I think it would be helpful to distinguish between wording that could be improved and cases where a citation does not verify the text. Sandbh (talk) 12:28, 13 July 2026 (UTC)

There are other errors, for instance both Au and Pt are quite electronegative. The page still is too much WP:SYNTH for me. Ldm1954 (talk) 13:21, 7 July 2026 (UTC)

@Ldm1954: There is no error re Au and Pt being quite electronegative, that I can see, since the table refers to metals having low to high EN values. Re WP:SYNTH, would you happen to have an example in mind? Thank you. Sandbh (talk) 11:46, 13 July 2026 (UTC)

Delete the two short "general properties" tables

The two short tables

1. Some general physical differences between elemental metals and nonmetals and
2. Some general chemistry-based properties of metals and nonmetals

are both very vague and poorly sourced. There are much more extensive versions of both much later. I suggest deleting them as they do not add anything. Ldm1954 (talk) 03:31, 13 July 2026 (UTC)

Both tables are intended to provide snapshot overviews of the applicable properties of nonmetals, and to pave the way for the general reader. The title of the second table is poorly expressed. Better would be “Some general chemical differences between elemental metals and nonmetals”. The second table’s heading has a footnote saying, “For a more specific assignment of chemical properties to the differing types of metals and nonmetals see the Comparison of chemical and physical properties section later in this article”, which is helpful but an equivalent needs to be added to the first table. The main body text accompanying the second table needs some adjustment since its logical flow doesn’t match the entries in the table. Sandbh (talk) 12:47, 13 July 2026 (UTC)

I agree with the deletion. This information is better suited for prose rather than a table with telegraph-like entries such as Shiny, colored or transparent; all but one solid or gaseous. Which one is not solid or gaseous? Prose allows expressing this more naturally, and this is already accomplished by the accompanying text. A summary is not needed because the lead section already serves that purpose, listing only the most relevant properties. Jähmefyysikko (talk) 11:36, 14 July 2026 (UTC)

Thanks, @Jähmefyysikko, for taking the time to explain your concerns. I would appreciate a little more time to consider them and to post some follow-up thoughts.

I remain committed to improving the article overall, with a view to eventually addressing the concerns represented by the source-checking tag. Since that tag has been in place for over a year nigh on two years, with relatively little progress since then towards its removal, I hope it would not cause any difficulty if the two general-properties tables remained in place for a little longer while the issues raised here are considered. Sandbh (talk) 12:59, 14 July 2026 (UTC)

Chemistry of the Non-Metals by Ralf Steudel (removed)

The physics-based section of our article used to say:

These definitions are equivalent to stating that metals conduct electricity at absolute zero, as suggested by Nevill Francis Mott, and the equivalent definition at other temperatures is also commonly used as in textbooks such as Chemistry of the Non-Metals by Ralf Steudel and work on metal–insulator transitions.

There are some problems with Steudel in that he writes:

"The strictest criterium to define nonmetals in contrast to metals is the electric conductivity. Typically, metals show a finite conductivity at ambient conditions, whereas the conductivity of nonmetals is close to zero. With this definition, 23 of the known chemical elements are nonmetals, and these are the subject of this textbook, namely hydrogen, boron, carbon, silicon, germanium, nitrogen, phosphorus, arsenic, the chalcogens, that is, oxygen through tellurium, as well as halogens and noble gases."

The first problem is that Steudel uses the qualifier typically, which provides him with an out to count carbon as a nonmetal even though carbon in its most stable form in ambient conditions, as graphite, is a good conductor of electricity. His definition is therefore not an equivalent since Mott's definition counts carbon as a metal. Second, he also counts arsenic as nonmetal even though its electrical conductivity in ambient conditions falls within the range of other elemental metals.

I have therefore removed mention of Steudel's text. Sandbh (talk) 00:53, 21 August 2026 (UTC)

Sandbh (talk) 00:53, 21 August 2026 (UTC)

I don't think this approach is incorrect. You are impeaching a source based on your own analysis of its content. A book by a university professor in the field of the professor from a reputable publisher would be reliable, absent published reliable evidence against it.

Steudel's definition is not strictly equal to the one claimed as by Yonezawa. Yonezawa says Mott says metals conduct at zero k; Steudel say nonmetal have close to zero conductivity in ambient conditions. These are very different claims so you are right to change the content. Per WP:NPOV both should be discussed.

The argument about graphite is really an argument against the concept of identifying elements as metals. Carbon is not a metal or a nonmetal. Graphite is a metal and diamond is not. Johnjbarton (talk) 03:14, 21 August 2026 (UTC)

In retrospect Johnjbarton, I think you're right. Steudel's approach is internally consistent. His "typically" qualifier applies to both metals and nonmetals. Thus carbon and arsenic, although they each have finite electrical conductivity, can still be counted by him as nonmetals, if he counts them as being atypical. He's a chemistry professor writing on the chemistry of the nonmetals, in an updated translation of the 5^th German edition of 2013, incorporating the literature up to spring 2019, and published by a reputable academic publisher.

The tension lies rather in citing him in a part of the article concerned with a physics-based approach to distinguishing metals and nonmetals.

Per WP:NPOV, I think Steudel may instead warrant mention in a more chemistry-based part of the article, given the relatively widespread classification of carbon and arsenic as nonmetals in the chemistry literature.

On carbon, I agree that in one sense it's neither intrinsically a metal nor a nonmetal since its allotropes differ. However, the properties of the elements—with the exception of phosphorus—are frequently discussed on the basis of their most stable forms at or near ambient conditions. On that basis, graphite is metallic, or a metal, from an electronic-structure standpoint (per User:Ldm1954), while carbon is conventionally classified as a nonmetal from a chemistry perspective.

Sandbh (talk) 12:21, 23 August 2026 (UTC)

References

Correlation does not imply causation

I think this article (and some others) can suffer from the classic problem that correlation does not imply causation. If we go back a century quantum mechanics was barely understood, and there was only the start of an understanding of chemical bonding. Hence early texts used observables to classify elements. Sometimes this still persists in high school or even some undergraduate texts, the standard lie-to-children. I will argue that we need to be very careful to avoid over-emphasis on derivative observable where there is correlation with the elements being classifiable as nonmetals, but no causation. As obvious examples:

• Electrical conductivity is a consequence of available delocalized states at the Fermi level and the Pauli exclusion principle.
• Ductility is a consequence of the crystal structure (number of slip systems) and the Peierls barrier.

These are just two. While the page should not be over complex, at the same time it should be rigorous. Ldm1954 (talk) 13:02, 24 August 2026 (UTC)

Density and electronegativity

I have added a short “Density and electronegativity” subsection to illustrate the textbook observation that nonmetals generally have low densities and high electronegativities. It gives numerical bounds for the usual nonmetal set and for the broader set including the six commonly recognised metalloids, based on the highest density and lowest electronegativity in each set.

For example, phosphorus has the lowest electronegativity (2.19) in the usual nonmetal set, so all members of that set have electronegativities of at least 2.19. The bounds describe the specified sets and depend on which elements are included. This accommodates varying treatments of the metalloids, some or all of which may be counted as nonmetals. Sandbh (talk) 05:59, 12 September 2026 (UTC)

The addition is not false, but I fail to see how it is encyclopedic. I think it fails WP:INDISCRIMINATE as part of what Wikipedia is not. That non metallic elements tend to be less dense is "correlation does not imply causation" and not that reliable; both Li and Be are less dense than graphite. Per WP:NOTEVERYTHING I think it should be removed, just because it is in some textbooks does not mean we should include it. To quote Dietrich Fischer-Dieskau:

"Brahms believed that there was no need to publish everything that Schubert ever wrote."

Ldm1954 (talk) 13:52, 12 September 2026 (UTC)

Assuming that the sourcing is correct, I think the paragraph is fine. It is short, does not overpromise, and it is a claim about the category nonmetal, exactly the kind of information we want in this article. The information is observational, more chemistry than physics, and I think the wording makes it clear this is not causal.

However it should not start "Several authors[180] have noted ...". This implies a controversial or idle claim. If reliable sources say what the paragraph says, then Wikipedia should summarize what they say and cite them.

I agree that if any of the sources call out exceptions (density of Li, Be for example) then representative exceptions should be added. Johnjbarton (talk) 16:47, 12 September 2026 (UTC)

The majority of the alkali and alkali earths are less dense than graphite, e.g. https://www.lenntech.com/periodic-chart-elements/density.htm Ldm1954 (talk) 16:58, 12 September 2026 (UTC)