Talk:Nonmetallic material

Disputed cite: A nonmetal has a gap

The general definition section starts...

"A nonmetal has a gap in the energy levels of the electrons at the Fermi level."

...citing chapter 19 of Ashcroft & Mermin's Solid State Physics (1976).

A&M chapter 19 says no such thing.

Alternatively, please provide the page number where they do.

They mention "nonmetals" twice in their 826-page book—on p. 2 and p. 60—and neither of these say anything about a nonmetal having a band gap. --- Sandbh (talk) 01:07, 28 June 2024 (UTC)

@Ldm1954: I manually reverted your edit [1]in which you removed the disputed cite tag for the following reason:

"Added Chpt 8 as well. This is clearly stated in the description, albeit not in those precise words but the same meaning, dispute was inappropriate."

Your reasoning is in breach of WP:NOR, no original research. The cited source does not explicitly makes the statement concerned i.e. that "A nonmetal has a gap in the energy levels of the electrons at the Fermi level." — Sandbh (talk) 13:31, 29 June 2024 (UTC)

@Ldm1954: After casting around, I feel that the following extract represents a good citable source for a physics-based definition of a nonmetal:

Materials are classified as metals or nonmetals depending on the behaviour of their conductance as temperature tends to absolute zero. For nonmetals, conductance tends to zero; for metals, it tends to a finite value. This can be rationalized given their electronic band structures. In nonmetals, the energy bands are either completely filled or completely empty, and in metals, are at least one band is partially filled." (p. 25-2)

Zabet-Khosousi J & Dhiriani A-A 2016, Coupling in metallic nanoparticles: Approaches to optical nanoparticles, in Sattler KD, Handbook of Nanophysics: Nanoparticles and Quantum Dots, CRC Press, Boca Raton, ISBN 978-1-4-200-7545-8, p. 25-2

I like this extract for its explicit use of the terms metals and nonmetals, in the same passage. That, and what Mott wrote. How does this look to you?

A quibble could be the expression "it tends to a finite value", depending on what the authors meant by "tends to". I understand that only about 1 in 5 metals show superconductivity in bulk form and ambient pressure. That 4 in 5 don't would appear to support a general tendency of showing a finite value.

--- Sandbh (talk) 06:11, 3 July 2024 (UTC)

@Johnjbarton: Fyi. --- Sandbh (talk) Sandbh (talk) 06:12, 3 July 2024 (UTC)

Direct quote from The Classification of Stars

@Ldm1954 and Headbomb: Before considering an RfC I'd like to see if we can resolve the disagreement over this quote between ourselves. The full quote is:

"'Metals' (a term which is used very equivocally). Stellar interior specialists use 'metals' to designate any element other than hydrogen and helium, and in consequence ‘metal abundance’ implies all elements other than the first two. For spectroscopists this is very misleading, because they use the word in the chemical sense. On the other hand photometrists, who observe combined effects of all lines (i.e. without distinguishing the different elements) often use this word 'metal abundance', in which case it may also include the effect of the hydrogen lines. It is important to make sure in each particular case what the author really meant."

The source is Jaschek, C; Jascheck, M (1990). The Classification of Stars. Cambridge: Cambridge University Press. p. 22. ISBN 978-0-521-26773-1.

Both of you have raised concerns about the last and first and last sentences of the quote to the effect that, "This is editorializing, and shouldn't be done", and "original research and verging on academic dishonesty."

There is no editorializing, OR or academic dishonesty applicable in the case of quoted content from a reliable source. — Sandbh (talk) 08:13, 30 June 2024 (UTC)

To Sandbh and Headbomb, I am going to reverse my comment a little as I had misunderstood who was doing what. The full quote has those two sentences so in a sense they are appropriate, as Sandbh is arguing. However they are both clear editorializing by Jaschek & Jaschek which is definite not appropriate, similar to Headbomb's edit.

I did not react when Sandbh added that paragraph as I thought it was harmless. However, on further reflection and reading the relevant pages on Google Books I think it is better to remove the whole paragraph. If included then their definitions of heavy elements, lanthanides etc should be included for context and rigor, a vast digression. Further, and most critically, they cite no sources for their statements which they attribute to others/a community. This is certainly editorializing and, IMO, should not be in any text, just as it is not allowed in Wikipedia. In many cases the rules are universal.

I don't see that paragraph as adding any useful information, so a propose Delete. Ldm1954 (talk) 08:54, 30 June 2024 (UTC)

I don't necessarily object to having a quote (with the irrelevant bits removed at least), but you are right that it doesn't really add anything to what's already there. Headbomb {t · c · p · b} 09:21, 30 June 2024 (UTC)

@Ldm1954 and Headbomb: Thanks for contributing to the discussion.

Ldm1954, I note your misunderstanding, which of course can happen from time to time.

Editorialising is a WP concept, referring to editorializing by WP editors.

It does not apply to reliable sources.

In this context, asserting that Jaschek & Jaschek are editorialising "which is definite [and] not appropriate", is meaningless.

Carlos Jaschek, an astrophysicist, and Mercedes Jaschek, a stellar astronomer and spectroscopist, are matter-of-factly laying out the sitation when it comes to the conception of what a metal is in their respective fields. For convenience, here's what they say:

'Metals' (a term which is used very equivocally). Stellar interior specialists use 'metals' to designate any element other than hydrogen and helium, and in consequence ‘metal abundance’ implies all elements other than the first two. For spectroscopists this is very misleading, because they use the word in the chemical sense. On the other hand photometrists, who observe combined effects of all lines (i.e. without distinguishing the different elements) often use this word 'metal abundance', in which case it may also include the effect of the hydrogen lines. It is important to make sure in each particular case what the author really meant.

J&J's introduction to this extract is:

"Since terminology has not always been clear, we summarize brielfy the most frequent groupings of elements."

If their paragraph on metals is included then there's no need to include their definitions of heavy elements, lanthanides etc since these topics are not relevant to the scope of the article.

The extract from J & J adds useful information by nuancing the understanding of metals in astronomy and related fields.

Could you please reconsider your positions. --- Sandbh (talk) 06:55, 1 July 2024 (UTC)

@Sandbh, I have full access to the book, as perhaps does @Headbomb. Editorializing in articles without support Is inappropriate, it does not matter whether it is on Wikipedia or in a book. It adds nothing extra, the key point is in the earlier paragraph about Fraunhofer and how his work was interpreted by Bunsen etc. As Wiki editors we don't just use sources, we have to gauge them for reliability, relevance and accuracy just as, for instance, top newspapers do. Ldm1954 (talk) 07:18, 1 July 2024 (UTC)

N.B. Sandbh, the rules in Wikipedia about sourcing are not very different from the rules in academia for student term papers, theses, journal articles etc, see for instance this article at Yale. Ldm1954 (talk) 08:05, 1 July 2024 (UTC)

@Ldm1954: I've raised the question at WP:ASTRONOMY. --- Sandbh (talk) 04:45, 2 July 2024 (UTC)

General definition: Citing Chemistry of the Non-Metals by Ralf Steudel

@Ldm1954: Steudel's definition of nonmetals is somewhat misleading in that he defines them as follows:

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.

OTOH, elsewhere in his book he treats carbon as a nonmetal, even though the conductivity of graphite exceeds that of some metals; and he treats arsenic as a nonmetal even though arsenic is a metallic conductor, with a conductivty exceeeding that of some other metals.

His reference to the chalcogens as "O through Te" is incorrect. "Chalcogens" is the name of Group 16, including Po and Lv, where Po is a metal and Lv is expected to be the same

His reference to the "halogens" is incorrect, since "halogens" is the name of Group 17, including At and Ts, both of which are expected to be metals. --- Sandbh (talk) 05:05, 2 July 2024 (UTC)

My apologies, but I do not think it is appropriate to selectively criticize a textbook which you are already citing. If there are multiple views then this should be clearly explained in a page, which is the current text. No definition is given priority, all are included (hopefully) with a neutral point of view.

In fact the definition of a metal via conductivity has a range of flavors as nicely described in Fumiko Yonezawa book which is where the Mott quote comes from. The key is whether there are states at Ef, which is equivalent to Mott's T=0 form, positive resistivity with T (if no anomalous lattice variations) etc. Ldm1954 (talk) 09:31, 2 July 2024 (UTC)

Honestly, I'm getting pretty fed up with this incessant obsession over what nonmetals are. It's extremely simple. Define what a metal is. A non-metal is not that. In practice, this refers to semiconductors and insulators. If you want room temperature elements only, it's those at the right handside of the periodic table, plus hydrogen. Headbomb {t · c · p · b} 16:07, 2 July 2024 (UTC)

100% in agreement with Headbomb Ldm1954 (talk) 16:23, 2 July 2024 (UTC)

Thanks Ldm1954. I feel it's OK to discuss a cited source in the spirit of academic debate, shared learning, and our endeavours to improve article quality. Electrical conductivity is mentioned four times in the list of 32 properties that have been used in attempts to distinguish metals and nonmetals. --- Sandbh (talk) 07:17, 3 July 2024 (UTC)

Why do you expect a monograph that's not about radiochemistry to consider polonium and astatine in detail? I daresay that in practice most chemists' experience with the chalcogen column ends with tellurium. Steudel is simply writing appropriately considering the focus of his book.

And as previously explained to you, Steudel understands allotropy. In fact he mentions the graphite thing for why applying the distinction to elements (as opposed to simple substances) is somewhat arbitrary (p. 266). Carbon and arsenic are common elements that have nonmetallic allotropes, so it's natural for him to include them for completeness. Did you expect him to write what's essentially a general-chemistry book without talking about carbon? Double sharp (talk) 03:28, 3 July 2024 (UTC)

Well, I didn't write that I expected Steudel to consider Po and At in detail. I did refer to his odd use of nomenclature. After 5 editions I would've thought that this could've been brought up to scratch.

On page p. 266 Steudel writes, "Of the elements of Group 14, carbon, silicon and germanium are typical nonmetals. The latter two, however, already show semiconductor properties in the solid state and the higher homologues tin and lead are metals."[Note 3]

Note [3] says, "This distinction is – as noted before – mostly arbitrary: graphite does show many of the hallmarks of a metal (e.g., conductivity), while α-tin is nonmetallic in most regards." It's odd that for graphite, its conductivity is a hallmark of metal, yet for α-tin, which is a semi-metal in the same sense, its conductitivity is not mentioned as the hallmark of a metal.

If he noted before that "This distinction" is mostly arbitrary, I was not able find any such mention.

Something I did not notice previously: On the same page as his "strictest criterium" to define nonmetals (p. 154), he writes, "All nonmetals, however, show more or less pronounced semiconducting properties at elevated temperatures (depending on the size of the band gap)."

So, is helium supposed to show semiconducting properties at elevated temperatures?

Still earlier, he writes:

"The ALLRED-ROCHOW electronegativities of the [main group] nonmetals are larger than 1.8, those of the metals are smaller than 1.5 and in the range 1.2−1.8 elements are found that exist as metallic and nonmetallic allotropes (metalloids)."

By this criterion, Be 1.58 , Al 1.61, Ga 1.76 and In 1.66 are metalloids; and Sn 1.82 and Sb 1.98 are nonmetals. Most peculiar.

--- Sandbh (talk) 07:31, 3 July 2024 (UTC)

It's not that odd for superheavy-element specialists to question whether Og is really a noble gas. This may be something similar, since Po and At are both (1) not investigable in most labs and (2) behave quite differently from their lighter congeners in many ways. Judging by articles like this one, if you do get chemists to think about astatine's position, they often do start admitting that it might not really be worth calling a "halogen". (Though I suspect that some of the problems may also be due to the fact that especially At can only be investigated under ultratrace conditions. Ultratrace iodine chemistry is already noticeably different from the situation at normal concentrations, because the equilibrium for the hydrolysis of I₂ starts noticeably favouring HIO instead of I₂.)

Alpha-tin was long thought to be a semiconductor, which could explain the statement. The others are indeed oddities, but I suppose such lapses on topics that aren't too relevant to the main point of the text are not unheard of, and are not fatal to reliability when it comes to the source's actual main topic. (For example, I assume that Steudel was thinking about the nonmetals that form condensed phases. To make it true for helium, you'd have to severely pressurise it first, and then it would actually be a correct statement.) Double sharp (talk) 08:50, 3 July 2024 (UTC)

The 2020 article on astatine, with its byline, "Is astatine more like a halogen or like a metal? This question awakens the inorganic chemist in everyone." strikes me as being a little shy on what's been reported in the literature. Physically, there've been quite a few forecasts that At'd be a metal. As a p-block metal, it could also be expected to show significant nonmetallic character, as is normally the case for metals in the p-block. I presume At would be regarded as a p-block halogen metal, it still being a member of group 17.

A further Steuedl oddity is on p. 155 where he says that the band gap, "is typically > 3 eV for insulators, 0.5–3 eV for semiconductors and nonexistent or well below 0.5 eV for metals." Since when did metals have band gaps? --- Sandbh (talk) 07:17, 6 July 2024 (UTC)

Sandbh, what is the change that you are proposing here? Jähmefyysikko (talk) 10:02, 6 July 2024 (UTC)

Depends on your definition, I suppose. Metals surely have gaps in their band structure. What distinguishes them is that the Fermi level of a metal does not lie within a band gap. So when people say "a metal has no band gap", they're loosely speaking about the absence of a gap between the valence and conduction bands.

Steudel might be thinking of semimetals as having small negative band gaps. In such cases, like alpha-Sn, it may be quite hard to pin down the exact sign of the band gap when it is so close to zero. Double sharp (talk) 10:00, 7 July 2024 (UTC)

Feedback on proposal (pro form crosspost)

Sandbh has made a suggestion about multiple articles renaming at Nonmetal proposal which would effect this page. I (Ldm1954) think it has some merit as a start to break an impass about names and content. As a first step I have suggested combining three of the articles proposed by Sandbh on materials, metallurgy and physics into one as they are the same. If interested, please vote either Accept Merge or Reject Merge at Nonmetal proposal. One small step to break the impass. Ldm1954 (talk) 07:30, 4 July 2024 (UTC)

Materials

Since this this article is supposed to be about materials, not matter in general, the sections on chemical elements and especially on astronomy are mostly out of scope. Should those sections be removed, and instead we would add a hatnote pointing to the dab page? Also, the introduction mostly discusses the term ("...refers to..."), while the article should be about the class of materials, not the terminology. Jähmefyysikko (talk) 11:35, 12 July 2024 (UTC)

The exact name of this article has changed, and may still change -- it used to be Nonmetallic materials and elements. The focus is on the general term nonmetal across all fields including vernacular use. The lede is I think OK.

Nonmetallic material, or in nontechnical terms a nonmetal, refers to materials which are not metals. Depending upon context it is used in slightly different ways.

This certainly includes elemental nonmetals, and I see no problem with including astonomy -- why cut it out?

Personally I am not so certain about the current Metal-insulator transition section; should it be expanded with a bit more higher level science such as band structures of VO2 above/below? I would also like to see the Properties expanded, suggestions from others please. Ldm1954 (talk) 12:08, 12 July 2024 (UTC)

N.B., the properties should almost certainly not include mechanical; there is Metal#Brittle elemental metal and metallic/nonmetallic polymers. The simple fcc metals are ductile because the Burgers vector for dislocations are small, and there are numerous slip planes. Ldm1954 (talk) 12:25, 12 July 2024 (UTC)

Ok yes, this can be approached in different ways.

1. we can decide the scope beforehand, and then try to figure out a suitable title, or
2. settle on the article name and then try to figure out what usually belongs under that title in various sources.

Option 1 can be a more creative endeavour, while for 2 we can study the literature to see how the term is used, and then decide what the central topic is. The option 2, which I took for granted in the above, would probably weight materials science textbooks quite heavily, making astronomy a side topic at best. Elements would be covered in some way that is appropriate for that topic.

To me, "Nonmetallic materials", where "material" is understood from an applied perspective as something than can be made into an object, would seem to me like a notable encyclopedic topic. I am not so sure about if terminology makes such a good article, especially if there are very few sources comparing different definitions. Jähmefyysikko (talk) 12:58, 12 July 2024 (UTC)

Stepping back slightly, this page started life after Johnjbarton raised some issues with the inclusiveness of Nonmetal, so a Nonmetal (physics) stub was created. I saw it via NPR then changed it to a more inclusive version, since the same definitions are used in materials science/metallurgy/physics and also by people such as Peter Edwards (chemist) in the article about Mott you cited. Also there is the vernacular definition, which I feel needs to be stated, plus the whole issue of metal-insulator transitions and strongly correlated systems, properties etc. At the moment this page is everything that is not in Nonmetal (chemistry), with a short section on the periodic table. Within reason duplication is OK IMHO. Later, maybe, pieces can be split off into their own page if there is enough new. For instance, I don't think the vernacular definition needs a page.

Nonmetallic materials and elements was a first pass name, which Johnjbarton and YBG switched to Nonmetallic materials. I am OK with that, but I don't think anyone thinks that will be the final name. I do think a simple disambiguation list is too simple, my opinion.

Perhaps we can cleanly construct what should be in first, then look again. For instance I just added a bit of many-body physics to the definition, trying not to get too complex.

N.B., we definitely should not make this into the materials engineering of nonmetallic materials, i.e. their applications. There are already articles on polymers, ceramics etc that cover this. For See also I think. Ldm1954 (talk) 17:24, 12 July 2024 (UTC)

Fundamentally "non" is difficult to work with. I would prefer "Nonmetal" as a Wikipedia:Summary style because of the diversity of topics. Disambiguation doesn't address the exclusion aspect, eg that insulators are nonmetals. However "Nonmetal" is taken. "Nonmetal (not elements)" "Nonmetals but not elements" are silly. So I don't have more to offer on titles. Johnjbarton (talk) 19:05, 12 July 2024 (UTC)

Ldm1954: Thanks for clarification, with this background I agree that we should not make an article on polymers etc. Some scattered thoughts:

• If the article was only about conduction and condensed matter only, the name could be Insulating phases of matter or similar. It is technical, but then conducting in general is quite advanced as a topic and the name would be in agreement with that. But such article would naturally have much more narrower scope than what we have here; the vernacular and other definitions on other fields would not fit there.
• About the current article: A nonmetal has a gap in the energy levels of the electrons at the Fermi level. More correct statement could be "a nonmetal has a finite activation energy for electron conduction". The current statement fails in e.g. amorphous semiconductors which do not have a gap in DOS, but where the states below the mobility edge are localized. For a strongly interacting system the single-particle energy level is not a well-defined concept, only the excitation spectrum is.
• Kohn (1964) proposed localization as a definition for an insulator, and there are some beautiful connections to the modern theory of polarization. [2]

Jähmefyysikko (talk) 20:59, 12 July 2024 (UTC)

I am OK with your change, that is more consistent with the R/T definition -- please go ahead. Of course then you have to define activation energy and (probably) have a Fermi-Dirac plot.

Polarization...let's not blow people's mind! I always like Resta's papers, but I have trouble with polarization, and I work in the area (flexoelectricity). That would be a good section of a "Nonmetallic conduction not for dummies" article. It would also involve fun topics such as polar metal and half-metal, and that is probably too short a list. Ldm1954 (talk) 21:52, 12 July 2024 (UTC)

@Jähmefyysikko How about at the start of the conduction adding

"The original approach to conduction and nonmetals was a band-structure with delocalized (i.e. spread out in space) electron states [ref Woods, maybe Bloch/Sommerfield but I am not sure which]".

Then later add a brief paragraph on localized states in amorphous or semiconductors and similar, perhaps using this image. Ldm1954 (talk) 10:51, 13 July 2024 (UTC)