Wikipedia talk:Featured article candidates/Nonmetal/archive7

Hydrogen behaving like a metal

On the FAC page, Mirokado wrote:

• §Unclassified nonmetals:

  I don't think the three examples of hydrogen behaving like a metal are very convincing. Being metallic is a bulk property deriving mainly from delocalised electrons and no band gap:

  Losing an electron to form compounds is something that metals do too, it doesn't really make hydrogen metallic

  Substituting for a alkali metal atom is a bit more convincing: how does that differ from the "forming a hydride" point?

  Some at least of the hydrides are much more like solutions in that the hydrogen can be readily extracted again (not for example possible with a gold ring ruined by mercury).

  In any case, hydrogen has no ability to exhibit bulk metallic properties on its own under normal conditions.

  Where hydrogen itself is thought to exhibit metallic properties is deep inside gas giants like Jupiter. That is what I was expecting when I first saw the mention of hydrogen as a metal. A suitable reference for this may be the following from Jupiter (DOI link):

  {{cite journal |last=Smoluchowski |first=R. |year=1971 |title=Metallic interiors and magnetic fields of Jupiter and Saturn |journal=The Astrophysical Journal |volume=166 |page=435 |doi=10.1086/150971  |bibcode=1971ApJ...166..435S |doi-access=free}}

  • It is quite proper to retain the current content despite the above when it is supported by reliable sources, but adding the mention of the high-pressure metallic phase would balance and clarify the paragraph. -- Mirokado (talk) 21:21, 6 October 2023 (UTC)

---

Please bear in mind that the article does not say that H is metallic. Rather, it says that H behaves in some respects like a metal, not that it is a metal.

I agree that H does not exhibit any bulk metallic properties on its own under normal conditions.

Since the article is based on the elements in ambient conditions, the purported existence of metallic hydrogen at close to 500 GPa is basically out of scope. Equally, the following nonmetals would become metals at that kind of pressure: B, Si, Ge; P; O, Se, Te; Cl, Br and I.

The alloy-like properties of metallic hydrides are as follows:

A. Solid solution: Like alloys, metallic hydrides can involve the formation of solid solutions. In the case of metallic hydrides, hydrogen atoms largely occupy interstitial sites within the crystal lattice of the metal. This can be seen as a solid solution of hydrogen in the metal lattice, similar to how alloying elements are incorporated into the crystal lattice of metals in alloys.

B. Variable composition: Metallic hydrides can have variable compositions. The amount of hydrogen that can be absorbed by the metal lattice can vary depending on factors like temperature and pressure. This is reminiscent of alloy systems where the composition can be adjusted to achieve specific properties.

C. Reversible absorption/desorption: At least some metallic hydrides can reversibly absorb and release hydrogen. This is akin to the reversible phase changes observed in some alloys under varying conditions.

D. Property tuning: Like alloys, the properties of metallic hydrides can be tuned by adjusting factors like hydrogen concentration and temperature. This allows for some control over their mechanical, electrical, and thermodynamic properties, similar to how alloys are designed for specific applications.

Describing metallic hydrides as "alloy-like" is consistent with them sharing some characteristics with alloys, particularly in terms of solid solution behavior and property modification.

*    *     *

The physics-based definition of a metal as having delocalized electrons and no band gap does not take account of chemical properties. Thus, arsenic has delocalized electrons and no band gap but a predominately nonmetallic chemistry.

From 1809 onwards, chemists realized that they also needed to consider chemical properties when working out which elements were metals and which nonmetals.

Chemically, H shows the following metal-like properties:

Similarities with the alkali metals

1. Electronic structure of H(1s¹) and the alkali metals (ns¹).

2. In chemical reactions, H and the alkali metals usually lose their single valence electrons; all usually have a valence of +1.

3. Both form similar compounds e.g. HCl and HNO3 and alkali metal chlorides (XCl) and nitrates (XNO₃); H can stand in for alkali metals in typical alkali metal molecular structures; see: "Hydrogen adopts alkali metal position". Here, H is clearly not behaving as a hydride.

4. H is capable of forming alloy-like hydrides with some transition metals.

5. During the electrolysis of water, H is liberated at the cathode, indicating its electropositive character akin to alkali metals during electrolysis of their molten salts.

6. H has an electron affinity of 73 kJ/mol cf. Li 60; Na 53; K 48

I’ve adjusted the hydrogen paragraph to more carefully refer to 2, 3 and 4. --- Sandbh (talk) 06:16, 9 October 2023 (UTC)

@Sandbh: If you're planning to take a mixed physics-and-chemistry approach to nonmetallicity (which is fine with me for an overview article like this, since many sources do it), then it seems to me that your treatment of astatine needs some clarification. You write that theoretical and experimental evidence suggests that it's a metal, but the 2013 article predicting that At is a metal is using the physics sense alone. If we look at the chemical properties of At, then obviously they are not known very well right now. What's more, what's known certainly does not strike me as being clearly weighted towards metallicity. For example, there is a large region of stability of At⁻ (aq) on the Pourbaix diagram, and in general astatine speciation doesn't look too different from what you'd expect from the halogen trend: Rayner-Canham's The Periodic Table calls At a nonmetal, just citing a Pourbaix diagram determination. (There is here a similarity with Po; both of them metallise due to relativistic effects. Otherwise they would probably both be semiconductors, and their chemical properties don't seem inspiring enough for people to call them clearly metals. Although for a while Ge was called a poorly conducting metal, so who knows?) I think that it might be better to say that physically At is expected to be a metal, chemically it seems to have both metallic and nonmetallic properties (you already basically say this in the notes, but I think it should be in the main text for balance with the previous point), and ultimately, we're not talking about it much simply because we are technologically unable to investigate it in as much detail as we'd like to. Double sharp (talk) 03:01, 10 October 2023 (UTC)

Thank you Double sharp.

Per your request, I edited the main body of the article so that it now reads:

"While astatine is capable of acting, in terms of its chemistry, both as a metal and a nonmetal it is expected to have the crystalline and electronic structure of a metal and is thereby precluded from the scope of this article.^[1][n3]"

Will this do?

As you note, the 2013 article is physics-based. It does mention early on that a qualitative argument for the solid being metallic had been made [13].

I'm not sure about Rayner-Canham (in this case) as he cites only:

•  SJ Hawkes, “Polonium and astatine are not semimetals,” J. Chem. Educ. 87(8), 782 (2010); and
•  D-C Sergentu et al., “Advances on the determination of the astatine Pourbaix diagram: Predomination of AtO(OH)2–over At− in basic conditions,” Chem. Eur. J. 22(9), 2964–2971 (2016).

I corresponded with Hawkes some years ago, pre-2013, and he conceded at that time that At was better counted as a metalloid.

It's curious that Rayner-Canham doesn't say anything about the 2013 article. Sandbh (talk) 00:29, 11 October 2023 (UTC)

@Sandbh: That's better indeed, but if you're disqualifying At by its crystal structure and band structure, questions arise about why As and Sb are not similarly excluded. (They share both with Bi, which we all agree is a metal. And in a way those two pnictogens are even better metals chemically than Po and At, as they are not known as simple aquated anions in water.) I think it's better just to say that At is excluded because we don't know much about it. Something like "Because of its extreme radioactivity, the properties of astatine are not known well enough to easily include it in the discussion. In terms of its chemistry, investigations on trace quantities of astatine have shown that it has properties characteristic of both metals and nonmetals; in terms of its physical properties, bulk astatine is predicted to have a crystal structure and band structure characteristic of a metal, though this cannot currently be experimentally verified."

And yes, that's why I only mentioned Rayner-Canham's second cite, which is about the Pourbaix diagram of At. Honestly, it doesn't look that different to that of the lighter halogens to me, especially when you consider that At⁰ must be disfavoured because of the ultratrace concentrations. So I agree with his point. This is not an article about At, though, which is why I suggested the above short version only. Double sharp (talk) 03:55, 11 October 2023 (UTC)

Double sharp: Thanks.

The crystal structure of At is expected to be FCC, which is close-packed, implying metallicity.

Whereas As and Sb are not close-packed: arsenic's packing efficiency = 38.5%; antimony = 41.

The appearance frequencies of As, Sb and Bi on lists of metalloids are 99, 87 and 5.9%.

I like your suggested form of words. I've edited the article so that it now reads as follows:

"Of the 118 known elements,^[11] roughly 20% are classified as nonmetals.^[12] Opinions differ as to the status of astatine. Its rarity and extreme radioactivity has resulted in it being frequently ignored in the literature.^[13] A comprehensive understanding of its properties is lacking and its classification remains uncertain. As a halogen it has usually been presumed to be a nonmetal.^[14] Chemically, studies on trace quantities of astatine, which are not necessarily reliable,^[15] have demonstrated characteristics of both metals and nonmetals.[16] Alternatively, given the near-metallic character of its lighter congener iodine,^{[n 3]} a succession of authors have speculated or presumed astatine would be a metal.^[18] A 2013 study based on relativistic physics concluded that it would be a monatomic metal with a close-packed crystalline structure, ^[19] although this has not been experimentally verified. Astatine is precluded from further consideration in this article due to uncertainty as to its behavior and status. The superheavy elements copernicium (element 112), flerovium (114), and oganesson (118) may turn out to be nonmetals; their status has not been confirmed.^[20]"

--- Sandbh (talk) 00:41, 13 October 2023 (UTC)

Double sharp: I hope this will work. --- Sandbh (talk) 00:42, 13 October 2023 (UTC)

@Sandbh: Thanks. Looks good to me. (I'd say that Cn, Fl, and Og may or may not be nonmetals, but that should be uncontroversial enough that I'll just change it myself.)

Since we agree re this wording, this is kind of moot, but as a point of clarification: my point about As and Sb was regarding their chemistry, rather than their physics. Po²⁻ (aq) is a thing, while As³⁻ (aq) isn't. Double sharp (talk) 09:20, 13 October 2023 (UTC)

Quotes from the literature re the nonmetallic chemistry of metalloids

• "The element arsenic possesses many of the physical properties of a metal, but in its chemical relations it is more allied to the non-metals; such elements as these are often distinguished by the name metalloids." (Newth 1894, pp. 7–8)

  --- Newth GS 1894, A text-book of inorganic chemistry, Longmans, Green, and Co, London

• "Usually, the metalloids possess the form or appearance of metals, but are more closely allied to the non-metals in their chemical behaviour." (Friend 1914, p. 9)

  --- Friend JN 1914, A text-book of inorganic chemistry, vol. 1, Charles Griffin and Company, London

• "They [metalloids] are distinguished from the typical nonmetals in…always giving compounds less acidic in character [underline added] than the corresponding compounds of the nonmetals." (Rochow 1966, p. 4)

  --- Rochow EG 1966, The metalloids, D. C. Heath and Company, Boston. Of metalloids he said they had some metallic properties e.g. the capacity to form organometallic compounds, and that they otherwise resembled nonmetals.

• "Whilst these heavier elements [Se, Te] look metallic they show the chemical properties of non-metals and therefore come into the category of "metalloids". (Sherwin & Weston 1966, p. 64)

  --- Sherwin E & Weston GJ 1966, The chemistry of the non-metallic elements, Pergamon Press, Oxford

• "metalloid…having the physical properties of metals and the chemical properties of nonmetals, e.g., As" (Grant 1969, p. 422)

  --- Grant J (ed.) 1969, Hackh's chemical dictionary [American and British usage], 4th ed., McGraw-Hill, New York

• "In general, the elemental metalloids, in their chemistry, resemble the nonmetals more than they do the metals." (Choppin & Johnsen 1972, p. 347)

  --- Choppin GR & Johnsen RH 1972, Introductory chemistry, Addison-Wesley, Reading, Massachusetts

• "Although the chemical behaviour of a metalloid is similar to that of the nonmetals in the same group, the physical characteristics are often more like those of metals…" (Huyser 1974, p. 468)

  --- Huyser ES 1974, General college chemistry, Heath, Lexington, Massachusetts

• "This book is concerned with the determination of pollutants in ambient air — in particular, the determination of certain toxic metals and metalloids (nonmetallic elements [underline added] that exhibit some of the properties of metals)." (Maccioli & Risby 1978, p. 1)

  --- Maccioli FJ & Risby TH 1978, Determination of toxic metals and metalloids in ambient air, Pennsylvania State University Press, University Park

• "Nonmetal: As used by chemists, this term includes two sets of elements; one group is made up of elements having little or no similarity to metals [noble gases, O, N, H, Cl, F, Br, C, P, S, I] and the other group of elements that are somewhat more like metals, especially in their electrical properties [B, Si, Ge, As, Sb, Se, Te, Po]. The latter group are semiconductors, sometimes referred to as metalloids although this term is obsolescent." (Hampel & Hawley 1982, pp. 204–205)

  --- Hampel CA & Hawley GG 1982, Glossary of chemical terms, 2nd ed., Van Nostrand Reinhold, New York

• "Metalloids are much more like nonmetals than metals." (Brady, Peck & Humiston 1986, p. 46)

  --- Brady JE, Peck L & Humiston GE 1986, Solutions manual for General chemistry: principles and structure, 4th ed., John Wiley & Sons, New York

• "These elements are classified as semiconducting elements, and are also known as metalloids or semimetals. The semiconducting elements resemble metals in appearance but are more like the nonmetals in chemical properties." (Moeller et al. 1989, p. 742)

  --- Moeller T, Bailar JC, Kleinberg J, Guss CO, Castellion ME & Metz C 1989, Chemistry, with Inorganic Qualitative Analysis, 3rd ed., Harcourt Brace Jovanovich, San Diego

• "In most respects, metalloids behave as nonmetals, both chemically and physically. However, in their most important physical property, electrical conductivity, they somewhat resemble metals." (Brady & Holum 1995, p. 61)

  --- Brady JE & Holum JR 1995, Chemistry: The study of matter and its changes, 2nd ed., John Wiley & Sons, New York

• "An examination of the Allen electronegativities shown in Table 1.1 (b) reveals that, on this basis, nitrogen and phosphorus are clearly non-metals, whereas arsenic is borderline non-metal/metalloid and antimony is within the metalloid band..." (Norman 1997, p. 7)

  --- Norman NC (ed.) 1997, Chemistry of arsenic, antimony and bismuth, Blackie Academic & Professional, London

• "A metalloid has the appearance and some of the physical properties of a metal but behaves chemically like a nonmetal." (Jones & Atkins 2000, p. 16)

  --- Jones L & Atkins P 2000, Chemistry: Molecules, matter, and change, 4th ed., W. H. Freeman and Company, New York

• "Metalloids typically act more like nonmetals than metals in their chemistry." (Young & Sessine 2000, p. 849)

  --- Young RV & Sessine S (eds) 2000, World of Chemistry, Gale Group, Farmington Hills, Michigan

• "…it can be said that over 70 of the 92 elements are metals; among the fewer than 22 remaining nonmetals, six are known as metalloids…" (Goffer 2007, p. 155)

  --- Goffer Z 2007, Chemical analysis: A series of monographs on analytical chemistry, 2nd ed., John Wiley & Sons, Hoboken

• "Some non-metals, such as As, Sb and Se which have the appearance and/or some of the properties of metals, but behave chemically like non-metals, have sometimes been considered within the group name 'heavy metals'. This is now considered to be too imprecise and so they are usually referred to as 'metalloids' to distinguish them from heavy metals." (Alloway 2013, p. 1)

  --- Alloway BJ 2013, Heavy metals in soils: Trace metals and metalloids in soils and their bioavailability, 3rd ed., Springer Science+Business Media, Dordrecht

• "metalloid: A non-metallic element such as silicon that possesses some of the properties of a metal." (Schaschke 2014, p. 236)

  --- Schaschke C 2014, A dictionary of chemical engineering, Oxford University Press, Oxford

• "Nonmetal: Any of a number of elements whose electronic structure, bonding characteristics, and consequent physical and chemical properties differ markedly from those of metals, particularly in respect to electronegativity and thermal and electrical conductivity. In general, nonmetals have very low to moderate conductivity and high electronegativity. The 25 elements classified as nonmetals may be considered in two groups (a) those having moderate electrical conductivity (semiconductors), all of which are solids [Sb, As, B, C, Ge, P, Po, Se, Si, S, Te], and (b) those having very low conductivity, many of which are gases [halogens, H, N, O, noble gases]. The semiconductors of group (a) were formerly called metalloids since they more nearly resemble metals than do the members of group (b), but this term in no longer used by chemists. The nonmetals are given below based on this subgrouping, though any such list is open to challenge." (Larrañaga, Lewis & Lewis 2016, p. 988)

  --- Larrañaga MD, Lewis RJ & Lewis RA (eds) 2016, Hawley's Condensed chemical dictionary, 16th ed., John Wiley & Sons, Hoboken, New Jersey

--- Sandbh (talk) 07:00, 18 October 2023 (UTC)

I have violated WP:TPO and lightly reformatted your list here to improve accessibility. As originally formatted, it would show up as 19 separate 1-item lists, each followed by unrelated indented text. As reformatted, it shows up as a single list of 19 items, each of which is followed by a subsidiary indented text. The visual appearance is virtually identical, but for those who use screen readers, the revised format makes the semantic structure clear. As to the lines with a bullet but no content, I am not sure how distracting they would be in a screen reader. By including not just an asterisk but also a colon, i think the screen reader would properly understand those lines as subsidiary to the bulleted line above. If I were doing this I would only have included three lines per block

( *"…" // *:---… // *: )

instead of the four lines I’ve used

( *"…" // *: // *:---… // *: )

to preserve the visual appearance of the wiki text as much as possible and so diffs show clearly what I’ve done. But either of these is a significant improvement IMO over what you had

( *"…" // blank line // ::---… // blank line ).

The improvement from this sort of formatting for bulleted lists is only apparent for those who use screen readers, but the same technique is immensely helpful for numbered lists. If you replaced the asterisk with a hash mark in your original, you would have seen 19 items all numbered “1”. But if you used hash signs in my reformatted version, the quotations would have been correctly numbered “1” thru “19”.

Pardon this long drawn-out blather. I think grasping the semantics of wiki markup is not immediately intuitive, but once grasped can be very helpful in understanding why odd things occur, like anomalous numbering; the accessibility for blind readers is an added benefit.

If my reformatting really grates on you, feel free to completely revert my edit including this comment.

YBG (talk) 13:31, 18 October 2023 (UTC)

@YBG: Thank you. --- Sandbh (talk) 23:12, 19 October 2023 (UTC)