User talk:Sandbh/Archive 2

when we go Sc-Y-La-Ac

We should put E121 below Ac in the extended periodic table, incidentally – despite the [Og]8s²8p¹ configuration (compare Sc, [Ar]3d¹4s²; Y, [Kr]4d¹5s²; La, [Xe]5d¹6s²; Ac, [Rn]6d¹7s²), the chemistry would not be all that different and would continue the trends down group 3 admirably well. (After our unraveling of the case of Lr, I have to wonder if the odd configurations of Rg and E121 – [Rn]5f¹⁴6d⁹7s² and [Og]8s²8p¹ respectively instead of the expected [Rn]5f¹⁴6d¹⁰7s¹ and [Og]7d¹8s² – would only appear in the gas-phase, and be perfectly normal in the condensed phase. But I can't find any condensed-phase predictions beyond Db.) Double sharp (talk) 07:54, 15 November 2016 (UTC)

The problem of where to begin the g-block is a little more vexed, since it seems to occur somewhat after Z = 122, the last element for which we have calculations. The only semi-solid figure I have sort-of found is Z = 125. I am tempted to just follow Fricke's table here and start a notional "superactinide" series where you would expect it to, just after eka-actinium, since the chemistries of the 5g and 6f elements are not really expected to be significantly different, and 7d and 8p are also supposed to partially fill there. There, if anywhere, does the block concept seem to break down if we follow the extrapolations by Fricke, Nefedov, and others. This superactinide series would continue until element 157; E158 should then be eka-rutherfordium, all the way to eka-oganesson at E172. Double sharp (talk) 08:05, 15 November 2016 (UTC)

P.S. BTW, with how much "superactinide" has been used in the literature, does that mean that E121 would have to be named "superactinium" (symbol Su, probably)? That would be entertaining, even if I find it a little unsatisfactory! Double sharp (talk) 08:07, 15 November 2016 (UTC)

Hmm. I presume the superactinides would end up getting called the E121ides, where E121 will not be superactinium! Let's see, if actinium was named after the Greek for beam or ray what would an eka-beam or -ray be? Sandbh (talk) 09:03, 15 November 2016 (UTC)

Looking at what happens in group 3, apart from the geographical scandium (Scandinavia) and yttrium (Ytterby), we have lanthanum (from λανθάνειν, to lie hidden) and actinium (from ακτίνος, ray). Rays lying hidden? What a pity roentgenium is taken! Double sharp (talk) 09:11, 15 November 2016 (UTC)

LOL! BTW I faintly suspect you might like my next edit to the Group 3 page re pesky conduction band structures. Gimme a few minutes to finish it. Sandbh (talk) 09:14, 15 November 2016 (UTC)

Your suspicions are right on the mark, as always! ^_^ Double sharp (talk) 09:36, 15 November 2016 (UTC)

Well, that's just the cat's pyjama's!! Sandbh (talk) 09:40, 15 November 2016 (UTC)

Thank you! (I had to click that.) Double sharp (talk) 09:54, 15 November 2016 (UTC)

idea for terminology to solve the d-block and f-block problems

Since the availability of d-electrons and f-electrons is important whether or not they are being filled, why not have these definitions:

• d-block element: one of the elements Sc–Zn; Y–Cd; La, Hf–Hg; Ac, Rf–Cn. These comprise the elements Sc, Y, La, and Ac in which the 3d, 4d, 5d, and 6d subshells first become occupied, and the remaining 27 elements which progressively fill these d-subshells.
• transition element: a d-block element which has an incomplete d-subshell, or which can form ions in which it has an incomplete d-subshell. This gives Sc–Cu; Y–Ag; La, Hf–Au; Ac, Rf–Rg.
• f-block element: one of the elements Ce–Lu; Th–Lr. These comprise the elements Ce and Th in which the 4f and 5f subshells first become occupied, and the remaining 26 elements which fill progressively fill these f-subshells. The first row of the f-block may be called the lanthanides and the second row the actinides, in homage to La and Ac, the elements just before them, and which they resemble due to the mostly chemically inactive nature of the f-electrons.
• inner transition element: an f-block element which has an incomplete f-subshell, or which can form ions in which it has an incomplete f-subshell. This gives Ce–Yb; Th–No.

I think this would also illustrate the similarity between Zn-Cd-Hg-Cn at the end of the d-block and Lu-Lr at the end of the f-block. Double sharp (talk) 03:20, 18 November 2016 (UTC)

At first reading, I like it. I see you say "4f and 5f subshells first become chemically active" and then, later, "due to the mostly chemically inactive nature of the f-electrons". This sounds contradictory? Sandbh (talk) 03:36, 18 November 2016 (UTC)

That's why I said "mostly" in the latter. I think I've fallen into the trap of using the same word for two different things: even though the f-electrons are not going to be ionised any further in, for example, Ho³⁺, the fact that they are there has effects on the magnetic and spectroscopic properties. Greenwood agrees: "Because the 4f electrons of lanthanide ions are largely buried in the inner core, they are effectively shielded from their chemical environments. As a result, spin–orbit coupling is much larger than the crystal field (of the order of 2000 cm⁻¹ compared to 100 cm⁻¹) and must be considered first. Note that this is precisely the reverse of the situation in the d-block elements where the d electrons are exposed directly to the influence of neighbouring groups and the crystal field is therefore much greater than the spin–orbit coupling." To further dissociate the two concepts I've changed the "block" definitions from "chemically active" to "occupied". Double sharp (talk) 03:50, 18 November 2016 (UTC)

Right, that's good then. I edited and trimmed your definitions. Sandbh (talk) 04:03, 18 November 2016 (UTC)

Thank you! Double sharp (talk) 04:08, 18 November 2016 (UTC)

Sorry if I take this out of its context, but if this is supposed to be displayed in Wiki mainspace, then I object on two of these. The thing characteristic for d-block elements is the d-shell being filled. Any reasonable introduction into this concept should start with that rather than the set of elements. Besides, I'd want at least "although other suggestions exist, the most common one is..." or something like that to show that there is at least some disagreement. We should describe and not define here. Same for f-block elements.

If the context is different, then my apologies.--R8R (talk) 13:26, 18 November 2016 (UTC)

Yes, none of this is meant for WP, where we are supposed to describe rather than prescribe. This is a suggestion for use outside WP. Double sharp (talk) 14:09, 18 November 2016 (UTC)

Aha, I see. Okay then. Sorry for the interruption.--R8R (talk) 14:44, 18 November 2016 (UTC)

No problem! Double sharp (talk) 15:22, 18 November 2016 (UTC)

looking at your old work

Post-transition metal seems GA-worthy at the very least and we may have overlooked it before... Double sharp (talk) 15:26, 19 November 2016 (UTC)

metallicities

See p.45. At is expected to be less metallic than Bi or Po (which are approaching the lower limits of metallicity), and be about as much so as Sb, so our metalloid classification seems sound. Ts is actually expected to be slightly more metallic than the clearly metallic Mc and Lv! So I think our current definition is sound for the 118 known elements. (Although I am still wondering whether the next noble gas, 172, would finally be metallised.) Double sharp (talk) 06:13, 4 December 2016 (UTC)

I somewhat doubt E172 would be metallic. From here I see it has an estimated atomic weight of 504 gm/mole, and density of 9 gm/cm, which gives a molar volume of 56 cm³/mol. In comparison, the molar volume of Rn is 50.5cm³/mol. According to the Goldhammer-Herzfeld criterion ratio, metallicity requires R/V >=1, where R = the molar refractivity of the gaseous state and V = the molar volume in the condensed state. Now, R = (4/3)Lπα, where L is the Avogadro number and α is the polarizahility. The table on p. 45 of Google books is from The Chemistry of Superheavy Elements, 2nd ed., edited by Schädel & Shaughnessy (eds). Plugging the estimated ionisation energy for E172 of 11.3 eV into the table on page 406 of Schädel and Shaughnessy I can see that α for E172 would be predicted to be marginally less than that of Rn. Given this means that R for E172 will be less than that of Rn, and that the molar volume of E172 is bigger than that of Rn, I see no prospect of a metallization catastrophe, barring relativistic funkiness. The polarizability of E172 would need to be at least (say) three times bigger for this to have any prospect of happening. Please check my line of reasoning.

The source of the p.45 table is "Eichler, B.: Das Flüchtigkeitsverhalten von Transactiniden im Bereich um Z = 114. Kernenergie 19, 307–311 (1976). Google scholar only gives a citation, with no link. The location of boron is interesting and reminded me of Greenwood's quote: " On the analogy between boron and metals, Greenwood[266] commented that: "The extent to which metallic elements mimic boron (in having fewer electrons than orbitals available for bonding) has been a fruitful cohering concept in the development of metalloborane chemistry ... Indeed, metals have been referred to as "honorary boron atoms" or even as "flexiboron atoms". The converse of this relationship is clearly also valid ..." ^_^

Ts would likely be a metal, would it not (and given the predicted metallic band structure for condensed At)?

BTW, in Schädel and Shaughnessy, it says, "If the elements Cn and Fl (element 114) have a noble-gas like character [54], then, in a fictitious solid state, they would form non-conducting colorless crystals." Does this ring true? Sandbh (talk) 11:37, 4 December 2016 (UTC)

I can find no flaws in your reasoning for E172, so I am confident about our current colouring! It rings true as a statement, but I would be cautious about drawing conclusions from it because later experiments seem to indicate that the premise is false. Cn and Fl act like unreactive metals, at least in their interaction with Au surfaces (though I'm sure you remember that paper speculating that Cn might be a semiconductor...). And yes, I found the location of boron (and to a lesser extent carbon) extremely interesting. (After all, carbon is lustrous, a good electrical conductor that behaves kind of like a metal with its delocalised electrons, and forming intercalation compounds that act like alloy analogues. I presume phosphorus is much lower because the standard state considered is white phosphorus, and not the graphite-like black phosphorus.) Since condensed At is probably metallic Ts should be a metal too. There were some old predictions (antedating the metallic At paper) predicting Ts to be semimetallic, but I think those are superseded now (because old predictions for At tended to assume that anything in group VIIB had to be a halogen). I'd still call At a metalloid, but more for its chemical than its physical properties, in which it is strangely two-faced as a halogen-metal hybrid. Double sharp (talk) 14:44, 4 December 2016 (UTC)

Polarisability of oganesson

P.S. This is also interesting because if you look at source 1 on the Og article, you will find the predicted α of Og given as over 54 au, almost double that of Rn and approaching that of Pb. But on Talk:Oganesson you appear to have done the GH calculations for that already, so the polarisability of Og would still be insufficient to induce a metallisation catastrophe. Double sharp (talk) 06:10, 12 December 2016 (UTC)

Yes, quite so. Og's not there yet. Sandbh (talk) 06:54, 12 December 2016 (UTC)

Happy Hogmanay!

		Happy Hogmanay!
		Wishing you and yours a Happy Hogmanay. May the year ahead be productive and harmonious. --John (talk) 21:29, 31 December 2016 (UTC)

Thank you John, nice to hear from you. Sandbh (talk) 10:38, 8 January 2017 (UTC)

Smart

I like this workaround. I've looked for an actual solution but couldn't find it. I've contacted DePiep; maybe DePiep will help. Anyway, in the meanwhile, this will also do; too bad I didn't come up with this myself.--R8R (talk) 22:43, 5 February 2017 (UTC)

Thank you. I hope De Piep will not mind me responding to your request. Sandbh (talk) 22:47, 5 February 2017 (UTC)

Lead FAC

John has issued an opposition decree based on prose quality, and he quoted a very strange run-on sentence indeed. I'll fix the highlighted case, but can I ask you to help me get through his criticism? I can certainly understand a possibility of such a comment (prose has constantly changed lately), but I expect you to have a greater capacity to deal with prose problems than I do. (Myself, I can only hope that this pretentious sentence was correct.)--R8R (talk) 17:32, 21 February 2017 (UTC)

Yes, I can review and adjust the prose. I'll start at the main body of the article and work through it paragraph by paragraph, so I expect it will take a little while. Sandbh (talk) 20:10, 21 February 2017 (UTC)

Thank you.--R8R (talk) 06:41, 22 February 2017 (UTC)

I've adjusted the prose, and reduced the count of John's words of concern from 31 to 3. I'll take a breather now, and see if I can look again tomorrow. Sandbh (talk) 06:26, 24 February 2017 (UTC)

I can only add how thankful I am. I have checked your edits and they seem great. This actually makes me want to rethink the need of linking words for any future writing.--R8R (talk) 10:01, 24 February 2017 (UTC)

Yes, I learnt to reduce the need for linking words by watching the way John copyedited Metalloid. I'm not as good as he is but I get the idea. Sandbh (talk) 11:50, 24 February 2017 (UTC)

One more question. Was the info deleted in this edit meant to be deleted and if so, why?--R8R (talk) 12:29, 24 February 2017 (UTC)

Will check tomorrow. Also want to look closer at mention of the Old Testament. Doesn't seem worth a mentioning anymore than any other historical source. Sandbh (talk) 12:33, 24 February 2017 (UTC)

I've had this feeling all along but I decided that we could go for a historical source. I don't have a strong opinion for either having it in or out. You can remove it if you want; just note that at the FAC page.--R8R (talk) 19:59, 24 February 2017 (UTC)

I turned the biblical cite into a note and to reduce citation clutter I merged two cites and separated them with a black four-pointed star. I'll see about doing that to other mutiple cites. Sandbh (talk) 22:54, 24 February 2017 (UTC)

It looks like I deleted that info you mentioned by mistake, but I see that you restored it so it should be good now. Sandbh (talk) 23:04, 24 February 2017 (UTC)

Pings

FYI, re your comments at WT:FAC, pings only work if you sign in the same edit, so some of your pings will definitely not have worked. If you delete them and your signature, and then re-add them in a separate edit, they should work. Best of luck with the FAC, by the way. Mike Christie (talk - contribs - library) 01:10, 28 February 2017 (UTC)

Lead refs

• Hello. Sorry things got off to a rough start. If you wish, I could change the referencing format to one similar to User:Lingzhi/sandbox. I believe I might be able to do it without extreme effort... But if not, then carry on!  Lingzhi ♦ (talk) 06:22, 28 February 2017 (UTC)

Admiration

Thank you.

I've seen your comments after the sudden FAC closure. I stand by your words and would've said so myself. I first saw this at 6 am; now that it's almost 6 pm, I finally have my time to write a comment, even though I planned to make tomorrow the day for that.

I've seen you elegantly end the skirmish. It would be difficult for me to stop there. That was why I didn't go over to add my comments, though I see what I would add. If you, however, think it's better I do (I definitely think this was done wrong procedure-wise; FAC ending up even with no consensus for promotion would be a fairer outcome than this), I will, for the sake of that this won't happen with anyone else if nothing else.

Whatever the answer, well done anyway.--R8R (talk) 14:36, 28 February 2017 (UTC)

Hi R8R Gtrs

I appreciate your post.

I feel that nothing productive would result if you added a comment to the FAC talk thread.

My focus is on getting the article to FA status.

Once that's out of the way I'll see if I want to revisit how our first FAC nomination was handled. Sandbh (talk) 06:13, 1 March 2017 (UTC)

I get the same feeling too, which is why I have not commented there either and have kept my personal annoyance to myself. At the moment, improving the article is the main priority, and being righteously indignant is far down the list. Double sharp (talk) 09:17, 1 March 2017 (UTC)

I expected this kind of a reply, and I think it's right; still, I felt obligated to ask.

I am still sort of uncertain about what would our next steps be. What do I do in the meanwhile? I see you do something; after a billion of re-readings, I still can't see what I could do apart from fussing. Well, I asked for more comments from those who raised some opposition. But what else is left?--R8R (talk) 10:09, 1 March 2017 (UTC)

Based on the comments at the FAC talk page, we apparently need to check for (a) "tracts [of text] unreferenced"; (2) unspecified issues with referencing and prose; (3) "too many sentences lack references" and another editor said (4) some of the websites we cite were questionable. I feel (1), (2) and (3) lack validity; and there could be something to (4). I just replaced one unreliable web ref with a cite from Emsley. We did assess the reliability of all of our references, and we thought that what we had was good enough but maybe one more look to see if there have been any new cites with dubious refs. Also, (5) there was an editor who posted on my talk page about errors in refs, so I need to follow up with them. And I'm going to review all of the recent edits made to this article by other editors, and check for prose suitability.

I'm not sure if Emsley is that much better, though. The annoying thing is that while he has a bibliography in Nature's Building Blocks (for example), it is never clear what is cited to what, and there are a few things in his book (e.g. natural Am, Cm, Bk, and Cf) that don't seem to actually be corroborated by any of the sources. Still, I suppose it is an improvement. Double sharp (talk) 12:30, 1 March 2017 (UTC)

So it's (4) and (5) I reckon, and then we try again. How does that seem? Sandbh (talk) 10:25, 1 March 2017 (UTC)

Seems pretty good and reasonable to me. Too bad I can't see myself helping you with this; yet if there's a way, please let me know. I've contacted those who left comments "lacking validity" (I agree on this assessment) to see if there is actually something; apart from this, this waiting currently seems dull to me. Hopefully there is something in that ref comment.--R8R (talk) 10:57, 1 March 2017 (UTC)

Oh, I forgot, there was a chemist at the FAC talk who appeared to have some concerns, too. Agree it's dull but what can we do? Doesn't seem much point doing anything apart from what we're doing now. Short-term pain, long-term gain. And lessons learnt. Sandbh (talk) 11:08, 1 March 2017 (UTC)

It's great you even turned this to your advantage by learning some lessons. But what would those be? I can't seem to find anything to learn here.--R8R (talk) 11:17, 1 March 2017 (UTC)

Ensure we think the article will pass John's prose standard. Bear in mind that there seem to be unwritten closure rules. Indicate our Support earlier, even if it is a conditional support, subject to wanting to see a few improvements, or say, Lean support, subject to whatever. Make sure the article is strong enough in the first place. I have a fair idea now of what is required. Sandbh (talk) 11:35, 1 March 2017 (UTC)

Good one about prose. As for early Support: makes sense, but how? even if we get Nergaal to do so at FAC2 here, this is hardly an universal recipe. Good to have it; but we may lack it.--R8R (talk) 12:07, 1 March 2017 (UTC)

It's just something to bear in mind; like preparing for a game of football. A team of individuals will be losers. A team of team members, will be more likely to be winners. It's just stuff we can discuss among ourselves in a more inclusive manner, before we go in. Either that or try the same thing again and fail = definition of insanity.

@R8R Gtrs: Do you have access to Ullmann's? If so, pls check section 4.2 re most European smelters now using direct smelting. Does the production section of our article need to be updated? Sandbh (talk) 12:15, 1 March 2017 (UTC)

Here's the relevant quote: "The major process for the production of primary lead from a sulfide concentrate is the sinter oxidation – blast furnace reduction route. Older processes involving direct oxidation of lead sulfide to lead or the roast reaction between lead sulfide and the oxidation products lead oxide and lead sulfate are now of little importance. In the last two decades new oxygen metallurgy processes featuring sulfide oxidation in a flash flame or by oxygen injection into a slag bath, followed by reduction of the lead oxide slag, have advanced to industrial application." Yes, we'll need an update.--R8R (talk) 12:33, 1 March 2017 (UTC)

Mine (doi:10.1002/14356007.a15_193.pub3) says, "In the past all smelting was undertaken by the smelting/sintering route followed by the blast furnace or ISF. However, although these processes are effective in producing lead, there are some major drawbacks associated with the techniques:

• There is much opportunity for pollution to occur during the operations, and therefore extensive gas cleaning equipment is necessary
• Operating the two separate stages is inefficient in terms of energy consumption and required plant
• The amount of coke required can be costly

For these reasons, much research has focused on developing methods of extracting lead directly from its ore in a one-stage process [11]. The main difficulty encountered in designing such a process was that if extraction is performed in a single vessel, either the metal obtained will have an undesirably high sulfur content, or the slag would have a high lead content. However, these concerns have been overcome, direct smelting is now the most common process for primary smelting, and in Europe most primary blast furnaces and ISF processes have now closed. Direct smelting is covered in detail in Section 4.4." Sandbh (talk) 12:44, 1 March 2017 (UTC)

Check your email. Sandbh (talk) 12:46, 1 March 2017 (UTC)

I see the oldest FAC still on the list of nominees is 8+ weeks old, whereas we were only there for about two weeks. Even though we are no longer there, the time it will take us to get the bronze star may not be any longer than 8+ weeks in total, once we get back on the FAC list. Sandbh (talk) 12:23, 1 March 2017 (UTC)

Old Testament

I added the mention of the book as an example of how lead was mentioned in a major literary work of the time. I wanted at best to have something like "in the Old Testament and the Iliad" to show how it was mentioned in description of everyday lives of people but I couldn't find anything. We now have biblical references, which is hardcore to me. If I had to choose between this and nothing, I would go for nothing.

Can we go back to what we had or forward to nothing?--R8R (talk) 10:28, 1 March 2017 (UTC)

Delete the thing for now. When we have some other reliable sources to ancient references to lead, add them back in then, in slow non-FAC time. There is no FAC urgency here. http://www.vanderkrogt.net/elements/element.php?sym=Pb mentions a couple of items but does not give any reliable sources that I can see. Sandbh (talk) 11:39, 1 March 2017 (UTC)

classifying the elements (your sandbox)

Greenwood and Earnshaw (1st edition, pp. 1051–1054) notes the existence of XeCl₄ and XeBr₂ (made from the β⁻ decay of the analogous iodine-129 interhalogen anions). So I suppose Cl and Br should both get a note about the possibility.

Xe–N bonds are known in FXe[N(SO₂)F₂] (Advances in Inorganic Chemistry, Vol. 46, p. 78). There are also Xe–C bonds, e.g. C₆F₅XeCl (also containing a Xe–Cl bond). Neither of these are from directly reacting the elements, but then neither are the Xe–O bonds. If you look at p. 93 of the same reference incidentally you will find evidence for radon oxoacids in the form of [HRnO₃]⁺ and [HRnO₄]⁻ (aq), and also that the relevant species in HF solution are RnF⁺ and Rn²⁺ (actual cationic chemistry!). (But then this is non-aqueous, so you could also argue for Xe²⁺ in SbF₅.)

(This isn't really relevant, but I wanted to include it anyway.) Calculations seem to imply also that [ArF]⁺ should have a stable ground state and should be stabilised with counter-anions like [AuF₆]⁻ and [SbF₆]⁻, and that the binding energy for [HC≡N–ArF]⁺ should be 160 kJ/mol (similar to the Kr analogue): opportunities for a modern-day Neil Bartlett?! ^_^ Double sharp (talk) 05:32, 5 March 2017 (UTC)

Thank you! I'll look closely at this. Sandbh (talk) 05:40, 5 March 2017 (UTC)

Wikipedia:Featured article candidates/Acne vulgaris/archive2

While I did not add additional comments since January 2017, I had been following the Acne FAC. I think it could be a FA someday, but still needs significant work. Your CE's were excellent! I am not good at copy editing, but knew enough to know the article was in bad shape. Next time it goes up for a FAC please email me and I will post a review again. Thank you! (just fyi: I am putting a similar note on a few user pages) --My Core Competency is Competency (talk) 13:36, 8 March 2017 (UTC)

Thank you, I appreciate the feedback. I'm looking forward to the renomination. Sandbh (talk) 22:02, 8 March 2017 (UTC)

Acne listed at Redirects for discussion

An editor has asked for a discussion to address the redirect Acne. Since you had some involvement with the Acne redirect, you might want to participate in the redirect discussion if you have not already done so. --My Core Competency is Competency (talk) 19:46, 9 March 2017 (UTC)

Acne and contraceptives

Hi Sand, I think I found the answer to reconcile your concern about acne medicamentosa (drug-induced acne) and contraceptives (helpful vs harmful) in this article https://www.ncbi.nlm.nih.gov/pubmed/19932324. This article states that the estrogen estradiol (at higher than physiologic doses and higher doses than what is found in combined oral contraceptive pills reduces sebum secretion (and would thus be expected to improve acne). In contrast, progestin (progesterone) hormones may have androgenic effects and be expected to worsen acne. This review states nearly all combination oral contraceptives are beneficial acne treatments and likely the estrogenic component offsets the androgenic part due to the progesterone component. Additionally, it depends on which progesterone is under discussion. Per the article, the progesterone drospirenone (derived from spironolactone which is also an acne treatment) (found in contraceptive products Yaz and Yasmin) has anti-androgen/anti-acne properties but progesterone norgestrel can cause acne flares. I hope this helps clarify the nuances of the relationship between COCs and acne since this was a continued concern of yours. This would mean that the initial sentence in the lead you were concerned about is still accurate (it says many types of birth control are effective as acne treatments and this is an accurate statement. For the sake of being comprehensive, I wouldn't be opposed to a brief mention that certain contraceptive products with offending progesterones (e.g., Norplant which has norgestrel) can worsen acne rather than help it in contrast to most contraceptive products. TylerDurden8823 (talk) 09:44, 11 March 2017 (UTC)

Thank you! Sandbh (talk) 09:48, 11 March 2017 (UTC)

FYI, I've made an adjustment to the acne article concisely explaining the preference for COCs with certain progestins with superior antiandrogenic effects in women who also have acne. This should address the concern you raised. When you get a chance, can you strike that out on your list on the talk page? Thanks! TylerDurden8823 (talk) 02:09, 27 October 2017 (UTC)

So done. Sandbh (talk) 04:55, 28 October 2017 (UTC)

your nonmetal classification

It's acting as a good reminder for me to plan out the rewrites of phosphorus and sulfur (which will take forever again). Also, now that all the halogens are GA I might get around to doing the other homogeneous group that isn't Li–Cs ^_^ (I suppose the noble gases are pretty homogeneous too, though in a rather degenerate way). Double sharp (talk) 16:33, 12 March 2017 (UTC)

That's a good one, the double line about the "other" (groan!) homogenous group. Now that you mention it, it is odd that there are more ostensibly heterogenous groups than there are homogenous groups (let's see, that would be 1, 3, 17 and 18)? (Even 1 is wobbly if you include H.) Last night I saw DePiep raised a concern about the cohesiveness of the intermediate nonmetals, a concern which is related to the hetero- v homogenous thing. This morning I was mulling upon what DePiep had said about "strong nonmetals" and that reminded me of thinking about the IN as being those in tne equivalent of a Goldilocks zone: not too rabid and not too laid back—more moderate kinds of folks. Taxonomically speaking, the nonmetals sure are an interesting bunch. Sandbh (talk) 22:20, 12 March 2017 (UTC)

I find the idea about the Goldilocks zone very amusing, because a Goldilocks zone for us necessitates O₂, which is definitely a "strong nonmetal" in your proposed classification! (Meanwhile, the hypothetical fluorine-breathing aliens are looking down at us from their observatories and thinking of us as complete wimps. ^_^)

LOL!

Yes, the resemblance seems to drop off abruptly outside 1, 2, 17, and 18. (17 would be more shaky if it wasn't for the short half-life of At allowing us to sweep it under the rug; 16 looks more homogeneous too without Po.) But then again the PT was originally constructed on the basis of similar valences; N, P, and V are a hilariously bad triad if you want similarity, but they all have +5 as their maximum oxidation state (and it is not an uncommon one). I think I have previously mentioned something about trends and gradual changes being the norm, instead of similarity: I wonder if the fact that chemistry teaching tends to start with the most similar groups (because, well, they're so much easier to teach) causes many people to think of similarity as the norm instead, when it quite clearly is not? Double sharp (talk) 04:49, 13 March 2017 (UTC)

This is a good conjecture! Starting with the most similar groups is perhaps the easiest way to start teaching so, sure, the alkali metals are contrasted with the halogens, and the combination of sodium with chlorine to give salt always gets rolled out. Sodium in water is another early favourite. Trends and gradual changes being the norm is also a relevant observation; still, similarities have their place with e.g. valence, as you say. And, talking the balcony view, we see similarities that enable us to distinguish metals (basic oxides, mostly) from nonmetals (acidic oxides, mostly) etc. Odd how we never read about this line of thinking, in quite the way we have been discussing it, in chemistry textbooks. Or maybe it is discussed but not all in the one place. Sandbh (talk) 05:09, 13 March 2017 (UTC)

I am pretty sure that I learnt all of this stuff from the textbooks, and I am equally sure that I did not learn it all from the same place! ^_^ The trouble with the balcony view, of course, is that you cannot get up there at once, and you have to learn a lot of stuff before you get there, preferably in multiple different ways for each thing. Similarly, you cannot give the student a bunch of helium balloons and lift them up there instantly. (Although I suppose starting with the noble gases might not be a bad idea, describing the elements up to Ca and how they "want" a noble-gas configuration.) You even have to take several detours along the way and then confess that you lied for more effective teaching: for example, I got the d-orbital explanation of hypervalence, at least at first. And when you get up to the top of the tower, you realise that it is still under construction.

Na in water is always a fun one. (Of course, everyone shows it, and then tries futilely to stop the more immature students from doing it. There are of course other fun and irresponsible possibilities for the immature, such as making H₂S. If asked, I will deny having done this. ^_^) Double sharp (talk) 06:59, 13 March 2017 (UTC)

(P.S. If it is not obvious, I would suggest that usually, the best way to reach the balcony is to take the route that past luminaries went in the past. Surely, if it's natural enough to see without prior knowledge, it should be even more natural to teach.) Double sharp (talk) 09:11, 14 March 2017 (UTC)

Barnstar!

		The Barnstar of Diligence
		Thank you for organizing the sources so clearly on smoking and acne. It has been a big help. SarahSV (talk) 23:58, 12 March 2017 (UTC)

Thank you Sarah! I appreciate your thoughtfulness. This was an unexpected but quite welcome surprise. -- Sandbh (talk) 00:59, 13 March 2017 (UTC)

Lead, again

Would you take a look at the new subsection? This probably could use some copyediting.--R8R (talk) 19:47, 16 March 2017 (UTC)

Will do. Sandbh (talk) 20:59, 16 March 2017 (UTC)

OK, that's done. I'm concerned about the last sentence: "This can be achieved by submerged fuel combustion or injection, reduction assisted by an electric furnace, or a combination of both." We should elaborate a bit as to what "submerged fuel combustion or injection" is. Sandbh (talk) 03:46, 20 March 2017 (UTC)

It seems intuitive to me, but come to think of it, yes, a good call. Will do.--R8R (talk) 10:48, 30 March 2017 (UTC)

I took a deep breath and now stand at the point where it needs no explanation. There is no uncommon word here an English speaker wouldn't understand and there's no special meaning in them.--R8R (talk) 21:42, 2 April 2017 (UTC)

I need a second opinion from you. Is it okay we currently don't have a galena pic? Nergaal asked for it and it seems it's the only comment of his I can consider unresolved. I am not yet sure if we should insert one. Of course it's better we did, but there's no room in the relevant On Earth section. We could remove the abundance graph. Not sure we should do it, though. Not sure we shouldn't, either.

After that, it's only one reply from Nergaal (see "is lead of any worry in fish?" in Talk:Lead if you're interested) and possibly subsequent action from there and I guess we could go for FAC again.

Also, I've wanted to write something to your classification proposal but I see there's much discussion already and I've spent my time reading and analyzing it. Do you mind I do it separately from it, bearing it in mind and referring to it, however?--R8R (talk) 21:42, 2 April 2017 (UTC)

I'm not fussed about the galena picture, pretty as it is. Since there is currently not enough room have you thought about a gallery, like the one in Heavy_metals#Environmental_heavy_metals? It looks like there would be room for one after the Compounds section. Galena could go there, as could red lead, lead II oxide, lead IV oxide, and, say, the infamous sugar of lead. The more I think about it, the more I'd like to see a gallery of this kind (with some interesting captions).

I did consider this once (years ago) but I read WP:GALLERY and it effectively discourages people from making these in our chemistry-related articles. I also think pictures are not too important here to make off-text galleries of them. Fluorine was an exception that I wouldn't make again: first, I didn't know about the rule back then and second, I didn't care that much about pictures, either. For comparison, the guideline suggests there is a (rare) good example of when you need a gallery: 1750–75 in Western fashion. It makes sense to have a gallery there: you can't describe a fashion style without many pictures, this is the entire point of the article. It's not our case, really.--R8R (talk) 01:49, 3 April 2017 (UTC)

Pesky lead beware, the FA cherry is upon you!

Before you go for FAC again, I recommend one more copy edit. I did one of these for one section. The rest need to be done too. I just haven't got around to it due to being busy with other things.

Now is the sort of a moment when you think, "okay, I've had enough!" I'll do what I should have done long ago (would probably do it sooner if the idea had crossed my mind earlier; as obvious as it is, it has only done so now) and learn to write good prose in English. There must be guides online on how to teach yourself to do that. Will try to find some spare time for this now as well. I'll probably need to refresh some grammar, too, anyway. This is sort of a skill I could find a use for IRL as well.

Even if I could start off right now, though, this should take time. In the meantime, can I ask you to go through? (no hurry needed) --R8R (talk) 01:49, 3 April 2017 (UTC)

Yep, I'll add it to the 'to do' list. Sandbh (talk) 01:59, 3 April 2017 (UTC)

I'm done. I've left quite a few citation needed, and clarification tags. I've asked at the science reference desk about the picture of "punched" lead around the connecting bar in the bridge picture. Once this lot is done, ping John to see if he will support. Then you're good to go. I'll support. (What an epic!) 07:13, 5 April 2017 (UTC)

Thank you! Your job was superb: I looked at some tags you added and thought, "yes, I want to tell that, too!"

I'll do it, maybe even today if I can.--R8R (talk) 08:49, 5 April 2017 (UTC)

Thank you, we're getting to cherry on top stage, which is always a pleasure. Sandbh (talk) 11:21, 5 April 2017 (UTC)

I've dealt with it. The Venice picture was good now that I've read the Science Desk reply but, alas, I couldn't find any literature on it and added a sculpture picture instead. We mention sculptures in the text and not these fixes, so it's a change fine with me. We're (almost; waiting for Nergaal) good to go now.--R8R (talk) 20:12, 7 April 2017 (UTC)

The statue image is extraordinary! I found some references to the use of lead as a movement dampening device in bridges so I'll see if I can do something with them and put the picture back in. It is a really good picture demonstrating a forgotten use of lead. Sandbh (talk) 00:25, 8 April 2017 (UTC)

Yes, please write separately on the classification proposal. Sandbh (talk) 01:13, 3 April 2017 (UTC)

Great. I will.--R8R (talk) 01:49, 3 April 2017 (UTC)

on metalloids as closer to nonmetals than metals

Thank you for quantifying my intuition at User:Double sharp#Chemistry: "So maybe the nonmetals Si, P, and S will come first. (Yes, yes, I know we colour Si as a metalloid, but you know what I mean!)" ^_-☆ 💖💖💖 (And now I finally have an excuse to use that emoji!)

Seriously, I have thought that way for most of the metalloid-ish elements for a while. They are all so reluctant to form cations...a few sections up (#redox potentials: classification of tennessine) I even discussed with you a classification for metallic behaviour in chemistry based on that, which would essentially ask the questions (1) "can it form simple cations?" and (2) "if it can also form simple anions, is their oxidation back to M⁰ spontaneous?" – I need (2) because of Au. (Incidentally, this would rule that At is not a metal, but that Ts should be one.)

Oh dear, it seems I forgot about the #redox potentials: classification of tennessine section. I presume it's been superseded by the tungsten discussion? There may be something in your attempt to distinguish between metals and nonmetals. In classification science, categories are usually defined by more than two criteria. Cld you add another criterion that addresses W? And what does your approach say about B, Si Ge, As, Sb and Te. Are they metals or nonmetals? Sandbh (talk) 01:19, 3 April 2017 (UTC)

Yes, the W problem kind of scotches this.

Indeed, when classifying just about anything, edge cases tend to scotch all ideas of a single organising principle. We always hear that metallic character increases as you go down the table, and this is true until you look carefully at the 4d and 5d metals. We always hear that metallic character decreases as you go to the right of the table; and then you see that As is actually a semimetal while Ge is only a semiconductor, so that the two of them look like they've been swapped!

I would split things into a set of chemical and a set of physical properties. To me, metallic chemistry implies a basic oxide and dissolving in acids to form salts; metallic physical properties imply high melting and boiling points, metallic lustre, malleability, ductility, hardness, good conduction of heat and electricity, and high density. It suffices to have a preponderance of these properties, because not many things can meet all of them. Sodium is an excellent metal from the chemical perspective, but is very lame from the physical perspective: it can't have a structural use because it's so soft and because it is too excellent a metal from the chemical perspective (imagine a sodium car: very fast, very light, very explody). Tungsten is great from the physical perspective but very lame from the chemical perspective. Not even iron, which I think is the poster child of the metals for most people, succeeds in meeting all of them, having an amphoteric oxide (kind of like a harder and denser aluminium). Copper might be one of the best, but its oxide is not all that basic.

I was going to write something about the nonmetals, but for the physical properties there's really not much to it but the negation of the metallic properties. Then iodine becomes a problem because it has a metallic lustre, even though it is famously a violet gas! (I wonder if astatine would be similar? Then it could look like a metal, but when it self-boiled you would see ominous black fumes pouring out. I'm not sure what that means for tennessine if anything. I mentally imagine it as behaving as a metal, at least not any worse than polonium, but that's just pure speculation. Okay, end of aside about the heavy halogens.)

I would say that nonmetals ought to react with metals to form simple salts and have acidic oxides that react with bases. The first criterion in this paragraph is perhaps a good one to exclude some things from being called nonmetals (because they don't do that).

As for looking at individual metalloids, I think we'd have to get a new section for this! ^_^ Double sharp (talk) 06:01, 3 April 2017 (UTC)

We shouldn't forget the group similarities transcending this, though. I would note after all that Greenwood covers each of the triads {Ge, Sn, Pb}, {As, Sb, Bi}, and {Se, Te, Po} in one chapter, and it doesn't feel strained like Holleman and Wiberg's chapter on {O, S, Se, Te, Po} does. Actually this is exactly the reason why I'm not sure how I would do groups like that when the story to tell breaks up so much.

(Speaking of which, why is the Sn article so terrible? Everything around it is better and it is just a yellow island of not being up to standards. I should put it on my plate, but realistically with silicon still in the works as disorganised Notepad documents on my desktop it's never going to get done in any reasonable amount of time...) Double sharp (talk) 15:47, 18 March 2017 (UTC)

the current discussion on metalloids

I said it over three years ago and I don't hesitate to say it again: thank you for always making me think.

Incidentally, I've replied to your recent post, and I am very happy that you refused my Na₂Te specialty. ^_^ Double sharp (talk) 04:40, 24 March 2017 (UTC)

Oh my, you're so quick! I was just dot-point summarising what Scerri says about the LST in his red book, when an e-mail came in from Wikipedia alerting me to this edit. I see I'll have to pull on my best thinking cap again, possibly augmented by some dog-walking thinking, sitting in front of the TV-mulling, and sleep-time processing. Bring it on I say and thank you for stress-testing this latest proposal. It will come out on the other side looking sparkly, spiffy, and groovy—I hope! Sandbh (talk) 05:07, 24 March 2017 (UTC)

As usual, you give me a lot to ponder. You may expect a reply sometime tomorrow... ^_^ Double sharp (talk) 14:59, 27 March 2017 (UTC)

Thank you. I've been trying to catch up with your posts, all of which have taken me down paths I hadn't previously thought about, at least not consciously. There is a tropical cyclone off the coast of North Queensland at the moment, which is a fair way north of me, and I feel like my proposal has been exposed to its own cyclone of scrutiny, which can only make it stronger (provided it doesn't fall over). I see DePiep has made another insightful contribution and I'll try and get started on a response to that one today. Sandbh (talk) 21:30, 27 March 2017 (UTC)

I'll probably only have time to write out that long response on looking for metallic properties as we go down the groups of the p-block in the weekend, but rest assured that I have already been thinking about it. ^_^ A nice side effect is that I am motivated to go for silicon again (I left it hanging last year to work on nitrogen first, IIRC). Double sharp (talk) 15:11, 20 April 2017 (UTC)

one advantage of bringing the transition metals a little further up your to-do list

is that almost all the d-block is ready.

On the definition of transition metals: I suppose we want to go for the widest possible one for the topic, even if not for the main article, so that would be IUPAC group 3–12. But then that was written at the time when all the Ln and An were considered to be in group 3, so we also need a lanthanide GA. We also need a gold GA, which is extremely scary. But dubnium is done and awaiting GAN, and silver should be ready to nominate by this weekend. ^_^ (Personally, though, I still think of 4–11 as the "good transition metals", with 3 being a little equivocal and 12 being not really transition at all.)

The other blocks are more spottily covered. The actinides are all done, but some lanthanides still await this – probably because working down all of them is really boring (I should finish Pr, because that is half-done in my other sandbox, but I just can't work up enough motivation to do it when Ag is far more interesting). The s-block has the historically important Ra (bad because the history spirals out of control) and the scary Mg and Ca to do. And if these blocks have lots of scary elements, the p-block is worse (although having taken care of the halogens last year, together with nitrogen across the New Year 2016/2017, maybe it'll be okay to finish Si, P, and S; it'll just take a while).

The most interesting thing I would want to do is to get a halogen GT. R8R took care of F, At, and Ts long ago; I made Cl, Br, and I GAs last year. Only the main article is missing. My concern is that just like the alkali metal article I will be unable to suppress the desire to keep writing and not stop. But now that Cl, Br, and I are all done it does feel closer than before. Double sharp (talk) 15:57, 7 April 2017 (UTC)

Lead FAC2

Finally, it begins! --R8R (talk) 19:00, 18 April 2017 (UTC)

heavier metalloids

On thinking about it a little I've realised that I am far more likely to think of B and Si as nonmetals than Ge, As, Se, Sb, Te, Po, and At, and then I remembered that you classified it the same way at heavy metals.

Incidentally on relooking at polonium there may well be enough nonmetallic properties there (e.g. it tends to follow S, Se, and Te in its metabolic pathways and its organic chem) that I am starting to mentally think of it as a metalloid again! Of course I still need to think about it during your IRL pause. ^_-☆ Double sharp (talk) 07:58, 26 April 2017 (UTC)

I expect I will be able to access Wikipedia during my break. More to follow. Sandbh (talk) 16:12, 26 April 2017 (UTC)

You can see B sometimes classed as a nonmetal reflected in Lists of metalloids where it has the lowest frequency of the elements most commonly recognised as metalloids. And there is Sb too, at 87% due to it being sometimes classed as a metal. I've never looked at the biological or organic chemistry of polonium. If an element has the band structure of a true metal (rather than a semimetal) then that currently represents a no go line for me in terms of treating an element such as Po as a metalloid. The fact that it forms a simple cation in aqueous solution is a rather strong indicator, too. I tend to think of polonium as appropriately classed as a post-transition metal however it would be good to know more about its living chemistry and how this compares to other metals.

The treatment of selenium as a heavy metal is a bit odd given it is usually counted as a nonmetal and that its chemistry, in general, is nonmetallic. I suppose this is an example of when it can be helpful to class something on the basis of a few of its properties and to class it as something else on the basis of its general properties. Sandbh (talk) 15:07, 27 April 2017 (UTC)

I am more than a bit suspicious about the highly charged cation Po⁴⁺ listed in the polonium article; I would not be surprised if this was really PoO₃²⁻ as Greenwood and Earnshaw have it on p. 755, where it would then fit with the tendency towards oxyanion formation among nonmetals; this is further corroborated by the most stable compounds of Po being the polonides and its oxides, fluorides, and chlorides, in almost exact parallels to Se and Te. I would not mind if a metal formed anions in salts, like Pt²⁻ and Au⁻, but the fact that the salts of Po²⁻ with the electropositive metals on the left side of the table are among polonium's most stable compounds gives me some pause. Meanwhile Po²⁺ is only really a thing in acidic solutions, according to Greenwood and Earnshaw on the same page.

The absence of polypolonium cations and anions may be more to do with absence of evidence than evidence of absence. Looking at the "salts", "polonium sulfate" is really 2PoO₂–SO₃, a double oxide, and much the same is true for most of its "oxoacid salts": this feels very much like metalloid behaviour to me, especially since organic Po compounds tend to be isostructural to the corresponding Te compounds (10.1002/9781119951438.eibc0182).

I think Po and At are really quite two-faced as elements: sure, they may be physically metallic, but chemically they are pretty much what you would expect for the chalcogen and halogen in period 6. If they have a few more metallic properties than their lighter congeners, then I think it is not too unexpected this far down the table, but that one or two properties may not really change the general "feel" of their chemistry. I think I've previously said that I'm a bit sceptical of "bright lines" drawn by one single property (e.g. metallic band structure – what would happen to bismuth, for instance?) and would like to look at several in common, and honestly I think Se, Te, and Po are all approximately straddling the border chemically: they create a normal "chalcogen" class. Physically, of course, Se is pretty clearly on the nonmetallic side, Po is pretty clearly on the metallic side, and Te is swinging slightly to the nonmetallic side. But chemically they all exhibit chalcogen-like behaviour, with slight variations in metallicity that don't really alter their generalised behaviour that much. Double sharp (talk) 15:29, 27 April 2017 (UTC)

I can only give a half-baked response as I don't have access to my usual resources. I don't know about the supposed +4 cation but Wiberg mentions the cherry-red +2 cation quite nonchalantly, as if there was nothing novel about it. Yes, this only forms in acid solution but that is not unusual. I will have to pass on the polonides and salts for want of references. Oxyanion formation would be consistent with post-transition metal behaviour. The only no go line I have is that an element with the band structure of a true metal, rather than a semimetal like bismuth, cannot be classed as a metalloid. If you gave me an example of a true metal that did not form a cation in aqueous solution and did not have at least one basic oxide, I would reconsider this no go rule. A semimetal like bismuth can be classed as a metal based on its metallic appearance, and capacity to form a simple cation in aqueous solution or its possession of a basic oxide. The general chemical behaviour of tellurium is regarded in the literature as nonmetallic. I will have to pass on perceptions of the general chemical behaviour polonium. Sandbh (talk) 00:57, 28 April 2017 (UTC)

Ta and W may be adequate examples of that phenomenon, if my memory serves me correctly. Double sharp (talk) 08:48, 28 April 2017 (UTC)

I'll have to think about what I meant to say here :( as Ta and W meet the lustrous appearance and close packing criteria. Sandbh (talk) 14:23, 28 April 2017 (UTC)

Having though about this some more, and relying on my limited understanding of solid state physics, the limits of which will soon become apparent, I should have said, "If you gave me an example of a true metal that did not have a close-packed crystalline structure, or formed a simple cation in aqueous solution, or had a basic oxide" then I'd reconsider the no calling true metals "metalloids" rule. Having said that, I'm now prompted to wonder why W, for example, apparently does not form a simple cation in aqueous solution nor has a basic oxide. These seem like fairly characteristic properties of metals, after all. Sandbh (talk) 16:12, 29 April 2017 (UTC)

I would note that β-Ta (metastable at STP) has a tetragonal crystal structure (not bcc), and that Pa may fail all three criteria as its structure is significantly distorted from bcc into a tetragonal crystal structure. ^_^ Sorry to continue poking holes in your criteria. Meanwhile, Po does not have a close-packed crystalline structure: actually post-transition metals don't and Al, Zn, Cd, Tl, and Pb are just the exceptions, and mind you, Zn and Cd have significant distortion from the ideal hcp structure too. The reverse case to W might be Sm, which is chemically a strong metal but has a rhombohedral crystal structure at STP, not turning into the normal hcp structure for lanthanide until 731 °C. Double sharp (talk) 16:33, 29 April 2017 (UTC)

Keep the counter examples coming! The definition was:

A lustrous appearance when freshly prepared or fractured and (a) has a densely-packed crystalline structure;1 or (b) forms a simple cation in aqueous solution;2 or (c) has a basic oxide.

   1 Hexagonal-close packed, face-centred cubic, α-lanthanum, α-samarium, body-centred tetragonal, or body-centred cubic
   2 Including aqua-cations such as [Bi(OH2)8]3+

Pa meets this criterion as far as I can see. For beta-Ta we need more information. Is it body-centred tetragonal or some other kind of tetragonal structure. If the latter we will need atoms per unit cell, and interatomic distances, in order to be able to work out packing efficiency. All the other metals meet the criteria, Zn and Cd easily at least on the basis of simple cation formation, as I recall. Sandbh (talk) 03:37, 30 April 2017 (UTC)

β-Ta is highly distorted with a structure reminiscent of β-U (tetragonal, space group P4₂/mnm, with 30 atoms per unit cell. The lattice constants are a = 10.194 Å, c = 5.313 Å; and the shortest Ta–Ta distance is 2.60 Å (there are many others, but the paper doesn't list the other values). The structure is shown at the end of the paper as figures 1a and 1b. Double sharp (talk) 05:04, 30 April 2017 (UTC)

So, the unit cell volume is a² x c = 635.88 and the volume occupied by the 30 atoms in the unit cell is no more than (presuming there is only one kind of atom in the crystal structure for this purpose unlike Mn for example which effectively has four different atom types in its structure) 4/3 x pi x 1.3³ x 30 = 276.1 which means the packing efficiency would be 276.1/635.88 = 43.4 per cent, which is Bi territory and certainly not close-packed!

I see from the paper that the structure involves five distinct sites but I tend to doubt that would make much difference to its packing efficiency.

So, we seem to have a metal (presumably) that does not have a close packed structure, does not form a cation in aqueous solution, and does not appear to have a basic oxide. The only thing that can be done is to see if Ta forms any lower oxides that might be basic (I tend to doubt it) or amend the definition in some way, perhaps by referring to a positive temperature of coefficient of resistivity (which could be hard without eliminating Pu and including As). Sandbh (talk) 16:47, 30 April 2017 (UTC)

Currently my only solution is to add another "or" criterion, namely, "or a density of at least 10 gm/cm³". Ionisation energy is another possibility although the cut off wouldn't be as round. And I'm inclined not to use electronegativity, since there is no universal electronegativity standard, although "or an electronegativity of no more than 1.5" could work. So the definition becomes based on five criteria again: one mandatory; four optional. Sandbh (talk) 03:27, 1 May 2017 (UTC)

The superheavies look like they are going to scotch two of them: an ionisation-energy-based criterion is going to call Og a metal (probably lower ionisation energy than Zn and Hg), and it is going to call Rg and Cn nonmetals because their ionisation energies are going to be close to those of Rn and Xe. Electronegativity may be a problem for Cn, looking at the trend down group 12 and noting that its hcp structure is expected to be similar to those of the noble gases. So density looks like the best thing we have at the moment. Mind you, copernicium seems to be expected to be so strange that I am not sure if it is a failing of a criterion to classify it as a metalloid or even a nonmetal. Double sharp (talk) 04:08, 1 May 2017 (UTC)

Yesterday I glanced through a chemistry textbook from 1942 and was reminded that metals were thought of as being characteristically heavy substances, so I feel better about adding something to the definition about this. Even our metal article says that metals in general have high densities. Sandbh (talk) 16:45, 6 May 2017 (UTC)

A "metal" is a chemical element that has a lustrous appearance when freshly prepared or fractured, and one or more of the following properties:

(a) a closely packed crystalline structure^

(b) a density of at least 10 gm/cm3

(c) simple cation formation in aqueous solution#

(d) a basic oxide.

   ^Hexagonal-close packed, face-centred cubic, α-lanthanum, α-samarium, body-centred tetragonal, or body-centred cubic
   #Including aqua-cations such as [Bi(OH2)8]3+

I suspect that if the [Be(H2O)4]2+ cation exists in appreciable concentrations in aqueous solution then the crystalline structure criterion could be eliminated and the density criterion changed to >= that of iron. Sandbh (talk) 19:05, 6 May 2017 (UTC)

And the basic oxide criterion could be eliminated too, which would make for quite a compact definition. Sandbh (talk) 02:14, 7 May 2017 (UTC)

Hmm. This article suggests the cation in question is indeed stable in aqueous solution ^_^ Sandbh (talk) 02:25, 7 May 2017 (UTC)

Disregard the struck out material. Zr does not form a simple cation in aqueous solution and is lighter than iron. Sandbh (talk) 03:42, 7 May 2017 (UTC)

There is something a bit unsettling about resorting to things as esoteric as cation formation or basic oxides when defining something that every man on the street intuitively understands. While I wouldn't want to resort to "I know it when I see it", nevertheless it would be nice if the primary part of the definition at least agreed with the everyman's understanding of what a metal is. Of course, more technical details are needed for things outside the core such as light metals or those on the edges. This is not unusual in taxonomy, that there is a simple definition that applies to the core members of a class, and more complicated details are needed as you go further from the center. Consider the definition of mammals and the first definition we normally learn (warm-blooded animals with hair or fur which bear live young) and the refinement needed when you get to the monotremes. We of course have a great deal of advantage over the 18th century Brits thought that the platypus pelt sent back to England had been sewn together in an elaborate hoax. (The comparison is not exactly parallel, but I do think it illustrative.) And consider the sign I recently noticed at my place of work, where "non metal" is used as a synonym for "not a metal". YBG (talk) 03:25, 7 May 2017 (UTC)

I find that while people do generally think of metals as heavy, they do not quite expect them to be as heavy as they can be or as light as they can be. Outside the range of about Ti to Pb, I think most people would be quite surprised (they are already often surprised by pure Al and pure Au). The root of the problem is that most people's ideas of metals do not come from most of the elements; and after high school chemistry, they tend to come from a different but no less incomplete subset of the elements.

It is intriguing how opposed the physical and chemical properties seem to be; chemically strong metals are usually physically weak, and vice versa. While it sounds ridiculous, the best exceptions to this rule of thumb I can think of are some of the rare earths and actinides(!), which are absolutely not household names apart from Th, U, and Pu – and those are some elements that most people will never encounter in person! Double sharp (talk) 03:51, 7 May 2017 (UTC)

I'm typing this on my phone. On YBGs comment I see what you are getting at. The approach I've taken here is based on the IAU definition of a planet. Hence it is more technical than the common understanding of what a metal is. I suppose this is likely to be similar to the askoxford definition: "A solid material which is typically hard, shiny, malleable, fusible, and ductile, with good electrical and thermal conductivity (e.g. iron, gold, silver, and aluminium, and alloys such as steel)." This is a problematic definition. Not many folks know what "fusible" means. Gold, silver and aluminium are soft, not hard.

So far I have this as an intro (it has its own issues since it mentions conductivity and deformability neither of which are cited in the more specific definition):

"Any of a class of elementary substances, as copper, silver, or gold, typically characterised by a lustrous appearance, conductivity, and deformability. As different metals show these characteristics to varying degrees the class is more specifically defined as follows…" Sandbh (talk) 23:43, 7 May 2017 (UTC)

Oh, and in this approach all other elements (H, He, B, C, N, O, F, Si, P, S, Cl, Ar, Ge, As, Se, Br, Kr, Sb, Te, I, Xe, Rn) are nonmetals, with metalloids distinguished by their weakly acidic or amphoteric oxides. Sandbh (talk)

That sounds good; starting with what people generally understand and then justifying the need for a more technical definition. I would change "elementary" to "elemental", wikilinking it as I presume it is meant to make this a subset of the class of elements. Oh, and do you suppose there is any way to include the sentence "Light metals are the platypus of the periodic table"? No, I suppose not. YBG (talk) 03:08, 8 May 2017 (UTC)

I recall this thread originated in my proposal to call metalloids "weak nonmetals (metalloids)" and now "metalloids (weak nonmetals)", one result of which was that Double sharp expressed an interest in what properties could be used to "define" metals. I'm not sure we would use it in Wikipedia. There is nothing quite like it in the literature the great bulk of which describes metals in general terms with few authors using even one quantitative criterion. I was going to say that it did not seem necessary to define metals generally and then specifically but I now suppose that elements are universal so no single domain such as chemistry can claim provenance over them so general + specific is good. I was worried about not saying anything about conductivity in the specific definition but I now think one needs to specifically define metals in another way, and once that is done, it informs the conductivity benchmarks for metals ie you go by the metals that have the lowest electrical conductivity and thermal conductivity. The density criterion then seems a little arbitrary but, then again, no one doubts that elements of that density or higher are metals (which is still somewhat of a weak argument). But it does appear to work for the purposes of addressing Double sharp's concerns.

Provisionally I am still of the view that the concept of a metalloid is academic laziness and the biggest obstacle to understanding the chemistry of the elements in question, and the chemistry of the nonmetals as a whole, even though a careful reading of the literature shows they behave chemically, in general, like nonmetals. Yet most of the literature glosses over the chemistry of the metalloids and does not consider them when discussing the chemistry of the nonmetals. Sandbh (talk) 03:02, 9 May 2017 (UTC)

Well, I actually think that it might help even more to ignore the metal-nonmetal line and only care about the groups outside the first-row anomaly. I find considering P, As, Sb, and Bi together to be very sensible and perhaps the ideal way to cover their chemistry, for example. Then the periodic trends down the table trump superficial similarities, rather like what we were talking about last year for the question of group 3. Indeed, I can't think of a single general inorganic chemistry text that does not do it this way, relegating metallicity to the backwater. ^_^ Double sharp (talk) 06:56, 9 May 2017 (UTC)

In fact, I might be bold enough to claim that the most important distinction in inorganic chemistry is not "metal vs nonmetal", but rather "main group vs transition", with the latter defined as groups 4 through 11 inclusive only and the former defined as everything else. Double sharp (talk) 04:42, 10 May 2017 (UTC)

A quick comment on the descriptive chemistry of group 5 is that, as I recall, P is always referred to as a nonmetal and Bi as a weak metal, and As and Sb are generally described as metalloids or sometimes the author resorts to hand waving and says they have "intermediate chemistry". A subsequent reading of the text of course shows that the chemistry of As and Sb is in fact generally nonmetallic. On inorganic chemistry I feel that the most fundamental distinction is between metals and nonmetals. Or at least that is what one learns first. I guess most of the chemistry of the TM, after all, involves nonmetals. Apologies for mistakes, am on my phone again. Sandbh (talk) 17:05, 10 May 2017 (UTC)

It's the trend that I think matters more: sure, P is very different from Bi, but you can see As and Sb as clear steps between them. It is kind of like Sc-Y-La writ large; there you start with a soft base and through an intermediate step you end up with a hard base. And all are quite distinguished from the TM via their common oxidation states being two apart instead of one apart, and despite the differences Greenwood and Earnshaw have no problem at all covering As, Sb, and Bi compounds in one chapter. Oh and there is no problem; I am also editing from my phone here. ^_^ Double sharp (talk) 00:58, 11 May 2017 (UTC)

I can't access G&E now but I remember they referred to As as a nonmetal or metalloid; and Sb as a metalloid; and I think they call Bi a metal. If they deal with the compounds of each of these elements in the same chapter then it makes me wonder about the relevance of their classifications. Now on my iPad :). Sandbh (talk) 01:46, 11 May 2017 (UTC)

Yeah, the G&E classification in the p-block keeps the period 4, 5, and 6 elements together in one chapter for groups IVB, VB, and VIB, even if it is cut in half by the metalloid line. R. Bruce King goes further in his Inorganic Chemistry of Main Group Elements, putting P-As-Sb-Bi as one chapter, and Si-Ge-Sn-Pb as one chapter, and O-S-Se-Te-Po(!) as one chapter. I get the feeling that the metallicity classification is more relevant to the physical properties of the pure elements; once one goes to the chemistry, the discontinuities get significantly smoothed over, so that even if Pb is a metal and Si is a semiconductor they are clearly part of the same group IVB from their tetravalence and reactions. Still on my phone. ^_-☆ Double sharp (talk) 02:04, 11 May 2017 (UTC)

An interesting article for you on delimiting the periodic table by groups (from the metallurgical perspective)

http://www.sciencedirect.com/science/article/pii/0001616079901044 Double sharp (talk) 14:45, 7 June 2017 (UTC)

Thank you. HEN Stone is an author of some repute and will be worth reading. Sandbh (talk) 00:35, 8 June 2017 (UTC)

You're welcome! I have found his classification quite inspiring and I can see why R. Bruce King adopted it for his Inorganic Chemistry of Main Group Elements (although I'm not sure how to relate it to general chemistry for the transition-element section). I've recognised that to support my position would require mounting a full-scale defence of the idea of treating metalloids as "stunted metals" (that is, treating boron for example as an overly small aluminium that is forced to be covalent when it would rather not); I've written the one for boron already at WT:ELEM#Boron, while the others (planned to go group-by-group) will still take some more time to organise my thoughts for. Double sharp (talk) 15:26, 12 June 2017 (UTC)

Complexes of the 4d and 5d transition metals

A while ago you gave me this quote from Parish's The metallic elements, p. 133:

"Study of the aquo-, hydroxo-, and oxo-complexes of the 4d- and 5d-metals is complicated by two factors. Firstly, when the metals are in relatively low oxidation states a great many ligands, including mostly simple anions, are able to displace water from the coordination sphere. In order to avoid hydrolysis it is usually necessary to work with acidic solutions, so that an excess of anions is present. The metal is usually found in anionic complexes which may or may not contain coordinated water, or hydroxo- or oxo-groups. Consequently, most redox-potential data refer to complexes rather than to aquated species, and the values vary with the anions present. In a few cases special care has been taken to use acids whose anions coordinate very weakly (ClO⁻
₄, NO⁻
₃), and simple aquated cations have been observed for Mo³⁺, Ru²⁺, Ru³⁺, Rh³⁺, Pd²⁺, Ag⁺, Au³⁺, Cd²⁺ and Hg²⁺. The second complicating factor is that, for the higher oxidation states in alkaline solution, polymeric species are formed which coexist over wide pH-ranges in complex series of slowly-attained equilibria. It is difficult to establish the nature of the species present and, as discussed…the species which can be isolated in solid salts may bear no relation to those present in solution."

Does he give a reason for this behaviour? Presumably there is something pretty obvious going on that I'm missing, which is why this effect is so localised. Either that or it really isn't obvious and that's why the 4d and 5d transition metals are shoved under the rug at high school. The tripositive cations Mo³⁺, Ru³⁺, and Rh³⁺ appear to have ionic radii in between the hard tripositive cations (Ln³⁺) and the borderline-soft ones like Fe³⁺; but alas, this great theory is scotched by Au³⁺, whose ionic radius is not that far from those of Yb³⁺ and Lu³⁺, and is surpassed by that of Bi³⁺. So I am somewhat at a loss here; Ge²⁺ appears to behave the same way as these transition metal cations, but including it as another member of this strange cluster does little to demystify it. Double sharp (talk) 15:14, 19 June 2017 (UTC)

I'll have a look at this a bit later today. On Ge²⁺ I mentioned that the evidence for the existence of such is rather contentious (I presume you saw my earlier comment about that) unless I've missed something in the more recent literature. On a related note I think you said something about "the average view" of categorisations or something like that and I was thinking about this concept earlier this morning. I'll see if I can post something about this a littler later on today, after attending to a few errands. Sandbh (talk) 23:21, 19 June 2017 (UTC)

It's fairly contentious, but from isostructural compounds of Cd^II and Sn^II I can believe that it exists in ionic lattices. Trying to make it in aqueous solution tends to result in other Ge^II complexes, at least according to metalloid, but this does not seem too far-removed from how Pd^II and Pt^II behave, for instance. So I can vaguely believe that it is a thing and we just haven't seen it yet; I find it much more difficult to believe for As³⁺, which is also sometimes referred to.

I might dare to say that the 3d transition metals are much more electropositive than the 4d and 5d metals. In the latter, I might only dare to claim electropositivity for Zr and Hf, and like Ti it only seems to show in the metallic state. Double sharp (talk) 23:38, 19 June 2017 (UTC)

Am looking at Parish now. Sandbh (talk) 23:41, 19 June 2017 (UTC)

Check your e-mail and let me know if you can read it. Sandbh (talk) 23:45, 19 June 2017 (UTC)

That will have to wait till later today when I get home, since I'm on my phone now (which is unfortunately not synced up to my Wikipedia-use email!). But thanks in advance! ^_-☆ Double sharp (talk) 03:32, 20 June 2017 (UTC)

Okay, I've got it. I'm reading the relevant pages now (and I should get that book ^_^). Double sharp (talk) 12:04, 20 June 2017 (UTC)

I can see nine copies going cheap on abe books (<$10 US inc. postage). Sandbh (talk) 12:10, 20 June 2017 (UTC)

Yes, broadly, the 3d metals appear to be more electropositive than the 4d metals which in turn are more electropositive than the 5d metals. Sandbh (talk) 00:53, 20 June 2017 (UTC)

Here's an attempt at an explanation: electropositivity goes up when the oxidation state goes down. The 3d metals from Mn onwards tend to only involve one or two d-electrons at the most in bonding in their common compounds (Fe³⁺, Cu²⁺) and very often none at all (Mn²⁺, Fe²⁺, Co²⁺, Ni²⁺). Even Cr³⁺ involves only two, though it is slightly easier to get all of them in play now. Whereas the 4d and 5d electrons are much more eager to play a role as valence electrons like the 5s and 6s ones. So I'm betting that this is once again because the 3d orbitals are small and too close to the electron core; indeed, 3p is close enough that they are almost totally drowned. (This is why the colours are stronger for complexes of the 3d metals than for the 4d and 5d metals; when ligands approach the former to bond with the 3d electrons, the repulsion from the core "stretches" the bond, and the poor overlapping of the 3d bands does the rest.) The result is lower oxidation states and higher electropositivity. I'll need to check this against Greenwood and Earnshaw later today, but if this is wrong, it should at least be non-trivially wrong. ^_^

The 4d shell appears to behave more similarly to the 5f shell, while 3d can usefully be compared to a more active 4f: it starts very much more active (Zr, Nb, Mo, Tc) before abruptly getting cold feet (Ru, Rh, Pd) and shrinking into the core almost one element too early at Ag. The 5d shell starts the same way thanks to the relativistic contraction of the 4f shell (Hf, Ta, W, Re), but then it gets to keep going because it is relativistically destabilised, so that even all those paired electrons are happy to join in the fun with Os, Ir, Pt, and Au, even if it isn't enough to coax them out for Hg! It is intriguing, then, that Pyykkö has speculated on the possibility of extremely high oxidation states like +12 in the analogous 6f row. (I really want to see how those elements would act, even more so than the 5g ones – although we should probably start with getting to know the 6d and 7p we already have better. ^_-☆) Double sharp (talk) 03:58, 20 June 2017 (UTC)

Having read his explanation it does not seem to be all that much different from what I just said...but I notice his bibliography has The Heavy Transition Elements which I actually can get ^_^ I'll be back! (And there is also the companion volume Complexes and First-Row Transition Elements.) ^_^ Double sharp (talk) 14:25, 20 June 2017 (UTC)

I know The Heavy Transition Elements but was somewhat disappointed with it. Sandbh (talk) 02:35, 21 June 2017 (UTC)

I am somewhat disappointed with both that and Parish, TBH; the latter at least gives reasons for the preference for high oxidation states (which are not too dissimilar to my speculation above, written before having read those pages you sent me), but what I was really looking for is why these aqua complexes like [Mo(H₂O)₆]³⁺ are so unstable, with so many ligands happily displacing water from the coordination sphere. I shall have to keep looking. Double sharp (talk) 03:09, 21 June 2017 (UTC)

I wonder: electronegative ligands should result in the s and d orbitals ending up with similar sizes, with the opposite effect as to the p-block elements. Hence in the d-block they should favour high oxidation states, and it is no wonder then that forcing them in low oxidation states like these aqua ions (with highly electronegative O as the electron pair donor) does not end well.

If this theory is correct, then the 4d and 5d metals in their low oxidation states should preferentially form complexes with poorly electronegative ligands for the most part. I wonder if this is true, but I am not currently home where I can check Greenwood and Earnshaw... Double sharp (talk) 03:33, 21 June 2017 (UTC)

Hmm, this actually seems fairly reasonable! Starting at group 5 (since Zr and Hf have little chemistry below the +4 state), Nb^III and Ta^III chemistry tends to involve S- and P-donor ligands. Most Mo^III and W^III complexes are with M≡M bonded species; among the known M₂X₆ compounds are X = NR₂, OR, CH₂SiMe₃, SAr, and SeAr – and they tend to be oxygen- and moisture-sensitive! Tc^III needs stabilisation by back-bonding ligands, and Re^III needs stabilisation from Re–Re bonding; Tc^II and Re^II are mostly confined to arsine and phosphine complexes. Ru^III is much more like Fe^III (I presume the stabilisation of the 4d orbitals after the half-filled shell is beginning to have more of an effect in the "normal" and non-relativistic 4d shell?), while Os^III again tends to prefer forming Os≡Os bonds. By the time we get to Rh^III and Ir^III, both elements are fairly similar to Co^III, while Rh^II and Ir^II either need metal–metal bonding or stabilisation by ligands like phosphines or C
₆Cl⁻
₅. Of Pd^II and Pt^II, GReenwood and Earnshaw (p. 1161) write, "Not many complexes are formed with O-donor ligands" (somehow listing the tetraaqua ions as among the important ones – go figure); these are essentially class-b cations anyway. For Au, of course Au–Au bonding with the metal in an average oxidation state below +1 is prolific. Au markedly prefers "the heavier donor atoms P, As and S" according to Greenwood and Earnshaw (p. 1196); and given that this is a survey of low oxidation states, I am content to consider Ag as a post-transition metal at heart.

I also wonder if the position of H₂O low down on the spectrochemical series has something to do with it; [Pt(H₂O)₆]²⁺ might not be a thing, but [Pt(NH₃)₆]²⁺ certainly is. Double sharp (talk) 15:28, 21 June 2017 (UTC)

Some comments on your sandbox rewrite of that paragraph in the N article

Given your signature high level of scholarship, I advance this with the greatest caution, but I do believe you may be going a bit too far in accentuating the similarities with carbon. Here are four things I find unsatisfying:

Thank you! I'm not sure if "accentuating" was the right word. I feel that "noting" would be closer to what I was attempting. But if it came over to you as "accentuating" then I may have some more editing to do.

Covalency

"Like carbon, nitrogen is limited to a maximum covalency of four." – yes, but that is only because it's in the second period. The difference is more with Si and P which can go much further (e.g. SiF²⁻
₆, PCl⁻
₆). The way nitrogen achieves four-coordination in things like NR⁺
₄ is by a combination of electron loss and bond formation (5 outer electrons minus 1 plus 4 single bonds makes 8), and certainly not while remaining neutral like C does; it is more like what O and F do in H₃O⁺ and H₂F⁺. Furthermore, oxygen can also reach four stably in oxatriquinane salts, and in superacidic solutions it really does become reasonable to talk about species like Me₄O²⁺ and even H₄O²⁺(!). The natural comparison, if you insist on bringing carbon into it, would be in carbonium ions. Then carbon has a maximum covalency of five in a unipositive cation, continuing the trend from nitrogen's four, oxygen's three, and fluorine's two; and it is not surprising that C is much less happy with it than N, O, or F.

Double sharp and Sandbh, by Rembrandt van Rijn (1628)

That C, like N, is limited to a maximum covalency of four is a statement made by Jones (1973, p. 162).^ I didn't know O could reach four in oxatriquinane salts, so thanks for bringing this fascinating development to my attention. I had a look for some information about such salts, rather than the ion, but haven't found anything yet. My impression so far is that this is chemistry at the extremes, rather than e.g. mundane ammonium chemistry in which N has a covalency of four, so I'm not sure that the comparison is on a level playing field. Similarly, what happens in superacid chemistry is a little less relevant, I think, since we tend to compare things in standard conditions.

The existence of the carbonium ion calls Jones' assertion into question but it is probably a fair enough statement for the mundane chemistry of carbon and nitrogen. I'll look more closely at the wording of this paragraph; it may start sprouting some footnotes.

^ Jones K 1973, "Nitrogen", in JC Bailar et al., Comprehensive inorganic chemistry, vol. 2, Pergamon Press, Oxford, pp. 147–388

Given that stable(!) pentacoordinate carbon can really be prepared, perhaps it is not quite as irrelevant as it might seem. ^_^ Double sharp (talk) 03:54, 26 June 2017 (UTC)

Perchance I happened upon a reference to "extremely stable" pentacoordinate nitrogen having been prepared in 1990. Hexacoordinate carbon was prepared in 1989 but the salt concerned, C₆(CH₃)₆(SbF₆)2•HSO₃F, is very unstable and easily decomposes, although single crystals could be handled in a cold nitrogen stream. Remarkably, it has been predicted that N, in common with B and C, should be able to form up to six(!) bonds with Au. I'm not yet sure as to what to make of all of this. I see from the article in question that even the fabled NF₅ is supposed to be able to exist. Sandbh (talk) 01:04, 28 June 2017 (UTC)

Incidentally, here are the doi's of two papers on H₄O²⁺: 10.1021/ja00265a031 and 10.1016/0009-2614(87)80490-6. There is also apparently H₃F²⁺. Here's another paper on these highly protonated species that should make most high-school chemistry students' eyes bleed ^_^ (and here is yet a fourth one). Double sharp (talk) 14:38, 29 June 2017 (UTC)

Nothing new, just some words I stumbled upon when looking at how the elements are classified: It's from Descriptive inorganic chemistry (2006, p. 29) by Rayner-Canham & Overton: "For chemists, however, the most important feature of an element is its pattern of chemical behaviour, in particular, its tendency toward covalent bond formation or its preference for cation formation." Sandbh (talk) 07:19, 12 July 2017 (UTC)

Hybridisation analogous to C

"In its compounds, particularly in biochemistry, it is capable of forming hybrid orbitals analogous to those seen in carbon. Examples include ammonia and amines (sp3 tetrahedral); histidine, purines, and pyrimidines (sp2 planar); and nitrogen gas and cyanide (sp linear)." – again, so can O and F. What exactly is H₂O if not sp³ tetrahedral hybridisation anyway? The only reason this is not so obvious is because more of the substituents are lone pairs. (Unfortunately the misconception that hybridisation is limited to carbon seems rather widespread.) Not that this is useful, because again the bonding as far down as PbEt₄ can be adequately described by hybridisation. It is a very good theory for a reason.

The full quote from Beckman et al.^ was, "In most organic molecules nitrogen readily forms hybrid orbitals directly analogous with carbon, forming, the familiar sp3 (with a pyramidal shape) as in ammonia and amines, sp2 (planar) found in histidine, purines, and pyramindines, and sp1 (linear) found in nitrogen gas (N2) and cyanide (HCN) This brand of nitrogen chemistry is familiar to biochemists because it resembles the chemistry of carbon." They made that comment in the context of considering where N was positioned i.e. between C and O. They don't have anything much to say about the comparison of nitrogen with oxygen. I'll look again at this paragraph, in light of your observations. At this point I'm re-reminded of Jones' observation (1973, p. 162) that the properties of N that he discussed are about what you'd expect for an element between C and O.

^ Beckman JS, Beckman TW, Chen J, Marshall PA & Freeman BA 1990, "The physiological and pathological chemistry of nitric oxide", in J Lancaster (ed.), Nitric oxide: principles and actions, Academic Press, San Diego, pp. 1–82

Update: I read some more, including about C, N, and O stereochemistry.^ Curiously, Cotton et al.^^ say that most C compounds are best regarded as organic chemicals, and that N forms an exceedingly large number of compounds, most of which are to be considered organic rather than inorganic. When it comes to O they say most inorganic chemistry is concerned with O compounds, if only in the sense that so much chemistry involves water. In this context, my guess now is that as C is the backbone of life and organic chemistry then, from a biochemist's perspective, Beckman et al. were observing that N stereochemistry, and presumably the nature of most of its compounds, would/should be familiar to the biochemist. Even so, I don't know if there is much in this and will look again at its relevance to my proposed edit to the N article.

^ Siekierski SC & Burgess J 2002, Concise chemistry of the elements, Horwood Publishing, Oxford, passim

^^ Cotton FA, Wilkinson G, Murillo CA & Bochmann M 1999, Advanced inorganic chemistry, 6th ed., John Wiley & Sons, New York, passim

It is not as if oxygen is rare in organic compounds; the carbonyl group and its derivatives alone would assure the opposite. All that means is that inorganic oxygen compounds are also pretty important, perhaps more so than inorganic nitrogen compounds. (Although NH₃ might give H₂O and SiO₂ a run for the money.) The stereochemistry of N in covalent compounds is not that different from that of C or O, anyway, and the perceived similarity of N to C over O is more that with one less electron than oxygen, nitrogen needs to form one more bond to be happy, and hence there is still a faint echo of the richness of carbon's stereochemistry around. Double sharp (talk) 03:57, 26 June 2017 (UTC)

Catenation

"Its proclivity for catenation is less than that of carbon but more than that of oxygen. The longest chain of nitrogen yet synthesised has eleven atoms (C is unlimited; O has an effective limit of three)." Except that catenation in N beyond the azides is really not the main thing in N chemistry, unless you like violently endothermic compounds like this one, so I don't see why N shouldn't be considered much different. In terms of "normal" chemistry, N and O are both effectively three for the most part; if we're talking all compounds, even the unbottlable ones, O should be at least five. I could just as well say that O is different from the halogens in that it can catenate at all, but that would just be because of an accident of electron configuration, which informs but does not fully dictate chemical behaviour. Anyway these differences between N (11), O (3 or 5), and F (2) are all small potatoes when you compare it to C (>6000 just in molecules).

Here I was trying to take an NPOV and stating the situation simply as it is (I read a reference saying nearly the same thing i.e. that N catenates more than O but less than C, without any value jugements, but I'm not sure if I noted the details; even G & E only note that catenation in N is more limited than for C^). The eleven atom N entity is relatively stable as far as I recall (the article said it has a relatively higher thermal stability and much lower sensitivity than N₈ and N₁₀ compounds); anything for O > 3, is not. I feel you are right about N and O being small potatoes compared to C. So, let's see if we can get some better wording here.

^ Greenwood NN & Earnshaw A 1998, Chemistry of the elements, 2nd ed., Butterworth-Heinemann, Oxford, p. 416

Update: I had a look in Cotton et al.^ They write that unlike C, N has little tendency to catenate (up to 8). For O they write that, as with N, catenation occurs only to a limited extent: up to 3, with the exception of four in the (unstable) O₄F₂. So, yes, I agree with your small potatoes analogy.

^ Cotton FA, Wilkinson G, Murillo CA & Bochmann M 1999, Advanced inorganic chemistry, 6th ed., John Wiley & Sons, New York, passim

Electronegativity, H-bonding and coord chem

"Nitrogen resembles oxygen in its capacity to form hydrogen bonds and coordination complexes by donating its lone pair of electrons. Its high electronegativity, while comparable to that of oxygen, has been described as "misleadingly high" (Phillips & Williams 1965, p. 609) on account of nitrogen's negative electron affinity." Not unless the electronegativities of Kr and Xe are also "misleadingly high". I advance this with the greatest caution, but I think Phillips & Williams are misunderstanding what electronegativity is really about. It is the ability for an atom to attract a pair of electrons that are already bonded to it, not that to attract a random floating electron that is just passing by. As a result it is no surprise that the electronegativity of N looks misleadingly high when you compare it irrelevantly to its electron affinity, but it looks spot-on when you consider the rich coordination chemistry of nitrogen and its capacity for hydrogen bonding. In fact, I would dare to say that it is the electron affinity which is misleading here, neglecting that once the electron is already bonded, electron affinity is not relevant, and all that matters is how close the nucleus is and how much effective nuclear charge the electron feels.

Double sharp (talk) 13:58, 24 June 2017 (UTC)

Nonmetal groupings (Synder 1966)

Electronegativity
High	O, F, Cl, Br, I
Intermediate	C, N, S, Se
Low	H, P

Ahh, the line about N's low EA was my quick interpretation written as I was half watching some TV, not that of Phillips & Williams! So I need to hunt down what is going on here. I have another reference discussing the validity of EN as a concept. It notes that bond strengths of the elements with one another are a pretty good indicator of EN.^ In my sandbox there is that table from Snyder,^^ who took the same approach, rating the elements according to their bond strengths with F, O, and Cl. The result is largely as expected, but again note how N does not live up to its EN. It doesn't deliver full value for money.

^ Williams AF 1979, A theoretical approach to inorganic chemistry, Springer-Verlag, Berlin, p. 186

^^ Synder MK 1966, Chemistry: Structure and reactions, Holt, Rinehart and Winston, New York, pp. 235–241, 242–243

Update: Massey (2000, p. 267)^ says, "It is possible for the Group 15 elements to achieve a rare gas electron configuration by accepting three electrons to form M^3– anions…this process is not very energetically favourable and, owing to strong inter electron repulsions, the formation of N^3– requires a huge 2130 kJ mol^–1…Electrons have more space on P, which lowers their mutual repulsion and results in the formation of P^3– requiring only about 1450kJ." So, basically, it seems to me that I was correct with my reference to electron affinity, albeit I could've worded it better.

^ Massey AG 2000, Main group chemistry, John Wiley & Sons, Chichester

Back to original commentary: I acknowledge that the capacity of N to engage in hydrogen bonding and coordination chemistry is attributed to its high electronegativity value. At the same time I note that C, P, and S are also capable of acting as ligands (Greenwood & Earnshaw described this aspect of P chemistry as "burgeoning");^ and that H-bonds formed by S, like those formed by N, are normally also considered to be strong, despite the lower electronegativity of S.^^

^ Greenwood NN & Earnshaw A 1998, Chemistry of the elements, 2nd ed., Butterworth-Heinemann, Oxford, pp. 485, 924, 665

^^ Gilli G & Gilli P 2009, The nature of the hydrogen bond: Outline of a comprehensive hydrogen bond theory, Oxford University Press, Oxford, pp. 29, 31

Further to S forming strong H-bonds, I was reminded that Greenwood & Earnshaw^ wrote that, "Nitrogen and sulfur are diagonally related…and might therefore be expected to have similar charge densities for similar coordination numbers…Likewise they have similar [sic] electronegatitivites (N 3.0, S 2.5) and these become even more similar when additional electron-withdrawing groups are bonded to the S atoms."

^ Greenwood NN & Earnshaw A 1998, Chemistry of the elements, 2nd ed., Butterworth-Heinemann, Oxford, p. 722

Yet the main difference is still between N- and O-donor ligands and P- and S-donor ligands a row further down: p. 665 of Greenwood and Earnshaw itself says "Ligands in which S acts as a donor atom are usually classified as class-b ligands ("soft" Lewis bases), in contrast to oxygen donor-atom ligands which tend to be class-a or hard (p. 909). The larger size of the S atom and the consequent greater deformability of its electron cloud give a qualitative rationalization of this difference and the possible participation of d_π orbitals in bonding to sulfur has also been invoked (see comparison of N and P, p. 416)." So once again it is the greater electronegativity of N and O vs. P and S that count and lead to distinct qualitative differences. I note that S needs a lot of electron-withdrawing groups to get anywhere near N, which is not surprising as electronegativity goes up together with the partial positive charge on the atom (which is why elements are more electronegative in higher oxidation states).

Regarding sulfur's ability to hydrogen bond, Greenwood and Earnshaw says on p. 682: "...comparison with the properties of water (p. 623) shows the absence of any appreciable H bonding in H₂S." Of course, hydrogen bonding is not quite as simple as high-school chemistry would have it, and certainly C, P, S, Cl, Br, and I can all do it, but only weakly (Greenwood and Earnshaw p. 53).

The comparison of N³⁻ and P³⁻ appears to miss something: are there actually any salt-like phosphides? Even with the most electropositive metals (Li, Na, Be, Mg, Zn, Cd, La, Ce, Th) the phosphides only have a little ionic character and there is a lot of metallic or covalent interaction (Greenwood and Earnshaw, p. 441). So even though they act as if they were really ionic phosphides, being hydrolysed by water or dilute acids to form phosphine, the bonding doesn't really cooperate. So here we see the difference: among the pnictogens, only nitrogen can handle forming a true trinegative ion (e.g. the alkaline earth nitrides), and that only because the nucleus is really close and can hold on to so many extra electrons tightly. So in a roundabout way we seem to have come back to nitrogen just being that much more electronegative than phosphorus. Certainly nitrogen suffers a little because of the stronger interelectronic repulsions that also result, but it seems that the gains of being small for electronegativity far outweigh the losses. Double sharp (talk) 03:46, 26 June 2017 (UTC)

Hi Double sharp, I owe you a more considered response. Off the cuff, I'd say that the electronegativity of N in the broadest sense (i.e rather than its assigned Pauling EN), is effectively greater than that of S, but it's a matter of degree not of kind. I'll ponder all of this some more. Sandbh (talk) 06:44, 26 June 2017 (UTC)

No, problem: take your time and I'll hold off on replying until that considered response comes. ^_^ Double sharp (talk) 06:47, 26 June 2017 (UTC)

LOL! Sandbh (talk) 06:52, 26 June 2017 (UTC)

Considered response to EN, H-bonding and coord chem

Ligands, and overlaps

Double sharp, yes, I agree there is a difference between N- and O-donor ligands and P- and S-donor ligands but I'm not too concerned about this. I felt it was enough to observe that C, P, and S are also capable of acting as ligands. If we drill down into the detail there will always be overlaps. For example, as I said earlier, what is an amphoteric metal like Be doing hiding among a bunch of supposedly alkaline metals? Hoping no one will notice? ^_^ What is S doing among the polyatomic nonmetals when the majority of its properties are more like that of a diatomic nonmetal? The chemistry of Ag is more like that of a transition metal so why do we categorise it as a transition metal? Et cetera. Sure, we should seek to minimise these anomalies but we should do this pragmatically rather than myopically—map the lay of the land, rather than be absorbed by the allure of its nooks and crannies. Sandbh (talk) 12:31, 28 June 2017 (UTC)

I would nevertheless argue that these examples all illustrate something important: because of trends, every category has one not-particularly-impressive member. Be is less than impressive as a metal, but extrapolating towards smaller and smaller sizes from Mg and Ca it fits rather well. Similarly, Ag is kind of what you would expect from the contraction of the 4d orbitals towards the end of that period (and even then, there are still Ag^II and Ag^III compounds if you get electronegative enough ligands, like the omnipresent fluoride). The fact that we include Be as an alkaline earth metal and Ag as a transition metal is because they fit better as slightly extreme members of those categories than as very extreme members of the others. Be would be a very strange post-transition metal, given that its valence electrons are very well-shielded; to some extent Al is in this situation too, hence the questions about it. Ag would also be a fairly strange post-transition metal: while it is true that Cu^I, Ag^I, and Au^I all act like what you would expect for group IB instead of group VIII in Mendeleev's scheme, the only reason why Ag looks any less like a transition metal is because the +1 state is more important for it than for Cu or Au. So I would like to suggest that N is more usefully regarded as a slightly tongue-tied corrosive nonmetal than a particularly active intermediate one. Double sharp (talk) 15:11, 28 June 2017 (UTC)

You seem to be agreeing with me i.e. "map the lay of the land, rather than be absorbed by the allure of its nooks and crannies"? Sandbh (talk) 06:49, 1 July 2017 (UTC)

I agree with that statement; I think the place we disagree is what exactly the lay is. I think I've often previously mentioned the idea of having a brief "first guess" of what an element should be like from the periodic table, before supplementing it with second-order course corrections, and the reason why I am happier to keep Ag as a transition metal and N among the stronger nonmetals (despite the fact that neither really delivers completely on the characteristic qualities without some cajoling, particularly from fluorine) is that I think the reasons why neither are particularly good fits for their categories are easier to conceive as second-order course corrections. Double sharp (talk) 14:21, 1 July 2017 (UTC)

Am I right in thinking we only have a substantial gentleman's disagreement about whether N is better classed as an intermediate nonmetal or as a corrosive nonmetal? Sandbh (talk) 02:21, 2 July 2017 (UTC)

Yes, of course! ^_^ Double sharp (talk) 03:09, 2 July 2017 (UTC)

Have I understood your first guess thought process correctly: (1) Brief "first guess" of what an element should be like from the periodic table; (2) If what the element is actually like doesn't live up to the first guess, look for "second-order" course corrections to justify the first guess? Sandbh (talk) 07:34, 2 July 2017 (UTC)

Something like that, though I would speak more of "if some properties of the element" rather than "what the element is actually like". Double sharp (talk) 09:34, 2 July 2017 (UTC)

I think I'm OK, for now, with this part of our discussion. I'll revisit if needs be. I'll head on over to the Electronegativity of N vis-à-vis its overall chemistry part, and pick up things there re sideshows. Sandbh (talk) 12:43, 2 July 2017 (UTC)

S and H-bonding

Further to S and H-bonds I had a closer look at Gilli & Gilli. S is apparently included among those elements that can form the strongest H-bonds, these being N, O, S, F, Cl, Br. That said, I do accept that S H-bonds are generally weaker than those of N.

On this topic, did you really mean to say (way back in April!), that, "once you do actual chemistry with it [N], instead of leaving it to pretend to be a noble gas and sip tea, the similarities with oxygen and fluorine seem more striking, with hydrogen bonding playing a minor role [italics added]"...? Sandbh (talk) 12:31, 28 June 2017 (UTC)

No, of course not: I meant "major role". And regarding the strong H-bond formers, I was under the impression that the vast majority of the literature would place N, O, and F way above all the others. Perhaps S, Cl, and Br lead the pack among the others but a short glance at H₂S vs NH₃ (and the latter has less hydrogen bonds per molecule!) confirms that S is just not in nitrogen's league here. At the very most it needs extreme conditions to approach nitrogen (I wonder if they similarly require strongly electron-withdrawing groups attached to the sulfur?). Double sharp (talk) 15:11, 28 June 2017 (UTC)

Here are some more illustrative quotes from the literature on the strength and importance of S H-bonding:

1. "The incorporation of thioamides within the backbone of linear peptides can produce subtle physical changes...thioamides were introduced...to yield the appropriate thiopeptides...NMR analysis...suggested that the γ-turn intramolecular hydrogen bonds for both molecules were generally weaker...than the γ-turn H-bond of the parent, while similar data suggested the β-turn H-bonds were at least as strong as in the all-amide parent."

2. "In summary…there are strong indications that C=S is a stronger hydrogen bonding acceptor than is C=O."

3. "…the thioamide can be protonated more effectively than the corresponding amide."

4. "…hydrogen bonding is important wherever hydrogen is covalently bonded to such highly electronegative atoms as nitrogen, oxygen, sulfur, or a halogen."

5. "Hydrogen bonding is generally thought to play an important role in tuning the electronic structure and reactivity of metal-sulfur sites in proteins."

¹ Sherman DB & Spatola AF 1990, "Compatibility of thioamides with reverse turn features: Synthesis and conformational analysis of two model cyclic pseudopeptides containing thioamides as backbone modifications", Journal of the American Chemical Society, vol. 112, no. 1, pp. 433–441 (433)

² Shaw RA, Kollát E, Hollósi M & Mantsch HH 1995, "Hydrogen bonding and isomerization in thioamide peptide derivatives", Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, vol. 51, no. 8, pp. 1399-1412 (1411)

³ Min BK, Lee H-J, Choi YS, Park J, Yoon C-J, Yu J_A 1998, "A comparative study on the hydrogen bonding ability of amide and thioamide using near IR spectroscopy", Journal of Molecular Structure, vol. 471, nos 1–3, pp. 283–288 (287)

⁴ Fox MA & Whitsell JK 2004, Organic chemistry, 3rd ed., Jones and Bartlett Publishers, Boston, p. 95

⁵ Dey A 2007, Nature of iron-sulfur bonds in electron transfer and catalytic active sites: Contribution to reactivity and the role of hydrogen bonding, Ph. D. thesis, Stanford University, Stanford, p. ix

While, as noted, I do accept that S H-bonds are generally weaker than those of N, the significance and strength of H-bonds is not to be downplayed. Sandbh (talk) 04:26, 1 July 2017 (UTC)

The case of thioamides, while no doubt quite spectacular, is essentially what I was suspecting: the lone pair on the N atom is significantly delocalised into the C–N bond, forming a partial double bond between it and the thiocarbonyl carbon. In fact, this happens to an even greater extent than for amides (with oxygen instead of sulfur). So the sulfur mostly ends up approximately single-bonded instead of double-bonded to the carbon, and thus is electron-deficient: perfect conditions to raise its electronegativity. So I don't see a reason to conclude for this that hydrogen bonds to sulfur are usually strong: I would instead conclude that they are strong only when sulfur is in a very electron-poor situation and is more electron-hungry than normal. It appears that only in exceptional circumstances like these does S start to rival N, and no doubt the same is true for Cl and Br. Double sharp (talk) 14:36, 1 July 2017 (UTC)

That's fine. I don't think there is anything more we need to discuss here. Sandbh (talk) 02:26, 2 July 2017 (UTC)

Anionic P, and N

On anionic P, Wiberg says there are many such phosphides (p. 835). Having said that, I noted earlier that the phosphides resemble in many ways the metal borides, carbides, and nitrides (Greenwood & Earnshaw 1998, pp. 490). Wiberg appears to round this out, observing that even the alkali and alkaline earth nitrides show considerable covalent character (p. 602). Sandbh (talk) 12:31, 28 June 2017 (UTC)

Nevertheless Greenwood & Earnshaw include a category of salt-like nitrides, while they do not do so for the phosphides. At the very most they mention on p. 491 that while there is some ionic character in compounds like Li₃P there is significant metallic or covalent interaction. If we cannot even expect simple ionicity from lithium, which works better for Li₃N, what hope is there? There do seem to be mostly ionic nitrides and even carbides (Be₂C, really!), but according to Greenwood and Earnshaw these do not appear to exist for borides and phosphides. Double sharp (talk) 15:11, 28 June 2017 (UTC)

G & E refer to the nitrides in question as "salt-like" (p. 417), which is significant. The "-like" suffix implies that these compounds are not true salts. The quotes around salt-like suggests that even "salt-like", in the context of nitrides, has its shortcomings. They add that, "It is possible to write ionic formulations of these compounds [the nitrides] using the species N₃^– though charge separation is unlikely to be complete" (p. 417).

Here are some other quotes from the literature:

• "Very little of the chemistry of the group 15 elements is that of simple ions. Although nitrides and phosphides that react with water are usually considered to contain N₃^– and P₃^– ions, electrostatic considerations make it doubtful whether these ionic formulation are correct."^
• "The nitrides and phosphides of the Group IA and IIA metals contains anions of high charge, which behave as strong bases. Therefore, they abstract protons from a variety of proton donors. The following reactions are typical:

Na₃P + 3 H₂O → 3 NaOH + PH₃

Mg₃N₂ + 6 H₂0 → 3 Mg(OH)₂ + 2 NH₃

Li₃N + 3 ROH → 3 LiOH + NH₃ ^^

^ Housecroft CE & Sharpe AG 2008, Inorganic chemistry, 3rd ed., Pearson Education Limited, Harlow, p. 433

^^ House JE & House KA 2015, Descriptive inorganic chemistry, 3rd ed., Elsevier, Amsterdam, p. 131

Overall, I expect it would probably be reasonable to speculate that the phosphides have smaller fraction of ionic binding than the nitrides.

This is all moot, in any event: As I noted earlier, "Very little of the chemistry of the group 15 elements in that of simple ions…nearly all the chemistry of the group…involves covalently bonded compounds."^ and, "Phosphides resemble in many ways the metal borides, carbides and nitrides…"^

^ Housecroft & Sharpe, p. 433, op. cit.

^^ Greenwood & Earnshaw, p. 490, op cit.

-- Sandbh (talk) 02:14, 1 July 2017 (UTC)

I would bear in mind of course that the ionic-bonding model can look very successful, even when being applied to nonsensical cases: you can reproduce the enthalpies of formation of BF₃, SiF₄, PF₅, and SF₆ quite reasonably by treating them as ionic, and you can reproduce the lattice energy of Li metal quite correctly if you treat its structure as Li⁺Li⁻ (Greenwood and Earnshaw, pp. 79–80). There is of course a transition towards metallic bonding along the series NaCl, Na₂O, Na₂S, Na₃P, Na₃As, Na₃Sb, Na₃Bi, Na (Greenwood and Earnshaw, p. 81). Furthermore the electronegativity that we love as our lifeline to predictions is changed not only by oxidation state but also by coordination (the famous high-school example being NH₃ and HCl reacting to give [NH₄]⁺[Cl]⁻). We simply need to say what we expect an ionic compound to act like and see whether the nitrides have a majority of these properties. While the vast majority of them do not, it seems to me that at least things like Ca₃N₂ do.

The behaviour of phosphides as if "P³⁻" anions were really there is a little inconclusive, as we know fully well how well wrong theories in chemistry can work. Nobody believes that d-orbitals are involved in hypervalent compounds anymore, but the theory is still a great predictor; we still talk about "H⁺" or "solvated protons" even though they are nothing of the sort, because thinking of them as a sort of aquo complex of H⁺ (perhaps trying to dress up as her self-proclaimed big sister Li⁺ ^_^) gives the right answers, albeit for the wrong reasons. I don't doubt that phosphorus behaves as though there were true trinegative phosphide anions there, but I'm not sure this proves that they were there; you'd need to look at the original compound, and while N³⁻ is perhaps a thing bonded to the most electropositive metals like Li, P³⁻ doesn't seem to be a thing even then.

It's worth noting that you could say much the same about the oxides: when Na₂O dissolves in water, the O²⁻ ions are such strong bases that they instantly abstract a proton from water and become OH⁻. This, of course, adds to the OH⁻ they made by abstracting a proton from water, so this is a slightly degenerate case, but the result is the same. Some of the carbides can similarly be classified as "methanides", "acetylides", and "propynides", but I'm not any more inclined to believe in the existence of a C⁴⁻ precursor than for the carbides. What clinches it for me for the nitrides is ionic conductivity by lithium ions in Li₃N – which seems to be clear proof that unlike C and P, nitrogen is just electronegative enough that despite the fact that it needs three electrons to get to the very stable Ne configuration, it can still form clearly ionic compounds with metals like Li on the extreme left of the periodic table. I don't see similar proof for any of the carbides and phosphides, at least. Double sharp (talk) 15:23, 1 July 2017 (UTC)

That's fine. I've previously said that the electronegativity of N in the broadest sense (i.e rather than its assigned Pauling EN), is effectively greater than that of S, but it's a matter of degree not of kind. You've agreed that the vast majority of nitrides aren't ionic compounds. I don't think there's anything more we need to discuss here. Sandbh (talk) 02:31, 2 July 2017 (UTC)

Electronegativity of N vis-à-vis its overall chemistry

I agree that N can be regarded as being more electronegative than S in some aspects.

When I then look at the overall chemistry of N and its compounds I remain of the view that the resemblance is more with rest of the nonmetals rather than the corrosive nonmetals. I base this on the following considerations about N (you've heard these already):

• its reluctance to form simple anions;
• the covalency of nearly all of its compounds;
• the kinetic restraints that are a hallmark of its chemistry and that of its compounds;
• that the average oxidising power of it and its species, in aqueous solution, is less than that of both iodine and of sulfur;
• that nonmetal displacement series list iodine and sulfur ahead of nitrogen; and
• that nitrides resemble the metal borides, carbides and phosphides in many ways.

You would think that with an electronegativity of 3.04, N would be a top-shelf candidate in terms of non-metallic character but it acts as if it has one hand tied behind its back. It delivers in some aspects but there are too few of these to constitute a pivotal mass of class-distinguishing features. By default, they become anomalies or boundary overlap points. Sandbh (talk) 12:31, 28 June 2017 (UTC)

I think the reason I remain uncomfortable with explaining nitrogen away as an intermediate nonmetal that sometimes rises above itself is that you would then expect it to act more like carbon or phosphorus, which it tends not to. Physically it is clearly much more allied to oxygen in its physical properties, for example. Thinking of N as a "kinetically hindered strong nonmetal" seems to be closer to what is actually going on. When nitrogen is being a good oxidising agent, i.e. in a high oxidation state, it needs to be bonded to more electronegative atoms, and the problem is that O and F are about the same size as N. The result is that these N species act kind of like the sterically hindered perchlorates: they show up better electrochemically than kinetically. And then you note that you would simultaneously expect a very strong N≡N bond. It seems to be easier to get at the true nature of nitrogen by starting with a "strong nonmetal" template and correcting it, than starting with a template like C or P and correcting it to strongness: we would then be hard put to explain where all that electronegativity is suddenly coming from in such common compounds like NH₃, where the H ligand is not particularly electron-withdrawing. Double sharp (talk) 15:11, 28 June 2017 (UTC)

It's good to know the suspected reason for your discomfort.

It may be helpful to reconsider a couple of your comments.

You wrote, "I think the reason I remain uncomfortable with explaining nitrogen away as an intermediate nonmetal that sometimes rises above itself is that you would then expect it to act more like carbon or phosphorus, which it tends not to." [italics added]

I listed a dot-point overview of the chemistry of N and its compounds, above, and it seems to me that N in fact very much tends to act like C or P or the rest of the intermediate nonmetals.

Earlier, in part 1, you appeared to be agreeing with me i.e. "map the lay of the land, rather than be absorbed by the allure of its nooks and crannies."

Accordingly, I've been advocating for the assignment of N to the intermediate nonmetals on something like an 80/20 basis i.e. the 80% or so of its behaviour that it has in common with the intermediate nonmetals, consistent with the dot-point overview.

OTOH you seem to have been advocating for the assignment of N to the corrosive nonmetals on something like a 20/80 basis i.e. the 20% or so of its behaviour that overlaps into the corrosive nonmetals category.

At this point it'd be good to hear further from you on the above observations, especially the apparent contradiction between paras 3 and 4 re how N acts. Sandbh (talk) 07:12, 1 July 2017 (UTC)

If N was really acting like C and P, I would expect to see (1) a giant polymeric structure, (2) very muted participation in hydrogen bonding, (3) the nonexistence of true salts, even if we go all the way and use something like barium as the cation, and (4) reduction potentials near zero for its most common species in aqueous solution. Instead we see none of these. Nitrogen's physical properties as one of the diatomic nonmetals are on a par with those of O₂; in its oxoanion chemistry it is quite on a level with Cl, comparing NO⁻
₃ with ClO⁻
₄, both thermodynamically strong oxidising agents which are inhibited kinetically – something you would expect for an atom as small as nitrogen. So it seems to me that the more weak aspects of nitrogen's nonmetallicity tend to be sideshows that result from its small size and multiple bonding in its normal allotrope. Double sharp (talk) 14:46, 1 July 2017 (UTC)

Double sharp, are you saying that the following properties—"the more weak aspects of nitrogen's nonmetallicity"—are sideshows(?):

1. its reluctance to form simple anions;
2. the covalency of nearly all of its compounds;
3. the kinetic restraints that are a hallmark of its chemistry and that of its compounds;
4. that the average oxidising power of it and its species, in aqueous solution, is less than that of both iodine and of sulfur;
5. that nonmetal displacement series list iodine and sulfur ahead of nitrogen; and
6. that nitrides resemble the metal borides, carbides and phosphides in many ways.

-- Sandbh (talk) 03:20, 2 July 2017 (UTC)

(I hope you don't mind that I'm taking the liberty of numbering your points to easily refer to them!)

Yes, because all of them are readily explained by its position in the periodic table. (1) is explained by it being rather far away from the noble-gas configuration (and it seems to me that the fact that it can form them at all is a qualitative difference already from P, As, Sb, and Bi), and (2) follows as a corollary (though I'm not sure if it is quite true because most nitrides are actually interstitial – again a consequence of its small size). This emphasis on interstitial compounds is part of (6) but the fact that N can actually form ions at all (there are quite a number of nitrido complexes, too) again seems to be qualitatively different from B, C, and P. (3) is common to most small atoms trying to be surrounded by lots of O and F ligands to raise their oxidation states, and (4) and (5) follow as a consequence. I am not sure how one would be able to go the other path and rationalise N's strongness in significant aspects (e.g. H-bonding) if one started from the point of view that it was a weak nonmetal at heart. Double sharp (talk) 06:49, 2 July 2017 (UTC)

Hmm. On the one hand the brief "first guess" from the periodic table is that N should be like a strong metal. On the other hand N has a whole swag of characteristics like an intermediate nonmetal which are readily explained by its position in the periodic table. I don't see the logic here. Sandbh (talk) 11:57, 3 July 2017 (UTC)

Oh dear, this is another case of the mild self-destructiveness of the periodic law, isn't it? ^_^

What I mean is that you would expect N from its position to prefer forming diatomic N≡N, kind of like oxygen next to it. Then you would conclude it to be likely a strongly electronegative nonmetal because it is small and the nucleus is very close. In both of these, N delivers completely (I've already mentioned lots of times why I think claims that N's electronegativity is "misleading" high are missing the point of electronegativity). From the first you would perhaps expect a bit of a kinetic barrier towards its reactions as well, and while the small size helps the electronegativity it makes it a bit slow to react as an oxidising agent when surrounded by O and F atoms. But the important property of electronegativity has already been given by the periodic law, and while N compounds may be slow to react, when they finally gets around to doing so they certainly act like strong oxidising agents. Double sharp (talk) 14:45, 3 July 2017 (UTC)

Yes, I think you are wisely in the right ball-park re the self-destructiveness of the periodic law.

Further to self-destructiveness, I said earlier (with italics added)

…I'm not so sure that I would describe the irregularities and anomalies in the periodic table as a concealed self-destructive assault on the principles that the PT was founded on, since these were loose in the first place. I like this quote attributed by Scerri (1996, p. 174)^ to Peter Nelson, "who often writes on matters of conceptual chemistry": "The application of periodic law is not simply a case of making logical deductions from basic principles as in the case of thermodynamics. It involves rather a fairly thorough knowledge of how the principles work out in practice - of exceptions, trends and patterns [e.g. the first row anomaly, anionic gold, rogue transition metals etc] - so that a particular deduction can be appraised and if necessary adjusted. In the limit the process becomes virtually intuitive"

So, yes, one would suspect from its position in the PT that N should have a high EN rating. And it turns out that it does. When appraising the overall behaviour of N one sees that, apart from e.g. H-bonding, and thanks to the ball and chain represented by properties 1–6, it fails to consistently live up to its EN, so much so that it turns out to be more like an intermediate nonmetal than a corrosive nonmetal. Thus, following appraisal, one's initial deduction is adjusted downwards—a subtraction if you like—and N is classed as an intermediate nonmetal, with some boundary overlaps into the corrosive nonmetals. Or, if I may borrow your own words when referring to Be, "…as a slightly extreme member of its category rather than as a very extreme member of another category."

In this vein, as you might know, how the overall nature of N plays out in the literature is provisionally outlined in the parsing the nonmetals section of my sandbox. Sandbh (talk) 04:42, 4 July 2017 (UTC)

Except that the parts of N chemistry that play out as you would expect for a strong nonmetal are among the most important parts of it, and in those cases it acts a lot more like O and F than C and P. I appreciate that some electronegativity differences look rather inflated – especially the apparent one between C and H – but N is pretty far from this. For from the electronegativity comes:

1. A capability for hydrogen bonding unmatched except by O and F;
2. The formation of coordination complexes more similar to those where O is the donor than when P is;
3. A great preference for multiple bonding over catenation, allowing for the richness of organic N chemistry, which is more similar to organic O chemistry than organic P chemistry.

I recognise that the inorganic chemistry of N is more covalent than that of O but it is not clear why we should expect it to be otherwise. Certainly it seems that we have a difference in kind from that of C and P which are never "truly" ionic at all, even if you go all the way and use cations like Ba²⁺. I also recognise that the oxoacid chemistry of N is much more restricted than that of P, but again it is not clear why we should expect this to be otherwise: the same is true for C compared to Si, and for O and F compared to S and Cl it even becomes vacuously true.

I also fail to see how kinetic restrictions have anything to do with electronegativity. Electronegativity just tells us how closely an atom would attract a bonding pair of electrons. So I don't see how this can be considered a case of N not living up to its electronegativity. If we take kinetic restraints into account for its most common species in aqueous solution, like you do in your sandbox table listing electrode potentials, then it looks like Cl is not living up to its electronegativity and, while still strong, should be placed below Br. But that is absurd, looking at their reactions, and should remind us what electronegativity can and cannot tell us.

It seems to me that in its covalent, organic chemistry, N still resembles a weakened oxygen more than it does a strengthened phosphorus. Even in its inorganic chemistry, because nitrides are usually interstitial while phosphides are often covalent or even metallic, it strikes me that there is still an essential difference dictated by size and not electronegativity that once again cuts N from P rather than from O. Double sharp (talk) 06:23, 4 July 2017 (UTC)

Hi Double sharp. Nice to be back. Have been on a mini-break.

My thoughts about N chemistry and where to draw the dividing line among the nonmetals are now going along the lines of fundamental v peripheral, noting I haven't completely made up my mind pending consultation with chemists and more research and discussion.

In the sense of attempting to divide the nonmetals I now tend to think that hydrogen bonding and ligand capacity are notable but not fundamental properties. I am reducing H-bonding in significance given it only occurs in the presence of H; and that S along with N, O, F, Cl and Br are apparently included among those elements that can form the strongest H-bonds, acknowledging that N is a better H-bonder than S. I am reducing ligand capacity in significance given the distinction between class A metals (including the lighter TMs in higher oxidation states) and class B metal ions (inc. the heavier TMs, and the lighter TMS in low oxidation states) and that while N is a strong ligand former with class A ions, P and S are strong ligand formers with class B ions.

The reference to organic v inorganic makes me think of Cotton et al's mention of so much N chemistry being thought of as organic whereas O was of such importance to inorganic chemistry. At the time I didn't think there was much in this worth pursuing and I still tend to think that way now, at the level abstraction we are dealing with, which is a first order division of the nonmetals.

I have nothing further to say on attempting to distinguish between N and C and P on the basis of supposedly ionic v covalent nitrides, carbides or phosphides. The overwhelmingly covalent nature of N compounds says it all, in view .

Kinetic restrictions may be less relevant. It is probably more fruitful to note that N by itself is a poor oxidant unlike O and the halogens, and that N needs the help of O (or F) before it can become a good oxidant.

Distinguishing between the interstitial nature of nitrides v the covalent nature of phosphides overlooks the interstitial nature of carbides. It is enough to note that, according to the literature, the phosphides "resemble in many ways the metal borides, carbides, and nitrides" (Greenwood & Earnshaw 1998, p. 490). Sandbh (talk) 01:01, 9 July 2017 (UTC)

What is important chemically is not just dependent on how many elements a phenomenon is observed with, but also how common these elements are. An anomaly that happens only with astatine is not particularly important given its fugitive nature and how F, Cl, Br, and I are so similar, for example. But hydrogen probably forms more compounds than any other element except for carbon. It seems to me that something that happens in the presence of H must happen very often and be a very serious deal.

You also seem to acknowledge this by noting that N chemistry is mostly "organic" (involving C), while O chemistry is also important to inorganic chemistry (not involving C), so I am not sure why C can have this special status but not the even more ubiquitous H. (And O is also important to organic chemistry, so I still don't consider this a perfect dichotomy.)

As I have said many times, O by itself is also kinetically hindered as an oxidant, because of the strength of the O=O bond and the spin mismatch. If it was a really unhindered oxidant like the halogens, we would all be on fire. We can see the importance of bond order here by noting that S reacts much more readily than O: it combines directly with all elements at room temperature (or slightly higher) except the noble gases, N, O, Te, I, Ir, Pt, and Au. (Similarly it reacts with O₃ but not O₂.) Again, this is because it is much easier to break the S–S bonds in S₈ than the O=O bonds in O₂. But we're not going to start calling sulfur a corrosive nonmetal, are we? Double sharp (talk) 06:10, 9 July 2017 (UTC)

Hmm, well, at the level of abstraction I think we are dealing with none of these observations matter.

How common an element is is not germane to its characterisation albeit this might be hard if it is quite scarce. H-bonding is an important phenomenon but (I hate using so many "but's" ^__^) in assessing the overall nature of an element I attempt to have regard to its reactions generally with the other elements, or at least I hope I do, rather than how it behaves in the presence of one element. In that sense, it seems to me that N's capacity to engage in H-bonding is a secondary consideration.

I didn't think there was much in the organic/inorganic distinction and I don't intend to pursue it.

I agree O is kinetically hindered, yet:

• it is highly combustible (a spark or a match will do), while N shrugs its shoulders and says, "you're wasting your time kiddies"
• O lends oxidative power to other nonmetals (nitrates, chlorates etc)
• N is otherwise a weak oxidant, whereas O and the halogens can speak for themselves.
• N was rightly regarded as as inert gas before the discovery of the noble gases
• not for nothing was oxidation called oxidation
• the relationship of O to corrosion has been appreciated for a long time.

On oxygen, G&E say it is, "an extremely reactive gas which vigorously oxidizes many elements directly, either at room temperature or above. Despite the high bond dissociation energy…these reactions are frequently exothermic" (p. 612).

On S they say it is, "a very reactive element especially at slightly elevated temperatures (which presumably facilitates cleavage of S-S bonds." They go on to note that S reacts F, Cl and Br; in the cold with the metals from groups 1, 2, 13, Sn, Pb, Bi, Cu, Ag and Hg. The Ln, An and the rest of the transition metals require heating (Ir, Pt and Au being impervious).

I can't see much difference here. I don't know but suspect that O will need the presence of moisture to oxidize some elements at room temperature; and I suspect that the reactions of S with some of those 20 metals will be triggered by the presence of micro sulphuric acid impurities.

G&E aside, and considering just the dot points, I feel it is pragmatically reasonable to not regard S as a corrosive nonmetal. Sandbh (talk) 12:46, 9 July 2017 (UTC)

However, the very fact that H is such a common element in compounds tells us a lot about its important role. It is not just that it is common, but that it is part of its very nature that it can form a lot of compounds. Not many elements can claim themselves to be the foundation of an entire concept in chemistry: yet H and C do that (hydrogen in acid–base chemistry and carbon of course in organic chemistry). It still seems to me that what elements get up to doing with H is pretty important for this reason.

If we want to be abstract about it, I would be slightly more happy with saying that all these nonmetals are more or less reactive depending on how strong their bonds to themselves in elemental form are, which is why you don't see the "abstract" trend in electronegativity followed. The barriers for breaking P–P and S–S bonds are much less than they are for breaking N≡N and O=O bonds. When we deal with atomic N and O, both are very reactive, because this barrier has been removed. That is why N is so much more reactive in its compounds than as a free element, and why hypothetical N allotropes have been investigated as possible extremely powerful propellants and explosives.

O tends to lend oxidative power because it is good at getting other elements in high oxidation states; F does the same thing. But then again Cl and Br are not that great at doing this, and given that the Pauling electronegativity of N is about that of Cl and Br, we could hardly expect anything else there. What is necessary is a significant gap in Pauling electronegativity to create significant bond polarity even with strong nonmetals in things like nitrates and perchlorates, and while I agree that it singles oxygen and fluorine out, I do not agree that it presents a good justification for the {O, F, Cl, Br, I} category as chlorine, bromine, and iodine are not good at this at all.

Regarding your last statement, I of course agree that it is reasonable to not regard S as a corrosive nonmetal. But I was rather interested in knowing why, especially given that there does not seem to be a lot of difference in the reactions, and amusingly that Ag reacts readily with S and not O. ^_^ Double sharp (talk) 13:11, 9 July 2017 (UTC)

P.S. Funny that Greenwood and Earnshaw on that page had just called Ge, As, and Sb nonmetals, and then proceeded to not say "groups 1, 2, 13, 14, and 15", but instead single out Sn, Pb, and Bi on their own. We just can't seem to decide how to deal with the metalloids, can we? ^_^ Double sharp (talk) 13:14, 9 July 2017 (UTC)

H-bonding is important yet is not considered, from my reading of the literature, when discussing the nonmetallic character of the elements whereas ionisation energy, electronegativity, reactivity/activity, oxidising power, simple anion formation, and electron affinity more often are.

Abstraction is appropriate when it attempts to summarise and draw together the most significant aspects of the topic at hand i.e. the nonmetallic character of the nonmetals, rather than individual aspects or anomalies (unless the matter is finely balanced, in which case you may wish to focus on an individual aspect or anomaly to tip the balance).

Cl, Br and I do well enough in the oxidation stakes compared to N, which is the runner up here. Of course, I wouldn't consider oxidising power in isolation, as per your comment, nor curious anomalies like S and Ag. I remember a chemist saying the interesting things occur in the margins, but that's not where the focus of his teaching was: generalities first, noting there may be some anomalies that will be taught later, or in a different course.

Yes, I regret to agree about the inconsistent way the literature tends to deal with the nature of metalloids when it is not that hard. A tripartite classification of the elements is fine for most purposes and corresponds with basic reality: the reactive alkali and alkaline earth metals at one end; the halogens at the other end, and meeting in the middle are some of the weaker post-transition metals, and the non-metallic metalloids. Sandbh (talk) 07:25, 10 July 2017 (UTC)

(1) H-bonding is nevertheless a manifestation of high electronegativity as mediated by the ability to take on a high partial negative charge, and so it becomes intimately related to one of the properties you mention. Nitrogen tends to act like you would expect from its Pauling electronegativity in compounds, which is as you would expect considering that electronegativity measures how much a bonding electron pair is attracted to the atom involved. If not we would have to consider the electronegativities of Kr, Xe, and Rn to be "misleadingly high", when they are actually completely accurate in rationalising their behaviour in their compounds. (And I would note that many of these criteria have a tendency to end up calling many of the d-block metals metalloids.)

(1) When I referenced Cl, Br, and I, I was going for the fact that dumping many Cl, Br, and I groups on an element does not tend to make a terribly good oxidising agent, because these groups are just not as electron-withdrawing as O and F are, IIRC. There's a good reason why most strong acids that don't involve O as peripheral groups use F insead (e.g. HBF₄, HPF₆, and famously HSbF₆). We can also see F as a big step up by comparing pKa values for acetic acid (4.76), iodoacetic acid (3.15), bromoacetic acid (2.86), chloroacetic acid (2.81), and fluoroacetic acid (2.66). (Granted, this inductive effect is not the biggest one, as the effect of delocalisation is much stronger still, but it is significant and is one way to get superacids like triflic acid. (BTW, are iodine and its compounds really that much more impressive than, for example, nitric acid?)

(3) Once again I would say that the crossover tends to happen with oxidation state. There seems to be a big drop in metallicity between Tl^I and Pb^II on the one hand and Bi^III and Po^IV on the other, for example: actually, some chemists (including Mendeleev in 1869!) seemed to consider Tl^I and Pb^II for quite some time to be reasonable heavier congeners to Cs^I and Ba^II(!). Double sharp (talk) 08:15, 10 July 2017 (UTC)

(1) We have spoken before about N's misleadingly high EN and I'm not seeing anything new in this regard. I don't dispute that N has a high EN rating; it's just that its overall electronegative character isn't of the same calibre as O and the halogens. It seems to me that the noble gases stand by themselves, just like the noble metals do. That aside, the EN's of Kr, Xe, and Rn may well be misleadingly high—they are reckoned to have EN's more like the chalcogens, noting I haven't fully digested the article in question.
(2) The standard reduction potentials of Cl, Br, and I exceed that of N. The average potentials of Cl, Br, and I and their compounds, exceeds that of N. There is no point in comparing a single aspect of N chemistry (nitric acid) with iodine and its compounds.
(3) These are not relevant considerations at the level of abstraction under consideration. Sandbh (talk) 13:19, 10 July 2017 (UTC)

For (2), I was thinking of compounds where Cl, Br, and I are the peripheral atoms, kind of like O and F in anions like PO³⁻
₄ and PF⁻
₆. Is that not what you were referring to? (Because if so I seem to have made a bit of a misunderstanding, though I would note that in this case Cl, Br, and I do not really hold a candle to O and F very well.)

No that wasn't what I was referring to. I was referring to the element and average standard reduction potentials in table 3 of my sandbox. Sandbh (talk) 02:00, 12 July 2017 (UTC)

As for (3), it depends on whether you want to start with metallicity as a basis or valency (and hence groups). I find it easier to derive metallicity from valency, so I would tend to start with the latter: it nicely puts the strong metals with low positive oxidation states (+1, +2, marginally +3), the strong nonmetals with low negative oxidation states (−1, sometimes −2, very rarely −3), and shrugs its shoulders in the middle to indicate the mess that results there in the middle of the p- and d-blocks. I am reading your linked article with interest with a mind to rationalise the singular case of the noble gases, where the standard oxidation state is 0! In cases like Xe^IV and Xe^VI we can usefully compare things with covalent I^III and I^V and obtain a reasonable picture here; but here, interpolating strangely between I⁻ and Cs⁺ to find Xe⁰, we have an interesting conundrum, given that the signature property of noble gas chemistry is a lack of chemistry (for He, Ne, and Ar, and while Kr, Xe, and Rn can form bonds they'd really rather you didn't get them to). Double sharp (talk) 14:43, 10 July 2017 (UTC)

Yes, I understand your approach. That's not the approach I've been taking, which is simpler. Yours has a level of sophistication that isn't required at the level of abstraction I think we are dealing with vis-à-vis the categorisation of the elements into broad categories. As you note, valency gets a bit complicated in the d-block. I think in terms of drop-in-the-water alkali metals on the left (e.g calcium), weak low melting metals on the right (tin), and respectable metals in the middle (iron). Same with nonmetals, working back from the inside right. I tend to think that's the approach most people take when trying to divide the elements into broad categories according to their shared characteristics. It all starts with the popularity of comparing the alkali metals with the halogens/noble gases. Sandbh (talk) 02:00, 12 July 2017 (UTC)

The comparison of chemistry would indeed be happiest pitting group 1 against group 17, but I am not sure how group 18 then comes into it, except as the ultimate goal and role model of both of the aforementioned pair. Comparing physical properties would certainly have group 18 on the one end with the weakest intermolecular forces (He is an extreme example and is the closest thing you'll ever find to a real ideal gas), but paradoxically it is in physical properties where the metal/metalloid/nonmetal trichotomy seems to reign more supremely, with semiconductors like Si in the middle forming a relatively well-defined patch, and then the best metals seem to be transition metals – especially things like W which are really quite poor from the chemical point of view (go figure). Fe is for sure a respectable metal, but as I have said, I wonder how much of that is because it is so much more common than anything else that we commonly see as a pure metal. As a result we end up judging metals by how much they act like Fe, and it is not surprising that apart from its fan club of Co and Ni, most metals cannot do that.

The difficulty of classification, to my mind, stems from this almost-total opposition between the chemical and the physical trends. I accept that my valency-based idea, going back to looking at electron configurations, is simultaneously too detailed and not detailed enough. A cursory look at valency makes it hard to predict what is going on with the 6p metals, for example in the anomalously weak metallic character of Tl^I, Pb^II, and most especially Bi^III, for example. (I wouldn't say it makes things hard in the transition metals; actually it seems to explain very well why Cr^III is so much better as a metal than Cr^VI, for example.) And a cursory look at physical properties makes it hard to explain just why the chemistry of H is more metallic than you would expect, or why so many of the 4d and 5d metals are so weak chemically. So I am beginning to suspect that, just as most texts do, starting with Aufbau and the electron configurations much earlier than considering metallicity, bonding, or any of general chemistry, is a very wise move. (For one, the differing cohesive effect of d and p electrons explains the otherwise troubling difference in metallicity between Mo and Te, for example.) We can then simply say: "here are your blocks; here are your groups; and everything shall come from that, so that whenever we find something, we can always explain it from those principles. You can see the metals here and the nonmetals over there; you can see some general trends in characteristics over here; you can see minute differences in some esoteric branch of chemistry over there; it doesn't matter, as they all shall fall under the jurisdiction of the s, p, d, and f orbitals". Double sharp (talk) 06:01, 12 July 2017 (UTC)

Agree, mostly. Sandbh (talk) 07:19, 12 July 2017 (UTC)

Fundamental properties of N

At the heart of the matter are the atomic properties of N and O, including those used to rationalise the behaviour of N and O atoms in chemically combined states. N does not rust metals; there are no anti-nitridants needing to be added to foods. At 100% O is too chemically active to live with comfortably; diluted down to 20%, thanks to the presence of N, these problems go away. N is reluctant to form anions; nearly all N compounds are covalent. O readily forms anions; the ionic chemistry of oxygen in the solid state "is a vast and important subject".

In N there is no comparably sweeping set of similarities with the general nature of the corrosive nonmetals. Sandbh (talk) 12:31, 28 June 2017 (UTC)

Tell that to Li and the group 2 metals! ^_^ I would prefer to see the N≡N molecule as a "de-reactivified" version of the isoelectronic CO, CN⁻, and NO⁺. All of these are significantly more reactive because there is now some polarity: again I tend to find that the unreactivity of nitrogen (and even the fact that it forms a simple N₂ molecule in the first place, rather than following its "older sisters" P and As) are easier to get if you start from a "highly electronegative nonmetal" base and apply second-order course corrections to it.

Even oxygen is sort of a "happy medium" between N and F. I've often joked about halogen-breathing life, but the problem with that is that the halogens are just too eager to react. The main reason is the unpaired electrons in O₂; triplet oxygen thus only reacts rather slowly with most organic molecules (which have spin-paired electrons), and that is why we haven't all spontaneously caught fire. Compared with F, Cl, Br, and even I, free oxygen seems a little bit wimpy! ^_^ Double sharp (talk) 15:11, 28 June 2017 (UTC)

I decline your gracious reference to Li and the group 2 metals on the grounds of this representing a distraction! There is no particular need to be concerned about Li given G & E refer to group 1 as "homogenous" (p. 68). The alkaline earth metals (AEM) are more noteworthy given the similarities of Be with Al and Mg with Zn. While I said earlier that we should seek to minimise these anomalies and do so pragmatically rather than myopically, I see no scope for doing anything here given the AEM are all metals. You even (effectively) acknowledge this when you said that we include Be as an AEM because it fits better as a slightly extreme member of its category rather than as a very extreme member of another category, an assessment with which I agree.^

^ And which applies just as well to N as an intermediate nonmetal!

I'm not sure why it's easier to start with N as a "highly electronegative nonmetal" and then knock it down by applying second-order negative corrections than it is to start with N as an "intermediate nonmetal" and then build it up by adding second-order positive corrections. Since when was subtraction easier than addition?

I don't think I need to say much more about oxygen, given these observations (you've see some of these before):

1. "The halogens and oxygen are the most active non-metals."

2. "By the end of 8th grade, students should know that…there are groups of elements that have similar properties, including highly reactive metals, less reactive metals, highly reactive nonmetals (such as chlorine, fluorine and oxygen) and some almost completely unreactive gases (such as helium and neon)."

3. "Of the nonmetals, oxygen and the halogens are highly reactive."

4. "…oxygen is a potent, reactive, and corrosive gas…liable to generate free radicals, highly reactive bleachlike compounds that burn up organic molecules. We are able to tolerate an oxygen-rich environment only because our cells possess complex biochemical mechanisms for suppressing its many harmful influences."

¹ Lee GL & Van Orden HO 1965, General chemistry: Inorganic and organic, 2nd ed., Saunders, Philadelphia, p. 19

² American Association for the Advancement of Science 1994, Benchmarks for science literacy, Oxford University Press, New York, p. 78

³ Frank DV, Miller S & Little JG 2004, Prentice Hall Science Explorer: Chemical Interactions, Prentice Hall, p. 19

⁴ Ball P 2013, Life's matrix: A biography of water, University of California Press, Berkeley, p. 232

-- Sandbh (talk) 00:10, 2 July 2017 (UTC)

The reason is that I can't see much of a reason why an element in that position should be strengthened as a nonmetal, whereas I can immediately come up with some good reasons to subtract nonmetallic power from it (the extreme negative charge required to attain the Ne configuration and the small size). Addition and subtraction are both well and good, but surely both should follow from well-established principles. It seems rather easier to me to say that N would have been strong in some respects if not for it being too small (hence handicapping its kinetic efficacy) and too far from its dearly desired Ne configuration (forcing it to compromise and avoid ionicity except when the metals are only too willing to give it its desires, like Li), than to say that N is at heart like P but sometimes decides to go full throttle for no obvious reason. It's not for nothing that N and P are so different that when chemists talk about group 15, they typically restrict "pnictogen" to mean P, As, Sb, and Bi only.

Regarding Li and the group 2 metals, I was responding to your statement "N does not rust metals". It does – at least those highly active ones. ^_^

I agree of course that we have evolved many mechanisms to tame oxygen. I just find it somewhat significant that we don't have any to tame the halogens, which seems to be a further step up. I have looked up some of the more speculative xenological material about life using HF or HCl as a solvent and there just keep seeming to be obstacles because F and Cl are just even more trigger-happy than O, and even Br and I seem to fit this description (I guess the reason why they weren't considered is that they are much rarer than F and Cl). This actually makes me suspect that even more than size, lowering the amount of negative charge that needs to be borne is the most important factor. Even oxygen is not too willing to stay as O²⁻: that species has such a huge proton affinity that it immediately grabs a proton from whatever solvent you are putting it in to become OH⁻. So only the halogens are really well-defined solvated X⁻ in solution. And as I've noted, while we have no doubt managed to tame oxygen, O₂ itself carries many of the seeds of this lessened reactivity compared to F₂ or Cl₂ – with perhaps the most obvious answer being that a double bond is stronger than a single bond. Double sharp (talk) 06:54, 2 July 2017 (UTC)

I'm still digesting the addition v subtraction idea.

Regarding N being at its heart like P, and N sometimes going full throttle for no obvious reasons, you're referring to things I've never said, and then adversely judging them. I don't see the logic here. I've never said that N is at its heart like P; I've never said that N sometimes goes full throttle for no obvious reasons. Slap me with a wet fish if I have.

I suppose we don't have to tame the halogens since there is so little of them in comparison to oxygen. Mind you, for all that "taming" of O by living systems, we still see the effects of O all around us in inanimate things, via rusting, tarnishing, rotting of wood and other materials, and sundry related phenomena. Sandbh (talk) 08:22, 3 July 2017 (UTC)

Well, a little bit further up you said that "I listed a dot-point overview of the chemistry of N and its compounds, above, and it seems to me that N in fact very much tends to act like C or P or the rest of the intermediate nonmetals.", considering the times it acts like O to be anomalies. My point is that it's not particularly clear how one would rationalise the times it acts like O in that case. If you suppose that N is really that weak, with an electronegativity more like C, why is it a strong hydrogen bonder all the time, for instance?

The effects of O₂ are still nonetheless tamed by the O=O bond energy and the spin mismatch of the involved electrons, which does not happen for the halogens. If you get atomic oxygen involved as a reactive intermediate, then things start to go as fast as they do usually with the halogens, but even that is also true of atomic nitrogen. We need only see the difference between sticking aluminium in oxygen vs sticking it in bromine, for example. Double sharp (talk) 10:30, 3 July 2017 (UTC)

Oh, I see. Perhaps I wasn't clear enough. When I said "that N in fact very much tends to act like C or P or the rest of the intermediate nonmetals" my intended meaning was that, with reference to the weaker nonmetallic properties of N, namely…

1. its reluctance to form simple anions;
2. the covalency of nearly all of its compounds;
3. the kinetic restraints that are a hallmark of its chemistry and that of its compounds;
4. that the average oxidising power of it and its species, in aqueous solution, is less than that of both iodine and of sulfur;
5. that nonmetal displacement series list iodine and sulfur ahead of nitrogen; and
6. that nitrides resemble the metal borides, carbides and phosphides in many ways

…it tends to act like the rest of the intermediate nonmetals, including C or P.

I owe you a response to your question re "how one would rationalise the times it acts like O in that case." I think this may be related to the addition v subtraction item, in respect of which I also owe you a considered response.

I don't have any more to say on oxygen, for now. The differences between oxygen and the other corrosive nonmetals are interesting but not germane, IMHO, to the broad level of classification science we are dealing with here. Sandbh (talk) 11:42, 3 July 2017 (UTC)

Here's an interesting site, which says the following on oxygen: "Although O₂ is a strong oxidizing agent, its oxidizing reactions are usually slow kinetically. Many flammable materials will exist in the presence of air until a flame or spark starts the reaction." So it seems to me that one cannot have kinetic barriers as a way to exclude N compounds while ignoring them when they impact the activity of oxygen. (Even the rusting of iron is not all that fast!)

He also seems to provide an example of considering metalloids like Si, Ge, As, Sb, Te, and At under the umbrella of nonmetals. Mind you, he doesn't include B, and he seems to also be including Bi and Po as nonmetals(!).

It is also amusing to see his table of nonmetal anion pK_a values and realise that Se²⁻ and Te²⁻ are actually not completely protonated by water (like the halide anions), unlike O²⁻ and S²⁻. This seems to suggest that "reluctance to form simple anions" is also partially about size and charge, and given that Se and Te are better at it than O and S I would suggest that it is not an important factor for denying N the status of a strong nonmetal. Double sharp (talk) 04:24, 6 July 2017 (UTC)

What do you know, it's actually from Gary Wulfsberg's Principles Of Descriptive Inorganic Chemistry (section 5.5). ^_^ Double sharp (talk) 14:20, 6 July 2017 (UTC)

It is odd that Wulfsberg appears to treat Bi and Po as nonmetals here, unless he was referring to the metalloid or nonmetallic aspects of their chemistry, given both are post-transition metals.

The aqueous basicity of the monatomic anions and their partially protonated forms, refers only to their capacity to compete with water for hydrogen ions, rather than to their capacity to form simple anions in the first place, or at least that is how I interpret it. It is worth looking into this further given the widespread acknowledgement in the literature on the capacity of O and the halogens to form simple ions. Sandbh (talk) 01:32, 9 July 2017 (UTC)

If you consider forming simple anions, then again we have a problem: only the halogens will form the simple anion according to their group number spontaneously if you dump some of their atoms in a gas of electrons. Oxygen, for instance, will quite happily form O⁻, but in such a situation it has no incentive to form O²⁻ because the negative charge it already has repels the electron and the reaction becomes endothermic: its second electron affinity is negative. Given that electron affinity refers quite simply to this – the amount of energy released by the addition of an electron to a gaseous atom or ion – it strikes me that if you are considering the anions the elements should form according to their group numbers, only the halogens will do it spontaneously (yes, even At).

Of course, we know that O²⁻ exists in chemistry in ionic lattices like CaO, because then the lattice energy released is high enough thanks to the high charges involved. But then N³⁻ also does in things like Ca₃N₂, so I am not sure what this proves if anything.

Incidentally, the fact that we do not see cations like C⁴⁺, Ge⁴⁺, and even Zr⁴⁺ in aqueous solution is for about the same reason: they are more acidic than water and protonate it immediately. Less acidic cations like Th⁴⁺ and Fe³⁺, and at a much lower extreme Mg²⁺, also protonate water but do it much more slowly. So if we consider it irrelevant that O²⁻ is too basic to exist in water, then by the same token it seems to be also irrelevant that Ge⁴⁺ is similarly too acidic, because you can certainly remove four electrons from gaseous germanium atoms if you want to! Analogies with Ti⁴⁺ in glasses certainly suggest the existence of Ge⁴⁺ as a cation.

As for Bi and Po, I suspect this is because they continue the trend down from Sb and Te reasonably well. After all, their most stable forms in neutral aqueous solutions are not like Tl and Pb (which form Tl⁺ and Pb²⁺ respectively), but rather Bi₂O₃ and HPoO⁻
₃. If memory serves well bismuth does not form simple cations in aqueous solution – its most famous oxocation [Bi₆(OH)₁₂]⁶⁺ is most reminiscent of [Ta₆Cl₁₂]²⁺ if anything. And it readily forms oxyanions such as BiO³⁻
₃ (in Li₃BiO₃) and Bi
₂O¹⁰⁻
₈ (in Ag₅BiO₄), even in its lower oxidation state of +3. The chemistry of Po is of course much less well-known, but given that it would usually be one oxidation state higher than Bi it would seem to be even more inclined to this kind of behaviour. I suspect the reason we are unwilling to call them metalloids is just like that for the "naughty" 4d and 5d metals like Ta and W: physically, bismuth and polonium are certainly metals, if not among the best ones like tantalum and tungsten are. But chemically speaking, they do not seem to be all that much more metallic than Sb and Te, but simply continue the trend down. Double sharp (talk) 03:43, 9 July 2017 (UTC)

Re simple anion chemistry, in Inorganic chemistry (Cox 2004, p. 146) the author writes: "Simple monatomic anions are formed only by the most electronegative elements, in groups 16 and 17 (e.g. O^2–, Cl^–). Although C and N form some compounds that could be formulated in this way (e.g. Li₃N and Al₄C₃), the ionic model is not very appropriate for these." Sandbh (talk) 03:11, 12 July 2017 (UTC)

Note how they say "groups 16 and 17" without qualification, confirming my suspicion that charge is the most important factor here. Given that Se²⁻, Te²⁻, and even Po²⁻ are all known, I hope you'll forgive me if I am not very convinced by the criterion of simple anion formation, if it means that nitrogen is not as good a nonmetal as polonium! The high coordination numbers of the nitrides of the heavy alkaline earths at the observed interatomic distances seems to indicate that the ionic bond model is not totally wrong-headed there, which I guess is the small size and electronegativity helping (with phosphorus and arsenic, you can forget about the compounds even trying to act as if they were ionic, even if the counter-cation is something like Na⁺), but this particular phenomenon seems like a job for the group number most of all, not metallicity. Double sharp (talk) 04:34, 12 July 2017 (UTC)

The wording is somewhat ambiguous. I take it to refer to the most EN elements i.e. O and the halogen nonmetals. I would not consider simple anionic chemistry in isolation. I think what I said before is still right: "Broadly, our organising principle is to categorise elements having regard to metallic or non-metallic character, and cross-cutting similarities in chemistry. Here, the character of an element is a composite of such chemical properties as activity, reactivity, electronegativity, acid-base character, and cationic/anionic tendencies, as influenced by relevant physical properties. Cross-cutting similarities in chemistry refers to e.g. the similarities in chemical properties that distinguish most of the transition metals." You can certainly find anomalies or overlaps if you focus on individual aspects of a broad categorisation scheme. Sandbh (talk) 05:29, 12 July 2017 (UTC)

I would say that the trouble comes from the fact that the periodic trends in the nonmetallic region are not really working in harmony. They favour small atoms for being more able to hold a negative charge, but also hinder them through kinetic restraints; and as one moves further away from group 0, anionic chemistry abruptly becomes much less likely between groups 16 and 15, and this isn't helped by the concomitant increase in multiple bonding as one goes to the left to get a stable octet configuration without much catenation. Since simple anions of the heavier chalcogens are known and can reasonably persist in water (H₂Te and friends evidently being acidic enough to lose both protons), I would say that "groups 16 and 17" is essentially correct, but that this must be included as just one of the factors influencing metallicity. Just because Se, Te, and Po are happy to form dianions does not mean they are necessarily strong as nonmetals; just because N forms its trianion very unwillingly does not mean it is necessarily weak. We mustn't forget also that one of the main reasons we're unhappy to call Po a metalloid or nonmetal is its physical properties (it has a metallic band structure, notwithstanding its chalcogen-like chemistry), and physically N₂ acts very much like what you would expect for a nonmetal that is very sure of its identity. ^_^ Double sharp (talk) 05:38, 12 July 2017 (UTC)

Agree (broadly). Sandbh (talk) 06:26, 12 July 2017 (UTC)

PS: Aiee! Another postscript. Earlier you said that, "why is it [N] a strong hydrogen bonder all the time, for instance?" I don't think you meant to say "all the time", did you? That is not my recollection of the literature. Sandbh (talk) 04:37, 12 July 2017 (UTC)

Maybe it's not strong all the time, but it's certainly strong a lot more often than not, isn't it? Double sharp (talk) 04:40, 12 July 2017 (UTC)

I don't know about the "a lot more often than not" but I have for sure agreed that, in general, N H-bonding could reasonably be expected to be stronger than S H-bonding, noting that the converse may be true at some other times. Sandbh (talk) 05:29, 12 July 2017 (UTC)

I'd say that N H-bonding can usually be taken to be strong by default, while S H-bonding can only safely be taken as such in the presence of one-off factors that increase the electronegativity of S in that particular compound. So strong N H-bonding can be considered the norm and strong S H-bonding somewhat of a second-order anomaly. Double sharp (talk) 05:41, 12 July 2017 (UTC)

Essentially agree at this time, subject to further research. Sandbh (talk) 06:28, 12 July 2017 (UTC)

Nitrogen vs phosphorus

Here, incidentally, is a choice quotes from Greenwood and Earnshaw supporting my vague idea that the real difference is between the second and third rows, and hence between N and P (instead of N and O):

• 'In much of its chemistry phosphorus stands in relation to nitrogen as sulfur does to oxygen. For example, whereas N₂ and O₂ are diatomic gases, P and S have many allotropic modifications which reflect the various modes of catenation adopted. Again, the ability of P and S to form multiple bonds to C, N and O, though it exists, is less highly developed than for N (p. 416), whereas the ability to form extended networks of –P–O–P–O– and –S–O–S–O– bonds is greater; this is well illustrated by comparing the oxides and oxoanions of N and P. "Valency expansion" is another point of difference between the elements of the first and second periods of the periodic table for, although compounds in which N has a formal oxidation state of +5 are known, no simple "single-bonded" species such as NF₅ and NCl₆⁻ have been prepared, analogous to PF₅ and PCl₆⁻. This finds interpretation in the availability of 3d orbitals for bonding in P (and S) but not for N (or O). The extremely important Wittig reaction for olefin synthesis (p. 545) is another manifestation of this property.' (pp. 473, 475)

This is why I find N is too different from P to talk about in the same league; P and S are just so much happier to catenate, and in fact the allotropy of P and especially S is quite unmatched by any element except C. The differences with As, Sb, and Bi are more a matter of kind "because of the countervailing influence of the underlying filled d and f orbitals" (p. 550). Double sharp (talk) 15:13, 4 July 2017 (UTC)

I think the distinction between the elements of the first second and second third periods of the periodic table is a second-order consideration at the level of abstraction we are dealing with. We are able to tolerate the co-location of B with the metalloids, even though it is a second row element, and the bizarre(?) difference in its chemistry compared to the other metalloids, but we cannot do so for N and P, never mind the broader similarity of N with the other intermediate nonmetals, analogous to the similarities of B with the other metalloids. I feel we should strive to accommodate these differences in our categorisation schemes where possible but not at the expense of pragmatism, in the first instance, over needless complexity.

I feel a little like I have to justify why P should not be in the same group as N, which it obviously is, yet the differences between N and P are accommodated nevertheless and understood by chemists.

You have got me thinking about this though. I'm looking forward to hearing from e.g. how Nelson, looks at this. Sandbh (talk) 03:49, 9 July 2017 (UTC)

But then I think boron is a bit too different to consider as a normal metalloid anyway because it's hypoelectronic. Thinking of it as a forcibly nonmetallised version of its congeners in group 3 and 13 is a slightly happier classification, since their most stable forms in water are all the "hydroxides" M(OH)₃ just like for boron (until you get to the really basic La and Ac, and marginally Y). Similarly, the inability of N to expand its octet, and its preference for multiple rather than single bonding just like O and F, seems to be a significant qualitative difference that speaks for a broader similarity of N with the other 2p elements instead of with P, for example. Double sharp (talk) 05:16, 9 July 2017 (UTC)

Boron is nevertheless treated as metalloid. The situation is effectively no different to treating N as an intermediate nonmetal. Sandbh (talk) 12:55, 9 July 2017 (UTC)

Nevertheless I think you could make a good argument that it shouldn't be one; note how Wulfsberg treats all those metalloids as nonmetals, but pointedly leaves boron out of it. Similarly I think nitrogen is not well-placed as one of the weaker nonmetals, because following these arguments would have us consider S potentially a better nonmetal than O as it has fewer kinetic barriers to reactivity. Double sharp (talk) 12:58, 9 July 2017 (UTC)

Well, that's interesting re boron. In Inorganic chemistry (2000), Wulfsberg refers to B, Si, Ge, As, Sb, Se and Te as semiconducting metalloids when he is discussing the physical properties of the elements. When he discusses the redox behaviour of the nonmetals he refers to the metalloids, including B, as nonmetals Like. On S compared to O we need to consider the broad sweep of applicable properties, rather than focussing on a specific difference. In this context, the analogy is not valid. Sandbh (talk) 04:41, 10 July 2017 (UTC)

Okay, but when I look at your critique at N, it seems to me that S is meeting all these stated criteria. For example:

1. the simple anion S²⁻ is well-known...
2. ...leading to a significant ionic chemistry in many metal sulfides;
3. it is large enough to avoid the kinetic restraints that are the hallmark of N^V and Cl^VII;
4. and its oxidising power and position in nonmetal displacement series appear to be ahead of nitrogen;
5. and finally, unlike for B, C, and N, you will not find interstitial sulfides.

So I remain somewhat confused as to S is not considered a corrosive nonmetal in your scheme. Double sharp (talk) 05:17, 10 July 2017 (UTC)

I've never thought about this, but it's an intriguing question. Will do some reading. Sandbh (talk) 07:38, 10 July 2017 (UTC)

This appears to be a bit of a distraction.

Phillips and Williams (1965b, p. 577) note that the range of ionic-type compounds of S with metals is markedly more restricted than with oxygen. For example, they group the Ln hydrides, borides, carbides, nitrides and sulfides together as, "they show considerable deviations from ionic behaviour" (1965a, p. 118).

S avoids kinetic restraints but this does not save it from being an intermediate nonmetal, given the following observations:

• "...sulfur is displaced slowly from a hydrogen sulfide solution by iodine." (Hogg & Bickel 1941, p. 426)
• "Sulfur should be a typical nonmetallic substance, but since the nonmetallic properties decrease as the atomic weight increases we should expect sulfur to be less active than oxygen. This decrease in chemical activity is so evident that we shall do well to contrast the activity of oxygen with the chemical inertia of sulfur." (Hopkins & Bailar 1956, p. 258)
• "Sulfur is not regarded as a particularly good oxidizing agent." (Rogers 2011, p. 504)

I once looked closer at the literature on interstitial compounds but didn't have much success. The concept seems to have been superseded by broader notions of a range of non-stoichiometric and stoichiometric phases (Matkoivch 1961, p. 1804). I say this with caution as I haven't looked too deeply into this aspect of structural chemistry.

Anyway I think it may be more important to consider that interstitial "compounds" of B, C and N are more generally considered as metallic "-ides". And the other kinds of B, C and N binary compounds tend to be regarded as salt-like (B?, C, N) or covalent. And then we note that sulfur similarly forms metallic, covalent and ionic sulfides; and that most oxides and halides of metals are ionic (Brown et al. 2000, p. 240) whereas this is not the case for the intermediate nonmetals.

There are references in the literature to interstitial sulfides (Searcy 1962, p. 57; Bartholomew & Farrauto 2011, p. 430) but I think such a phenomenon might be too marginal to worry about.

So, while C and N can form interstitial compounds due to their small size, rather unlike S, I tend to think this is small beer, overall, compared to the overall similarities of the carbides, nitrides and sulfides. Even O is supposed to be able to form interstitial compounds (Matkovich 1961, p. 1804), so I think the "S largely can't, O can", thing is likely to be overlap, anomaly, or drill-down territory.

• Bartholomew CH & Farrauto RJ 2011, Fundamentals of industrial catalytic processes, 2nd ed., John Wiley & Sons, Hoboken: "Iron-group metals (Co, Fe, Ni) and Mo form interstitial borides, carbides, nitrides and sulfides..."
• Brown et al. 2012, Chemistry: The central science, 12th ed., Prentice Hall, Boston
• Hogg JC & Bickel CL 1941, Elementary general chemistry, D Van Nostrand Company, New York
• Hopkins BS & Bailar JC 1956, General chemistry for colleges, 5th ed., DC Heath and Co., Boston
• Matkovich VI 1961, "Interstitial compounds of boron", Journal of the American Chemical Society, vol. 83, no. 8, pp. 1804–1806
• Phillips CSG & Williams RJP 1965a, Inorganic chemistry, vol. 1, Principles and non-metals, Clarendon Press, Oxford
• —— 1965b, Inorganic chemistry, vol. 2, Metals, Clarendon Press, Oxford
• Rogers GE 2011, Descriptive, inorganic, coordination, & solid-state chemistry, 3rd ed., Brooks/Cole Cengage Learning, Belmont CA
• Searcy AW 1962, "High-temperature inorganic chemistry", in FA Cotton (ed.), Progress in inorganic chemistry, vol. 3, John Wiley & Sons, New York, pp. 49–127

--- Sandbh (talk) 05:30, 11 July 2017 (UTC)

(1) Given that the point of forming interstitial compounds is that the smaller element's atoms must be able to fit between those of the bigger one, it is not surprising that H, B, C, N, and O can do this while S cannot. If one is studying this sort of compound, then the main generalisation is the size difference between 2p and 3p elements. I think this is another case when the route you ought to take to rationalise things differs along with what you might be trying to show: you might want to start with atomic and ionic radii, oxidation states, metallicity, electronegativity, bond energies, or electron configuration, and it is not clear which is the "best" way to start, or even if that makes sense. A combination of all factors seems to be the best option here, even if it smacks of being a non-option, instead of focusing to much on metallic character or bond energies in the pure element at the expense of how it tends to actually behave in compounds.

(2) Regarding the sulfides, Greenwood and Earnshaw (p. 679) note that the sulfides of La, Ce, Pr, Nd, Sm, Eu, Tb, and Ho adopt the NaCl-type 6:6 structure, and do not yet show the lower coordination numbers or the layer-lattice structures characteristic of increaing covalency (e.g. Be, Zn, Cd, Hg, Mn, Pt). In any case it seems that given the presence of NaCl-style monosulfides of metals like Pb that sulfur is more adept at forming a simple anion than nitrogen, as one would expect from noting that S needs to gain fewer electrons than N to be happy with a noble gas configuration. There are certainly metallic and covalent sulfides; but there do not seem to be interstitial ones, though semiconducting(!) ones appear to exist.

(3) I find the statement about the chemical activities of oxygen and sulfur rather strange. Certainly oxygen reacts with more things, but sulfur tends to do it a lot more quickly. So are we considering kinetic hindrance or not? If we are then S seems to deserve corrosive status; if not, then the case against N is weakened. Double sharp (talk) 14:54, 11 July 2017 (UTC)

(1) Agree
(2) Agree
(3) The authors say that S has less attraction for electrons than does O e.g. O and S combine readily with H to form water and hydrogen sulfide but that the formation of water takes place readily with the liberation of 57,820 cals of heat per gram-molecular weight, while the formation of the sulfide does not proceed spontaneously and liberates only 5,260 cal. On page 262 they show a picture of a rather large stockpile of S in the open air, ready for shipment, with the following caption, "Hugh vats a quarter of a mile long hold over 2,000,000 tons [!]. The sulfur is broken up by blasting[!] and loaded into cars by trains." Later they say, "The ready tendency of sulfur to combine with oxygen presents a difficulty in the out-of-door-storage of sulfur. When loosely piled, the combined effect of sun and rain produces a small quantity of sulfuric acid which is troublesome when the sulfur is to be used for certain purposes. When the sulfur is stored in a huge monolithic block, like that shown…this oxidation is reduced to a minimum." Sandbh (talk) 02:40, 12 July 2017 (UTC)

PS: I forget to add that, as I commented earlier, "Kinetic restrictions may be less relevant." Sandbh (talk) 04:33, 12 July 2017 (UTC)

I shall never cease to be amazed at the lax safety regulations of the 1950s. ^_-☆

I am rather impressed by the formation of sulfuric acid here though: usually oxidation of S by O proceeds only as far as SO₂, and the reaction to form SO₃ is painfully slow. So it usually needs a catalyst if you want to get it practically in quantity, but if you leave S around in the open for long enough you do need to take that into account. So talking about O₂ as a kinetically hindered oxidant is perhaps appropriate in the laboratory but not outside it: another case where the boundaries are rightfully blurred depending on just which case we might be considering. So more and more I am suspecting that while O is certainly strong to overcome its strong O=O bond and spin mismatch in reactions, perhaps it is still not the fairest comparison since S and the halogens do not need to overcome this obstacle for the most part.

And are we counting the reactivity of sulfur oxides and oxoacids as a point for S or for O? These are merely tetrahedrally coordinated and hence aren't that kinetically hindered, but that is because adding oxygens raises the oxidation state two at a time while adding fluorines only does so one at a time; so octahedral SF₆ is really quite inert (no space)! As one would have expected, SeF₆ and TeF₆ to an even larger extent lack this problem. So two trends are crashing to each other here: smaller size leads to greater electronegativity but also lower coordination number and increasing kinetic hindrance, and poor N is caught in the crossfire. So I suspect that we shall have to decide what is more important here. Double sharp (talk) 04:00, 12 July 2017 (UTC)

I was flabbergasted the first time I saw that picture, but had forgotten where, and I'm still surprised by it.

On sulfuric acid formation I'm not surprised. In supposedly pure sulfur it's been blamed for causing explosions in pyrotechnics factories. I once had a mixture of S and NaClO go off spontaneously---I could feel the container was getting warm---which I suspect may have been due to this phenomenon.

Analogously to the N oxides (from what I remember of them) and oxoacids, I'd be inclined to count this as a point for O. Sandbh (talk) 05:02, 12 July 2017 (UTC)

OK, but then given the relative inactivity of SF₆ we would have to bring kinetic hindrance into it again to rationalise it because F is famously stronger than O, right? Double sharp (talk) 05:09, 12 July 2017 (UTC)

I think that would be reasonable, with one caveat, which is that the expression "kinetic factors" can cover a lot, hand-waving-ish style, and I'm not sure that attributing something to kinetic factors may be the full story or even the right story. I've seen "kinetic factors" refer to e.g. rates of reaction, steric factors, activation energies, species mobility, bond breaking and changes in coordination of nearest neighbour atoms or ions, polarity, reactivity or mass transport, and concentration and ratio of the reagents. At other times I've seen e.g. bond breaking referred to as a thermodynamic factor. I've also noted that: "Very little of the chemistry of the group 15 elements in that of simple ions…nearly all the chemistry of the group…involves covalently bonded compounds. The thermochemical basis of the chemistry of such species is much harder to establish than that of ionic compounds. In addition, they are much more likely to be kinetically inert, both to substitution reactions…and to oxidation or reduction when these processes involved making or breaking covalent bonds, as well as the transfer of electrons." --- Housecroft CE & Sharpe AG 2008, Inorganic chemistry, 3rd ed., Prentice Hall, Harlow. So, in the case of group 5 "kinetic factors" affect more than size-challenged N, or its next door neighbour to S i.e., P.

All that said, let's see where we can get this going. Sandbh (talk) 06:07, 12 July 2017 (UTC)

Let's stick to talking about steric hindrance then, as it is the active factor here for rationalising the relative inactivity of many N and Cl species, as well as comparing SF₆ to SeF₆ and TeF₆.

The group 15 elements being unhappy to form simple anions is again just a consequence of going too far from the noble gas column; group 16 only needs two electrons instead of three and is much more eager to form ions, even for Se, Te, and Po. Regarding oxidation or reduction, this statement somewhat surprises me, since P^III tends to quite readily grab a passing O atom and double-bond to it to form P^V IIRC. It is true that As and Bi are quite unhappy to be in the +5 state but that is due to the 3d and 4f contractions; unfortunately I don't remember the chemistry of Sb very well, but it seems to be much happier to be in the +5 state than the +3 state, like P. Double sharp (talk) 06:54, 12 July 2017 (UTC)

P.S. This assertion of the lack of ionic character in the pnictogens becomes really quite striking once we remember that Bi is one of them too. I've already remarked on the lack of ionicity in Bi^III, even though you would naïvely expect a lot from comparing its "ionic radius" with that of La^III, which it somehow manages to equal! I regret to say that I should get to know the pnictogens better, but this is really reinforcing my idea, as I slowly get to understand them, that bismuth is chemically on very shaky ground as a metal.

And given that we call things like tantalum metals mostly because of their physical properties, and Bi not only lacks a close-packed structure but also even a fully metallic band structure, I think the case for Bi and Po as metalloids is starting to look more and more reasonable! Double sharp (talk) 07:04, 12 July 2017 (UTC)

The other fanciful alternative

Is to shove O out of the corrosive nonmetals, and call F, Cl, Br and I halogen nonmetals. That would just increase the overlap in nonmetallic character between the two divisions of nonmetals. The corrosive nonmetals become a subgroup that we don't show on the big table, but we could for example comment that O and the halogen nonmetals are corrosive nonmetals. A bit like H and the alkali metals comprise group 1. Then we have to accommodate O's major league oxidising power (yes it may be kinetically restrained but the numbers don't show that) perhaps by contrasting the rate of oxidation of O vs that of the halogen nonmetals. And then we have to figure out what to call H, C, N, O, P, S and Se, apart from other nonmetals. I would suggest…

• Metalloid
• Grade B nonmetal
• Halogen nonmetal
• Noble gas

…but for having a hard time conceiving of O as a Grade B nonmetal. Anything I have considered as a more interesting category name for the nonmetals in between the metalloids and the halogen nonmetals looks too much like neologism. Which is why I have always liked intermediate nonmetals, because it doesn't look too fancy, yet fits geographically, and in terms of overall nonmetallic character.

Incidentally I see that the IUPAC Gold Book provides a more general definition of a class (a) metal ion, as "A metal ion that combines preferentially with ligands containing ligating atoms that are the lightest of their Periodic Group", and a class (b) metal as one "that combines preferentially with ligands containing ligating atoms other than the lightest of their Periodic Group."

Yet another example of category that needs to refer to another category to define itself. I know we sometimes bag IUPAC but really, do we have to knock ourselves out too? Sandbh (talk) 04:30, 9 July 2017 (UTC)

That's not what I meant by "a category must be definable without making reference to another category". I was intending to refer to cases like "intermediate nonmetal" where you had no choice but to say "an intermediate nonmetal is something in between a metalloid and a corrosive nonmetal", and there are no unique properties that define the category by itself. Referring to individual other elements (e.g. 2p vs 3p) is not a problem in my opinion, since we are giving a short, defining property here. (Otherwise, what exactly is the IAU doing saying that a planet must orbit around a star? That is perfectly fine to me; what would not be fine is saying that a planet is something intermediate between an asteroid and a star.) Double sharp (talk) 05:13, 9 July 2017 (UTC)

Here is an IAU style definition of the categories in question:

“ The IUPAC resolves that nonmetals, apart from the noble gases, be defined into three categories in the following way:

(1) A "corrosive nonmetal"¹ is a nonmetal that: (a) has a first ionisation energy of 1,000 kJ/mol or more, (b) has an electronegativity of 2.6 or more,² and (c) has a standard electrode potential of 0.5 V or more.³

(2) A "metalloid"⁴ is a nonmetal that: (a) has a first ionisation energy of less than 1,000 kJ/mol, (b) has an electronegativity of 2.2 or less, and (c) has a standard electrode potential of less than 0.5 V.

(3) All other nonmetals⁵ shall be referred to collectively as "intermediate nonmetals".⁵

Footnotes:

¹ The corrosive nonmetals are: oxygen, fluorine, chlorine, bromine, and iodine.
² revised Pauling scale
³ at pH 0
⁴ The metalloids are: boron, silicon, germanium, arsenic, antimony, tellurium, and astatine.
⁵ The intermediate nonmetals are: hydrogen, carbon, nitrogen, phosphorus, sulfur, and selenium.

---

The IUPAC further resolves:

A project will be established to consider if astatine is better regarded as a post-transition metal.

”

Does this, or would something like it, look OK? Sandbh (talk) 04:12, 10 July 2017 (UTC)

(1) Well, one thing I always liked about the IAU definition of a planet is its hierarchical nature. You just took three criteria: (α) being in orbit around the Sun, (β) being large enough for its shape to be dictated by self-gravitation, and (γ) being large enough to clear its neighbourhood. Then each group is easily seen as being one step up from the previous, since SSSBs fulfil only (α), dwarf planets fulfil (α) and (β), and planets fulfil (α), (β), and (γ). I would love to be able to do that here.

(2) My biggest problem is that it is not terribly clear why the categories are bounded the way they are. A number like 1000 kJ/mol may look nice decimally, but it is not clear why it should correspond to any qualitative change of behaviour: and with the other delimiters we see a lot of elements that are not far from the bounds. For instance, why do we break S (2.58) away from I (2.66), but not Cl (3.16) from O (3.44)?

(3) And are we using a consistent set of electrode potentials? If we consider a consistent set, such as going from the elements to the more or less protonated forms of the hydrides like Wulfsberg does, then "0.5 V" for the metalloids is not very strong as a statement and may not put a finger on what is going on with them: in actuality their electrode potentials for this are all significantly negative, including Se though excluding Si. Their unwillingness to truly be in negative oxidation states for the most part seems to be one reason for calling them metalloids, as this is not standard nonmetallic behaviour. Double sharp (talk) 05:32, 10 July 2017 (UTC)

(1) The hierarchical thing can done here except I'm so sure it will look as clean, but let have a look at it. It just means that the metalloids meet none of corrosive nonmetal criteria, that the intermediate nonmetals meet only one or two, and that the corrosive nonmetals meet all three. Sandbh (talk) 12:47, 10 July 2017 (UTC)

Hierarchical definition
As requested:

A "corrosive nonmetal" is a nonmetal that (a) has a first ionisation energy of 1,000 kJ/mol or more, (b) has an electronegativity of 2.2 or more, and (c) has a standard electrode potential of 1.0 V or more.

An "intermediate nonmetal" is a nonmetal that (a) has a first ionisation energy of 1,000 kJ/mol or more, or (b) has an electronegativity of 2.2 or more, and (c) has a standard electrode potential of less than 1.0 V.

All other nonmetals shall be referred to collectively as "metalloids".

--- Sandbh (talk) 01:13, 12 July 2017 (UTC)

---

(2) The short answer as to why the categories are bounded/defined the way they are is that they accomodate already known contours laid down in the literature. Scerri sent us a paper about ideas being more often accepted on the basis of accommodating pre-existing knowledge rather than predicting new knowledge, and I'm not essentially doing anything different here. Sandbh (talk) 10:09, 10 July 2017 (UTC)

I haven't seen Scerri's paper yet, but I can look at it within a few hours when I get back. Nevertheless I would suggest that a really good theory or means of classification should not just conform to the data by setting parameters, but provide some sort of clear justification for the parameters, and it is not clear where the figure 2.6 comes from for electronegativity. Wulfsberg in his classification above seems to be following instead the demarcations 1.9–2.2; 2.2–2.8; 2.8–4.0 (IIRC; I remember the 2.8 clearly, and the first one is also in the Pergamon Press series in the volume on the pnictogens, but I can't quite remember if I saw all of this in one place); whether this is better or worse than your scheme is hard to say, because there's not much in the difference, and I only see a rather smooth ladder down of electronegativity perhaps in the order {F}; a gap; {O}; a gap; {N, Cl, Br}; a gap; {I}; {C, S, Se}; a gap; {H, P, As, At}; {Sb, Te}; {(B), Si, Ge, Sn, Bi, Po}. (The last few are a funny case because he seems to realise near the end that he's let in Sn, Bi, and Po as nonmetals like Si and Ge, and then feels obliged to keep all of them in the table.) There are certainly gaps, but they all seem to show up in the wrong place. And even here it is not clear wherein lies the difference between S and I, for example; sulfur seems to show up surprisingly strongly as a nonmetal than would have been expected from these criteria, while iodine doesn't look all that impressive.

If a theory succeeds in explaining one observed effect (forgetting that I am not sure whether we have any yet here), it may be a coincidence, and no one would publish the paper. The reason Planet Nine got published, for example, is that it ended up explaining more things than it was originally intended to do (e.g. perpendicular orbits like Drac); and later some effects of it (e.g. an aligned population with P9 complementing the anti-aligned population) were seen after they had been predicted. A theory that does not predict anything new and only explains one thing, and requires some fudging of parameters beforehand, does not appear to be very much at ease under Occam's famously sharp razor. Double sharp (talk) 10:50, 10 July 2017 (UTC)

The only justification I have for the parameter types are that they are commonly associated in the literature with metallic character and I wanted more than one, consistent with good classification science. Consistent with Occam's razor, I see no need to have more than three parameters. We don't have much in Wikipedia about classification science that I could find so there's not much more I can say. I did find this kernel in the Taxonomy (general) article: "Folk taxonomies of organisms have been found in large part to agree with scientific classification, at least for the larger and more obvious species, which means that it is not the case that folk taxonomies are based purely on utilitarian characteristics." So there may be hope.

The values of the parameters have been retrofitted with the category contours found in the literature. If 2.6 looks too precise, we could round it down to 2.5. That then makes 2.2 look a bit too precise but then the literature commonly refers to metalloids having ENs around 2.0, i.e. 1.8 to 2.2, so that's where the 2.2 comes from.

Iodine meets all three parameters; S meets one, or two if we round the EN parameter to 2.5.

The definition is not a theory it's only a way of structuring knowledge. Sandbh (talk) 12:21, 11 July 2017 (UTC)

Electrode potentials

(3) I'm not sure there is an issue with the electrode potentials? They seem to be a consistent set showing the elements with their lowest oxidation numbers. The metalloids are the most metallic of the nonmetals so we would expect their potentials to be the lowest of the nonmetals. Even H, P, and Se have negative potentials. Sandbh (talk) 10:59, 10 July 2017 (UTC)

Firstly, I would think then that a lower value than 0.5 V might be better for defining what the metalloids are actually like. I am not really worried about H, P, and Se considering that their chemistry often seems like it is halfway to being that of a metalloid: Se is classified as such fairly often. Secondly, what then stops us from declaring Bi and Po to be metalloids, especially since the potential for Bi is close to that of Te and is actually less negative than that of Ge? Double sharp (talk) 11:04, 10 July 2017 (UTC)

I don't want to go lower than 0.5 V given I'm using a figure of 0.3 for At. I never saw an issue about some nonmetals having negative potentials in the first place! Bi and Po are metals, as per the definition we thrashed out in archive 27. Sandbh (talk) 13:29, 10 July 2017 (UTC)

Well, then perhaps that value for At should give us a warning about taking too metallic a view of it! ^_^ Though if a nonmetal has a negative potential, that seems to indicate that it is unhappy to exist in negative oxidation states, since these species are basically protonated forms of the supposed simple anions (e.g. PH₃ is supposedly triprotonated P³⁻, never mind that the latter does not really exist).

As for that definition, I still remain vaguely unsatisfied with it, as it does not provide a set of criteria which all the elements we like to consider metals to follow. (Also, aren't a majority of those five criteria about physical instead of chemical properties? If we are looking at that, the case for metalloids as nonmetals becomes weaker than it was already.) The general picture of Po^IV and Po^VI chemistry accords more with what we know of Te; only as Po^II does cationic behaviour start to show itself (and it tends to oxidise rather quickly to Po^IV anyway). As for bismuth, it doesn't even form a simple cation in aqueous solution (despite the fact that the Bi^III ionic radius is supposely as large as the La^III one, which should already make us wonder if this is really "Bi³⁺"), its electronegativity is almost the same as that of As and Sb, and it doesn't even have a metallic band structure. So I am seriously starting to find myself agreeing with Parish (1966, p. 179) who adds Bi and Po (but not At) to the list of six commonly accepted metalloids. Double sharp (talk) 14:38, 10 July 2017 (UTC)

Negative potentials for those nonmetals are what I would expect for weaker nonmetals.

I forgot that we sorted out the metal definition earlier on this page, not in archive 27. It has one mandatory criterion (it looks like a metal) plus four optional criteria, two physical and two chemical. It does cover all the elements we consider as metals, including Bi and Po. Sandbh (talk) 11:16, 11 July 2017 (UTC)

I still find myself vaguely uncomfortable with the idea of optional criteria, so I hope you won't mind if we put this bit on pause until I find a more elegant formulation (at least to myself) of how I tend to think of what a metal ought to be. ^_^ Double sharp (talk) 14:41, 11 July 2017 (UTC)

Good. The definition of a metal isn't the main game here. I harbour serious doubts that it will be possible to define a metal without one or more optional criteria. Sandbh (talk) 02:04, 12 July 2017 (UTC)

I am more positive, because the simple criterion "has a metallic band structure" almost perfectly recreates the borders I would want for the category: the only irritant for me is polonium, which is even less metallic than bismuth which is right next to it. (I continue to be rather sceptical of the claims for metallic condensed astatine, for reasons you've heard already.) My scepticism for Bi and Po is that unlike, say, Ta and W in the d-block, they not only lack the chemistry but also the physical properties of metals for the most part. Double sharp (talk) 14:29, 14 July 2017 (UTC)

P.S. I'm amused to note BTW that Ralf Steudel's Chemistry of the Non-Metals considers the following elements: {H; B; C, Si, Ge; N, P, As; O, S, Se, Te, Po; F, Cl, Br, I, At; He, Ne, Ar, Kr, Xe, Rn}. Double sharp (talk) 14:29, 14 July 2017 (UTC)

Metallic band structure
Double sharp, I suggest you eschew a single criterion definition. Per Jones (2010, p. 169), who was writing on the role of classification in science and whether Pluto could be regarded as a planet, "Classes are usually defined by more than two attributes." (Pluto: Sentinel of the outer Solar System). I've seen single criteria used for the purpose of distinguishing metals but have never paid much attention to them since which one is choosen is too arbitrary. Electrical conductivity is a popular one, and I suspect I've seen one use +ve v −ve temperature coefficient of resistivity. The Goldhammer-Herzfeld ratio is another one. Parish, with his two criteria, was an exception. The only authors who got close to getting this right were Masterton and Slowinski (1977, p. 160) who used ionisation energy, electronegativity, and band structure to attempt to distinguish metalloids.

Polonium
I was never able to get a bead from the literature as to why some authors regarded Po as a metalloid, other than its ability to form polonides. What do we know about polonium? It has a silvery, metallic appearance; it conducts electricity like a metal; it has the electronic band structure of a metal; its enthalpy of fusion is near the average for close-packed metals; it is soluble in acids, forming the rose-coloured Po²⁺ cation and displacing hydrogen; many polonium salts are known; and the oxide (PoO₂), which assumes the fluorite structure more typical of ionic compounds/metallic oxides, is predominately basic in nature.

According to our own metalloid article, "Whether polonium is ductile or brittle is unclear. It is predicted to be ductile based on its calculated elastic constants. It has a simple cubic crystalline structure. Such a structure has few slip systems and "leads to very low ductility and hence low fracture resistance"

Polonium also has some intermediate or nonmetallic properties. It has an intermediate coordination number, electronegativity, ionisation energy and metallicity ratio; and it can form anionic polonides, volatile and easily hydrolysed halides (which are soluble in organic solvents), and a volatile and unstable hydride (PoH₂). Most of the latter properties are characteristic of the heavier noble metals or post-transition metals.

If the elements are categorised on the basis of whether they are judged to exhibit a preponderance of metallic or nonmetallic properties then I suggest that the weight of evidence, in the case of polonium, falls on the metal side of the line. A parallel might be drawn with gold, which exhibits several nonmetallic properties, including auride formation, yet is universally categorised as a metal on account of its distinctive metallic properties.

Bismuth
Bismuth is tough to regard as something other than a metal because it forms a cation in aqueous solution, and has an undeniably basic oxide. Some of the literature recognises bismuth is a puny metal, but a metal nonetheless. For the few authors that regarded bismuth as a metalloids I could never figure out why they did so apart from one, who suggested treating metalloids as the semimetals and semiconductors, which would therefore include C and P i.e. another unsatisfactory single criterion approach. Looking at lists of metalloids I see that aluminium is more likely to regarded as a metalloid than bismuth by a close to 2:1 margin, and the notion of aluminium being a metalloid has been rightly disputed.

Here is what I have from the literature on bismuth:

• "… arsenic is mainly an acid-forming element, and is therefore a non-metal, while antimony is both acid-forming and base-forming, and bismuth is base-forming." (Smith 1906, p. 707)
• "Antimony… is of more metallic appearance than arsenic, but, although it has some of the properties of the metals (lustre, electrical and thermal conductivity), in its chemical behaviour it is closely connected with arsenic and phosphorus… Bismuth …has no non-metallic characters [sic] and may be considered as a metal, as it forms no gaseous hydrogen derivative and its oxide has basic characteristics." (Molinari 1920, pp. 426, 792)
• "Antimony… is more nonmetallic than metallic… bismuth… more nearly approaches a metal in physical and chemical properties." (Norris & Young 1938, p. 529)
• "Alkyls and aryls of phosphorus, arsenic and antimony are named as derivatives of the hydrides, phosphine, arsine and stibine, while the bismuth compounds are named as derivatives of the metal." (Rochow, Hurd & Lewis 1957, p. 199)
• "As and Sb sulphides may be dissolved as thioacids, but this is not true for the more basic Bi₂S₃." (Phillips & Williams 1965, p. 636)
• "When non-metallic elements react with the oxidizing acids, acidic oxides or acids are formed… The trisulphides of arsenic and antimony are acidic, forming salts with yellow ammonium sulphide and alkali, while that of bismuth is typical of a metal." (Moody 1969, pp. 267, 321)
• "The Nitrogen Group… the elements range from the electronegative non-metal nitrogen, to the very weakly electropositive metal bismuth, via the semi-metals arsenic and antimony." (Kneen, Rogers & Simpson 1972, p. 403).
• "Interest centres on the trend from non-metallic to metallic properties with increasing atomic weight. Thus there are many parallels between phosphorus and arsenic, but considerably fewer between phosphorus and bismuth, which is a typical B metal like tin or lead. Arsenic and antimony are important largely because of their intermediate or metalloid character…" (Smith 1973, p. 547)
• Bailar et al. (1984, p. 951) refer to bismuth as being, "the least 'metallic' metal in its physical properties… brittle rather than malleable, and [with] …the lowest electrical conductivity of all metals." Which element has the lowest electrical conductivity is debatable but bismuth is certainly in the lowest cohort.
• "In terms of the criteria discussed at the beginning of chapter 20, bismuth is more logically considered a metal rather than a nonmetal. Bismuth usually appears in the +3 oxidation state; there is little tendency to attain the higher +5 oxidation state common to phosphorus. The common oxide of bismuth is Bi₂O₃. This substance is insoluble in water or basic solution but is soluble in acidic solution. It thus is classified as a basic anhydride. As we have seen, the oxides of metals characteristically behave as basic anhydrides." (Brown & LeMay 1985, pp. 708–9)
• "Arsenic and antimony are frequently referred to as metalloids, implying a partially metallic character. Bismuth… more properly fits the description of a true metal." (Borg & Dienes 1992, pp. 29–30)
• "Arsenic and antimony, isostructrual with black phosphorus, are also good examples of the complex relationship between bonding and physical properties. Their network forms are stable under normal conditions, melting at 816º C (under 39 bar) and 631 ºC respectively; but arsenic sublimes at 615º C (1 bar) to give As₄ molecules in the gas phase, while antimony requires a temperature of 1600 ºC before it boils. Clearly arsenic is more like covalent molecules of As₄ in a solid held together by van der Waals forces, than antimony, which is leaning towards being a metal… Both bismuth and polonium are ‘near-metals’." (Laing 1993, p. 163)
• "In the nitrogen family, we move from nonmetals that form acidic oxides—nitrogen and phosphorus—to metalloids that form amphoteric oxides—arsenic and antimony—to the last element—bismuth—that is barely a metal and forms a basic oxide." (Brady & Holum 1996, p. 61)
• "Arsenic and antimony are classified as metalloids or semi-metals and bismuth is a typical B sub-group (post-transition-element) metal like tin and lead." (Greenwood & Earnshaw 1997, p. 548)
• "Antimony is found in forms analogous to those of arsenic… The most common antimony chalcogenide ore is stibinite (antimonite), Sb₂S₃… Some metal antimonides are breithauptite = NiSb… and dyscrasite (antimony silver) = Ag₃Sb… Bismuth is not found in large amounts in nature… Unlike arsenic and antimony, it is not found as anionic bismuthides." (Wiberg 2001, p. 757)
• "In addition to compounds with cationic bismuth, there are a few metal bismuthides M_mB_n (M usually an alkali or alkaline earth metal)… Because of the metallic nature of bismuth, even the alkali and alkaline earth bismuthides have definite metallic properties. For example, LiBi and NaBi are superconducting… The other bismuthides, such as PtBi₂, MnBi, NiBi, IrBi and PtBi, are even more metallic." (Wiberg 2001, p. 768)
• Hoffman (2004) refers to bismuth as "a poor metal, on the verge of being a semiconductor." Arsenic and antimony are likewise close to being semiconductors but bismuth is the closest, in that sense.
• "Bismuth(III) oxide occurs naturally as bismite and is formed when Bi combines with O₂ on heating. In contrast to earlier members of group 15, molecular species are not observed for Bi₂O₃ and the structure is more like that of a typical metal oxide." (Housecroft & Sharpe 2008, p. 474)

Steudel
Like. According to GB he mentions metalloids twice including "metalloidal" P and As. I think he is taking the view that the metalloids generally behave chemically as nonmetals, whilst acknowledging that some of them are called metalloids. I see the e-book version is available for Oz $102 (no thank you!) v $10 from ABE. William Jolly's The chemistry of the non-metals (1966) is the same as Steudel, except he includes Sb as a nonmetal Like since "Although the cationic tendencies of Sb(III) suggest metallic behaviour [commenting earlier that SbO⁺ can exist in aqueous solution], most of the chemistry of antimony is characteristic of a non-metal" (p. 107). He does not, however, mention metalloids or semimetals.

Bismuth references

• Bailar JC, Moeller T, Kleinberg J, Guss CO, Castellion ME & Metz C 1984, Chemistry, 2nd ed., Academic Press, Orlando.
• Borg RJ & Dienes GJ 1992, The physical chemistry of solids, Academic Press, Boston.
• Brady JE & Hulum JR 1996, Descriptive chemistry of the elements, John Wiley, New York.
• Brown TL & LeMay HE 1985, Chemistry: The central science, 3rd ed., Prentice/Hall International, London.
• Greenwood NN & Earnshaw A 1997, Chemistry of the elements, 2nd ed., Butterworth-Heinemann, Oxford.
• Hoffman P 2004, Semimetal surfaces, viewed 15 July 2017.
• Housecroft CE & Sharpe AG 2008, Inorganic chemistry, 3rd ed., Pearson, Harlow.
• Kneen WR, Rogers MJW & Simpson P 1972, Chemistry: Facts, patterns, and principles, Addison-Wesley, London.
• Laing M 1993, ‘A tetrahedron of bonding’, Education in chemistry, November, pp. 160–163.
• Molinari E 1920, Treatise on general and industrial inorganic chemistry, 2nd ed., J & A Churchill, London.
• Moody B 1969, Comparative inorganic chemistry, 2nd ed., Edward Arnold, London.
• Norris JF & Young RC 1938, A textbook of inorganic chemistry for colleges, 2nd ed., McGraw-Hill, New York.
• Phillips CSG & Williams RJP 1965, Inorganic chemistry, vol. 1, Principles and non-metals, Oxford, London.
• Rochow EG, Hurd DT & Lewis RN 1957, The chemistry of organometallic compounds, John Wiley, New York.
• Smith A 1906, Introduction to general inorganic chemistry, Century, New York.
• Smith JD 1973, "Arsenic, antimony and bismuth", in JC Bailar, HJ Emeléus, R Nyholm, AF Trotman-Dickenson (eds), Comprehensive inorganic chemistry, vol. 2, Pergamon, Oxford.
• Wiberg N 2001, Inorganic chemistry, translation from the 34th German edition (1995), Academic Press, San Diego.

--- Sandbh (talk) 05:07, 15 July 2017 (UTC)

Bismuth, polonium, and astatine

I'm not too keen on a single-criterion definition either. However the fact that a single criterion can correlate so well with what we tend to think of as metallic behaviour chemically surely augurs well for the coincidence of these varying criteria, that should obviate the need for optional criteria.

Regarding bismuth, surely it gives one pause that:

1. The resistivity of bismuth (120 μΩ·cm) is actually higher than some commercial resistors at 100 μΩ·cm, and is even higher than arsenic (33 μΩ·cm) and antimony (41.7 μΩ·cm). (Greenwood & Earnshaw p. 552)
2. Like antimony, bismuth is too brittle to work at room temperature like a metal. (Greenwood & Earnshaw p. 550)
3. How the organobismuth compounds are named is irrelevant (or else La would similarly not be a lanthanide), but rather their behaviour is; the strength of covalent linkages decreases down the group as P > As > Sb > Bi, resulting in the instability of BiH₃ and its derivatives, which is fairly different from the relative stability of organothallium and organolead compounds. (Greenwood and Earnshaw p. 553)
4. Incidentally, your 1920 source appears to be unaware of the existence of bismuthine, even though the famous Marsh test will detect it as surely as it will detect arsine and stibine, forming mirrors of As, Sb, and Bi. (Hollemann & Wiberg, as cited by our article on bismuthine)
5. The argument of bismuth preferring the lower oxidation state rather than the higher would be rather more convincing if it could actually be convinced to form a simple cation in aqueous solution like Tl⁺ and Pb²⁺. Instead, even though Bi³⁺ would have an ionic radius similar to La³⁺(!!), it is not only much less basic than lanthanum, but even in neutral perchlorate solutions the main species is still the oxocation [Bi₆O₆]⁺ or a hydrated form thereof; already at pH 1 only hydrolysed species exist. At pH 5.5–7, bismuth exists as Bi₂O₃ like antimony (forming Sb₂O₃), unlike thallium and lead which form Tl⁺ and Pb²⁺ even here;
6. Even in the bismuthides, there is clearly some ionicity present as well supplementing the metallic interactions; otherwise it would be difficult to explain the high melting point of Na₃Bi (840 °C), for example, compared with those of Na (98 °C) and Bi (271 °C). Conversely, even in the arsenides and stibides, things like LiAs, NaSb, and KSb clearly show metallic lustre and electrical conductivity, despite the similarity of the As and Sb chains to those of Se and Te; so the line between metallicity and nonmetallicity in As, Sb, and Bi intermetallic compounds is fuzzy at best and runs through all three elements. (Greenwood and Earnshaw, p. 554–5)
7. Differences in bond type in compounds such as AsI₃, SbI₃, and BiI₃ are surely overblown considering the similarities in electronegativity (As 2.18, Sb 2.05, Bi 2.02, in which there would paradoxically seem to be more difference between As and Sb than between Sb and Bi). A simpler explanation simply arises from the size of the atoms and their correspondingly different preference in coordination number. In any case, among the bismuth halides, only BiF₃ is actually ionic with tricapped trigonal prismatic Bi. Already BiCl₃ is composed of discrete molecules just like halides of antimony, showing once again that only fluorine can get bismuth out of its covalent preference; Bi^III then ends up being just as bad as Sn^IV at forming ionic compounds and is actually worse than Sn^II, for which SnCl₂ at least has interlinked {SnCl₃} units. (Greenwood and Earnshaw, p. 559–60)
8. The complex chemistry of Bi is mostly analogous to that of Sb, with many isostructural compounds. (Greenwood and Earnshaw, p. 565–567)
9. Bi₂O₃ surely cannot be very impressive as an ionic compound, given that it happens to be a semiconductor. The fact that it reacts with Li₂O and even Ag₂O to give bismuth oxyanion compounds with ions like BiO³⁻
   ₃ and Bi
   ₂O¹⁰⁻
   ₈ shows that it is not even wholly basic, rather lying slightly on the basic side of amphotericity. Certainly Bi(OH)₃ is basic enough to happily dissolve in acid, but unless the pH is kept absurdly low oxo-salts immediately precipitate out. The preference for Bi to form oxocations and exist in aqueous solution as the insoluble oxide is most analogous to Sb. (Greenwood and Earnshaw, pp. 575–6)
10. Attempts to thermally dehydrate supposed "Bi³⁺" salts end up yielding complex oxocation salts, usually involving [Bi₆O₆]⁶⁺. The existence of Bi₂(SO₄)₃ seems to be at about the same level of that of its Sb analogue, since both are very readily hydrolysed to basic salts. (Greenwood and Earnshaw, p. 591) I have to wonder if something similar to those supposed "Zr⁴⁺" salts is going on here, where the cation itself turned out to be a hydrolysed oxocation. It is quite evident that Bi cannot even take a minuscule +3 charge without losing most of its cationic behaviour.

Given all this, I would think that it is more useful to consider Bi as a close relation to As and Sb: certainly more metallic, but not vastly so, and probably not enough to actually deserve to be called a metal, given its lack of anything like real metallic physical properties. I would actually be a lot happier to take Ge apart from Sn and Pb than to take Bi apart from As and Sb, at least after reading their respective G&E chapters. Double sharp (talk) 08:59, 15 July 2017 (UTC)

P.S. I haven't touched on polonium very much because we do not have anywhere near as good a picture of it as we have for bismuth. Nevertheless I would note that while dissolving Po in water initially yields Po²⁺, this rapidly reduces water further to form Po^IV derivatives, mostly things like PoO²⁻
₃ and HPoO⁻
₃, so this would seem more to be along the lines of Pa^IV rapidly becoming Pa^V (and then pulling a disappearing act via hydrolysis). The volatile hydride is also something common to most of the p-block (including As, Sb, and Bi, as well as Se, Te, and Po) and is not very similar to the noble metals which instead show a gap in hydride formation. I agree that Po^II salts are pretty ionic but it tends to prefer to exist as Po^IV, and the difference is similar to that between Sn^II and Sn^IV. BTW, Greenwood and Earnshaw describes PoO₂ on p. 780 as amphoteric, though on the more basic side of it and certainly more basic than TeO₂. Double sharp (talk) 09:09, 15 July 2017 (UTC)

OK, I checked Bagnall (10.1524/ract.1983.32.13.153): the existence of Po^IV oxoacid salts is rather spotty. There is no Po(NO₃)₄; while a solvated form with N₂O₄ exists, this needs to be kept in dinitrogen tetroxide itself or else it decomposes to a basic nitrate. While Po(SO₄)₂ does exist, it can only be made with rather concentrated sulfuric acid (>0.5 N) or else basic 2PoO₂·SO₃ is formed instead: the latter is analogous to the behaviour of Te. And similarly for the chromate; the selenate salt is not even known in any form other than the basic one. As for supposed Po^II salts like "PoSO₃" and "PoSeO₃", their stoichiometry is under dispute and Bagnall cautiously refers to them as a sulfoxide and a selenoxide instead. The "phosphate" may well actually be PoO₂·H₃PO₄ instead of Po₃(PO₄)₄. What little is known of organopolonium chemistry is a pretty much exact analogue of organotellurium chemistry. Furthermore the amphoterism of PoO₂ is evident by the formation of polonate(IV) compounds upon reaction with KOH and even Bi₂O₃(!!), as do V₂O₅ and Ta₂O₅. (The most amusing reaction though is with WO₃, which seems to incredibly form a polonium(IV) tungstate instead of a tungsten(VI) polonate, further underscoring just how absurdly weak tungsten is chemically as a metal despite its physical strength.)

This does not inspire much confidence in me about its metallicity, on the face of it! Double sharp (talk) 13:07, 15 July 2017 (UTC)

Regarding astatine, I feel somewhat vindicated by this paper (10.1524/ract.1989.47.23.105) which notes that [I(H₂O)_n]⁺ ought to be stable enough to exist when only trace quantities of iodine are present, along the lines of the typical concentrations at which At is experimented on. Certainly At⁺ should be more stable than I⁺, but equally similarly given that there is no "I⁺" in bulk iodine chemistry I am likewise skeptical of the prospect of "At⁺". I would expect AtOH (hypoastatous acid), and perhaps that would be weak enough to protonate to AtOH⁺
₂, but I am not terribly convinced that that really counts as a solvated At⁺ cation any more than H₃O⁺ is solvated H⁺. I honestly suspect that Bi, Po, and At are more metallic than their chemistry, if I may joke a little. Double sharp (talk) 13:35, 15 July 2017 (UTC)

This is impressive work. I'll have a closer look. In the meantime, I forgot some things:

In The Metalloids (1966, p. 7), Rochow wrote: "Since its (Bi) conductance is only 1.4% that of silver, and since in its amphoteric behavior it often [sic] resembles arsenic and antimony, it sometimes is grouped with the metalloids."

I think that to be recognised as an amphoteric oxide (at least for the purposes of classing an element as a metalloid), amphoterism has to be demonstrated in aqueous solution rather than via fusion with a basic oxide.

PoO₂ has the fluorite structure this normally being associated with an ionic oxide.

In The Aqueous Chemistry of the Elements (2010, p. 241) Schwietzer & Pesterfield's Pourbaix diagram for Po shows the presence of a colourless Po^–2 (aq) ion at pH > 12.8. I've never been able to find a primary source for the existence of this anion. On the next page they say, "The metallic character of Po is indicated by the numerous salts it forms…The polonium(II) salts are readily oxidized, often by the effects of the radioactivity, and the polonium(IV) salts are subject to hydrolysis." Sandbh (talk) 12:00, 16 July 2017‎ (UTC)

Like I said, we do not have a very complete picture for polonium. So I would be reluctant to consider absence of evidence as evidence of absence; there is similarly a lack of polypolonides analogous to polytellurides, but given how well Po tends to pattern with Te, and the difficulty of making such compounds due to the instability of their precursors to radiolysis, there are simpler explanations. While the results of fusing PoO₂ with basic oxides are not the final nail in the coffin they are cetainly suggestive, given that Bagnall also notes that adding dilute aqueous alkali to an aqueous solution of Po^IV produces a hydrated oxide with feebly acidic properties, with K_a = [PoO²⁻
₃]/[OH⁻] = 8.2 × 10⁻⁵ at 22 °C. Given that Po⁴⁺ is a large cation I am not surprised at the fluorite structure, but this is perhaps less an illustration of ionic character than atomic radius given SnO₂ and PbO₂ (Sn 1.96, Pb 1.87) have a rutile structure while SiO₂ (Si 1.90) has the quartz structure, and GeO₂ (Ge 2.01) forms both modifications (in comparison, Po has electronegativity 2.0).

Bagnall mentions an aqueous Po²⁻ anion himself when he discusses electrochemical media in alkaline media, and if anyone would know this about Po chemistry he surely would. ^_^ Certainly it is probably difficult to characterise, just like H₂Po, but given the trends down group 16 I would have difficulty conceiving of it not existing. It is certainly well-characterised in the more ionic polonides such as K₂Po, at the very least, and when dissolved in aqueous solution it should be essentially nonbasic and persist as such if it were not for radioactive self-oxidation. Double sharp (talk) 14:53, 16 July 2017 (UTC)

Earlier update

I think this alternative may be a case of pragmatism gone too far but I'll think some more about it. I probably have gone too far: the level of abstraction is the key consideration, not the technical subtleties. Chemists will grasp this straight away, I reckon, just like they can accommodate the fact that N and P are in the same periodic table group. Sandbh (talk) 04:42, 9 July 2017 (UTC)

Lead FAC – reference check complete

I've fixed or otherwise closed all of the referencing issues at the lead FAC, and pinged both you and the coordinators. Really, this article can't wait any longer. Parcly Taxel 16:20, 2 July 2017 (UTC)

Thank you. But please do not worry too much about when it ends already. There has been a lot of support and no opposition. At dome point, an FAC coordinator indicated they wanted to promote the article and they didn't because there was a review open; it is still open. There are no signs of this ending with not getting the bronze star, so don't worry (I don't). Common courtesy, however, hints we should let the last reviewer finish their review.--R8R (talk) 18:40, 2 July 2017 (UTC)

An atlas of Pourbaix diagrams

You may find this helpful. ^_^ Double sharp (talk) 14:40, 10 July 2017 (UTC)

(Mind you, while having all this stuff in one place is a good start, for some of the less common elements you may want to take its pronouncements with some grains of salt and research things yourself too; at least that's the impression I get.) Double sharp (talk) 03:11, 11 July 2017 (UTC)

Thanks! Agree. Working through the rest of your comments. Sandbh (talk) 05:32, 11 July 2017 (UTC)

glass-forming oxides for the table in nonmetal

It seems that we may need to add H₂O. Double sharp (talk) 09:33, 20 July 2017 (UTC)

Far out and very cool. I'd overlooked the fact that water is an oxide of hydrogen. So added. Thank you! Sandbh (talk) 00:49, 21 July 2017 (UTC)

Thank you! BTW, aren't there also (weaker) H bonds to C, P, and I, if my memory of Greenwood and Earnshaw's list in the hydrogen chapter serves well? Double sharp (talk) 02:31, 21 July 2017 (UTC)

Yes, I need to do some more reading about this = get that book by Gilli G & Gilli. At the moment I'm trying to figure out how to respond to the posts at our project page. Category nomenclature thoughts (picture furrowed brow) have been occupying me a lot. Sandbh (talk) 03:25, 21 July 2017 (UTC)

nonmetal redraft

It may require a lot of moving stuff around, but why not just treat all the nonmetals together for the bios (since you put O and S in the same section)? Then you could organise the bios by groups, putting all of group 18 as one bio, and for example cover B between H and C, {Si, Ge} after C, {As, Sb, (Bi)} after P, {Te, (Po)} after Se, and At after I. It seems to be a fairly consistent approach with what you have in the sandbox, if I have successfully divined your intentions. ^_-☆ Double sharp (talk) 05:53, 1 August 2017 (UTC)

That's an interesting suggestion. The sections would then look like this:

Categories

…Metalloids

…Other nonmetals

…Corrosive nonmetals

…Alternative categories

Comparison of properties

Nonmetals (and metalloids) by group

…Group 1

…Group 13

…Group 14

…Group 15

…Group 16

…Group 17

Allotropes

This looks like a good way to cover both aspects of the nonmetals i.e. by type and by group. I'll take it on board. Thanks! Sandbh (talk) 00:48, 2 August 2017 (UTC)

Why, thank you for considering it. I am wondering though: where would you want to cover the noble gases? Because they are both a "type" category and a "group" category (group 18). Double sharp (talk) 02:01, 2 August 2017 (UTC)

Reorganisation done. I presume it looks OK now, including treating the NG both as a type and a group. So, yes, will need bios for all the metalloids and all the NG. And maybe add something about the curious diagonal relationship between B and Si, aside from their metalloid status. Sandbh (talk) 02:53, 2 August 2017 (UTC)

Wonderful!

BTW, I notice you mention Wulfsberg as one of the schemes. Actually this may be a bit different from the others in that it contains a group for the metalloids as nonmetals as well: his EN delimiters for the nonmetals in groups 13 to 17 are 1.9–2.2, 2.2–2.8, and 2.8–4.0 IIRC. We'll have to find some way to put that in the chart, as well as note his inclusion of Bi and Po as nonmetals.

I wonder if that would necessitate putting little bios of Bi and Po too? I suppose they might provide useful comparative cases, but while Po is pretty often considered a metalloid (rightly or wrongly) the same is not really true of Bi. Yet I would feel uncomfortable about mentioning that Wulfsberg includes Bi in a showcase classification example as a nonmetal, without saying anything further about why he might have chosen to do this. Double sharp (talk) 07:23, 2 August 2017 (UTC)

I have Wulsberg's two books and will be able to check them by no later than early next week. I suspect we could accomodate his inclusion of Bi and Po by providing our own carefully worded commentary. Sandbh (talk) 10:02, 2 August 2017 (UTC)

Nonmetals show more variability in their properties than do metals

That single sentence, located in your sandbox nonmetal article, has caused me to rethink my desire to have only two nonmetal categories. I haven't got time right now to add anything to the discussion, but I thought you might like to know. YBG (talk) 23:20, 23 August 2017 (UTC)

This is not too surprising, mind you, given that metals must perforce be more similar to each other by virtue of sharing the very properties that make them metals, whereas every nonmetal fails to meet some of them in its own particular way. This is so close to the Anna Karenina principle that one wonders how the metals manage to outnumber the nonmetals anyway, and indeed to get that accepted result you need to focus only on one or two criteria, mostly physical, in which case the situation does not apply (but we lose all the talk about typical chemical metallicity). Double sharp (talk) 23:40, 23 August 2017 (UTC)

Thank you YBG. That is good and welcome news.

Double sharp, my comments to follow… Sandbh (talk) 04:00, 24 August 2017 (UTC)

But you really do need to get rid of the "other nonmetals" name -- that is a show-stopper for me.

Completely off-topic, we drove about 50 miles south early Monday morning to have a dekko at the spectacle in the sky; it was absolutely worth it. YBG (talk) 06:14, 24 August 2017 (UTC)

It must have been! ^_^ I have only yet seen a partial eclipse, but not a total one. It is still interesting to see sharp shadows and a bright Sun with a light level that suggests an overcast day, but that cannot possibly beat the sudden darkness of totality. Double sharp (talk) 06:21, 24 August 2017 (UTC)

Back to business, the classification science is more important than nomenclature. So I'll be quite happy to get rid of other nonmetals, even if that doesn't happen until after the draft has reached an acceptable quality (because I have no better ideas/haven't found any good alternatives at this time other than previous suggestions). Sandbh (talk) 07:02, 24 August 2017 (UTC)

Of course, I don't think we have problems with "other nonmetals" as a provisional name. ^_^

What is worrying me more is YBG's observation combined with Tolstoy's. If being a metal demands that one has a number of properties, and being a metalloid or nonmetal only demands the negation of at least one of them, then it should be easier to be a nonmetal than a metal. Yet we routinely classify there as being more metals than nonmetals. This suggests to me that many of the attributes that we use to define metallicity are not defining but rather suggestive properties that hold in most, but not all, cases. Most metals have relatively high densities, but those in groups 1 and 2 (plus Eu and Yb, I guess) do not. Most metals have relatively high melting points, but Hg is one spectacular exception, and it is certainly not alone. Most metals conduct electricity like you would expect, but Bi doesn't (it actually resists more than most commercial resistors). Most metals have significant cationic chemistry, except for that huge gap in the 4d and 5d elements. My worry is that this may mean that we have it backwards and nonmetallicity is actually the more fundamental concept that we step away from to varying degrees, and not metallicity, despite the names! Double sharp (talk) 07:13, 24 August 2017 (UTC)

Getting rid of the "other nonmetal" term

The challenge has been to come up with a descriptive phrase that one can adequately hang H, C, N, P, S, and Se under. As my sandbox draft currently says, "The other nonmetals have a diverse range of individual physical and chemical properties." So we have H representing the chemistry of the proton; C conducting electricity better than some metals; N being relatively inert and a weak oxidising agent despite having the fourth highest electronegativity in the periodic table (and yet the cause of this inertness can be exploited explosively); P being far more abundant in its most unstable form than its most stable form, not to mention being pliable; S showing some metallic character, either intrinsically or in its compounds with other nonmetals e.g. the malleability of plastic sulfur and the lustrous-bronze appearance and metallic conductivity of polysulfur nitride, SN_x; and Se being counted as a metalloid, especially in the environmental literature.

In this light I propose to refer to these other nonmetals by using the descriptive phrase, heterogenic nonmetals, from the Greek, 1. "hetero-" +2. "gen" + 3. "-ic", as follows:

1. The other of two, other, different; a formative of many scientific and other terms
2. An adjective suffix which has two different uses: (1) giving the sense "born in a certain place or condition", (2) giving the sense "of a (specified) kind", of another kind
3. This was in Gr. one of the commonest of suffixes, forming adjs., with the sense "after the manner of", "of the nature of", "pertaining to," "of".

The nonmetal article would then read, in parts, something like this:

Based on shared attributes, the nonmetals can be divided into the three categories of heterogenic nonmetal, corrosive nonmetal, and noble gas. The first two of these category names are descriptive phrases rather than IUPAC-approved collective names…

…

There are six heterogenic nonmetals: hydrogen (H), carbon (C), nitrogen (N), phosphorus (P), sulfur (S), and selenium (Se). In periodic table terms they largely occupy a position between the weakly nonmetallic metalloids to the left and the strongly nonmetallic corrosive nonmetals to the right. They are called heterogenic nonmetals in light of the diverse range of their individual physical and chemical properties. They are nevertheless characterised by having moderate to high ionisation energies, low to high electron affinities, moderate to high electronegativity values; being relatively poor to moderately strong oxidising agents; and having a tendency to forming predominately covalent compounds with metals.

How does that look? I believe I like this better than intermediate nonmetal, and I suspect (hope) that it finally captures the heart of the matter. Sandbh (talk) 06:44, 26 August 2017 (UTC)

Well, you've just invalidated the comments I just made on the project page, but no worries, mate.

I'd generally like to avoid inventing new words, and would instead create a new technical phrase out of two existing words. So to take up the meaning and etymology you are proposing, I'd prefer the term heterogeneous nonmetals.

But do you have something against using oxidative and reductive as proposed by Parcly Taxel? YBG (talk) 07:28, 26 August 2017 (UTC)

Heterogenic is a real word, and shorter than heterogenous. I tend to agree with Double sharp's earlier comment in archive 28, re "I agree with the point, though not very much with the names: surely S is also rather oxidative as a silver-slayer?" More concerning for me, I think, is that hydrogen is a reducing agent when it reacts with non-metals and an oxidizing agent when it reacts with metals. Sandbh (talk) 12:11, 26 August 2017 (UTC)

It is a new one to me. According to wiktionary, heterogenic has only one definition, "of, or relating to the genes of different species", which seems completely non-applicable, etymology notwithstanding. However, heterogeneous (alt spelling heterogenous) means "Diverse in kind or nature; composed of diverse parts". I prefer it, not only because its meaning seems more spot-on but also because it seems it would be more likely to be known since (a) it also has two other definitions used in chemistry, (b) it is analogous to "homogenous" which the general reader would recognize from homogenized milk, and (c) it is a more common word being in the top 30% according to Merriam Webster. In general I think shorter is better, but I think these other points are weightier, so I stand by my preference. Your thoughts? YBG (talk) 17:16, 26 August 2017 (UTC)

I suspect I thought of heterogeneous a while ago but dismissed it for reasons I can no longer clearly recollect—possibly because I was looking for something shorter or less generic. [Post-coffee, it just occurred to me that I may also have dismissed "heterogeneous" as being too cumbersome(?) as a substitute for other.] Looking back at archive 28 I see that hetero nonmetals did in fact make it onto the crazy table. Most recently I thought of calling them heterogen nonmetals but I dismissed that as sounding too group-like e.g. chalcogen, halogen, and we're not talking about a group here.

Re the Wiktionary meaning of heterogenic, that is its meaning in one particular context e.g. "Pollen grains with such different alleles are described as heterogenic." (Lewis D 1947, "Competition and dominance of incompatibility alleles in diploid pollen", doi:10.1038/hdy.1947.5), rather than the actual meaning of the word. Merriam Webster is closer to the mark: "derived from or involving individuals of a different species", and in this case the nonmetals come from groups 1, 14, 15 and 16.

Looking at GB, and please excuse the heterogenic array that follows, I can see other examples of heterogenic e.g. 1. "Natural organic matter in both ground and surface water consists of a complex heterogenic mixture including humic or fulvic acids…"; 2. "In general, titanium and aluminium create a TiAl₃ compound for heterogenic nuclei points in aluminium liquid"; 3. "Microliquation is observed in all heterogenic alloys, for instance tin bronzes…"; 4. "Due to the wide spectrum of industrial chemicals which are synthesized at chemical production sites, the chemical composition of effluent samples is very heterogenic." 5. "More and more reaction conditions are being developed that enable the use of heterogenic catalysts and a range of solid supports are being explored."; 6. "The measuring length of the used gauges is 140 mm, which is suitable for measurements in heterogenic materials like concrete." 7. "The collection of glass pieces from the monastery's pharmacy is characterised by being numerous and heterogenic." 8. "However, the tourism industry is not one homogenous industry but rather a heterogenic collection of different industries and industry branches, from transport and hotels to experience-based services and cultural offerings…" 9. "Essential criteria for sustainable urban developments are heterogenic patterns of land use e. g. density, flexibility of urban spaces, and a mixture of features." 10. "The following specialized informative material includes heterogenic statics, the atomistic formation of metallic crystals, diffusion and the various physical properties of metals." 11. "Never before in history were there coalitions like the one of our enemies, composed of such heterogenic elements with completely contradictory goals."

I agree heterogeneous is a more common word, and I think "hetero-" or "heterogen-" [ad. Gr. ἑτερογενής of different kinds, f. ἑτερο- hetero- + γένος, γενε- kind] would be pretty well understood by most general readers.

If I look back on what I've written it seems to me that "heterogeneous" is something one would more appropriately use in a sentence like, "The museum contained a heterogeneous collection of curiosities", whereas "heterogenic" is something you would use to be a bit more specific, such as, "Carbon is an example of a heterogenic nonmetal" (i.e. it belongs to the family of heterogenic nonmetals), rather than "Carbon is an an example of a heterogeneous nonmetal", which doesn't really tell me anything; carbon is carbon, after all.

So, I think I still like heterogenic (simpler to spell, too). Sandbh (talk) 04:44, 27 August 2017 (UTC)

PS. I had a look in the Oxford English Dictionary for words ending in "-geneous", "-genous", and "-genic". There were 16 of the former, 181 of the middle, and 240 of the latter. Not much in it.

Pardon me for being blunt about it – I respect your work very much, as always – but this is one point that I think is crying out to be addressed. Is there anything "heterogenic" implies that isn't just "the word 'other' dressed up in Greek to look less silly as a category"? ^_^ Double sharp (talk) 05:13, 27 August 2017 (UTC)

Hi Double sharp! Nobody likes "other nonmetals" because it doesn't tell you anything. The same thing applied to "intermediate nonmetals" IIRC. "Heterogenic nonmetals" attempts to convey that the nonmetals involved are (1) different from the rest of the nonmetals; (2) different even among themselves compared to other kinds of nonmetals; yet (3) have enough similarities to support a descriptive-phrase category name that is more specific than the leftover "other nonmetals" and which thereby eliminates the notion that they are (collectively) forgotten nonmetals. While "other" can be read to mean "different from the rest," I don't think it conveys the same sense of internal difference that is conveyed by "heterogenic". And I think, as per my selection of GB extracts, that heterogenic has a pretty good record of being used to describe things having a differential nature (I tried substituting "other" for "heterogenic in those quotes, and it doesn't work very well at all). Sandbh (talk) 06:19, 27 August 2017 (UTC)

(edit conflict) @Double sharp: Thank you! Bluntness which AGF often helps force a clarification.

What is common in the 'hetero-' words is not at all like 'other'; that is to say, at its root, it isn't saying that a thing is different from the rest, but that a thing (or collection) is varigated within itself. Thus, what is being stated by these category names is NOT that H/C/N/P/S/Se are different from O/F/Cl/Br/I and He/Ne/Ar/Kr/Xe/Rn/Og, but rather that each of H/C/N/P/S/Se are quite distinct from each other, that is to say, that H/C/N/P/S/Se constitute a heterogeneous collection whereas O/F/Cl/Br/I and He/Ne/Ar/Kr/Xe/Rn/Og are fairly homogeneous collections.

But even if these terms were merely Greek window dressing for the term "other", that would be an improvement. One of the major difficulties with the term "other nonmetals" as a category name is that whenever the term appears in a sentence, the reader's first instinct is to parse it to mean "nonmetals other than the ones that have already been mentioned in this sentence or paragraph" instead of "a category of nonmetals called 'other nonmetals'". So even if the proposed term were just a foreign term meaning "other", it would be a great improvement.

YBG (talk) 06:31, 27 August 2017 (UTC)

I don't know if I like it but could we call them diverse nonmetals? Merriam-Webster defines diverse as "differing from one another; composed of distinct or unlike elements or qualities". Sandbh (talk) 13:12, 28 August 2017 (UTC)

That's essentially an English translation of heterogeneous, but is suffers from a similar problem: it doesn't sound like a category, just like an adjective. YBG (talk) 18:30, 28 August 2017 (UTC)

Do you have an example of the kind of thing you have in mind? Alkali as in alkali metal, and transition as in transition metals are adjectives. The only reason they sound like category names is that they are recognised category names. Incidentally, the first translation my Cassell's Latin-English dictionary gives for "other" is "as adj. = different, alius, diversus. Sandbh (talk) 23:05, 28 August 2017 (UTC)

But that is precisely the problem: unless they are well-established, adjectival names like this don't sound like category names. So the impression that naturally comes to mind with "diverse nonmetals" is simply the adjectival one. Here it is perhaps even worse, because "diverse" does not make much sense when it points to only one referrent: what to make of a sentence like "phosphorus is a diverse nonmetal"? Double sharp (talk) 23:36, 28 August 2017 (UTC)

(edit conflict) Would it help if I said I know it when I see it? No, I expect not. It seems that maybe what is key here is something like the goldilocks zone, not too common and not too uncommon. Too uncommon, and the reader won't get any information without clicking the link. Too common, and the reader might think that it is up to the reader to figure out what the author intends. What we want is an adjective "X" so that when the reader sees "X nonmetals", it is intuitively understood to mean "the collection of nonmetals known as X nonmetals". "Other nonmetals" sounds like the author is referring to the immediate context of the sentence or paragraph, as though the author meant "the nonmetals that haven't yet been mentioned in this list/sentence/paragraph" instead of meaning "the collection of nonmetals known as Other Nonmetals". "Diverse" makes it sound too general, like the author meant "an unspecified (possibly random) collection of nonmetals" instead of "the collection of nonmetals known as Diverse Nonmetals".

At a gut level, the difference seems to me to be intuitively similar to the difference between restrictive and nonrestrictive clauses in English grammar. Restrictive clauses generally begin with "that" as for example, "noble gasses that eight outer shell electron include Ne, Ar, Kr, Xe, Rn, and Og". Nonrestrictive clauses generally begin with "which", e.g., "noble gasses, which have filled outer electron shells, include He, Ne, Ar, Kr, Xe, Rn, and Og". The parallel with restrictive and nonrestrictive clauses is very tenuous, so if it doesn't make any sense to you, just ignore this paragraph.

YBG (talk) 00:11, 29 August 2017 (UTC)

@Double sharp: I agree in general with what you've said about adjectival names not sounding like category names. But it seems to me that this is less true with latin- or greek-sounding adjectives. We are conditioned to treat latin and greek derived adjectives or phrases as technical terms but we treat the good old anglo-saxon adjectives just as adjectives. YBG (talk) 00:17, 29 August 2017 (UTC)

The Latin- and Greek-derived adjectives do sound slightly more impressive, but I think that there is also a significant contribution from tradition. I mean, I fully agree that "transition metal" is problematic if transition means what it usually does, but by this point "transition" in chemistry has gained a special meaning of "d-orbitals in use", so it does not feel problematic unless you think too hard about it. Double sharp (talk) 02:01, 29 August 2017 (UTC)

By all means, let's avoid hard thinking - except to keep our reader from unproductive hard thinking. YBG (talk) 04:44, 29 August 2017 (UTC)

Stage	Category name	aka
Incipient	Metalloid	Junior nonmetals
Primitive	Foundation nonmetal Formative nonmetal	Rabble nonmetals
Mature	Corrosive nonmetal	Psycho nonmetals
Dormant	Noble gas	Cup-of-tea nonmetals
The progression in nonmetallic character, from left to right, would then look like this.

Could we call them foundation nonmetals since they represent the first or beginning set of "true" nonmetals after the metalloids? And they represent a kind of benchmark for comparing the metalloids and the corrosive nonmetals. I'm coincidentally reminded of Asmiov's Foundation series trilogy, Foundation, Foundation and Empire, and Second Foundation.

Another possibility is formative nonmetals. One meaning of this given by MW is "relating to, or characterized by formative effects or formation".

I think both adjectives convey notions of things that may be more half-baked than the more mature model seen in the corrosive nonmetals (hence greater variability in IE, EA, and EN).

Which reads better(?):

1. "Based on shared attributes, the nonmetals can be divided into the three categories of foundation nonmetal, corrosive nonmetal, and noble gas;"

or

2. "Based on shared attributes, the nonmetals can be divided into the three categories of formative nonmetal, corrosive nonmetal, and noble gas."

I find it hard to decide. We already have an "-ion" category in the transition metals, so that is good. OTOH, the simplicity of the two "-ives" is good, too. "Formative" is shorter by one letter. Formative sounds more friendly to me for reducing agents ("may I share my electrons with you?"); whereas corrosive is apt for the oxidising agents ("handover your electrons!"). I think I like "formative" for reasons I'm not fully conscious of.

Sandbh (talk) 05:39, 31 August 2017 (UTC)

The trouble I find with calling these the foundation nonmetals, is that not only does it seem fairly opaque without the benefit of "foundation" having received a distinct meaning in chemistry like "transition", but also the nonmetals which display the nonmetallic properties most clearly would rather be the corrosive nonmetals. And if we are talking aboutthe formation of nonmetallic character, surely this has already happened when we reach the metalloids, and you could make a case for it already at the post-transition metals (e.g. Sn, Bi, Po, except that I am not sure that the latter two should not be considered metalloids anyway).

I must confess that Parcly's idea of oxidative and reductive nonmetals is growing on me a great deal, for despite not being completely true (e.g. H), it is quite reasonable at the broad-brushstrokes level. (After all, most metal hydrides are not really ionic, and H is somewhat more akin to Li than F in its usual behaviour.) Double sharp (talk) 06:15, 31 August 2017 (UTC)

One principle I would want to suggest for consideration is that a category must be defined by the characteristics its members share, rather than those that they don't. Otherwise it seems to be no better than a leftovers bag. Double sharp (talk) 06:17, 31 August 2017 (UTC)

I tend to agree, especially at the broad-brushstrokes level. Subject to YBG's thoughts I believe we have enough content to write up something for our project talk page, yes? Sandbh (talk) 06:31, 31 August 2017 (UTC)

Progress, hopefully, at last! Sandbh (talk) 06:35, 31 August 2017 (UTC)

Sandbh's chart reminds me of the term "borderline nonmetals" which I coined for the metalloids back when we were thinking of merging them into the nonmetal supercategory. I still thke that term and think it would work very well for the metalloids, but that's hardly sufficient grounds to modify how many supercategories we have. On the other hand, maybe if we think of the metalloids as "borderline metals", we could use the term "borderline nonmetals" for the non-noble, non-corrosive nonmetals (NNNCNM).

This is in direct contrast to the terms "formative" and "foundational" which place the NNNCNM not on the fringes of the nonmetals, but rather at the very heart of them, that is to say, that all other NM build upon their basis. I'm not sure what I think of this idea. Are H/C/N/P/S/Se really the quintessential nonmetals? But if that is the idea that we're trying to get across, then maybe "basic nonmetals" would be preferable - except of course for the obvious fact that "basic" already has a well-defined chemistry-specific meaning. Sigh.

I agree 100% with Double sharp's assessment that categories should be named for what they are, not for what they aren't. Of course, the well-accepted term "nonmetal" clearly violates that rule! Nevertheless, sometimes a category name helps place the category in the broader context, for example, "transition metal".

<side discussion> When I first learned the term "transition metal" decades ago, it was in the context of seeing a large PT on the wall of my high school science classroom. The term made perfect sense, because these metals bridged the gap between what are today called groups 2 and 13. And seeing the then-much-fewer footnoted elements called "inner transition metals" immediately conjured up for me the image of something very like "our" 32-column PT. And of course, I also envisioned a 40 column PT with inner-inner TM, but then how could I possi8bly have known about the relativistic carnival sideshow that is now expected to follow Og? </side discussion>

What about looking for something that contrasts with "corrosive"? My electronic thesaurus from Redmond gives these antonyms: gentle, mild, calm, kind, tender, moderate, placid, temperate.

Some of these don't quite seem apropos, but I'll throw into the had those that seem at least sort-of-ok, along with the two mentioned earlier in this post:

• borderline nonmetals
• basic nonmetals
• mild nonmetals
• moderate nonmetals
• temporate nonmetals

Do any of these strike you fancy? YBG (talk) 22:19, 31 August 2017 (UTC)

No. I don't think the other nonmetals are borderline; they are distinctly nonmetallic albeit a motley crew. Basic is unsuitable for the reason you mentioned. Mild and temperate sound too much like "weak", which is what the metalloids are (weakly nonmetallic). Moderate tends to suffer from the same issue that "intermediate" does. We don't want something the contrasts with "corrosive". We want something that lies between "corrosive" and "passive"—except that I haven't been able to find such a word. And anything that would fit the bill would tend to suggest that there must be some kinds of nonmetals that are weaker, and there are, but we call these metalloids, and maintain them as a distinct super-category. It is the same problem that we get with the sequence metalloid, physiochemically moderate nonmetal and corrosive nonmetal. It looks like a lopsided taxonomy, because the brain expects to see a third category of weak nonmetal to complement the corrosive nonmetal category, and there isn't one. I'm quite happy with reductive nonmetal, and oxidative nonmetal as broad-brush descriptive category names, consistent with how these nonmetals are conceived of in the literature. For example

"It should be noted that border-line cases exist where an element shows both metallic and non-metallic properties. (a) Metals act only as reducing agents, whereas nonmetals may act either as reducing agents or as oxidizing agents." (College Chemistry 1956, p. 342).

The "-ive" suffix means, "that performs or tends toward or serves to accomplish an indicated action esp. regularly or lastingly" or "having a tendency to, having the nature, character, or quality of, given to (some action)". The tendency flavour of the "-ive" suffix nicely accommodates the fact that e.g. H and S are sometime capable of acting as oxidants, but their overall tendency is to act as reducing agents. What do you think of reductive and oxidative? Sandbh (talk) 00:00, 1 September 2017 (UTC)

I was quite happy with oxidative and reductive, but I gave up on it when the exceptions of H and S were mentioned. But if as you say the oxidative nature of H and S are actually exceptions to their normal reductive nature, I think that pretty much reduces the objection - shall we say it oxidizes it to ashes? Much better overall than "formative" or "foundational" YBG (talk) 04:14, 1 September 2017 (UTC)

Good, will go ahead and post something to our talk page. Sandbh (talk) 05:27, 1 September 2017 (UTC)

New NNNM colors

As I wrote in WT:ELEMENTS#New_colors, new categories should have new colors. If you can agree in general (better talked at WT:ELEM not here), could you allow me to edit all current live proposals/sandboxes (like this) into these new numbers/colors? Ask me if things are unclear or confusing. -DePiep (talk) 21:15, 25 August 2017 (UTC)

Homogeneity of metallicity categories

Just for laughs, how would you order the metallicity categories in order from most to least homogenous? My first guess would be that the noble gasses, lanthanides and actinides would be the most homogenous and metalloids and "other" nonmetals are the least homogenous. What say you? YBG (talk) 22:00, 28 August 2017 (UTC)

The alkali metals surely belong together with the most homogenous categories, and because the useful halogens are included with them it seems clear that the corrosive nonmetals are up there too. The alkaline earth metals are, I guess, one little step down from this, with the transition metals somewhere in the middle. Double sharp (talk) 23:38, 28 August 2017 (UTC)

Maybe something like this: Noble gases; Alkali metals; Lanthanides; Alkaline earth metals; Corrosive nonmetals; Actinides; Transition metals; Post-transition metals; Metalloids; Other nonmetals. Sandbh (talk) 01:40, 29 August 2017 (UTC)

I agree, except that I'd swap the lanthanides and the alkali metals. Double sharp (talk) 01:57, 29 August 2017 (UTC)

Your proposal

Hi, even though I oppose your proposal I would like thank you for your time and effort you put into creating it. I would also like to advise you not to get disheartened if it does get rejected. Your enthusiasm and effort are exactly what this encyclopedia needs. Kind regards EvilxFish (talk) 11:07, 12 September 2017 (UTC)

@EvilxFish:Thank you. In the event that there were just two nonmetal categories these would be "chemically active nonmetal" and "noble gas". How does the former name strike you? Sandbh (talk) 00:25, 13 September 2017 (UTC)

Personally I am fine with categorizing noble gases as such as this is widely accepted in chemical communication and is a clear hard category. As for the former whereas it is true some non-metals are called "more" or "less" chemically active I don't believe it is a common thing to split the periodic table into such categories (nor would it be easy as its a continuum not a hard break), just in the same way one wouldn't split it into electronegative and electropositive categories. My main objection though is the fact the proposed hard categorization isn't used in chemical communication. EvilxFish (talk) 18:00, 13 September 2017 (UTC)

@EvilxFish: What category name do we then give to the non noble gas nonmetals i.e., H, C, N, O, F, P, S, Cl, Se, Br, I, bearing mind that we categorise At as a metalloid.

@Sandbh: F, Cl, Br, I are halogens (I believe At is both a metalloid and a halogen but need to check that). O,S,Se are known as the Chalcogens. N,P are Pnictogens. As for H it's not easy to decide where it lives. As for C I am not sure what category on the periodic table it would fall into but I could look it up if you want to know. Regards EvilxFish (talk) 12:22, 14 September 2017 (UTC)

May I also recommend having a play with this EvilxFish (talk) 12:26, 14 September 2017 (UTC)

Electron affinity (data page)

Hello. Now that you have carefully destroyed all the information that the page contained on i) the uncertainties attached to the experimental electron affinities ii) the isotope shift of those electron affinities for which it was so painfully determined, what shall we do? What does "temporary removal is isotopic forms of H, C, O and S" (one of your comments on August 12) mean, when I see that today the information has not yet come back? Do you realize the quantity of potentially useful data you erased in a single day from this page, which had taken years to put together? Chrisanion (talk) 14:25, 26 September 2017 (UTC)

Good morning. I apologise for (a) neglecting to restore the isotopic shift values; and (b) if my edits appeared to you as if I was destroying the page by leaving out the uncertainty values. My good faith intent was to create a sortable table by element, and the uncertainty values did not allow me to do this. Could we restore the original top table with the newly determined value for Zr?

Sandbh (talk) 02:05, 27 September 2017 (UTC)

You can use sortkeys via {{sort}} to sort values with uncertainties. Double sharp (talk) 04:03, 27 September 2017 (UTC)

Tx. It should be possible to use the current table and reinstate the uncertainty values, and isotopic shift figures. Sandbh (talk) 07:06, 27 September 2017 (UTC)

Please! You would be welcome to do that. Chrisanion (talk) 12:34, 27 September 2017 (UTC)

Well, I believe all of the missing data has been restored. @Double sharp: Is there a way of getting the negative values sorting properly? Sandbh (talk) 04:37, 28 September 2017 (UTC)

IIRC using a hyphen as the minus sign (e.g. -5) in the sortkey should work. Double sharp (talk) 06:01, 28 September 2017 (UTC)

It only half works in that it sorts the negative numbers in reverse order rather than in ascending order. So you end up with -0.5, -0.7, -1.0, -1.5, -2.5, 0.017, 0.025, 0.05, 0.1 rather than -2.5, -1.5, -1.0, -0.7, -0.5, 0.017, 0.025, 0.05, 0.1 -- Sandbh (talk) 03:27, 29 September 2017 (UTC)

My memory may have failed me on this, because I don't think I've needed to sort negatives very often. Maybe it works with plus signs on the positives, though I haven't tried it either. Mind you, I doubt there are that many negatives, so using fake sortkeys as a workaround should solve the problem. Double sharp (talk) 06:49, 29 September 2017 (UTC)

It should all be working now after an unbelievable amount of messing around with code. Give me the simple functionality of an excel spreadsheet any day. Sandbh (talk) 06:29, 30 September 2017 (UTC)

Precious anniversary

Precious

Three years!

--Gerda Arendt (talk) 07:02, 4 October 2017 (UTC)

Four years now! --Gerda Arendt (talk) 06:17, 4 October 2018 (UTC)

... and five --Gerda Arendt (talk) 10:22, 4 October 2019 (UTC)

Vitriole

Hi Sandbh! Would you help me figure how to reference this concept of vitriole? It exists in French, German, and Russian, to say the least. (For example, see de:Vitriole.) It is basically a hydrated sulfate salt. How do I refer to that in English in a historical context?--R8R (talk) 15:49, 22 October 2017 (UTC)

This is going to be very confusing, since in English vitriol is an archaic word for sulfuric acid... Double sharp (talk) 16:01, 22 October 2017 (UTC)

I just checked, "vitriole" is plural from "vitriol," which, to be precise, is a hydrated sulfate of some divalent metals. Anyway, is there such a concept in English and how can we reference it other than just using this foreign word in italics?--R8R (talk) 20:04, 22 October 2017 (UTC)

I just checked Wiktionary and Merriam-Webster. It does appear that this concept exists in English language, or at least it certainly did some time ago. Authors from the 19th century openly use the word vitriol in historical contexts, so we must be good to use it. Problem solved, thank you.--R8R (talk) 22:10, 22 October 2017 (UTC)

Here's part of the entry from the Oxford English Dictionary. First quote is from 1386:

vitriol, n.

(ˈvɪtrɪəl)

Forms: 4–5 vitriole, 5 vit-, vytreole, 5–6 vytryol(e, 6–7 vitrioll (6 -olle), 5– vitriol; 5–7 vitriall, 6–7 vitrial, 7 vitraell.

[a. OF. (also F.) vitriol 13th c.; = Sp. and Pg. vitriolo, It. vetriolo, -iuolo, vitriolo, -iuolo, -ivuolo) or directly ad. med.L. vitriolum (Albertus Magnus) f. vitrum glass.]

1.1 One or other of various native or artificial sulphates of metals (see 2 and 3) used in the arts or medicinally, esp. sulphate of iron: a.1.a Used in sing. without article.

   c 1386 Chaucer Can. Yeom. Prol. & T. 255 Vnslekked lym, chalk,‥Poudres diuerse, asshes,‥Cered pottes, sal peter, vitriole.    14‥ Voc. in Wr.-Wülcker 579 Draganti, vytryole, or coporose.    a 1425 tr. Arderne's Treat. Fistula, etc. 40 Puluerez of alume, zucarine brent, of attrament, and of vitriol.    1471 Ripley Comp. Alch. Adm. iv. in Ashm. (1652) 190 Also I wrought in Sulphur and in Vitriall, Whych folys doe call the Grene Lyon.    1527 Andrew Brunswyke's Distyll. Waters F j b, Halfe an ounce of vytryol wherof the ynke is made.    1599 A. M. tr. Gabelhouer's Bk. Physicke 317/1 Bloodstenchinge. Take of the best Vitriolle, beate it smalle, and boulte it through a fine cloth.    1612 Woodall Surg. Mate Wks. (1653) 210 Copperas or Vitriol‥is a mineral salt which‥doth farre excel many other kinds of salts.    1681 tr. Belon's Myst. Physick Introd. 38 Those Acides, and acrimonious Particles of the Salt and Vitriol which had caused its Sublimation.    1718 Quincy Compl. Disp. 8 The last is what is forced from Vinegar, Vitriol, and such like acid Substances.    1728 Chambers Cycl. s.v., The Antients give the Name Chalcitis, or Chalcite, to native Vitriol;‥which is a kind of mineral Stone, of a reddish Colour.    1756–7 tr. Keysler's Trav. (1760) III. 124 Besides sulphur, vitriol is also made here, of a sapphire colour.    1854 Ronalds & Richardson Chem. Technol. (ed. 2) I. 359 The chloride of calcium melting easily in the still, enables the whole of the acetic acid to be evolved at a lower temperature than when vitriol is employed.    1879 McCarthy Own Times xviii. II. 26 The use of vitriol was recommended among other destructive agencies.

Or see here: https://en.oxforddictionaries.com/definition/vitriol Sandbh (talk) 00:54, 23 October 2017 (UTC)

I think we'll be fine, provided we mark the term as being rather archaic in this sense. Today vitriol is more often used in the figurative sense, which comes more likely from the "sulfuric acid" sense, so it might confuse at first without parenthetical explanation (it certainly perplexed me). Double sharp (talk) 08:00, 24 October 2017 (UTC)

Sandbh, thank you. This confirms my thinking. Double sharp, when introducing the term to the article, I added the first mention as "vitriole (sulfates)" and changed that to "vitriole (hydrated sulfates)" later. Have you just overlooked that or is this not too clear somehow?--R8R (talk) 11:50, 24 October 2017 (UTC)

I think I just saw the second line about blue and green vitriole, to be honest. I don't know how on earth I missed that, so I don't have any further problems with it. Double sharp (talk) 13:48, 24 October 2017 (UTC)

Very well then.--R8R (talk) 14:33, 24 October 2017 (UTC)

OED also says, "2.2 With distinguishing epithets: a.2.a With adjs. of colour. blue vitriol, green vitriol, red vitriol, white vitriol, sulphate of copper, iron, cobalt, and zinc respectively." Oh, and which article are we talking about? I see the vitriol article lists more types. Sandbh (talk) 22:09, 24 October 2017 (UTC)

It's on the aluminium article, which it seems R8R has started his FA campaign for. ^_^ Double sharp (talk) 02:09, 25 October 2017 (UTC)

Thanks for the hint. Somehow, it didn't occur to me I could check if there is such an article in en.wiki, I only judged from how de.wiki, ru.wiki and fr.wiki had no interwikis to en.wiki.--R8R (talk) 07:51, 25 October 2017 (UTC)

This opens up a whole new are of thought for me. We often mention the etymology of element names, that is to say, how non-Chemical words came to be borrowed, combined, or modified to have chemical meanings. Vitriol is an example of the reverse direction, how a word with a chemical meaning came to be used with a non-chemical meaning. Could be turned into a fascinating list! YBG (talk) 16:15, 24 October 2017 (UTC)

With regards to the categorisation proposal

Hi, I wasn't sure where to add to the discussion, unfortunately I have been unavailable for the past few weeks and as you are the proposer for the amendment may I simply express my opinion here?

• -> Group 14 Carbon family
• -> Group 15 (nitrogen and down) Pnictcogens
• -> Group 16 Chalcogens
• -> Group 17 Halogens.

Based on the RSC periodic table. That way you don't have the problem of the "other non-metals" category but also you have something that is accepted by the chemical community and does not have ambiguity/subjectivity. Sorry for missing the discussion and also for not really knowing where to put this in it. Much love EvilxFish (talk) 15:52, 11 November 2017 (UTC)

This is a good question and I'm still thinking about it. In the meantime, what will happen to group 13, and to hydrogen? Presumably H would have its own colour, as would group 13? Sandbh (talk) 11:56, 12 November 2017 (UTC)

I think I've also previously suggested the idea of colouring by groups only, half tongue-in-cheek and half to seriously get us thinking about why it seems to sometimes result in a knee-jerk rejection: see Wikipedia_talk:WikiProject_Elements/Archive_27#Groups_trumping_metallicity for example. It is true that all serious general chemistry books organise elements by groups; it is likewise true that they have to break up the groups into several chapters (or quasi-chapters) because N is too different from P to be lumped in with the same bunch. (P is easier to lump in with As, Sb, and Bi.) If I say any more right now it'll turn into a rehash of the old discussions on WT:ELEM and your talk page, so I'll stop here. Double sharp (talk) 15:00, 12 November 2017 (UTC)

It is certainly true that the elements can be categorised by group, but there is a lot more going on than just that.

You are correct but as you pointed out group categorisation is used quite widely by chemists. That being said you are right the chemistry of fluorine is rather different to chlorine for example. But then no matter how you try and classify elements you will always run into that issue. EvilxFish (talk) 15:18, 16 November 2017 (UTC)

As a whole I would say that the literature distinguishes between metals and nonmetals, together with a fair amount of coverage as to what happens when the metals and nonmetals meet; and that it also covers at least the main group elements on a group by group basis.

So, I could support colouring by groups provided our table had five borderlines: one borderline line between groups 2 and 3 to separate the pre-transition metals from the transition metals; one between groups 11 and 12 to separate the transition metals from the post-transition metals; one on either side of the metalloids; and one between groups 17 and 18 to separate the reactive nonmetals from the noble gas nonmetals. Of course there would also be the challenge of establishing an aesthetically pleasing 13-colour colour scheme; the colour scheme used by the RSC table is woeful.

Here are examples of periodic tables with two borderlines, and four borderlines. That is the kind of thing I have in mind. See also p. 233 of the latter reference for Klemm's division of the elements into true metals, meta-metals, semi-metals, and non-metals, again using (just three) borderlines. Sandbh (talk) 03:16, 13 November 2017 (UTC)

I wasn't suggesting a simple categorisation by group but rather as a way to eliminate the "other non-metals" category that seems to upset some people. Naturally one would also include divisions for metals (and into that transition metals etc) and semi-metals. Hydrogen is a difficult one, it doesn't really belong anywhere. My main objection (as it has always has been) is that whatever system is implemented it should be one which is used by chemists and not a new one that has been created by wikipedia. I will leave finding an aesthetically pleasing colour scheme in your capable hands, I have the artistic talent of a Sciaridae. Kind regards EvilxFish (talk) 15:18, 16 November 2017 (UTC)

@EvilxFish: I don't understand what you're suggesting. If you eliminate the "other non-metals" category what do you propose to call the nonmetals in question, namely H, C, N, O, F, P, S, Cl, Se, Br, I? Going by the RSC table it would be something like this:

Hydrogen

Carbon

Pnictogen nonmetal

Chalcogen nonmetal

Halogen nonmetal

OTOH you say you are not suggesting a simple categorisation by group. Maybe it is this(?):

Hydrogen

Group 14–16 nonmetal

Halogen nonmetal

Or:

Hydrogen

Group 14–17 nonmetal

If none of the above, could you please list what nonmetal categories you have in mind? I'm confused. Sandbh (talk) 09:50, 17 November 2017 (UTC)

Hi sorry for not being clear, I was proposing the first option though I would suggest "Pnictogen" not "Pnictogen non-metal" for example. As well as this one can have "transition metals" "Alkaline earth metals", "metalloids " (which aren't just down the group like Pnictogen), etc. That way you reduce the need for anything that doesn't fall within a category (which will make some people happy) and they are widely accepted categories (which makes people like me happy). EvilxFish (talk) 10:22, 27 November 2017 (UTC)

@EvilxFish: Thank you. I understand now. We would need to retain "Pnictogen non-metal". That is because, as well as N and P being pnictogens, As and Sb (metalloids), and Bi (a post transition metal) are also pnictogens. If we just called N and P pnictogens, that would not be right since As, Sb, and Bi are also pnictogens. So you can hopefully see what the problem is. Sandbh (talk) 10:31, 27 November 2017 (UTC)

Ah, very true. So either one of two solutions may be applied, the first is as you suggested. The second is to have certain elements belong to two categories simultaneously. I'm not bothered either way. Kind regards EvilxFish (talk) 10:33, 27 November 2017 (UTC)

What's this?

Please take a look at this picture. You can see aluminum in the foreground; but what are these shapes/objects made of it? I could only come up with "cylinders." Perhaps you can do better? I used "rolls" in the article. Not sure if that's correct at all, though.--R8R (talk) 15:41, 12 November 2017 (UTC)

They might be extrusion billets (i.e. cylindrical ingots) rather than the sheet ingots (rectangular) in the background. I can find references to extrusion billets of up to 870 mm in diameter and these look about that size. See here for pics. Sandbh (talk) 03:39, 13 November 2017 (UTC)

Makes perfect sense. I'll use this, thank you very much. (I'm a little sad I didn't come up with this myself but even this concept of extrusion somehow didn't pop up in my mind. Bravo to you then!)--R8R (talk) 13:14, 13 November 2017 (UTC)

History of aluminium

Hi! If you've got a spare minute, could you please look at the article? It was promoted to the GA status just recently but as of now, it looks very FA-y to me. Maybe because I have never yet written an article dedicated to one aspect of a story of an element. I really could use a fresh couple of eyes to confirm or refute my dreamful thinking. If you could find the minute, that would be appreciated.--R8R (talk) 11:27, 28 November 2017 (UTC)

Will do. Sandbh (talk) 11:13, 30 November 2017 (UTC)

Thank you very much.--R8R (talk) 11:21, 30 November 2017 (UTC)

Will look later today. Sandbh (talk) 05:38, 7 December 2017 (UTC)

I'm worried about this edit. Triggered by YBG's comment, I sought to look deeper into the problem and I found out alum is not necessarily potassium alum. In fact, there were times when ammonium alum was the primary kind of alum. (You can find a citation supporting this statement on my talk page.) I seem not to imply in the article that we are talking about one particular kind of alum; once, at contrary, I say that alum may be synthesized with various monovalent cations: "He realized that adding soda, potash, or alkali to a solution of the new earth in sulfuric acid yielded alum,[15]" which contradicts the idea that alum is potassium alum only. The question of there being multiple alums is very minor and not particularly relevant to this discussion as both ammonium and potassium alums have the aluminum we and the scientists were looking for.--R8R (talk) 09:21, 21 December 2017 (UTC)

In fact, I'm worried about all added formulas. In general, I didn't add the formulas myself because I think it's interesting to watch the story unfold as the scientists saw it: they did not know the formulas. We later tell the formula of alumina at the moment it is discovered, so this important formula is not left out. The formula for alunite is not really needed rather distracts us from the point: we only need to tell the reader that this is an aluminum mineral and details are leading us astray from the point. And as for alum (the formula for which, come to think of it, we don't really need either), it is complicated as there are multiple important alums.--R8R (talk) 09:29, 21 December 2017 (UTC)

FA-y check up

@R8R: I've compiled the topic sentences from the main body of the article. I should be able to read these and see the flow and logic of the article, without going into the detail of the rest of each paragraph. Each topic sentence should focus on one theme which is explored in the rest of the paragraph. I have added comments where I have some concerns about your topic sentences.

When you have considered and responded to my comments let me know and I'll do the next steps of drilling down into each paragraph. Sandbh (talk) 07:05, 19 December 2017 (UTC)

Early history
The history of aluminium has been shaped by usage of alum.

After the Crusades, alum was a subject of international commerce;[6] it was indispensable in European fabric industry.

Establishing nature of alum
The nature of alum remained unknown. 

In 1754, German chemist Andreas Sigismund Marggraf synthesized the earth of alum by boiling clay in sulfuric acid and subsequently adding potash.

In 1782, French chemist Antoine Lavoisier wrote that he considered highly probable that alumina was an oxide of a metal which had an affinity for oxygen so strong that no known reducing agents could overcome it.

Comment: “alumina” should be mentioned in the first topic sentence; otherwise the relationship between “earth of alum” and “alumina” (in the second topic sentence) is not clear.

I don't quite understand; I've added a note so it is clear these two terms have the same meaning?--R8R (talk) 10:41, 19 December 2017 (UTC)

Synthesis of metal
In 1760, French chemist Theodor Baron de Henouville declared he believed alumina was a metallic earth and first attempted to reduce it to its metal, at which he was unsuccessful. 

Comment: the mention of 1760 makes this topic sentence chronologically out of place, compared to the previous 1765 and 1782 lead sentences.

It seems to make more sense to slightly disabide the chronology to establish clearer logical associations, so rather main topics go consecutively and events within those topics go chronologically. That makes the text easier to read: by tge end, you have a clear understanding of each major aspect in aluminum history.--R8R (talk) 10:41, 19 December 2017 (UTC)

In 1790, Austrian chemists Anton Leopold Ruprecht and Matteo Tondi repeated Baron's experiments, significantly increasing the temperatures; they found small metallic particles, which they believed to be the sought-after metal, but later experiments by other chemists showed these were only iron phosphide from impurities in charcoal and bone ash.

In 1807, British chemist Humphry Davy attempted to electrolyze alumina with alkaline batteries; in fact, he did electrolyze it, but the metal formed contained alkali metals potassium and sodium and Davy had no means to separate the desired metal from these two

In 1813, American chemist Benjamin Silliman repeated Hare's experiment and while he did at one moment obtain small granules of the sought-after metal, it almost immediately burned

Comment: who is Hare?

An American chemist. He had been mentioned by this point?--R8R (talk) 10:41, 19 December 2017 (UTC)

In general, I understand the notion behind the comment but I don't see a way to address it without breaking the text flow.--R8R (talk) 10:12, 20 December 2017 (UTC)

Production of the metal was first claimed in 1824 by Danish physicist and chemist Hans Christian Ørsted. 

Berzelius attempted to isolate the metal in 1825; he carefully washed the potassium ana--R8R (talk) 10:41, 19 December 2017 (UTC)log of the base salt in cryolite in a crucible.

Comment: need to make clearer that Orsetd was unsuccessful despite his claim, thus the relevance of talking about Berzelius.

Oersted was not unsuccessful and he may be indeed considered the disoverer. He only lacked recognition as such and apparently didn't seek to get it.--R8R (talk) 10:41, 19 December 2017 (UTC)

German chemist Friedrich Wöhler visited Ørsted in 1827.

Comment: This topic sentence does not flow very well after the previous two topic sentences. It doesn’t give enough context.

Removed the story on the trip to Denmark.--R8R (talk) 10:12, 20 December 2017 (UTC)

Rare metal
As Wöhler's method could not yield large amounts of aluminium, the metal remained rare; its cost exceeded that of gold

Deville's smelter moved in 1856–1857 to La Glacière, Nanterre, and finally to Salindres. 

Comment: what is the relevance of Deville's smelter; there is no flow from the previous topic sentence.

Deville's smelter was the most significant aluminium source of all at the time, and at many times the only source I was able to identify. I haven't seen anyone directly say this was the case but it appears to be the case and I give more information on other production sites later.

Moved a sentence from the previous para to the beginning of this one. Does it look better now?--R8R (talk) 10:12, 20 December 2017 (UTC)

Other chemists also sought to industrialize the production of aluminium.

At the next fair in Paris in 1867, the visitors were presented with aluminium wire and foil; by the time of the next fair in 1878, aluminium had become a symbol of the future

Electrolytic production
Aluminium was first synthesized electrolytically in 1854 independently by Deville and German chemist Robert Wilhelm Bunsen.

Comment: chronological flow does not work very well, compared to previous topic sentence.

same as above--R8R (talk) 10:41, 19 December 2017 (UTC)

The first industrial large-scale production method was independently developed by French engineer Paul Héroult and American engineer Charles Martin Hall; it is now known as the Hall–Héroult process.

At the same time, Hall invented the same process and successfully tested it; he then sought to employ it for a large-scale production; for that, however, the existing smelter would have to radically change their production methods, which they were not willing to do in part because a mass production aluminium would then immediately drop the price of the metal.

Comment: seems to duplicate part of the previous topic sentence.

I don't see how? The previous topic sentence was about Heroult, and this one is about Hall.--R8R (talk) 10:12, 20 December 2017 (UTC)

The Hall–Héroult process converts alumina into the metal; Austrian chemist Carl Joseph Bayer discovered a way of purifying bauxite to yield alumina in 1889, now known as the Bayer process.

Comment: the two parts of this topic sentence would flow better if they were the other way around.

Could be the case if we were talking about the production process, but we're talking about history and this is when the "chronology within each topic rules" rule applies.--R8R (talk) 10:41, 19 December 2017 (UTC)

Mass usage
Prices of aluminium dropped, and aluminium had become widely used in jewelry, many everyday items, eyeglass frames, and optical instruments by the early 1890s. 

During the first half of the 20th century, the real price for aluminium continuously fell from $14,000 in 1900 to $2,340 in 1948 (in 1998 United States dollars), with some exceptions such as the sharp price rise during the World War I

Exchange commodity
In the beginning of the second half of the 20th century, the space race began. 

Comment: hard to see how this pertains to aluminium becoming an exchange commodity.

True. I don't particularly like that header but haven't been able to invent a better one.--R8R (talk) 10:41, 19 December 2017 (UTC)

By 1955, the market had been mostly divided by the Six Majors: Alcoa (successor of Hall's Pittsburgh Reduction Company), Alcan (originated as a part of that company), Reynolds, Kaiser, Pechiney (successor of Pechiney and the Compagnie d’Alais et de la Camargue that bought Deville's smelter), and Alusuisse (successor of Héroult's Aluminium Industrie Aktien Gesellschaft), with their combined share of the market equaling 86%. 

In the 1970s, the increased demand for aluminium made it an exchange commodity; it entered the London Metal Exchange, world's oldest industrial metal exchange, in 1978.

This along with the change of tariffs and taxes started redistribution of the shares of world producers: while the United States, the Soviet Union, and Japan accounted for nearly 60% of the world's primary production in 1972 (and their combined share of consumption of primary aluminium was also close to 60%),[73] their combined share only slightly exceeded 10% in 2012.

Comment The last "combined share" seems ambiguous - is it production or consumption? I can think of several ways to rearrange this, depending on whether the 1972 consumption was >60% or <60% and depending on what the 2012 combined share of consumption was. YBG (talk) 12:17, 20 December 2017 (UTC)

The world output continued to grow: in 2013, the annual production of aluminium exceeded 50,000,000 metric tons. In 2015, it was record 57,500,000 tons.

--- Sandbh (talk) 07:05, 19 December 2017 (UTC)

Thorium FAC

It's on right now! I assume you don't have too much spare time on your hands, so I didn't ask earlier, but could you please review the article now?

If you cannot help me with both history of aluminium and this at the same time, then please do this.--R8R (talk) 13:00, 20 December 2017 (UTC)

Wow, That's a biggie. I can find time to do it. Sandbh (talk) 09:00, 21 December 2017 (UTC)

Done! A formidable piece of work. Sandbh (talk) 02:14, 31 December 2017 (UTC)

Your nth-order course corrections to the main-group/transition division

I hope you don't mind me fiddling a little with your sandbox, though my second edit was meant for discussion: do you think Be needs to be called out as an exceptional sd-bloc element (well, okay, much less exceptional than H)? I think you could make a very good case for it. Also, do you think Al is that different from Ga that it deserves to be considered an exception? Of course Al has similarities to both the Sc group and the Ga group, but we do put it over Ga and not Sc in the table after all. (I'm not opposed to broadening this to talking about the bifurcations in groups II and III if you want to bring them in as comparisons to clarify things. ^_^) Double sharp (talk) 16:40, 13 January 2018 (UTC)

Feel free to hop around in my sandbox!

I'm happy to call out Be given it lies on the sp frontier---it's almost a poor metal.

On Al I had in mind that it's sometimes counted as a pre-transition metal, given its electropositivity. There's also the mini-bio for Al in the metalloid article:

Stott[435] labels aluminium as a weak metal. It has the physical properties of a metal but some of the chemical properties of a nonmetal. Steele[436] notes the paradoxical chemical behaviour of aluminium: "It resembles a weak metal in its amphoteric oxide and in the covalent character of many of its compounds ... Yet it is a highly electropositive metal ... [with] a high negative electrode potential". Moody[437] says that, "aluminium is on the 'diagonal borderland' between metals and non-metals in the chemical sense."

Ga is plainly a post-transition ps metal. No special mention required.

On merged blocks I see Eric once supported a table with an sp block on the left and the rest of the p block on the right; see http://philsci-archive.pitt.edu/3095/1/Scerri%27s_New_PT_%2B_pics.doc -- Sandbh (talk) 01:59, 14 January 2018 (UTC)

I think I wasn't very clear about Al, but what I'm trying to get at is that if you focus on the properties of Al where it looks like a weak metal, it does not look very different from Ga. Ga also has an amphoteric oxide and its compounds usually have much covalent character. There are gallates like there are aluminates, and the directional bonding that is incipient in the structure of Al is fully present in Ga. (Even the In crystal structure is distorted.) So if you look at these properties Al looks like a reasonable little sister of Ga and In (Tl is a bit different but is clearly in the same group). I agree that Al is a strongly electropositive metal, but it's also commonly amphoteric like Ga; you don't see this kind of behaviour from Sc. So I would think this might be more of a 4th-order (sorry) course correction; Al can sometimes act like a pre-transition metal ought to because removing its three outer electrons gets you straight to a noble-gas core, but its chemistry seems to be largely that of a bona fide post-transition metal as I see it. I reckon you could probably say the same of Be, which has an even more strongly negative electrode potential, except that it's already in the sd-block and so what looks anomalous are the times the diagonal relationship asserts itself and it acts like Al instead of Mg. (You could take Stott, Steele, and Moody's comments about Al and apply them to Be, and I don't think anyone would notice that anything was changed. ^_^)

Well, this has become a bit of a ramble, and I still have to take a closer look at Be and B to see where I think they fit in all of this. I guess we could say that the reason we put Be with Mg and not with Zn is because we expect Be to be a feeble base because basicity increases as the group is descended (hence that argument for La). I think we'd have to make the argument that Al is closer to Ga than it is to Sc, because otherwise the question becomes why you can't ignore the s²p configuration for better chemistry. (I mean, we already do it for Lr, albeit this is a gas-phase configuration.) Incidentally I still think that there should be some mention on how group 13 differs from the groups beyond it in the p-block (because they are hypoelectronic instead of hyperelectronic); I guess metallic bonding is sort of a generalised version of what B is up to (metallic bonding is many-electron many-centre bonding ^_^). But going more into this is going to result in another ramble. Double sharp (talk) 07:34, 14 January 2018 (UTC)

Regarding the split-block table Scerri once supported: I still don't find the atomic-number triads argument all that convincing. Actually, I do not find it convincing at all. Among all the atomic number triads we are already only considering the ones that make chemical sense: hence 32 as the average of 14 and 50 is acceptable, but 32 as the average of 7 and 57 isn't. So this essentially is already subordinated to chemical facts. It therefore seems to me that moving elements on the basis of creating more triads (e.g. H above F) is rather putting the cart before the horse again. I'm happy to point to it as an exhibit showing that there is no need to keep the blocks pristine like the Janet table would have us do, while distancing myself from the justifications given for it. ^_^ Double sharp (talk) 07:39, 14 January 2018 (UTC)

P.S. Doesn't the first-row anomaly need to be somewhere in there too? Or more generally primogenic repulsion? Based on that I'd be more willing to say that the behaviour of the An is the norm and that of the Ln ought to be the exception. Then again that would probably have us say the same about 3d vs. 4d and 5d, which I'm sure is backwards from the normal perspective. ^_^ Double sharp (talk) 07:47, 14 January 2018 (UTC)

Silicon GAN

Incidentally, would you mind saving the Si GAN from development hell by taking it on? ^_^ That might give me some motivation to go for Sn and finish the group. I'm sure you'll be able to write a more coherent account by this fascinating if very complicated element than I can (we mostly handwaved about it in high school IIRC). Double sharp (talk) 07:52, 14 January 2018 (UTC)

Post-transition metal: blurbs for copernicium through tennessine

I've wanted to write the blurbs for Cn through Ts to complete this, but since so little is known, we're certainly going to get into predictions. Now the trouble with the predictions is that they are not the sort that can write us a blurb about these properties: for instance, looking at flerovium and comparing what is predicted there with your blurb on Pb, the only thing that I can take over and say is that flerovites should be a thing just like plumbites (which implies that FlO is an amphoteric oxide, should it actually exist, since FlO₂ is predicted to be grossly unstable). This is about the same reason why I keep wanting to reserve judgement for Po and At, even though I would not be very surprised anymore if Po turned out significantly more metallic than Bi (which seems likely; it's not that different from As being more metallic than Ge). Yet the article feels incomplete if it mentions Cn through Ts and then proceeds not to talk about them at all. I'm considering covering all six of these transactinides in a single blurb, something like:

"Copernicium, nihonium, flerovium, moscovium, livermorium, and tennessine are little-studied: only the first three have had their chemistry preliminarily investigated. In most cases, they are expected to behave similarly to their lighter congeners, although relativistic effects would tend to favour lower oxidation states and increase the chemical nobility of copernicium through livermorium. Due to their closed-shell configurations, copernicium and flerovium may in some ways behave like the noble gases: both are expected to go further than mercury in weak metallic bonding and be gaseous at room temperature."

Well, that was just written off the top of my head. What do you think? Is there something else we might add? Double sharp (talk) 14:13, 15 January 2018 (UTC)

This is rather good! Please proceed. Sandbh (talk) 08:51, 22 January 2018 (UTC)

Word smithing, accept or reject as you see fit:

"Copernicium, nihonium, flerovium, moscovium, livermorium, and tennessine are little-studied: only the first three have had a preliminary investication of their chemistry. Most are expected to behave similarly to their lighter congeners, although relativistic effects would tend to favour lower oxidation states and increase their chemical nobility ^{OR: increase the chemical nobility of all but tennessine}. Due to their closed-shell configurations, copernicium and flerovium may in some ways behave like the noble gases: both are expected to go further than mercury in weak metallic bonding and be gaseous at room temperature."

YBG (talk) 10:15, 22 January 2018 (UTC)

I like this one which is somewhat of a combination of the two:

"Copernicium, nihonium, flerovium, moscovium, livermorium, and tennessine are little-studied: only the first three have had a preliminary investigation of their chemistry. Most are expected to behave similarly to their lighter congeners, although relativistic effects would tend to favour lower oxidation states and increase the chemical nobility of copernicium through livermorium. Due to their closed-shell configurations, copernicium and flerovium may in some ways behave like the noble gases: both are expected to go further than mercury in the weakness of their metallic bonding and be gaseous at room temperature." -- Sandbh (talk)

I like Sandbh's version and will add it. Many thanks to both of you for your help! Double sharp (talk) 00:20, 23 January 2018 (UTC)

A couple of notes: (1) "copernicium through livormorium" actually includes all but one of the elements listed, but this is not apparent from reading; this seems rather unfortunate. (2) Without context, "lighter cogeners" and "chemical nobility" cry out for wikilinks. YBG (talk) 02:00, 23 January 2018 (UTC)

OK; I've addressed both of these now. Double sharp (talk) 04:58, 23 January 2018 (UTC)

Looks nice, thank you very much! One other thing, I was a bit surprised to see these elements not described in the section about their group, i.e., Cn with group 12, Nh with group 13, Fl with group 14, Mc with group 15, Lv with group 16, and Ts with group 17. Was there a reason for not doing this? No need to make any changes, I'm just wondering why this organization was chosen. YBG (talk) 06:07, 23 January 2018 (UTC)

The reason is in my OP above: if I were to put them in those sections, you'd expect Fl (for example) to get a full paragraph by itself along the lines of Pb, but we simply don't know enough about its chemistry to do that. Why, we haven't even predicted enough about its chemistry to do that: if solving the puzzle of At is like solving a jigsaw with no corners or edges, solving the puzzle of Ts is like solving a jigsaw with no pieces at all. The main thing we know about Cn through Ts is how much we don't know, and I felt therefore that they should be treated as a group, which demands a separate "Transactinide" section. You can see this problem in action in my current FA push for nihonium (feel free to comment on the PR, BTW!), where History is by far the longest section. Hope this helps explain things! ^_^ Double sharp (talk) 07:40, 23 January 2018 (UTC)

Thank you, explains the situation quite well. As long as we have a note at the beginning of the section, do you think it would help the reader to mention something there? Alternately, could the section be reorganized into period-by-period subsections instead of group-by-group? That would trade two singletons (Po,At) for one (Al). I note that only group 11 and 12 have a real summary section - group 13's "summary" section is wholly devoted to Al - group 14 & 15 make no pretense of a summary section - and of course, a summary has no meaning for groups 16 and 17. As I glance through this, I'm vaguely uncomfortable by the amount of text that is in single-element paragraphs. Although these sections are admittedly all about the metallic character of the particular elements, it seems rather too much about the individual trees with precious little said about the groves or forests. One way to rectify this would be to completely reorganize the section so that the organizing principle is not the individual elements, but rather the individual properties, for example, one section about Mohs hardness telling how it varies throughout the PTM and why. Of course, this comment applies equally to most of the element collection articles - the articles about the periods and groups, for example. Please consider these comments to be the idle ramblings of someone too much time but not enough to actually do some editing. I would be most encouraged if my thoughts were considered, but whether they are implemented is of little concern to me. YBG (talk) 15:16, 23 January 2018 (UTC)

I still think the groups are a better way to handle things than the periods, except of course in period 7 where the unifying property is "dunno". ^_^ It's just much easier to do general comparative chemistry when electron configurations and hence valences stay constant. I'll see what I can do to write blurbs on the groups. IIRC last year when the megadiscussion on the nonmetals was in its stride I was planning a look at Ag through Xe to see the increasing nonmetallic character, restricting silver of course to Ag^I; I'll see if I mightn't get something out of that idea. Double sharp (talk) 10:36, 24 January 2018 (UTC)

Yea, I get that, and concede that having sections be about groups generally works better than having sections be about periods. But please do consider my other idea. The current sections are composed of paragraphs about elements which are in turn composed of sentences about properties. Why not have the sections be composed of paragraphs about properties and the paragraphs be composed of sentences that show the progression of the property values from element to element down the group. YBG (talk) 10:47, 24 January 2018 (UTC)

I'm thinking about it: it's definitely tougher to do, but it's a great idea (thanks!) and it's definitely worth trying. I actually think it might overlap a bit with my Ag-to-Xe idea, because that implies judging them against some sort of standard of metallicity, which implies looking at properties. But I'll have to mull over it a bit more to figure out how to start. Double sharp (talk) 11:26, 24 January 2018 (UTC)

BTW, your edit here makes me wonder if these transactinide PTMs are not really just "class D metals". After all: after the "class A metals" and "class B metals", the late period 6 metals (Au through Po are astonishingly weak) and have IIRC been called "class C metals" (do you remember the source which gave that classification? I remember you showed it to me once unless I am very much mistaken), as they receive a double-whammy in the 4f and 5d contractions. Now the late period 7 metals Cn through Ts come with the 5f and 6d contractions plus relativistic effects thanks to the appearance of a quasi-closed shell at Fl, so that especially Nh and Fl are very bad homologues of Tl and Pb. It seems to me that metals like Cn and Fl that are likely gases at room temperature don't even deserve a class-C ranking. In that case they would not really so much be a "superheavy zoo of weirdness", but kind of what you would expect when metallicity is being assaulted from all sides but isn't gone yet. Of course this is just blind speculation so far (and probably will remain so for a while). Double sharp (talk) 07:22, 18 February 2018 (UTC)

Agree, that source was Phillips CSG & Williams RJP 1966, Inorganic chemistry: II Metals, Clarendon Press, Oxford, p. 459: "In some respects these elements [Au to Po] might also be classed as super-B or C metals." Sandbh (talk) 09:06, 18 February 2018‎ (UTC)

Ah, thanks to that I found a more complete quote from Wikipedia talk:WikiProject Elements/Archive 27: "Because of the increase of nuclear charge across each of the transition series, the B metals are distinguished from the early A metals by their much weaker tendency to form ions or to form compounds with the non-metals. The ions themselves have high electron affinities and tend to seek out, wherever possible, polarisable anions or ligands. This feature is particularly marked in the final row of the B metals, Au, Hg, Tl, Pb, Bi, and Po, where the nuclear charge has been built up across the lanthanide as well as the third transition series. In some respects these elements might almost be classed as super-B or C metals." I think I greatly prefer the letter-based nomenclature: "D metals" certainly seems a lot better than "super-duper-B metals". ^_^ This is of course presuming that we don't consider Cn likely being a semiconductor to be the last straw! Double sharp (talk) 13:27, 18 February 2018 (UTC)

Based on my chats with Droog Andrey at WT:ELEM I have written some fuller blurbs for Rg through Og at User:Double sharp/sandbox. Almost all the assertions can be found in various pages on the element articles, although a few are only present in the periodic table he publishes (+3 oxidation state for Cn, a few predictions of oxide basicity/acidity). What do you think of this? Double sharp (talk) 08:15, 21 April 2018 (UTC)

The bios you have written here in your sandbox extraordinary! In which article will you publish them? For the predictions that are only present in Droog Andrey's periodic table I would keep these and add a "citation required". That will give us the best of both worlds. Sandbh (talk) 03:52, 22 April 2018 (UTC)

I see you'll have to update the Cp bio. Sandbh (talk) 03:54, 22 April 2018 (UTC)

They'll go in post-transition metal, once I get around to citing them up in your signature style. ^_^ I've already updated the copernicium bio to read that it's predicted to be a metal with a closed-packed bcc structure (see the bottommost section on your talk page). Double sharp (talk) 04:08, 22 April 2018 (UTC)

@Double sharp: Please go ahead and add them in there now. Add a note to the talk page saying references to follow :) Sandbh (talk) 04:26, 22 April 2018 (UTC)

@Sandbh: I've added them. Any chance we could get a quick update of File:Post-transition metals.png to colour in Rg and Og as green? (Even though Rg has a d⁹s² configuration its predicted chemistry seems to be similar enough to the known chemistry of Au to put them together.) Also it seems a bit odd to put arsenic in the article and not colour it in on the table (and speaking of the metalloids, perhaps selenium and tellurium ought to be given blurbs; if they are sometimes considered heavy metals, then they'd also be post-transition metals by that definition, even though I would disagree with such a classification). Double sharp (talk) 04:37, 22 April 2018 (UTC)

@Double sharp: I can update the template soonish. I'll probably colour Og differently to show that even with a citation required tag, it's speculation on our part---much more so than the other super-heavies. The inclusion of As in the article may be a mistake since there is no citation of it as a PTM that I can recall---will have a look at this. There are no references that I'm aware of that treat Se or Te as PTMs hence their omission. Sandbh (talk) 04:55, 22 April 2018 (UTC)

@Sandbh: Alexandru Babalan includes As, Se, and Te as PTMs in From Chemical Topology to Three-Dimensional Geometry, p. 397ff. As he is discussing Zintl-phase compounds he is presumably classifying them as such because they form similar Zintl-phase compounds to those of the metals proper in the p-block. So, it's been done. ^_^ Double sharp (talk) 05:09, 22 April 2018 (UTC)

@Double sharp: I've also added a blurb on oganesson to your sandbox on the nonmetals. BTW, if astatine is in there, why isn't polonium? I know we don't classify it as a metalloid for good reasons, but they are included as metalloids with about equal frequency. Double sharp (talk) 05:20, 22 April 2018 (UTC)

I'll have a look at that reference. Arsenic I'm OK with but I'm dubious about Se and Te, since they are both conductors. The Og blurb is VG. I'd be reluctant to include polonium in light of its fully metallic band structure. Sandbh (talk) 05:26, 22 April 2018 (UTC)

I've just realised that I've contradicted myself, since I've included germanium as a PTM, even though it is a semiconductor. More work required on my part. Sandbh (talk) 05:44, 22 April 2018 (UTC)

Neither Cn, nor Fl should be gaseous at RT I believe. They are just too heavy. Droog Andrey (talk) 11:03, 22 April 2018 (UTC)

@Droog Andrey: There have been experiments done on this for copernicium suggesting a boiling point lower than that of mercury; the error bars fall below room temperature. Similar thermographical experiments for flerovium have been mentioned in this presentation by Yuri Oganessian, suggesting a boiling point of around −60 °C, although I can't find the paper where this comes from. I presume that, like the copernicium value, this is from using the linear relationship between ΔH_ads (for single atoms or molecules) and ΔH_subl (for a bulk quantity), and then calculating the boiling point from the latter. The first paper on Cn incidentally mentions that "The recent quantum chemical methods seem to underestimate the elemental volatility of element 112." Double sharp (talk) 15:10, 23 April 2018 (UTC)

@Double sharp: I've just made some DFT calculations, obtaining r_e = 3.504Å, D_e = 0.06 eV for Cn₂ (c.f. 10.1016/j.cplett.2014.10.048), and r_e = 3.656Å, D_e = 0.03 eV for Hg₂ (c.f. 10.1063/1.3354976). Now I'm firing at icosahedral Cn₁₃ and Hg₁₃; the results will appear in a day or two. Droog Andrey (talk) 21:12, 25 April 2018 (UTC)

@Droog Andrey: Interesting! I'll look up those papers in the meantime while awaiting your further results. ^_^ Double sharp (talk) 23:37, 25 April 2018 (UTC)

@Double sharp: Binding energies appeared to be 2.76 eV for Cn₁₃ and 1.25 eV for Hg₁₃, while inner interatomic distances appeared to be 3.425Å for Cn₁₃ and 3.490Å for Hg₁₃. So the binding energy advantage of Cn over Hg grows with the number of atoms in the cluster in spite of the faster bond collapsing for Hg. I believe it happens due to larger contribution of 7p compared to 6p, so I expect that the volatility of Cn will be barely higher than for Hg. Fl could be indeed well more volatile than Hg, but it's still hard to believe in boiling point going below room temperature. Droog Andrey (talk) 04:12, 26 April 2018 (UTC)

@Droog Andrey: That's certainly intriguing. In fairness we should note that the Cn and Fl boiling point data is calculated indirectly based on sublimation enthalpies, which is itself calculated based on adsorption enthalpies of single atoms (not bulk Cn or Fl) by an observed linear correlation to sublimation enthalpies of the bulk substance. Thus there are very large error bars and while I am very excited about having even this data I'm aware that it's only half-experimental and that there is a lot of room for improvement. (I also wonder if the observed too-high volatility of copernicium might just be because we are looking at single atoms.) I'll do some de-accenting of these first calculated reports in post-transition metal, copernicium, and flerovium. Based on your calculations I think it might be reasonable to predict Cn and Fl to be liquid metals like Hg, but I don't dare to put that in until you publish them. ^_^ Double sharp (talk) 04:39, 26 April 2018 (UTC)

This is fairly late, but I've now added blurbs for selenium and tellurium to post-transition metal. Citations, like those for the superheavies, are forthcoming when I can spare the time to look through all of my WT:ELEM posts. ^_^ Double sharp (talk) 13:50, 14 May 2018 (UTC)

Regarding your old request for electron affinities for the superheavies

With the help of DMacks I found a value for Rg; I also added values I found for Nh and Mc to Electron affinity (data page). Cn and Fl are stated in the same reference to have no electron affinity, but since it just gives "0" for them just like it does for Hg and Rn instead of a negative number, I have not added these. Double sharp (talk) 14:48, 30 January 2018 (UTC)

I also found a value for Lv and added it. Double sharp (talk) 15:01, 30 January 2018 (UTC)

Thank you Double sharp, even though it has been a long time coming on my part. I appreciate your interest and long memory. Sandbh (talk) 07:01, 18 February 2018 (UTC)

You're very welcome! Double sharp (talk) 07:04, 18 February 2018 (UTC)

Regarding the blocs

I think it might be useful to call out a splitting in the blocs at groups 6 and 14 as a further refinement. For the transition elements and post-transition elements respectively, these groups represent the boundary between hypo- and hyperelectronic elements (considering the four lobes of an sp³ manifold for the post-transitions, and the octahedral clusters prevalent in early transition metal chemistry). Of course the transition one is not as clear-cut (the break is certainly later in the 5d metals than in the 3d and 4d metals). There's a reasonable short account in Chapter 9 of R. Bruce King's Applications of Graph Theory and Topology in Inorganic Cluster and Coordination Chemistry.

I guess the principle that basicity and ionicity go up with increasing size and decreasing oxidation state is also useful as a guide on which way to make the nth-order course corrections. Double sharp (talk) 05:33, 20 February 2018 (UTC)

Further to this talk section

It strikes me that the condensed-phase configurations are an excellent second-order course correction to the first-order explanations that the gas-phase configurations are, and I am sorry I did not remember this earlier. I should really salvage most of these fruitful chats from good old 2016 to the talk page! Double sharp (talk) 06:26, 20 February 2018 (UTC)

P.S. When I have time I'll probably write some of them up as longer essays and present them for your critique! ^_^ Lanthanum by itself has made up a whole big essay, but I also want to return to some of the other topics we have discussed: first-row anomalies (Novalis in chemistry ^_^), Fajans' rules as applied to ionicity, basicity, and electronegativity, condensed-phase configurations (and really, the predictive power of electron configurations), hypo- and hyperelectronicity, the analogies between main-group and transition chemistry, just how metallic the metalloids really are, astatine, the superheavies, and the balance involved in all of this (Mozart in chemistry ^_^). I suspect that trawling through our discussions I shall find enough material to make a whole book out of, and I'm pretty sure we can do something similar with the usual principles of organic chemistry, which really is the singularity of C producing a perfect presentation of the principles of periodicity (I'm not sorry at all for the alliteration). ^_-☆ Double sharp (talk) 06:40, 20 February 2018 (UTC)

Goldhammer-Herzfeld ratio of oganesson

If you have a little bit of time: how close is the GH ratio of Og to unity? I just remembered that Po has a value just below unity (0.95) but is nevertheless metallised by relativistic effects (presumably 6p_3/2 destabilisation). It would be interesting to see on that basis how close Og is, given that relativistic effects seem to be proportional to the square of the atomic number and should thus be about twice as strong for Og than for Po. (Astatine and tennessine are already above 1 from the start, right?) Double sharp (talk) 03:42, 14 March 2018 (UTC)

For astatine, based on its atomic polarisability, molar refractivity, and molar volume, I get a GH ratio value of less than 1.0, suggesting a molecular structure. This happens to be consistent with scalar-relativistic DFT calculations for At of a molecular structure. As you know, the inclusion of spin-orbit and dispersion effects then result in condensed astatine being expected to be a monatomic metal with an FCC structure. Sandbh (talk) 23:58, 16 March 2018 (UTC)

Let me try working it out. Xe and Rn have values around 0.3. The polarisability of Og is about twice that of Rn (pushing things up to about 0.6), while the molar volume is 294/5 = 58.8 g/mol, which is more than that of Rn (50.5 g/mol) and pulls things back down to about 0.5. So, in other words, while Og has gotten closer, it is very unlikely that fcc Og with the Rn structure would be metallic. So the only prospect for a structurally metallic Og is if it has another allotrope that has a higher packing density, which would create a situation rather like that of α-Sn and β-Sn. Double sharp (talk) 15:00, 14 March 2018 (UTC)

An interesting question is to look at copernicium and flerovium as well, as they also have closed-shell structures. Cn has an ideal-like hcp structure and a density of around 23.7 g/cm³ (about what you would expect calculating from Cd and correcting for the non-ideal crystal structure it has). Thus its molar volume is about 285/23.7 = 12.0 g/mol (compare Hg 14.82 g/mol) and its polarisability is close to that of Hg. This implies that Cn has an even higher Goldhammer-Herzfeld ratio than Hg (which is already strongly metallic), and yet calculations predict it to be a semiconductor. Flerovium has a hcp structure and a density of around 14 g/cm³; this gives a molar volume of about 289/14 = 20.6 g/mol (compare Pb 18.3 g/mol) and its polarisability is about two-thirds that of Pb. That would seem to be enough to push down its ratio from the approximately 1.25 on Pb (eyeballing the graph) to about 0.75 for Fl. And yet, all levels of theory predict that Fl is at least a semimetal (like Bi) and is probably a metal in its band structure (10.1103/PhysRevB.82.155116). I am willing to conclude therefore that the GH ratio is not giving useful results in the 7th period where relativistic effects rule. Unfortunately the one article I can find about the solid state of Og simply extrapolates the fcc structure from the lighter Rn and thus I think that the best answer for Og that can be given now is "we simply don't know". (I'll copy this comment to WT:ELEM.) Double sharp (talk) 15:19, 14 March 2018 (UTC)

P.S. If Og does turn out to be a metal (which I find likely now) then we would presumably have to rename the category noble gases to monatomic nonmetals, because Og would be a metal in the noble gas group (like how At and presumably Ts are metals in the halogen group). Double sharp (talk) 01:26, 15 March 2018 (UTC)

Perhaps Pauling's term for the noble gases namely argonons (from the Greek word ἀργόν, neuter singular form of ἀργός meaning "lazy" or "inactive") might make a come back. [That would cover metallic Og, too.] Sandbh (talk) 22:23, 16 March 2018 (UTC)

OTOH, I concur with your earlier observation that, based only on its GH ratio, Og is unlikely to be a metal. Sandbh (talk) 00:02, 17 March 2018 (UTC)

I don't think Og would be inactive enough to be called an argonon either. Regardless of whether the metallisation catastrophe happens, the 7p_3/2 orbitals are expected to be so greatly destabilised that they approach the energy levels of the outer 8s orbitals (that's why Og has a positive EA) and are quite far from the inert [Rn]6d¹⁰7s²7p_1/2² flerovium core. Og would then quite reasonably have a valence core of four electrons and the predicted ionic character of OgF₂ and even OgF₄ suggests rather the chemistry of Sn than that of Ge in group 14, even if it does turn out to be a nonmetal. Its first two ionisation energies are expected to be close to those of Os. Double sharp (talk) 02:43, 17 March 2018 (UTC)

Well perhaps we could focus on their position at the end of the periodic table, with their at least notionally full valence shells, and call them the telogens (Greek τέλος (télos, “end; complete"). But I think the word telogen is already used in chemistry. Perhaps "plerogens" then (Gr. πλήρης, πληρο- full). Something like that. Or plenogens, which is Latin-Greek. Sandbh (talk) 03:44, 17 March 2018 (UTC)

I really like plerogens and it's a terrible shame we can't use it on Wikipedia. ^_^ Failing that I think common usage of the term noble gas already assumes that the element is actually noble and a gas at STP, so we could yet weasel our way out of this for Og. (OTOH, this might bite us back for Cn and Fl.) Double sharp (talk) 04:01, 17 March 2018 (UTC)

P.S. New predictions on bulk copernicium suggest that it will likely be a metal after all, with a body-centered cubic structure. Double sharp (talk) 08:23, 21 April 2018 (UTC)

2013 update 2018a

Thanks. And compliments for the quality, by text and by cooperation. -DePiep (talk) 19:35, 25 April 2018 (UTC)

Interesting findings on Chemistry Stack Exchange

Here's very nice argument I saw on Chemistry Stack Exchange explaining why most of the Madelung violations are in the second rows of the blocks. Ideally, in the d-block elements, (n−1)d is always below ns in energy. Now the 3d orbitals have no radial nodes, so they are very small, and hence there is more repulsion between the d-electrons and it becomes energetically worth it to alleviate this repulsion by promoting some electrons to 4s: hence Fe has configuration [Ar]3d⁶4s², not [Ar]3d⁸. The main exceptions are then for the half- or fully-filled shells at Cr or Cu, which make the promotion not quite as worth it, so that only one rather two electrons get promoted to 4s. The 4d orbitals are larger and so it is usually not as worth it to do this: most of them only promote one rather than two electrons (Tc instead being a sort of anti-exception for its half-filled shell), and palladium even gets away with not promoting any. (Same thing for 4f and 5f, although the 5f situation is heightened relativistically for the first half of the series and dampened similarly for the second half). Now when you get to 5d the relativistic stabilisation of orbitals with lower angular momentum (i.e. 6s) starts to become more important and hence almost all of them are now willing to promote both electrons. Indeed, if you look at the M⁺ and M²⁺ configurations for the 5d metals (leaving out La vs. Lu as contentious), a sizeable number of them give up a 5d electron first (Hf, Re, Ir) or second (Os, Pt, Au) – though then the elephant in the room is why Pt and Au, which have the largest relativistic effects, promote an electron to 6s. I guess we could say that the close-to-fully-filled shell still counts for something for Au and simply appeal to the concurrent destabilisation of 5d making them so close together for Pt that it doesn't really matter. And indeed, this continues for the 6d metals, when the 7s subshell is so stabilised that Madelung's rule is followed exactly from Rf to Cn.

Also, here's a case in the lanthanides (stability of some complexes) where it's Gd which falls off the trend line and not Eu. Double sharp (talk) 16:24, 28 April 2018 (UTC)

Your email

I replied to your email two days ago. I'm really sorry for taking a long time to do so; I've been rather busy during the last couple of weeks. Double sharp (talk) 08:07, 17 June 2018 (UTC)

Got it thanks. No need to be sorry. I've been busy too :) Sandbh (talk) 11:00, 17 June 2018 (UTC)

Raw data, please

Could you send me the raw data behind File:EN & SEP of nonmetallic elementsF.png? Ta! YBG (talk) 07:31, 19 June 2018 (UTC)

Will do as soon as I can. Sandbh (talk) 04:00, 21 June 2018 (UTC)

Done. Sandbh (talk) 12:21, 22 June 2018 (UTC)

Continuing on simple cations etc

I dont think that the elements project are necessarily interested in our mutual interest in detailed chemistry. So I copied part of the conversation and responded.

@Sandbh: Helpful comments. I do like the idea of a preface. Giving the range of oxidation states and mentioning something about important compounds might be one way to give some insights into the element. I am unfamiliar with the meaning of a "simple cation", sounds suspect or archaic. The astatine article might be unrepresentative since its chemistry is so unimportant, not much context since no one uses it. Related to the metals, I have never (or v rarely) heard any inorganicker refer electronegativity except in the case of Au. But these perceptions do not mean that I am correct and might reflect my blindspots. But we can agree on some brief overview statement. Cheers,--Smokefoot (talk) 13:14, 11 July 2018 (UTC)

Simple ion is another word for a monoatomic ion Christian75 (talk) 17:04, 13 July 2018 (UTC)

@Smokefoot: Thank you. What I said was partly wrong; I must have written it in my sleep. Molybdenum does form a simple cation (aq), and the standard electrode potential of molybdenum (–0.20) is negative, not positive. Regarding the meaning of "simple cation", and further to Christian75's comment above, Parish (The metallic elements 1977, pp. 113, 133) discusses the aqueous chemistry of the 4d- and 5d- metals (excluding group 3) and notes that only in a few cases are "simple" aquated cations known. In a similar vein, and in writing about the hydrolysis of cations in aqueous solution, Smith (Inorganic substances: A Prelude to the study of descriptive inorganic chemistry 2000, p. 173) refers to simple cations as "Mⁿ⁺(aq)".

I was surprised about your observation that inorganic chemists don't, or very rarely, refer to electronegativity (but for Au). In scanning the molybdenum article I saw that the Physical properties section made mention of the electronegativity of molybdenum as well as some other chemical properties. For clarity, I've extracted all of these chemical properties and placed them in their own section. And I added a citation-supported sentence about molybdenum's disinclination to form a cation in aqueous solution.

Perhaps the important thing is whether mentioning electronegativity in the preface would add any value, or reader interest. For example, it could be added that "among the non-noble metals, only tungsten has a higher electronegativity."

I feel that flourishes such as this, together with molybdenum's refractoriness; its reluctant cationic nature; and seemingly complex chemistry, have the potential to make the article more engaging. Sandbh (talk) 06:11, 14 July 2018 (UTC)

The electronegativities are not interesting to inorganic chemists (well, at least this one) since the ligands have such an overwhelming influence. IP's are a little more interesting because of the trends wrt stabilization of higher oxidation states. But again all those parameters are washed out with the complexes. One cannot say much useful about W(CO)6 and WCl6 based on EN or IP.

"Simple cations" is probably an old-timey term when the scope of coordination chemistry was limited and chemists were restricted to water as a solvent. Being polyprotic and tending often to oligomerize, metal aquo complexes are complicated. But that is my prejudice. For my own work, we strongly avoid water.

Mo does have such complexes: [Mo(aq)₆]3+ and dimolybdate [Mo₂(aq)_{8 or 10?}]⁴⁺. We need a dimolybdenum article.

In any case, nice chatting with you and keep up the good work.--Smokefoot (talk) 16:24, 14 July 2018 (UTC)

Better talk & improve this wiki through WT:ELEMENTS. -DePiep (talk) 23:41, 21 July 2018 (UTC)

Metal

I am absolutely grateful to you for trying to take on such a broad topic. I definitely want you to get to FA and intend to help you with a review of mine but I am unable to do so at the moment as I'm on vacation and I'm genuinely afraid I may have forgotten by the time I get back home. So if I haven't started a review in, say, two weeks, could you please ping me?--R8R (talk) 08:06, 11 August 2018 (UTC)

Thank you R8R. I have mixed views about getting the article to FA status but let's see how things pan out. I've enjoyed working on it so for although some sections were quite trying, especially getting the Electrical section into something readable, and particularly the energy states picture. I'll ping you in two weeks. Sandbh (talk) 07:17, 12 August 2018 (UTC)

You're very welcome. I think the FA status is actually a good thing to try to go for as the accompanying review provides comments of probably highest quality achievable in Wikipedia, not to mention all the accompanying work you do to get to the bronze star. I agree it's still not the best thing there could be but that's what we get to have. Since you (as I imagine it) really try to go for high-quality articles, I could also suggest you eventually try to go for a WikiJournal of Science publication; I got me one over late 2017 and the first half of 2018. That gave me really good comments from pros from the field and (I imagine) that's what you'd like also. In the meantime, have you considered a Wikipedia peer review? I imagine you'd get a lot of comments for such a broad article.

As for myself, I'm now back at home and I hope to begin my review in the beginning of the next week.--R8R (talk) 07:16, 17 August 2018 (UTC)

A peer review was certainly on my mind. It needs more work before it gets to that stage. I still have at least some structuring work to do. And the history section needs more work. I'd forgotten about the WikiJournal of Science; thanks for the reminder. Sandbh (talk) 07:32, 17 August 2018 (UTC)

Copying within Wikipedia requires attribution

Hi Sandbh. It looks like you copied some content from Bronze to Metal. When you copy from one Wikipedia article to another, you need to provide attribution. This is done by saying in your edit summary that the material was copied, and where you got it. Please have a look at this edit summary for an example of how it is done. Please let me know if you have any questions, or have a look at WP:Copying within Wikipedia for more information. Thanks, — Ninja Diannaa (Talk) 16:08, 16 August 2018 (UTC)

Thank you Ninja. I wasn't aware of that policy, and will observe it from here on. Sandbh (talk) 23:04, 16 August 2018 (UTC)

poly-, di-, mon- atomic

I think I've improved the wording in nonmetal § Alternative categories, but I'd appreciate another set of eyes. YBG (talk) 03:22, 30 August 2018 (UTC)

Yes, you did; I've refined the words some more. For your consideration. Sandbh (talk) 08:37, 30 August 2018 (UTC)

400 ly

this is a fascinating statistic, but what it actually illuminates is not the earth's core, but our lack of intuition about large numbers and volume. Like the surprising fact that a cubic yard is 27 cubic feet or a cubic km is a billion cubic meters. YBG (talk) 14:16, 8 September 2018 (UTC)

Upon reflection, although I did check my calculations, it seems like an unbelievable figure. Are you able to check it? And yes, I too have a hard time getting my head around large numbers. Even one million is a struggle. Sandbh (talk) 22:10, 8 September 2018 (UTC)

Lessee. The volume of the Earth = 1.08321×10¹² km³. The volume of the outercore is 15% of that = 1.625×10¹¹ km³ = 1.625×10²⁰ m³. If we divide that by the 25 m ² base we get 6.5×10¹⁸ m. If we divide that by one light year of 9.461×10¹⁵ m, we get 6.87×10² = 687 light years (so I was low). Sandbh (talk) 10:31, 10 September 2018 (UTC)

Your request has been answered

@Sandbh: Hey sandbh, I just wanted to let you know that I am able to help you out in your attempt to bring "Metal" to FA standard; I posted earlier about what projects the community was working on in the Wikiproject Elements talk page, and you said you would be happy to accept contributions. I will be able to do so, and because you are the chief editor for Metal in a sense, I wanted to check with you to see exactly what you would consider to be the most vital section that needs fixing, and I will try my best to contribute in that area. You may believe that some sections are further along than others, and I don't want to "contribute" in areas that are not in need of contribution. Just let me know what you want me to do to help, and I'll try my best to do it. In the meantime, I've been putting in some time trying to raise Chromium to FA, but I am able to lend my hand here as well. UtopianPoyzin (talk) 14:18, 8 October 2018 (UTC)

@UtopianPoyzin: Thanks for your offer. I don't think there is any particular section needing fixing, but I could be wrong. I believe what needs to happen next is:

• a read through to check the article content is understandable, that the article flows OK, and that the lead appropriately reflects the main body of the article
• checking the images for alt-text inclusions
• citations need to be added
• wikilinks need to be added.

R8R has suggested hiving off a few of the larger sections into their own articles. Do you see a need for this?

If you see anything interesting here, feel free to have a go. I'll see if I can find time to have a look at the Cr article. Sandbh (talk) 01:09, 9 October 2018 (UTC)

@Sandbh: Sorry if I missed most of the groundwork for the article creation, but I'd love to read through it in retrospect and make small edits anywhere I feel it is necessary. I'm not that great with citations and citation format; however, I personally am able to add in the wikilinks. Once you feel like its 100% locked down, I'll also give my complete coverage of the article if you do decide to create a peer review for Metal (which I personally would). Anyway, I just finished reviewing History of aluminium, and if I continue at the speed which I did that (90 minutes, yeesh), I won't have time to review Metal. I'll likely revisit Chromium and head off for the day, and take a look at Metal tomorrow. Expect a review if the peer review isn't up yet! UtopianPoyzin (talk) 01:36, 9 October 2018 (UTC)

@UtopianPoyzin: Thank you. Reading it through and making small edits anywhere you feel necessary will help. Sandbh (talk) 02:37, 9 October 2018 (UTC)

Periodic table and TFA

re your post [1] on 25th: your point "2. The project nominates an article." (the project being WP:ELEMENT, I understand). Actually, it was a TFA coordinator who proposed Periodic table: [2] (Dank, see the timeline I added in that talkpage). It was not proposed or suggested initially by WP:ELEM editors. That said, no big deal (and no problem just weakening your argument a tiny bit). -DePiep (talk) 16:48, 27 December 2018 (UTC)

Re (non)metal(oid)s

I found the note you wrote over at WT:ELEM interesting: Yes, minimising the number of exceptions is the criterion, since it is quantitative and applies to the whole of the table. Wondering if this might affect your thoughts about your attempt to define metals based on various properties in the spreadsheet you e-mailed me long ago. YBG (talk) 18:08, 27 December 2018 (UTC)

Bad link

The link you added here doesn't work. It looks like you copied it from someplace where the abbreviated the URL by adding ellipses. YBG (talk) 05:58, 30 December 2018 (UTC)

Managed to fix it. Looks like you copied the abbreviated link text out of google. I grabbed it by right-clicking on the document title and selecting "Copy link address". YBG (talk) 06:02, 30 December 2018 (UTC)

Just read the lede paragraph of the article, but I'll have to wait to look at the whole thing. What I'm hoping for is not just an answer to the specific Group 3 question, but whether there might be something instructive about how such questions in general ought to be decided. Now the answer to that meta-question should be a very useful indeed. YBG (talk) 06:07, 30 December 2018 (UTC)

@YBG: Thank you! Been working on a response to Double Sharp, hence not much activity from me recently. Sandbh (talk) 06:23, 30 December 2018 (UTC)

Happy Australia Day!

As I see, it is already the twenty-seventh in much of Australia. I can only hope that you live in a timezone that makes this congratulation still timely, or at least that it's the effort that counts: after all, it is only 18:15 back at my place and I somehow forgot to take into account that time is different throughout the globe.

By the way, when you mentioned that people often go on vacation from Christmas through Australia Day, I was a little puzzled at first. Why would people massively take their vacations when it's so cold outside? It took me a few seconds to remember that not only is Australia much closer to the equator, but is also in the Southern hemisphere :) --R8R (talk) 15:13, 26 January 2019 (UTC)

The flag flying

Thank you. I remember seeing this post at the time then I got caught up in other things, even though I wanted to say something. Oh yes. Things go crazy in the period leading up to Xmas Day, as everyone wants everything done before Xmas. Then it's "tools down" mostly until Australia Day, which is 26 January. During the break there is cricket, and tennis, and New Years Day festivities, and a lot of people go on holidays, down to the beach etc. The kids go back to school about a week or so after Australia Day, and Parliament resumes shortly thereafter. And there are also droughts, flooding rains, and bushfires to manage (this time in Tasmania, of all places). It is a relaxing time of year but can also be challenging in some parts of the country. Not forgetting this year's mass fish kills. Sandbh (talk) 23:36, 19 February 2019 (UTC)

Where'd that come from

Please check this edit. You seem to have added a stray URL that broke section heading format. Not sure what was intended or I'd try to fix it myself. YBG (talk) 14:47, 13 February 2019 (UTC)

(talk page stalker) @YBG: I've removed the stray URL, as it appears later in Sandbh's comment. Double sharp (talk) 00:25, 14 February 2019 (UTC)

Yes, it was a stray edit. Thanks. Sandbh (talk) 00:26, 14 February 2019 (UTC)

classifying Nh and Fl as metalloids

I've seen your email and responded briefly, but I wanted to mention the paper (10.1103/PhysRevA.99.022110) here because I think it'd be useful to cite for WP (feel free to delete this from your talk page if you want to keep it between us). It's cool to see a citeable prediction of Ts and Og as metals; it's also cool to see the swap in properties between groups 13/17 and 14/18 in the 7th period. I expect it won't be long before we can't rationalise the Madelung placement at all and we have to adopt something like Fricke's, Nefedov's, or Pyykkö's displacements for period 8. ^_^ Double sharp (talk) 15:03, 19 February 2019 (UTC)

P.S. I see your 2013 J. Chem. Ed. article got pride of place as cite number 1 on it. ^_^ Double sharp (talk) 15:03, 19 February 2019 (UTC)

Giovanni Dienstmann moved to draftspace

You've provided one source, which is to a book whose contents are not searchable online, without a page number in the reference. Please add enough citations to meet WP:GNG, and page numbers for book references, then feel free to move the article back to the mainspace (no need to bother with the AfC system, but that stuff gets automatically added when this move to draft tool is used). _signed, Rosguill ^talk 05:33, 23 July 2019 (UTC)

If you remember the argument about s-block carbonyls from User:Sandbh/Group 3...

So they do exist after all for Ca, Sr, and Ba! Double sharp (talk) 03:15, 5 August 2019 (UTC)

Yes, thank you. I saw that article a little while ago. I suppose we shouldn't be that surprised given the influence of the empty d-shells on Ca, Sr and Ba. It doesn't make any difference to what we said about the carbonyl argument in our IUPAC submission. We said that the carbonyl argument would call into question the periodicity of groups 1 and 2 but I don't think that's an argument anymore. Sandbh (talk) 00:59, 8 August 2019 (UTC)

Yeah, our argument worked regardless of whether group 2 carbonyls actually existed, but it is undoubtedly more satisfying to know that they do! ^_-☆

It is interesting to note that while Ca, Sr, and Ba show significant d involvement Zn, Cd, and Hg do not. And yet the latter three are universally regarded as d-block elements because Ca, Sr, and Ba are far closer to the rest of the s-block than Zn, Cd, and Hg (the first three show the characteristic pre-transition properties). This is basically the argument I would use for Sc-Y-La-Ac; yes, while La has more f involvement than Lu, it is weak for both of them. But it is fairly clear that Lu acts more the other nine 5d metals than La does, just looking at its basic properties. Double sharp (talk) 03:48, 8 August 2019 (UTC)

For Ca, Sr, Ba I think the fact they have zero d electrons is a huge mark against them. For Zn, Cd, and Hg, I think G & E mention what few d-block properties they do have, and this presumably seems to be (barely) good enough (?) to count them as d-block metals even though they behave predominately like PTMs. As ions, they also have d10 configurations, similar to Lu having an f14 configuration. You could hardly call them p-block metals, could you? They are just greedy metals, since they closed their d10 shells prematurely. Ditto calling Ca, Sr, and Ba as d-block metals.

I'd be wary arguing that Lu acts more like the other nine 5d metals, when you take into account:

• The oft-cited Spedding and Beadry (1968, p. 377) who wrote: "Since metallic lutetium resembles closely erbium and holmium, except that it melts at a slightly higher temperature and is essentially non-magnetic, the details of producing, purifying and fabricating it are almost identical with those described under Holmium."
• Restrepo's work on the binary compounds of the elements found that La shares similarities with transition metals and with lanthanides, while Lu is more similar to lanthanides than transition metals. Thus, according to him, La is better positioned under Y than is the case for Lu.
• Glawe et al. (2016) came to essentially the same conclusion.

I'm OK calling La a d-block metal, like Na is an s-block metal. I'm OK calling Lu a (greedy) f-block metal, given its ionic configuration of f14.

All that said, the bifurcated group 3 option, which it turns out has been around for a while, is still my first choice.

Glawe H, Sanna A, Gross EKU & Marques MAL 2916 “The optimal one dimensional periodic table: a modified Pettifor chemical scale from data mining”, New Journal of Physics, vol. 18, 093011

Sandbh (talk) 04:57, 8 August 2019 (UTC)

Placeholder I'll try to respond tomorrow by Sunday (I should know by now that saying a specific date is a good way to jinx things...) Double sharp (talk) 16:26, 10 August 2019 (UTC). Double sharp (talk) 15:49, 8 August 2019 (UTC)

OK, here we go. Ca, Sr, and Ba being d⁰ is indeed a strike, but not a totally unprecedented one, as Th is also f⁰ and everyone seems to agree that it is an f-block element for lack of anywhere better to put it and because it has f-involvement in its chemistry. Well, Ca, Sr, and Ba have d-involvement in their chemistry too! ^_^ Zn, Cd, and Hg get in the d-block instead for a number of reasons instead:

1. Physically speaking, the group 12 elements are fairly obviously closer to the sure d-block elements in groups 3 through 11 than the group 2 elements are, mostly because the group 11 elements already show the characteristic post-transition metal properties physically.
2. The alkaline earth metals form hard cations, whereas Zn²⁺ and Cd²⁺ are intermediate and Hg²⁺ is definitely soft, which makes them resemble the late transition metals. In particular Zn, as Greenwood and Earnshaw remark, is much more willing to form covalent compounds than Mg, and is perfectly willing to form complexes with N- and S-donor ligands as well as O-donor ligands.

Obviously, you cannot put Zn, Cd, and Hg in the p-block as they are p⁰ and don't show significant p-orbital involvement in their chemistry. So the question is simply whether they or Ca, Sr, and Ba should get the second s-block position in rows 4, 5, and 6. The preceding arguments show that Zn, Cd, and Hg fit better in the d-block; and there are many arguments for Ca, Sr, Ba in the s-block, such as their hardness, their often analogous behaviour to the alkali metals, and uniting the unique situation of having a few valence electrons in an s-subshell and nothing else outside a hard and totally unbreachable noble-gas core with group 1. But I would not say that Zn, Cd, and Hg "prematurely" closed their d-shells. They filled them exactly when they should have if we just count electrons; it's Cu, Ag, and Au that have done so a bit early.

Lu is a different case from Zn; you can legitimately call it a d-block element because it does have d¹ outside the f¹⁴s² closed shells. Zn does not have such an extra electron. The arguments you list regarding Lu are not exactly relevant here. First of all, Restrepo (as I noted before) is concerned about stoichiometry, which implies valence, and thus it is obvious that similarities with elements from different groups like Lu vs. Hf will be ignored. But secondly, the procedure is not to only look at which of La and Lu are more similar to the other lanthanides: obviously this is going to be totally inconclusive because of how similar the lanthanides are to each other. We need to ask what we did for the analysis of Zn, Cd, and Hg above, i.e.: what are the characteristic properties of the d-block elements, specifically the nine sure 5d elements from hafnium to mercury inclusive, and does La or Lu display more of those properties? In every conclusive case the answer is Lu.

1. The 5d metals are usually small cations due to the lanthanide contraction; so Lu³⁺ obviously fits better than the La³⁺ as it comes when the lanthanide contraction is at its end. (In fact Lu³⁺ has a similar ionic radius to Au³⁺.)
2. The early 4f metals often show rather large coordination numbers uncharacteristic of the d-block metals, because the latter are not large enough. Again, Lu (being the smallest lanthanide) is the closest fit here.
3. The 5d metals are more covalent than ionic in most of their chemistry, and their oxides and hydroxides mostly acidic or at least amphoteric: as expected from Fajans' rules, Lu more closely approximates this behaviour than La, as Lu(OH)₃ is (barely) amphoteric.
4. Most of the sure 5d metals are intermediate or soft cations (Hf being the only exception); clearly Lu is closer, as it is the smallest and hence softest lanthanide cation.
5. The 5d metals are very dense, and the early ones are very hard and often refractory. They are also mostly unreactive in the bulk form. Lutetium, being the hardest, densest, most refractory and least reactive of the lanthanides, is obviously closer to this kind of behaviour than lanthanum, a soft metal which corrodes in the air.

This is more or less what settles it for me: even though sometimes a Sc-Y-La trend looks better, sometimes a Sc-Y-Lu trend looks better, and both La and Lu are full of similarities with the other thirteen lanthanides, it is hands-down obvious that lutetium is a better fit with the other nine 5d metals than lanthanum is. We can for sure put an extra 3 above the La-Ac small column to alert people that the Sc-Y-La trend is also worth looking at, but I submit that the case for Lu under Y is clear-cut because it leads to a more homogeneous d-block. The issue of which elements should be in group 3 should be decided by the prospective group 3 elements and hence can remain unresolved and probably unresolvable, but the issue of which elements should go in the d-block should be decided by looking at everything in the d-block, and that provides an answer. So for me, yes, La and Ac can also be extra group 3 elements just as Lu and Lr are, but the d-block positions below Y in the table must be occupied by Lu and Lr. (Obviously we have to extrapolate a lot to do this analysis for Lr, but the actinide contraction by itself ensures that the points I wrote above for Lu are almost certainly valid for Lr as well when you compare whether actinium or lawrencium acts more like the sure 6d metals from rutherfordium to copernicium inclusive.) Double sharp (talk) 15:25, 11 August 2019 (UTC)

I'll try and look at this in more depth, subject to study commitments.

There is something nagging away at me about the fact that we know why Lu has the similarities it has to the nine 5d metals i.e. the Ln contraction. So, yes, if La is under Y then I'd expect weirdness to ensue, once the Ln contraction, which runs from Ce to Lu, concludes.

The d-block becomes less homogenous due to the Ln contraction. It becomes more homogenous, by one element, if La is under Y. This gives rise to a controversial split-d block. However, it can be argued that the somewhat greater complexity of a split-d block is merely a reflection of the actual complexity of chemistry itself (adapted from here).

Placing a 3 above La in an Lu table washes this away. However, I'd like to express the above more clearly, at least for my own understanding. I'll get back to that. Sandbh (talk) 12:34, 14 August 2019 (UTC)

The fact that you first see the collapse of 4f at Ce is merely an illustration of delayed collapse that gets common for elements with higher atomic numbers: witness how 5f doesn't collapse below 6d until Pa, 6d doesn't collapse below 7p until Rf, and the collapses of 7d, 6f, and 5g below 8p seem to be staggered respectively to E122, E123, and E125 respectively (E121 should be [Og]8s²8p¹; the foregoing atomic numbers are when the first 7d, 6f, and 5g electrons are expected to appear in the ground-state neutral gaseous atoms respectively.)

The importance of the Ln contraction over most contractions in the periodic table is that the lanthanides all have a dominant +3 oxidation state across the series, and therefore a very salient difference between them is the exact size of those Ln³⁺ cations. As such La³⁺ and Lu³⁺ are both highly relevant to a discussion of the lanthanide contraction as the extreme 4f⁰ and 4f¹⁴ cases, and as expected La³⁺ appears as the largest and the hardest tripositive Ln cation, surpassing Ce³⁺. As such claiming that the Ln contraction happens from Ce to Lu instead of from La to Lu is rather missing the point IMHO about why we like to talk so much about this contraction. Lu is smaller than Yb, which is smaller than Tm, and so on until Pr is smaller than Ce; why should we not extend it and say that Ce is also smaller than La, when all fifteen are so chemically similar?

Every 5d element from Hf to Hg obviously has to deal with the aftermath of the lanthanide contraction, as their configurations include a [Xe]4f¹⁴ core. As such the d-block obviously becomes more homogeneous if Lu is included as the first 5d metal rather than La, not less, as Lu follows the other nine in having this core and the concomitant Ln contraction effects while La does not. Putting Lu in the d-block, all ten of its columns are affected by the Ln and An contractions, whereas putting La in the d-block makes it have one column unaffected and the other nine affected. Not only that, but since the Ln configuration almost perfectly cancels out the expected expansion going from the 4d to the 5d elements, it makes the d-block even more homogeneous by creating "twin pairs" of Y/Lu, Zr/Hf, Nb/Ta, Mo/W, and Tc/Re, where the behaviour of the 4d and 5d element in the same group is near-identical. Putting La under Y makes us lose one such twin pair. The aftermath of such contractions is not weird; the same thing happens already with the 4p elements Ga–Kr once the 3d contraction finishes, and as expected the p-block is made more homogeneous by that by virtue of the atomic radii of Ga–Kr being brought down to become close to those of Al–Ar. More generally we can expect to see incomplete screening effects absolutely everywhere in the periodic table with the exception of periods 1 and 2 (the first has no screening and the second has excellent screening due to 1s² outshining even the filled p-shells of the noble gases), so it is not exactly weird, even if the 3d and 4f ones are more striking than the other ones. In some sense it is the elements we know and love from H to Ne that have the weird and extreme properties (see that wonderful radial nodes paper at 10.1002/jcc.20522 again) – which is, of course, why we know and love them (and indeed run on them to a great extent). ^_^ Double sharp (talk) 13:56, 14 August 2019 (UTC)

P.S. The excellent screening of 1s² even compared to the noble-gas ns²np⁶ cores from n = 2 onwards, which is responsible for why Li is too electropositive for its position in the periodic table (for example), is actually the best argument I've yet come across for putting He over Be. Mind you I don't recall seeing it put forward in this capacity, but it's interesting nonetheless. (It ties in well with why Ne is more inert than He, which is an argument that has been raised to support He over Be, but I feel this one has more force because it affects the following elements like Li and Na, and isn't simply a matter of complaining about who gets the crown when He, Ne, and Ar at typical conditions are all absolutely inert.) Double sharp (talk) 14:16, 14 August 2019 (UTC)

But wait!

18-electron complexes questioned Sandbh (talk) 12:11, 13 August 2019 (UTC)

As I understand it, the doubt is not whether M(CO)₈ exists for M = Ca, Sr, Ba, but whether the bonding in them is really most naturally explained by invoking d² configurations like for transition metal complexes, rather than simply considering them to be saline compounds formulated M²⁺[(CO)₈]²⁻ (as expected, since Ca is a good reducing agent whose chemistry is mostly Ca²⁺, to say nothing of the heavier two). Our old argument is still valid (and was valid anyway even before we knew about those compounds), because it was simply showing how counting towards the next noble gas doesn't work well for the s-block. The pro-Sc-Y-Lu argument that we were rebutting was that La needs 29 electrons to achieve a [Rn] configuration while Lu only needs 15, the same as Sc and Y do to achieve [Kr] and [Xe] configurations respectively; we countered that by noting that by that logic Ba, which needs 30 electrons to achieve a [Rn] configuration, cannot be placed below Sr, which needs only 16 to achieve a [Xe] configuration, which is absurd. (Of course this argument is minor and merely demonstrates that positioning elements in the periodic table is more subtle than simply counting along.) Double sharp (talk) 12:39, 13 August 2019 (UTC)

Copernicium

Might be an insulator with a band gap of 6.4±0.2 V! Compare Rn at 7.1&nbsp:V! o_O Double sharp (talk) 16:59, 15 October 2019 (UTC)

That's rather extraordinary. Then again, it could be interpreted as relativistic effects causing a reduction in metallic character going down group 11.

And how would we categorise Cn? A relativistic nonmetal? Or call He, Ne, Ar, Kr, Xe, Rn, and Cn the noble fluids or the noble nonmetals?

I like noble nonmetals in analogy to the noble metals. Sandbh (talk) 22:27, 25 November 2019 (UTC)

Draft:Giovanni Dienstmann concern

Hi there, I'm HasteurBot. I just wanted to let you know that Draft:Giovanni Dienstmann, a page you created, has not been edited in 5 months. The Articles for Creation space is not an indefinite storage location for content that is not appropriate for articlespace.

If your submission is not edited soon, it could be nominated for deletion. If you would like to attempt to save it, you will need to improve it.

You may request Userfication of the content if it meets requirements.

If the deletion has already occured, instructions on how you may be able to retrieve it are available at WP:REFUND/G13.

Thank you for your attention. HasteurBot (talk) 01:26, 26 December 2019 (UTC)

An unrelated note re Group 3

Part of the discussion about group 3 relates to which set of four elements matches better with groups 2 and 4. This got me thinking of the question:

For each N, does Group N correlate better with Group N-1 or with Group N+1.

Does this question even make any sense? Does it prompt any curiosity? Does it provide any insights? Just wondering. YBG (talk) 08:58, 10 January 2020 (UTC)

Ta, potentially there is something in this. Here are my guesses. I’m not seeing a pattern yet.

s-block

1 is closer to 2

2 is closer to 1

d-block

3 is closer to 2

4 is closer to 5

6?

7 is closer to 6

---

8-10 are closely related

---

11 is closer to 10

12 is closer to 13

p-block

13 is closer to 12

---

14 is perhaps as close to 13 as it is to 15

---

15 is closer to 16

16 is closer to 15

17 is closer to 16

---

18 is as close to 17 as it is to 1

---

Sandbh (talk) 21:37, 10 January 2020 (UTC)

You put a ? for 6 but forgot about 5 so I've put ? for both. Please verify that I've captured what you mean here:

		s-block			d-block											p-block
Predecessor		18	1		2	3	4	5	6	7	8	9	10	11		12	13	14	15	16	17
Closer?			**		**		**		**		*	**	**			**	*		**	**	*
Group in focus		1	2		3	4	5	6	7	8	9	10	11	12		13	14	15	16	17	18
Closer?		**				**		**		**	*			**			*	**			*
Successor		2	3		4	5	6	7	8	9	10	11	12	13		14	15	16	17	18	1
		s-block			d-block											p-block

YBG (talk) 06:32, 11 January 2020 (UTC)

It looks about right now. Sandbh (talk) 08:23, 11 January 2020 (UTC)

Here's another representation:

	Arrows point away from less similar adjacent group and towards more similar adjacent group
	←← more like previous group						←→ equally similar to both						→→ more like following group
Blocks	s		d										p
Groups	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18
Similarities	→→	←←	←←	→→	←←	→→	←←	→→	←→	←←	←←	→→	←←	←→	→→	←←	←←	←→
"Peaks"			↑		↑		↑				↑			↑				↑
"Valleys"	↓			↓		↓			↓			↓			↓

YBG (talk) 08:58, 11 January 2020 (UTC)

Here is the reason why I think this might be interesting.

Blocks	s		d										p						Blocks
Groups	1	2	3LaAc	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	Groups
Similarities	→→	←←	??	→→	←←	→→	←←	→→	←→	←←	←←	→→	←←	←→	→→	←←	←←	←→	Similarities
Blocks	s		d										p						Blocks
Groups	1	2	3LuLr	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	Groups
Similarities	→→	←←	??	→→	←←	→→	←←	→→	←→	←←	←←	→→	←←	←→	→→	←←	←←	←→	Similarities
Blocks	s		d										p						Blocks

Or maybe not so interesting. YBG (talk) 03:00, 13 January 2020 (UTC)

I'm taking a break

And in light of that it seemed appropriate for me to write two things to you. One is that I, for aforementioned reasons, will not be able to respond to you during the continuation of the discussion on your draft. So you might want to close the review soon enough or wait for my return, which is probably going to be somewhere between the middle of the next week and the end of the month.

The other thing is that I wanted to give you a general comment on the review to sort of compensate for my temporary absence. I generally hold that the conclusion is hasty. Not because I don't prefer the conclusion myself---more on that later---but because I genuinely believe a proper case for it has not been made, at least not in the paper as it is. And many of my questions are not an attack per se to sink the paper, but rather a hunch on what's missing in your argumentation and I believe would generally aid it. Could such a case be made in principle? I don't know. Or at least I'm hesitant to say so. I believe that most arguments you give, perhaps all save the popular usage one, are either insignificant or can be easily turned back to support -Lu-Lr instead, which again makes the conclusion a matter of preference. I recall that there was once (rather long ago, I think: IIRC Double sharp had not switched to -La-Ac by then) a discussion at WT:ELEM in which there were a few unique arguments that could not be mirrored, and if I were to get to the bottom of this, I'd try to find that discussion from long ago and revisit those arguments. But as a general comment, it was to a great extent thanks to your critique that I learned to look at the other side of the argument (to look past contiguous blocks, for instance), so I'm asking you to do the same and try to not withstand attacks for a person who holds a different point of view, but rather to convince an informed but undecided reader. I think you could do the former but I'm not so sure about the latter. I've sincerely tried to help your case by my questions rather than attack it, but I think that a complete argumentation might render many arguments unresolute.

I also fear, and I hope to be wrong at that, that the discussion with Double sharp is not helping you much---not because DS is not giving you reasonable arguments but rather because, as it appears to me from the sidelines of your discussion, you two cannot truly hear each other. I would be inclined to think that a complete consideration would include his arguments too, and then try to compare each one hand-by-hand by the same metrics. If I were to summarize in one sentence how your paper could be improved, it's that---it could use a comparison of both sides of each argument by the same metrics. I don't know if that would yield the same result or whether it would yield a definitive result at all, but if a result was to be gained this way, it would stand a much greater chance to persuade me, even if into the option I don't usually fancy. The same, I believe, is also to be said of your intended readers.

I wish you luck with the paper.--R8R (talk) 03:36, 4 February 2020 (UTC)

Some words

First of all, I have taken some time away from the issue and calmed myself down. So I have some words now.

1. I have gone overboard, and I apologise deeply for it.
2. As such, I have withdrawn the RFC.
3. I still disagree with your approach and continue to believe that Lu under Y is significantly better. However, I will endeavour to keep this a respectful disagreement and not make it descend into what it previously turned into.
4. Since for Wikipedia the factor that most textbooks currently still show a "La" in the box under Y is still pretty relevant, I have decided that it is best to let go of the issue. The reason behind my decision is that I realised that I have never gotten terribly worked up about He over Ne even after I adopted the view about a year ago that He over Be is better. That's mostly because there's no chance WP is going to show that now. Therefore, in hindsight, it is probably not worth getting worked up about this one either. If I believe it is right, that's fine and well, but this is not the place to start evangelising for the change. Others have started doing it already in the right places.
5. Therefore, I would be completely satisfied if the relevant articles (i.e. periodic table, group 3 element, lanthanum, lutetium, actinium, lawrencium) were updated to include a brief and neutral footnote (in the first two cases, a paragraph or section) stating that there is a dispute, rather like the one I wrote for helium. I hope that solution is fine with you.
6. I do not think it will be productive to start another RFC even after your article is published, because the above situation will almost certainly not have changed significantly. Therefore, I withdraw the idea to do it as well.
7. I also do not think it will be very productive to start discussing this again right now, because it's by now quite clear where our approaches differ. I think it would be better to just agree to disagree for now. We each have a right to our own viewpoints, but let's be neutral on Wikipedia. It will be more productive for the reader if we invest our WP time on actual articles instead.
8. In the event that the IUPAC project decides in favour of Lu under Y, and only if that happens, I plan to start drafting an RFC. But I will endeavour to make it neutral. To that aim, I would like to request collaborating on the writing of such an RFC with you, if that situation should happen. But this is all conditional on that event, of course.

Again, I apologise for my previous behaviour.

Respectfully, Double sharp (talk) 09:37, 21 July 2020 (UTC)

Regarding your sandbox summary

I am happy with most of your redrafted section. However I have two comments that make me not totally happy with it. One small and one big.

The small one. The solid state electron configurations of the lanthanides are, at least as far as what we can say on Wikipedia, disputed. Even in 2016 Gschneidner was still talking about 4f hybridisation in the lanthanides including La. Maybe some others doubt it. But I do not think this can be considered a consensus to count anomalies against, and therefore I am not very happy with the last two paragraphs of "Gas phase v condensed phase".

No drama. We could add something about this.

The big one. I feel you miss a chance to clarify to the reader one rather salient difference between the two approaches. From our discussion it seems that the notion of the differentiating electron, and the ground-state gas-phase electron configuration that they are based on, is important to the case for La. However, neither of these notions play any large part in the case for Lu. When Jensen wrote about what you called the thorium conundrum, under "The placement of lanthanum and actinium", he did not refer to differentiating electrons. Indeed the words "differentiating electron" do not appear at all in his article. Rather he refers to the valence configuration without regard to what electron is added from the previous element. Therefore I think the counterargument you mention from Scerri, which talks about differentiating electrons, does not directly address the thorium conundrum in the way that Jensen puts it. Additionally Jensen does not only refer to thorium, but also lutetium: at stake is that he clearly considers the f electrons of Lu to be core electrons, so neither La nor Lu have valence f electrons in the kinds and numbers of valence electrons and/or vacancies (which implies looking at valence orbitals). And if you remember Jensen's reply to Lavelle, you can see that Jensen considers the periodic table to be based on idealized rather than actual electron configurations. (In support of his statement it may be observed dispassionately that Cu and Ag have different differentiating electrons, and Ni and Pd have different ground-state gas-phase electron configurations, but all generally accepted layouts of the periodic table reflect neither of these.)

Therefore I feel that mentioning that the Lu case is based primarily on valence orbitals and idealised configurations rather than differentiating electrons in the ground-state gas-phase configurations would help clarify the reader's understanding that there is this difference between the two approaches: they appear to start by considering different things to be important. Please correct me if I have mischaracterised the importance of differentiating electrons and ground-state gas-phase configurations to the La approach.

Therefore, I feel that a statement about the chemical activity of the f orbitals of La and Lu should be included, because this is something that is important for the Lu case just as differentiating electrons and the ground-state electron configuration seem to be important for the La case. The argument that La and Ac have non-hydrogenic low-lying f orbitals appears in both Hamilton and Jensen, and indeed Jensen goes on to add that Lu and Lr do not have these. Valery Tsimmerman says the exact same thing in his 2018 paper. Note, this is not to convince you, it is simply saying that this is what Lu advocates in the literature refer to.

Hence I recommend a sentence or two about how 4f involvement has been suggested and disputed as an explanation for some of lanthanum's properties be added, as should a statement about what calculations say re whether 4f is a core orbital for lutetium and the extent and existence of 4f contribution to the bonding for La.

(Incidentally, Tsimmerman's term symbol argument about each block ending in ¹S₀ only in a Lu table, and Eu and Am showing the expected term symbol for the half-filled f elements, seems like it would be a reasonable addition because it can be stated so briefly like I just did. But it is not a must-have; only the above two, I feel, are the big issues.) Double sharp (talk) 14:18, 22 July 2020 (UTC)

I see no drama here, either. I'll look closer at the draft to see how your feedback could be addressed. Sandbh (talk) 02:20, 23 July 2020 (UTC)

@Double sharp: Thank you. The purpose of my sandbox summary was only to see what it looked like without the comments. I was surprised to see the word counts were so close. I haven't yet looked closely at the two summaries to see if they do a fair job of capturing the key arguments. I intend to do that shortly, after attending to a few other matters. Sandbh (talk) 02:12, 23 July 2020 (UTC)

OK, no problem.

I also notice I forgot to mention something: it would probably be a good idea to explicitly define the differentiating electron, at least in a footnote. I think we have gone over by now the problem that the definition is not completely obvious for a case like zirconium [Kr]4d²5s² vs niobium [Kr]4d⁴5s¹. Double sharp (talk) 03:23, 23 July 2020 (UTC)

@Double sharp: OK. I forgot to mention the old word count in the PT article was ~300 words for the two options. That seems appropriate. We spoke about moving the magnum opuses for each of the two options into the Group 3 article (I recall you had no objection). Sandbh (talk) 05:41, 23 July 2020 (UTC)

I indeed have no objection. I think the section in the periodic table article should cover the basic arguments and state where they are coming from. Hence my suggestions above. Double sharp (talk) 08:20, 23 July 2020 (UTC)

Working towards a new 18-column navigation table for Wikipedia

Hi Sandbh. Thanks for contributing to my Talk page discussion on a new navigation table for the Periodic Table. Your first comment was that, like me, you think the 32-column version is pretty poor as it gives such tiny cells for a user to click onto, especially when in an element's infobox. Hence we are now trying to design an 18-column replacement. We should certainly bear in mind R&R's view that whatever we come up with should be used consistently across en:Wikipedia for other similar templates but I don't think that implies necessarily being consistent with the literature, as by definition that's impossible if you want a single navigation table when the Group 3 debate (and similar debates) show there is no such thing. As I implied elsewhere, we are designing a standard navigation table for Wikipedia, not trying to create the internationally accepted version of a printed table such as the one currently (badly) given on IUPAC's website.

I have no axe to grind about what the details of the new navigation table are. I just want a new and improved standard for Wikipedia! It seems to me as an organic chemist not involved in the Group 3 debate that the current version 3 in my thread is pretty good. If you disagree, please try to say what you would like instead. I take the point, for example, that version 1 had no extra separator: that's because of my lack of skill at editing the template — I certainly intended there would be one in the final version. There has to me some momentum now for creating a new 18-column table to become {{compact periodic table}} and a final push should see us get there :-) Michael D. Turnbull (talk) 10:47, 23 July 2020 (UTC)

​Hi Michael

The La form is the most common in the literature by a 4:1 margin compared to the 15-column f-block form with both La and Lu under Y, and by a 4:1 margin compared to the Lu form.

The Group 3 debate (and similar debates) has not shown there is no such thing.

Wikipedia is a tertiary source or encyclopedia that reflects what the literature says.

The table shown on the IUPAC website is not an internationally accepted version. There is no WP:RELIABLE evidence supporting this assertion.

If we are to have an 18-column table in our info-box header then it should be an La table, consistent with the most common form in the literature. Sandbh (talk) 13:25, 23 July 2020 (UTC)

Thanks, that clarifies your views for me. I've added a couple of thoughts in the main thread, and am now suggesting an "or" solution that I know you won't like very much but may be prepared to accept. I suggest you make further comments in that thread. Michael D. Turnbull (talk) 13:35, 23 July 2020 (UTC)

Thank you

Thank you for the link to this wonderful Scerri (2020) article [3]. Right into my interest :-) It might also trigger me into a reply somewherehow. -DePiep (talk) 19:13, 27 July 2020 (UTC)

WP:ANI

There is currently a discussion at Wikipedia:Administrators' noticeboard/Incidents regarding an issue with which you may have been involved. Thank you. Double sharp (talk) 02:08, 4 August 2020 (UTC)

Notice of Dispute resolution noticeboard discussion

This message is being sent to let you know of a discussion at the Wikipedia:Dispute resolution noticeboard regarding a content dispute discussion you may have participated in. Content disputes can hold up article development and make editing difficult for editors. You are not required to participate, but you are both invited and encouraged to help this dispute come to a resolution. The thread is "Periodic table".The discussion is about the topic Periodic table.

Please join us to help form a consensus. Thank you!

--Double sharp (talk) 08:36, 4 August 2020 (UTC)

Slander

https://en.wikipedia.org/w/index.php?title=Wikipedia%3ADispute_resolution_noticeboard&type=revision&diff=971625510&oldid=971624957

Nice work. Slandering me in a dispute resolution forum. One more for WP:ANI. Sandbh (talk) 08:33, 7 August 2020 (UTC)

In the interest of peace

In the interest of peace, and because indeed after an hour I feel that I went too far again: I have removed my comments at WP:DRN and WT:ELEM. For WT:ELEM, that involved deleting your replies as well, as they don't really make sense if I withdraw my own comments; if you would prefer any other way to handle this, let me know.

That is not to say that I agree with your stances on chemistry or your interpretations of sources. I still most certainly don't. But: all that fighting it does is make me unnecessarily unhappy, and I will soon have no time to do it. WP and chemistry are just not that important in my real life anymore. Moreover: there are already topics I avoid because I have strong opinions and do not want to fight. So, I feel it will be best to treat this as one more from now on.

Therefore, I apologise for my behaviour, even though I still disagree strongly with your ideas, and I will as I said leave the subject of chemical periodicity on WP and simply not comment any further on the subject at WT:ELEM. That way we can part in the most amicable manner possible at this moment, I hope, even as our viewpoints strongly differ.

Because a second RFC will inflame this situation, and will itself take time for me to monitor the progress of: I withdraw the idea and log out. Double sharp (talk) 14:17, 7 August 2020 (UTC)

P.S. Since I apparently forgot about it before logging out; I also removed the sections on Dreigorich's talk page that are not very courteous on my part. Right, back to Wikibreaking unless there is anything else. ;) Double sharp (talk) 06:59, 12 August 2020 (UTC)

Coactive nonmetals and halogen nonmetals​

Thank you for asking my opinion. regarding the term "coactive". I have never encountered use of this term, so I have no opinion on the merits or otherwise on its use.

I will add only this: there are flaws in all PT classifications other the one based on electronic structure. The advantage of the standard classification (s, p, d, La, Ac, noble) is that it provides a framework within which the "normal" trends and deviations from the normal can be discussed.

N.b. I did a search for the term "coactive" in WP. It is used in other disciplines and contexts, so I would vote against its use in chemistry. Petergans (talk) 08:51, 10 August 2020 (UTC) Petergans (talk) 08:51, 10 August 2020 (UTC)

Thank you Peter. Good to hear from you after so long. The standard classification is interesting. Perhaps the p category is a bit unwieldy as it includes metals, “metalloids”, and nonmetals. At least s, d, La, Ac are all metals, and “noble” is all nonmetals, or maybe not depending on what Og turns out to be. I do take your point that the heterogenous nature of the p block could still be discussed within the context of the overall s, p, d, f etc framework.

Speaking personally, use in other disciplines seems not to be so much of a consideration. I have in mind “transition” and “noble”. Thus, “transition radiation”; “noble polyhedron”. There is “reactive” as well: “reactive nonmetals”, “reactive mind”, “reactive programming”. Sandbh (talk) 13:10, 10 August 2020 (UTC)

A question

I noticed a particular emoticon ^_^ in several of your recent talk posts, and was a bit disoriented because I was thinking that our mutual friend Double sharp was the only WP:ELEM participant to regularly use it. Am I correct that (a) DS has been using this emoticon for a long time and (b) you recently started?. Or are my little grey cells starting to misfire? YBG (talk) 06:19, 1 September 2020 (UTC)

You're right on both counts ^_^ --- Sandbh (talk) 07:02, 1 September 2020 (UTC)

The little grey cells, they are magnifique monsiuer! Sandbh (talk) 07:05, 1 September 2020 (UTC)

The little gray cells would no doubt approve, but Captain Hastings would react differently and Miss Lemon would be totally befuddled. At 25% Felicity and 75% Arthur, I shall attempt to overcome my negative reaction and get used to it. YBG (talk) 07:52, 1 September 2020 (UTC)

LOL! Sandbh (talk) 08:04, 1 September 2020 (UTC)

Legend options gallery

When I posted my gallery of legend options, I was focusing on format and nothing else. Although I considered changing the content to something more of my liking, I backed off and chose instead to use your content so as to focus on appearance.

Unfortunately, you interpreted this as me agreeing with your content. Alas, misunderstanding is so easy!

Many thanks for your collaborations and contributions! YBG (talk) 20:43, 26 September 2020 (UTC)

Regarding categories

Out of interest: if you were writing something yourself outside WP about periodicity, how would you draw the table and colour it in?

Mine can naturally be found here. But I'm curious about yours. Double sharp (talk) 14:41, 27 September 2020 (UTC)

@Double sharp: I'll look closer at yours in due course. Between you, YBG, WP:ANI, WP:ELEM, my Group 3 article, the article with Eugen, my pre-halogen article, the current article I'm working on (RATE group names), and RL, I'm stretched, and not complaining.

I feel there are three aspects at play. 1 is the approach to introducing the table. 2 is the table itself. 3 is the contextual explanation.

For #2, I expect I'd colour-organise the table into the four complementary pairs:

Metallic	Nonmetallic
Active metals − Pre-transition metals (inc. Al) − Rare earths − Actinides	Halogen nonmetals
Transition metals	Pre-halogen metals (aka coactive nonmetals)
Noble metals	Noble gases
PTM (aka frontier or poor metals)	Metalloids (aka poor nonmetals)

I'd at least include the labels alkali metals, and alkaline earth metals in order to illustrate the components of the PTM. I expect I'd include your *excellent* IUPAC notes, as way of leading into #3. The annotation next to the colour category box for the rare earths would include a comment that the lanthanides are cerium to lutetium (in this case).

Noble metals would have the same colour as the transition metals, yet be separately marked out in some manner.

For the apparently out-of-place sequence of the noble gases compared to noble metals I'd explain that Nature is just like that. Whatever expectations we may have for "pure" symmetry are unfounded. Examples are all around us. That is the texture of the world etc. This could lead into a discussion as to why the noble metals fall where they are v where the noble gases end up.

This layout is founded on didactic utility namely the use of tradition; analogy; descriptive chemistry; systematics; symmetry; contrast; and aspects of learning theory.

I place some, not overriding, importance in the 7±2 aspiration per YGB. "Simple, but no simpler".

1. An approach to introducing the periodic table

Double sharp, something like this:

Step	Item
1	The linear arrangement of the elements, from 1 to 118—mention of Mendeleev, and Henry Bent's coloured line
2	Wound into a flat spiral per Philip Stewart's Chemical Galaxy—possible mention of Gustav Hinrichs
3	Transformed into a hexagonal equivalent, per Jeff Moran
4	Wrapped into a helix, by way of the AA---possible mention of de Chancourtois
5	The top down view of the AA: see Figure 1.
6	The top down view can be unwrapped into the 32-column Janet form—possible mention of the Madelung Rule
7	The Janet form can be rolled into a Telluric Remix, per Philip Stewart
8	The Telluric Remix can be disassembled into a tetrahedron inside a cube—mention of Jess Tauber and Larry Tsimmerman
9	The tetrahedron inside a cube can be unpacked into the ADOMAH form, per Tsimmerman
10	The ADOMAH form can be rearranged into the conventional 18-column form—possible mention of Deming, and the five contextual options: i. IUPAC; ii. La—per Seaborg; and Lavelle; possible mention of ACS Inorganic Chemistry Division logo, with its split d-block iii. La out of Z order—per Eric; iv. Lu—per Jensen; and Eric; and v. La and Lu—per Mendeleev’s mosaic; Coryell; Silberberg; me.

Figure 1: Top down view of the Alexander Arrangement of the elements

Since periodic tables or systems form a continuum-like series of representations, each of the five options set out at step 10 will have their uses, provided the relevant context is flagged. --- Sandbh (talk) 03:47, 28 September 2020 (UTC)

I'm not sure I have time to really summarise it soon, but a written-out version of how I'd teach a bunch of students about periodicity is also on my pages at User:Double sharp/Teaching periodicity (it came out of the argument on group 3, but it's not combative). It is interesting to see the differences. So you can either read it now if you feel you need to cool down from things, or you can wait for me to try and summarise it. Remember the Pascal quote: "I have made this longer than usual because I have not had time to make it shorter." XD

Hmm, so you start with Janet and ADOMAH. That's very interesting because my approach doesn't even mention ADOMAH at all, and only mentions Janet briefly as an alternative and argues against it. And it's also interesting that you discuss the symmetrical forms with the tetrahedron and cube whereas I don't mention them at all. Maybe you are looking at it via a symmetry break from an ideal? That's quite interesting too because I'm the one who ends up with the more symmetric form despite symmetry having very little to do with how I actually get there. Interestingly I didn't even feel any need to give (i), (iii), and (v) any time, and only just argue against (ii) and for (iv). It's a gentlemen's disagreement. I like knowing what the other side is all about, and probably that interests you too. ^_^ Double sharp (talk) 11:14, 28 September 2020 (UTC)

@Double sharp: Normally I wouldn’t have given the spiral(s) and the 3D forms much of a look. I got to talk to, and argue with, Roy Alexander, Valery Tsimmerman, Jess Tauber, Jeff Moran, and Philip Stewart, and Eric along the way, as part of trying to sort out Group 3. I remember reading something about the pro’s and con’s of 3D forms. I managed to finally decipher Bent‘s book by reading it backwards. After that the relationships between the different forms sort of fell into place, including DIM’s line, Janet, and ADOMAH.

The symmetry aspect wasn’t so much of a factor, because I could accomodate it, and asymmetry, into the sequence. Bent played a huge part since he was such a strong advocate for, effectively, the Janet form yet he was equally as strong an advocate for there being no form better than another: they each have their uses. Nobody that I know realises this. They all focus on his passion for LST. Yet it’s all there in his book. Sandbh (talk) 12:09, 29 September 2020 (UTC)

@Sandbh: It's very interesting to hear about this all from you. Especially because my version is so very different: no 3D forms, no DIM line, no Janet, no ADOMAH, no symmetry. Almost everything you mention on the way is absent from my approach. That's why it's good to ask and hear about other viewpoints, I think. ^_^ I wonder how common proportionately the views "there is an ideal form" and "there is no form better than another" are.

Here's a crack at trying to summarise my version. Hopefully you find this as interesting as I found yours, despite it being very far away in spirit. It's actually difficult to do this because I view groups and blocks as something fundamental based on atomic structure, and so don't start from a line of elements at all, but already from a fixed 2D arrangement. So already from the beginning, there's only one form. So, extremely different. ^_^

Step	Item
1	A conception of the elements as characterised by: (i) the atomic number Z, and; (ii) the electronic structure. (Yes, that means that at very high pressures where electronic structure changes, I would argue we have new elements and need a new periodic table. But for the elements we have at standard conditions I'd still argue for one ideal form.)
2	Relating (i) to the horizontal structure of the table (the periods, going in order of increasing Z), and (ii) to the vertical structure (the groups, uniting elements with homologous valence subshells and occupancies), thus making the 2D table format appear as particularly natural.
3	The special position of the noble gases with stable arrangements, and how other elements at this basic level can have their chemistry related to them. This makes the period break natural. At this point, the form of the table as He-Be + Sc-Y-Lu, with the break in periods from the noble gas to the alkali metal, is fixed; no elements can be moved without violating one of the first three items here.
4	Emphasis that the characteristics (i) and (ii) are something like the genotypes of the elements, and that the road to rationalising their phenotypes from there can be complicated. (Examples: F-Cl-Br-I-At physically differ a lot, C-Si-Ge-Sn-Pb chemically differ a lot.) Emphasis that the table is based on the genotypes, not the phenotypes. "You can choose your friends, but you can't choose your brothers."

The reason why I consider both (i) and (ii) to be necessary to define what an element is is that if we just have atomic number, we have no way to determine the period lengths. In a 2D universe (Flatland), we would equally well have atomic numbers, yet the period lengths would be quite different. The element with Z = 24 becomes a noble gas in Flatland; I would argue that this is not the same element as our chromium. So something would have to be added. The electronic structure (structure of the electronic cloud), being equally a part of atomic structure as the proton number, seems to be appropriately fundamental. It seems more fundamental than all chemical properties because it's inherent in the atom and that's exactly what drives them: the electrons are the bearers of chemical properties.

Now, although it's not strictly part of justifying the table, what I'd then want to do is use the table as a hook to explain chemistry and global trends. So I cover that below. But strictly speaking, the justification of my table is from the above points alone (well, 3 is justified by Klechkovsky and Ostrovsky per 7).

5	A basic covering of chemistry and periodicity, using only the principal quantum number and the 2-8-18-32 shell sequence, for elements hydrogen through argon (Z = 1 through 18). Helium naturally appears in group II because it has two valence electrons. Basic periodic trends such as electronegativity, first ionisation energy, etc. Basic level electron configurations such as "2, 8, 6" for sulfur. Basic concepts also such as the ionic-covalent continuum of bonding, metallic bonding, bonding and structure, oxidation states, stoichiometry, reactivity will be covered here. Fajans' rules, rather than group divides, are given as a rationalisation. (As we are going to argon only here, group divides are counterproductive because the line for cations changes: H+ but not He2+, Be2+ but not B3+, Al3+ but not Si4+.)
6	Rationalisation of why argon ends its period (rather than it going on to eighteen elements) by the introduction of subshells. Upgrade to gas-phase ground-state electron configurations like [Ne] 3s2 3p4 for sulfur, and extension to elements Z = 19 and 20 (potassium and calcium).
7	The Klechkovsky rule + Ostrovsky's statement about the position of the s orbitals (the break in energies being approximately 1s, 2s 2p, 3s 3p, 4s 3d 4p, 5s 4d 5p, 6s 4f 5d 6p, 7s 5f 6d 7p rather than strictly following n+l). Here I'd just mention the derivation exists and of course not go through it. Explain that different orbitals can be close in energy and hybridise strongly. Briefly mention that the energy gap before each s orbital is the justification for why the noble gases appear where they do.
8	Basics of transition metals: Z = 21 through 30 (scandium through zinc). Explain that chemical environment can alter the configuration now and that the exact ones will not be important or tested. Upgrade from gas-phase ground-state configurations to fuzzy configurations like [Ar] (3d 4s 4p)10 for nickel. Basic ideas of transition metal chemistry.
9	Extension to Z = 54 (xenon). General periodicity: definition of blocks, patterns across blocks such as contractions, cutting the blocks into halves, even vs odd period effects, physical metallicity, secondary relationships. Effects of interelectronic repulsion and size applied to d orbitals.
10	Extension to Z = 86 (radon). Effects of interelectronic repulsion and size applied to f orbitals. Basic ideas of lanthanide (defined as La-Yb, for me Lu is a transition metal only) chemistry. Discussion of the -La-Ac vs -Lu-Lr conundrum. The elements are naturally covered with La-Yb / Ac-No as the f block, because of f orbital involvement, for consistency of the general trends covered in 9, and because La and Ac are outliers for the d block. The old Ce-Lu / Th-Lr is covered as something common but viewed as a historical misunderstanding in this paradigm; Sc-Y-La is rationalised as another secondary relationship covered in 9.
11	Relativistic effects on 5d and 6p elements and beyond. Basic ideas of actinide chemistry. Extension to Z = 118 (oganesson), discussing relativistic effects on chemistry as a perturbation of the original structure (explaining why for once elements can use their inner subshells in Fr-Ra and be ionised past the noble gas configuration at 119). Emphasis on genotype vs phenotype when it comes to the bad homologies of Hg to Rn with Cn to Og.
12	Discussion of history and alternative forms of the table and how what they have to say can be subsumed under the umbrella of secondary-relationships. Advocacy of He-Be + Sc-Y-Lu as the optimal periodic table on the grounds that its secondary relationships show the best of all those other forms (B-Al-Sc, Sc-Y-La, He-Ne, H-F, C-Si-Ti, Be-Mg-Zn, etc.).
13	Brief discussion of the current limits: elements beyond 118 and how they may fit into the table, with something about their predicted properties.

Depending on exactly how much time I have, I might not obviously get to do all of that. Double sharp (talk) 12:45, 29 September 2020 (UTC)

@Double sharp: I'll look at this closely. In my dreams I imagine a grand unified version of our two approaches.

Re, "how common proportionately the views "there is an ideal form" and "there is no form better than another" are."

I'd never heard of the view there is an ideal form until I learnt about this notion from Eric. My impression is there are few such proponents. Maybe Gary Katz, and Tsimmerman. Even Eric no longer supports this idea. DIM didn't so much, as far as I can recall. That said, I think he bewailed the lack of sophistication in the mathematics of the day. The 1,000+ periodic systems that Mark Leach shows, each with their own value, casts some doubt on notion of an ideal system. What does "ideal" mean, anyway, in the context? Surely it is subjective, human contingent? Sandbh (talk) 22:54, 29 September 2020 (UTC)

Looking forward to it!

Well, you can add me to the list of proponents of the view that there is an ideal form. ^_^ I think, in hindsight, that I misphrased the second view. There rather seems to be a continuum. Maybe most people don't think there is an ideal form; nonetheless I think most people would agree that some forms are better than others. After all: a form of the periodic table that looked like

H  He Li Be B
C  N  O  F  Ne
Na Mg Al Si P
S  Cl Ar K  Ca
...
Lv Ts Og

might be a bit cute, but it seems to me very weird to say it's on a par with the standard one. The five-groups don't mean anything, all this really seems to be is just a line of elements; to me it's a bit self-evidently worse than the standard form. Something like how we can say, the classification of eight major planets makes sense, the classification that everything round is a planet also makes sense, but we can generally agree that there isn't a sensible classification that gets you the eight major planets plus Pluto. So maybe the other view is more like "some forms are better than others, but there's none that is the best at everything".

I'd say that an ideal form means one that is better than any other one at the important metrics. To me those are being (i) internally consistent across all elements, (ii) based on the most fundamental properties of the elements, and (iii) allows one to rationalise everything else about them from there. That seems in keeping with the generally reductionist bent of science. In DA's words from the DRN I filed and then basically withdrew (my apologies for all that, again), "The periodic table is not a chart to describe elements, but a system where they are naturally organized." But I think this may be contentious. ^_^

Regardless, I think part of the challenge from alternative forms is answered by my 12 above: because many of those alternative forms are about what I consider secondary relationships, I can answer "you don't need the form, you just need to understand the secondary relationships". That is, showing a Sc-Y-Lu table does not necessarily mean that you have to treat the relationship of Sc and Y to La as anathema or nonexistent; it only means that you treat it on a par with something like that of B and Al to Sc. So in some sense, it's trying to create a form that is best at everything by taking all the sensible extra relationships and showing how they can be read from it indirectly. So while I think you'd go for an approach more like "well, in some cases Sc-Y-Lu is more useful and in some cases Sc-Y-La is, so let's show both forms as needed", I'd go rather for "Sc-Y-Lu comes out more fundamentally from my preferred approach, but Sc-Y-La is also useful, so let's see how we can use the Sc-Y-Lu table as well to deduce the Sc-Y-La". Somehow that sits better with me because the trouble with Sc-Y-La alone is that it's difficult to justify to my satisfaction why it should be OK to shift this but not some other stuff; that makes me unhappy because I want total internal consistency of principles across the entire table. But viewed as one of a tapestry of secondary relationships that further enlivens the strict block-based form and can be found by counting columns (all three elements are three columns from the left edge), I have absolutely no objection. Double sharp (talk) 00:06, 30 September 2020 (UTC)

One small thought on the viewpoints

I suspect it's not quite a matter of one ideal PT vs many are OK, but rather something more like a continuum. If we restrict to group 3, say, you could distinguish five points on a sort of continuum of views:

1. Sc-Y-Lu isn't useful in any context, whereas Sc-Y-La is. Calls for Sc-Y-Lu are misguided and should be thrown in the fire unread.
2. Sc-Y-La and Sc-Y-Lu can both be useful in their contexts, but Sc-Y-La is more useful. Calls for general Sc-Y-Lu are misguided, but within their context such tables can be useful.
3. Sc-Y-La and Sc-Y-Lu are about equally useful. There's no reason to abolish general Sc-Y-La, everything is OK in its context.
4. Sc-Y-La and Sc-Y-Lu can both be useful in their contexts, but Sc-Y-Lu is more useful. Calls for general Sc-Y-Lu make sense and should be supported, but Sc-Y-La should be remembered still.
5. Sc-Y-La isn't useful in any context, whereas Sc-Y-Lu is. Calls for general Sc-Y-Lu not only make sense, but should be adopted this instant, with Sc-Y-La being thrown in the historical dustbin and incinerated.

Yes, there is some intentionally comic exaggeration there at the edges, but then again they are meant to be edges. (I guess I am probably at 41⁄3 or so; I'd argue for deprecation of Sc-Y-La but I would still argue that it should be remembered as part of history as a good hook for explaining secondary relationships.) ^_^ So maybe the interesting thing is not the popularity of 1 or 5 vs 3, but the popularity of 2 or 4 vs 3. They're along the lines of:

1. There's many PT's and every single one is equally OK provided the context is provided;
2. There's many PT's, but some are better than others even if we account for the context;
3. There's many PT's, but one is uniquely better than others across all contexts. (That's the "ideal PT" view.)

Personally I suspect the "many PT's are OK" view is closer to (2) than (1) just above: my five-group example above shows why I find (1) rather unbelieveable as written. I suppose that many would argue against (3) by pointing out that showing something must inevitably mean something else is shown worse. But I think that if you consider a table whose general principles can unite both opposing things and show them, (3) looks suddenly less implausible, and that's more or less why my view is closer to (3): one can imagine that such a thing exists. Double sharp (talk) 00:04, 1 October 2020 (UTC)

Precious anniversary

Precious

Six years!

--Gerda Arendt (talk) 06:23, 4 October 2020 (UTC)

E-mail

Yes, I did. I can't imagine why it didn't come through. I can see yours okay. Maybe you should check your preferences and see whether you have the correct e-mail address set up for yourself. It's nothing to worry about, just something I wanted to let you know. Deb (talk) 13:05, 5 October 2020 (UTC)

I can't explain that. Never had a problem before. Deb (talk) 13:18, 5 October 2020 (UTC)

Oops

In special:diff/984604613, you said

There are far fewer sources arguing in favour of Lu simply because La is the accepted standard; effectively no one has been convinced by the Lu arguments; and effectively no one has seen the need to argue for La.

I think you meant

There are far fewer sources arguing in favour of La simply because La is the accepted standard; effectively no one has been convinced by the Lu arguments; and effectively no one has seen the need to argue for La.

When I first read it, I thought there was another place you used the wrong Lx, but I couldn't find it again. Might be worth you re-reading. YBG (talk) 01:34, 21 October 2020 (UTC)

@YBG: That is a nice pick up thank you. I'll go and revisit how I expressed that. Sandbh (talk) 05:18, 21 October 2020 (UTC)

Notification (ANI)

There is currently a discussion at Wikipedia:Administrators' noticeboard/Incidents regarding an issue with which you may have been involved. The thread is Trouble at WP:ELEM, round 3: conduct of User:Sandbh. Thank you. Double sharp (talk) 21:50, 23 October 2020 (UTC)

Your recent post

I saw your recent post and would be happy to respond, just not there, as I think it would be best to avoid threaded discussions in WT:ELEM § Opinions. If you would like to repost or move your question someplace outside that section, I will be glad to respond briefly. BTW, your response seems to indicate that I did a poor job of explaining myself. YBG (talk) 05:49, 25 October 2020 (UTC)

@YBG: VG. Done, here. Sandbh (talk) 07:18, 25 October 2020 (UTC)

Because you thanked me

Sandbh, you thanked me for one of my recent edits, so here is a heart-felt...  YOU'RE WELCOME! It's a pleasure, and I hope you have a lot of fun while you edit this inspiring encyclopedia phenomenon! I dream of horses (Contribs) Please notify me after replying off my talk page. Thank you.

22:34, 26 October 2020 (UTC)I dream of horses (Contribs) Please notify me after replying off my talk page. Thank you. 05:47, 5 October 2020 (UTC)

ANI

Hi. Please stop pinging me from the ANI page.

If and when I want to read your posts, I can find them by myself. Thanks. -DePiep (talk) 02:26, 31 October 2020 (UTC)