Misplaced Pages

Revision as of 05:34, 15 April 2020

Elements Project‑class

This page is supported by WikiProject Elements, which gives a central approach to the chemical elements and their isotopes on Misplaced Pages. Please participate by editing this page, or visit the project page for more details.ElementsWikipedia:WikiProject ElementsTemplate:WikiProject Elementschemical elements Project This page does not require a rating on Misplaced Pages's content assessment scale.

Misplaced Pages:Misplaced Pages Signpost/WikiProject used Misplaced Pages:Misplaced Pages Signpost/WikiProject used

Archives

Archive 1 Archive 2 Archive 3 Archive 4 Archive 5 Archive 6 Archive 7 Archive 8 Archive 9 Archive 10 Archive 11 Archive 12 Archive 13 Archive 14 Archive 15 Archive 16 Archive 17 Archive 18 Archive 19 Archive 20 Archive 21 Archive 22 Archive 23 Archive 24 Archive 25 Archive 26 Archive 27 Archive 28 Archive 29 Archive 30 Archive 31 Archive 32 Archive 33 Archive 34 Archive 35 Archive 36 Archive 37 Archive 38 Archive 39 Archive 40 Archive 41 Archive 42 Archive 43 Archive 44 Archive 45 Archive 46 Archive 47 Archive 48 Archive 49 Archive 50 Archive 51 Archive 52 Archive 53 Archive 54 Archive 55 Archive 56 Archive 57 Archive 58 Archive 59 Archive 60 Archive 61 Archive 62 Archive 63

This page has archives. Sections older than 60 days may be automatically archived by Lowercase sigmabot III when more than 3 sections are present.

This is the talk page for discussing WikiProject Elements and anything related to its purposes and tasks.

Put new text under old text. Click here to start a new topic. New to Misplaced Pages? Welcome! Learn to edit; get help. Assume good faith Be polite and avoid personal attacks Be welcoming to newcomers Seek dispute resolution if needed

Archives: Index, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63Auto-archiving period: 2 months

FA	A	GA	B	C	Start	Stub	FL	List	Category	Disambig	Draft	File	Portal	Project	Redirect	Template	NA	???	Total
29	0	96	103	127	95	34	0	172	307	3	0	117	1	22	8,465	116	11	0	9,698

31

31 featured content items: Featured articles (30) H, He, O, F, Zn, Ge, Y, Nb, Tc, Xe, Cs, Pb, At, Fr, Th, U, Pu, Cf, Db, Hs, Nh, Ts, Og, noble gas, metalloid, periodic table, heavy metals, radiocarbon dating, history of aluminium, island of stability Featured topics (1) period 1 elements Signpost interviews: 2011, 2013 view edit Detailed list Article quality in the periodic table Other high-quality articles

Palladium Good Article Reassessment

Hello everyone, I wanted to let you all know that I have nominated Palladium for a good article reassessment. The associated discussion page is: Talk:Palladium/GA2. Utopes _{(talk / cont)} 21:28, 23 February 2020 (UTC)

Pinging @Double sharp: and @ComplexRational:. This may have slipped under the radar due to the larger discussion happening above, but I thought you might be interested in this reassessment. Utopes _{(talk / cont)} 19:58, 24 February 2020 (UTC)

@Utopes: The article does indeed seem to be suffering a bad case of citation-needed-itis. Unfortunately, rewriting this properly would require time and research that I don't see myself having for a while (the last time I did a full-blown rewrite to get to GA standard was for Ca and Si and that's years ago now, not to mention that I am no longer very happy with how I did Si; since then I've only been slowly doing FA's)...like for V, though, it may not be bad if it loses the green plus. At the very least it may well attract somebody new like you to fix it with Greenwood and Earnshaw! Double sharp (talk) 20:21, 24 February 2020 (UTC)

@Utopes: I second Double sharp's comments. I must note that several sections also seem underdeveloped (in addition to citations needed) and could use a rewrite. With the exception of isotopes, I wouldn't be able to commit much to it at the moment, but I'd be willing to help here and there if anyone else is willing to take charge. If it's not GA now (by the criteria), it's not GA, but that of course doesn't mean it can't become GA again at some point. ComplexRational (talk) 23:27, 24 February 2020 (UTC)

I had similar thoughts when I nominated the article for a Good Article Reassessment. However, because I do feel like I would need to re-write the article as opposed to removing simple tags, I was skeptical of whether this (and Vanadium too for that matter) were truly "Good Articles". They were accepted nearly 9 years ago as Good Articles, however I don't believe they have held up well after all this time, and I don't think I can easily fix the changes. Utopes _{(talk / cont)} 00:32, 25 February 2020 (UTC)

@Double sharp:, @R8R:, @ComplexRational:, I think the main problem we are seeing is that a large range in quality exists among this project's Good Articles. For an extreme example, Vanadium and Hassium are both assessed as Good Articles, however we can agree that the article on Hassium is of a higher caliber than the former article. Now, this much makes sense, as Hassium is being actively worked on to become a FA, and Vanadium is at risk of losing GA status. Yet, at least to me, labeling both of these articles as "Good" seems disingenuous to the project in the sense that the two articles aren't of the same quality. Now, I am aware that it has been established to discontinue the usage of "A-Class" among WP:ELEMENTS articles. However... I believe that it would make sense for the A-Class criteria to be used in certain instances, such as when an article has undergone a peer review or was a part of an FAN that did not succeed. That way it can be clear what the "cream of the crop" is among the WP:ELEM articles. I am aware that many of you are probably fed up with discussion about re-implementing the A-Class criteria, but I'm not asking to change half of the GA articles into A-Class instead. Alternatively, I was thinking about labeling articles as A-Class that have undergone substantial improvement beyond becoming a GA, AND have had high levels of community involvement from a FAN or a peer review. I can only think of a handful of articles that would meet this criteria. Thoughts? Utopes _{(talk / cont)} 00:58, 25 February 2020 (UTC)

@Utopes: This sounds great, I had this idea some time ago myself. However, if my memory serves me right, the problem was that we don’t have enough people to maintain an A-class review (like, say, the Military History project famously does), so it has to be an individual assessment by one user as they see fit, just like we do with, B, C, or Start classes, and there’s the issue of whether it’s okay to have a higher-tier class given out just like that. I’d like to have some criteria that are close to those of a FA, to show this article is not far from the FA status, but those articles are not very numerous. I am not sure whether that addresses your concern, but this can be thought through. But if there is such a desire, I think formal criteria can be written (or borrowed).—R8R (talk) 06:14, 25 February 2020 (UTC)

My worry about reinstating A-class is precisely that we don't really have enough members and activity to justify it. (Matter of fact, I'm not even sure that C-class is necessary.) That's why I think that perhaps it might not be such a bad thing if some of those old GA's get GAR'd (not all at once, obviously). If one of us has time to fix it then, then that's a good outcome! And if not, maybe someone later will fix it because GA seems like a good target. Partly thanks to our old spamming, many of the non-GA elements are extremely daunting important elements and tasks (tin? gold? phosphorus? oh boy), and this may look pretty scary. But if you look in the d block there are a lot of GA's that are really C or B class articles, I'd think, that could definitely use some improvement if you have some time and a source or two handy. And you can always ask R8R or me for tips on finding sources, since we have been doing this for a while even if our activity has taken a back seat to our real lives. ;)

Also: I think R8R really has a point that an all-green table may end up staying that way, because a GA already looks pretty decent if it's a real GA. So perhaps we should do a precedent, showing that GA is not and should not be "the end", by taking an old GA from really long ago (maybe so long ago that none of us worked on it) and taking it to FA. And it should be a really important element that we do it for. Vanadium and palladium would both make excellent choices for this, actually. (When I thought about this first a while ago my thoughts ran to tungsten, but either of these transition metals would work too. I know Utopes wants to do chromium, so this idea may well go somewhere in the near future. ^_^) These TM's are kind of really useful for the average reader. I know, I know, we've done a lot of stuff on superheavies, there is a beautiful human element there, and there are beautifully cool predictions there (you know, not only Hs, but also Cn and Fl, ought to become FA's), but it's true that for almost everybody flerovium does not have even one tenth of a percent of the importance of its heavier homologue lead (which is not a very good homologue anyway). So this would be something good to do.

P.S. Regarding old GA's, I think carbon and thulium need a good deal of work. Double sharp (talk) 14:20, 25 February 2020 (UTC)

I think we may try the following: we may promote Hs to A (I think we can agree it's rather good right now) and then if there are any more near-FA articles, we can promote them, too. GA in itself is not a very high standard to my liking; the article is good for a collaboration of people doing this as a hobby and there's that (then again, there is ambiguity because the standard now is higher then it once was but there is reluctance to demote those articles promoted long ago); I think that aluminium may be an A once it's a GA. Maybe a better delineation would be a good thing given the big gap between old GA and FA. We could generalize this principle and say A is anything that comfortably fits into the current GA standard. Then again, there needs to be anything resembling a consensus for that.--R8R (talk) 14:56, 8 March 2020 (UTC)

Introduction into superheavy elements

Hello everyone,

you may have seen the new Hassium#Introduction section, which serves as an introduction into the basic concepts about synthesis of superheavy elements. In a discussion with Double sharp, both he and I agreed the introduction is best kept brief and a more extensive introduction should be put into the main superheavy element article (see Superheavy_element#Introduction). The idea is that the reader does actually get some understanding before getting to read the whole article while a more extensive discussion is slightly off the topic of the individual element, and should be kept with the general concept. The reader can reader a brief introduction, get interested, and read the full version in the appropriate place; if only a link is provide, the reader is not nearly as likely to click it.

The brief introduction for hassium could be of great use for all superheavy element articles, but the question then arises: how do we keep the brief introduction up to date in all at least fifteen articles? It seems that the appropriate solution would be transcluding that bit into all articles on individual superheavy elements, but from where if superheavy element is to have the longer version? I'd very much like to hear comments from you on this. Possibly someone wants to make a more general comment on the topic of the introduction into superheavy elements in general, in which case I'm all ears.

Pinging @Double sharp, ComplexRational, Sandbh, YBG, Droog Andrey, DePiep, Дрейгорич, and Burzuchius: please see the question and leave a comment. Sorry if I forgot to ping anyone.--R8R (talk) 10:22, 16 March 2020 (UTC)

Agree. Droog Andrey (talk) 10:41, 16 March 2020 (UTC)

Great to hear that. Do you have any idea on where the transcluded section should be stored? I have so far considered transuranium element or leaving it in hassium and transcluding the section from there. I don't like the second option, but it's an option nonetheless.--R8R (talk) 14:05, 16 March 2020 (UTC)

I think Superheavy_element#Introduction is the best choice. Droog Andrey (talk) 15:04, 16 March 2020 (UTC)

That is for sure, that's where the longer proper introduction will be. What I'm looking for is the place for the shorter introduction that is supposed to interest the reader enough so that they read the longer introduction at Superheavy element#Introduction.--R8R (talk) 15:33, 16 March 2020 (UTC)

I feel that the introduction should definitely be focused on the element itself instead of its synthesis. A brief mention might be okay, but really, it should go in the History or Discovery section. For an introduction, I would propose sticking to the known properties of the element - keep it simple for the non-technical readers. ― Дрейгорич / Dreigorich Talk 14:59, 16 March 2020 (UTC)

The aspects of synthesis described should be pertinent to each element. I believe a short introduction like the one in hassium is helpful to give context, but I would not recommend transcluding it (the short section) because not everything relevant to hassium may be relevant to every other SHE, and we don't want any articles straying too far off topic. We need to make (likely minor) adaptations on a case-by-case basis, with the aggregate, detailed description in Superheavy element#Introduction. ComplexRational (talk) 16:04, 16 March 2020 (UTC)

Re @Дрейгорич: you see, synthesis is absolutely crucial to every superheavy element (SHE), much more so than to a stable element. There are two major differences: one is that you can only produce SHEs in ridiculously small quantities compared to the stable elements, which makes understanding how the minuscule amount you get at all so important. The other thing is that SHEs don't live very long, they are created, they last for a short while so that you, if you're lucky, run an experiment or two with it, and then they're gone. And even as for those chemistry experiments, every chemistry experiment starts as a nuclear physics experiment. So that's why creation is such a big part of the whole topic of the superheavy elements, and that's why we need to introduce it. I share your idea that article should be kept as simple as possible---but not simpler than that, or we interfere with the idea of writing an encyclopedia otherwise---and the non-technical readers are aided by a brief explanation of how things work and an incentive to read a larger description in a different article.

Re @ComplexRational: please go ahead and look at the shorter introduction scrupulously. What, if any, modification would it possibly need for any other SHE article?--R8R (talk) 20:05, 16 March 2020 (UTC)

As it stands, any changes would be minor, since the same technique is generally used from Rf onwards and other techniques have not yet yielded any new SHEs, and decay properties are generally consistent (though SF branches aren't significant in the heaviest elements, but even this will likely change as more isotopes are synthesized). I'd say this current version is a pretty solid baseline. ComplexRational (talk) 20:45, 16 March 2020 (UTC)

The point that I'm trying to get to is that there is not going to be any change at all. Nothing in the short (or long) introduction is specific to hassium or any element; it merely explains the basic principles. Can be reused in a different SHE article very easily. I'm certain about that and I welcome you to challenge this assumption if you disagree.--R8R (talk) 20:49, 16 March 2020 (UTC)

Yep. The question is how much is too much? Too little? It should be enough to be understandable that there's not much known about it, but not so much that it clogs up the introduction. Some For some elements with little known, it will be more critical than others. Not much is known about either astatine or francium, but a synthesis section for either one makes little sense - stay to what chemical properties are known. For something like hassium, it would be more critical, as there's really almost nothing to go off of except some guesses. ― Дрейгорич / Dreigorich Talk 20:20, 16 March 2020 (UTC)

I thought you were talking about the lead at first... :-/. Dumb me. The method I'd argue could be something like an introduction and a link to Superheavy element for more info, like Main article: Superheavy element. ― Дрейгорич / Dreigorich Talk 20:23, 16 March 2020 (UTC)

That is precisely what is done at Hassium#Introduction.--R8R (talk) 20:38, 16 March 2020 (UTC)

I think then that would be a great introduction, though I wonder if it's needed on 15-20+ articles. Surely it would get too repetitive after a while? ― Дрейгорич / Dreigorich Talk 22:26, 16 March 2020 (UTC)

@R8R: This is what I was getting at as well. The content is great, but I'm a little hesitant to include verbatim copies of this section in so many articles. In several of my GA reviews (for E124 and E126), this was explicitly not recommended; I had to do a little wordsmithing and tie it back into the article's main topic (but in the case of Hs, this latter point is less necessary). ComplexRational (talk) 01:02, 17 March 2020 (UTC)

Aha, I get it. This is a reasonable concern. I have considered it myself. I don’t think, however, that the proposed solution makes the problem too bad. Consider this: suppose you wanted to read the article on darmstadtium. It does not have that introduction, and you would never guess that you could find one in hassium. The introduction is helpful for hassium and it would be equally helpful for darmstadtium if the article on the latter had it. But it’s not there, and there is no hint there is one at hassium.

This is not to say, however, that I dismiss your problem: the section was originally much longer (see the current long introduction). While too much repetition might not seem very appealing, much of that repetition has already cut out and is kept in one place (in Superheavy element). I think it is a great improvement over keeping the long section in an article on an individual element.

I did once have your line of thinking myself, so I certainly understand the legitimacy of the concern here. However, it is also important that we are not writing a book where we may need the text not to be too repetitive between chapters (or in our case, articles); we write first and foremost an encyclopedia, and it is important that if some useful bit is added in one article that all other articles also get it, hence the need for transclusion. If I were facing such a problem at a GAN, I would say that to the reviewer.—R8R (talk) 12:14, 17 March 2020 (UTC)

I think just a link under "Introduction" is sufficient. Repeating one and the same section in 15 articles seems silly. But if you want to do it, you can make the introduction a Misplaced Pages template. Burzuchius (talk) 14:08, 17 March 2020 (UTC)

I conclude from the above that there is no strong objection to the proposal, and I've been WP:BOLD to advance the introduction to the four present-day FAs: two of them written by myself (tennessine, dubnium) and two by other editors (oganesson, nihonium). I think the introduction looks great in every one of these four. If there are no objections, I'll add it to all elements from 104 onward.--R8R (talk) 22:36, 27 March 2020 (UTC)

@R8R: It looks great to me, so I think there is no problem in putting it in all elements from 104 onwards (maybe even 102 onwards, since No and Lr were discovered by hot fusion with light ions just like Rf, Db, and Sg). Double sharp (talk) 06:57, 28 March 2020 (UTC)

I agree with Дрейгорич / Dreigorich re, "I feel that the introduction should definitely be focused on the element itself instead of its synthesis. A brief mention might be okay, but really, it should go in the History or Discovery section. For an introduction, I would propose sticking to the known properties of the element - keep it simple for the non-technical readers."

I don't support a transclusion into each article, nor a template.

I had a similar problem with the Metalloid FA. I solved it by putting a blurb at the start of the relevant sections and before the text e.g.:

The focus of this section is on the recognised metalloids. Elements less often recognised as metalloids are ordinarily classified as either metals or nonmetals; some of these are included here for comparative purposes.

See here and here.

Just after the start of each introduction you could put a blurb like this:

For an introduction to the basic concepts of the synthesis of superheavy elements see Superheavy element/Short introduction

That's all that's needed. Sandbh (talk) 03:38, 6 April 2020 (UTC)

I have for the time being been bold and put the short introduction into the articles on elements from 104 to 118. I closely considered the idea of not doing that, and I decided to go ahead anyway. The introduction itself is applicable not only to the topic of superheavy elements in general, but also to each individual element. The big notable thing about superheavy elements is that synthesis is of paramount importance for the latter; no topic of a regular element can claim the same with possible exceptions of hydrogen and helium, or uranium and plutonium (and I am still inclined to say not in either case). To quote Christoph Düllmann, a very big name in the area, "Every chemistry experiment starts as a nuclear physics experiment." The topic would not have even been to begin with if it weren't for synthesis and nuclear reactions. This is unlike the case of stable elements, where you could do chemistry of even fluorine long before you could synthesize fluorine gas. This is also why every article on a superheavy elements starts with a section of history of discovery, not one of its properties.

I promise to think some more on the topic, but for now, I am satisfied with this thinking. Whether the same should be done for elements 102 and 103 as well as 119+ is up for deliberation.--R8R (talk) 13:37, 11 April 2020 (UTC)

For 102 and 103, we might have to mention a few different techniques in relation to the history and discovery; they were discovered in light-ion bombardment before cold and hot fusion really were studied. But it might be doable with some small tweaks, and there would be a pretty clear benefit for readers. And for 119+, I'd say for now to use an abridged version at most because they are of course undiscovered, and we don't even know if new techniques will be required to synthesize further isotopes; I'd hold off on those a bit longer, but I'm open to further ideas. ComplexRational (talk) 16:24, 11 April 2020 (UTC)

The former is also true of 104-106, though. Double sharp (talk) 17:06, 11 April 2020 (UTC)

I thought that was precisely hot fusion?--R8R (talk) 17:48, 11 April 2020 (UTC)

@R8R: Well, that will teach me not to write late in my time zone. ;) Yes, it is hot fusion. But my point rather stands; there is not really a difference between how 102 and 103 were synthesised, from how 104, 105, and 106 were synthesised. Double sharp (talk) 03:16, 12 April 2020 (UTC)

@Double sharp and ComplexRational: okay, let's say we want to have that intro there, too. Then the question comes, how do we adapt it? The big difference I see is that it is generally agreed that the set of superheavy elements starts with Z = 104, and the intro starts with the words "A superheavy atomic nucleus is created in a nuclear reaction that combines two other nuclei of unequal size into one," and the transclusion tag also points that it is transcluded from a SHE article, which is potentially confusing.--R8R (talk) 12:04, 12 April 2020 (UTC)

@R8R: Since 102 and 103 usually are not labeled "superheavy elements", transcluding the introduction would not be ideal and violate the "usual" definition given in superheavy element. We certainly could do a manual copy-and-paste of the relevant synthesis bit, perhaps also noting how neutron capture and alpha irradiation were judged to be insufficient and introducing the technique of light-ion bombardment (as 102 and 103 were the first discovered elements pioneering this technique). Some of the history is applicable, but all the description on terminology and decay modes is not (for example, EC is still a major decay mode here, while it is rare or only theorized for heavier elements, and SF plays a larger role here as the island of stability is nowhere near these nuclides). ComplexRational (talk) 16:26, 12 April 2020 (UTC)

The decay mode consideration is a good one, thank you. That didn't cross my mind.

If there is going to be variation of the text, we need to establish over which axes would those variations go (decay modes with inclusion or non-inclusion of the island of stability, anything else?) and how far do we want to go back (do we want to include, say, californium?)

On a brief inspection of the relevant infoboxes, I don't see too much electron capture in the most important isotopes of nobelium and lawrencium. Should we bother in the brief introduction with it at all?--R8R (talk) 18:53, 12 April 2020 (UTC)

Since No and Lr are not usually considered "superheavy", I would propose that the wording be mostly retained, but edited to simply talk about the heaviest elements on the table instead (so it would be a manual copy-paste, adding some text about how the neutron capture up to Fm and the alpha irradiation that worked for Md would come up dry here – it's fairly obvious why, because of the Fm wall and the fact that you can't get a target beyond Es). We can add a mention of electron capture to the decay mode. Double sharp (talk) 13:07, 14 April 2020 (UTC)

Use a template?

• re the "from where ": simple, if we want to do this, a template is the best place. I see no reason to transclude it from, say, hassium. Template is where we editors can keep it stable easily. (I thought is was ill advised at Misplaced Pages (considered bad practice) to transclude body text anyway, but I cannot find that statement anywehere so it must be absent). A more detailed puzzle might be how to make the block of text nicely flow in each article's flow. It might attach subtle strings (requirements re good writing). -DePiep (talk) 07:14, 18 March 2020 (UTC)

  I strongly disagree with putting the introduction into a template. I feel about this very strongly and I will try to explain why. The very concept of Misplaced Pages revolves around the idea that it is easy to edit for anyone. You need no knowledge of anything; just follow the markup and you'll be good. This was further improved when visual editing came to being. Now, of course, there are some limitations. It is advisable to conceal some complicated templates from the public, so that the markup is not stored in the article text and does not confuse anyone, and the thing is easier to edit (including those who understand well how the markup works). Some topics are too tempting for people to ruin them (like Donald Trump, Muhammad, Jews, whatever it is that gets people to lose their minds), so some limitations are in place. Here, however, we are talking about uncontroversial encyclopedic content, rather than a supplement to that. It would be a very bad idea to make it difficult to edit and this would contradict the very principle of Misplaced Pages without giving anything in return (like protection from vandalism) that could compensate for that. It is already bad enough that it needs to be stored in one place for consistency, but at least clear guidance where to go to can be given (see the first section of NATO bombing of Yugoslavia for an example).

  I generally see why you wouldn't want to transclude that from hassium; that's where the question about where the section would actually be stored comes from. I'd be eager to listen to suggestions on that as long as it's not a template.

  To be perfectly clear, templates are fine but they are not used to store encyclopedic content (as opposed to anything that supplements it) in them.--R8R (talk) 21:26, 18 March 2020 (UTC)

You could still make a template and subst: it instead, or even introduce it from hassium to the other articles using subst:. That way, no encyclopedic content is stored remotely (each article/section can still individually be edited), and major changes can be easily transferred with a semi-automated run of 15 subst:s. ComplexRational (talk) 21:45, 18 March 2020 (UTC)

User:ComplexRational: subst: is the oppposite ofwhat is proposed here. It actually fixes the transcluded text into today's version; later edits (to the source text, think the Hassium section), those edits are not appearing in other articles. -DePiep (talk) 07:05, 19 March 2020 (UTC)

I am well aware of that and how subst: works. I only proposed it because this thread seems intent on establishing a standard introduction, while I stand by my point of keeping it open to minor adjustments in each article–so there is no need for centralized editing, room for changes specific to each element if necessary, and reviewers or scrutineers do not get annoyed with the repetition of text in 15 articles. Subst: would be a one-off establishment with these provisions. But of course, if transclusion is preferred, I will not object to a template, subpage, or whatever the preferred practice is for article text. ComplexRational (talk) 19:23, 19 March 2020 (UTC)

• Last buttons first: "templates are fine but they are not used to store encyclopedic content" (source? I could not find this): Yes they are, for exactly the reason you raise in this question: repetition of information (here).
• In short, once you want to transclude text, template cannot be discarded beforehand.
• Now on the ease of editing. My solution is, while putting the text block in a template: in top, add a link like This section:
  • view
  • edit
  . In Misplaced Pages, this is commonly accepted (also for encyclopedic text too).

-DePiep (talk) 07:29, 19 March 2020 (UTC)

Technical implementation

I've been looking on how to implement this technically (transclude the section from Hassium#Introduction).

• Route One. Basically: when not a template, add the namespace to the page name in the {{...}} notation (when page is not in the Template: namespace). So write in the target page: {{User:DePiep/sandbox}} Articles just get a colon {{:Argentina}}

Then, in the source article (Hassium) add these tags to select parts to be transcluded:

Lede of hassium text here <onlyinclude>==Intoducton== A superheavy ... images, refs, &tc. ... have been made.</onlyinclude> Rest of Hassium text here

This is examplified in Misplaced Pages:Transclusion#Pages_with_a_common_section, third example: article sectionJoseph Gordon-Levitt#HitRecord imports the lede of article HitRecord.

• Other Route. As the very same section in WP:Transclusion says in its intro: "When two pages need to discuss the same material in the same way, they can share a section. This involves creating a third page and transcluding that page onto both pages". I note it says "page", in an unspecified/any namespace; I take "discussion" more widely, to include article body text.

So this third page could be a template. Adding a well-crafted v-t-e links would solve editability.

Importantly, I claim that this is way more easy to grasp for editors than the "<-- To edit this section in ], go to ] section -->"; plus the extra presciptive comment in that hassium section. We all know how bad such instructions are read or understood, if at all.

• My conclusion. While writing this, I came to think of this technical solution, in case we do want that central section:

The Introduction section content is in page Superheavy element/Short introduction. Part of mainspace, and quite a logic title without the why-in-Hassium?-distraction. Of course it has some v-t-e box.

To me, this is equally sound as using a template. I reject the in-hassium-article construction. -DePiep (talk) 08:08, 19 March 2020 (UTC)

I did not realize a subpage was a feasible solution. We could do that; seems like a great to me since that's allowed. Thank you for bringing that up.

Some readers who could edit a main space page would be afraid to edit a template page. As far as we can stay in the main space, that's good.--R8R (talk) 10:19, 19 March 2020 (UTC)

Trivial note: actually, in mainspace this is not considered a "subpage"; it is considered a stand-alone article (example: AC/DC). Probably, such a page could have <onlyinclude> tags to allow extra info (or even make formally into a full stand-alone article). Some wiki-purists might frown upon this, but we may have good arguments to do so. -DePiep (talk) 10:59, 19 March 2020 (UTC)

• Currently: an editor has moved the article content into template Template:Superheavy element introduction(edit talk links history) . Workings is the same. I see no need or gain in fighting this choice, but that's just me.

On a sidenote, for inspiration: an introductionary article, esp into difficult topics like SHE or QM, is well received. In March I was involved in "getting some, any, corona-related FA article speedily as TFA". Guess what: on March 27, the article Introduction(!) to viruses was at MainPage and got >100k hits (3x24h). Also, reading the article made me into a virologist, which is nice to be these days. Next stop for me: QM-specialist. -DePiep (talk) 00:39, 2 April 2020 (UTC)

• FFS, R8R, stop messing around. The Template is Great. -DePiep (talk) 19:06, 4 April 2020 (UTC)

  Please strike this. And why is the template great? ComplexRational (talk) 21:20, 4 April 2020 (UTC)

• OK then The table is great, by aim & setup. -DePiep (talk) 21:32, 4 April 2020 (UTC)

I'm rather surprised that a clarification is needed, but since it is, I'm happy to explain my reasoning.

First of all, the rules consider having the common section as a separate page as a fine possibility. To quote Misplaced Pages:Transclusion,

When two pages need to discuss the same material in the same way, they can share a section. This involves creating a third page and transcluding that page onto both pages. This third page may be a page in its own right or a subpage of either of the other two, and if the first it may be placed in the same namespace as the other pages or in template namespace.

What I'm opting for is a third page in the same namespace as the other pages. That's one of the possibilities outlined in the relevant technical page. It is, I note, not a guideline, so if you can find a guideline that says otherwise, maybe you can correct me. However, I want a guideline before making such a statement. "R8R, ffs what are you doing, template is fine" is no explanation. I can't respond to it; there is nothing to say to, "that's good reasoning, I stand corrected" or "that's not good reasoning, here's why." There is no reasoning at all.

Second, this guideline has been brought up already, in this very section. When I moved the section where I did, it didn't cause any controversy back then. Where this is coming from is genuinely puzzling to me, given that I'm merely moving it to the original location. I did announce where it would go; nobody protested.

Third, I find it very important so I will reiterate: editing Misplaced Pages should be as easy for an editor as possible. A small "edit" link does not explain why the usual "edit" link in the header of the section does not work; it is more puzzling. The template I added explains that the text is stored elsewhere. I hope that you will undo this edit of yours.

Fourth, I have already said this too, but I will be happy to reiterate: the text contains encyclopedic text as opposed to anything that compliments it, and encyclopedic text should be stored in the main space. The very fact there is that green space below (template documentation) was unsettling to me when I was learning Misplaced Pages and the markup. Only later did I learn how to work with it. This is not what most readers should be able to do, they should not feel disenfranchised by ending up in a "technical" zone. To add to that, Misplaced Pages:Template namespace, which is a guideline, says, "Templates should not normally be used to store article text, as this makes it more difficult to edit the content. They should also not be used to "collapse" or "hide" content from the reader." We are clearly talking about article text here; why should we not store it in the main space?

Fifth, by moving the page back to its original title, I undid the moving, that's when discussion begins. I actually tried to start it even earlier, here, but my call for discussion was not returned. This point is merely for making clear what the actual order of events was and what is the "revert" in the BRD scheme.--R8R (talk) 22:57, 4 April 2020 (UTC)

Edits

Not about the concept of centralising this text. But about (future) edits, as in regular article improvements.

• As it stands in Hassium#Introduction now, it needs adjustments into the article text flow. For example, a section title "Introduction" can only pertain to "Hassium". Instead, the section better be like "Introduction of superheavy elements". -DePiep (talk) 15:07, 29 March 2020 (UTC)

What do you mean by article text flow? Is this similar to what I described above about slight differences and adjustments for each SHE? I'm not opposed to a section title change if that's all this is about, though. ComplexRational (talk) 16:04, 29 March 2020 (UTC)

Yes, it is the same. I'm not sure if your proposal (tailored adjustments per element) can be achieved. When simple, we could start using parameters... But varying sentences? Re this, I am not too familiar with the topic to propose content edits. Anyway, my "article text flow" here is about connections between sections. At first reading, I saw the same isolation in section Hassium#Cold_fusion (no hassium-specific text in there either). -DePiep (talk) 16:14, 29 March 2020 (UTC)

Shorter: think what is best in the TOC: "Introduction" or "Introduction of superheavy elements" (or "The creation of SH elements")? -DePiep (talk) 16:17, 29 March 2020 (UTC)

Alright, that makes sense. Given the nature of the content, I'm inclined to think "introduction to superheavy elements" or similar is more appropriate so readers are not misled into believing that each element has an "introduction". Finer content changes (varying sentences) don't seem necessary at the moment; we'll discuss the technical nature of that if and when it is necessary (e.g. I think when 119 and 120 are discovered, and even more so beyond that, a few small changes might be in order). ComplexRational (talk) 16:23, 29 March 2020 (UTC)

"List of chemical elements" redesign

I am working on an overhaul of the table in List of chemical elements#List. If you are interested, please join the talk here. -DePiep (talk) 15:12, 29 March 2020 (UTC)

Group 3 StateOfMatter please

Introduction

This page is nearing the 1M bitsize, the Group 3 discussion is running for some 3 months now.

It would be useful if the discussion could be condensed but not frozen into an intermediate StateOfMatter overview: those points, arguments, results that should be kept and reused. Gaseous parts can be let go. When we do nothing, all will be lost and no consequences can be implemented.

Ideas? Suggestions? For starters, probably we need an ~outsider, at least not involved contributors ;-). -DePiep (talk) 17:23, 2 April 2020 (UTC)

@Double sharp, Sandbh, R8R, Droog Andrey, and Dreigorich: pinging the most active participants. Are there any sections (particularly from the top, with the last comments in January or early February) that can be closed and archived relatively soon? Since I haven't commented much (and haven't been swayed one way or the other, nor would I be particularly willing to get involved in a heated debate if I shared my preexisting opinions on the matter), I could try to write short summaries of those sections, but I'd like to be sure those threads are closed first and perhaps input on which parts can be omitted entirely from any such summaries.

+1 to DePiep that the page is uncomfortably long to navigate and that the Group 3 discussion completely dwarfs anything else on this talk page. The ~2.8 kB I archived earlier was barely anything by comparison, so more needs to go soon, and again I'd rather not demolish something if it's still being built. ComplexRational (talk) 21:27, 2 April 2020 (UTC)

(CR, I can handle a long Talk ;-). There are two 400-day talk links in top. My first concern is the reasoning & arguing we want to keep. Producing sound, learnful and usable conclusions). -DePiep (talk) 21:59, 2 April 2020 (UTC)

yep. Maybe start a new ==level section, TOC first? Gives opportunity to redesign the setup, support the reasoning lines. (ie, TOC from scratch & heliview, or drone as it is called today). Worth making all this productive.

Oh and why not invite an outsider from further away, say WP:PHYSICS, WP:CHEM, WP:MATH—they are smart!, and even "All PT is physics" will boil down to ... maths, WP:MILHIST. -DePiep (talk) 21:51, 2 April 2020 (UTC)

I've archived the whole thing into Archive 42. Double sharp (talk) 10:17, 7 April 2020 (UTC)

Main lines table

Well, here is a summary of what seem to be the main lines of argumentation so far:

I have decided to terminate my participation in the main thread, as I will lack the time for it from tomorrow and there is no point in repeating myself more than I already have been doing. I will only fill in responses in my column 4 here, if Sandbh adds anything else to column 3. Double sharp (talk) 06:41, 5 April 2020 (UTC)

@Double sharp and DePiep: That's good. My column 3 is complete. Over to you, subject to your time commitments. An excellent summary, from which I'll attempt to provide DePiep with an immediate overview. Sandbh (talk) 08:20, 5 April 2020 (UTC)

@Sandbh: I've completed the table with my column 4 (somewhat in a rush, to get it done before other RL concerns won't allow it). So I think we've summarised everything. Double sharp (talk) 08:25, 5 April 2020 (UTC)

Immediate overview

Background (18 bullets)

1. Group 3 has been Sc-Y-La-Ac since about the 1920s
2. A few chemists in the 1920's and 1930's assigned Lu rather than La to group 3
3. In the 1940s it was realised that placing Lu in the f-block was ostensibly wrong since the 4f sub-shell was completed at Yb, rather than at Lu, as previously thought
4. Since this discovery did not change anything of the chemistry of Lu, nothing happened; Lu stayed where it was the end of the f-block
5. DS contends that this is itself somewhat problematic, because once the premises are falsified the conclusion has to at least be derived by alternative means. In other words, reasons based on electron configuration for the placement of La under Y can no longer be admitted, and this should have resulted in the composition of group 3 being relooked at.

6. From about the 1950s onwards a few physicists (mainly) argued that some properties of Lu suggested it'd be better placed under Y instead of La
7. These arguments were mostly based on a single property, and did not attract significant attention
8. Jensen (1983) published a rather longer article arguing for Lu in group 3, and reiterating some of the past arguments
9. Not much notice was taken of Jensen, although he caused some non-critical excitement among a few chemists, and some more papers supporting Sc-Y-Lu-Lr appeared
10. DS and I (2017) comprehensively "dismissed" Jensen's article in our submission to IUPAC, noting many of his arguments were unbalanced

11. DS (2020) later retracted his dismissal, once he learnt more in 2018 from Droog Andrey about the issue. Previously, we dismissed as flawed Jensen's arguments that group 3 only showed trends similar to group 4+ if Sc-Y-Lu was chosen, because a Sc-Y-La group 3 has trends similar to groups 1 and 2, which we thought were chemically closer to group 3: but DS considers now that actually Sc and Y are chemically intermediate between group 2 and 4, and so that this should not be a strong argument after all, and that electronic bases are sounder than chemical bases for reasons that he will consider later in this summary.
12. Scerri and Parsons (2018) assessed Jensen as being too selective in his arguments
13. Scerri (chair of the IUPAC project looking at this issue) has said several times that the group 3 question can't be resolved by comparisons of the properties of La and Lu
14. DS contends that this is simply due to the fact that the road from electron configurations to chemistry is complicated. Without considering electronic structure, the group II question (whether Be and Mg should go over Ca or over Zn) that previously was an issue before WWII would be equally unresolvable. And there would also be questions about B-Al-Sc: see footnote 1. Once chemically relevant electron configurations across all chemically plausible environments are considered, we recover the ideal electron configurations of the Madelung rule the presence of valence f involvement for La and Ac and its absence for Lu and Lr is decisive for DS.
15. As it is for Jensen (2009), see footnote 2

16. The IUPAC project team surveyed nearly 200 chemistry texts (a survey to which Sandbh contributed) and established a 4:1:1 distribution between Sc-Y-La-Ac, Sc-Y-Lu-Lr, and Sc-Y-*-** (the last option not being considered as a future possibility by the current IUPAC project team)
17. DS contends the IUPAC project team are likely to have misinterpreted these chemistry texts. A table with the asterisks * and ** for "following elements" in the same cells as La and Ac, part of a group 3 column (as shown to the right), is literally speaking a Sc-Y-*-** table in terms of what elements it claims to be in group 3, not a Sc-Y-La-Ac table. Under this interpretation, the lead of Sc-Y-La-Ac over Sc-Y-Lu-Lr is weakened.
18. DS also notes the majority of sources whose focus is the group 3 issue support Sc-Y-Lu-Lr, and this statistic is more meaningful wrt the dispute than the literature that is not focusing on this. The idea of d orbital hybridisation as an explanation of hypervalence for 3p elements is alive and well in many texts, but obviously the literature we should reflect is that in which it was debunked by specialists (quantum chemists) back in the 1990s. This case serves as evidence that the textbook literature may well be behind the latest scientific knowledge, and further weakens the significance of the survey of the IUPAC project team.

__________
Regarding group II, Jensen wrote in 2003: "From a chemical point of view, Zn and Cd most resemble Be and Mg, not only in terms of their atomic radii, ionic radii, and electronegativities (Table 4), but also in terms of the structures of their binary compounds and in their ability to form complex ions with a wide variety of oxygen and nitrogen donor ligands (including complex hydrates and amines). Indeed, prior to the introduction of electronic periodic tables, the similarity between Be and Mg and Zn and Cd was often considered to be greater than the similarity between Be and Mg and the rest of the alkaline earth metals (Ca–Ra). Many inorganic texts written before the Second World War placed their discussion of the chemistry of Be and Mg in the chapter dealing with the Zn subgroup rather than in the chapter dealing with the Ca subgroup, and the same is true of many older periodic tables, including those originally proposed by Mendeleev (34, 35)." Rayner-Canham (2011) also quotes Greenwood and Earnshaw (1997) in their contention "that magnesium is atypical of group 2 (though beryllium is even more so)".

Regarding group III, Rayner-Canham (2012) wrote: "In this pair of groups, Greenwood and Earnshaw (1997) have discussed the way in which aluminum can be considered as belonging to Group 3 as much as to Group 13, particularly in its physical properties. The Canadian geochemist, Habashi, has suggested that there are so many similarities between aluminum and scandium that aluminum’s place in the Periodic Table should actually be shifted to Group 3 (Habashi 2010). In terms of the electron configuration of the tripositive ions, one would indeed expect that Al (electron configuration, ) would resemble Sc (electron configuration, ) more than Ga (electron configuration, 3d). In terms of their comparative solution behaviour, scandium(III) resembles both aluminum(III) and gallium(III). For each ion, the free hydrated cation exists only in acidic solution. On addition of hydroxide ion to the respective cation, the hydroxides are produced as gelatinous precipitates. Each of the hydroxides re-dissolve in excess base to give an anionic hydroxo-complex, M(OH)₄. There does seem to be a triangular relationship between these three elements. However, scandium does more closely resemble aluminum rather than gallium in its chemistry. If hydrogen sulfide is bubbled through a solution of the respective cation, scandium ion gives a precipitate of scandium hydroxide, while aluminum ion gives a precipitate of aluminum hydroxide. By contrast, gallium ion gives a precipitate of gallium(III) sulfide. Also, scandium and aluminum both form carbides, while gallium does not."

DS notes that this suggests an analogy to why Sc-Y-La has often passed without comment: Y and La both have noble gas configurations, which grants them some second-order similarities, whereas Lu does not. The fact, however, that this argument is not used to make group 3 start B-Al-Sc, or group 4 go C-Si-Ti-Zr-Ce-Th, suggests to him that more important considerations are at play for element placement.

__________
Jensen wrote: "...it is inconsistent for Lavelle to dismiss (ref 1, Note 4) the (n – 1)d ns valence configuration of Th as an inconvenient irregularity that should not affect its assignment to the f-block as an idealized (n – 2)f ns element and then turn around and insist that it is absolutely verboten to entertain the idea that La and Ac may have similar irregular valence configurations corresponding to an idealized (n – 2)f ns valence configuration. After all both elements have low-lying empty f orbitals, which is more than can be said for Lu and Lr. Indeed, more than a quarter of the elements in the d- and f-blocks have irregular valence configurations and in several instances these irregularities apply to the majority of the elements within a given group. The simple fact is that the periodic table is based on idealized electronic configurations rather than on actual configurations and in this fashion functions in chemistry much as the ideal gas law or the concepts of ideal crystals and ideal solutions." Although Lavelle replied to Jensen's article, this point was left unaddressed.

Current state of affairs (39 bullets)

Philosophical

1. Sandbh considers a difference between DS and him to be that DS likes to drill down into details whereas Sandbh focuses on the broad contours of each situation
2. DS disputes this difference. To DS, both him and Sandbh consider a broad-strokes generalisation with the minimal complexity needed for a good framework. DS considers Sandbh's approach to have too little complexity, because DS contends that it fails to work when generalised to the whole table.
3. Sandbh contends that local patterns and generalisations remain valid and useful

Oxidation states

4. Frex, Sandbh focuses on the most common oxidation states of the elements e.g. Zn = +2, Yb = +3
5. Sandbh calls this the simplest sufficient complexity that derives useful information
6. DS focuses on e.g. Zn vs. Yb
7. Here +2 for Yb is not the most common oxidation state; it decomposes water, and thus only Yb occurs in aqueous solution.
8. DS notes that by this logic one cannot find the proper start of the 3d contraction either: the most common oxidation states of Sc, Ti, and V, according to Wulfsberg, are Sc, Ti, and V, which have no d electrons to cause any contraction! The situation for 4d and 5d is even worse with the group oxidation state being the main one till Tc and Re! +2 is the only reasonable shared state to apply to transition elements (d, f, and g block) as it corresponds to removing the outer s electrons; the only other reasonable choice is neutral atoms, which makes sense for main group elements (s and p blocks) too.

Lanthanide contraction

9. We differ on what is meant by the lanthanide contraction (LC), which is caused by the progressive occupancy of the 4f sub-shell
10. I (Sandbh) take the LC to span the trivalent cations of Ce to Lu i.e. Ce 4f to Lu 4f, which accords with Goldschmidt who discovered the LC in 1925
11. DS takes the LC to run along the neutral Ln atoms, to accord with the way all other contractions (e.g. scandide) are defined. (Otherwise, the scandide contraction cannot start at Sc because its major oxidation state is Sc 3d.) In this case La may have 4f occupancy in chemical environments, and the contraction ends at Yb 4f6s which is the last element that uses the 4f electrons for chemistry. Notably, Greenwood & Earnshaw (G&E) include La in the LC.
12. G&E instead say La is "included for completeness" (p. 1232) i.e. not because it is a part of the LC—since La has no contraction-causing 4f valence electron—but rather to have something to compare the ionic radius of Ce to
13. DS notes the idea that La is not strictly speaking a Ln is basically obsolete now and doesn't make chemical sense given how similar it is to Ce-Lu, so it should be included in the chemical Ln contraction as well as the electronic one

14. Sandbh says there are broadly comparable contractions in each period of the PT, including the actinide contraction and the scandide contraction
15. DS notes that those contractions do not have the chemical significance of the Ln contraction, only the electronic significance; they are more important for their impact on the subsequent elements, precisely because of the lack of a common oxidation state. (Appealing to a common +3 state for the actinide contraction requires considering Th, Pa, and U that decompose water; for other contractions it only gets worse.) They are not comparable to the chemical Ln contraction that spreads across La-Lu.

Patterns

16. I (Sandbh) say patterns seen only in parts of the periodic table e.g. the well-known diagonal relationships between Li-Mg, Be-Al, and B-Si, are valid in considering periodic trends and relationships
17. Rayner-Canham, a UK professor of chemistry, has written extensively on such patterns and a book of his on the same topic is due to be published on 25 August 2020
18. Rayner-Canham, in his Inorganic Chemist's Periodic Table (which notably is Sc-Y-Lu), only colours in the patterns when they appear, and is totally comfortable with the idea that these patterns are only local, calling isodiagonality just "one of the valid linkages among the chemical elements". He recognises many different patterns and understands that for different elements, different ones will be more important. Sandbh agrees.
19. DS says patterns that exist only partially "cannot be admitted as foundational principles for the PT". He notes that such patterns are drawn on the PT, including by Rayner-Canham, only after the basic structure has already been erected by some other means, that by definition of periodicity must apply to all elements.

The popular form: Sc-Y-La-Ac

20. I (Sandbh) say La in group 3 is backed up by 100 years of chemistry practice, and that most new ideas are wrong
21. To some degree, resisting new ideas is healthy because most new ideas are wrong and established concepts are often backed up by lots of experiments and data. Scientists had a right to be skeptical, at first, of the ideas of Darwin and Mitchell. There were phenomena neither could explain and both of them got parts of the story wrong. As a result, it took about 17 years for Mitchell’s ideas and more than a half-century for Darwin’s ideas to become accepted.
22. DS notes that Lu in group 3 has history dating to the 1920s, too, and most chemists who seriously considered the group 3 question have supported it. It is not a new idea, and there is a lot of data supporting it.
23. DS says the fact the La form is the most popular does not make it right, citing how the myth of d orbital involvement in hypervalent main group compounds seems to be refusing to die even though it was refuted in the 1990s. At some point, the weight of evidence for a new idea becomes strong enough that the old idea has to be rejected, e.g. the end of phlogiston.

4f involvement in Lu

24. Sandbh says Lu has 4f bond strengthening involvement (p. 8); see footnote 3
25. Thus, excluding the 14 4f electrons of Lu makes the bond length too long i.e. the 4f electrons contribute to bond shortening or strengthening. The 5sp sub-shell also plays a (stronger) role here.
26. DS considers this to be a misunderstanding. In computational chemistry, many inner orbitals must be included to get quantitatively correct answers, but it doesn't mean they have significant contribution to the bonding. The fact that 4f here has an even weaker effect than the core 5s and 5p orbitals (part of the xenon core!) strongly suggests that 4f valence involvement in Lu is, for practical purposes, zero. Lu has no significant f involvement (not even to cohesively strengthen bonds like d in Zn). See footnote 4.

Differentiating electrons

27. Sandbh says the d/e concept is easily understood and has been used consistently by various authors since Ebel (AFAIK) first introduced the term in 1938
28. DS says they should be ignored as they cannot be defined consistently for cases like V 3d4s vs Cr 3d4s, and are chemically irrelevant as within the energies of chemical bonds, many different electron configurations can exist, and depending on the chemical environment a chemically bound atom may exhibit any one of them. He also notes that the process involved in defining DE's, "add a proton and an electron to get the next element", is more the domain of nuclear physics than chemistry with the huge energies involved that will surely cleave all chemical bonds.

A continuum of properties

29. DS says the properties of the elements form a continuum and it isn't possible to draw dividing lines therein that will be relevant in all circumstances.
30. I (Sandbh) say the properties of the elements form a semi-continuum and that is possible to draw useful and informative dividing lines, as chemists have done since the time of Hinrichs (1869), including those that lie along group boundaries
31. DS does not dispute that dividing lines may be useful and informative, having drawn some of them himself with the caveats that depending on the context it may be valuable to draw them elsewhere. Sandbh agrees.
32. DS says that group divides, besides group 18 vs. group 1, are never very useful or informative in general (see footnote 5), and contends that generalisations of Fajans' rules (i.e. acidity increases as electronegativity and charge increase and atomic radius decreases) are more fruitful as a broad-strokes approach. When group divides appear, they are simply a result of a localised coincidence that is easily destroyed by expanding the boundaries to all significant chemistry (e.g. the group divide between group 3 and 4 disappears completely once we leave periods 4–6, or stop insisting on comparing +3 compounds to +4 compounds).
33. Such generalisations of Fajans' rules appear in the toolbox of chemists (e.g. Wulfsberg's Principles of Descriptive Inorganic Chemistry); group divides are at least significantly rarer due to their locality. DS thus refers back to his contention that the basic layout of the periodic table must be obtained through means that generalise to all elements.

Delayed start of filling of the 4f sub-shell.

34. Sandbh says the delayed start is widely if not universally recognised in chemistry and that e.g. it results in the double periodicity in the Ln, as first observed by Klemm (1929) and confirmed by the well-regarded Russian chemist Shchukarev (1974); the same double periodicity, to a weaker extent, is seen in the actinides
35. DS says this is not relevant to chemistry, just as that of 5f is not (no one doubts that Th, with zero 5f electrons in the ground state, is an f element). The delayed start of filling of 4f and 5f still allows La, Ac, and Th to display involvement of that subshell as a valence subshell for chemical purposes, resulting in La metal having the second-highest 4f involvement of all the lanthanides (Gschneidner): paradoxically, because the collapse is swift, elements before the start of filling are likely to show bigger involvement of that subshell than the ones after!
36. DS notes that a consistent reflexion of delayed collapses would anyway result in a "staggered" f block starting at Ce and Pa for its first and second rows, not a Sc-Y-La table; and it would also lead to an unsustainable situation once we open the 8th row, where calculations say 8p is first occupied at E121, followed by 7d at E122, followed by 6f at E123 or E124, and finally 5g only at E125!

Double periodicity.

37. DS notes that double periodicity of the Ln and An, from clear analogies of the stability of the +2 oxidation state to the 3d elements, supports Sc-Y-Lu. He quotes standard electrode potentials to support his point (see footnote 6).
38. Sandbh notes the special status of the half-filled and filled valence sub-shells in the cations, at Mn and Zn; and analogously at Gd and Lu.
39. DS contends that Sandbh's previous point is not supported by the data. Mn and Zn are local maxima, as expected: Mn and Zn reach a half- or fully-filled 3d subshell, so by high-school chemistry it must be more difficult to have electrons break free of those. Fe adds an electron on top of a half-filled 3d subshell, and the trend as expected goes down there. We expect the half- and fully-filled subshells in the f rows to also form maxima for the same reason, which appear at Eu/Am and Yb/No. By Ga, Lu, and Lr, the drowning of 3d/4f/5f into the core has gotten to the point that the new easily removed third electron must be coming from somewhere else.

__________
"Freezing the outer closed shells of Lu causes large errors in the bond lengths (too long) and bond energies (too small). The freezing of the outer 5s5p Xe noble-gas semicore shell is more serious than the freezing of the inner 4f semivalence shell." (So apparently 5s and 5p show bigger involvement than 4f!)

__________
Droog Andrey, a computational chemist, offered as an example that for Ni complexes, only 1s, 2s, and 2p could be ignored as core electrons without significant consequences: in other words, at least the outermost core orbitals must be included, but that doesn't make them have significant valence involvement. A similar situation happens for the inclusion of d orbitals for hypervalent compounds, see Errol G. Lewars' Modeling Marvels (p. 59): "Including d functions in a basis set for calculations on hypercoordinate compounds may improve the accuracy of the results (this can easily be tested by comparison with known molecules), but this does not mean that physical d orbitals (whatever that may mean) are involved: the orbitals may merely be acting as polarization functions, skewing the s and p orbitals in more propitious directions".

__________
Take basicity of oxidation states for an example: witness how the maximum basic oxidation state moves from +1 in period 2, to +2 in period 3, to +3 in periods 4 through 6, to +5 in period 7, rather than having a group divide. One can see similar trends for example in the structures and degree of hydrolysis of chlorides in water, in which group divides are also absent. The appearance of a "group divide" in the specific situation when only d-block group oxidation states are considered stems from a coincidence in the d block only: the increased size going from 3d to 4d is partially counteracted by the fact that the EN doesn't go down very much (in fact it sometimes goes up), and there is no increased size going from 4d to 5d because of the Ln contraction (this added to an increase in electronegativity). Therefore, generalised Fajans' rules (acidity increases as electronegativity and charge increase and atomic radius decreases) predict that the group divide from 3d to 6d will move really slowly, as it in fact does: Sc is amphoteric, its heavier congeners are basic; but the first basic group 4 cation is Rf. Once we get the EN drop and size increase back, we can go up to fantastic heights: Th and Pa are basic cations, as their EN is lower and their size is greater. And once the charge stops increasing, the group divide disappears: compare Sc, Ti, V, and we can see there isn't a significant difference (all three form chlorides that dissolve in water rather than hydrolyse, whereas the chlorides of Ti and V hydrolyse). And once we bring the size down to a minimum, we see already that BeCl₂ and BCl₃ are hydrolysing in water, even coming before the +3 vs +4 line, and the only reason CCl₄ is spared that fate is steric hindrance.

So let's apply the general theory to what happens in the early period 5 chlorides (EN_KK(Cl) = 3.06):

Rb (ENKK = 0.77)	Sr (ENKK = 1.05)	Y (ENKK = 1.28)	Zr (ENKK = 1.35)	Nb (ENKK = 1.44)	Mo (ENKK = 1.53)
RbCl	SrCl2	YCl3	ZrCl4	NbCl5	MoCl6
3D ionic lattice, regular	3D ionic lattice, deformed rutile structure (undistorted rutile structure pictured, as I can't find a picture for the SrCl2 structure)	layered 2D polymers like AlCl3	linear 1D polymers	dimeric molecule	monomeric molecule
			ZrCl3	NbCl4	MoCl5
			3D polymers	1D polymers	dimeric molecule
					MoCl4
					1D polymers (diagram)
					MoCl3
					layered 2D polymers like AlCl3
					MoCl2
					metal cluster structure

Closing statements

Limited to 400 words

Sandbh

I'm comfortable with my arguments for Sc-Y-La. They're well-supported by the literature.

Re Double sharp’s arguments, I have reservations about their presumptiveness and complexity. On several occasions my arguments were said to be a based on a misunderstanding or misinterpretation by the cited authors; history needing to be revisited; or were countered by introducing novel interpretations not supported by the literature. The complexity of Double sharp’s arguments speaks for itself.

There’s no silver bullet solution to the group 3 question. So we have to argue it out using qualitative or quantitative arguments.

Rather than dwelling upon the minutiae of the individual properties of La and Lu, I attempted to take more of a helicopter view. That meant examining the group in the context of its surrounds; the congruity of the f-block; patterns seen elsewhere in the periodic table; the periodic law; and global considerations. Along the way I entertained a few more detailed (ancillary) arguments where I felt these were required to provide context, were novel, or provide useful insights.

An interesting argument Double sharp raised with me, which I didn’t address, was that Lu below Y would result in a more homogenous d-block. That may be so. Yet the periodic table isn’t so much about striving for homogeneity as it is about periodic trends. And it’s these trends that result in the diversity we see among, for example, the transition metals, and the p-block with its mixture of metals, non-metals, including metalloids, halogens, and noble gases.

The funny thing about Lu under Y is that it looks nice—more symmetric or regular if you will—but it disrupts some cool chemistry-based patterns (as seen elsewhere in the periodic table) and introduces further irregularities of its own. Double sharp contends these patterns are too local or are that the irregularities are insignificant. I argue they are part of the rich texture of the periodic table, and that we here to draw and wonder at the periodic table as it is, not how we think it should appear

From a Platonic symmetry perspective and perhaps that of physics, and on some grounds of regularity but not on others, it can be argued that Lu is better placed under Y. But not from a chemistry perspective, or at least not as well.

My unqualified respect for Double sharp the person, and as a valued peer, stands.

--- Sandbh (talk) 04:13, 8 April 2020 (UTC)

Double sharp

No Sc-Y-La argument presented fulfills what a theory of chemistry underlying the periodic table must have: consistent applicability to every element.

Complexity is required to holistically understand why chemistry alone cannot solve the group 3 issue.

Shall we observe ScCl₃ vs TiCl₄, call it a fundamental group 3-4 divide, ignoring its disappearance for ScCl₃ vs TiCl₃, shifting past the +4 state in the 5f row, and before the +2 state for periods 2 and 3?

Exclude La 4f from the Ln contraction, claiming predominant oxidation states, but let the scandide contraction start at Sc 3d?

Criticise water-reducing Yb, and equally use Th, Pa, and U for the actinide contraction?

Insist half-filled subshells on the E trend give peaks in the d rows, but troughs in the f rows?

No! An end to "one argument here, one argument there, never the twain shall meet". Foundational periodicity must know no boundaries. Shall we be denied following Mendeleev, projecting into the unknown, where no boundaries are charted? Shall we holistically seek working generalisations, or argument barrages at each other's throats within milliseconds without artificial leashes?

Ever-reliable electronic structure, that solved formerly unsolvable Be-Mg-Zn and B-Al-Sc, truly is the "helicopter view": logically deriving continuous trends from theory, without artificial divides. The PT's rich structure is only derivable from holistic criteria. Locality is the next level!

We draw a first-order generalisation, not take one secondary relationship out of many strong ones, to honour as it never deserved!

Most textbooks still explain hypervalence with d-orbitals, though it was debunked decades ago. Why should we follow them if the literature knows better?

Shall we endlessly copy mistakes of a century ago, when electron filling wasn't understood, and 4f was a one-off interruption of the d block? Emphasise the ground-state delayed start of 4f/5f forever, when it means nothing for chemistry, their ends still happen at Yb/No, blinding ourselves to the reproachful counterexamples of thorium and the looming 8th row?

Or shall we progress and heed luminaries: Seaborg, Jensen, Schwarz, Wulfsberg, and Jørgensen, who long ago exposed the hollowness of ground-state DEs? Shall we deny the crown of useful generalisations to ideal configurations, given willingly to ideal gases, crystals, and solutions?

Respecting Sandbh, I support literature and logic, not debunked, self-contradicting tradition. Double sharp (talk) 07:43, 8 April 2020 (UTC)

What next?

• Double sharp and Sandbh to conclude edits on immediate summary
• Ask DePiep what he makes of it
• Ask other members of our project what they make of the immediate summary
• Hope for IUPAC project team report and recommendation this side of post-COVID-19 Xmas
• WP:RFC by Double sharp in July?

Sandbh (talk) 12:46, 6 April 2020 (UTC)

@Sandbh: Seems like a good plan to me; yes, I plan an RFC in July. (I am taking a little bit of extra time only to finish this summary; after that, I think my work on convincing people from the literature should be done, and we will see what people think while I take my wikibreak.) That way, it should be possible for everyone to comment on this shrunken overview, instead of the original doorstopper. Double sharp (talk) 12:50, 6 April 2020 (UTC)

@Double sharp: Nice. Sandbh (talk) 12:53, 6 April 2020 (UTC)

The more I look at this, the more I think we should also ask for (separately, to avoid clouding the issue) group 12 to be coloured as TMs again, as direct d involvement stops mostly after group 12 instead of group 11 (doi:10.1016/0368-2048(75)80018-1; although there is a drop from group 11 to group 12, it is only at group 13 that d involvement becomes purely incomplete screening effects). The Zn group is not that far removed from the Cu group, chemically. But not in this RFC. Double sharp (talk) 05:42, 7 April 2020 (UTC)

Commentary

I still don't see a single La argument with a leg to stand on. Especially the point about sources who really consider the question mostly plumping for Lu tells me that an RFC is the right way to go here to correct the group 3 situation. (We can correct the group 2 situation and move helium there once we get the critical mass: right now I think we are in the equivalent situation as Jensen writing in 1982, with the advocacy of Grochala, Grandinetti, and the late Henry Bent.) Points 15 and 20 (in the right-hand column) are why I think an RFC is justified. Double sharp (talk) 16:47, 3 April 2020 (UTC)

As for how much of this is still active: mostly just the bottom of it at this point, from "Falsifiability" onwards, I guess. (I've just archived another big chunk into archive 42, which is now 1.3M...) Since Sandbh has recently stated that he won't address old ground, and the old ground is the key to our being in disagreement, I am not sure if there is anything new left that we need to finish addressing. I am also not sure how much inviting outsiders will help, because frankly the material required to tackle this subject is rather specialised, and unless your focus is rare earth chemistry (so these elements are the relevant ones, and even then it may not work so well because intraperiod resemblances must also be brought in) or the foundational aspects of general inorganic chemistry (i.e. regularities across the whole table, not just one region of it), it probably doesn't matter much to you. If we bring in an outsider, it should be one whose expertise is on one of these chemical branches.

So probably in the future an RFC should start, except that by mid-April I shall be a lot busier and therefore maybe the great big 2nd RFC to fix group 3 will have to be delayed until July. So perhaps what we should do is:

1. Tie up whatever loose ends we see fit to tie up from this thread within the next few days (that depends on what Sandbh wants to go over or address):
2. Archive the rest of it (I think, next week, since my reserve of free time for this should end very soon), so that we all have some time-out for passions about this to not flare so much:
3. And I'll be back later to start the giant RFC when my free time returns.

I still would like to thank Sandbh for the discussion, since it has definitely sharpened and greatly improved my approach to the periodic table. I still disagree with him, of course. ^_^ And I would also like to thank Droog Andrey for initially showing me the more advanced chemistry required to understand why the La table must be rejected. And also R8R, Dreigorich, and all other present and past participants in the group 3 debate, ranging all the way back those we cite who published about this in the last century. Double sharp (talk) 02:23, 3 April 2020 (UTC)

@DePiep, ComplexRational, Sandbh, R8R, Droog Andrey, and Дрейгорич: Pinging everybody. Double sharp (talk) 05:14, 3 April 2020 (UTC)

Responses etc

Heck, just reading this discussion improved my understanding of the theoretical structure of the periodic table, and how it ties in to the practical decisions that need to be made regarding chemical families. I talked a lot with Double sharp on my own talk page about the structure of the periodic table. I also agree that this has gone on for quite some time now. Thanks to everyone who attended this amazing not-quite-a-TED-Talk-but-it-might-as-well-have-been-one kind of thing. ― Дрейгорич / Dreigorich Talk 07:42, 3 April 2020 (UTC)

@Дрейгорич: Thank you!

There have been a few last posts to the discussion by both of us since your comment, but I think I've just written the last one I'll have time for in the near future. So, you know where to go if you want a group 3 archive binge (my sincerest apologies for linking to TV Tropes, which will certainly cause another one ^_^). ;) Double sharp (talk) 04:52, 5 April 2020 (UTC)

I’m a few responses behind recent posts. I’ll get to those shortly, and check for o/s threads. I don’t mind revisiting old ground as long as we don’t recycle it. I’ll see if I can extract the nub of these issues and address them that way or incorporate them into DS’ table. It’s good to be able to load our page without cacking my iPad. Sandbh (talk) 10:34, 3 April 2020 (UTC)

DS's table looks useful, then I strongly suggest we do not contunue arguing in this thread (so better ignore prose that starts I still don't see a single argument that .... This thread is *not* to settle an outcome. It is to give overview. -DePiep (talk) 11:46, 3 April 2020 (UTC)

OK, struck and replaced by an overview statement that points 15 and 20 (my column) are why I suggest an RFC to happen. Double sharp (talk) 16:47, 3 April 2020 (UTC)

@DePiep: Forgot to ping you, sorry. Double sharp (talk) 16:47, 3 April 2020 (UTC)

Thx. rereading me: sounding harsh, while trying to be clear & short... -DePiep (talk) 16:50, 3 April 2020 (UTC)

• The #Main lines table, a great initiative and setup, is descending into a battle. Whatever the cost, I will not allow a 'shortened' repeat of the Grand discussion. Must I explain? Well here it goes: the #Main lines table shall not be a battlefield. Leave it to others. @Double sharp and Sandbh: -DePiep (talk) 18:06, 4 April 2020 (UTC)
  • @DePiep: So far as I can see, there is nothing in the main lines table that has not been raised and discussed in the main thread already, so it is still a good summary. It just looks like a heated argument replay because it is summarising what already was a heated argument by quoting each side's arguments. Double sharp (talk) 02:34, 5 April 2020 (UTC)

Response to immediate summary

@DePiep, ComplexRational, R8R, Droog Andrey, Дрейгорич, YBG, 1.02 editor, Kpgjhpjm, UtopianPoyzin, and Paintspot:

Double sharp and I have provided an overview of the group 3 situation in a Background section (18 numbered bullets), a Current state of affairs section (39 numbered bullets), and closing statements (400 words each).

Your thoughts, if any, welcome. Sandbh (talk) 03:05, 9 April 2020 (UTC)

Can we have a TL;DR of the TL;DR? ― Дрейгорич / Dreigorich Talk 03:52, 9 April 2020 (UTC)

@Дрейгорич: Unfortunately, I start to doubt that it's possible to cut any further... Double sharp (talk) 04:05, 9 April 2020 (UTC)

TLDR of the TLDR

Double sharp

Although against my better judgement, here is a try:

• Sandbh thinks the La table is backed up by a century of practice and notes, consistent with historical experience, that most new ideas are wrong; but I note that it has been challenged since almost the very beginning of that practice by authors supporting the Lu table.
• I note most authors who seriously consider the situation end up supporting Lu; that the whole point of science is the ability to say "we were wrong" when new evidence comes in; and that textbooks are usually behind the latest literature by quite a while (just look at how many resources still give the d orbital explanation for 3p element hypervalence, refuted in the 1990s).
• Sandbh thinks I'm being too complicated. I think that being less complicated won't fit the facts.
• Sandbh thinks I'm choosing unrepresentative oxidation states like Yb. I say that by referring to the scandide and actinide contractions, he's implicitly doing so because there's no oxidation state that will be representative for all of Sc-Zn or all of Ac-Lr.
  • Clarification: Reading this, one may get the impression that I don't use comparable standards. In fact, for the scandide contraction, we can use +2 since that is the only state common to all of them (aside from +1), and the most stable for six of ten of them. For the An, +3 is common to all of them and is the most stable state for 8 of 14 of them. Sandbh (talk) 01:35, 10 April 2020 (UTC)
    • Well, just look at 4d then. There's no such oxidation state to pick. Double sharp (talk) 03:13, 10 April 2020 (UTC)
      • ​Hey, you were the one to add the 4d metals to the table! Sure, there is no common oxidation state, yet +2 is the only one common to all of them (aside from +1). Like you said, the data that could be gathered for the 4d metals has gaps at Zr and Pd, but there is enough to see analogous periodicity is still present, although weaker (as usual) for later rows. I even somewhat support what you said re once we get rid of the s electrons we can at least see where the half-filled and filled states are. Sandbh (talk) 05:40, 10 April 2020 (UTC)
        • @Sandbh: Ah, but my whole point of picking +2 has nothing to do with common oxidation states. It has everything to do with getting rid of the s electrons only. That's why even for the lanthanides I insist on +2 and not +3; you want to get rid of the s electrons only to see where the half-filled and filled states are, like you agree in your last sentence. It's the only measure that is both fair and gets us to the point of what is going on. And I trust that my having added 4d is not a problem, since I'm sure you agree the 4d contraction runs Y through Cd, and we had better be able to use the same tools consistently to discover that. Double sharp (talk) 05:42, 10 April 2020 (UTC)
          • @Double sharp: In the Ln you need to get rid of ds. 4d is awkward since effectively no-one talks about a 4d contraction occurring along the 4d metals per se, since they have no common, most stable oxidation state. Instead, the contraction is considered in terms of its knock-on consequences, from Cd or In onwards. Whereas with the Ln they have a common most stable oxidation state of +3 and you get to see the contraction as it occurs along the Ln trivalent cations, from Ce to Ln, as well as its knock-on impact starting with Hf(IV). Sandbh (talk) 06:01, 10 April 2020 (UTC)
            • @Sandbh: You're almost getting it. ;) If you apply the logic "they have no common, most stable oxidation state", you will see immediately that the same is true for 3d and 5f. You can pluck out +2 and +3 that are the most stable for just over half of them only, but anyone can see that there is significant opposition for which it doesn't work (it's literally 6-4 and 8-6 respectively). So that confirms what I've been telling you: there is nothing comparable to the Ln contraction, because that's the only contraction for which both the direct and the knock-on effects are important.
            • Now, for ds: this assumes already that one d electron is hanging up and filling early. We both know that's not so, just look at the electron configurations. No, please don't tell me to look at the condensed-phase ones, because then we also have p occupancy filling early. And in chemical environments, configurations switch around, what matters is the number of electrons you can get and whether you can force them into a half-filled or fully-filled situation plus the s orbital. Then we can easily see that whole idea self-contradicts stability considerations, because a supposed half-filled fds should be unstable because of that singly filled d orbital above. So we should be looking at fds becoming stable fds. So by this logic, the stable ones should be not half-filled and fully-filled subshells f and f, but ideal f and f that can rearrange themselves into a more stable arrangement. Isn't this at the very least strange? Eu and Yb show all the signs of half- and fully-filledness in trends (they match the positions of Mn and Zn), but a Sc-Y-La table denies them that position and puts them as almost-half and almost-full shells which now get the special treatment. Well, by that logic we may also start each period in group 2 and end it at group 1. Double sharp (talk) 06:10, 10 April 2020 (UTC)

​This is a level of complexity that isn't warranted or required. P occupancy is noise, compared to the main game. fds, as seen in metallic Gd, is very stable going by Klemm's crude analogy of Gd to a noble gas. There is no such thing as fds. There is instead fs in metallic Eu. Or, as you know, in it's most chemically stable +3 state, f That's why trivalent Gd f gets the special treatment, not Eu. Sandbh (talk) 07:06, 10 April 2020 (UTC)

@Sandbh: p occupancy proves that condensed-phase band occupancy is too complicated to really use. You have to consider instead the number of electrons available. Metallic Gd has three electrons mostly delocalised, so the important thing is that you have basically a sea of Gd 4f ions with three delocalised electrons per atom. For Eu you have a sea of Eu 4f ions. We can clearly see that fds isn't stable at all by itself until you get rid of that d electron. Whereas fds just needs the s electrons to be take care of. It's therefore absolutely clear that Eu fds is the one with the half-filled configuration just like Mn ds here: they just have an outer s shell, and a half filled inner shell that resists attempts to take more electrons out of it (though it is possible).

@Double sharp: Yes, agree on metallic Gd. For metallic Eu it's predominantly Eu 4f with some Eu 4f which is weird but gives a hint as to what's going on. In actual chemistry, Eu is the most stable cation with f, quite unlike the most stable divalent form of Mn, which is d. Here the analog to Mn is trivalent Gd as d. Sandbh (talk) 10:21, 13 April 2020 (UTC)

@Sandbh: You keep focusing on most stable oxidation states. That is not the point. In actual chemistry, Tc and Re are not the most stable oxidation states, but that is not the point either. As we know because we have no qualms asserting that Tc and Re are the elements at the halfway point of the block. The point is just getting rid of the outer s subshell and creating equal terms for everybody. Because the s shell is the only covering shell that persists throughout, and we want the +2 state to ionise it away. Double sharp (talk) 10:24, 13 April 2020 (UTC)

BTW, you say there is no fds (true), but if you consider ds to be removed from each lanthanide, then you are implying that that and fds are the ideal configurations of Eu and Yb respectively. Never mind that they, on trends, show exact analogies to Mn and Zn in 3d, not Cr and Cu. Double sharp (talk) 07:29, 10 April 2020 (UTC)

@Double sharp: I'm not implying anything. The most stable, chemistry-based configurations for Eu and Yb are trivalent f and f, as you know. Sandbh (talk) 10:21, 13 April 2020 (UTC)

@Sandbh: As above. Double sharp (talk) 10:24, 13 April 2020 (UTC)

@Double sharp: Comparing ions with s ionised away is one thing. We may observe that the ionic radius of divalent Ti is 0.86 and that of Zn is 0.74 and that therefore, on this valid basis, there is a 14% contraction from Ti to Zn. If you are going to persist with invalidly comparing the less stable divalent Eu and Yb states with the most stable divalent Mn, rather than using the most stable trivalent forms of Eu and Yb, we have nothing further to talk about, on this point. Sandbh (talk) 12:00, 13 April 2020 (UTC)

@Sandbh: The absolute stability of the +2 oxidation state is not the relevant issue here. Otherwise we will never be able to find the 4d and 5d contractions, and as I said below, you must erect the PT's structure through a means that encompasses every element. It is simply used because it corresponds to the removal of the s electrons and exposes only the filling of the characteristic subshell. That's why I am only comparing relative stability of +3 vs. +2, knowing that reaching a half- or fully-filled subshell ought to stabilise +2 beyond what would linearly be expected. As I wrote four days ago: "As reiterated above, the point of using +2 everywhere is not because it's a common oxidation state (because it isn't), but because it corresponds to getting rid of the outer s subshell. So then you can look in isolation at the filling subshell and the stability of the half-filled one. If you want +3, that implies you have an outer 5d6s all the time. Which obviously doesn't happen. ;)" (Nope, not even in the condensed phase, because of p occupancy.) Double sharp (talk) 13:03, 13 April 2020 (UTC)

Double sharp (talk) 07:45, 10 April 2020 (UTC)

Let's focus on the chemically relevant properties.

Looking at the standard reduction potentials for the 3d and 4f metals, divalent Mn wants to hold on to its d configuration, as trivalent Gd wants to hold on to its f configuration:

For the 4f metals, the most stable oxidation state is +3. Ostensibly +2 is the benchmark for the 3d metals since that is the only state common to all of them (aside from +1), and the most stable for six of ten of them. As you note, the data for the 4d metals has gaps at Zr and Pd, but there is enough to see double periodicity is still present although weaker, as usual, for later rows. For the 5f metals the +3 state is common to all of them and the most stable for 8 of 14.

It seems to me that had the 4f sub-shell started to fill at La rather than Ce, then it would have been Eu rather than Gd with the trivalent f configuration. For the 4f metals, I take the spikes at Gd and Lu to be a chemical consequence of the delayed start of filling of the 4f sub-shell. The further delayed start of filling of the 5f shell does not appear to have any additional chemical consequences.

The parallels between the d- and f-block metals are quite pleasing. Sandbh (talk) 07:52, 13 April 2020 (UTC)

@Sandbh: You only get the parallel by artificially shifting your boundaries. To quote Droog Andrey again: "you compare diactions with trications, so that's not a surprise you obtain analogies like Ca-Zn vs. La-Lu and Sr-Cd vs. Ac-Lr. So that's just an old "+3 state" argument. My point is that we should use equal terms for such a comparison." It must be the +2 state for everybody, that's the only sensible one because outer ns are chemically speaking always ionised. It's not about whether that is a common oxidation state or not, it's purely about removing those s electrons and focusing on the characteristic subshell filling. Likewise, for the s and p blocks I would go for the 0 oxidation state, even if most elements don't want to be in it as free elements, simply because there is no covering subshell to get rid of there. It's a matter of electronic structure only. (And, besides, if 4f started in La, what you would expect is for Eu to be 4f as it would be seven f electrons above = 6s. Well, it actually is.)

Just look at the peaks and troughs of your chart. You say the parallels from the d block to the f block are quite pleasing. Well, they are: Mn/Tc and Zn/Cd show up as obvious peaks in the d block that end each half-series. And similarly Eu/Am and Yb/No show up as such in the f block. And yet you go and highlight the following elements instead and destroy the parallel. Well, let me paraphrase something from the old group 3 submission that I withdraw my support from: while I agree with the spirit of finding parallels, by that measure your support for Sc-Y-La is totally misguided, and the argument very strongly and elegantly supports Sc-Y-Lu. As is supported by absolutely every trend, when plotted, suggesting double periodicity as La-Eu and Gd-Yb families. Double sharp (talk) 10:17, 13 April 2020 (UTC)

@Double sharp: Headslap ^_^. There is no "artificial" shifting of boundaries. This is the doing of Nature, and the delayed start of filling of the f sub-shells. We should indeed use equal terms for the comparison just like all the authors I cited do. +2 is the most common stable state among the 3d metals. +3 is the uniformly mostly stable state among the Ln. Using a less stable state for Eu and Yb is not a level playing field. Next you will be telling me that alkali metals are nonmetals given they can form –1 anions, never mind their most common oxidation state is +1. OK, I'm exaggerating but the analogy is there. Many things become possible when we start referring to less stable oxidation states but we don't use these to make generalisations of the kind you are advancing.

Your bold assertion that, "as is supported by absolutely every trend, when plotted, suggesting double periodicity as La-Eu and Gd-Yb families" reminds me of Flat Earthers who explain a round Earth the same way they explain anything else that contradicts their narrative. They simply ignore or discredit or reinterpret any evidence of a round Earth. I'm not having a go at you, and I'm not annoyed. I find it quite funny that this happens to nearly everything I present, regardless of how many sources I provide. Praise be I still have my sense of humour. I hope you do to as we strive for common ground. Sandbh (talk) 10:44, 13 April 2020 (UTC)

For the 9001st time, there is no such thing as a delayed start of filling of the f subshells. It only comes about because of a double standard, in which the ability for thorium to show f involvement despite its 5f6d7s ground-state gas-phase configuration is accepted to put it in the f block, and the ability for lanthanum to show it with 4f5d6s and actinium to show it with 5f6d7s is vociferously denied. Just like Jensen noted.

So, let's be consistent: we are talking about electronic filling, and stability of oxidation states is only a relative thing here that we do in order to look for the stabilising effect of a half-filled subshell. It does not matter one bit to me that +2 is not the most common stable state among the 4d metals: I use it anyway because it creates consistency and corresponds to the totally sensible removal of the s subshells. Common oxidation states are irrelevant for this problem precisely because they get us absolutely nowhere with 4d and 5d and we have to pick consistent criteria: you must erect the PT's structure through a means that encompasses every element. So we are just comparing the relative stability of the +3 and the +2 states, knowing that a half or fully filled subshell will increase the stability of the +2 state. The fact that Mn vs. Eu and Zn vs. Yb fits so well as an analogy makes it perverse to argue that instead we should have ds hanging up instead, creating a totally unprecedented split block and destroying the nice pattern in favour of a shifted one. And the final nail in the coffin for that idea comes from plotting every single trend. Look at melting points. Look at boiling points. Look at densities. Look at phase transition temperatures. Look at standard reduction potentials. Look at electronegativity. There is no case at all where Eu and Yb match the positions of Cr and Cu, they always match those of Mn and Zn. That's all the confirmation you could ever ask for that the analogy of Sc-Zn to Ce-Lu is totally incorrect, and that the right analogy is Sc-Zn to La-Yb.

Let me quote Droog Andrey yet again: "You build a list of La arguments, a list of references, a list of authors who are wrong. But could you make a little analysis of what you have built?" Double sharp (talk) 13:03, 13 April 2020 (UTC)

You know, it is excellent that Droog Andrey has already addressed all of this, because it saves me time from drafting new responses that I need for RL:

Double sharp (talk) 07:53, 10 April 2020 (UTC)

@Double sharp: I've addressed Droog Andrey's arguments before. He isn't comparing like with like. The dips he is referring to as weird are an outcome of the delayed start of filling of the f sub-shells. This is seen in the gas phase, the metallic phase, and the ionic phase. The argument about starting every period with a group 2 element and ending it with a group 1 element is baseless, since their is no delayed start anywhere else (not yet, anyway). The trends are indeed about real compounds. The boiling points of the Ln have nothing significant to do with their chemistry, in the context of our huge thread. Sandbh (talk) 10:58, 13 April 2020 (UTC)

As I just wrote above: "For the 9001st time, there is no such thing as a delayed start of filling of the f subshells. It only comes about because of a double standard, in which the ability for thorium to show f involvement despite its 5f6d7s ground-state gas-phase configuration is accepted to put it in the f block, and the ability for lanthanum to show it with 4f5d6s and actinium to show it with 5f6d7s is vociferously denied. Just like Jensen noted." So that whole argument collapses like a house of cards. There are only two consistent options here: either there is no delay (the chemically sound approach), and there's no reason for such strange dips; or there is a delay (the formal and irrelevant approach), in which case we note that the 4f row starts at Ce, but the 5f row starts at Pa. Well, the 5f increased delay explains the weird dips at antepenultimate members of the families Pa-Bk and Cf-Rf; everything is equally clear, isn't it?

And as I just wrote below: "I wonder why Greenwood & Earnshaw discuss melting points then, since their book is called Chemistry of the Elements? An explanation is, of course, not far off: what controls melting points is just the strength of bonds or intermolecular forces, which goes back to the electronic structure that underpins chemistry as well. There's not really a difference between how you explain melting and boiling points and how you explain chemical reactivity on this basis: they are both reflexions of the atomic structure and both are universally considered as periodic trends." That's exactly why bringing in physical properties is also important. Double sharp (talk) 13:07, 13 April 2020 (UTC)

@Double sharp: I'm puzzled as to why, on 9,001 occasions ^_^, you ignore the delayed start of filling phenomenon. It's widely discussed in chemistry and physics texbooks. It can be seen after La in period 6 and after Th in period 7. What is the issue with strange dips? There are plenty of other strange dips for Ln properties. Likewise, sometimes physical properties correlate with chemical properties, sometimes they don't. We're here to draw Nature as it is, not how we believe it should be. Sandbh (talk) 05:10, 14 April 2020 (UTC)

@Sandbh: Two reasons:

1. It has precisely zero impact on real chemistry. The end of 4f and 5f involvement as valence orbitals still happens at Yb and No, as expected if there was no delay. The elements that show up on the trends similarly to Mn and Zn in 3d are always Eu/Am and Yb/No, never Gd/Cm and Lu/Lr, as expected if there was no delay. La, Ac, and Th all can easily use low-lying 4f and 5f orbitals for hybridisation, as expected if there was no delay. And, actually, that last point is precisely why the delayed start of filling phenomenon is totally irrelevant. Because as has been widely discussed here already, backed up by the literature, the important thing is not which of many chemically relevant configurations lucks out into becoming the ground state in a gaseous atom by itself, but which ones are low enough in energy to be chemically relevant, and when you look at that there is no delayed start of filling phenomenon at all.
2. And even if it was important enough to reflect on the periodic table, it would not at all support a Sc-Y-La-Ac table. It would support a table where group 3 was Sc-Y-La-Ac and group 4 was Ti-Zr-Hf-Th. The first 5f electron appears in the ground state at Pa, so consistently reflecting the delayed start of filling (as irrelevant as it is) would mean that the f block has to be inserted between Th and Db. After all, Th has incumbency over Rf for the ds configuration.

So arguments for Sc-Y-La-Ac and nothing else based on delayed start of filling are not only focusing on an irrelevancy, they are even misguided. A real "delayed start" periodic table would have to look like

H                                                        He
Li Be                                     B  C  N  O  F  Ne
Na Mg                                     Al Si P  S  Cl Ar
K  Ca Sc    Ti    V  Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr
Rb Sr Y     Zr    Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I  Xe
Cs Ba La *  Hf    Ta W  Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn
Fr Ra Ac    Th ** Db Sg Bh Hs Mt Ds Rg Cn Nh Fl Mc Lv Ts Og

         *  Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
            ** Pa U  Np Pu Am Cm Bk Cf Es Fm Md No Lr Rf

which is evidently not the traditional Sc-Y-La-Ac layout. Well, it is certainly less symmetrical, and the f block no longer starts in a vertical column. But, as you say, "We're here to draw Nature as it is, not how we believe it should be." So why are you not consistently using this argument to support this table, since Nature gives us thorium with no 5f electrons, and you appear to believe that the delayed start of filling phenomenon is actually important? There's only two consistent options here: either argue with delayed starts and go for the table above, or drop them and go for something else. (But of course, dropping them for the good reason that they're chemically irrelevant immediately knocks out a lot of the arguments for Sc-Y-La.) Double sharp (talk) 06:46, 14 April 2020 (UTC)

• I say the only constant oxidation states you can sensibly pick are 0 and +2, not +3, because you have to use identical standards for at least the d and f elements (transition elements). 0 is obvious; +2 at least ideally means removing the outer s electrons (that are almost always removed anyway chemically). Nothing else makes much chemical sense (otherwise, we can equally well say we have to draw the period break between groups 1 and 2, just looking at 2nd ionisation energies everywhere).
• Sandbh thinks the Ln contraction should be measured from the filling of Ce to Lu as the 4f electrons fill from 4f to 4f.
  • Clarification: That is a misrepresentation of what I say. The contraction starts with Ce 4f and finishes at Lu 4f, having passed through Yb 4f. Of course, to measure the contraction in Ce, you have to compare it to La, which has no 4f contraction. Sandbh (talk) 01:35, 10 April 2020 (UTC)
    • My rebuttal still stands. Measuring contractions anywhere else requires bringing in weird oxidation states. You say below that you're relaxing it to oxidation states that at least the majority of the elements have as the most stable to allow +2 for 3d – very well, we still cannot find the start of the 4d or 5d contraction. That's why we need to use +2 everywhere because even if it's not stable, it takes away those s electrons and lets the stability of a half-filled or fully-filled subshell shine through. Double sharp (talk) 03:12, 10 April 2020 (UTC)
• I say you can't find the start of any other contraction that way: Sc has no d electron, Th has no f electron, those are the usual oxidation states. You have to pick neutral atoms or +2 cations, and then Yb is the last one using its 4f electrons for anything.
• Sandbh says the Ln contraction isn't a one-off, pointing to the scandide and actinide contractions as comparable
• I note that they're only comparable electronically, not chemically, because there's no characteristic oxidation state to match across them.
• Sandbh quotes a paper to support 4f involvement in Lu; I noted that the paper does not actually support that, as it suggests 4f involvement to be weaker than the involvement of the 5s and 5p orbitals (which are part of the core).
  • Clarification: It is nevertheless true that the paper supports 4f involvement in Lu, even if weaker than 5s and 5p. Sandbh (talk) 01:35, 10 April 2020 (UTC)
    • We are going to have involvement of every single orbital, speaking pedantically, because there is a non-zero chance a random 1s electron in a Lu atom somewhere on Earth is going to find itself far enough away from the nucleus to participate in bonding. There is also a non-zero chance that it appears on the Moon. Or on the planet Skyron in the Andromeda Galaxy, for that matter. It is just that at some point you have to draw a line because the chance ends up being ridiculously tiny. (After all, you refer to semi-continuums yourself, this is not a foreign concept.) And being lower than subshells which everyone agrees are in the core and not participating significantly is pretty damning. Double sharp (talk) 03:12, 10 April 2020 (UTC)
• Sandbh wants to use local patterns like diagonal relationships or supposed group divides (that only exist if you look at just the right oxidation states); I say those only come after you've erected the PT's structure through a means that encompasses every element.
  • Clarification: I agree with Double sharp's last sentence. Sandbh (talk) 01:35, 10 April 2020 (UTC)
    • So why refer to diagonal relationships to support Sc-Y-La? If you agree with my last sentence, then that means you have to erect the PT's structure first through global means, and then only then note that diagonal relationships (which are not global means, applying as they are only to 29 out of 118 elements) are cool. And since we agree that Al should not go over Sc even though that creates one more diagonal relationship Be-Al-Ti that Rayner-Canham has noted, consistency demands that diagonal relationships not be brought out to defend Sc-Y-La. We can only defend it through global means; the problem is that there aren't any such things that really defend Sc-Y-La if you look globally, as I note in my other points. Double sharp (talk) 03:27, 10 April 2020 (UTC)
• Sandbh mentions DE's, and hence focuses on the delayed start of 4f; I say you can't define them (what is the DE between V 3d4s and Cr 3d4s, anyway?), and that they're irrelevant (chemically bound atoms aren't usually in their ground-state gas-phase configurations)
  • Clarification: The delayed start of 4f is seen in the free gas in a vacuum phase; the metallic phase, and the ionic state. Sandbh (talk) 01:35, 10 April 2020 (UTC)
    • Nope. Realistically speaking, free gas atoms don't matter one bit for chemistry. In the metallic phase, as already demonstrated, La has some f occupancy just like Th does. For ions, let me bring out another of Droog Andrey's old arguments; in practice you do not have a real +3 naked cation in a chemical environment, charge will be transferred and it will be closer to +2 on the La atom than +3. Now I just remark that promotion energy 5d to 4f for La is only 0.9 eV. You can think of it as whether it's more profitable to attach an electron to La 4f or 5d, and 5d is only barely preferred in a vacuum; chemical environments can easily provide the push to swing it to 4f. Meanwhile, promotion energy for Yb 4f to 5d is 4.1 eV, and yet we are sure that 5d is doing something in Yb. There's no delayed 4f start in real chemistry. ;) Double sharp (talk) 03:12, 10 April 2020 (UTC)
• I also note that by that logic, the start of the f block must be staggered (4f starts at Ce in the ground state, but 5f starts at Pa), and the 8th row becomes a horrific mess (we start occupying 8s at E119, 8p at E121, 7d at E122, 6f at E123 or E124, and 5g at E125), so you don't get a Sc-Y-La table anyway from this argument, just an unsustainable situation
• I want to use the idealised electron configurations that match chemically active valence subshells; electronic considerations are the only thing that makes it possible to resolve the group II (Be-Mg-Ca or -Zn?) and III (B-Al-Sc or -Ga?) questions, and we should use them again here. (If we just look at chemical considerations, Al is actually closer to Sc than Ga!)
• I claim (following Jensen) that there's a double standard in saying that Ac 6d7s can't be an f element with the wrong configuration (which can be corrected in chemical environments, of course), but Th 6d7s can
• Sandbh thinks that group divides can be useful, and that group 3 vs. 4 is one, and that that supports Sc-Y-La to make it match groups 1 and 2
  • Clarification: No, I'm not saying a 3-4 divide "makes it match" anything. What I'm saying is the overall chemical behaviour of group 3 most closely resembles that of groups 1-2. Further, trends going down Sc-Y-La resemble those going down groups 1-2, whereas the trends going down Sc-Y-Lu most closely resemble those of groups 4-5+. There's nothing new here; it was noted by Jensen, and partly by Scerri. As the latter asserted, "Chemically similar groups should be close together, either as vertical groups or horizontal triads, with links between related elements clearly visible" (2004).
    • This is just my poor wording in an attempt to be concise, I think. What I meant to say is that I read you as saying that group 3 vs group 4 shows a group divide (viz. chemical difference), and therefore that groups 1-3 are more chemically similar and should be collocated, and so that group 3 should follow a trend like group 1 and 2 (i.e. Sc-Y-La). I think this is not too far from what you mean; if it's not, please correct me again.
    • Now, the statement about the trends is completely accurate. The only problem is that the overall chemical behaviour of group 3 does not closely resemble groups 1 and 2 alone. In actuality, most of those characteristic pre-transition properties are equally shared by the heavy group 4 and sometimes even group 5 elements: Zr, Hf, Rf, Nb, Ta, Db. They strongly prefer the group oxidation state, in which they are d and don't have transition properties to speak of. And the aqueous chemistry of Zr, Hf, and Rf strongly resembles that of the tetravalent actinides Th through Pu which have what is basically pre-transition chemistry, thanks to their high electropositivity. So, if we claim group 3 as trivalent versions of the group 1 and 2 metals, we can equally well claim group 4 from Zr onwards as their tetravalent version along with a bunch of actinides.
    • Meanwhile, the physical properties of Sc and Y are those of normal transition metals, more resembling group 4. This resemblance only works well if Lu and Lr go under Y, of course. Not to mention that the poor coordination chemistry of group 3 also resembles that of group 4, not only groups 1 and 2. So, there's still no group 3 vs. 4 divide, there's still no stronger resemblance of group 3 to group 2 vs. group 4, and so the rest of the argument crumbles. Double sharp (talk) 03:20, 10 April 2020 (UTC)
• I say group divides are never useful (well, except noble gas to alkali metal that is the period divide) and that the real thing going on is generalised Fajans' rules (acidity and covalency increase as charge and electronegative increase and as radius decreases). As expected from those rules, such divides always disappear once we're not in periods 4-6, or if we stop insisting on group oxidation states. So we should ignore them.
• Sandbh thinks the double periodicity of electrode potentials supports the f block starting at Ce, whereas I note that comparisons with the d block clearly show it must start at La.
• Sandbh disputes this, arguing for +2 as a benchmark oxidation state for 3d (stable for a majority) and +3 for 4f. This, however, overlooks the inapplicability of this idea to 4d and 5d. By this logic, we cannot discover that Tc and Re are the "halfway" elements for 4d and 5, because for Tc, +2 isn't a common oxidation state. For Re, it's not even water-stable.
• As reiterated above, the point of using +2 everywhere is not because it's a common oxidation state (because it isn't), but because it corresponds to getting rid of the outer s subshell. So then you can look in isolation at the filling subshell and the stability of the half-filled one. If you want +3, that implies you have an outer 5d6s all the time. Which obviously doesn't happen. ;)

Detail about it can be found above and in the footnotes, but this should be a start that almost fits in one screen. Double sharp (talk) 04:16, 9 April 2020 (UTC)

Cool, thanks. This makes perfect sense. Well said. Team Lu for me! ― Дрейгорич / Dreigorich Talk 04:23, 9 April 2020 (UTC)

@Дрейгорич: You're welcome! I made a few little changes after you commented, but I don't think they change much. (Oh, and: DE = differentiating electron.)

Of course, this summary is dependent on whether Sandbh agrees if I have characterised his position accurately. Double sharp (talk) 04:26, 9 April 2020 (UTC)

May I suggest section title "... intermediate ...", as in: halfway the process? Thanks. -DePiep (talk) 05:29, 9 April 2020 (UTC)

Ye gods! A TLDR of a TLDR? Really? Head slap! Sandbh (talk) 09:38, 9 April 2020 (UTC)

@Дрейгорич: There have been some additions, in which Sandbh clarified his stand, and I rebut it again. Double sharp (talk) 03:20, 10 April 2020 (UTC)

Reread. ― Дрейгорич / Dreigorich Talk 08:33, 10 April 2020 (UTC)

Sandbh

Here it is, my TLDR of a TLDR, written with the benefit of having read Double sharp’s version.

Over the past 100 years a few arguments have been made for Lu. Jensen gave it a red hot go, with several arguments.

None of these went anywhere. Jensen's paper caused a frisson of excitement among a few authors. Not one of these authors, aside from Scerri (a world authority on the periodic table) looked critically at Jensen’s paper. Scerri concluded that Jensen was too selective and that his (Jensen's) arguments don’t cut the mustard. I agree.

A few people (not necessarily in this forum) feel that I sound like an old cloistered cleric, quoting chapter and verse, in order to defend the established tradition. In fact people familiar with me and my work know I can come up with, and promote, some far out ideas. So my support for La has nothing to do with defending the faith, so to speak.

As per Double sharp's quotation of Droog Andrey, arguments in support of La are mutually reinforcing, consistent with the nature of a periodic table as an integrated, complex structure whereas Lu in group 3 unravels this rich tapestry of chemical relationships.

One example is diagonal relationships. The classics are Li-Mg, Be-Al, and B-Si. Such relationships, which encompass another 20 elements (including La) are found in all blocks, and cut across 12 groups. Double sharp dismisses the relevance of such relationships. He says they only apply to 29 out of 118 elements, and don’t constitute a majority. Never mind their presence in all blocks and 12 groups! Extending this analogy we can say that nonmetals must be even more irrelevant since, according to us, there are only 17 of them, they cut across a mere four groups, and appear in just two blocks!

Another example is that I noted the special status of the half-filled and filled valence sub-shells in the cations, at Mn and Zn; and analogously at Gd and Lu (bullets 37–39, here). This analogy relies on +2 as the benchmark oxidation state for the 3d metals since that is the only state common to all of them (aside from +1), and the most stable for six of ten of them, including Mn and Zn. For the 4f metals, the most stable state is +3, and common to them all. Thus, Mn and Zn want to keep their +2 state, just like Gd and Lu want to keep their +3 states.

Double sharp responded by saying my argument is not supported by the data. He says, "We expect the half- and fully-filled subshells in the f rows to also form maxima…which appear at Eu/Am and Yb/No."

Now, Double sharp is happy to support divalent Mn and Zn as the relevant 3d metals since +2 is their most stable oxidation state, but when it comes to the 4f metals he supports divalent Eu and Yb even though +2 is not their most stable oxidation state! Thus, apples are comparable to oranges. Really?

Go figure, I say. Sandbh (talk) 01:53, 10 April 2020 (UTC)

@Sandbh: Added a rebuttal of the +2 vs +3 argument above to my TL;DR. Just look at 4d or 5d, there's no majority +2 state anymore, but we're still sure where the half- and fully-filled ones after we get rid of the s orbitals. Double sharp (talk) 03:02, 10 April 2020 (UTC)

@Double sharp: Same applies to the Ln once you get rid of their ds electrons. We can see where the half- and fully-filled ones are, as per Klemm (1929), Endres (1932), Shchukarev (1974), Ternstrom (1976), Rokhlin (2003). Sandbh (talk) 05:48, 10 April 2020 (UTC)

@Sandbh: If you go by ds as the covering subshell, you would expect a half-filled or fully-filled configuration (fds or fds) to be unstable still because of that unpaired d electron: it's not a closed shell, unlike what happens with Mn ds and Zn ds. That is unlike anything else in the PT, and is another argument against this chemically weird idea that one d electron hangs up before the other nine get to fill.

There's no need to invoke ds as the ideal configuration for the Ln to explain the +3 oxidation state. Chemically to get a +3 state you are taking something out of the f orbitals ideally, that's not a problem. It's totally analogous to what Fe does: Fe is ds, to get +2 it gives away the s electrons to get ds, to get +3 it throws in a d electron to get ds. It's not the slightest problem to posit most of the lanthanides doing the exact same thing. Double sharp (talk) 05:51, 10 April 2020 (UTC)

@Double sharp: Well, I'm not invoking anything, aside from condensed and ionic configurations. You know the condensed configuration of e.g. Nd is 4f5d6s and that the ionic Nd configuration is 4f. There's no need to invoke or posit "throwing in" an f electron. Sandbh (talk) 06:42, 10 April 2020 (UTC)

@Sandbh: There is no integer occupancy of subshells in the condensed phase, all you have is electronic band structure. In chemical environments, configurations can switch around, what matters is the number of electrons you can get and whether you can force them into a half-filled or fully-filled situation plus the s orbital. You can absolutely have an f electron promoted, or not; depending on pressure, Eu and Yb will change their mind on whether to be divalent or trivalent in the metallic phase, for example. It's exactly what I've been saying about the interplay between 4f and 4f configurations (which, as usual for most electronic features of the lanthanides, doesn't exist for Lu).

My explanation of Ln +3 is absolutely standard, see this resource from the Open University: there isn't a difference between how the 3d elements use their 3d electrons and how the 4f elements use their 4f ones. Once the atom is significantly ionised, it's no longer profitable to bring out more such electrons for bonding. That just happens sooner for 4f than 3d, probably because 4f ends up contracting quickly through an electron with higher electron density (4s+d+p instead of just 3s+p), and because 4f was already penetrating the core quite a bit more than 3d was penetrating the core. You can see that once radius increases for 5f (where there are radial nodes), we can get higher oxidation states.

So, that's precisely why the true analogy is from Mn (5 electrons plus an s-orbital) to Eu (7 electrons plus an s-orbital), similarly Zn to Yb. Just look at where the melting points drop drastically (Mn and Zn, same as Eu and Yb) when the half- and fully-filled subshell is more reluctant to be ionised. Double sharp (talk) 06:55, 10 April 2020 (UTC)

@Double sharp: And what does the melting point have to do with the actual chemistry of the elements involved, as per the comparative tables of standard reduction potentials I posted? And what does pressure have to do with our periodic table of the chemical elements in ambient conditions?

I've seen that unattributed and unsupported paper before. It strikes me as more complexity than is required, as if the author was unaware of the Ln configurations in the metallic phase. Why doesn't it apply to the metallic phase? Sandbh (talk) 12:12, 13 April 2020 (UTC)

@Sandbh: I wonder why Greenwood & Earnshaw discuss melting points then, since their book is called Chemistry of the Elements? An explanation is, of course, not far off: what controls melting points is just the strength of bonds or intermolecular forces, which goes back to the electronic structure that underpins chemistry as well. There's not really a difference between how you explain melting and boiling points and how you explain chemical reactivity on this basis: they are both reflexions of the atomic structure and both are universally considered as periodic trends.

Condensed-phase configurations are going to be mixtures of many configurations anyway and obviously don't fit your criterion of simplest sufficient complexity. Better to just consider the total occupancy of chemically active subshells, as that's the one thing that stays constant and easily explains the situation. Which supports a Lu table all over again. Double sharp (talk) 12:52, 13 April 2020 (UTC)

Triple TLDR: trying to get to the heart of the disagreement

So far as I can see, the points that Sandbh commented on reveal two major points of difference between us:

• A significant argument line for Sc-Y-La appears to boil down to a supposed delayed start of the f orbital, to explain the otherwise weird deformations in trends from other blocks (Sc-Mn and Fe-Zn form double periodicities that look more like La-Eu and Gd-Yb respectively, than Ce-Gd and Tb-Lu with their weird dips at penultimate members of the family).
• I assert that there is no such thing in real chemistry. Not unless you invoke a double standard that admits f involvement for Th 5f6d7s, and not for Ac 5f6d7s or La 4f5d6s. Not to mention that a delayed start would also imply a delayed end, and 4f in Lu or 5f in Lr is none other than a core subshell.
• Another significant argument for Sc-Y-La appears to boil down to the assertion that the group 3 metals (and the lanthanides) have chemistry most similar to groups 1 and 2, being basically trivalent versions of those metals; therefore, since Sc-Y-La gives a group-2-like trend, and Sc-Y-Lu gives a group-4-like trend, the former is to be preferred.
• I note that Zr, Hf, Rf, and Th are essentially tetravalent versions of the group 1 and 2 metals, too. And we can draw analogies easily between the organometallic chemistries of Sc and Ti (where lower oxidation states are stabilised for both, and both organometallic chemistries are dominated by cyclopentadienyls). In general, we will have strongly pre-transition-like chemistry for not just the metallic s elements, but also the f elements, group 3, heavy group 4 (Zr-Rf), with heavy group 5 (Nb-Db) as a bridge to more transition-like chemistry.

As a clarification of my stand on the issue, I consider Sc-Y-La a valid piece of secondary periodicity, just like B-Al-Sc or Be-Mg-Zn; but I think that considering it primary periodicity to be reflected by the PT is inconsistent with the real principles underlying the PT, which are about idealised electronic structure, whose predicted numbers of valence electrons and types of valence subshells are retained in real chemistry. Double sharp (talk) 13:26, 13 April 2020 (UTC)

Appendix: MP, BP & standard reduction potentials

@Double sharp:

Too much importance is placed on gas phase configurations.

Here Mn is 3d4s and Eu is 4f6s. Half-filled sub-shells all round.

We can see an associated phenomenon in the melting and boiling points (per your tables):

Unlike the rest of the Ln, Eu and Yb delocalise only their two s electrons, so their MP and BP dip. The other metallic Ln delocalise three electrons.

So, due to the gas phase configurations and e.g. the MP and BP dips, people think just as 3d runs from Sc to Zn, so 4f must run from La to Yb.

Now, in aqueous solution—where real chemistry occures—Mn likes to be Mn i.e. d. And Eu likes to be Eu i.e. d. So the analogy is lost; it is rather Gd which likes to be d, as Gd.

Why then does Eu f with its half-filled sub-shell stability like to lose an electron and become Eu f? Has Eu gone mad?

No. Apparently the energy required to ionise Eu f to Eu f is more than offset by the gain in hydration energy.

As another example, Na could hypothetically become Na and form NaCl. Theoretically such a compound would have a significantly larger lattice energy than NaCl. However, sodium's second ionisation energy is prohibitively large, which means it stays put in the +1 oxidation state.

So the chemistry-based analogy is Mn and Gd, with 3d going from Sc to Zn, and 4f going from Ce to Lu.

How do you see this? Sandbh (talk) 03:36, 15 April 2020 (UTC)

It be a template

Following the talk here, at #Introduction_into_superheavy_elements, it is that Hassium#Intoduction be a separate, re-used body of text.

As it happens, that body of text is to be a template, not a "part of article text be reused". Currently, there is a shifting around between pages without any use.

Al things short: the content will be in template {{Superheavy element introduction‎‎}}, not in mainspace Superheavy element/Short introduction. Not.

If and when R8R continues to disrupt this clear & sound process, I must consider a block request. -DePiep (talk) 22:42, 4 April 2020 (UTC)

Ask for it any time. We'll see how far the request goes.--R8R (talk) 22:59, 4 April 2020 (UTC)

WP:BOOMERANG. ComplexRational (talk) 23:49, 4 April 2020 (UTC)

"Boomerang" is not an argument. You can do better. -DePiep (talk) 23:59, 4 April 2020 (UTC)

I'm not arguing anything. My point, as explained in that essay, is that such a block request could very well end up with you being blocked again for incivility (for months, if not indefinitely; I read exactly what happened last time) and the problem(s) still unresolved. No admin will turn a blind eye to this, and neither will I, so I suggest you not ask for anything and turn your attention to other, more pressing matters. ComplexRational (talk) 00:13, 5 April 2020 (UTC)

I'm not arguing anything. My point ... is ... LOL. -DePiep (talk) 00:28, 5 April 2020 (UTC)

@DePiep: When you say, "I will not allow" when DS and I are attempting to provide you with an "immediate" overview, I feel annoyed. When you ask for an "immediate" overview of a conversation DS and I've had for some months, I feel like you are treating us like servants, without consideration of the effort required to respond to your "demand". When you unilaterally fold an ongoing conversation I feel quite annoyed.

I ask you to behave in a civil manner, as a peer amongst equals, not as a king of the realm. Sandbh (talk) 03:53, 6 April 2020 (UTC)

You are right. I apologise. -DePiep (talk) 05:22, 6 April 2020 (UTC)

I thank you all here for checking me in this. -DePiep (talk) 18:53, 8 April 2020 (UTC)

@DePiep: You're very welcome DePiep. Sandbh (talk) 05:34, 15 April 2020 (UTC)

File:Mercury spectrum visible.png =? File:Barium spectrum visible.png

Hello,

An email we received on OTRS suggests that both pictures are identical.

Would someone be able to check if they are correct?

Best,

--AntonierCH (talk) 13:36, 5 April 2020 (UTC)

@AntonierCH: It indeed appears to be the case. Although the files differ in size by 20 bytes, I compared them via this tool and found no difference. I am not very experienced with this sort of stuff, but if I interpret the data from the National Institute of Science and Technology correctly, then this picture is likely not to be that of mercury, as the orange and red lines tell. And it's probably not barium either, because that would have a very strong green line in its spectrum at 554 nm. It sure beats me what this is. Perhaps if someone has more knowledge on this matter and can comment, that would be great, but for now, I am concerned about the whole set of spectra.--R8R (talk) 16:47, 5 April 2020 (UTC)

@R8R: I took a sample from another source - I have confirmed (at least 99.999% sure) that the image shows barium's spectrum. However, the intensities of the lines are a bit off. The bright green line (the last thick green line closest to yellow in the Misplaced Pages image) is a bit dim here. It could be a limitation of the software used to draw the lines. But the fainter lines are all showing up in exactly the correct places for barium. ― Дрейгорич / Dreigorich Talk 21:22, 5 April 2020 (UTC)

They are both cited to a Matlab function that is documented to come with table of spectral data. The README does not cite the sourse of the data. I cannot download the file to see the values or run it (to see if the results match the images) or if there is a cite somewhere internally. Commons has about 100 spectra cited to that program among the 160ish in commons:Category:Atomic spectra. DMacks (talk) 05:36, 6 April 2020 (UTC)

Did anyone ping User:McZusatz who uploaded the files? Long-inactive on commons but active on enwiki as recently as December. DMacks (talk) 05:38, 6 April 2020 (UTC)

@DMacks: I did on his Commons talk page, here. AntonierCH (talk) 13:46, 6 April 2020 (UTC)

Perfect thanks, and thanks also for starting c:Commons:Deletion requests/File:Mercury spectrum visible.png. I fixed the link to the user's notification in your message...stray slash char. DMacks (talk) 04:05, 10 April 2020 (UTC)

Orbital radius, EA, and isodiagonality

Two properties
While researching double periodicity, I happened upon an obscure article, which simply correlates electron affinity (EA) with orbital radius, and in so doing reproduces the broad contours of the periodic table (p. 362). Having never thought much about the value or significance of EA, and its absence of easily discernible trends, I was suitably astonished.

• Godovikov AA and Hariya Y 1987, "The connection between the properties of elements and compounds; mineralogical-crystallochemical classification of elements", Jour. Fac. Sci., Hokkaido Univ., ser. IV, vol. 22, no. 2, pp. 357-385.

The authors left out the Ln and An and stopped at Bi. They were sitting on a gold mine but provided no further analysis.

Development
I added the data up to Lr, updated the EA values, and redrew their graph. It's a thing of beauty and wonderment in its simplest sufficient complexity and its return on investment.

I’ve appended 39 observations, covering all 103 elements.

Isodiagonality
The authors refer to the following new diagonal relationships (pp. 370-372):

Chemical

• Na-Ca-Y
• La(Ln)-Th: "This aids in explaining the isomorphism between the named elements in the complex oxides, thorianite." ^
• Ga-Sn
• Zn-In

^If so, this would be congruent with Ca-Y-Ce; according to us, thorianite was so named on account of its high percentage of Th; it also contains the oxides of U, La, Ce, Pr and Nd (weird!)

Solid solution

• Mn-Ru
• Fe-Rh-Pt
• Ru-Ir
• Co-Pd

Mineralogical (and retrograde at that)

• S-As
• Se-Sb
• Te-Bi

Conclusion
So there it is, just two properties which account for nearly everything, and possibly eleven new diagonal relationships.

Observations

1. Very good correspondence with natural categories
2. Largely linear trends seen along main groups; two switchbacks seen in group 13; also falloffs (6p sub-shell) seen in groups 14-17
3. First row anomalies seen for Li (in amphoteric territory), Be (ditto), C (misaligned), N (in noble gas territory), O (misaligned), F (ditto) and He (ditto)
4. For group 13, the whole group is anomalous, no doubt due to the scandide contraction impacting Ga and the double whammy of the lanthanide and 5d contraction impacting Tl
5. Nitrogen was called a noble gas before the discovery of the real noble gases and appropriately enough falls into that territory

6. Rn is metallic enough to show cationic behaviour and falls just outside of noble gas territory
7. F and O are the most corrosive of the corrosive nonmetals
8. The rest of the corrosive nonmetals (Cl, Br and I) are nicely distributed, across the border from F
9. The rest of the simple and complex anions, funnily enough, comprise the intermediate nonmetals
10. The metalloids are nicely aligned; Ge falls a little outside of the metalloid line, being still occasionally referred to as a metal; Sb, being the most metallic of the metalloids falls outside the border; At is inside; Po is just outside

11. Pd is located among the nonmetals due to its absence of 5s electrons; see https://pubs.rsc.org/en/content/articlelanding/2013/dt/c3dt50599e#!divAbstract
12. The proximity of H to Pd is astonishing given the latter’s capacity to adsorb the former
13. The post-transition metals (PTM) form an "archipelago of amphoterism" bounded by transition metals: Ni and C to the west; Fe and Re to the south; V, Tc and W to the east; noble metals to the north
14. Curiously, Zn, Cd, and Hg are collocated with Be, and distant from the PTM and the TM proper (aside from Mn)
15. Zn is shown as amphoteric, which it is. Cd is shown as cationic but is not too far away from amphoteric territory; it does show amphoterism, reluctantly; Hg is shown as amphoteric which is the case, weakly, for HgO, as is the congener sulfide HgS, which forms anionic thiomercurates (such as Na2HgS2 and BaHgS3) in strongly basic solutions

16. The ostensibly noble metals are nicely delineated; Ag is anomalous given its greater reactivity; Cu, as a coinage metal, is a little further away
17. The proximity of Au and Pt to the halogen line is remarkable given the former’s capacity to form monovalent anions
18. The ferromagnetic metals (Fe-Co-Ni) form a nice line
19. The TM from groups 4-12 form switchback patterns e.g. Ti-Zr and the switchback to Hf
20. The refractory metals, Nb, Ta, Mo, W and Re are in a wedge formation

21. Tc is the central element of the periodic table in terms of mean radius and EA values; V is close, Cr is a little further away
22. Ti is just inside the basic cation line; while Ti(IV) is amphoteric, Ti3+ is ionic
23. Sc-Y-La shows a main group pattern up to La, when there is a switchback to Ac
24. Sc-Y-Lu-Lr shows a TM switch back pattern
25. La, and to lesser extent Ce are rather separated from the rest of the Ln, consistent with Restrepo

26. Sc and Lu are close to the amphoteric territory and are both in fact, weakly amphoteric
27. The post-cerium Ln and An (but for Th) all fall within basic cation territory
28. EA values for the An are estimates and need to be treated with due caution
29. The light actinides (Th to Cm) occupy a tight locus, with the exception of Th, where the 5f collapse is thought to occur, and Pu, which sits on the border of 5f delocalisation and localisation
30. While the light actinides U to Cm are shown as being cationic they are all known in amphoteric forms

31. The heavy actinides, Bk to Lr, are widely dispersed
32. All the Ln, bar Tm, are located within close proximity of the light An locus; Tm is the least abundant stable Ln
33. The gap between La and Ce, and rest of the Ln is consistent with Restrepo’s findings
34. Nobelium in this edition of the chart falls off the bottom, having a radius 1.58 (cf Es) and an EA of -2.33
35. There is an extraordinary alignment between He and the Group 2 metals

36. Magnesium is on the cationic-amphoteric boundary; some of its compounds show appreciable covalent character
37. Li, being the least basic of the alkali metals, is located just outside the alkalic zone; Li compounds are known for their covalent properties
38. The reversal of the positions of Fr and Cs is consistent with Cs being the most electronegative metal
39. A similar, weaker pattern is seen with Ba and Ra.

I'll add this to our EA article in due course. Sandbh (talk) 05:56, 13 April 2020 (UTC)

Looks interesting, but the data should be carefully checked. Orbital radii of transition metals are hard to calculate, and the values from reference books are questionable as a rule. Negative electron affinities are also unclear because they have no physical sense; they go closer to zero when more diffuse functions are added to the basis set. I would not recommend to use this chart as it is. Droog Andrey (talk) 07:55, 13 April 2020 (UTC)

@Droog Andrey: Sigh. The article giving the orbital radii has been cited 514 times. Only 77 times since 2016. The EA values are from our own article, every single source of which I personally checked. Just what level of evidentiary support do you expect? Do you feel I post these things without doing due diligence? Do you feel I'm a pissed off by your Holier-than-thou attitude? I sure am. Sandbh (talk) 08:09, 13 April 2020 (UTC)

I feel I'm more willing to believe Droog Andrey's misgivings because he is a computational chemist and neither of us are. So, @Droog Andrey:: could you explain to us why orbital radii are problematic for the transition metals?

P.S. If you are going to draw trends, I suggest again what Wulfsberg does: predict degree of hydrolysis of cations from what are basically the generalised Fajans' rules I have been advocating. Double sharp (talk) 10:09, 13 April 2020 (UTC)

@Sandbh: It is not a matter of due diligence. It is a matter of understanding whether or not the things you are diligently plotting make chemical sense. I was doing the same thing until Droog Andrey explained the problems with our old Sc-Y-La approach, for example. And I feel that in this case I want to learn from him, and not just count citations, when I bet you can find tons of citations for the d-orbital explanation of SF₆ that is totally wrong. Double sharp (talk) 10:11, 13 April 2020 (UTC)

@Double sharp: I was recently asked by a well-respected chemist/physicist for some help for them and their colleagues correcting an article. They and their colleagues don't get how we work in that they make some edits and then get reverted. They don't understand that reputation without citing sources won't work. It's a good thing to learn from people like Droog Andrey, and I welcome his participation. On the due diligence thing, I do it with much more care and effort than a professional saying something on the basis of their discipline—since I have no science qualification—and that includes checking with professionals in the field.

What I don't respect in this forum is professionals saying trust me, I'm an expert (not that Droog Andrey has done this) or making assumptions about how much work and research I may have conducted before posting. Amateurs have made significant contributions to chemistry. When a chemist says, "that can't be done" they do so from within their frame of reference, whereas non-professionals have no such blinkers, and say, "why not?".

Cross-discipline academic expertise (which I have) can bring fresh thinking to the table, and that can only be a good thing

On "whether or not the things you are diligently plotting make chemical sense". Really? You're familiar with my work here and my two FA. I gave the details of the article in which the original plot appeared. I checked Google Scholar for other items by Godovikov and saw he has written a lot in this field. I looked at his and my plot and asked myself: is there anything here untoward? Not to my eye. I checked the works of Sneath; Restrepo; Leach; and Schwarz. I couldn't find any inconsistencies. Hell, I even checked my own peer-reviewed work. Again, nothing untoward. I saw the Pd anomaly, and hunted down the explanation.

You know I've been published in peer-reviewed academic journals. I do my research. The same goes for my edits to Misplaced Pages.

All this counts for nothing in Misplaced Pages. So I cite my damned sources.

I've never looked for citations for the d-orbital explanation of SF₆ but I've read of the controversy and would check my sources. Sandbh (talk) 11:46, 13 April 2020 (UTC)

@Sandbh: The Pd anomaly is, in fact, exactly why I agree with Droog Andrey that this chart shouldn't be used as it is. Because chemically speaking, Pd is no different from any other 4d metal: 4d, 5s, and 5p are all active. The ground-state electron configuration is just an accident that is partly brought on by the higher 4d-5s gap compared to 3d-4s, and is easily corrected in actual chemistry. Much better to look at atomic or ionic radii when this totally disappears; that should give better results.

Citing sources is important, but there is also a need to understand the material so that you know which sources are likely to be more accurate. That's why when I know I don't understand something that comes up here, I ask someone who's likely to know more. As happened for taking into account orbitals in computational chemistry and for intermetallics in archive 42. Double sharp (talk) 12:46, 13 April 2020 (UTC)

@Double sharp: Words almost fail me.

There's one friggin' anomaly out of 118 = < 1%. I provided a reference explaining the anomaly. Here's another: "The Pd-anomaly is clearly seen experimentally…The small radius of Pd, relative to all its neighbors in the Periodic Table, is a consequence of its outermost electron density deriving predominantly from d-levels, instead of s-levels."

Never mind the other > 99% of data points. Never mind the explanation. In bullet #12 I said, "The proximity of H to Pd is astonishing given the latter’s capacity to adsorb the former." Of course, there is nothing going on here! Professors Andrey and Sharp know better.

Anomalies prompt curiosity and inquiry. That's interesting. What might be going on here? "Bah, humbug" says Scrooge McDouble sharp. "Move on!" "There's nothing to see here!" Never mind my 38 other dot point observations.

While we're at it, let's remove the MP cross EN chart from the post-transition metal article, since that contains at least one anomaly. Oh, and let's remove the EN v standard reduction potential chart from the nonmetal article, since that contains one anomaly. Even better let's have dumbing down project to remove all other interesting perspectives we attempt to add to our articles. Yeah, that'll be good for the world.

Hey! Let's remove the periodic tables from our periodic table article, since each of those tables contains at least one anomaly!

You've done this to me before when you argued for Sb as a metal rather than a metalloid. Never mind the 194 references making up the list of metalloid lists showing Sb was included in 87% of these. Never mind my peer-reviewed article on metalloids published in JChemEd, now with 21 citations. And hey, the EA cross OR chart supports Sb as the most metallic of the metalloids. Did you say anything about that? No, of course not, the important thing is to immediately side with Droog Andrey.

You did it to me again when I mentioned the "magic thread" running through the intermediate nonmetals. You undermined that.

You did it to me again when I suggested N should be part of the intermediate nonmetals. You disagreed.

What happened subsequently?

Each of these three items formed part of my peer-reviewed article published in Foundations of Chemistry, with > 800 accesses to date.

More recently you unilaterally removed my contribution to our periodic table article, on La and Lu in group 3, referring to it as "fringe science". Really? Never mind I provided a reference from a well-respected chemist writing in a well-known chemistry textbook. Never that the composition of group 3 is of topical interest to IUPAC. Never mind that there are other sources supporting this option. Did you do any research to see if there were any other such sources? No. Did you add a disputed/discuss tag? No. You decided to unilaterally remove my contribution. Never mind that our periodic table article seeks to provide a good survey of the options on offer for the various controversies. Thanks a lot for imposing your righteousness view upon the world, not. Do you know what self-knowledge is? You need some of that to appreciate how you came over.

As a fellow member of this project I feel your agreement with Droog Andrey, in this instance, is beyond belief; unreasonable by any standard of Misplaced Pages reasonableness, let alone an ordinary reasonable person; disrespectful, and incivil. How about stepping back, taking at look at what you wrote, and asking yourself how that contributed to mutual respect and camaraderie among project team members? Sandbh (talk) 08:01, 14 April 2020 (UTC)

@Sandbh: As I explained in archive 42: the single source putting both La and Lu in group 3 (and not all the other Ln as well) that you mention (Silberberg) is not even consistent about doing that in his own periodic table. (Yes, I actually looked at the book!) And Silberberg was the only source mentioned as doing this. It certainly seems fringe to me.

The Pd 5s band is also occupied and contributing in the metal, and that is why it is paramagnetic. "The low-lying levels are hybrids of s-like states with s and d-like Pd states" (doi:10.1016/0038-1098(80)90218-5). (So much for the idea that the proximity of Pd to H on this chart explains Pd's astonishing ability to take up H, when that proximity only comes about apparently because of that weird gas-phase configuration that chemically means nothing, as usual for gas-phase configuration anomalies. I agree that that situation is interesting, but as an explanation this doesn't seem to work based on what I see from papers.) And I suggested alternative measures that get rid of such anomalies and look more closely at what is universally recognised as significant since Fajans: atomic and ionic size.

Why can't I agree with Droog Andrey if he gives a well-argued reason for his disagreement with you? Not to mention that me disagreeing with you about N as part of the intermediate nonmetals had nothing to do with Droog Andrey's opinion, as I recall. Not to mention that I pointed out to him that Sb and Bi were semimetals and had weak metallic chemical properties (and he agreed that Sb and Bi were not true metals); the point which I agreed with him on was that they were both closer to metals than nonmetals. I don't agree with him just because he said something, but because his arguments convinced me. When I argue for Sb as a metal, that is in the context of a metal-nonmetal dichotomy (without metalloids): if you include metalloids, I would include Sb as one indeed, recognising that a metalloid category surely just means elements around the border between more metallic and more nonmetallic behaviour. It's not a dichotomy, there are more metallic elements and more nonmetallic elements, each have characteristic properties, and you have some that are close to the border like Sb really is; there's a difference between placement decisions for a three-way split and a two-way split. I have the right to my own analysis, particularly if it is backed up by actual chemists and actual texts. Which it always is. And I have the right to disagree with you, pointing to understanding of chemistry from the literature to back up my stand, just as you have the right to disagree with me. Please remember what R8R mentioned to us: the truth is born in argument.

If you seek to characterise my actions this way, and if this is going to be your response to disagreement – well, that is regrettable. In that case I think further discussion between us in the near future will not be productive since tempers are clearly flaring on both sides. So, I suggest we both cool this discussion off, stop trying to summarise our things in a TL;DR (because it seems abundantly obvious that we are not going to be able to agree at the moment), and resume this when I do a July RFC that should bring more people in to comment. Double sharp (talk) 09:48, 14 April 2020 (UTC)

@Double sharp: Time is a friend so I'm not as annoyed now as I was yesterday. What riled me was the work I put into checking the article and updating the chart, only for the two of you to more or less immediately dismiss all of it based on one anomaly, never mind the other merits. Normally I'm quite diplomatic. Yesterday I decided I'd had enough of the kind of behaviour that's caused me to feel the way I felt, and for the first time decided to let rip as if to say, "enough BS, no more Mr nice guy; I'm not putting up with this crap anymore."

Anyway. Back to business.

I raised about two dozen items in my post setting out my feelings on, and giving examples of, your behaviour towards me.

As usual you adopt a defensive mode, choosing to ignore twenty of these.

You couldn't even bother to acknowledge the unilaterality of your removal of my edit re Silberberg. Nice way to treat a fellow project member.

For Pd, and as usual, you misquote me and go into defensive mode. I never said "the proximity of Pd to H on this chart explains Pd's astonishing ability to take up H". I said, the proximity is interesting and that I wondered what was going on. No acknowledgement of the extra citation I provided explaining the smaller size of Pd.

You are of course entitled to disagree with me. That said I was stunned by the rapidity with which you agreed with DA (not a citation in sight). On the N question, and the Sb question I was complaining about your consistent pattern of "self-righteousness" behaviour, which never goes anywhere. Meanwhile my positions appear in the peer-reviewed academic literature.

The truth is not born in argument. It's born in discussion to find out what is right.

All that said, I've found our argument-discussions to be a worthy and richly rewarding learning experience for me, and I hope for you, and our contributions to Misplaced Pages. Don't read too much in my venting. I hope you can learn something from me as to my feelings on how to work collegiately and respectively. I look forward to continuing to work with you.

Please don't feel undue concern about what I've said. We're nowhere near FJ's unacceptable standard of behaviour. I only ask for respect from a fellow project member. Sandbh (talk) 03:21, 15 April 2020 (UTC)

Isolation of an elusive phosphatetrahedrane

The first synthesis of a PC₃ unit arranged in a tetrahedron, here.

This is quite something since plain all-carbon tetrahedrane (CH)₄ has never been isolated. P was used in light of its capacity to form tetrahedral molecules, and the similarity of some of its properties to those of C, as noted in our nonmetal article, now updated. Sandbh (talk) 00:14, 14 April 2020 (UTC)

Careful...this was the tri-tert-butyl derivative of the C₃P core, the same type of derivative of which the C₄ core was made over 40 years ago (tetra-tert-butyl derivative). and it seems like the new molecule was made using an electronically equivalent final closure step as the second-generation synthesis of the all-carbon case. It's great to make unusual molecules (and especially to expand the scope of core atoms and the number of sterically shielding groups needed to be able to isolate it. But it's quite unlike what your "quite something since..." comment seems to imply. DMacks (talk) 05:25, 14 April 2020 (UTC)

@DMacks: Thank you for the feedback. Could you have a look at the nonmetal article, here, and let me know if my mention of this development is appropriate? It's in the "carbon and phosphorus paragraph", last two sentences. I tried to tone it down. See also this item in the Chemistry World weekly newsletter. Sandbh (talk) 08:10, 14 April 2020 (UTC)

• Project-Class chemical elements pages
• NA-importance chemical elements pages
• WikiProject Elements articles