Talk:Periodic table/Archive 16

This is an archive of past discussions about Periodic table. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 10

←

Archive 14

Archive 15

Archive 16

Archive 17

Updated Periodic Table template

I have updated as per IUPAC the periodic table template with a modified template in which Lu and Lr are placed in their respective lanthanoids and actinoids row. If there are no objections, in a couple of days, I will place the update on main. --Robertiki (talk) 14:12, 14 October 2022 (UTC)

I'm afraid you have strayed into one of the most controversial aspects of the article. If you consult the Talk Page archives, you'll find endless discussion of what is called the Group 3 problem. It is covered in the main article at Periodic table#Group 3 and a spin-out article Group 3 element (especially Group 3 element#Dispute on composition. The issue is not the creation of a template, it is what the template should show to reflect the consensus of Wikipedia editors. I would summarize the current consensus as saying that we should wait until IUPAC's committee on the Group 3 issue (see item 7 on this IUPAC page) has reached a conclusion. They have been discussing this since 2015 and as far as I know have yet to publish the "answer": indeed they may have ceased trying! Meanwhile, I doubt you will achieve agreement to alter the template. Mike Turnbull (talk) 14:48, 14 October 2022 (UTC)

I would see the IUPAC proposal as a compromise. I made a second version under the first, restoring the blue color for Lu and Lr (I guess I made a mistake). So we have period 6 and 7 of group 3 undetermined and leaving the contested elements all under the relative list. In this way we do not take a position other at saying that there is no conclusion of what exactly to place in group 3. My source is the IUPAC 2018 table. --Robertiki (talk) 17:36, 14 October 2022 (UTC)

Yes, that's the same table as in the link to the IUPAC website I supplied. Unfortunately, your second version does take a position, namely that Lu and Lr go under Sc and Y (rather than La and Ac doing so). That's the heart of the controversy! We need to wait to see what other interested editors such as DePiep and Double sharp think. Mike Turnbull (talk) 20:38, 14 October 2022 (UTC)

In 2021, IUPAC released a provisional report concluding that the Sc-Y-Lu-Lr composition of group 3 was the preferable one. When it came out we had a discussion on WT:ELEM and agreed to reflect that, and that ended a time period when WP:ELEM members (including myself) were arguing about it way too much (and I wasn't being civil about it at times, which I regret). I'd rather not return to that time, and a point in favour of the current situation is that (1) it arose through a consensus and (2) it totally stopped the arguments that Mike Turnbull mentions; they have not flared up since. It is true though (and odd) that IUPAC still hasn't updated their website to actually show it.

The history is mostly indeed summarised in Periodic table#Group 3 and Group 3#Dispute on composition. But just to be clear: showing Sc-Y-*-** is not "not taking a position". It is rather a third side that claims all lanthanides and actinides belong to group 3. IUPAC indeed intended it as a compromise when it was introduced back in 1988 (although oddly enough the IUPAC publication that announced its introduction then immediately went on to praise Sc-Y-Lu-Lr and explain how to display it), but that's not how in practice it has been treated. The choice is not between some neutral position and other POV's, because there is no neutral position here: it's between which POV you show for the template and which ones you talk about in the text. That is unless you show three or four periodic tables everywhere – and we do here in the dedicated section about the dispute (Periodic table#Group 3) – but I would find that a bit silly for, say, another article that needs a PT for illustration, but whose topic is far away from what this dispute really is about: small differences in properties between some rare earths. Double sharp (talk) 22:10, 14 October 2022 (UTC)

First quick reply: yes, as Double sharp says: the 2021 report is the most substantiated and recent position as preferred presentation. That is why enwiki back then decided to use that one as our topmost article stance. Other (scientific) claims re group 3 (such as older ones and the as yet underpreferred ones) have a place in the encyclopedia, but only as dedicated topic. Not the default PT. Note that we write "preferred presentation", not "THE PT form".

I stress that the PT on the IUPAC site has no authority as scientific claim. It has editorial expressions, which we are free to ignore. DePiep (talk) 13:39, 15 October 2022 (UTC)

BTW as a sidenote: I object to using "lanthanides and actinides" here as an alternative to "group 3" and "f-block" class issues. The loose and impractical class definition of Ln's and An's actually stems from the group 3 &tc unclarities, and so do not make an new, independent classification/clarification base. (Also, they are a sloppy leftover of the PT-expansion with Seaborg discoveries in the 1940s, as a tail & asterisk subordering). Is why these words do not appear in our generic PTs here, long time. "lanthanides, actinides" is used actually in other classifyonig schemes, in reltation to are earths, metallishness, heavy metals etc (sub)classifications. They are in diferent sets. -DePiep (talk) 14:02, 15 October 2022 (UTC)

I was not aware of the current dispute. And that is not good, because looking at the table, without reading all the body, I would think Lu and Lr as group 3 elements. I have changed the second version of the table reflecting that, changing color of the disputed elements to pink. The leggend should add “d/p-block” or a “disputed” to reflect that there is no agreement, so to make aware the occasional browser of the fact. I hope that I have correctly understood what is on stake. --Robertiki (talk) 14:11, 15 October 2022 (UTC)

No blaming or anything intended. You absolutely have a right to clarification and explanation of the backgrounds of current situation. Best take is Double sharps 22:10 post above. I note that your current use of the asterisks, graphically by definition it states that all Ln's and Ac's are under (stretched) the Sc and Y elements. More group 3 variants at Template:Periodic table (micro) § Group 3 variants. Have a nice edit. User:DePiep 14:27, 15 October 2022 (late sign)

And that is not good, because looking at the table, without reading all the body, I would think Lu and Lr as group 3 elements. – well yes, but looking at your revised table, without reading all the body, and if I magically forgot everything I knew about this, I would think that all lanthanides and actinides live under scandium and yttrium. That is, I would think that not only La/Ac and Lu/Lr are group 3 elements, but also things like uranium. Which is itself a side in the dispute, as I explained in my 22:10 post, and as DePiep has just noted. So while I applaud your attempt, I think that what it shows is still not neutral either. This is precisely why we felt the need to have a "preferred form": you can't show all three group 3 options in one table. At least not without creating a hybrid that no one uses.

The dispute is somewhat analogous to the one about whether hypervalent molecules like SF₆ really involve d-orbitals. Textbooks often say yes (copying it each other and some understandable if mistaken early theorisation), there is some argument in the literature, but there's quite a literature consensus that thinks the textbooks are wrong. Similarly here, textbooks often show La under Y (or IUPAC's compromise), there is some argument in the literature, but most people arguing about it think the textbooks are wrong. (And both issues even have raised enough tempers to create some wonderful sarcasm in the literature, e.g. hypervalence, group 3) Taking Lu under Y as the "preferred form" is thus essentially citing the literature to counter the textbooks – but I think that is not a bad thing when that literature has been going strong since 1948 (Landau and Lifshitz were apparently the first to argue this in their physics course, so it's not a new-fangled revolution), and that it's even less of a bad thing when IUPAC's most recent statement is to agree with the literature. Double sharp (talk) 23:12, 15 October 2022 (UTC)

Clarifications

Welcome Robertiki and Michael D. Turnbull.

With respect, and in a polite and calm manner, I beg to differ with my WP:ELEM colleagues Double sharp and DePiep.

Regarding...

"In 2021, IUPAC released a provisional report concluding that the Sc-Y-Lu-Lr composition of group 3 was the preferable one." (Double sharp)

...the provisional report made no such conclusion. All it said was:

"Perhaps [italics added] a compromise could [italics added] be reached on the table depicted as figure 2 [with Lu-Lr in group 3] since it achieves three desiderata."

Nor did IUPAC per se "release" it, i.e. in some sort of officially endorsed capacity. The report was instead published in the newsmagazine Chemistry International, which has news about IUPAC, its chemists, publications, recommendations, conferences and the work of its Divisions and Committees along with special feature articles published quarterly.

Regarding...

"...the 2021 report is the most substantiated and recent position as preferred presentation." (DePiep)

No, the report has not been substantiated. No, it is not the most recent position on Lu-Lr as the preferred presentation. The most recent article addressing this question was published in 2022, here, by Simon Cotton and team. Cotton is an expert on lanthanide chemistry. Their conclusion(?):

"From the chemistry viewpoint there is no compelling reason for arranging Lu rather than La in one triad with Sc and Y."

Regarding...

"Taking Lu under Y as the "preferred form" is thus essentially citing the literature to counter the textbooks – but I think that is not a bad thing when that literature has been going strong since 1948 (Landau and Lifshitz were apparently the first to argue this in their physics course, so it's not a new-fangled revolution)…" (Double sharp)

There is also a fair amount of earlier literature supporting La-Ac in group 3. For example:

"The location and composition of Group 3"^ (2021), with 111 references, as cited by Cotton et al. (COI declaration: this article was written by me);
From 2020, we may recall https://pubs.acs.org/doi/10.1021/acsomega.0c00256 which supported La in group 3 on the grounds of a comprehensive analysis of electronegativity scales and considerations; and
Restrepo (2018, pp. 94–95) (a former member of the IUPAC Group 3 working group, no less) supported La under Y on the basis of stoichiometric considerations.

^ This article includes a critique of Landau and Lifshitz (1948) which found that support for Lu-Lr was a misinterpretation of their position.

Regarding...

"...and that it's even less of a bad thing when IUPAC's most recent statement is to agree with the literature." (Double sharp)

No, as a provisional report, it was not an IUPAC considered, accepted and endorsed statement.

The facts of the matter are that the IUPAC project team, based on a literature survey as at Dec 2019, found that the La-Ac form was the most popular.

Nothing has changed since then.

The consensus arrived at by WP:ELEM is not reflective of the literature-based popularity of the La-Ac form. A more accurate reflection of the literature would be to show the La-Ac form, together with commentary as to the arguments put forward in the literature for that form as well as the Lu-Lr and *-** forms. Sandbh (talk) 04:14, 16 October 2022 (UTC)

For comparison, I understand that the IUPAC/IUPAP Transfermium Working Group concluded its assessment and disbanded in 1991, whereas the results they came up with were only published in 1993.

Now, the IUPAC Group 3 Project ceased at the very end of 2021, so perhaps we can look forward to a final report in 2023. Even then the report, if it is released by IUPAC, will have to go out for consultation with the chemistry community before it is finally put up for formal IUPAC endorsement, which may or may not eventuate in 2024.

Here at Wikipedia I feel that a better approach, in keeping with our role as encyclopedia, would have been to wait for the final IUPAC recommendation (or non-recommendation). Sandbh (talk) 04:14, 16 October 2022 (UTC)

My "the 2021 report is the most substantiated and recent position" is to be read in context, which is explicit about IUPAC & Scerri's research. So read the as their. I do not understand why Sandbh secondguesses the IUPAC "release": it is a scientific publication, and so to be considered as such full stop. Point is, still in context, this is different from claiming that the IUPAC presented PT is "The Official IUPAC PT Form" (quad non). Third, I note that OP and the informal IUPAC PT both refer to the 15-column wide Sc, Y cells per graphical form. That is different from the option Sandbh advocates here (without recognising that difference). It looks like Sandbh is opening a different discussion. -DePiep (talk) 06:49, 16 October 2022 (UTC)

I agree with what DePiep has written. Sandbh is arguing in favour of Sc-Y-La here, whereas the OP was arguing in favour of Sc-Y-*. Though I do have a few comments about Sandbh's post:

On IUPAC's actual website the 2021 report is explicitly stated to be intended as a brief progress report from this task group. And also intended as a call for feedback or suggestions from members of IUPAC and other readers. So it seems the "going out for consultation" must have already started.

It is a cited statement in this article that the majority of work in the literature specifically on this subject supports Lu in group 3, not La. Cited to Jensen, a member of the IUPAC project as well: much of the literature that has appeared on this subject in the 32 years since I published the initial paper has in fact tended to support the proposed reassignment of Lu (paper). And that quite agrees with what I found checking it myself (User:Double sharp/Group 3 sources, though I haven't updated it well since 2021 when the IUPAC report came out). Naturally we cannot expect unanimity, but there's clearly a strong trend towards support for Lu under Y among sources focusing on this issue. The "literature survey" you link to from 2019 is a survey of textbooks, and I think I already explained why that is a different situation. We do not say SF₆ has 3d hybridisation either, right?

Cotton (1) is not a member of the IUPAC project specifically meant to give a recommendation and (2) does not seem consistent about whether he wants La under Y or * under Y in his paper: he writes both the case for replacing La by Lu in Group 3 has not been made on structural grounds, suggesting he wants to keep La there, and the Periodic Table as presented currently, not least in the IUPAC form, gives the most appropriate description of the chemistry of Group 3 and of the lanthanides, suggesting he wants to keep all the lanthanides there. I agree he is an expert on lanthanide chemistry, but I don't see how what he's written can be used as a good argument when he contradicts himself on what his preferred form actually is. BTW, I find his logic very strange here: The structural chemistry of Lu most closely matches that of Y and not of La, so we feel that the case for replacing La by Lu in Group 3 has not been made on structural grounds.

P.S. You didn't seem to mind it being Sc-Y-Lu-Lr when somebody asked about it on this talk page last year. :) Double sharp (talk) 09:37, 16 October 2022 (UTC)

re Sandbh's notion the literature-based popularity [sic] of the La-Ac form: such "popularity" has no place in science, and I don't understand why Sandbh even considers mentioning it. Arguments, not tallying. On top of this, the tallying ignores the nontrivial fact that the group 3 issue has risen only recently (in the PT time scale), e.g. because of new research findings, and then to tally pre-new-knowledge claims unqualified is a priori a methodological misfitting. After all, this is why the Scerri-group was installed. -DePiep (talk) 10:46, 16 October 2022 (UTC)

Made (lower table) another table with disputed color also for third and fourth element in group 3. --Robertiki (talk) 23:38, 16 October 2022 (UTC)

That still looks like all the lanthanides and actinides are in group 3, and it's only the block assignments that are disputed. Double sharp (talk) 08:14, 17 October 2022 (UTC)

As Double sharp writes, Robertiki. Having the aterisks below Sc, Y as you do implies a 15-column wide cell for those two. See the 06:49 PT image above. Expect to be pointed to this error again if you do not try to find out this issue. DePiep (talk) 06:17, 18 October 2022 (UTC)

I’ve been following the thread and intend to post some more comments in reply to Double sharp, subject to some RL commitments, by no later than the end of this week.

In the meantime, the elephant in the room is that the IUPAC provisional report on the progress of the project concluded that there were no objective means to make a decision in favour of either La or Lu. This is despite all the arguments presented in the literature pro-Lu (or pro-La). We are therefore obliged to observe the status quo i.e. La-Ac as the most popular form in the literature. If and when IUPAC endorses the final report of the Group 3 project team, and the recommendation contained therein as to the composition of group 3, that will be the time to review the WP template. I believe this is what Michael D. Turnbull was getting at. Sandbh (talk) 10:52, 17 October 2022 (UTC)

Concluding that the decision cannot be made on objective grounds is not the same as saying that a decision is not being made at all. In the next sentence, the IUPAC report seems to say that this situation makes it more important for them to make a decision: This situation makes it more important that IUPAC should make a ruling on the question which in the final analysis is one of convention rather than one that can be decided on objective scientific grounds. It also concluded that the Lu-Lr form was the only one obeying the stated desiderata. And you seem to be saying that it is not the time to "review the WP template", while arguing for just that (i.e. changing it to something that isn't the Lu-Lr form it currently is at), so maybe I have misunderstood something. And as I mentioned earlier, you are still using "literature" to mean both (1) papers specifically saying something about the issue with an argument or at least a reference to one and (2) textbooks which mostly don't do that. Finally, I agree with DePiep that what you are advancing is a different topic to the OP's. Double sharp (talk) 11:44, 17 October 2022 (UTC)

Thanks Double sharp for your thoughts.

I’m not currently advocating a review of the decision by WP:ELEM to adopt the Lu form. Rather my aim is to discuss the issues and background, and to clarify the situation including understanding the views of other editors, including the OPs. As they proposed replacing the current template with the 15 wide Ln/An form I felt that the topic merited a broader perspective.

Of course, the Group 3 project can conclude that it is important for IUPAC to make a decision. If and when IUPAC do, that is the time for WP to act. In the meantime, as an encyclopaedia, we should show the La form, since that is the most popular as per the survey reported by the project.

The three desiderata are those that the project mentions in its “provisional” report. Whether these desiderata make it into the final report to IUPAC, and whether they wash with IUPAC as a whole remains to be seen.

I am arguing that the consensus reached by WP:ELEM to adopt the Lu form was premature, non-encyclopaedic and non-objective.

Yes, by “the literature” I am referring to reliable sources as defined by WP, including journal articles and textbooks. Sandbh (talk) 02:41, 18 October 2022 (UTC)

No that is not what Mike Turnbull wrote. Instead, MT wrote a clarification of and support for the current enwiki consensus. Also, that's not a correct way to extract 'support' from a post. DePiep (talk) 13:15, 17 October 2022 (UTC)

Oh, this is what Mike wrote: “I would summarize the current consensus as saying that we should wait until IUPAC's committee on the Group 3 issue (see item 7 on this IUPAC page) has reached a conclusion. They have been discussing this since 2015 and as far as I know have yet to publish the "answer": indeed they may have ceased trying!” The committee did not reach a conclusion. All they did was release a provisional report, with fluffy words like “perhaps” and “could”, and to ask for more feedback i.e. they have not made a final recommendation to IUPAC, which is what they were set up to do. Mike can choose his own response or non-response to my ping. Sandbh (talk) 22:57, 17 October 2022 (UTC)

First, why the indirect talk of someones post? Extracting/constructing a !vote? Speak for oneself should do. DePiep (talk) 06:05, 18 October 2022 (UTC)

Second, MT replies to a different proposal (OP's). Even worse, afterwards you changed the topic and then here you have the audacity to reconstruct MT's reply into a sort of support for an altered point. You also ignored the .. the consensus of Wikipedia editors .. part that doesn't fit your reconstruction. As for talkpage flow, all this can be ignored (and preferably redacted/removed by writer ie you to keep discussion clear & clean). -DePiep (talk) 06:13, 18 October 2022 (UTC)

The subject of the thread is replacement of the PT template. My posts have addressed this topic. Sandbh (talk) 06:38, 18 October 2022 (UTC)

As anticipated, this thread shows @Robertiki why I doubted agreement to alter the template would be achieved. Let's all agree to de-escalate any WP:BATTLEGROUND and await the most likely external factor which might alter the existing consensus here: IUPAC's committee reaching a published conclusion. This article is getting about 20,000 page views a day and, in my opinion, already does a pretty good job of explaining the history of the group 3 problem and how the alternative views lead to different presentations in a table. Mike Turnbull (talk) 10:23, 18 October 2022 (UTC)

Thanks @Michael D. Turnbull: I feel there is no WP:BATTLEGROUND to de-escalate. We were having a discussion as to the basis for, and background to, the adoption of the current template. It has been a fruitful, illuminating exchange.

I concur with your sentiment to await IUPAC's committee reaching a published conclusion, subject to such conclusion being endorsed by IUPAC as a whole. That is to say, IUPAC makes a decision to recommend the composition of group 3.

Until that time we are obliged to observe the literature, in which the La form is the traditional, most popular form.

The consensus reached at WP:ELEM talk to adopt the Lu form was/is un-encyclopaedic, and a breach of WP:NOV.

What follows are some pertinent extracts from the history of the group 3 question:

1. "It is also true that the fact that most textbooks discuss Sc, Y, La, and Ac together is, for Wikipedia, probably definitive at this moment.

Respectfully, Double sharp" (talk) 08:51, 21 July 2020 (UTC)

https://wiki.riteme.site/wiki/Talk:Periodic_table/Archive_13

2. "However, since IUPAC is working on it, I will wait till they make their decision before proposing that we change back (if they choose Sc-Y-Lu-Lr) or not (if they choose Sc-Y-La-Ac)."Double sharp (talk) 09:13, 10 September 2020 (UTC)

Comment by me: IUPAC have made no such decision.

https://wiki.riteme.site/wiki/Template_talk:Periodic_table#Double_sharp's_comments

3. "WP is still hampered, in my opinion, by a preliminary report with recommendations / suggestions inconsistent with RS, that does not yet have the imprimatur of IUPAC itself. RS like textbooks will act on this (or not) in future editions, so any new consensus on a convention will take time to evolve.

EdChem (talk) 23:48, 21 January 2021"

Comment: No such consensus has evolved.

https://wiki.riteme.site/wiki/Wikipedia_talk:WikiProject_Elements/Archive_59#Group_3_news_from_IUPAC

4. "Extremely strong oppose (I have to make this prominent to balance and counteract the supports above) Just about all the arguments are irelevant and we should just use the traditional form with La and Ac under Y. If needed we can use the place holder * or **."Graeme Bartlett (talk) 07:52, 21 July 2020 (UTC)

https://wiki.riteme.site/wiki/Talk:Periodic_table/Archive_13

In the preceding extracts, I note the mention of most textbooks showing the La-Ac form; the absence of an IUPAC decision on the recommended composition of Group 3; the absence of a consensus in the literature in favour of the Lu form; and the recognition of La as the traditional form.

Since the Group 3 Pandora's Box has been re-opened I propose to calmly redo the informal poll that was conducted in view of the publication of the IUPAC provisional report on group 3, and which (inappropriately in my view) resulted in the adoption of the Lu form:

https://wiki.riteme.site/wiki/Wikipedia_talk:WikiProject_Elements/Archive_59#Polling

— Sandbh (talk) 07:06, 19 October 2022 (UTC)

First, bringing in an elephant in the room and after misreading Turnbulls posts, also reframing the OP, then self-concluding I feel there is no WP:BATTLEGROUND to de-escalate: it gives me the impression that the talk is being played with. Missing is a basic reasooning say starting from A, with argument flow self-corrections fair enough; instead it is hijacking arguments to redirect and never return. -DePiep (talk) 09:46, 19 October 2022 (UTC)

The subject of the thread is replacement of the PT template. My posts have addressed this topic. Sandbh (talk) 01:50, 20 October 2022 (UTC)

No. It was not about juist any replacement, Sandbh, but a replacement with a different content than what you promote. IOW, you changed topic on the fly. Then you misquoted arguments from the other topic, the one you forked from, to make some point in your new thread. DePiep (talk) 06:19, 24 October 2022 (UTC)

To poll or not to poll; that is the question

@DePiep: It was initiated by Sandbh's comment of 07:06, 19 October 2022 (UTC), near the bottom of the previous section: Since the Group 3 Pandora's Box has been re-opened I propose to calmly redo the informal poll that was conducted in view of the publication of the IUPAC provisional report on group 3, and which (inappropriately in my view) resulted in the adoption of the Lu form. Double sharp (talk) 11:30, 30 October 2022 (UTC)

@Sandbh: I don't see a reason to redo the informal poll because I don't see evidence that the issues are substantially different from 2021 (when the IUPAC preliminary report was already around), or if any minds have actually changed since it was conducted. (Whereas the 2020 quote you have from me is from a different situation because the IUPAC preliminary report of 2021 was not out yet.) I think made it pretty clear in the opening statement that the report was "provisional", and that while it leaned in support of Sc-Y-Lu-Lr, it did not do so that firmly. Though maybe you can judge it based on what I wrote:

The last time we ran this was in Wikipedia talk:WikiProject Elements/Archive 52, after which we had an agreement to not raise the issue again until the IUPAC project responds. Now they have, albeit as a provisional report (presumably, a real one is forthcoming). Group 3 as Sc-Y-Lu-Lr appears to be supported by their wording, albeit not that firmly (perhaps this will be made clearer in the real report).

And I quoted the parts of the report where it said we have concluded that there is no objective means to adjudicate between group 3 consisting of Sc, Y, La and Ac or as Sc, Y, Lu and Lr and Perhaps a compromise could be reached on the table depicted as figure 2 [Sc-Y-Lu-Lr]. So, even if you don't agree with how people voted, I think people did not vote based on a misunderstanding of what the IUPAC report was and what it said. That was clearly presented.

Anyway, such a poll/discussion is done when people have different ideas about what policy obliges or allows us to do, as is evident here (you favour the textbooks as an argument, I favour the literature plus IUPAC report leanings as a subsidiary argument). Surely there can be legitimate differences in opinion; policy does not micromanage to the extent that there is only one way to write an article. I understand you were unable to !vote in that poll, but 7−1 isn't that much weaker than 7−0 as a consensus.

With that said, I like your expansion of the article to include more content showing a Sc-Y-La-Ac argument. It is in the literature and it's simple enough to use as a flagship to explain what such arguments look like. So why don't we carry on improving and expanding the article in ways we can agree on, as we just did? Double sharp (talk) 21:40, 21 October 2022 (UTC)

@Double sharp: Thanks for the opportunity to discuss this.

1. EdChem honed in on the issue when he wrote:

"WP is still hampered, in my opinion, by a preliminary report with recommendations / suggestions inconsistent with RS, that does not yet have the imprimatur of IUPAC itself. RS like textbooks will act on this (or not) in future editions, so any new consensus on a convention will take time to evolve."

That is the current situation; while some RS debate the composition of group 3, the current consensus apparent from the entirety of RS is for Sc-Y-La-Lu.

As EdChem said, even if IUPAC makes a formal recommendation it will take a while for RS to act or not act on it. For example, I recall that when IUPAC recommended abandoning use of the word "metalloid", usage in the literature subsequently increased six-fold. So much so, that this recommendation was subsequently dropped by IUPAC.

Given WP is an encyclopedia, we are obliged to show Sc-Y-La-Ac until there is a change in the consensus of RS.

2. It looks like the history of the Group 3 debate is nicely summarised and balanced in the periodic table article. I'll have a closer look to check, and to see if further La-Ac content would be helpful. Sandbh (talk) 00:13, 24 October 2022 (UTC)

I don't agree that Sc-Y-La-Ac is the consensus of RS, for reasons I already stated: sources focusing on the issue tend to overwhelmingly support Sc-Y-Lu-Lr. Therefore, it looks exactly like the SF₆ situation where textbooks often keep giving the 3d hybridisation explanation while nearly everyone in the literature focusing on this kind of thing objects. There is also the fact that the 2021 IUPAC report does not mention that Sc-Y-La-Ac is more popular in textbooks, suggesting that they do not consider a popularity contest of textbooks to be a binding argument. Double sharp (talk) 12:06, 24 October 2022 (UTC)

It seems there may be need to work out what is meant by consensus of RS? Yes, I agree with you that the IUPAC project apparently did not consider the popularity of Sc-Y-La-Ac to be binding argument. That said, they are not bound by WP policies whereas we are. That is a key difference. What the IUPAC Group 3 project did conclude was that there was no objective way of settling the question. On this basis, the "sources focusing on the issue" and tending to "overwhelmingly support Sc-Y-Lu-Lr" mean naught. The remaining option is the three desiderata mentioned in the Group 3 provisional report. So be it. If Sc-Y-Lu-Lr gets endorsed on this basis by IUAPC as a whole then we'll have to wait and see, as EdChem said, if consensus develops in RS, noting the checkered history of IUPAC recommendations being accepted within the chemistry community.

As I understand it, the SF₆ situation is not so clear. As our article on hypervalent molecule notes:

"In 1990, Magnusson published a seminal work definitively excluding the significance of d-orbital hybridization in the bonding of hypervalent compounds of second-row elements. This had long been a point of contention and confusion in describing these molecules using molecular orbital theory. Part of the confusion here originates from the fact that one must include d-functions in the basis sets used to describe these compounds (or else unreasonably high energies and distorted geometries result), and the contribution of the d-function to the molecular wavefunction is large. These facts were historically interpreted to mean that d-orbitals must be involved in bonding. However, Magnusson concludes in his work that d-orbital involvement is not implicated in hypervalency.^[1]

Nevertheless, a 2013 study showed that although the Pimentel ionic model best accounts for the bonding of hypervalent species, the energetic contribution of an expanded octet structure is also not null. In this modern valence bond theory study of the bonding of xenon difluoride, it was found that ionic structures account for about 81% of the overall wavefunction, of which 70% arises from ionic structures employing only the p orbital on xenon while 11% arises from ionic structures employing an

\mathrm {sd} _{z^{2}}

hybrid on xenon. The contribution of a formally hypervalent structure employing an orbital of sp³d hybridization on xenon accounts for 11% of the wavefunction, with a diradical contribution making up the remaining 8%. The 11% sp³d contribution results in a net stabilization of the molecule by 7.2 kcal (30 kJ) mol⁻¹,^[2] a minor but significant fraction of the total energy of the total bond energy (64 kcal (270 kJ) mol⁻¹).^[3] Other studies have similarly found minor but non-negligible energetic contributions from expanded octet structures in SF₆ (17%) and XeF₆ (14%).^[4]"

Per the IUPAC Group 3 project's provisional report, no article or report has been published definitively excluding Sc-Y-La-Ac in favour of Sc-Y-Lu-Lr. Sandbh (talk) 06:28, 25 October 2022 (UTC)

We don't just have to cite which option the PT has, but also what reasoning there is for it. As such, a source like your average textbook merely giving a PT (but not explaining why their group 3 is like that) may be reliable for many other things, but for this it's just irrelevant in my opinion. We couldn't cite a statement from such a source because it doesn't give one. I point out what IUPAC used on the grounds that it suggests that they share this opinion (after all, they gathered all that data, and then for some reason decided not to use it), although you are correct that they are not bound by our policies.

Anyway, just because the decision can't be made on objective grounds doesn't mean it can't be made on any grounds. The IUPAC report gives its desiderata, after all. It's not as if all the sources in the literature that I referred to claimed that Sc-Y-Lu-Lr was somehow objectively the correct choice, although it's true that some did. But the ones that don't claim it's an objectively decidable matter nonetheless did say things in favour of Sc-Y-Lu-Lr anyway, e.g. Jørgensen (1973) saying Lu "can most conveniently be considered as the first member of the 5d series", or Johnson (1984) saying that both definitions (La-Yb and Ce-Lu f-blocks) have matching drawbacks but that he prefers the first. That seems like they came to a similar conclusion as IUPAC in 2021, that the problem is one of convention, but that there can nevertheless be reasons to prefer one convention over another. And indeed Wulfsberg (2000) even called Sc-Y-Lu-Lr the "metallurgist's convention", citing Jensen and Scerri who would later end up on the IUPAC project. So I don't see how the IUPAC 2021 provisional report makes such sources irrelevant when they are pretty much of the same opinion. I'd need more time to see how many of those sources make such statements instead of saying it can be decided objectively, but considering that Scerri used to think it could be decided objectively and used pretty much the same desiderata anyway then (and that's even from the IUPAC newsmagazine too), I'm not even sure how relevant that is to the count (indeed the 1988 IUPAC report seems to think that it can be decided through physics, chemistry, and electronics instead). Surely there can be legitimate differences in opinion between sources about whether the table is "right" or just a "convention", but that doesn't affect which form is displayed or even necessarily what arguments are used to defend the chosen convention. Double sharp (talk) 08:23, 25 October 2022 (UTC)

To be clear: irrespective of the wording (expanded on below), any such poll is unacceptable as base for encyclopedic judgement on inclusion & presentation. For starters, the (inappropriately in my view) by Sandbh (and other posts, eg in the wordings below) says it is POV. This is the encyclopedia, not a peer reviewed paper-in-development that could redo a researched report. @Double sharp:. I note that this issue (of Sandbh pushing POV & doing paper reviews) occurs more often & with more MOS-capital issues on this talkage & topic. So, it's not just a simple wording thing. -DePiep (talk) 07:29, 23 November 2022 (UTC)

DePiep: Such an "unacceptable" POV poll resulted in the PT in the lede being changed from the La form to the Lu form.

You'll have an opportunity to have your say, as will all other interested parties, when the proposed poll is published. Sandbh (talk) 06:48, 24 November 2022 (UTC)

No. It has been pointed out to you throughout. For starters, we do not build this encyclopedia by ~~votes~~. There is no "we [who?] will let you vote later". Your textbooks is not the RS you take them for. See also this Talk:Lists of metalloids/Archive 1 § Wandering authors (2014) etc. etc. DePiep (talk) 08:01, 24 November 2022 (UTC)

Poll wording

Good. I agree that just because the decision can't be made on objective grounds doesn't mean it can't be made on any grounds. As far as WP goes, and as an encyclopedia, we are then obliged to make the decision according to the consensus in RS. The poll question would look like this:

"Should the default depiction of the periodic table, as shown in the lede of the periodic table article, show group 3 as Sc-Y-La-Ac? It currently shows Group 3 as Sc-Y-Lu-Lr.

Considerations

Textbooks most commonly show group 3 as Sc-Y-La-Ac.^[5]
That said, there are more arguments in the critical literature favouring group 3 as Sc-Y-Lu-Lr than there are arguments supporting Sc-Y-La-Ac.
However, a 2021 provisional report from the IUPAC group 3 project found there was no objective way of deciding in favour of either Sc-Y-La-Ac or Sc-Y-Lu-Lr.
As an encyclopedia we are obliged, therefore, to reflect the consensus of textbooks, in which Sc-Y-La-Ac is the most frequently appearing form.
If and when IUPAC endorses a formal recommendation as to the composition of group 3 we would have to wait to see if their recommendation came to represent a consensus in RS, before acting on it.
This is important as not all IUPAC-endorsed recommendations are necessarily accepted by the chemistry community."

As a draft, how does this look? Sandbh (talk) 01:26, 28 October 2022 (UTC)

This is not a neutral opening statement. Your statement is interpreting all the data in a way that supports Sc-Y-La-Ac, when I could and have been making arguments that it doesn't necessarily have to be interpreted that way. For instance, you omit the facts that (1) the textbooks don't justify their placement and (2) while the IUPAC group 3 project found that there was no objective way of making the decision it nonetheless wrote in favour of Sc-Y-Lu-Lr for other reasons anyway. Also, looking at that survey, you're also omitting the fact that IUPAC shows the data divided by decades. While Sc-Y-La-Ac is indeed overwhelmingly popular from the 1970s to the 1990s, it certainly isn't in the 2000s and 2010s. In the 2010s it doesn't even manage a majority (33 textbooks surveyed, and it's only that way in 16 of them). So it seems to me that your assumption that "textbooks most commonly shows group 3 as Sc-Y-La-Ac" might very well not be true anymore (allowing for the "might" because three years have elapsed from 2019 to 2022, though the trend is suggestive), and I'd object to combining all the data from that long ago when (1) IUPAC never combined the data, leaving it split by decade and (2) key literature statements on this that chemists paid attention to only started to be published in the 1980s (Jensen), which is after the survey starts. Now, even if you disagree with my interpretation here, I think it's clear that these points could potentially change how people look at this data. Finally, your statement takes your interpretation of what policy obliges us to do and takes it as if it is a matter of fact in your fourth bullet point, when the fact that we're having a discussion demonstrates that the situation isn't that clear-cut. You're of course free to make those points, but I think they should be in your !vote rather than in your statement. Double sharp (talk) 09:57, 28 October 2022 (UTC)

Double sharp, Thanks.

In the second column of the below table is a direct cut, as written by you, from the periodic table article (with the addition of three more recent citations). It is consistent with your earlier observation in the third column:

Current	Proposed	Double sharp (talk) 08:51, 21 July 2020 (UTC)
Textbooks most commonly show group 3 as Sc-Y-La-Ac.^[6]	Published periodic tables most commonly show lanthanum and actinium in group 3.^[7]^[8]^[9]^[10]	"It is also true that the fact that most textbooks discuss Sc, Y, La, and Ac together is, for Wikipedia, probably definitive at this moment."

How does this now look?

We have further corroboration from Simon Cotton (a lanthanide chemistry expert) and team who wrote, in 2022, that:

"Traditionally, scandium, yttrium and lanthanum have been considered as a triad due to their occupation of successive vertically arranged boxes in Group 3 of the Periodic Table."

Contrary to your interpretation, there is no Wikipedia requirement for textbooks, or wall charts for that matter (since Scerri also mentions these in his 2020 book), to justify their choice.

No, as I mentioned previously, the IUPAC Group 3 project did not find "in favour" of Lu-Lr. Instead, what they wrote was:

"Perhaps [italics added] a compromise could [italics added] be reached on the table depicted as figure 2 [with Lu-Lr in group 3] since it achieves three desiderata."

The numbers reported by the IUPAC Group 3 project were...

Frequency of group 3 options appearing in textbooks
Decade	La-Ac	Lu-Lr	-*	Total books
2000's	38	14	9	61
2010's	16	6	11	33

...that is, La-Ac was the most popular in either decade.

Finally, on my fourth dot point, I based this on what respected editor EdChem wrote, as I mentioned previously...

"WP is still hampered, in my opinion, by a preliminary report with recommendations / suggestions inconsistent with RS, that does not yet have the imprimatur of IUPAC itself. RS like textbooks will act on this (or not) in future editions, so any new consensus on a convention will take time to evolve.

EdChem (talk) 23:48, 21 January 2021"

...and no such consensus has evolved.

— Sandbh (talk) 13:10, 30 October 2022 (UTC)

Evidently, I changed my mind. Note that 2020 is before the IUPAC preliminary report came out, so there is a good reason for me to have done so. Also, 16 out of 33 is a plurality but clearly not a majority. Most textbooks published in the 2010s show something that is not Sc-Y-La. I've edited the article accordingly to match the 2019 IUPAC findings.

WP has a requirement that information must be cited. We are not just showing a form, but also saying the reasons for it. A source that shows a form but does not provide a reason can surely cite the first bit of that, but not the second. And besides, textbooks are sometimes even confused and show different forms in different places, as Jensen has pointed out (and he's on the IUPAC project). If such confused cases include Cotton & Wilkinson and Housecroft & Sharpe, then what should we expect from the others?

Saying that "perhaps a compromise could be reached on" Sc-Y-Lu is obviously a suggestion in favour of it. It is clearly a suggestion about whether Sc-Y-La or Sc-Y-Lu is preferable and it certainly isn't against Sc-Y-Lu. Also, I agree that IUPAC is not always listened to, but there has literally never been a time AFAICS that IUPAC has said anything in favour of Sc-Y-La.

And could you please not repeat the same arguments over and over? I've repeatedly explained to you why that's not the only possible interpretation, and the consensus from last year indicates that most people who were polled agree with that. Just because people don't agree with you doesn't mean their thinking is un-encyclopaedic, and a breach of WP:NOV as you wrote. To quote WP:IDIDNTHEARTHAT, Believing that you have a valid point does not confer upon you the right to act as though your point must be accepted by the community when you have been told that it is not accepted. The community's rejection of your idea is not proof that they have failed to hear you. Stop writing, listen, and consider what the other editors are telling you. Make a strong effort to see their side of the debate, and work on finding points of agreement. Do not confuse "hearing" with "agreeing with". Double sharp (talk) 13:46, 30 October 2022 (UTC)

BTW, here's an editorial reason why I think it's better to leave it the way it is. (Not a comment on your opening statement.) Even in textbooks, my impression is that it's common to refer to Madelung's rule when discussing how the subshells are filled and therefore how the table is built up. This is consistent with Sc-Y-Lu-Lr: 4f comes before 5d. If you make it Sc-Y-La-Ac, you either have to leave an inconsistency in the article, or you have to explain why the Madelung rule is violated just for that. The first is clearly undesirable and the second immediately runs into trying to cite the arguments, where as I keep pointing out, the consensus of RS is against Sc-Y-La-Ac. If you try citing statements about how that particular Madelung violation justifies rupturing the d-block, what you'll instead find is a few sources like Lavelle and your article saying it does, in contention against a whole bunch like Jensen and Scerri saying it doesn't. I find that a bit awkward. And similarly if it's Sc-Y-*-**, there then needs to be an explanation of how there can be 15 f-block elements when there's only 14 spaces in the 4f shell to avoid another inconsistency with saying the Madelung rule is behind the PT's shape; yet most sources referring to this contradiction use it as an argument against Sc-Y-*-**. Double sharp (talk) 10:00, 28 October 2022 (UTC)

P.S. I didn't even OR this up, even if it's an argument based on what's easier to say and not contradict yourself – it matches the Madelung-rule argument made by Thyssen and Binnemans (2011) as well as Scerri and Parsons (2018). The latter being relevant as Scerri is on the group 3 project, too. Double sharp (talk) 15:48, 30 October 2022 (UTC)

@Double sharp: Oh, if we have gotten to where you feel the need to ask me to not repeat my arguments then I feel there is not much value anymore to continuing the discussion about poll wording.

There is no basis for saying that " 'perhaps a compromise could be reached on' Sc-Y-Lu is obviously [italics added] a suggestion in favour of it", aside from hopeful or wishful thinking. None us know anything for certain about the workings of the project nor the background to the wording of their provisional report. The only thing we know for sure is from the talk Eric gave at the Chemical Education Conference, in Padova, Italy, Sept 3rd, 2021 https://www.youtube.com/watch?v=3Kh1gOjLlUk ? In response to a question from the floor, Eric said:

"i happen to be the chairperson of this 53:12 working group we have not resolved this 53:14 issue there's many many disagreements"

I would surmise that IUPAC has never said anything in favour of Sc-Y-La because they never felt a need to, given La-Ac is (and still is) the most popular form in RS. You may recall FJ's mention of the IUPAC minutes showing how much hostility there was to making a recommendation in favour of the composition of Group 3. Organisational politics at play. Just like, I speculate, was the reason why the project was shut down so abruptly, and the Inorganic Chemistry Division of the ACS reverted back to their depiction of La-Ac, in their newsletter masthead.

I was not able to participate in last year's poll and put my pov. Nor, at the time, would I have been able to do so in a considered manner given the events occurring back then. With the benefit of the passage of time, I feel I have been able to a better job now.

At the time of last year's poll I feel that the issues I have raised in this thread, in terms of un-encyclopaedic, and a breach of WP:NOV were not given due consideration, as alluded to by e.g. EdChem, and Graeme Bartlett.

Hence my proposal for a new poll, noting that, IMH and polite O, the arguments you have presented in support of Lu-Lr, creative as they are :), have not come to grips with un-encyclopaedic, and a breach of WP:NOV.

I have not been in a hurry to "put" the new poll. I will see how it goes from here in terms of timing. Sandbh (talk) 06:53, 5 November 2022 (UTC)

@Sandbh: I'll just respond to what I think is new here, if you don't mind. :)

If saying "perhaps a compromise could be reached on" Lu-Lr is not mildly leaning in favour of it, then I don't know what it is. It certainly isn't sitting on the fence and it certainly isn't leaning away from Lu-Lr either. I am aware that there's a ton of behind-the-scenes politicking but the fact of the matter is that that is what was published.

IUPAC has previously considered the group 3 issue in 1988, when deciding on the 1-18 numbering of the groups. If they decided that Sc-Y-La-Ac was better, they could easily have said so. Yet they didn't do that. Instead they said:

In the "Red Book" which will appear in 1988 the same arrangement was chosen for the elements of the scandium group as in the periodic table as originally proposed by CNIC and subsequently published by VCH Verlagsgesellschaft, Weinheim.
It is a compromise. According to the electron configurations of the elements, the scandium group consists of the elements

Sc, Y, Lu, Lr.
This was pointed out as early as 1959 by L.D. Landau (ref. 20) and later by other authors (ref. 13, 14, 20 to 25). Most periodic tables in textbooks and classrooms, however, list Sc, Y, La, and Ac as elements of the scandium group and designate the elements Ce to Lu and Th to Lr as lanthanides and actinides, respectively. The historical background for this arrangement is given in a paper by W.B. Jensen (ref. 21). Based upon their electronic configurations and their chemical and physical properties, the elements La to Yb and Ac to No should be inserted between barium and lutetium and between radium and lawrencium or for practical reasons be listed at the bottom of the table. The series La to Yb and Ac to No then, however, cannot be named correctly as lanthanides and actinides since they contain the elements lanthanum and actinium and not only elements similar to lanthanum and actinium as is purported by the ending -ide (or -oid according to an earlier IUPAC recommendation).

Incidentally, this is a fairly odd way to introduce a compromise. It certainly reads as though they'd rather have made it Sc-Y-Lu-Lr totally, but some kind of politicking and perhaps a need to not make all textbooks look wrong stopped them. And in the 32-column form they next published, it was in fact Sc-Y-Lu-Lr! (The 18-column was Sc-Y-*-**.)

And I've never seen a IUPAC PT (at least from 1990 onward, haven't looked earlier) showing Sc-Y-La. They either show Sc-Y-* or Sc-Y-Lu – and in the 32-column form they have shown Sc-Y-Lu (1990 Red Book, 2009 CI article discussing the forms) as far as I'm aware. That is itself indicative. Double sharp (talk) 11:20, 5 November 2022 (UTC)

Can someone recap the OPs where this originates from? What is the actual proposal & later forms being discussed? (I understand is has left Clasrifications).Clarification some time ago by Sandbh. -DePiep (talk) 11:59, 29 October 2022 (UTC)

Of course, as long as this wiki has troubles in describing the scientific situation itself, there is no gain is polling (ouch) a presentation anyway. DePiep (talk) 06:45, 15 November 2022 (UTC)

References

^ Magnusson, E. (1990). "Hypercoordinate molecules of second-row elements: d functions or d orbitals?". J. Am. Chem. Soc. 112 (22): 7940–7951. doi:10.1021/ja00178a014.
^ Braïda, Benoît; Hiberty, Philippe C. (2013-04-07). "The essential role of charge-shift bonding in hypervalent prototype XeF2" (PDF). Nature Chemistry. 5 (5): 417–422. Bibcode:2013NatCh...5..417B. doi:10.1038/nchem.1619. ISSN 1755-4330. PMID 23609093.
^ H., Cockett, A. (2013). The Chemistry of the Monatomic Gases : Pergamon Texts in Inorganic Chemistry. Smith, K. C., Bartlett, Neil. Saint Louis: Elsevier Science. ISBN 9781483157368. OCLC 953379200.{{cite book}}: CS1 maint: multiple names: authors list (link)
^ Lein, Matthias; Frenking, Gernot (2005-01-01). "The nature of the chemical bond in the light of an energy decomposition analysis". Theory and Applications of Computational Chemistry: 291–372. doi:10.1016/B978-044451719-7/50056-1. ISBN 9780444517197.
^ Scerri, ER (2019). "Dec 2019 survey of about 200 university textbooks". IUPAC Group 3 project.; Scerri, ER (2020). The Periodic Table, Its Story and Its Significance (2 ed.). New York: Oxford University Press. p. 392.
^ Scerri, ER (2019). "Dec 2019 survey of about 200 university textbooks". IUPAC Group 3 project.; Scerri, ER (2020). The Periodic Table, Its Story and Its Significance (2 ed.). New York: Oxford University Press. p. 392.
^ William B. Jensen (1982). "The Positions of Lanthanum (Actinium) and Lutetium (Lawrencium) in the Periodic Table". J. Chem. Educ. 59 (8): 634–636. Bibcode:1982JChEd..59..634J. doi:10.1021/ed059p634.
^ Scerri, ER (2019). "Dec 2019 survey of about 200 university textbooks". IUPAC Group 3 project.
^ Scerri, ER (2020). The Periodic Table, Its Story and Its Significance (2 ed.). New York: Oxford University Press. pp. 377, 392.
^ Cotton (and team), S (2022). "A comparison of the structural chemistry of scandium, yttrium, lanthanum and lutetium: A contribution to the group 3 debate". Coordination Chemistry Reviews. 455. doi:10.1016/j.ccr.2021.214366.

Poll wording, proposed

Version 1

Here it is:

Given the factors listed below, should the default depiction of the periodic table shown in the lede of the periodic table article show group 3 as either [A] Sc-Y-La-Ac; or [B] Sc-Y-Lu-Lr?

In 2016 IUPAC began a project to make recommendations to IUPAC on the composition of group 3.
A 2021 provisional progress report from the project concluded there was no objective way of adjudicating in favour of either [A] or [B].
The report added "perhaps" group 3 "could" be shown as [B], since this would avoid a split d-block.
Published periodic tables most commonly show group 3 as [A], per the IUPAC survey results image.
Relevant policy includes WP:ENC, WP:NPOV, WP:DUE.

How does it look? --- Sandbh (talk) 02:34, 8 November 2022 (UTC)

@Sandbh: You are still omitting important information, in my opinion:

The fact that there is no objective reason to make a decision does not mean that it cannot be made on other grounds. IUPAC specifically notes in its report that the decision is one of convention, and actually argues that that makes it more important to make a decision. So your wording looks to me as if this fact downplays the significance of their decision, when that's not at all how the report interpreted it.
In 4 you are omitting the fact that 16 out of 33 (the figures for the 2010s) is not a majority. That is a bit hard to see from your graph too, because it does not give a guide-line for 50% (though it does for 40% and 60%). So, most periodic tables nowadays are not Sc-Y-La, and just saying Sc-Y-La is the most common is true but by no means the full story. There are other omissions that I could live with, but this is the one that I feel is most important, as omitting it leads to a situation where the information provided is true, but not the full story.

I also think that people do not need to be explicitly referred to relevant policy. It is a given.

So I would say that your opening statement still does not neutrally describe the situation. If a poll is to be done, then I think we should present all the facts, and let people make an informed decision. Double sharp (talk) 09:43, 8 November 2022 (UTC)

@Double sharp: Thanks. The challenge IMO is to keep the wording concise, and to not go into excessive detail. I’ll give some more thought to your feedback. Sandbh (talk) 09:04, 9 November 2022 (UTC)

@Double sharp: I agree with the sentiment of your first point, and’ll consider the wording. I’ll add a 50% line to the chart. I included the policy because at the previous poll it was seemingly not taken account of, especially WP:ENC. More later. Sandbh (talk) 10:00, 9 November 2022 (UTC)

Version 2

@Double sharp: Updated wording, below, for your thoughts. Sandbh (talk) 11:12, 9 November 2022 (UTC)

Given the factors listed below, should the default periodic table in the lede of the periodic table article show group 3 as either [A] Sc-Y-La-Ac; or [B] Sc-Y-Lu-Lr?

In 2016 IUPAC began a project to make recommendations to IUPAC on the composition of group 3.
A 2021 provisional progress report by the project found there was no objective way of adjudicating in favour of either [A] or [B], and added it was therefore important for IUPAC to decide.
The report added "perhaps" group 3 "could" be shown as [B], as this would avoid a split d-block.
IUPAC has made no decision about group 3; the majority of published periodic tables show group 3 as [A], per the IUPAC survey results image.
Relevant policy includes WP:ENC, WP:NPOV, WP:DUE.

@Sandbh: I still don't agree with #5, but that's a repeat.

For #4, I feel this must be put into context. IUPAC has not made a formal statement about how to draw the periodic table as a whole. Singling out group 3, in my view, gives the wrong impression about that. Moreover, the entire point of the project was to make an exception to this for group 3, because there is variation in the literature.

Version 3

Here's something I'd be more comfortable with, assuming the poll goes through at all. I think it is also better to state the three desiderata listed by IUPAC rather than listing just one. I think the addition also makes it clear that both options have been around for a while, and that it is not new.

IUPAC has not taken a formal position recommending any particular form of the periodic table.
The question of whether group 3 in the periodic table consists of [A] Sc-Y-La-Ac or [B] Sc-Y-Lu-Lr has been debated in the literature since at least 1921, and both forms can be found throughout the literature.
The matter was first brought before the IUPAC General Assembly in 1985, when an argument to change from [A] to [B] was presented. A following 1988 IUPAC report wrote that According to the electron configurations of the elements, the scandium group consists of the elements Sc, Y, Lu, Lr ... Based upon their electronic configurations and their chemical and physical properties, the elements La to Yb and Ac to No should be inserted between barium and lutetium and between radium and lawrencium or for practical reasons be listed at the bottom of the table. However, it was also stated that the following Red Book would show a compromise arrangement.
Because "students and instructors are typically puzzled" by the variation, IUPAC established a project in 2015 to deliver a recommendation on whether group 3 should be [A] or [B].
A 2021 provisional progress report by the project found there was no objective way of adjudicating in favour of either [A] or [B], and added that it was therefore important for IUPAC to make a ruling as a matter of convention.
The provisional report wrote "Perhaps a compromise could be reached on" [B], on the basis of three desiderata. First, it displays all the elements in order of increasing atomic number. Secondly, it avoids splitting the d-block into two highly uneven portions, and thirdly, it depicts all the blocks of the periodic table in accordance with the underlying quantum mechanical account of the periodic table which calls for 2, 6, 10 and 14 orbitals to occur in the extra-nuclear electron-shells.
In the course of the project, IUPAC surveyed periodic tables found in university textbooks from the 1970s to the 2010s. [A] was the most common form in each decade, but in the 2010s it was a plurality but not a majority.

Of course, I still doubt that there is a strong case to do a re-poll, because this amounts to repeating all the information that was already presented at last year's one. But I think that delivers pretty much all the relevant info. (In particular, the quote from 1988 is needed to justify the asterisks.) Double sharp (talk) 11:32, 9 November 2022 (UTC)

@Double sharp: Thanks. It strikes me as being too long, so I'll see if I can reduce the length. Subject to my concern about length, my first look impressions are:

Looks good.
OK
The reasoning given in the IUPAC General Assembly paper in 1985 is, IMO, seriously weak, flawed, unbalanced and challengeable. But I am not sure what to do about that (it is no wonder they instead adopted the compromise form). IUPACs' decision to "adopt" the compromise form is also contradicted by #1.
Looks good
Looks good
Looks good
Looks good, and needs the graph added.

On redoing the poll the case for doing so (as I mentioned) is that during the last poll, WP policy was seemingly not taken account of, especially WP:ENC, and WP:NPOV and WP:DUE.

On a related matter I don't intend to run the poll until the wording of the group 3 section of the periodic table article has been set aright.

I'll give all of this some further thought. Sandbh (talk) 23:09, 9 November 2022 (UTC)

@Sandbh: Re the apparent contradiction between #1 and #3, the IUPAC newsmagazine Chemistry International published an article in 2009 making it clear that IUPAC hasn't made any recommendations on the PT apart from numbering the groups from 1 to 18. Apparently displaying a so-called "IUPAC Periodic Table of the Elements" does not constitute a formal recommendation. Yes, this is weird, but it is also their clearly stated position.

The "adoption" of that compromise seems to have been a bit haphazard as well, since in the 1990 Red Book, the 8- and 18-column tables were Sc-Y-*-**, and the 32-column table was Sc-Y-Lu-Lr. Same mix from that 2009 article I linked above. There's even an 18-column Sc-Y-Lu-Lr table on p. 16 of the January 2019 issue of Chemistry International. That said, I haven't seen any Sc-Y-La-Ac from them; only Sc-Y-*-**, and less commonly Sc-Y-Lu-Lr. I note that 32-column tables get consistently illustrated as Sc-Y-Lu-Lr when they (rarely) appear in IUPAC publications. That might have something to do with how authors seem to be unsure about what exactly Sc-Y-*-** is saying. Some seem to think it means "all the Ln and An go in group 3 under Y", and others seem to think it means "group 3 has Sc and Y, and it stops there". In the former case this entails expanding the widths of the Sc and Y cells to 15 cells wide, and all this awkwardness perhaps explains why they do something else in 32-column.

Exactly how strong the reasoning in 1985 was is not relevant IMHO. It's not our job to evaluate it. The point is that it was presented to IUPAC and so we know from that that IUPAC has been aware of the group 3 situation since at least then. I think that's an important date in the chronology. Double sharp (talk) 23:57, 9 November 2022 (UTC)

Version 4

@Double sharp:, thanks. I edited your proposed words with the following result, for your feedback:

Given the factors listed below, should the default periodic table (PT) in the lede of the PT article show group 3 as either [A] Sc-Y-La-Ac; or [B] Sc-Y-Lu-Lr?

1. IUPAC has not recommended any form of PT.

2. Whether group 3 should be [A] Sc-Y-La-Ac or [B] Sc-Y-Lu-Lr has been debated since at least 1921; both variations occur in the literature.

3. As "students and instructors are typically puzzled" by this variation, IUPAC started a project in 2015 to give a recommendation to IUPAC on whether group 3 should be [A] or [B].

4. A 2021 provisional progress report by the project found there was no objective way of adjudicating in favour of either [A] or [B], and added it was therefore important for IUPAC to make a recommendation, which in the final analysis is one of convention rather than one that can be decided on objective scientific grounds. The report added that:

Perhaps a compromise could be reached on" [B], on the basis of three desiderata. First, it displays all the elements in order of increasing atomic number. Secondly, it avoids splitting the d-block into two highly uneven portions, and thirdly, it depicts all the blocks of the periodic table in accordance with the underlying quantum mechanical account of the periodic table which calls for 2, 6, 10 and 14 orbitals to occur in the extra-nuclear electron-shells.

5. As part of their work the project team surveyed PTs in university textbooks from the 1970s to the 2010s. [A] was the most common form in each decade, as seen in the accompanying chart, but in the 2010s it was a plurality but not a majority.

6. Relevant WP policy considerations include WP:ENC, WP:NPOV, WP:DUE.

I removed your #3 (argument presented to IUPAC General Assembly 37 years ago) in light of my #4 above re the project team finding no objective basis to adjudicate. On this point, the project description says:

"Various forms of evidence have been put forward in support of each version of group 3...The task force will aim to evaluate all this evidence [underline added] in order to reach a conclusion that encompasses these different approaches."

Given this, none of the evidence-based arguments merit mentioning in the proposed poll wording. Sandbh (talk) 05:55, 10 November 2022 (UTC)

@Sandbh: I generally agree with this, but like I said, it's not about 1985 being an "evidence-based argument". It's about IUPAC having been aware of this issue and having said things since earlier than 2016. It seems your objection is to 1985, not 1988 (which doesn't give a specific argument and is the one that most clearly states how the footnote is supposed to work): so how about this?

1. IUPAC has not recommended any form of PT.

2. Whether group 3 should be [A] Sc-Y-La-Ac or [B] Sc-Y-Lu-Lr has been debated since at least 1921; both variations occur in the literature.

3. A 1988 IUPAC report wrote that According to the electron configurations of the elements, the scandium group consists of the elements Sc, Y, Lu, Lr ... Based upon their electronic configurations and their chemical and physical properties, the elements La to Yb and Ac to No should be inserted between barium and lutetium and between radium and lawrencium or for practical reasons be listed at the bottom of the table. However, it was also stated that the following Red Book would show a compromise arrangement.

4. As "students and instructors are typically puzzled" by this variation, IUPAC started a project in 2015 to give a recommendation to IUPAC on whether group 3 should be [A] or [B].

5. A 2021 provisional progress report by the project found there was no objective way of adjudicating in favour of either [A] or [B], and added it was therefore important for IUPAC to make a recommendation, which in the final analysis is one of convention rather than one that can be decided on objective scientific grounds. The report added that:

Perhaps a compromise could be reached on" [B], on the basis of three desiderata. First, it displays all the elements in order of increasing atomic number. Secondly, it avoids splitting the d-block into two highly uneven portions, and thirdly, it depicts all the blocks of the periodic table in accordance with the underlying quantum mechanical account of the periodic table which calls for 2, 6, 10 and 14 orbitals to occur in the extra-nuclear electron-shells.

6. As part of their work the project team surveyed PTs in university textbooks from the 1970s to the 2010s. [A] was the most common form in each decade, as seen in the accompanying chart, but in the 2010s it was a plurality but not a majority.

7. Relevant WP policy considerations include WP:ENC, WP:NPOV, WP:DUE.

(As you know, I don't think the last bullet is necessary, but I won't object if it stays.) Double sharp (talk) 11:16, 10 November 2022 (UTC)

@Double sharp: I'm OK with all bar #3, which I suggest trimming and wording as follows:

"3. A 1988 IUPAC report on the background to the 1-18 group numbering scheme briefly mentioned the composition of group 3. It said the group should be Sc-Y-Lu-Lr, citing several sources. The report lacked objectivity since it did not cite any sources supporting Sc-Y-La-Ac. In the event, the next edition of the IUPAC Red Book showed the Sc-Y-*-** form."

More specifically:

In about 3,000 words the 1988 report details the history and thinking behind the decision to adopt the 1-18 group numbering system, yet the composition of group 3 gets just 240 words without any details.
Indeed, the title of the article, "New Notations in the Periodic Table", gives no clue that the Sc-Y-La-Ac v Sc-Y-Lu-Lr issue will be mentioned in passing.
It is unbalanced and lacks objectivity since it only gives Lu-Lr citations, including L & L (1959) whose argument is unclear; Christyakow (1968) whom Scerri criticised; Luder (1970) who bases his argument on differentiating electron (d/e) considerations, overlooking that fact that an La-Ac table has one less d/e discrepancy overall; and Jensen (1982) whom Scerri described as selective. Even Jensen (1982) wrote, "The current consensus...places lanthanum and actinium in this position".
Since Fluck seems to have been familiar with the non-English literature what about e.g. Trifonov (1970) or Shchukarev (1974); or in the English, Dash (1967)?
Compounding the lack of objectivity, Fluck cites one his own articles in support of Sc-Y-Lu-Lr.

All that said, I feel that the suggested rewording will suffice. Sandbh (talk) 23:38, 13 November 2022 (UTC)

@Sandbh: I think we're almost at an agreement. I still feel like saying "The report lacked objectivity" runs a bit too close to editorialising for my taste, and I'd rather just give a dispassionate description. Also, the next edition of the IUPAC Red Book showed both Sc-Y-*-** (p. 282) and Sc-Y-Lu-Lr on facing pages (p. 283).

So how's this for you?

"3. A 1988 IUPAC report on the background to the 1-18 group numbering scheme briefly mentioned the composition of group 3. It said the group should be Sc-Y-Lu-Lr, citing several sources. The report did not cite any sources supporting Sc-Y-La-Ac. In the event, the next edition of the IUPAC Red Book showed both the Sc-Y-*-** and Sc-Y-Lu-Lr forms."

Double sharp (talk) 23:44, 13 November 2022 (UTC)

P.S. It might not be in the exact letter of WP:COI, but since WP:SELFCITE goes to a section on that page, it seems to me to be in its spirit. So, for full transparency (since you've referred to your own paper on this page), I think you might want to consider disclosing in your proposed statement that you recently published a paper in support of the Sc-Y-La-Ac option. Double sharp (talk) 21:18, 14 November 2022 (UTC)

Double sharp: I tuned your good words a little:

"3. A 1988 IUPAC report on the background to the 1-18 group numbering scheme briefly mentioned the composition of group 3. It said the group should be Sc-Y-Lu-Lr, citing several sources in favour of this option but not citing any sources supporting Sc-Y-La-Ac. In the event, the next edition of the IUPAC Red Book showed the 18-column Sc-Y-*-** form on its inside cover, and showed in the appendix 8- and 18-column Sc-Y-*-** forms and a 32-column Sc-Y-Lu-Lr form.

On COI, if we do that then I feel we should flag that the chairperson of the IUPAC Group 3 project has written extensively in favour of Sc-Y-Lu-Lr. Sandbh (talk) 05:51, 15 November 2022 (UTC)

@Sandbh: I just found new information presented publicly by Scerri, that I've summarised at #News from Scerri (2022), that has bearings on what comes after the 2021 report. So I suggest we discuss there. Double sharp (talk) 20:55, 15 November 2022 (UTC)

Concluding: subthread closed

As of 13:05, 1 December 2022 (UTC), this thread is closed for further implementation. In general, the Group 3 topic has concluded, and that includes this poll venue, generally as "no major changes". Apart from this Talkpage thread, discussion also took place at User talk:Sandbh (version as of now; concluding thread 26Nov2022). -DePiep (talk) 13:05, 1 December 2022 (UTC)

Belated response to DS

Double sharp, my belated response to your earlier comments [2] follows. I have placed it here, together with your comments, in order to maintain chronological order.

DS: "So it seems the "going out for consultation" must have already started."

No, the only consultation is on the provisional report. There has been no consultation on an IUPAC cleared final report.

DS: "It is a cited statement in this article that the majority of work in the literature specifically on this subject supports Lu in group 3, not La. Cited to Jensen, a member of the IUPAC project as well: “much of the literature that has appeared on this subject in the 32 years since I published the initial paper has in fact tended to support the proposed reassignment of Lu” (paper)."

The question is not how much of the literature has supported Lu. It is instead the calibre of the arguments. But that is neither here nor there since the IUPAC project, of whom Jensen is a member, concluded that there was no objective basis to decide between the two options.

DS: "And that quite agrees with what I found checking it myself (User:Double sharp/Group 3 sources, though I haven't updated it well since 2021 when the IUPAC report came out). Naturally we cannot expect unanimity, but there's clearly a strong trend towards support for Lu under Y among sources focusing on this issue. The "literature survey" you link to from 2019 is a survey of textbooks, and I think I already explained why that is a different situation. We do not say SF6 has 3d hybridisation either, right?"

SF6 is a separate topic that is not relevant to this discussion given such hybridisation is (reasonably) well-recognised as a textbook error.

DS: "Cotton (1) is not a member of the IUPAC project specifically meant to give a recommendation and (2) does not seem consistent about whether he wants La under Y or * under Y in his paper: he writes both “the case for replacing La by Lu in Group 3 has not been made on structural grounds”, suggesting he wants to keep La there, and “the Periodic Table as presented currently, not least in the IUPAC form, gives the most appropriate description of the chemistry of Group 3 and of the lanthanides”, suggesting he wants to keep all the lanthanides there. I agree he is an expert on lanthanide chemistry, but I don't see how what he's written can be used as a good argument when he contradicts himself on what his preferred form actually is. BTW, I find his logic very strange here: “The structural chemistry of Lu most closely matches that of Y and not of La, so we feel that the case for replacing La by Lu in Group 3 has not been made on structural grounds.”"

In the case of Cotton I merely wished to note that arguments in support of La continue to appear in the literature.

Nevertheless I note Cotton et al. write, as we all know: "Traditionally, scandium, yttrium and lanthanum have been considered as a triad due to their occupation of successive vertically arranged boxes in Group 3 of the Periodic Table."

What Cotton et al. write is straightforward and not contradictory.

First, they write: "In general, the size of successive atoms increases as groups are descended in the Periodic Table , although with heavier elements, including La - Lu, relativistic effects counterbalance this to some extent [11,12], and it is remarkable that atomic radii vary by only a factor of approximately two over the whole Table. Nonetheless, yttrium is larger than scandium, and lanthanum is larger than yttrium, but in the elements immediately following lanthanum this trend is reversed.”

Second, they write: "So, in the discussion of whether the order of the elements for Group 3 should be Sc, Y and La, or whether Lu should replace La in the group, this detailed review of the structural chemistry indicates that there is a logical progression in chemistry as the size of the Sc, Y and La ions increase. The structural chemistry of Lu most closely matches that of Y and not of La, so we feel that the case for replacing La by Lu in Group 3 has not been made on structural grounds.”

That is to say, they recognise the validity of the increase in size going down group 3. Since Lu is impacted by the Ln contraction, starting at Ce and finishing at Lu, the latter does not fit into group 3.

Thirdly, they conclude, "In the absence of more compelling evidence, the Periodic Table as presented currently, not least in the IUPAC form, gives the most appropriate description of the chemistry of Group 3 and of the lanthanides.”

Note they use the expression “not the least”. So they are saying the IUPAC form (at the least) or the traditional La form give an appropriate description of the chemistry of Group 3 and the Ln.

DS: "P.S. You didn't seem to mind it being Sc-Y-Lu-Lr when somebody asked about it on this talk page last year." :) Double sharp (talk) 09:37, 16 October 2022 (UTC)

Your archive fu is impressive.

Yes, I’m equivocal about the group 3 question but strive to be principled about what WP, as an encyclopaedia, should show. Sandbh (talk) 10:52, 18 October 2022 (UTC)

Yet last year you used your own OR argument to support Sc-Y-Lu on the talk page. Double sharp (talk) 15:37, 30 October 2022 (UTC)

@Double sharp: Yes; thanks for the reminder. At that time it was an impressive argument based on a rubric that only applies to the f-block. As we have discussed, the other argument concerns the electron configurations in the trivalent cations, where regularity is better with group 3 as Sc-Y-La-Ac. Further, as Wulfsberg has observed, the configurations of the ions are more important than the gas phase configurations. Sandbh (talk) 23:29, 7 November 2022 (UTC)

@Sandbh: But the point is that that argument was OR, which seems inconsistent with being, as you say, "principled about what WP, as an encyclopaedia, should show". We clearly disagree about what is most appropriate to show on WP, but surely we should be arguing based on sources?

But let me reply to the ion argument. The thing is, cations inevitably attract some electron density from the surrounding chemical environment. Greenwood and Earnshaw already explain this on p. 80 for the alkali metals: The compounds which most nearly fit the classical conception of ionic bonding are the alkali metal halides. However, even here, one must ask to what extent it is reasonable to maintain that positively charged cations M⁺ with favourably directed vacant p orbitals remain uncoordinated by the surrounding anionic ligands X⁻ to form extended (bridged) complexes. Such interaction would be expected to increase from Cs⁺ to Li⁺ and from F⁻ to I⁻ (why?) and would place some electron density between the cation and anion. ... Data from Li⁺ are shown in Fig. 4.2a from which it is clear that the Li⁺ ion is no longer spherical and that the electron density, while it falls to a low value between the ions, does not beome zero. ... It may be significant that adoption of the NaCl structure by all these compounds maximizes the p orbital overlap along the orthogonal x, y, and z-directions, and so favours molecular orbital formation in these directions. Naturally for higher-charged Ln³⁺ we can expect even more of that, and Xu et al. investigated it in 2013 for LnF₃. They found a constant dative F⁻(2p) bonding into the Ln(5d) shells, and for the early lanthanides (including La) even into the Ln(4f) shell. Per table 3, "La³⁺" in LaF₃ was found to have a +1.77 charge and population +0.34 of f and f-polarisation functions (vs insignificant +0.00 for "Cs⁺" in CsF and +0.08 for "Ba²⁺" in BaF₂). Noting the high electronegativity of F, this means that in a real chemical environment, Ln³⁺ is too high a charge, and that "La³⁺ 4f⁰" is an oversimplification for a real environment. As G&E is a standard text, presumably this is why it's more common to argue based on neutral atoms than cations, since cations in a real environment will have a smaller charge than the ionic model would suggest. I have no idea what Wulfsberg has in mind, unless maybe it's Ln²⁺ as a more plausible real charge based on such analyses (indeed most Madelung anomalies vanish once you get down to +2). Double sharp (talk) 23:49, 7 November 2022 (UTC)

@Double sharp: Thanks. Yes, my argument was OR, which I understand we are able to engage in on a talk page; as OR it didn't make it into the article space. Actually, on the OR question, it's no more than simple fact-based arithmetic so there's an argument it would be OK somewhere in article space.

The ion oxidation state argument is a first order argument based on the ubiquity and remarkable usefulness of the oxidation number concept in chemistry. Per G&E it is of course possible to consider the nuances of higher order models.

BTW, I found this article and this article discussing Pt⁺¹⁰. And this one claiming Cs⁺⁹. Sandbh (talk) 01:33, 8 November 2022 (UTC)

@Sandbh: It is more than simple fact-based arithmetic. You are referring to oxidation numbers, but you then quote configurations of ions with that number as a charge. The very article you link to says that oxidation number and charge are not the same thing and that for that reason, oxidation states are sometimes called formal oxidation states: Hence, the oft-encountered term, formal oxidation state, and OS is often represented using Roman numerals (I, II, III etc.) to distinguish it from a real charge for which Arabic numerals (1, 2, 3 etc.) are employed. So it is not just WP:CALC: you are taking an extra step that requires RS support and, worse, seems to directly contradict the RS you are using to support it.

The talk page is meant as a forum aimed at article improvement, not airing personal views. Articles must be based on RS, so OR cannot be used to improve the article. This discussion is even precisely exemplifying why not by showing OR in contradiction with the RS used to back it up. Therefore, there doesn't seem to be any point to discussing it on the talk page.

The sources in your last paragraph are appreciated in my capacity as somebody who is interested in Cs chemistry. However, in my capacity as somebody who is discussing how this article, on WP, should handle group 3, which neither Cs nor Pt is in, I don't see the relevance. Double sharp (talk) 09:55, 8 November 2022 (UTC)

@Double sharp: Thanks for the ping. Yes, oxidation numbers and charge numbers of ions are interchangeable in the literature, at the first order level. This is what Simon Cotton does when he gives the electron configurations and oxidation numbers of Ln⁺³ and An⁺³ cations. Yes, it is sometimes useful, to distinguish between the two, at a higher order of modelling, in order to further appreciate the fine nuances of a situation, as I acknowledged, and as that source I cited does. None of this is a personal view; it is what occurs in the literature. I mentioned the two other sources, in passing, out of incidental courtesy, since the topic of the thread was oxidation numbers. Sandbh (talk) 06:57, 10 November 2022 (UTC)

@Sandbh: I agree that this is the case to a first order level, as you say. However, it is not clear to me that it is being used at that first order level when it is brought into the group 3 dispute. You've mentioned Wulfsberg, who indeed says that "ions are more important than isolated gaseous atoms for nearly all atoms". But in the very same paragraph he supports Sc-Y-Lu-Lr as group 3. Or you can look at Jørgensen (p. 228), who has said that Madelung anomalies are irrelevant and points to the fact that nearly all the anomalies disappear in cations M²⁺ through M⁶⁺ (so once again he's looking at ions rather than gaseous atoms). But he also thought that Lu is most conveniently considered as the first 5d element. That seems to indicate to me that it's not obvious or generally accepted that it's enough to stop at the first-order "La³⁺ 4f⁰" idea. Because as you've noted, stopping at the first order would seem to naturally support Sc-Y-La-Ac, but they mention it and then support Sc-Y-Lu-Lr anyway. That cannot involve stopping at the first order.

As I mentioned, your sources are appreciated. That said, I'd like to request that such links that are not quite relevant to an ongoing conversation are put on my user talk page in future. Not only does that keep the conversation more on-topic, but it also helps me to find those sources later when I need them. :) Double sharp (talk) 16:11, 10 November 2022 (UTC)

@Double sharp: Thanks. We noted in our joint 2017 submission to the IUPAC group 3 project the inexplicable discrepancy between what Wulfsberg said on the one hand about the importance of ions, and on the other hand, his depiction of group 3. Wulfsberg cites Cherkesov (1984); Jensen (1982); Chistyakov (1968); Merz & Ulmer (1967); and Hamilton (1965). Jensen's article was criticised by Scerri as being selective. Chistyakov was subject to criticism from Scerri for only looking at the second IE. We too criticised Chistyakov for only comparing Sc-Y-Lu to the trends in groups 4-8 and ignoring what was happening in groups 1−2. Merz and Ulmer were contradicted by Bergwall (1966). Hamilton (inconclusively) said, "There are at least several intermetallic compounds where the compound with La has a different crystal structure from the corresponding compounds with Sc, Y, and Lu." This is not entirely unexpected. Given its position in the next row down in the periodic table, lanthanum atoms are larger than scandium or yttrium atoms. It would therefore stand to reason that they should not be able to pack in a crystal structure in quite the same way as the smaller scandium, yttrium, or lutetium atoms. He also relied on excited state spectra, completely overlooking what happens in group 2, with Ca-Sr-Ba.

Of course, Chistyakov; Merz & Ulmer; and Hamilton were cited by Jensen, so why Wulfsberg felt the need to recite them is inexplicable.

The only new citation Wulfsberg gave was Cherkesov (1984), who says that Al-Sc-Y-Lu and Lr are close analogues, but so too are Y-La-Ac. He concludes by saying there is no reason why Lu-Lr could not go under Sc. I'll ping you a copy of this article for your assessment.

Jørgensen neither provides a PT showing the composition of group 3 nor any grounds, that I could find, for saying Lu can most conveniently be considered as the first member of the 5d series. Sandbh (talk) 07:20, 13 November 2022 (UTC)

@Sandbh: I'll respond at your talk page. Double sharp (talk) 13:41, 13 November 2022 (UTC)

Fringe

In this thread, and before & elsewhere, it has been pointed out that Sandbh is applying sloppy reasoning. It keeps touching WP:OWN, WP:FRINGE, !votecounting (both in-wiki and superfluous tallying). At the moment, WP:SELFPUB has been added to the article [3][4], well before this thread had conclused and while it was actually discussing the very topic, e.g. by Double sharp. (skipping current ref #29Cao for now). Also, the editor/author repeatedly declares to "clarify", correct the text, while no consensus has been reached about the general encyclopedic building block. The base reasoning behind it/in it, both in talkpage and in article(!), is based on fog, obfuscation, deviation, thread-reconstuction, incomplete reading, and more. Not fleshed out, not main discourse (more like an evolving paper in preparation). Also long-term drip-drip changing aka gaslighting. None or few theads lead to a stable, encyclopedic, readable intro-arguments-conclusion paragraph part. For starters, we will have to remove the SELFPUBLISH promotions. Altogether, it leads to fringe, own, selfpub article parts, which do not deserve the encyclopedia.

-DePiep (talk) 11:45, 24 October 2022 (UTC) (late sign, so reping: Sandbh Double sharp)

I don't object to the article addition, as while Sandbh is indeed citing himself, that article is in a peer-reviewed journal. But if others object, I suggest citing Lavelle for a Sc-Y-La-Ac argument, since he is a IUPAC project member. Double sharp (talk) 12:12, 24 October 2022 (UTC)

Yes, it seems fine as per WP:SELFCITE. Mike Turnbull (talk) 15:20, 24 October 2022 (UTC)

@Double sharp and Michael D. Turnbull: Thank you. — Sandbh (talk) 03:32, 25 October 2022 (UTC)

It's wider. @Double sharp and Michael D. Turnbull: As for WP:SELFPUB as isolated issue here, it states if the information in question is suitable for inclusion, someone else will probably have published it in independent, reliable sources. Obviously, such mainstreaming is not the case. Then, such selfpub as a reply to the 2021 IUPAC (Scerri) report is a red flag for WP:UNDUE weight. It does not establish (quite the opposite so) that the furious debate is stable fact worth inclusion. This for the single edit of self-citing.

I note that the whole wider paragraph and environment (statements, claims, issue approach) is based on the Sandbh-self. This includes casting doubt & challenging on the Scerri report, in the (scientifically/encyclopedically) immature way as is. Inserting the self-published vision re Group 3 already started way before the Scerry 2021 report was published Vernon (2020). So, if that position were noteworthy, Scerry (2021) would have incorporated it in some way; since that is not the case, the theory has a duty to prove itself to be non-fringe still. IOW: not in Scerry 2021, so not relevant enough, and no ground to add it later (e.g., in here) on as such.

Further, this OP—as told by section title Fringe—is about a wider range than just a selfquote (actually, two then). That the topic "Composition of Group 3" is not the same as disputing the Scerry report, as is introduced here. I already mentioned their talkpage-relation (e.g. switching frivolously between an ongoing, inconcluded talkpage discussion and article edits; frivolously forking discussion inconsistently and confusingly, giving rise to cherry-picked arguments), WP:FRINGE, WP:UNDUEWEIGHT (details that better belong in Group 3 article, which is lagging behind or differing btw). We should also consider the drip-drip editing by Sandbh, & over various articles—notably but noty solely in nonmetal, instead of single-edits. The total picture does not look great, and not encyclopedically. Thereby the once-stable article is degraded away from its FA-checks by these imbalances.

So, let's reconsider the wider picture of edits instead of sungle-dripped ones, and whether they are an improvement, encyclopedically, that belong in there. -DePiep (talk) 11:43, 29 October 2022 (UTC)

One more note on talkpage etiquette. Since I explicitly address issues with talkflows &tc from OP onwards, I have little patience with evasive reply forms. As usual, if something is unclear, the best form is to ask clarification first instead of going offroad. -DePiep (talk) 11:43, 29 October 2022 (UTC)

@DePiep I think you are confusing WP:SELFPUB with WP:SELFCITE. The former is frowned upon as sourcing in Wikipedia articles while the latter is usually fine (as in this case). doi:10.1007/s10698-020-09384-2 was published in Foundations of Chemistry which is a peer-reviewed journal that happens to have Scerri as its editor-in-chief. In my (admittedly non-expert in this topic) opinion that article is not WP:FRINGE since it is a review article of other people's work, nor is it WP:UNDUE to include it as a recent analysis of this controversial topic. If the composition of Group 3 was agreed by all but a tiny minority on the fringe, IUPAC would have had no need to set up their committee to investigate the issue. Mike Turnbull (talk) 13:08, 29 October 2022 (UTC)

@Michael D. Turnbull: Thank you. — Sandbh (talk) 06:57, 30 October 2022 (UTC)

Consider using H:THANKS. Unhelpful Talkpage flow is an issue. DePiep (talk) 08:00, 30 October 2022 (UTC)

Ha Ha, the irony. Polyamorph (talk) 08:08, 30 October 2022 (UTC)

Since this has been deemed "unhelpful, complaining about interrupting the flow of a talk page by *checks notes* interrupting the flow of talk, is mildly amusing. As you were. Polyamorph (talk) 08:24, 30 October 2022 (UTC)

Incomprehensible, and still not explained how this would help the Talk. To ignore. DePiep (talk) 10:59, 30 October 2022 (UTC)

You are very rude! Polyamorph (talk) 12:13, 30 October 2022 (UTC)

After doing some further research, I think DePiep has a point here. Vernon (2020) is not just summarising other people's arguments. The ten arguments he lists for Sc-Y-La are stated as being the author's arguments: My arguments are presented as an interlocking progression of concepts and relationships. So they are new contributions to the literature. (An exception might be IX which is based on an argument by Shchukarev, but it appears to be augmented with Vernon's interpretation.) In that case I feel that in this overview article, the arguments that should be listed should rather be less recent and more influential (an article published two years ago hardly has time to get cited as much), and have a somewhat greater connexion to the body (IUPAC) that was supposed to issue a statement on this problem. As far as I can see this has neither. (Scerri has cited it as [35] of this article of his, but not on anything related to group 3.) I would therefore prefer to replace it with Lavelle's argument (that Sc-Y-La is better because neither La nor Ac actually have an f-electron in their ground-state gas-phase configurations), as he is a member of that IUPAC project. In fact, that's maybe even a better option because Lavelle, Jensen, and Scerri (all on the IUPAC project) then proceeded to write some rebuttals in the pages of J. Chem. Ed. and Found. Chem. in 2009 and 2015, providing a microcosm of the dispute that is directly connected to the IUPAC group 3 project. And it's also the way Geoff Rayner-Canham describes it in his recent book The Periodic Table: Past, Present, and Future (2020), mostly following the Lavelle–Jensen–Scerri debate; I feel the existence of this secondary source supports the idea to do it this way. Double sharp (talk) 14:13, 30 October 2022 (UTC)

@Michael D. Turnbull: [06:57]: You seem to think that I am disputing the sources, as not being RS? That's not the case.

And no, I am not confusing these two SELF-guidelines. I quoted "if the information in ..." directly from WP:SELFPUB, and by itself that quote describies a reasoning for scrutiny. I add: from the same wlink: if the information in question is suitable for inclusion, someone else will probably have published it in independent, reliable sources applies. And also, that is: next to SELFPUB, WP:SELFCITE applies. Both are an extension on WP:RS, being careful.

About you writing peer-reviewed journal .. [again, I am not disputing "RS" by itself] .. that happens to have Scerri as its editor-in-chief. Right! So, Scerry already in 2020 knew about Vernons peer-reviewed claims (of course, he would have known anyway without being the editor). So, these must have been digested by the IUPAC commission on Group 3! Obviously, it did not change their IUPAC (2021) paper. Full stop. Vernons paper & statements reduced to lower importance.

Then recently, the same author starts editing their —already and explicitly rejected— scientific, peer-reviewed WP:POINTs into this wiki. Multiple uppercases apply, such as the SELFs, and I can repeat the title we are typing under: WP:FRINGE.

All in all, author/WPeditor Sandbh is pushing and pulling the IUPAC (2021) report here, cherry picking wordings etc. to get a wiggle in. That is the wider issue with these Group 3 edits. And, it leaves the section is a sub-FA-standard state (& illegible, rambling logic IMO), irrespective of the number of "but it is in a RS"'s in there. @Sandbh and Double sharp: DePiep (talk) 12:00, 9 November 2022 (UTC)

I've replaced Vernon with a summary of the Lavelle–Jensen–Scerri debate, based on what Rayner-Canham presents. Double sharp (talk) 18:01, 30 October 2022 (UTC)

It seems from looking at some sources that "fringeness" depends on whether you're talking about the 18-column or the 32-column table. Geoff Rayner-Canham in his recent book The Periodic Table: Past, Present, and Future (2020) says that in the 32-column form, there is no issue (p. 72), and Lu and Lr fall naturally in place into group 3; it's only in the 18-column form, according to him, that there is an issue. (In his book he uses the 18-column form and Sc-Y-*, except on the cover where he uses Sc-Y-Lu.) Based on this, indeed it would be true that "the composition of Group 3 was agreed by all but a tiny minority on the fringe" (quoting Mike Turnbull above or rather to reuse his wording to describe that hypothetical for the 18-column form, and certainly not intending to imply that he believes that's true there Double sharp (talk) 00:17, 4 November 2022 (UTC)), with Sc-Y-Lu being the typically agreed composition – but only so far as the 32-column table is concerned. I was surprised by this, since I always thought they were supposed to be the same thing with or without some cut-and-pasting, but it matches how IUPAC's 1990 Red Book is willing to use its compromise Sc-Y-* for the 18-column form, but not for the 32-column form, where it uses Sc-Y-Lu. The same was done in a 2009 article in IUPAC's newsmagazine Chemistry International. A brief sampling of the sources arguing about this seems to concur, in that most arguments showing a form show the 18-column one, unless the 32-column one is being brought in to make a point based on the belief that they're supposed to be the same thing, like the IUPAC 2021 report does. But that might be confounded by the fact that the 18-column form is a lot more common to see than the 32-column one. Anyway, this is an RS, and it seems to accord with what I've seen, so I've added a sentence. (Rayner-Canham's book received a favourable review from Scerri, chair of the IUPAC project.)

That may be of interest since we tend to use both forms interchangeably, per what IUPAC 2021 seems to have in mind. I've added Rayner-Canham's point to the article in one sentence. Double sharp (talk) 15:35, 30 October 2022 (UTC)

@Double sharp: Mike Turnbull has been quoted absent of context. He certainly did not imply that "the composition of Group 3 was agreed by all but a tiny minority on the fringe". Rather, what he wrote was:

"If the composition of Group 3 was agreed by all but a tiny minority on the fringe, IUPAC would have had no need to set up their committee to investigate the issue."

Rayner-Canham's "opinion" that in the 32-column form there is no issue and that group 3 naturally falls into place as Sc-Y-La-Ac, is contradicted by the literature.

1. A referee who reviewed a recent OUP book proposal of mine, a week or so ago, referred to:

"...the standard classical publication by John W Van Spronsen. The Periodic System. A History of the First Hundred Years (1969)."

Here, the inside cover of Van Spronsen features a 32 column table with group 3 shown as Sc-Y-La-Ac.

2. Professor Sir Martyn Poliakoff's take on this matter, at the start of his YouTube presentation shows what he refers to as an "ultra-modern" periodic table, 32-column form, with group 3 as Sc-Y-La-Ac.

3. Hill & Holman (2011, p. 158), in their chapter on "The periodic table and periodicity", show "The modern periodic table (wide form)", with 32-columns, and group 3 as Sc-Y-La-Ac.

Hill G & Holman J 2011, Chemistry in Context, 6th ed., Nelsom Thornes, Cheltenham

Your original line of thinking, "I always thought they were supposed to be the same thing with or without some cut-and-pasting" is on the mark.

--- Sandbh (talk) 23:58, 3 November 2022 (UTC)

@Sandbh: The point of my quote of Mike Turnbull was to refer to the hypothetical situation he mentioned as being possibly real for the 32-column table, but yes, I can see how it looks like quoting out of context. My apologies, and I've added some small text to clarify my intention.

1969 doesn't seem too relevant to me, since most of the discourse on the debate was later than that. Do you have any more sources than 2 or 3? I'd like to see a few more than that before I'm persuaded that Rayner-Canham (surely an RS himself) is wrong. Two could be a few mavericks: a few more books or articles showing a 32-column La-Ac table, preferably published this century, would persuade me. Double sharp (talk) 00:15, 4 November 2022 (UTC)

@Double sharp and Sandbh: Deducting that a switch between 18-column and 32-column PT by itself is a change in statement re group 3 is incorrect, and OR. As we have noted thousands of times on these pages, the table graph itself states the group 3 constitution. When GRC writes [only] in the 32-column form, there is no issue (above), then they have used wrongly-drawn or scientifically different tables. It's simple: if you can't draw your claim in an 18-column PT, then don't use an 18-column PT. (Mendeleev never said: "My 7-column PT proves that X cannot not exist". Seaborg never said "All my PTs are in Roman font, sometines in italics, so the primordial elements stem from Italy"). Really, editorial/drawing choices & mistakes do not make an argument.

Whatever statement GRC or others pull from the graphs: if it is not based in the paper text, the statement is useless. Your PT is drawn wrong or meningless wrt the point made. Here, it still is about: what is GRC or any other scientist actually stating wrt group 3, not wrt the font choice. DePiep (talk) 12:22, 5 November 2022 (UTC)

@DePiep: GRC wrote: The first two members are self-evident: scandium and yttrium. But which are the subsequent members of the Group? Using the long (32-column) form of the Periodic Table, there is no issue: lutetium and lawrencium fall naturally in place. That doesn't sound very far from the IUPAC report's argument, which itself starts by expanding the table into 32 columns: Sc-Y-Lu-Lr meets the desiderata, which could reasonably be said to be "natural" (to use GRC's word), whereas Sc-Y-La-Ac requires either breaking the Z order or breaking up the d-block. And later he even explains that by paraphrasing Scerri: Scerri pointed out that if the long form of the Periodic Table was written out, then keeping the f- and d-blocks intact, lutetium and lawrencium would naturally occur beneath scandium and yttrium. The Sc–Y–La–Ac alternative could only be structured if, in the long form, the d-block elements of Group 3 (scandium and yttrium) were shifted over next to Group 2, breaking up the transition metal series. (GRC p. 77).

So I suspect what GRC means is that among authors who show the 18-column table, there is a dispute: he wrote However, the 18-column form of the Periodic Table is ubiquitous. The issue becomes which set of 14 elements are pulled out and placed beneath, or whether all 15 elements are placed beneath. But among authors who show the 32-column table, the solution is obvious, and so there isn't a dispute. This accords with what Scerri suggested in 2012: Scerri wrote I suggest that any reluctance to accept this grouping [Sc-Y-Lu-Lr] as opposed to the more frequently seen grouping of Sc, Y, La, and Ac (as shown in figure 5) stems entirely from a reluctance to display the periodic table in its 32-column format. It would be nice to see some more 32-column tables to verify what GRC wrote, but I don't insist on it, since GRC is a RS. Double sharp (talk) 19:43, 5 November 2022 (UTC)

(I admit that GRC shows different forms of group 3 throughout his book. But he's explicitly discussing, comparing, and contrasting the merits of many different forms of the table, so he's probably not being confused.) Double sharp (talk) 19:41, 5 November 2022 (UTC)

The constitution of group 3 is not based on the author's choice between 18- or 32-column periodic table.

That doesn't sound very far from the IUPAC report's argument [Ds]— is a declaration of OR. Also, mxing this with "(to use GRC's word)" [Ds] is manipulating the sources into your POV aka WP:SYNTHESIS.

IUPAC 2021 is not making any claim on "18- column is an argument for claim XYZ".

The word "natural" that keeps popping up: not an argument. If it is not elaborated, it is unusable. (Would you accept that "Well, He in group 2 makes a nice natural rectagle so ..."??). For example, an existing RL RS argument is that the Z order should not be broken. That is evaluated, but not by saying "hard to draw so not true".

You are to prove that the way of drawing the PT is a scientific argument for the group 3 constitution.

The issue is wider than this point. For example, why these elaborations in this article, and not in its natural home Group 3? Why not reread the paragraph//.section on being encyclopediacally right? I maintain that the text is going offtrack into sidetalk and hypothesis, poorly sourced and now with OR, SYNTHESIS, and more.

DePiep (talk) 11:01, 6 November 2022 (UTC), 11:31, 6 November 2022 (UTC)

@DePiep: I do not understand you. GRC explicitly explained what he meant by "natural" on p. 77, as I quoted: Scerri pointed out that if the long form of the Periodic Table was written out, then keeping the f- and d-blocks intact, lutetium and lawrencium would naturally occur beneath scandium and yttrium. The Sc–Y–La–Ac alternative could only be structured if, in the long form, the d-block elements of Group 3 (scandium and yttrium) were shifted over next to Group 2, breaking up the transition metal series. He even draws on p. 79 the long-form tables in Sc-Y-La and Sc-Y-Lu form just like the IUPAC report does. Exactly how is this different from the IUPAC report saying: This difference is far more apparent if the periodic table is displayed in an even more expanded 32-column format which incorporates the f-block into the main body of the table. If Lu and Lr appear in group 3, as they do in figure 4, the d-block consists of a continuous sequence of 10 elements. On the other hand, if group 3 consists of Sc, Y, La and Ac, as it does in figure 1, the d-block rows now appear to be split in a very uneven fashion (fig 5). For example, in period 6 we find La (considered as a d-block element) followed by a sequence of 14 f-block elements from Ce to Lu followed by a sequence of nine d-block elements from Hf to Hg. Double sharp (talk) 11:49, 6 November 2022 (UTC)

"How to draw group 3 in the the PT" is not an argument. Nor is, how ~~nice~~"narural" the result looks. Still you and Sandh have written that in the article. Literally.

Furthermore, you are not replying to the points nor the gist of my thread. Actually, the fact that you quote from the source as if I'm disputing that, is missing the whole point. DePiep (talk) 12:05, 6 November 2022 (UTC)

@DePiep: I quote the sources because they are the ones explicitly bringing 18 vs 32 columns into it. IUPAC 2021 report: In the tables shown in figures 1 to 3, group 3 is shown as containing either Sc, Y, La and Ac or Sc, Y, Lu and Lr or just Sc and Y. While the first two options seem to be equally plausible on an 18-column representation, there is one difference that the representation masks somewhat. This difference is far more apparent if the periodic table is displayed in an even more expanded 32-column format which incorporates the f-block into the main body of the table. Scerri 2012 article for IUPAC newsmagazine: I suggest that any reluctance to accept this grouping as opposed to the more frequently seen grouping of Sc, Y, La, and Ac (as shown in figure 5) stems entirely from a reluctance to display the periodic table in its 32-column format. Double sharp (talk) 13:25, 6 November 2022 (UTC)

@Double sharp and DePiep: I understand what you are saying, DePiep: "The constitution of group 3 is not based on the author's choice between 18- or 32-column periodic table." I concur. GRC's statement that there is no issue in a 32-column table since Lu and Lr "fit naturally" as the heavier group 3 elements does not hold up since it is a byproduct of keeping the d-block intact (as GRC says) when there is no inviolable requirement to do so, in a similar way that the s-block is not kept intact when He is placed over Ne. Same thing happens if H is placed over F, as it sometimes is. Double sharp: On a related note, Hill, Holman and Hulme 2017, Chemistry in Context, 7th ed., Oxford University Press, p. 179, show a 32-column table with group 3 as Sc-Y-La-Ac, and with the caption, "Modern periodic table (wide form)". H is shown floating, detached from the main body of the table. And I could not find where Scerri said Lu-Lr would "naturally occur" beneath scandium and yttrium. Sandbh (talk) 13:01, 6 November 2022 (UTC)

@Sandbh: Scerri called Sc-Y-Lu-Lr the "natural choice" in his 2012 article for Chemistry International. In addition to arranging all the elements in a more correct sequence of increasing atomic numbers, the decision to move to a long-form or 32-column table forces the periodic table designer towards just one possible option regarding the question of which elements to place in group 3. The natural choice, turns out to be the placement of Lu and Lr into group 3, as seen in figure 4, because the other option fails to maintain an orderly increasing sequence. Double sharp (talk) 13:19, 6 November 2022 (UTC)

But since DePiep and you are agreeing, I've removed the sentence under discussion. Double sharp (talk) 15:42, 6 November 2022 (UTC)

Thank you Double sharp. I now suspect the reviewer was quoting van Spronsen as a classic historical work, since the sample chapter he was looking at was referring to categorisation approaches to the elements dating from antiquity to the present day.

I know that the Division of Inorganic Chemistry of the ACS still uses the 32-column form with group 3 as La-Ac, in their monthly newsletter and in ad hoc announcements. They did temporarily switch to the 32-column form with Lu-Lr in group 3, after Eric spoke to them. However, that switch was shortly thereafter reverted back to the La-Ac form. Curiously their face book page shows the 32-column form with Lu-Lr. I don't know the reason for the discrepancy.

I have not kept track of 32-column tables, since they appear so infrequently. Sandbh (talk) 05:58, 5 November 2022 (UTC)

@Sandbh: Thank you. I see you edited Rayner-Canham's statement to directly quote him, and I think we have a version of the section at present that we both agree on. Double sharp (talk) 19:46, 5 November 2022 (UTC)

@Double sharp: Thanks. I've (re-added) mention of Vernon's analysis of L&L, and the electron configurations of trivalent Ln and An cations, but this time as footnotes instead. Sandbh (talk) 02:33, 6 November 2022 (UTC)

Case in point supporting GRC: Theodore Gray's PT website. He lets you pick and choose multiple forms of the PT for the front page, including all three group 3 options in the 18-column form. But the only wide 32-column form of the standard PT he lets you pick is the one with group 3 as Sc-Y-Lu-Lr. (His header on each page is 32-column and shows Sc-Y-Lu-Lr too.) Double sharp (talk) 20:43, 5 November 2022 (UTC)

Kudos to Double sharp

For your magnificent list of sources addressing the Group 3 question. Sandbh (talk) 06:50, 26 October 2022 (UTC)

Thanks! Double sharp (talk) 15:44, 30 October 2022 (UTC)

The IUPAC 2021 report on Group 3

"The constitution of group 3 of the periodic table". 2015.
Eric Scerri (2021). "Provisional Report on Discussions on Group 3 of The Periodic Table". Chemistry International. doi:10.1515/ci-2021-0115.

A timeline of positions on the group 3 issue that IUPAC published

Based on searching through old publications of IUPAC, I have added a timeline of positions on the group 3 issue published by IUPAC to my list of sources. I have included a few publications that, while not actually published by IUPAC, were mentioned on the IUPAC group 3 project page as part of the project's progress. I have added brief mentions of the documents published by IUPAC or presented to IUPAC that were not already in the article.

The first entry is from 1985. It doesn't appear that IUPAC has supported Sc-Y-La-Ac in any way, at least since about 1990. Particularly not the split d-block version of it, which the 2016 announcement of the project doesn't even list as an option. Double sharp (talk) 17:57, 8 November 2022 (UTC)

Why background stories in this article? How does this make the section encyclopedic? How do you editors know this argument-picking is not OR or SYNTHESIS? What you are doing is evaluating paper(s) (IUPAC 2021?) & making POV POINTs, but not describing RL RS groups 3. Other issues might apply too, as I said it is wider & FRINGE. Yes that's a lot of capitals—which by itself is an issue with the article texts. IOW, chaotic text by many angles, we're not talking typos here. Why not write to Scerri directly?

What is missing is serious self-criticism in editing Periodic table § Group 3. For crying out loud, why doesn't the UNDUE text volume by itself not trigger questions & self-criticism? You do not have any idea of what the general line is that should/best be conveyed here.

It does not even belong in this article, it's OT. Maybe in Group 3 (if other tests were passed). More likely: in a new article "History of groug 3 ..." as an evaluation of. BTW, this is putting the FA status at risk. DePiep (talk) 20:10, 8 November 2022 (UTC)

I didn't pick arguments. I just paraphrased what each source said at each point while preserving the meaning (obviously, you're not supposed to copy it verbatim for copyright reasons). Aside from some notes that I didn't add, I based the text wholly on:

Scerri, The Periodic Table, Its Story and Its Significance.
Thyssen and Binneman's historical review of how the rare earths were accommodated from Mendeleev to the present, in Handbook on the Physics and Chemistry of Rare Earths.
Geoff Rayner-Canham's chemical review of periodic table issues in The Periodic Table: Past, Present, Future.
articles and reports published by IUPAC.

Those are all secondary sources. The arguments I added are exactly the ones found in there.

But let's work towards some kind of consensus. I do agree that this is getting a bit long for this article. So I'll copy the relevant bits of this over to the group 3 article and begin work cutting this down. Double sharp (talk) 22:49, 8 November 2022 (UTC)

Thanks for taking care to reply.

Now, re "picking" arguments. I did not say that to PINPOINT individual statements let alone their phrasing. Incidentally as in § Frame above, the argument used is "but it is an RS"—re one UC objection I have not made. It is about the wider frame: the paragraphs, section, and so the resulting, intended encyclopedic thread. I miss the self-criticism in the article rebuilding. As noted with Sandbh's many, longterm & nontrivial edits in the nonmetals: unbalanced if not UNDUE. (Over there, Sandbh has singlehandedly degraded metalloids into a subclass of nonmetals; not a pinpoint case, though phase-first-sentence is a case). Now the edits here have started 3 weeks ago with FUD about the published IUPAC report. As if to establish a dismissal of that report through enwiki. Not sure if this false start has been removed altogether (it is wider) or just rephrased ('found a different RS').

I will not propose a different article or section. Nor will I check the dayly edit bursts like the twelve-edit-versions. First of all, because every change that does not fit the Sandbh aim (ie, POV/OWN/..) is overflooded with hectobytes of talkpage forked into sideissues while rarely building into consensus or stable article-quality reasoning, and more in less-helpfulls, but not editing selfcriticism. An argument for a certain in-article statement can take thrice its amount of space at the talkpage, and still be incomprehensible. That is: illegible as article text. The article fails to make it clear to the Reader. (Similar: over at nonmetals, User:John is invited to edit non-acceptable claims into the same but with high-quality grammar; but "remain true to the sources" sure).

At first, when I read this section title "A timeline of positions on the group 3 issue that IUPAC", I though that was an article title proposal, an inviting one at that. Second thought: why do we have to provide that IUPAC timeline at all? Why would we evaluate IUPAC at enwiki? The wikipediatic approach is: such an article is needed, being FORKed from Group 3, but with title adjusted to like "Group 3 composition positions over time" (secondary IUPAC sections expected).

As said, it is wider. DePiep (talk) 06:55, 9 November 2022 (UTC)

@DePiep: I think we agree on a separate article. We have things like lists of metalloids as a precedent: this is equally a list of positions taken by sources throughout history, forked from the main metalloid article. It can be developed pretty easily from the list I already have at User:Double sharp/Group 3 sources. Personally, I think that the ones coming from IUPAC ought to be highlighted. Double sharp (talk) 08:13, 9 November 2022 (UTC)

Thumb is up here. The metalloids list is just a list (used for tallying), still a good entrance. That new article is challenging and inviting as to encyclopedic aim; count me in for spell checking. Articles PT and Gr3 both can enjoy its summary of course. Am I wrong in expecting that IUPAC/Scerry (2021) will be recognisable thoughout? DePiep (talk) 08:23, 9 November 2022 (UTC)

@DePiep: I planned to make it so, though off the top of my head I'm not 100% sure how to do it yet. Maybe highlight sources that give the same argument that IUPAC/Scerri (2021) uses? Or also count argument type (chemistry vs physics vs electronics vs IUPAC/Scerri 2021 desiderata?). At the very least I guess the lede of such an article (maybe List of arguments used for the composition of group 3) should start off with IUPAC/Scerri (2021) and the particular argument that it used. Double sharp (talk) 08:25, 9 November 2022 (UTC)

Personally, I'd like to learn from it the various scientific views (elconfig etc). How much is the Reader audience at home with orbit & energy? Assuming lower=better. If I finally can understand those from the article, you can enjoy payed retirement from wiki.

TBH, after these sections IUPAC 2021 will be easy to add (so I change opinion re this: better be at the end), as overview & recap & desiderata. Also helpful clarifying tipping the Favoured Gr3 Presentation scale (as we have read & done).

IMO, even in IUPAC (2021) the 18/32 discussion is only a sideshow: addressing the old school scientists who grew up pre-Seaborgy; some of them are good wiki editors though. DePiep (talk) 08:45, 9 November 2022 (UTC)

@DePiep: I'll start drafting something and let you know when I think it's in a good enough state for you to take a look. :) Double sharp (talk) 10:59, 9 November 2022 (UTC)

@DePiep: I wrote a first draft for such a separate article, based on text I had prepared previously. To ascertain that the arguments in question have been advocated by significant groups or individuals, I've only included arguments that either (1) somebody involved with IUPAC deliberations on this made/quoted/cited, or at least (2) are in secondary sources (rather than primary research papers) and have been agreed with by somebody else. Also I have not included sources that contradict themselves, obviously. Per your excellent suggestion, I have placed IUPAC 2021 at the end: the last section collects everything in IUPAC publications and culminates in the 2021 report. (Some citations are not completely filled out yet, but the text is at least presentable enough for reading.)

I've tried to assume low, but of course I might not have done that well enough. So I await your comments, positive and negative alike. That's how we improve things after all. :)

P.S. I would like to have Trifonov and Shchukarev for their Sc-Y-La arguments, but I can't find their papers (they're old, from the 1970s) and it doesn't appear that their arguments had much influence over the debate (Scerri does not mention them in his book on the PT's history). So, OK not to have them in my view, though I'd still personally be interested in getting copies to read. Double sharp (talk) 21:29, 9 November 2022 (UTC)

Great (so, you can write such articles in one day? That endangers your wiki-$retirement!).

I planned to read it thoroughly later on. I like the ambition and the setup, glad we have the same in mind. At last, there is space enough for all RS views.

Section outline? Don't know if it is good encyclopedic style, but we could open each section with an short outline? "From atomic electron configurations, A and B is advocated/argued." Done afterwards anyway.

Top illustration. In top, the question is to be made clear, not the answers. Skipping Mendeleev/19th-century forms here. We could open with a 18-column form with Sc-Y-*-* encircled (immediately recognisable!), plus question mark(ing). Right below, a single PT with both options (=2 Sc-Y positions) present & encircled, with question mark(ing) between. Best in 32-column. The number "3" is present, part of the question marking. Graphic: in b/w should do. PTs may be cut-off ~right half, as long as situation is recognisable and corresponding btwn the two PTs. Essential symbols readable. Looking for time to make sketches.

Other illustrations: later more. I hope we can find (create) great ones. At least one per section. DePiep (talk) 05:00, 10 November 2022 (UTC)

In this article, I think there is no a priori objection to cite Vernon, as any other RS. There is space and article intention to give argument evaluations, as with any other RS. It's more about Due Weight in here. (The difference with PT article is, that in PT we expect fleshed out main RS status/es, not disputes). @Sandbh: DePiep (talk) 05:11, 10 November 2022 (UTC)

@Double sharp and DePiep: Is this draft article going to form part of the Group 3 article? Sandbh (talk) 05:30, 10 November 2022 (UTC)

Not to my understanding. It is to be a separate full article. Consistent summaries to appear in PT (short) and G3 (wider). These summaries do not need to reflect actual dispute(s), just RL mainstream w/due weight for that level. Reflecting IUPAC 2021. DePiep (talk) 05:36, 10 November 2022 (UTC)

@DePiep: The IUPAC source additions have been moved to group 3 element per your suggestion. This makes a small addition only to the pre-14 October 2022 version. (So, in fact, much of what I added here was already in the main article; that was my mistake, I'd forgotten about it. The main change really to the group 3 element article is citing about three more IUPAC documents and correcting the start date of IUPAC thinking about this from 1988 to 1985.)

As for this article (periodic table), I have restored the 11 October 2022 version of this section that predates either my or Sandbh's additions. Since that was stable for a long time, it should serve fine while this is sorted out on the talk page.

Feel free to question me again if you feel this is still not suitable. Surely I'm not perfect, and I welcome your opinion. In particular, I'm also willing to restore the pre-October 2022 version of the group 3 element article if you think that's suitable. Double sharp (talk) 23:07, 8 November 2022 (UTC)

@Double sharp: With respect, I’ve reverted your edit that resulted in the article being restored to the 11 October version. You and I have been working very well, calmly, patiently, respectfully, and successfully to develop a more balanced article, and I see no good reason to undo that one month’s worth of careful and thoughtful work. I understand that if there are any unresolved concerns about the article we are now obliged to attempt to seek consensus. Sandbh (talk) 09:37, 9 November 2022 (UTC)

So, the 11_Oct/10_Nov diff [5] is in the recent phrase, now removed: 2021, which found there was no objective basis to resolve the question but that "perhaps a compromise could be reached on" the table with lutetium and lawrencium in group 3. I can support the removal (ie revert) because the quote is confusing "compromise" with "various scientific classifications". IUPAC (2021) does not compromise on these classifications, it compomises on the classification to present as 'preferred' (IOW, the compromise is wrt the presentation preferrence). That is: not a compromise on the (absent) "objective basis".

In the text, this one occurrence of "compromise" is conflicting with the other five surrounding occurrences in that section. The other "compromise"s refer to a 1988 statement. No reason for this mixup is given, nor why a 44 year old, in 2021 dismissed option, would be brought up here—at all. Add Red Book versions as argument, 8-column PT (hey, Mendeleev drew the 18-column form himself already in 1900).

Incidentally, IUPAC (2021) does acknowledge a 'Sc-Y-0-0' group 3 (and so, a 15-column f-block), but explicitly as a scientific interest form. Not as a compromise then.

My objections are against the multiple confusing mixups in the section, against the needless sidetracks visited, against the and against the general line of the section. Bringing up the compromises this way, the 8-columns Mendeleevian PT, pre-Scerri Red Book: nothing is made clear. I still have no idea what the setion is trying to say. Instead, the text shouts "hey, IUPAC-is-compromising", "gotcha", "Scerri is so unclear/incomplete/..".

While, IUPAC (2021) has presented that description on a plate for us. DePiep (talk) 16:59, 11 November 2022 (UTC)

@DePiep: That's because the section is giving a summary of the history as well as the current status. This is fairly important, as without the history, it's impossible to understand why the dispute started in the first place. If the electron configurations had been measured properly in the beginning, for example, there probably would never have been a dispute at all.

That said, I can see the point of moving it to the History section. The History section already mentions how the rare earths' classification changed from Mendeleev onward as part of turning things from his original to the current form, and this is a natural extension of that. Maybe I'll try redrafting it that way and present it for your opinion. I'll probably make an edit, then self-revert so you can see the diff. :) Double sharp (talk) 17:26, 11 November 2022 (UTC)

re paragraph 1: in this PT article, its History is not required nor needed. (Especially not since people still think it is caused by the PT graphics). And "elconfig measured right" ... are you sure? IUPAC 2021 is saying that there are interests in a classification. That's not about erroneous measurements right? Simple: there are two (three) different classification principles today, so here is my section draft: (1) describe them and (2) c/p why IUPAC ends up with that one preference. 150 words max.

The remarks in the report re PT graphics is a background for the "dispute" (as in: personal reluctancy to accept a classicication, ie scientists clinging to their unthoughtfully-drawn Group3 PT, ca. 1965). But not a background for the science. Forget the graphs, write down the classifications (as the Report mentions them). The electrons don't know about our graphs.

re paragraph 2: So no, not to History section either. (And, as long as the misguided graphic-is-the-point is in there, it does not even belong in the Group 3 article). The rare earths: otherstuff.

And nobody has thought of removing MT. Is there a secret challenge going on to enter cat names in an FA? DePiep (talk) 18:28, 11 November 2022 (UTC)

@DePiep: Yes, I'm quite sure about "elconfig measured right" in the context of the history. On p. 393 of Eric Scerri's book The Periodic Table: Its Story and Its Significance, he describes the early history of the group 3 thing. According to him, it goes: (1) elconfigs are measured, people take that as a sign that Lu should be f-block; (2) remeasurements prove the elconfigs were wrong, Landau and Lifshitz point out that that suggests that Lu should be d-block. Then after that a whole lot of people start arguing. So yes, that's exactly how it started. Yes, Scerri writes later that "appealing to specific chemical or physical data" cannot conclusively resolve the problem (i.e. you can't use elconfigs that way), just as the 2021 IUPAC report says. But for the history section, what matters is that people at that point thought you could do that. We cover Bohr's initial mistakes and Mendeleev having to add noble gases in the History section, so I don't see why this mistake can't be mentioned as a historical error. It is one small step in the story of how we got from Mendeleev to today's version.

Of course, this is only the overview article, so I've kept it very short. Double sharp (talk) 18:52, 11 November 2022 (UTC)

P.S. Managed to get through that in three sentences in "History"; that should be short enough. Double sharp (talk) 19:10, 11 November 2022 (UTC)

If it is an error, why does IUPAC 2021 keep it on the table as an option? (albeit less favoured for scholarly & tiger reasons; as it does with Sc-Y-0-0). And anyway, this article can mention it as a "controversy" (ie, opposing scientific reasons), not as historic errors (not "history of group 3" then). There are other articles for that. DePiep (talk) 12:54, 13 November 2022 (UTC)

@DePiep: It started from an error, but then some people started arguing that it might be right for other reasons. Double sharp (talk) 21:52, 13 November 2022 (UTC)

@Sandbh and Double sharp: quick reply. I don't have time these days to dive in the necessary sources & enwiki versions. My questions, some new, are: (1) Ds (not IUPAC 2021 afaik) is introducing the "error" claim. (2) Sc-Y-0-0 is an third option in IUPAC, with little preference but scientifically valid (another tiger). (3) I'm not sure how the extended history of all this really helps clarifying current situation, ency-wise. (4) Eloquence of text. DePiep (talk) 04:16, 14 November 2022 (UTC)

@DePiep: Understood. I'll await when you have time.

For when you do, though: Scerri states on p. 393 of The Periodic Table: Its Story and Its Significance that Sc-Y-La-Ac arose from an error in measuring elconfigs. The quote:

Some early determinations on the elements ytterbium (70) and lutetium (71) indicated the following electronic configurations for their atoms:
Yb [Xe] 4f¹³ 5d¹ 6s²
Lu [Xe] 4f¹⁴ 5d¹ 6s²

As a consequence, it was concluded that lutetium should mark the end of the lanthanide series. In addition, since the discovery of lutetium had taken place at about the same time as the discovery of several other rare earth elements, it was generally thought that lutetium was also a rare earth element.
In 1937, William Meggers and Bourdon Scribner published an article in which they reported that contrary to earlier observations the configuration of ytterbium should be assigned as

Yb [Xe] 4f¹⁴ 6s²

Nevertheless, the authors did not comment on any possible ramifications for placement of the subsequent element lutetium in the periodic table. If ytterbium possesses 14 f-electrons, rather than 13 as formerly believed, it can genuinely be thought of as the final rare earth element. Consequently, the next element, lutetium, can be regarded as a d-block element, thus placing it under scandium and yttrium in group 3. In the years following this discovery, a few books and published periodic tables incorporated the newly assigned configuration of ytterbium but still refrained from discussing any possible change to the position of lutetium.
The first statement that these configurations provided grounds for regarding lutetium as a d-block rather than as an f-block element comes from the classic book on quantum mechanics by Lev Landau and Evgenii Lifshitz in 1959. The authors stated: “In books on chemistry, lutetium is also placed with the rare earth elements. This, however is incorrect, since the 4f shell is complete in lutetium.” This recommended change was then independently rediscovered by a number of other authors, working in different subdisciplines, but none of them seemed to have any impact on the way the periodic table was presented in textbooks or elsewhere, with the possible exception of the work of W.F. Luder. Several of the authors who proposed that lutetium should replace the element lanthanum in group 3 were physicists, a factor that may have contributed to their being ignored by the chemical community.

So I'm not particularly surprised that while the 2021 report does state that there may be a reason for Sc-Y-*-** in specialised scientific contexts, it doesn't say so for Sc-Y-La-Ac. That started, after all, from an error. It just happens to be widespread in textbooks. Double sharp (talk) 11:41, 14 November 2022 (UTC)

@Double sharp: Sc-Y-La-Ac did not start from an error. La was discovered in 1839. Mendeleev published his first periodic table in 1869. La subsequently came to be associated with Group 3, along with scandium and yttrium (Thyssen and Binnemans 2011, p. 36). Lu was not discovered until 1907. It was not until 1937 that the error as to Lu's e/c was noticed. La and Lu thereby each had a d- differentiating electron and, ostensibly, an equal claim to the periodic table position under yttrium, in Group 3. Hence L&L's depiction of La above Lu, and both being part of the d-block. Sandbh (talk) 23:00, 14 November 2022 (UTC)

@Sandbh: That's quite a way to oversimplify the situation. The passage in Thyssen and Binnemans you refer to is talking about the 8-column table of Mendeleev. There was then no "Group 3" but "Group III", and it started with B, not with Sc. And later in his life, Mendeleev didn't just place La in group III, but also Yb, and still stubbornly refused to recognise the lanthanides when they wouldn't fit his system. This clearly has no significant bearing on the modern version, which does not mix the A and B groups (i.e. the Reihen have since been unfolded into periods).

As Thyssen and Binnemans note, the first long-form tables published according to the intergroup methodology (i.e. the Ln are not all stuck in one place or scattered around) were Bassett's and Werner's. Both were Sc-Y-Lu, in the sense that La was placed in its own column, a blank space was left below Y, and the d-block was not split. Thyssen and Binnemans interpret that quite naturally as Sc-Y-Lu on p. 81: For instance, in the periodic table of Werner (1905a,b), there is an open place below yttrium at the position where lutetium is expected, but it should be realized that at that time lutetium had not yet been discovered (this was in 1907). However, Werner did not consider lanthanum as a homologue of yttrium, because of the differences in chemical properties between these two elements.

So I see no reason to doubt Scerri's account, in which the La form began with the incorrect configurations. ~~Chemists before that were often putting Lu under Y. Not exclusively, as shown by Smith (1927), but the big names Alfred Werner (1905) and Charles Rugeley Bury (1921) were doing it.~~ The first Lu tables predate that, and come from big names like Alfred Werner (1905) and Charles Rugeley Bury (1921). Niels Bohr's system (1922) circles Ce–Yb as rare earths, so he's more like Landau and Lifshitz and puts both La and Lu as transition elements. That is still not Sc-Y-La-Ac, which still can be seen to arrive later on the scene than Sc-Y-Lu. Double sharp (talk) 23:08, 14 November 2022 (UTC)

@Double sharp: The passage in T&B (p. 36) was, "At the beginning of the twentieth century, only scandium, yttrium, lanthanum, and cerium had been accommodated in the periodic system with some certainty, but all other rare earths remained homeless."

I'm not aware of any evidence supporting the notion that, "chemists before that were often [italics added] putting Lu under Y." OTOH, I support Jensen's assertion that, "a number [italics added] of chemists in the 1920's and 1930's assigned lutetium rather than lanthanum to group IIIB", with him providing two specific examples. Sandbh (talk) 05:34, 15 November 2022 (UTC)

@Sandbh: Again, this is the 8-column version of the periodic table that T&B are then talking about. At that point boron and aluminium were also regarded to be in the same "group III" as scandium. Note that "group III" and the modern "group 3" are not the same thing. So this is, like I said, not relevant to the modern periodic system.

I've revised the statement, looking through Mark Leach's database. Regardless, it still does not appear that Sc-Y-La came first. The accommodation of the rare earths was confused for a long time, with some proto-Sc-Y-*-**'s, some * going across both group III and group IV, and occasionally considering the rare earths to be between Ce and Ta exclusive instead (Black and Conant). Bassett and Werner seem to be clear indications that Sc-Y-Lu came earlier than Sc-Y-La.

P.S. Other post-Janet Lu tables before 1950: Gardner & Mazzucchelli (1930), Bacher & Goudsmith (1932), Zmaczynski (1937), Gamow (1940), Catalan (1949). Double sharp (talk) 16:12, 15 November 2022 (UTC)

I must add (5), Double sharp, that the continuing "let's poll on this" drip-drip multikbyte POV'wiggling push is not exactly a "question" any more. Let's call it a "worry". DePiep (talk) 06:50, 15 November 2022 (UTC)

@DePiep: It seemed to me that Sandbh intends to start a poll regardless of how either you or I feel about that. Therefore, I felt it would be more effective to suggest changes to his proposed wording of such a thing in the interest of presenting the information neutrally. Naturally I still don't think it makes a convincing case for the change he seems to be proposing (a reversion back to Sc-Y-La), but I can always oppose it afterwards.

Or at least that's how I thought the best way to proceed was in that situation, though I'd be glad to hear your opinion as well. Double sharp (talk) 16:12, 15 November 2022 (UTC)

@Double sharp: Thanks for your help in reaching a neutral form of wording for the poll. Sandbh (talk) 05:54, 16 November 2022 (UTC)

@Double sharp: Re where you wrote:

"...Scerri states on p. 393 of The Periodic Table: Its Story and Its Significance that Sc-Y-La-Ac arose from an error in measuring elconfigs...."

No, Eric did not write that Sc-Y-La-Ac arose from an error in measuring elconfig. Sc-Y-La-Ac was around way before the Lu elconfig error was realised in 1937. And whether it was an an 8-col, 18-col or more-col table is irrelevant. Groups were around before then. It is irrelevant whether B-Al were or were not included together with Sc-Y-La.

What the Lu error established was that La and Lu had an equal claim to the position under Y given it turned out they were both 5d elements. As such, neither is an error.

Brauner (1882) is the earliest table I have been able to find that associates (B, Al,) Sc, Y and La. --- Sandbh (talk) 06:24, 16 November 2022 (UTC)

You are supporting my, Jensen's, and Scerri's point. Sc-Y-La-Ac occurred before it was realised that this is an error. So it did originate in the time when people thought that the lanthanides all had a d-electron.

As Jensen points out, and actually the IUPAC report also alludes to by referring to Th, it still constitutes an error of consistency to use such logic to place La under Y. Th doesn't have an f-electron either and its d²s² configuration makes it the first element to match hafnium in group 4. Yet nobody has put it in the d-block since Seaborg, proving Jensen and Scerri's double standard. As Thyssen remarks, neither does anybody end the d-block early at Cu, Pd, and Au just because that's where the first d¹⁰ appears (and never goes away). No other Madelung exception has any significance for periodic table placement whatsoever, except for La and Ac, and it's not even consistently applied to the element next door (Th).

In Brauner's table, the A vs B alignment puts boron with Sc, Y, and La in group IIIA. Then it thinks Yb is their next congener. Then it thinks Tb and Sm might be heavier congeners of Cr and Mn. This clearly has no significant ramifications for the modern form. Double sharp (talk) 11:53, 19 November 2022 (UTC)

@Double sharp: What Landau and Lifshitz wrote (p. 245) was:

"In books on chemistry, Lu is also usually placed with the rare-earth elements. This, however, is incorrect, since the 4f shell is complete in Lu; it must therefore be placed in the platinum group."

They show what they refer to as the "iron", "palladium", and "platinum" groups as follows (p. 244):

Iron group
Sc Ti V  Cr Mn Fe Co Ni
Palladium group
Y  Zr Nb Mo Tc Ru Rh Pd
Platinum group 
La
Lu Hf Ta W  Re Os Ir Pt

On page 246 they list the rare-earth elements as Ce−Yb.

IOW, Scerri wrote only half the story, neglecting to point out that L&L showed both La and Lu being part of the d-block.

Contemporary corroboration is provided by Frye (1949, p. 4):

"La, the first member of the series, has no 4f-electrons and is not considered a rare earth by some spectroscopists."

Frye, H.G.: A scheme of analysis for the ceric rare earths. University of the Pacific, Thesis (1949). https://scholarlycommons.pacific.edu/uop_etds/1093

And from Collier’s Encyclopedia (1958):

"La, 57, is excluded by spectroscopists because it has no electron in the fourth shell and, therefore, has a markedly different spectrum from that shown by the other members of the group. Lu, 71, is sometimes excluded from the rare earth group because its fourth shell is filled completely."

--- Sandbh (talk) 01:00, 14 November 2022 (UTC)

@Sandbh: Landau & Lifshitz are no longer in the article in the current version. But even if they did not mean that Lu should replace La, that is pretty much how physicists in the decades after interpreted them. So maybe Matthias should have proud after all, that his group was the first set of physicists to suggest Sc-Y-Lu in the modern form. (Werner suggested it before, but he was a chemist.)

But I changed the group 3 element article.

P.S. Spectroscopically, helium is an alkaline earth, and thorium at least requires some apologetic handwaving. Spectroscopists are working in one of the few fields when anomalous configurations actually matter. So, this runs straight back into what the IUPAC report pointed out: Moreover, one may readily concede that an element such as thorium does not actually possess any f-orbital electrons and yet it is classified as being among the f-block elements in all five of the periodic table representations shown in figures 1 to 5. Double sharp (talk) 01:09, 14 November 2022 (UTC)

@DePiep: Here's a shortening attempt. Sections of interest are "Electron shells" (part of "History") and "Group 3" (of course). Double sharp (talk) 18:57, 11 November 2022 (UTC)

Basis for indeterminate finding of IUPAC group 3 project

See #News from Scerri (2022) (I disagree with the view suggested by this section title). Double sharp (talk) 20:58, 15 November 2022 (UTC)

I was referring to the fact that the IUPAC group 3 project found there was no objective way of adjudicating the group 3 question. There is naught in the "News from Scerri (2022) item" that contradicts this. Sandbh (talk) 05:50, 16 November 2022 (UTC)

Here are three contributory factors.

1. This chemistry stack exchange page appears to shed some light on the question:

"Q: Are the elements La and Ac considered to be in the d block or the f block of the periodic table?

A: The real lesson here is that the boundaries between "blocks" of the Periodic Table, like the boundary between "strong" and "weak" acids or bases or even between what is a stable compound and what isn't, is not sharp. Some other examples of a rough, spotty, changeable real world:

Most simple magnesium compounds, even the best known hydride and boride compounds, are primarily ionic, but when they have covalent character the magnesium often bonds tetrahedrally as if using 3𝑝 as well as 3𝑠 valence orbitals. See for instance the coordinated structure given here for methylmagnesium chloride in THF. Beryllium shows this effect even more prominently in its wider variety of covalent compounds.

Calcium could be called a 𝑑-block element when it bonds with its 3𝑑 orbitals in this calcium(I) compound (and yes it is +1, showing multiple oxidation states like a transition element).

Cerium, the second element is the lantanide series, does some straddling of its own between 𝑑 and 𝑓 blocks as its valence in the metal is changeable between 3 (with a core-like 5𝑓 electron) and 4 (with this electron engaged in the bonding). See Johanssen et al. 2 and the WP article citing this reference.

Among the actinides there is also thorium, which appears to involve only 7𝑠 and 6𝑑 valence electrons in the gas phase but brings in 5𝑓 orbitals in the metal (see this answer and the references therein)."

Johansson, Börje; Luo, Wei; Li, Sa; Ahuja, Rajeev (17 September 2014). "Cerium; Crystal structure and position in the periodic table". Scientific Reports. 4: 6398. Bibcode:2014NatSR...4E6398J. https://doi.org/10.1038/srep06398.

Krieck, Sven; Görls, Helmar; Westerhausen, Matthias (2010). "Mechanistic elucidation of the formation of the inverse Ca(I) sandwich complex [(thf)3Ca(μ-C6H3-1,3,5-Ph3)Ca(thf)3] and stability of aryl-substituted phenylcalcium xomplexes". Journal of the American Chemical Society. 132 (35): 12492–12501. https://doi.org/10.1021/ja105534w.

On a related, note Sanderson (1960, p. 8) wrote:

"If a d electron, for example, can easily behave like an f electron, or vice versa, the argument as to the exact ground state configuration becomes relatively unimportant."

That is to say, both La and Lu can relatively easily behave as if they were f elements never mind their 4f⁰5d¹6s² and 4f¹⁴5d¹6s² formal configs.

Sanderson RT 1960, Chemical Periodicity, Reinhold, New York

The chemistry stack exchange response and Sanderson show that the presence of this or that kind of electron, especially in marginal cases at the borders of putative blocks is less than a major consideration.

2. Each of the two options has something to offer and each comes with advantages and drawbacks.

For the La-Ac form, many of the advantages were set out in my peer reviewed article that appeared in Foundations of Chemistry (13 citations). As well, each block starts with the first appearance of the relevant electron.

For the disadvantages, I suppose the biggest is that it requires a split d-block, if shown in 32-column form.^ And Sc-Y-La-Ac messes up the regularity of spin multiplicity.

^ I qualify this by noting that a split between groups 1 to 3 and 4 accords with a transition from largely ionic chemistry to covalent chemistry.

For the Lu-Lr form, some advantages I can think of are:

if the 4f row is shown as La to Yb, the number of f electrons in each atom corresponds to its position in the row, in all bar three cases (La, Ce and Gd);
the situation in the 5f row is a little more involved, but still pretty regular;^^
a cohesive d-block in the 32-column form;
more amenable (?) to being mathematised.

^^ For Ac–No, the actinide series can be divided into three relatively cohesive sets:

Ac and Th, which each have a number of d electrons equal to their position;
U and Am, which have a mix of d and f electrons; and
Pu to No, in which the number of f electrons in each atom corresponds to its position in the row in all bar one case (Cm).

Some drawbacks are:

an extra differentiating electron discrepancy;
Sc-Y-Lu-Ac is inconsistent with stoichiometric considerations;
reduced regularity of term symbols;
disaggregated lanthanide contraction.

3. Over the past 85 years, since it was found that La and Lu each had a d differentiating electron and (ostensibly) an equal claim to the position under Y in group 3, nobody has been able to mount a sufficiently compelling case for the superiority of either Sc-Y-La-Ac or Sc-Y-Lu-Lr over the other. In the category of "insufficiently compelling" I include here "making such a decision on the basis of convention".

Re 3, compelling to whom? Where is the RS? If we're talking about "compelling to authors focused on the group 3 question", then most found the arguments that Sc-Y-Lu-Lr is superior compelling. If we're talking about "compelling to textbook authors", then per Jensen's 1982 article it's unlikely that they even were aware of any arguments at all, so they couldn't have found any of them compelling or not compelling. On the other hand, we have exactly an example of a relevant organisation for whom making a decision based on convention was compelling: IUPAC, which after all started a project to make a decision on behalf of it, and was happy with the report that said exactly that it is a convention. The IUPAC report itself says that electronic configurations and block assignment is an approximation anyway, so I'm not sure how your 1 is relevant. And where are all your supposed advantages and disadvantages from 2 in the IUPAC report? That report went out of its way to note that some specialist relativistic quantum chemistry considerations support Sc-Y-*-**. It did no such thing for Sc-Y-La-Ac. And finally, what do you think most definitions in science are, anyway? Deciding what a planet is, as opposed to a dwarf planet or brown dwarf, is a convention. Choosing a point to divide semiconductors from insulators is equally a convention. Numbering the groups from 1 to 18 is, even more relevantly, a convention. That does not render the convention insufficiently compelling to adopt. In the second case we even see exactly why conventions are a quite compelling reason, namely that if that had not been made as a convention, we would still be in the situation where I write "group VIB" and nobody has any idea whether I mean the chalcogens or the chromium group.

BTW, why is Johansson et al. (a theoretical paper) compelling to you about Ce 4f, while Wittig isn't compelling to you about La 4f? They're both theoretical – after all, what could "experimental" mean when it comes to the orbital approximation, which is a model anyway? And Wittig wasn't even alone in his suggestion or first: Gardner and Smith 1964 suggested it before. It's still been advocated in modern chemistry, noting that Gschneidner similarly calculated La with and without 4f in his very last article from 2016, using the melting-point difference to justify his contention that La metal has some 4f valence involvement). And like I said on your talk page, La actually needs a smaller excitation energy to occupy 4f (1.88 eV) than Be needs for 2p (2.72 eV). I see a double standard. Double sharp (talk) 22:23, 18 November 2022 (UTC)

Revising the "Current questions" subsections

@DePiep: Based on your comments today, I've done some more reworking, based on the description given in the IUPAC 2021 report. What do you think of it?

I'd also like to ask you for your opinion about the other sections in "Current questions", since I've just taken a look at them. Currently, all three are 3 paragraphs long, so we're moving towards balance. However, I wonder now if the last section ("Superheavy elements") should be there at all. Scientists have talked about the possibility of moving the last few elements around if their chemistry turns out different from expected, and vary between being in favour of the idea or being against it. But the chemistry has not actually been investigated and so nobody has actually proposed any actual changes, and the positions of the heaviest elements are currently the same everywhere. So, there is no actual "dispute", just some musing in the literature. Is that even worth mentioning at all?

(Period 1 has at least been discussed in IUPAC publications, so that one clearly deserves a place here.) Double sharp (talk) 22:12, 13 November 2022 (UTC)

As in my 04:16 reply above: 4 Qs; <0 time for literature; hope I can respond to all, content, later this week. DePiep (talk) 04:18, 14 November 2022 (UTC)

Well, I decided to remove this section, because it is about a dispute that does not exist. Double sharp (talk) 20:40, 22 November 2022 (UTC)

News from Scerri (2022)

@Sandbh and DePiep: Eric Scerri released some relevant IUPAC correspondence in a lecture at the University of New Hampshire this year (start watching at 52:54 for context). He said: "we suggested that it should be option number two, with lutetium and lawrencium in group 3 rather than lanthanum and actinium ... IUPAC basically said 'well, yes we agree with you; however, if we're going to change group 3, this requires changing the periodic table as a whole'". Since Scerri clearly says that the conclusion of the report was to suggest Sc-Y-Lu-Lr, and that IUPAC agrees with the conclusion, there are no grounds for saying that the report has not been substantiated. So Sandbh's statements in #Clarifications appear to be contradicted by the actual chair of the project.

From this and the correspondence he released (which is surely just a fragment of all of it) we can glean some info already. The IUPAC Inorganic Chemistry Division (Division II) said, per the released correspondence, in April 2021 that it is "happy with this result". Presumably that is the 2021 report which was published in the IUPAC newsmagazine that January, per Scerri's summary wording yes we agree with you and the fact that that's the last update on the group 3 project page. It is just that the change affects the periodic table as a whole, so per the released correspondence, the project seems to have concluded because it expanded beyond its scope of issuing a recommendation on group 3 alone to dealing with "the entire table", not because of any scientific objections to its conclusion (kind of an organisational reason rather than a scientific one). That agrees with what the project report said: What has also become apparent is that this question cannot be treated independently of that of the form of the periodic table as a whole. It also appears that IUPAC don't consider it urgent to "do anything" (presumably, change the so-called IUPAC periodic table, since that's what Scerri then talks about) "until there are new elements added to the periodic table" (at which point it generally changes anyway). Scerri does worry in his lecture that this means it's been left hanging, and suspects that the delay might be because IUPAC is afraid of the sort of situation like when Pluto was reclassified and everybody "freaks out" that their "favourite periodic table" has been changed, in his words. Then again, "until there are new elements added" is a pretty clear condition for the hanging to end.

Not that that matters much anyway, because Scerri also mentions that the so-called IUPAC periodic table on the IUPAC website is not even IUPAC-approved at the moment (a point that IUPAC themselves made back in 2009). If it is not IUPAC-approved, then the fact that it's not updated to reflect this conclusion does not seem to be that important. The important thing is that the report's conclusion in favour of Sc-Y-Lu-Lr was published in the IUPAC newsmagazine and has evidently been accepted by the relevant IUPAC authorities: that's definitive. The only question left "hanging", to use Scerri's word, has nothing to do with disputing its conclusions, but with when IUPAC wants to get around to updating the "IUPAC periodic table" that is not even IUPAC-approved anyway. Noting that IUPAC's periodic table coloured by discovery is also not updated to their latest atomic weights, and that they also haven't updated their website to give the final criteria for element discovery (it still links only to the provisional report), the only delay left strikes me as a bit of a nothingburger. Better to rely on IUPAC reports than whatever their websites say, especially if tables on the IUPAC website are not IUPAC-approved and the website often takes very long to be updated. They do tend to update it more urgently when new elements are discovered, which generally checks out with the released correspondence. Double sharp (talk) 20:57, 15 November 2022 (UTC)

[A small P.S. note, added later: and besides, we always put the discovered superheavies on the periodic table before they appeared on the so-called IUPAC periodic table, so there's even a precedent for not taking this non-IUPAC-approved table too seriously.] Double sharp (talk) 22:37, 16 November 2022 (UTC)

@Michael D. Turnbull and Robertiki: As you've participated in this thread, I think this bit of news Scerri released this year would interest you too. Double sharp (talk) 21:00, 15 November 2022 (UTC)

Thanks Ds for noting this, & so carefully. -DePiep (talk) 06:22, 16 November 2022 (UTC)

@Double sharp: Thanks very much for bringing this to our attention.

To correct and clarify a few points.

1. You wrote:

"Eric said "we suggested that it should be option number two, with lutetium and lawrencium in group 3 rather than lanthanum and actinium."

Eric is blatantly wrong. In the provisional report he wrote: "Perhaps a compromise could be reached on [Sc-Y-Lu-Lr]."

2. You wrote:

"IUPAC agrees with the conclusion..."

No, that is not what the email (presented by Eric) said. It just says that one of IUPAC's divisions is happy with the result of the work of the Group 3 project as set out in the provisional report.

Note too, that the email is not even from an IUPAC address. Instead, it is from the originator at Chalmers University of Technology, Sweden.

3. You wrote:

"There are no grounds for saying that the report has not been substantiated."

Untrue. One internal e-mail via a non-IUPAC address, referring to one division of IUPAC, to Scerri, does not represent formal substantiation by IUPAC as a whole.

3a. Further, the email Eric presented is from Lars Öhrström who was a member of the IUPAC Group 3 project team! Aside from the conflict-of-interest-warning-light flashing bright red, Lars Öhrström does not represent IUPAC as a whole.

3b. Lars further wrote, in the undated addition:

"It may even be premature to do anything right now."

Quite so.

3c. Lars added:

"Students will always be annoyed when the answer is not "yes" or "no" but "on one hand..." However, there is value in learning that is how things are in real life."

Quite so, consistent with the current IUPAC periodic table which can be interpreted either way.

4. You're spot on when you wrote:

"Scerri does worry in his lecture that this means it's been left hanging, and suspects that the delay might be because IUPAC is afraid of the sort of situation like when Pluto was reclassified and everybody "freaks out" that their "favourite periodic table" has been changed, in his words."

Among other reasons, this is why the decision to change the PT in the lede of the PT article to Sc-Y-Lu-Lr was premature.

5. You wrote:

"The important thing is that the report's conclusion in favour of Sc-Y-Lu-Lr was published in the IUPAC newsmagazine and has evidently been accepted by the relevant IUPAC authorities: that's definitive.

No, the "provisional" report did not conclude in favour of Sc-Y-Lu-Lr. All they wrote was, "Perhaps a compromise could be reached on [Sc-Y-Lu-Lr]." That falls short of a conclusion in favour of Sc-Y-Lu-Lr.

No the "provisional" report has not been accepted by the relevant IUPAC authorities. All that has happened is that a member of the Group 3 project, from one of IUPAC's divisions, in a non-IUPAC email, has said that the division is happy with the result. No, that is not definitive.

6. I further note Eric's presentation was given in April 2021. If that much weight can be allocated to the email in question then I would've expected Eric to have (metaphorically) shouted it from the rooftops---which he didn't.

7. As EdChem wrote:

"WP is still hampered, in my opinion, by a preliminary report with recommendations / suggestions inconsistent with RS, that does not yet have the imprimatur of IUPAC itself. RS like textbooks will act on this (or not) in future editions, so any new consensus on a convention will take time to evolve.

EdChem (talk) 23:48, 21 January 2021"

Comment: This is still the case; no such consensus has evolved.

--- Sandbh (talk) 04:43, 16 November 2022 (UTC)

Re: Thanks Scerri & Double sharp for introducing the Pluto analogy. With this, the "was based on an error" is much more easy to grasp. Now let me find my recorded diffs that said so (like "pre-Seaborgy"), and those stating that the IUPAC-sitepublished PT is not decisive nor a statement. Our task: write that in ~100 words max, here. And let's start the rename E94 project. -DePiep (talk) 07:24, 16 November 2022 (UTC)

Instead of trying to square the plutonic circle (an ellipse but anyway): how does a Left Step PT resolve the question? DePiep (talk) 08:56, 16 November 2022 (UTC)

Well, the left step periodic table was not recommended by IUPAC. Scerri personally thinks it's the ideal version, though. It would also resolve the problem if it was adopted, because it is in complete adherence to Madelung's rule, as noted at Alternative periodic tables#Left-step periodic table (Janet, 1928): that says 6s (Cs and Ba) comes before 4f (La–Yb) comes before 5d (Lu–Hg). No split d-block is possible there. Double sharp (talk) 09:47, 16 November 2022 (UTC)

Thanks, though I thought the "[is/not] recommended by IUPAC" and ".. as published on their website" source quotes as quality we are abandoning, except as substantiated papered Reports. DePiep (talk) 12:04, 16 November 2022 (UTC)

@Sandbh: Scerri has clearly not released all the correspondence. He specifically said that "I [he] won't go through the details here". Therefore, he is privy to information that we are not privy to, and what he says about the saga of this at IUPAC is clearly more reliable than what we might glean from the limited amount he has revealed. I think it is unjustified to say that Scerri is "blatantly wrong" about what he wrote. Presumably he is in the best position to know what he meant. If I say at the dinner table "Could someone perhaps pass me the salt", it does not mean I don't want it. It just means I'm being politely indirect about it. Likewise, when Scerri writes "Perhaps a compromise could be reached on [Sc-Y-Lu-Lr]", that doesn't mean the project hasn't decided on it. Particularly not when Scerri clarifies in a public forum that it has.

Lars Öhrström's non-IUPAC email address is his email specifically listed on his page on IUPAC's website as a IUPAC member, so it doesn't seem a cause for serious concern. He is on the Inorganic Chemistry Division of IUPAC whose comments he relayed to Scerri ("From Div II we are happy with this result" – note the "we"), and so he is in a perfect position to have been speaking for the relevant IUPAC division. The idea that "the current IUPAC periodic table...can be interpreted either way", as you write, is odd. Setting aside the fact that it's not IUPAC-approved: no it can't, it clearly says that all lanthanides and actinides are placed in group 3. It even helpfully puts the words "lanthanoids" and "actinoids" in those places. The report, which you agree that the relevant IUPAC division is happy with per your point 2, explicitly makes that clear by graphing that version (Fig. 3) as a separate form from either Sc-Y-La or Sc-Y-Lu. In particular, even if it was an official IUPAC-approved PT (which it isn't), it is still not evidence in favour of the Sc-Y-La form you have been championing in this discussion. If IUPAC does not do anything, the "IUPAC periodic table" that is not IUPAC-approved anyway would show Sc-Y-*-** still; if we on WP did not do anything, we would be showing Sc-Y-La which you have been advocating a return to in this thread, and wrote a long external journal article in favour of. Those are two entirely different things, and in 3b and 4 you do not mention this difference at all. And finally, you say that Scerri's presentation was given in "April 2021" when the title of the very video says "March 2022". Double sharp (talk) 09:47, 16 November 2022 (UTC)

@Double sharp: Thanks, sorry, I stand corrected; Scerri gave the talk in March 2022.

The appearance of the IUPAC table does not imply all Ln and An are placed in group 3. That is just one (dubious) interpretation. As Jensen (who was a member of the IUPAC group 3 project) wrote of this interpretation:

"I can hardly believe that a modern inorganic chemist would advocate such an antiquated interpretation of these elements, unless, as noted above, they have lost all contact between the underlying premises of their periodic table and the facts of chemistry."

The fact is that the table appearing on the IUPAC web site has contributed to the confusion as to the composition of Group 3. OTOH, as I indicated that Lars wrote, maybe this is not such a bad thing.

Please do not conflate separate ideas and then imply they are somehow connected with my support for La-Ac (appearing in the lede of the PT article) . Of course, as you wrote, "that" version (Fig. 3) is separate from from either Sc-Y-La or Sc-Y-Lu. And as you wrote, it is neither IUPAC-approved nor is it evidence in favour of Sc-Y-La, nor have I ever claimed to the contrary.

I repeat again, where you wrote: "The important thing is that the report's conclusion in favour of Sc-Y-Lu-Lr..." is false. All the report said was:

"Perhaps a compromise could be reached on the table depicted as figure 2 [with Lu-Lr in group 3] since it achieves three desiderata."

The fact is there is no officially released advice from IUPAC pertaining to the composition of group 3.

Further, according to Scerri, from the transcript of the presentation:

"so i'm afraid that we this has not been resolved it's been left hanging and i'm afraid it's i this is just a personal opinion which i hesitate to say publicly but i think they're they're copping out iupac is afraid of a pluto situation or something like it where a major change to the periodic table occurs and everybody freaks out and says no no you've destroyed my favorite periodic table."

I repeat: The fact is there is no officially released advice from IUPAC (as a whole) pertaining to the composition of group 3.

PS: I note at 51:39 of the presentation that Eric says "some" textbooks show the split d-block version. In fact, per the project team's own research, the split d-block form appears in the majority of textbooks from 1972 to 2010, and in a plurality of textbooks during the 2010s, up to 2018. Sandbh (talk) 04:16, 17 November 2022 (UTC)

@Sandbh: the table appearing on the IUPAC web site has contributed to the confusion: I think you are talking about your own situation. Time and time again you are claiming that textbook tallying can be an argument for confirming/rejecting the IUPAC 2021 report.

Recap: current problem in the article is not to somehow falsify the IUPAC 2021 report (which is you have not arrived at halfway even on this talkpage even, BTW). The current quest for the article its to write the encyclopedy to reflect existing scientific situation. Current section re Group 3 is not doing anything of that. (plus, the notions that (17Nov-A) Group 3 element is the place where this should be handled in the first place, (17Nov-B) possibly with wider backgrounds such as history, and (17-NovC) the IUPAC-named option Sc-Y-0-0 is still absent; all encyclopedic not in-article POV-bickering).

This has been posted for you multiple times recently here, but still you are not responding to or resolving. DePiep (talk) 07:56, 17 November 2022 (UTC)

DePiep, what do you consider missing from the current version? I can look into it and start writing a draft to address your concerns, but first I'd like to clarify them a bit more. I think A is addressed by the hatnote saying that Group 3 element is the main article, and I just added a sentence to explain C to Periodic table#Group 3, but what do you have in mind for B? Double sharp (talk) 11:33, 17 November 2022 (UTC)

eh, Double sharp, "missing" from the current version? That would be: someone should add self-criticism. I can't be shorter without repeating myself, like from 07:56. As the Wernher von Braun team said when designing Saturnus: "We must add some lightness".

It needs rewriting from the bottom up. Or: simply back to the Feb/Mar 2020 version. DePiep (talk) 05:38, 18 November 2022 (UTC)

@DePiep: I don't think the Feb 2020 version could work, as it's even longer. I recently significantly shortened it, so that it now mentions the versions in one paragraph and summarises the argumentation in one as well. I really don't see how to shorten it any further myself while still at least explaining why the argument exists. It is of course very possible that I can't figure out how to do this myself just because I've been looking at this text for too long, which is why I'm asking for your opinion. (Maybe I should ask some others, too.) Double sharp (talk) 22:10, 18 November 2022 (UTC)

Well, I only mentioned that Feb2020 as a idea-of-reference: the version incorporating the IUPAC2021 report, and before it was questioned on this talk. That is: before the gotcha-,-IUPAC wording came in via this talk (ca. 14 Oct 2022).

New, at least to me, is the Pluto-situation you/Scerri explained in this News-From-section. I'd surely expect that in, as it is clarifying a lot. Then, I've mentioned that full history nor PTgraphic-as-proof is not needed, also as a frame of mind. But I better not write or edit the full sentences & science describing it; lack of English, lack of science. That too is why I'd need more time to respond to deeper content. I do can react to the impression of paragraphs, eg, 'is this how I myself can understand it?'. I note that I also use the word "encyclopedic", referencing an October 2019(?; can't find now) talkpost from a non-member. On the article in general, I noticed that over last years it has become very specialistic (physical atom models). Something to point talk about elsewhere & elsetime.

Main point is that if we agree on the basic requirements (both wrt ins and outs), the result will be good for starters. More eyes are welcome sure. DePiep (talk) 05:59, 19 November 2022 (UTC)

@Double sharp: a simple request. The wording Brauner's asteroid hypothesis are not common(-ly) accepted nor used, and are not easily linkable(!). So we better describe instead of naming? For me, a reuse of the IUPAC phrasing like "third option, relevant for specialistic relativistic approach" would do. Link to Group 3 element-section would solve a lot, and be in place here. DePiep (talk) 06:10, 19 November 2022 (UTC)

@DePiep: Brauner's asteroid hypothesis was already mentioned in the history, under Periodic table#Mendeleev, since he was one of the consolidators of the periodic law. I've added a short parenthetical description "(which gave all the rare earths one place in the table)"; is that better?

I've altered the "main article" link to Group 3 element#Dispute on composition. And I think I've explained the "Pluto" better this time. Double sharp (talk) 11:38, 19 November 2022 (UTC)

@Sandbh: You appear to have overlooked the difference in your previous post, unless I read you wrong. Under 4, you wrote your view that "the decision to change the PT in the lede of the PT article to Sc-Y-Lu-Lr was premature". In 3b, you quoted Öhrström when he said "It may even be premature to do anything right now", and approvingly noted "Quite so". But if IUPAC did not do anything, then the non-IUPAC-approved "IUPAC periodic table" would show Sc-Y-*-**. If we did not do anything, then we would be showing Sc-Y-La-Ac. These are two different things.

I have already explained why the words "Perhaps" and "could" do not necessarily mean the report is not saying what it appears to me to be saying. And we have the author's word backing up the interpretation I have given. Double sharp (talk) 11:39, 17 November 2022 (UTC)

@Double sharp: (i) Que? I don't understand your point re me overlooking something. I do agree with you about what I wrote.

(ii) Earlier you referred to something I wrote running a bit too close to editorialising for your taste. I feel the same way about your interpretation of "perhaps" and "could". The provisional report, which is only a progress report, and which understandably lacks a firm, bold, and clear recommendation, is the only published RS we have so far. As well as a possible conflict of interest, something else Scerri said in his presentation was not accurate. He said "some" tables show Sc-Y-La-Ac. Well, in fact, according to the Group 3 project's own research, most tables did so in the four decades from 1972 to 2010, and a plurality of tables did so in the 2010's up to 2018. That is something more than "some". Sandbh (talk) 03:27, 18 November 2022 (UTC)

@DePiep: Thank you for posting your apparent concerns. You wrote:

the table appearing on the IUPAC web site has contributed to the confusion: I think you are talking about your own situation.

The situation is referred to in the periodic table article:

"A compromise can also sometimes be found, in which the spaces below yttrium are left blank. This is a reversion to Brauner's asteroid hypothesis,[53] and leaves it ambiguous if the group only contains scandium and yttrium,[179] or if it also extends to include all thirty lanthanides and actinides.[53]

You further wrote:

Time and time again you [Sandbh] are claiming that textbook tallying can be an argument for confirming/rejecting the IUPAC 2021 report.

No, I have never claimed this. No, I have never done so on a time and time again basis. If you have any evidence to the contrary please post it here.

Instead, what I have said (in response to comments posted by yourself and Double sharp) is:

The IUPAC report is a provisional report by the project team commissioned by IUPAC to make a recommendation to IUPAC on the composition of group 3.
The provisional report has not made a recommendation as such nor has IUPAC as a whole indicated its formal acceptance of the report, nor has IUPAC as a whole consulted with the chemistry community, nor has IUPAC as a whole published a formal recommendation on the composition of group.
The IUPAC group 3 project team conducted a textbook tallying survey and found that La-Ac was the most popular form over the four decades from 1972 to 2010, and represented a plurality during the 2010's (up to 2018).
Given the above three items, and since WP is an encyclopaedia, and university textbooks are RS, there is a case to be put that the lede PT in the PT article should be Sc-Y-La-Ac, per WP:ENC, WP:NPOV, and WP:DUE.
In the event that IUPAC as a whole ever publishes a formal recommendation as to the composition of group 3 then, as EdChem posted:

"WP is still hampered, in my opinion, by a preliminary report with recommendations / suggestions inconsistent with RS, that does not yet have the imprimatur of IUPAC itself. RS like textbooks will act on this (or not) in future editions, so any new consensus on a convention will take time to evolve."

~~What is it that you are unable to understand here? Sandbh (talk) 02:06, 18 November 2022 (UTC)~~

Sandbh: What is it that you are unable to understand here?. So after being invited to read & learn WP:NOTGETTINGIT, Sandbh you come back with an incivility. Please redact yself. -DePiep (talk) 05:31, 18 November 2022 (UTC)

DePiep I have struck out my comments. I apologise if I came over in an incivil manner. That was not my intent. Sandbh (talk) 05:37, 18 November 2022 (UTC)

Happy: Re the periodic table article I am reasonably happy with the way it currently discusses the group 3 issue, thanks to the endeavours of Double sharp. Sandbh (talk) 02:06, 18 November 2022 (UTC)

I am glad to hear that you are happy with the current discussion of group 3.

1. But I will very much be unhappy with a reversion of the lede PT to Sc-Y-La-Ac. This would not represent the majority opinion of current textbooks (noting that it dropped below 50%, not to mention that if we looked at textbook conventions, then judging from this A-level UK syllabus we're going to have a whole lot of them teaching that scandium and zinc are supposedly not transition metals and using hilarities like "ethanedioic acid" that nobody actually working on those things uses), the relevant literature focusing on the group 3 problem (self-explanatory), relevant monographs on the table as a whole (Scerri's two books on the PT), or the relevant IUPAC reports of 1988 and 2021 (equally self-explanatory, despite your refusal to accept what Scerri says about what he himself meant), or the recent unanimous (noting how rare that probably is on WP) 2021 consensus to follow the IUPAC report in full knowledge that it was provisional (and nothing has changed since).

2. Moreover, while the IUPAC report at least notes that there is a niche topic (relativistic quantum chemistry of superheavy elements) where Sc-Y-*-** makes sense, it does no such thing for Sc-Y-La-Ac.

3. Therefore, it appears to have no basis whatsoever for returning except for you continually pushing it against everyone else who is happy with the status quo (the OP hasn't responded to my most recent ping, originally wanted Sc-Y-*-** per the non-IUPAC-approved "IUPAC periodic table" instead of Sc-Y-La-Ac, and anyway removed the planned Sc-Y-*-** form from sandbox development with an edit summary of "Resolved").

4. And we equally have the WP:ALUM precedent of using IUPAC decisions when sources don't agree, even if they are not formally published as recommendations: the 2005 Red Book mentions aluminum and cesium as commonly used alternatives, but it uses aluminium and caesium as primary spellings in the main table, and those are what we use to ensure consistency despite the fact that (of course) more sources use American spellings.

5. And if we pedantically waited for updates to the non-IUPAC-approved IUPAC periodic table for everything, then we would not have had elements 113, 115, 117, and 118 (which all the relevant literature agreed were known) on the table with temporary "unun" names till 2015, yet we clearly did.

6. In summary, I will oppose any attempt to change the standard PT throughout WP to anything that is not Sc-Y-Lu as the IUPAC reports and the majority of the relevant sources per WP:CONTEXTMATTERS support, especially since it seems to be just a clear one-against-all given the strength of that unanimous 2021 consensus.

7. And given that you are selectively quoting Scerri in your poll draft in your sandbox, I will note that if you put up your poll (which there is no consensus to perform) in that form, I will most certainly point out that its opening statement is non-neutral as it does not quote Scerri with full context. Double sharp (talk) 02:17, 18 November 2022 (UTC)

@Double sharp: I've added a section break, and spacing, for ease of editing and comprehension. Revert if you wish.

@Double sharp: I've numbered your paragraphs to make it easier to follow the thread and my comments. Revert if you wish.

NB. The addition of a specifically named sub-section break was reverted by DePiep. I have now added a non-specific arbitrary break, for ease of editing purposes. Sandbh (talk) 05:32, 18 November 2022 (UTC)

NB. My non-specific arbitrary break, was reverted by DePiep. Sandbh (talk) 05:40, 18 November 2022 (UTC)

@Double sharp:

1. Sc-Y-La-Ac represent the majority of textbooks in the 2000s and, as you pointed out to me, a plurality of textbooks in the 2010s (up to 2018). Of the three options it is the most common.

There is nothing remarkable about teaching Sc and Zn are not transition metals, as is the approach of Rayner-Canham in his 2020 book, The Periodic Table: Past, Present and Future, (pp. 152–153), described by Eric Scerri as "an absolutely wonderful book".

As the 2021 provisional report of the IUPAC group 3 project found there is no objective way of adjudicating between the two versions, what you refer to as, "the relevant literature focusing on the group 3 problem (self-explanatory), relevant monographs on the table as a whole (Scerri's two books on the PT), or the relevant IUPAC reports of 1988" is therefore irrelevant.

Further, what Scerri says about what he himself meant is contradictory, in that the report of the IUPAC Group 3 project team, of which Scerri was but one member, is not consistent with what he said in a public forum with no IUPAC imprimatur.

The recent "unanimous (noting how rare that probably is on WP) 2021 consensus to follow the IUPAC report" violated WP:ENC, WP:NPOV; and WP:DUE.

2. Sc-Y-La-Ac makes sense from what Scerri & Parsons wrote in 2018:

"For the purpose of selecting an optimal periodic table we prefer to consider block membership as a global property in which we focus on the predominant differentiating electron."

This supports La in group 3, since such a table has one less differentiating electron discrepancy than an Lu table. But of course this is irrelevant given the 2021 provisional report of the IUPAC group 3 project.

3. The basis for Sc-Y-La-Ac potentially returning is, among other things, WP:ENC, WP:NPOV; and WP:DUE. Of course, this will be contingent upon achieving consensus via an RFC, which is by no means guaranteed.

4. I am fine with using the WP:ALUM precedent given how trivial a matter this and in the absence of a survey of university textbook usage, noting the English wp is hosted in the US AFAIK.

5. I am fine with mentioning elements 113, 115, 117, and 118 given RS referring to their existence and knowing that textbooks will invariably incorporate their addition, unlike the situation, necessarily, with formal IUPAC recommendations.

6. WP:CONTEXTMATTERS extends to the finding of the IUPAC 2021 provisional report that there is no objective method of adjudicating the question. That being so, WP:ENC, WP:NPOV and WP:DUE point to Sc-Y-La-Ac, since this is the form appearing in a majority/plurality of university textbooks.

7. What is in my sandbox can only be regarded as a work in progress. Thank you for raising your concerns about the wording. I will take this on and seek your opinion. I wasn't aware I needed a consensus to conduct an RFC.

--- Sandbh (talk) 06:43, 18 November 2022 (UTC)

@Sandbh: Let me just reply to 3 and 7, because to me that is the crux of the matter. You are correct that you do not need a consensus to conduct an RFC. Indeed, the purpose of an RFC is precisely to find out what consensus is. However, in the 2021 poll, everyone was made aware that the report was provisional. Despite that, the consensus went through unanimously anyway. So, there is clear and obvious evidence as to what the consensus is based on pretty much exactly the same information that we have now. (OK, we have Scerri's 2022 statement, but that if anything strengthens it: Scerri interprets it as the project suggesting Sc-Y-Lu-Lr, the presenter says it's a recommendation and he doesn't disagree, and his worries about IUPAC dragging their feet seem to be about updating the periodic table as a whole and the obvious contradiction of having a "IUPAC Periodic Table" that is not IUPAC-approved. Not about IUPAC rejecting the report's conclusion.) Yet that doesn't stop you from calling the result of the consensus "inappropriate". If the poll reruns now, it will be on pretty much the exact same information. So, how are we supposed to believe that if you try it again, and you don't get your way, you're not going to keep disputing the consensus because you don't like the result?

P.S. You may be amused to hear that 113, 115, 117, and 118 are still not in the periodic table given in the 2021 Singapore syllabus. So I think I have some reason to doubt that textbooks will always keep up-to-date there. Good for anyone who does, but I don't think it can be guaranteed. BTW, what do textbooks give for the electron configuration of Lr? And do they think astatine is a metal or not? :) Double sharp (talk) 12:12, 18 November 2022 (UTC)

DePiep, I have written a new draft to replace the current section on group 3 that you have criticised. It is probably too long, but it is easier for me to write something and cut later. I also think it's better to start asking you at this stage so that I know what you think should be kept, what you think should be cut, and what you think should be rewritten.

I have tried to do it this way:

One opening paragraph introducing all formats;
One paragraph explaining the scientific basis for each of the three forms. A little bit of history is unavoidable, because Scerri's analogy with Pluto is exact. Initially people thought Pluto was really big because they thought it was perturbing Neptune, and so that it should be a planet; but measurements kept coming in and it turned out to be really small, and people then started arguing based on that that it shouldn't be a planet after all. So here: initially people thought Sc-Y-La was correct because the electron configurations appeared to match better that way; but new measurements came in and it turned out that nearly all those configuration measurements were wrong, and people then started arguing based on that that it should be Sc-Y-Lu after all. Finally, just like Pluto, it turned out to be a matter of convention and how you want to pick your definition to best suit most users, as the IUPAC project noted. As inspired by Pluto#Planet X disproved, I left the details (electron configurations as known in 1934 vs 2019) to a side table.
A final paragraph detailing the conclusions of the IUPAC project.

This way, the rough structure can remain the same, while each paragraph gets shortened as needed.

I used the publications of the IUPAC project members (Scerri, Jensen, and Lavelle) as the sources, as they were the people chosen to make the decision on behalf of IUPAC. Consequently, I have left out a ton of arguments based on specific chemical and physical properties that have been used in the literature, firstly because including them would be obviously UNDUE, and secondly because Scerri already made the point in his book that they are all inconclusive. All of those are at User:Double sharp/Dispute on the composition of group 3 anyway as a separate draft not to be considered for this article.

I am not seeking Sandbh's commentary on this version, because I think the previous discussion has already made it clear that it's impossible to find a version that will satisfy both me and him: we already clearly have extremely different viewpoints on what the project report actually means, and since referring to what the project chair said in a lecture doesn't convince him, I don't think anything will. However, I'd like to at least get something that more than one of us can agree with, and I agree with DePiep's previously stated concerns. So let's start with the improvements. :) Double sharp (talk) 01:08, 18 November 2022 (UTC)

@Double sharp: As I wrote, I was reasonably happy with the periodic table article currently discussed the group 3 issue. I don't therefore understand why "it's impossible to find a version that will satisfy both me and him". Sandbh (talk) 06:48, 18 November 2022 (UTC)

@Sandbh: The article includes the lede table. I won't be happy unless it's Sc-Y-Lu per the IUPAC project and the consensus to follow it, and you clearly won't be happy unless it's Sc-Y-La. Unless that's wrong, I think my point stands. Double sharp (talk) 12:13, 18 November 2022 (UTC)

To all: irrespective of my opinion about the section, there is no ground to describe or redo a perceived academic dispute in this article.It does not clarify anything for the Reader. DePiep (talk) 12:33, 18 November 2022 (UTC)

@DePiep: Then what do you propose? The group 3 thing is an academic dispute that eventually got adjudicated by IUPAC. Is there then anything wrong with the current version of Periodic table#Group 3, in your opinion? The way I see it, it very briefly describes all the forms, mentions the problem causing the dispute, and then jumps immediately to the IUPAC project. Double sharp (talk) 13:09, 18 November 2022 (UTC)

The edits I reverted quoted two singled out words. Not from the scientists involved, not to clarify the situation, but to attack the secondary report. This article is not the place to discuss the Report. Since it is described, as you say, that is enough. (IOW, if IUPAC 2021 could be disputed, but not through this article.). hth. DePiep (talk) 14:16, 18 November 2022 (UTC)

I mean, writing "perhaps" and "could" (quotes included) is more of a gotcha approach. And this, after the more complete quotes that are already in there. DePiep (talk) 14:34, 18 November 2022 (UTC)

@DePiep: OK, thanks for clarifying. Is this wording better? Double sharp (talk) 14:39, 18 November 2022 (UTC)

yeah like it. DePiep (talk) 14:54, 18 November 2022 (UTC)

Though must say, the tq-text I just wrote above are the actual spellings I reverted. i.e., two individual words. It was not about the longer quote you edited out just now. Up to you. DePiep (talk) 14:59, 18 November 2022 (UTC)

@DePiep: OK, thanks for such a quick response. :) I'm happy with "suggested" since that's what Scerri said in his lecture.

Is the rest of the Periodic table#Group 3 section fine with you? Double sharp (talk) 15:07, 18 November 2022 (UTC)

That's the deeper research/read I have to postpone. Sorry. DePiep (talk) 15:10, 18 November 2022 (UTC)

@DePiep: No problem, take your time.

I've updated the section in the main article group 3 element, where some of the history belongs anyway, and based it on the publications of the IUPAC project members. For now I have left the section in this periodic table article alone, pending your deeper research/read. :) Double sharp (talk) 15:28, 18 November 2022 (UTC)

Sandbh you are breaking edit & discussion carefulness here. You are ignoring existing & building consensus here. I strongly suggest you revert yourself. (also, forking this very topic on my talkpage ie making this personal (as you did just now) is not the way to go. To be clear: first revert, then engage in existing talk. @Double sharp:. -DePiep (talk) 07:34, 23 November 2022 (UTC) (late sign my 06:57 UTC post +redo pinging @Sandbh and Double sharp:)

One of the earlier points worth repeating, as Sandbh refers to it in their editsummary: it is not just 'It's a in the source so I can use it in my advocacy". It is out of context (that context already discussed & in the section), and even in the source, an argument for a different reasoning (not the graph). Sort of IN-RS OUTOFCONTEXT CHERRYPICKING. DePiep (talk) 07:40, 23 November 2022 (UTC)

DePiep: In fact I was citing from the source exactly as it reads:

"Perhaps a compromise could be reached on the table depicted" [i.e. as Sc-Y-Lu-Lr].

There is nothing out of context here, neither is there cherrypicking. Those words in the provisional report go the heart of the matter namely that the group 3 project found there was no objective way to adjudicate the matter and that, in this light (as the report prints in black on white, on page 33) "Perhaps a compromise could be reached...". --- Sandbh (talk) 07:03, 24 November 2022 (UTC)

q.e.d.

You are not following nor building the article section thread. You are not digesting the talk that has been build. It has been explained to you. MOS:IN-RS OUTOFCONTEXT CHERRYPICKING says: "don't cut-and-paste reliable sources partially or out of that sources reasoning". DePiep (talk) 07:36, 24 November 2022 (UTC)

+ late @Double sharp:. Thanks for providing your wellwritten quote. DePiep (talk) 07:38, 24 November 2022 (UTC)

Some MOS-updating

The thread seems to be quieter now, except for some spillover on User talk:Sandbh: it appears that the since-2021 consensus to follow the IUPAC provisional report makes everybody happy except Sandbh, who is still cherry-picking the report, reading it selectively, and refusing to believe Scerri's clarification of what he meant despite the fact that Scerri is exactly the one who wrote that report. The OP seems to have withdrawn the proposal.

So since this seems to be the consensus, and this has already come up on this talk page about a year ago, I think all this is best put in the MOS so we don't have to go through an endless discussion again. To that end I have put in a section "The periodic table" at Wikipedia:Naming_conventions_(chemistry)#The_periodic_table, based on what seems to be general current practice: it covers (1) naming groups following IUPAC, (2) the scandium group following the 2021 IUPAC report, (3) 18 vs 32 columns interchangeably but necessarily making the same scientific statement, and (4) default colouring by blocks. Of these (2) and (4) have been consistently followed since 2021 and (1) and (3) since even earlier. I realise (4) is not actually connected to this topic, but since it is not under dispute but equally affects the appearances of our periodic tables by default, it seems like a reasonable thing to add mention of.

Comments welcome. Double sharp (talk) 02:13, 27 November 2022 (UTC)

@DePiep: Double sharp (talk) 02:22, 27 November 2022 (UTC)

I fully support these conclusions and subsequent changes. Thanks for the effort. DePiep (talk) 08:26, 27 November 2022 (UTC)

@DePiep: Thank you! I have added WP:PERIODIC, WP:PERIODICTABLE, WP:SCANDIUM, and WP:GROUP3 as shortcuts. (The first two link to the whole section, and the second two specifically to the one about group 3.) Double sharp (talk) 11:44, 27 November 2022 (UTC)

...and I added a couple of illustrations to make it exactly clear what was meant. Double sharp (talk) 11:54, 27 November 2022 (UTC)

This from my talkpage by Sandbh [6]. I will reply here.

==Wikipedia:Naming conventions (chemistry): Your reversion of my revert==
DePiep
With regard to your reversion [7] of my manual reversion [8] of the edits to Wikipedia:Naming conventions (chemistry), I reverted the content as it had nothing to do with naming conventions, and indicated this in the reason given for the edit.
Since the content I reverted has nothing to do with naming conventions I seek your agreement to (re)remove that content.
thank you,
— Sandbh (talk) 00:49, 2 December 2022 (UTC)

@Sandbh:. I use my words, YMMV. (1) the discussion has closed. You have had all opportunity and attention from other editors to make your point—it did not come through. (2) Wrapped up, all recent group 3-treads including those at your talkpage, lead the same result: text adjustments all right obviously in the articles, but no change in our general approach of group 3 descriptions & wikian preferences in our encyclopedia. IOW: no consensus for the change. (3) Time, reply size and patience spend by editors has been generous. (4) The MOS is part of this outcome. Double sharp is right to lay down the enwiki early 2021 consensus as established MOS. The MOS status is also given to our change from element classification coloring to block coloring (also 2021 IIRC [9]). Personally, I applaud this MOSsification by Double sharp, being more smart than bold, as it prevents biannual repetition of same discussions. (5) While maybe not exactly "naming convention", as a MOS it is allright. It is precise enough for me; we could alter that title, or split. (6) Change of MOS is possible via established routes as always, although trying to do so by prolonguing these same discussions I do not recommend, for example for reasons of reflection needed. -DePiep (talk) 08:50, 2 December 2022 (UTC)

I think it is most revealing that Sandbh did not remove the bit about using categories instead of blocks, just the bits having to do with group 3, even though both are arguably not quite "naming conventions". It is quite evident therefore that this may not be Sandbh's real reason. Combined with the difference between Nov 2021 (no endless complaints from Sandbh when he thought it was OK for WP to show Sc-Y-Lu) and Nov 2022 (endless stream of them now that he thinks it's not), it seems to me that it's rather that he cannot accept that neither the IUPAC project nor the consensus of WP editors agreed with him. "I can't possibly be wrong; all of them must be". Double sharp (talk) 10:57, 2 December 2022 (UTC)

I am trying to close this down without the need for BF questionings. So far, I think this can be done. Personals could detract, and make closure more difficult to explain for unfamiliar eyes. DePiep (talk) 11:42, 2 December 2022 (UTC)

@DePiep: Okay, I see your point. I have struck my comment, because it was made in annoyance and poorly phrased. Double sharp (talk) 11:48, 2 December 2022 (UTC)

@DePiep: I reverted the original edit since the content involved had nothing to do with naming conventions. You re-reverted my edit for no valid reason. Subsequently I posted to your talk page seeking agreement that the content involved had nothing to do with with naming conventions and seeking your agreement to (re)remove that content.

Since you haven't agreed to my request for the reversion of content unrelated to the page in question, in accordance with wp policy, I'll seek the involvement of an admin. I'll leave a notice on your talk page. ---- Sandbh (talk) 03:45, 4 December 2022 (UTC)

Don't forget to clarify (a) why you went to a personal talkpage instead of follow-up on an already existing section named § Some MOS-updating, and (b) why you ignore my 6-pointed reply here. DePiep (talk) 08:04, 4 December 2022 (UTC)

DePiep: (a) As I understand it I'm allowed to raise the matter at either a personal talk page (where I started the thread) or an article talk page; and (b) I mentioned your 6-pointed reply in my request for admin help, on my talk page. Sandbh (talk) 12:04, 4 December 2022 (UTC)

"allowed to"? -- nobody said otherwise. "I mentioned"? -- you did not reply. Anyway, be careful. DePiep (talk) 12:31, 4 December 2022 (UTC)

@Robertiki and Michael D. Turnbull: concluding; in my words (YMMV).

Looks like that for now this threads have concluded. "Group 3 is Sc-Y-Lu-Lr" is the preferred version to present (unchanged since enwiki started applying Scerri (2021) then); other versions to be discussed in-encyclopedia at the appropriate places (like Group 3 (periodic table)-sections). Discussion took place here and at this usertalkpage. To me, the discussion brought new depths and views (for example, the old group 3-version being comparable to the "Pluto is not a planet"-restlessness in science). Main conclusions, as established since Scerri (2021), have ended up in MOS:PERIODICTABLE. Future changes can be proposed through regular channels, though repetition is to be prevented. -DePiep (talk) 11:29, 12 December 2022 (UTC)

[ReferenceA-1] Magnusson, E. (1990). "Hypercoordinate molecules of second-row elements: d functions or d orbitals?". J. Am. Chem. Soc. 112 (22): 7940–7951. doi:10.1021/ja00178a014.

[2] Braïda, Benoît; Hiberty, Philippe C. (2013-04-07). "The essential role of charge-shift bonding in hypervalent prototype XeF2" (PDF). Nature Chemistry. 5 (5): 417–422. Bibcode:2013NatCh...5..417B. doi:10.1038/nchem.1619. ISSN 1755-4330. PMID 23609093.

[3] H., Cockett, A. (2013). The Chemistry of the Monatomic Gases : Pergamon Texts in Inorganic Chemistry. Smith, K. C., Bartlett, Neil. Saint Louis: Elsevier Science. ISBN 9781483157368. OCLC 953379200.{{cite book}}: CS1 maint: multiple names: authors list (link)

[4] Lein, Matthias; Frenking, Gernot (2005-01-01). "The nature of the chemical bond in the light of an energy decomposition analysis". Theory and Applications of Computational Chemistry: 291–372. doi:10.1016/B978-044451719-7/50056-1. ISBN 9780444517197.

[5] Scerri, ER (2019). "Dec 2019 survey of about 200 university textbooks". IUPAC Group 3 project.; Scerri, ER (2020). The Periodic Table, Its Story and Its Significance (2 ed.). New York: Oxford University Press. p. 392.

[6] Scerri, ER (2019). "Dec 2019 survey of about 200 university textbooks". IUPAC Group 3 project.; Scerri, ER (2020). The Periodic Table, Its Story and Its Significance (2 ed.). New York: Oxford University Press. p. 392.

[Jensen1982-7] William B. Jensen (1982). "The Positions of Lanthanum (Actinium) and Lutetium (Lawrencium) in the Periodic Table". J. Chem. Educ. 59 (8): 634–636. Bibcode:1982JChEd..59..634J. doi:10.1021/ed059p634.

[8] Scerri, ER (2019). "Dec 2019 survey of about 200 university textbooks". IUPAC Group 3 project.

[9] Scerri, ER (2020). The Periodic Table, Its Story and Its Significance (2 ed.). New York: Oxford University Press. pp. 377, 392.

[10] Cotton (and team), S (2022). "A comparison of the structural chemistry of scandium, yttrium, lanthanum and lutetium: A contribution to the group 3 debate". Coordination Chemistry Reviews. 455. doi:10.1016/j.ccr.2021.214366.

[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]