Wikipedia talk:WikiProject Elements

This page has archives. Sections older than 60 days may be auto-archived by if there are more than 4.

Article alerts

Templates for discussion

02 Jul 2025 – Template:Infobox periodic table group/footer (talk · edit · hist) TfDed by Izno (t · c) was closed; see discussion
02 Jul 2025 – Template:Infobox periodic table group/header (talk · edit · hist) TfDed by Izno (t · c) was closed; see discussion

Redirects for discussion

07 Aug 2025 – Ca+2 (talk · edit · hist) →Calcium was RfDed by 1234qwer1234qwer4 (t · c); see discussion
07 Aug 2025 – Ca(2+) (talk · edit · hist) →Calcium was RfDed by 1234qwer1234qwer4 (t · c); see discussion

Peer reviews

06 Aug 2025 – Aluminium (talk · edit · hist) has been put up for PR by R8R (t · c); see discussion

Updated daily by AAlertBot — Discuss? / Report bug? / Request feature?
Click to watch (Subscribe via

RSS

Atom) · Find Article Alerts for other topics!

FA	A	GA	B	C	Start	Stub	FL	List	Category	Disambig	Draft	File	Portal	Project	Redirect	Template	NA	???	Total
29	0	97	102	129	96	33	0	172	320	3	1	117	1	22	8,906	228	10	0	10,266

TrophyBox

One of your project's articles has been selected for improvement!

Latest comment: 1 month ago1 comment1 person in discussion

Hello,
Please note that Magnesium, which is within this project's scope, has been selected as one of the Articles for improvement. The article is scheduled to appear on Wikipedia's Community portal in the "Articles for improvement" section for one week, beginning today. Everyone is encouraged to collaborate to improve the article. Thanks, and happy editing!
_{Delivered by — MusikBot ^talk 00:05, 30 June 2025 (UTC) on behalf of the AFI team}Reply

Nucleus Composition

Latest comment: 1 month ago3 comments2 people in discussion

I think I need to learn more before I try discussing this issue *laughs nervously* Sorry about that! Blackwoodm (talk) 02:46, 11 July 2025 (UTC)Reply

The Lithium § Isotopes section summarizes Lithium isotopes. So I don't know what more you want. Johnjbarton (talk) 02:59, 11 July 2025 (UTC)Reply

You're right! I realized pretty much everything I said in my post was completely ignorant, and I would have deleted it entirely had I known how. I didn't realize anyone had even responded. Thanks. Blackwoodm (talk) 03:06, 11 July 2025 (UTC)Reply

NUBASE2020

Latest comment: 3 days ago9 comments7 people in discussion

Updating the lists of isotopes to NUBASE2020 is now complete. There may still be new data after the March 2021 publication date that isn't cited, including unmentioned new isomers, but newer isotopes (as well as data from the same publications) are believed to be accounted for. –LaundryPizza03 (d c̄) 00:32, 18 July 2025 (UTC)Reply

OK. Now I am trying to extend this update to pages on the elements, on individual isotopes, and finally to List of nuclides (which can be reformatted when it has all updated data). Some of this, such as keeping half-lives consistent (up to rounding) one would wish were automatic, though I know that's not necessarily an easy task. But in order to do this, I have to use the same standards: Nubase2020 should be the sole source for nuclear data it contains (unless there is more recent work to cite) and superseded all older sources, less importantly the latest versions of AME and CIAAW for atomic weights. It should not be needed to give a citation to Nubase for every sentence that uses a fact from it (as is not done on the isotope pages) as that would be not particularly helpful and difficult to make consistent. In general there is more 'junk' to clear out than on the isotope list pages; and trivia is no less trivia for having a citation. So that's what the IP edits to such pages are, and I'd appreciate being, if not helped, then at least not hindered in dealing with this mess.

I'd like to ask also: is there a complete list of all isotopes that currently have their own pages? This would be useful to check. 2601:441:8500:B870:0:0:0:F1A7 (talk) 18:27, 27 July 2025 (UTC)Reply

If needed I would like to draw the attention of someone capable of addressing the issue to Template_talk:Infobox_isotopes_(meta)#Atomic_weight_references. I know that should not seem urgent but it has been three years since the original discussion of the matter. 2601:441:8500:B870:0:0:0:F1A7 (talk) 02:36, 28 July 2025 (UTC)Reply

I have not found any way of determining what isotopes/nuclides currently have their own page. I was considering changing the link for each isotope at List of nuclides to the isotope rather than the element so as to ensure (barring any isomer-designation problems) that every such page is linked to, and may have to adopt this solution though I don't consider it ideal to send users to the middle of the table. 2601:441:8500:B870:E0BA:CF17:F8D7:7D9C (talk) 13:49, 17 August 2025 (UTC)Reply

Perhaps if you could elaborate on the reason you want to determine which isotopes/nuclides have pages we can make better suggestions. Links to redirects named for isotopes would not be distinguishable from links to pages and a sea of red in the List of nuclides would not be great in my opinion. I don't understand the issue of "middle of the table". Johnjbarton (talk) 20:03, 17 August 2025 (UTC)Reply

Category:Isotope content page was created some time ago to answer this exact question. ^Complex/_Rational 23:34, 17 August 2025 (UTC)Reply

Thank you. That category is even maintained automatically. My goal here was to link each entry on List of nuclides to the individual page if it exists, and otherwise to the main 'Isotopes of ...' page. Linking all to the individual isotopes results (as I'm sure you know) in not red links but in links that jump into the middle of the isotopes page at the same nuclide the user just looked at; though consistent with some other isotope links on Wikipedia it seemed to me possibly confusing and less helpful. 2601:441:8500:B870:2881:42FE:869A:88DE (talk) 12:22, 18 August 2025 (UTC)Reply

Ok thanks that is clearer. I think a user clicking on ²⁹Si should land on Isotopes_of_silicon#Silicon-29, not the intro to Isotopes_of_silicon. Johnjbarton (talk) 15:33, 18 August 2025 (UTC)Reply

There are 64 of the 987 nuclides in that list of pages, not all of which (and none of the stable ones) are linked now. I have also noticed/remembered that the link will redirect to the section of the isotopes page in lieu of a separate page, and perhaps even more isotopes have that. Presumably even more of them should have one or the other, so redirecting in every case might not be terrible, even though I still don't see how it's more useful to be put in the middle of the list. 2601:441:8500:B870:B95E:3BB6:B1A7:3A16 (talk) 03:48, 25 August 2025 (UTC)Reply

Removal of NUBASE2020 sources

Latest comment: 29 days ago15 comments3 people in discussion

Editor @73.228.195.198 has been removing NUBASE2020 references on many pages with no or minimal edit summaries. @CodeTalker and I have reverted some of these, but the IP editor undoes the delete. Here are some examples: Samarium Lutetium Rhenium.

I asked for an explanation in my edit summaries but what I got was a post on Talk page that asserts they are justified in the edits and don't need to explain them. I do not agree with the removal of some references and I consider the edit summaries to be inadequate. Johnjbarton (talk) 18:57, 27 July 2025 (UTC)Reply

This is part of what I mentioned in the previous section, not a new topic. I am not removing Nubase2020, but attempting to remove sources _other than_ Nubase2020, since that is the appropriate source for nuclear data it contains. Nubase2020 is already cited at least once on the page (in the infobox) and I believe, as I've mentioned several times now, that Nubase2020 does not need to be cited for each individual statement, but (as on the isotope pages) just once, although I'm not particular as to where. Such a "general reference" principle could be overridden by consensus, but I know of none established here that would. In any case, even if you believe that you should be able to override it, you could show good faith, at least once, by adding the appropriate citations rather than reverting the whole edit - which restores information and often sources that should be obsoleted by Nubase2020.

In addition, my post on your Talk page attempted to explain this in detail and without unnecessary hostility and is not accurately characterized as you did above. Your post here shows that you do understand, in a general sense, the purpose of my edits and can't pretend to ignorance. 2601:441:8500:B870:0:0:0:F1A7 (talk) 20:12, 27 July 2025 (UTC)Reply

I do not understand the purpose of the poorly documented edits by the anonymous IP user 73.228.195.198. Those edits are deleting sources and should be reverted. The standards for sourcing are set by Wikipedia policy. Johnjbarton (talk) 20:24, 27 July 2025 (UTC)Reply

I comment, ignoring the questionable statement above, that Johnjbarton has revised one of the element pages I edited - Caesium, not one listed above. Unlike my suggestion, and despite his complaint, he added no references to Nubase2020; I added the general reference. However, he restored this passage (references removed for clarity):

Several of caesium isotopes are synthesized from lighter elements by the slow neutron capture process (S-process) inside old stars and by the R-process in supernova explosions.

I had good reason to remove this, as it is a general statement about nucleosynthesis with no particular relevance to caesium or its isotopes, and other elements have no corresponding statement. In addition it contains errors: grammar (several of) and capitalizing S- and R-processes, which is non-standard; it seems he was not interested in why I might have removed it. I do not try to remove against consensus, but I do not feel that there is or is likely to be meaningful consensus to keep statements like that as they stand. 73.228.195.198 (talk) 23:58, 27 July 2025 (UTC)Reply

You claim:

"it seems he was not interested in why I might have removed it."

directly contrary to my posts here and on my Talk page. I am interested in why you removed it. That is why asked for appropriate edit summaries. Your edit that removed sources from Caesium has a summary

* "isotope correction/cleanup edit - added section reference to Nubase making it clear information is sourced ... Tag: references removed"

Nothing at all about "Why" the sources were removed.

The addition of this line

"All nuclear data not otherwise stated is from the standard source:"

citing NUBASE2020 is contrary to Wikipedia policy which specifically calls for inline citations. This approach is impractical as well: future additions of unsourced content will be incorrect assumed to be sourced. Johnjbarton (talk) 00:55, 28 July 2025 (UTC)Reply

I added an inline NUBASE2020 source to Caesium in the only place in the isotopes section where it was needed. Johnjbarton (talk) 01:11, 28 July 2025 (UTC)Reply

- I have now explained "why", regardless of what you were or weren't thinking. Your silence in that regard can reasonably be interpreted as not having any reason to retain that sentence, which leads to the same conclusion again: that you had no interest in "why" unless it could be used to attack me. Further, sources are naturally removed when the information they are used to support is removed, as should be obvious. When I did remove that sentence, its two sources were also removed, and that isn't a separate act.

- The practice of giving the 'general reference', as you know, was specifically endorsed by PhilKnight after I asked him to give an explicit answer, so as not to mistakenly believe he had. The fact that you added an "inline" citation at the end of the paragraph, rather than each sentence, shows you agree that it's not reasonable to do for every sentence - but placement once for the paragraph is open to the same objection that you give (how seriously I can't tell) to the placement once for the section; namely, that unsourced content could appear to be sourced. I don't consider that a serious objection (especially if one actually knows what sort of information is in Nubase and can check quickly), but you should be judged by the standards you claim.

- Other points have already been made on your on or PhilKnight's page, in posts you at least should have read. 2601:441:8500:B870:0:0:0:F1A7 (talk) 02:32, 28 July 2025 (UTC)Reply

I assume you are referring to @PhilKnight's post on User_talk:73.228.195.198#July_2025 which records that you were blocked for exactly the reason I am complaining about here: "making numeric changes without explanation or sourcing." PhilKnight as admin agreed to unblock. The reply does not endorse your "general references" and a single editors endorsement does not weigh against a site-wide policy.

I've fixed the damage on Samarium. Please stop making changes to elements that add this incorrect form of referencing. Johnjbarton (talk) 02:46, 28 July 2025 (UTC)Reply

You know that the post I referred to is this one, because you replied to it showing you understood. Your reference to a different one is misleading and apparently only an excuse to mention my block, which he described in the log as 'mistaken'; I will not claim that he endorsed my actions completely by unblocking, but I can fairly assume he did not strongly oppose them. 73.228.195.198 (talk) 03:02, 28 July 2025 (UTC)Reply

As a way of moving forward, I offer to fix the sourcing in the isotope section of any of our main element article that you work on. Just ping me, I don't follow every element. Johnjbarton (talk) 22:31, 28 July 2025 (UTC)Reply

I do not really mind if you change the form of sourcing on any element; unlike you, I am not obsessed with that. But this edit you just made was not what you say; it reverted content (including the half-lives that I am most careful to keep consistent) and restored inappropriate sources (an old version of Nubase, and the 'source' Neufcourt2019 that is easily seen not to support anything in the article). I do care about the quality of sourcing, as should be clear from my edits. 73.228.195.198 (talk) 00:57, 29 July 2025 (UTC)Reply

I appreciate your efforts to try to improve these articles. However, Wikipedia is a community effort. At least three other editors also reverted your NUBASE2020-related edits. I had to spend quite a bit of time to repair the damage you did and all you have dished is additional argumentation. Being civil is not that hard, give it a try. Johnjbarton (talk) 16:39, 29 July 2025 (UTC)Reply

I am working alone here, not by choice, not in a 'community effort'. I'd have to point out that I don't think your own civility is very exemplary, though I'm not yet going to dredge through your history - argument about civility is rather pointless anyway. I just found three more elements where you had added/restored old versions of standard sources, which further shows that you are making more work for me - that is the whole reason I found your reverting objectionable to start with. 2601:441:8500:B870:0:0:0:F1A7 (talk) 22:34, 29 July 2025 (UTC)Reply

Under the additional IP address @2601:441:8500:B870:0:0:0:F1A7 the poorly or unexplained removal of sources has continued. The result is poor sourcing. Many of the Elements articles are FA or GA articles and these changes are damaging their quality. Is this what we want? Some editors that have reverted similar edits include @CodeTalker @Isaidnoway @Arcrev1 @PhilKnight @DMacks.

In my opinion we should ask this editor to stop editing rather than remove sources in existing articles without discussion. Johnjbarton (talk) 23:37, 29 July 2025 (UTC)Reply

Referring to me in the third person is definitely rude, you know my IP is not static, and including PhilKnight (at least) in the above list is clearly misleading - I do not see how it could not be intentionally so. 2601:441:8500:B870:0:0:0:F1A7 (talk) 23:42, 29 July 2025 (UTC)Reply

Suitability of NUBASE2020 as a source.

Latest comment: 23 days ago12 comments4 people in discussion

In the process of looking a many edits related to the isotope sections of the primary articles on the elements I have looked into the NUBASE2020 source:

Kondev, F. G.; Wang, M.; Huang, W. J.; Naimi, S.; Audi, G. (2021). "The NUBASE2020 evaluation of nuclear properties" (PDF). Chinese Physics C. 45 (3): 030001. doi:10.1088/1674-1137/abddae.

This is clearly a reliable source for aspects of individual isotopes, but I find that is being applied in a way that I believe to be inconsistent with Wikipedia's goal of WP:Verifiability: "verifiability means that people can check that facts or claims correspond to reliable sources."

The NUBASE2020 source is a long table with at least one row for every isotope. I have no problem with the use of this data for particular isotopes. Note that the table is organized by mass, not by atomic number and include isomers.

However, many articles use this source to verify summaries across the isotopes. Some examples:

There are 22 known isotopes of phosphorus, ranging from ²⁶P to ⁴⁷P.
Naturally occurring samarium is composed of five stable isotopes: ¹⁴⁴Sm, ¹⁴⁹Sm, ¹⁵⁰Sm, ¹⁵²Sm and ¹⁵⁴Sm, and two extremely long-lived radioisotopes, ¹⁴⁷Sm (half-life t_1/2 = 1.066×10¹¹ years) and ¹⁴⁸Sm (6.3×10¹⁵ years), with ¹⁵²Sm being the most abundant (26.75%)
Lighter unstable isotopes of samarium mainly decay by electron capture to promethium, while heavier ones beta decay to europium.

How would we verify these claims against the NUBASE table? Counting all of the phosphorus rows seems very error prone. Verifying the decay paths is similarly an exercise in analysis.

If these summaries are known to be straightforward to verify, then I think we should share that knowledge so every editor can verify the content. The verification process could be documented in the {{NUBASE2020}} template. Johnjbarton (talk) 00:23, 30 July 2025 (UTC)Reply

They aren't easy to verify in that sense. But there is no better source for nuclear data, other than our own lists of isotopes, which put Nubase into a more accessible form and improve on it as possible. That's why it was important to finish updating them. For that reason, our citations of Nubase on any other pages are really superfluous, done only to satisfy policy that prevents us from citing ourselves. From our lists (which are no less accurate than Nubase2020 is), those things are easily seen and I would expect people to verify them that way. 2601:441:8500:B870:0:0:0:F1A7 (talk) 02:16, 30 July 2025 (UTC)Reply

Thanks, in my view you are confirming that NUBASE does not verify these claims, they are original research. To provide encyclopedic value, data should be put in context with explanations referenced to independent sources.

While I appreciate your enthusiasm, I think this kind of goal is a mirage and disservice to readers. Wikipedia is intrinsically unreliable by design. NUBASE2020 and similar published reviews are written by experts and professionally reviewed. Wikipedia will never replace them and we should not attempt to do so. Quite the opposite: our role is summarizing sources, aiming for a broad view and the best sources. Concentrating the isotope sections on NUBASE2020 is contrary to Wikipedia. We should use NUBASE for what it is good for, data on specific isotopes and we should summarize other sources for notable overviews of elemental isotopes. Johnjbarton (talk) 23:38, 30 July 2025 (UTC)Reply

In fact, there are diagrams in the first few pages of NUBASE that represent trends in nuclides and do not require OR or SYNTH to verify. The chart of nuclides on page 11 (page 12 of the PDF document), for instance, illustrates the most common decay mode of each nuclide, and one can easily read any of those charts to determine the lightest and heaviest isotope of each element. And when there are exceptions, most often for isotopes discovered more recently than NUBASE was published, there is an additional citation present.

Regarding naturally occurring isotopes and atomic weights, the abundance for each isotope can be readily verified in the NUBASE data and I believe it falls within the purview of WP:CALC that the sums of the abundances for all naturally occurring isotopes must add up to 1 (100%). Similarly, determining decay products for each stated decay mode is also a routine calculation, so it is sufficient to cite either the decay mode or the decay product by itself. ^Complex/_Rational 01:55, 31 July 2025 (UTC)Reply

Ok thanks, I added Template:NUBASE2020/doc#Verification based on your post. Johnjbarton (talk) 14:48, 31 July 2025 (UTC)Reply

(written before ComplexRational's edit, with which I concur.)

They are not original research. Verifiability does not require the verification be easy: we are allowed to use sources that are not online, which are probably even harder for most readers to check. But all the facts are verifiable if you really want to verify them. Wikipedia is used by many as a scientific reference (as was just admitted to me on this talk page, "all scientists use Wikipedia to some degree") - we are used because we are extremely convenient, fairly reliable, and pretty comprehensive, and users all the time are trying to make us more reliable and comprehensive. It is not acknowledging that that is more 'a disservice to our readers'.

Nubase2020 is not the only source we can use, but in practice it is the only compilation we should use for what it covers, as no others are found sufficiently accurate. Other than that, Wikipedia itself can in fact provide a better overview than any other source, and it addition can more reliably remain up to date. 73.228.195.198 (talk) 02:07, 31 July 2025 (UTC)Reply

Wikipedia is convenient and comprehensive, but AI summaries beat it in those categories. They are similar in reliability in my experience. The big remaining advantage of Wikipedia is sources, which AI so far completely fails. (I have found that if you push the chat AI hard enough for sources it will eventually give you a list from a Wikipedia page). That is another reason for us to take special care with sourcing. Johnjbarton (talk) 16:44, 31 July 2025 (UTC)Reply

Having familiarity with Google AI, as everyone does today, and other common AIs available to the public, it is difficult to take such a comparison seriously. AI summaries are often inaccurate (sometimes grossly so) and far from comprehensive; one reason is that AIs are incapable of saying "I don't know" or otherwise not providing information in response to a question. At that point it should be clear that our human editors still do a better job than AIs. Sourcing is a big part of it but it's using sources, more than citing them, that matters to overall utility. If an AI were programmed to give a proper citation for all sources, the quality of its output would not be improved - because that wouldn't make it more able to distinguish what sources are worth using. It is human judgement in one form or another that makes Wikipedia - or any reference work - more than the sum of its parts. 73.228.195.198 (talk) 04:28, 2 August 2025 (UTC)Reply

Relatively few Wikipedia articles are edited by experts; we should and do rely on reviews with many citations. I think AI uses roughly the same criteria.

There are not many studies on AI use vs other sources, but they are starting to appear. They match my informal take that users will accept the AI results for the most part.

Google users are less likely to click on links when an AI summary appears in the results

AI summaries cause ‘devastating’ drop in audiences, online news media told Johnjbarton (talk) 16:12, 2 August 2025 (UTC)Reply

I cited my own experience with AI, rather than appealing to such studies, for a reason: that's what I know. I don't know how others use Google, nor do I know how often, when they stop after reading an AI summary, they are right to do so given what they want to know. But I myself have not found Google AI to ever give me an answer that is correct and that I consider satisfactory, while Wikipedia has often done so (even without following any of the citations). I suppose I can't give a proof that Wikipedia is better than AI, but I think the fact that we still appeal to both readers and potential editors is reasonably good evidence.

As for Wikipedia editors being 'experts', that is in fact a matter of degree. Of course few editors are publicly-recognized experts in the field they edit, but they likely have more expertise than the average person simply because of their interest, especially in technical subjects as here. Very rarely would I want to argue (as I have not here) that their expertise is sufficient by itself to make sourcing completely unnecessary or that their original research is worth including, but it is sufficient to matter when it comes to making judgements about sourcing and what is original research or likely so. Verifiability to sources and editorial judgement by Wikipedians are not contradictory, but complementary, and both are equally necessary to the utility we do have.

It is telling that the place where collaborative editing breaks down is not where there are issues of sourcing, but on controversial topics on which there is usually an abundance of sources on both sides, and that is chiefly because contributors there lack any expertise or do not use judgement because of their bias. It is on such matters that improvement is most difficult relative to interest in the subject. Having good faith, which is the same as believing one is acting in good faith, is not sufficient to render one able to make improvements. And I believe that you, too, have good faith in that sense. 73.228.195.198 (talk) 05:46, 4 August 2025 (UTC)Reply

Again we have completely different experiences. I find that most controversies are arise when self-proclaimed experts want content different from a balanced view of the sources. Johnjbarton (talk) 15:57, 4 August 2025 (UTC)Reply

This would be, I suppose, the experience when dealing with fringe theory proponents, which you must have encountered, but I was thinking of Wikipedia in general. But both fringe-theory people and the more conventional POV warriors either do not understand the purpose of Wikipedia or deliberately fail to apply it, which is really just a restatement of my points above that you did not dispute, and I assume understand. 73.228.195.198 (talk) 05:06, 5 August 2025 (UTC)Reply

PR of aluminium

Latest comment: 21 days ago1 comment1 person in discussion

Hi all! I'd like to continue where I left off several years ago and bring aluminium to the FA standard. When I left, I thought we were close to the FA standard, but I particularly want more input from other readers/editors to ensure this top-importance article is good to go to the FAC. Any comments are welcome at Wikipedia:Peer_review/Aluminium/archive1. Cheers--R8R (talk) 18:20, 6 August 2025 (UTC)Reply

Isomer database

Latest comment: 10 hours ago2 comments2 people in discussion

The Discovery of Nuclides Project recently launched a new isomer sub-database for nuclear isomers with half-lives longer than 100 ns. However, coverage of this database is currently very incomplete — most of the post-NUBASE2020 discoveries have an entry, including an isomer of silver that is not in our listings, but ^26mNa does not. –LaundryPizza03 (d c̄) 01:21, 22 August 2025 (UTC)Reply

Can we get the total number of isomers? A minor problem I've noticed on Wikipedia is that the number of 'known nuclides' did not include isomers, being only ~3300, whereas List of nuclides does include isomers in its count of 987, so they are not truly comparable. These numbers are also used on a few other pages. It would be useful to have a reference for the total known nuclides (including isomers) better than counting the entries in Nubase. 73.228.195.198 (talk) 23:55, 27 August 2025 (UTC)Reply

Add topic