Wikipedia talk:WikiProject Elements

From Wikipedia, the free encyclopedia
  (Redirected from Wikipedia talk:ELEM)
Jump to: navigation, search
(edit · watch · refresh · history · Old notices · RC for main articles · Article alerts)
WikiProject Elements (Rated Project-class)
WikiProject icon This page is supported by WikiProject Elements, which gives a central approach to the chemical elements and their isotopes on Wikipedia. Please participate by editing this article, or visit the project page for more details.
 Project  This page does not require a rating on the quality scale.
Article alerts

Featured article candidates
Good article nominees
Requested moves

Featured article FA  A-Class article A   GA  B-Class article B  C-Class article C  Start-Class article Start  Stub-Class article Stub  Featured list FL   List  Wikipedia Book Book  Category page Category  Disambiguation page Disambig   Draft   File   Portal   Project  Redirect page Redirect   Template   NA   ???  Total
25 0 91 35 98 85 32 1 176 131 265 5 0 115 1 16 3,374 230 0 1 4,681

IUPAC on group 3[edit]

For future reference, I note the IUPAC project here. Methinks this is worth a top mentioning.

  • Chair of the project is Eric Scerri.
  • Apparently IUPAC has abandoned the Sc/Y/*/** option altogether. Group 3 does not consist of Sc, Y, all lanthanides and actinides (would be 32 elements in total).
  • It states that the presentation form (in 18- or 32 columns) is not related to the content (especially not the group 3 constitution).
  • It nicely uses the descriptive "32-column" wording, not the ambiguously defined "long form" wording. -DePiep (talk) 08:30, 25 November 2016 (UTC)

General notes, added later:

  • The outcome will possibly imply a change of the periodic table structure, away from the 1945 Seaborg version (that puts all lanthanides and actinides in group 3). That would be huge in PT history. -DePiep (talk) 08:55, 26 December 2016 (UTC)
  • When announcing the four new element names and symbols, IUPAC has published this this version, dated 28 November, 2016. It is still showing the Seaborg Group 3 constitution (Sc/Y/*/**). -DePiep (talk) 08:48, 18 January 2017 (UTC)
  • The RfC on group 3 has closed. -DePiep (talk) 08:48, 18 January 2017 (UTC)
Thank you DePiep, for posting this here. On a slow day, I happened upon this "news item(?)" discussing the placement of Lr in the periodic table. It isn't very good but is interesting in that it mentions Jensen's, Lavelle's and Scerri's views on the placement question (but see also here, for a different opinion by Jensen), and quotes Jan Reedijk, president of IUPAC's inorganic chemistry division, on how long it might take IUPAC to make up their mind, once Scerri's project team makes their recommendation:

Reedijk, of IUPAC, looks forward to Scerri's report. But he cautions that IUPAC's deliberations will probably be slow. When IUPAC proposed modifying the periodic table's column numbering in 1985, it took about five years to decide and another 10 to 15 years for chemists to adopt the changes, Reedijk says.

Joy. Sandbh (talk) 04:26, 26 December 2016 (UTC)
Interesting link indeed (from May 2015, so when the Lr ionization energy was first published. Scerri's project team had not even started then).
Wrt the "5+15 years" remark, I can say this. Back when the group numbers changed, there was an alternative valid set (two even: US and European style). These sets are still valid. Current group numbering was just an improvement/change, that took a long time to replace because there was no urgent need to do so. However. In the case of Group 3 constitution, the current form will be declared invalid. Today no source can be found to claim that group 3 has those 32 Sc/Y/*/** elements. So it must be replaced into one form or another. Now, even the latest IUPAC PT (28 November 2016, adding the four new names) says old style Sc/Y/*/** because the Scerri-project has not published yet. Unless the Scerri-project has some surprise up their sleeve, when they publish the Sc/Y/*/** graphic will be illegal. Publishing it will be a scientific fraud. IUPAC will have to publish a new version, or two.
We at Wikipedia can easily adopt the new version(s). For reasons we know, we are free to pick a preferred version etc. Enwiki will publish a correct new version. When, over 15 years, the scientists finally have agreed on the obvious, all young scholars (under 25 y) already will have learned the correct, legal version(s) by heart—from us. -DePiep (talk) 08:44, 26 December 2016 (UTC)
About IUPAC writing 'itselve'
<rant>Has no one else cringed at "The task group will only concern itselve [sic] ..."? I checked, and it is actually on the IUPAC website. By my en-us ear, it should be "The task group will only concern itself ..." but it may be that en-uk would prefer "The task group will only concern themselves ..." but unless I'm missing something, "itselve" would )be equally unacceptable in any ENGVAR. Does someone have a contact who could suggest they fix this? It certainly makes the IUPAC look to be a bit illiterate. </rant> YBG (talk) 18:37, 26 December 2016 (UTC)
Send an email to the task group Chair, Eric Scerri. Sandbh (talk) 21:35, 26 December 2016 (UTC)
Why didn't I think of such an obvious thing? (Probably 'cause I was wrapped up in my rant.) I'll e-mail him in a day or two. YBG (talk) 21:47, 26 December 2016 (UTC)
I am beginning to get why such a process can take 15 year. -DePiep (talk) 00:17, 27 December 2016 (UTC)
I checked, and it is actually on the IUPAC website. [YBG]. Y'know YBG, I do can copy/paste. -DePiep (talk) 02:04, 14 January 2017 (UTC).
👍 Like I generally can also, but then my cursor goes all weird on my and I wind up typing someplace I don't expect, which is fine when I notice it but when I don't, I've been known to unintentionally edit other folks contributions to talk pages. Not facebook not like thumbs down.pngDislike . YBG (talk) 02:20, 14 January 2017 (UTC)
Allow me to make this singular winning point. On average, 45% of my 1000 edits are spelling fixes. -DePiep (talk) 02:40, 14 January 2017 (UTC)
Do you have a WP:RS for that factoid? Or is it WP:OR? Oh, wait a minute. It doesn't meet WP:N. All I can say is I'm glad no one is keeping a running count of the spelling errors I create. Cheers! YBG (talk) 04:20, 14 January 2017 (UTC)
  • Does someone know a good source (wiki article) for that "Glenn T. Seaborg" (1945) table I keep mentioning? The one when Seaborg first puts the LN and AN over there, that way? -DePiep (talk) 02:57, 14 January 2017 (UTC)
There is a good (non-wiki) source here (p. 128). He appears to treat the lanthanides as Ce–Lu and the actinides analogously (Th onwards). Sandbh (talk) 06:02, 17 January 2017 (UTC)
Interesting enough. This is the first time that I can think of when I see not only hydrogen but also aluminum taking two positions in the table instead of the usual one. The intent is clear; that's just interesting. Thank you for sharing.--R8R (talk) 21:37, 19 March 2017 (UTC)
I'm looking forward to seeing you mention that as well when you get to Al! There are several good chemical reasons for thinking of Al in that position as well as the standard one. Double sharp (talk) 02:06, 23 March 2017 (UTC)
Thanks. -DePiep (talk) 21:11, 16 February 2017 (UTC)
(Intentionally undated post, just to keep this from being archived. DP).

WikiJournal of Science promotion[edit]

WikiJournal of Science logo.svg

The WikiJournal of Science is a start-up academic journal which aims to provide a new mechanism for ensuring the accuracy of Wikipedia's scientific content. It is part of a WikiJournal User Group that includes the flagship WikiJournal of Medicine.[1][2]. Like Wiki.J.Med, it intends to bridge the academia-Wikipedia gap by encouraging contributions by non-Wikipedians, and by putting content through peer review before integrating it into Wikipedia.

Since it is just starting out, it is looking for contributors in two main areas:


  • See submissions through external academic peer review
  • Format accepted articles
  • Promote the journal


  • Original articles on topics that don't yet have a Wikipedia page, or only a stub/start
  • Wikipedia articles that you are willing to see through external peer review (either solo or as in a group, process analagous to GA / FA review)
  • Image articles, based around an important medical image or summary diagram

If you're interested, please come and discuss the project on the journal's talk page, or the general discussion page for the WikiJournal User group.

  1. ^ Shafee, T; Das, D; Masukume, G; Häggström, M. "WikiJournal of Medicine, the first Wikipedia-integrated academic journal". WikiJournal of Medicine. 4. doi:10.15347/wjm/2017.001. 
  2. ^ "Wikiversity Journal: A new user group". The Signpost. 2016-06-15. 

T.Shafee(Evo&Evo)talk 10:39, 24 January 2017 (UTC)

+X and +Y in Isotopes of a element articles[edit]

What is meant by "+X" or "+Y" in a excitation energy column? Please see Isotopes_of_rhenium. There are also some other confusing values - without a explanation - in this article, including "non-exists" for 168mRe and "0(100)# kev" for 172mRe. (talk) 22:57, 4 February 2017 (UTC)

"+X" or "+Y" probably mean that the excitation energy is not well-known, so the figure is probably a lower bound. "0(100)" probably means a notional "000 ± 100": since the "#" means it is extrapolated from trends somehow, I assume the source's extrapolation gives a silly figure of 0, but the uncertainty would make half of its possible values fall in a sensible range. I have no clue why that is listed for 168mRe, since it certainly does exist (the link is to the paper describing its first identification). Or is the idea that no data was available in the source? I'll need to go hunt through some nuclear the meantime, here is a level scheme for 168Re. Double sharp (talk) 04:19, 5 February 2017 (UTC)
And the plot thickens: 168mRe is not in NUBASE 2012 at all! Double sharp (talk) 04:53, 5 February 2017 (UTC)
According to Wikipedia, the Lutetium has some exciting (pun intended) meta states: 153m1,153m2,153m3Lu can be converted to 153Yb without β+ decay. I cannot find any sources to these daughters and if I understand correctly, the IT decay should only emit some high energy gamma rays and daughter should be same nucleus at lower state. I made some edits to Isotopes_of_lutetium table. (talk) 23:54, 5 February 2017 (UTC)

Supposed closure of Lead FAC[edit]

See here for the nomination and closure, and here for my response. A morning needlessly wasted. Sandbh (talk) 00:25, 28 February 2017 (UTC)

Check in and Kudos![edit]

15 years ago I started this project with the goal of turning the Periodic table by the quality blue. I see now that it is almost all green with a lot of blue. Great work and keep it up! I no longer have time to help, but continue to be proud of what everybody who has contributed to this project has done. --mav (reviews needed) 03:19, 1 March 2017 (UTC)

Thank you so much! We still have to turn all those greens blue, though! ^_^ Double sharp (talk) 03:50, 1 March 2017 (UTC)

If only I could read German…[edit]

103rd edition (2016) of Holleman-Wiberg's Inorganic Chemistry; only 2,622 pages. Sandbh (talk) 06:06, 3 March 2017‎ (UTC)

Disputed edits in discoveries[edit]

In recent weeks, User:Squee3 has changed discovery details of several elements' infoboxes (contributions)

Today I had to revert a Template:Infobox technium change, back to the year that was plainly in the article's source. Earlier, a back-and-forth at Template:Infobox arsenic (es saying 'As far as I can tell' as a source). In both examples the infobox deviates from the article body text & sources. Last month, in Timeline_of_chemical_element_discoveries only one source was added (again deviating from Tc article source).

Seeing that the edits are badly sourced if at all, I propose that all their edits in this area are reviewed, and that Squee3 be notified that no unsourced edits are disputed a priori and so must be discussed. -DePiep (talk) 09:07, 3 March 2017 (UTC)

I think Talk:Bromine#Discovery year may shed a little light on the problem. S/he seems to be working on some kind of project in which s/he can only write down one discovery year for each element. This of course necessitates some ingenious rationalisations of history, because in this case Löwig discovered Br in 1825, Balard discovered Br in 1826 and published his discovery the same year, and Löwig published his in 1827. Surely we cannot usually know for how long people were working on something (relevant recent xkcd), so in general the date of publication is surely more important, for that is when the result was made known to the scientific community as a whole. (Obviously, the ancient elements need to be handled differently; a deliberate use or knowledge of the pure element should suffice in this case.)
There are also some cases where in retrospect, it turns out that the team which claimed to have discovered a new element was honestly mistaken (e.g. nobelium), that two people independently found an element at the same time but one had much purer samples (e.g. lutetium), and that somebody had indeed found a new element, just not the one he thought he had found (e.g. Masataka Ogawa on rhenium). Clearly, there is no way you can get all of that across in a small infobox field.
So I would think that the simplification that is necessary in rationalising history this way actually does a great disservice to interested readers, even further than the necessary one that is needed in filling the infobox field. I may have to admit on the other hand that most readers are not that interested, and that they just want a date to put in a colourful "science project" that does not involve any actual science or understanding. Still, I would argue that the "History" section ought to give the detailed account, while we should base the dates in the infobox field on published sources. Double sharp (talk) 09:55, 3 March 2017 (UTC)

when TFAs can be rerun...[edit]

Since there is now a discussion about that, a great date for rerunning periodic table would be 6 March 2019 (150th anniversary of Mendeleyev presenting his first periodic table to the Russian Chemical Society). Double sharp (talk) 09:23, 6 March 2017 (UTC)

Sure. Should be Mach 6, 1869 by Western calender (which also puts the October Revolution in November), to make 150. Not O.S. -DePiep (talk) 15:44, 6 March 2017 (UTC)
On first publication.
So Glenn Seaborg mentioned that date, unqualified wrt OS/NS calendar. From Scerri, 2007 (chapter 4 (Mendeleev, The Crucial Discovery). pp 105– and footnotes).
On backside of an unrelated letter, Feb 17, 1869 Julian calendar (OS) (the Cheese factory letter), earliest sketch of the PT as published. In today's European Georgian calendar: Mar 1, 1869. Looks like the letter was written (dated) the same day M. received & used its backside.
On that same date (stated/unsourced by Scerri): manuscript of his full PT (Scerri fig 4.2), 'horizontal groups' i.e., transposed rows/cols compared to current day 'vertical groups'. (please always add descriptive word groups or periods when using this description; a table always has something horizontal, not is horizontal).
M.'s announcement (publication) of this PT: M. had made 200 prints in Russian, sent to chemicists in Europe. "N.A. Menshutkin communicated the initial discovery to the new Russian Chemical Society on March 6". Journal publication (Scerri fig 4.3): same month, "1869".
A question about this Scerri section (Scerri 20017, p 106 in my copy, more recent isbn-13 978-0-19-530573-9):
"This first published periodic table version of Mendeleev's (figure 4.3?) contains divisions into main and subgroups... first column ...". However, this seems to be about fig 4.4!, a 1871 publication. That 'first column' being ~group 1 (transposed, modern). All in all this 1871 paragraph looks out of place as Scerri is describing 1869 happenings. -DePiep (talk) 10:36, 15 March 2017 (UTC)

Standard atomic weight in Wikidata[edit]

Working on relative atomic mass and CIAAW. From there, over at Wikidata I am proposing to add two Properties for elements:

d:Wikidata:Property proposal/standard atomic weight
d:Wikidata:Property proposal/conventional atomic weight

-DePiep (talk) 20:27, 7 March 2017 (UTC)

Also, here at enwiki I have put content into standard atomic weight (copy/pasted from relative a.m. to start with). -DePiep (talk) 23:46, 9 March 2017 (UTC)

Wikidata timely thoughts[edit]

A general note from me. What is happening now in Wikidata is serious matter (however you and I dislike it). also '' and '' (projected) will pull data freely from Wikidata (legally, and no source need be mentioned). Wikidata will be the source google pages pull this data from (and not enwiki any more). The problem is (for me at least): how to work with Wikidata as I am used to work with enwiki? At least for the standard atomic weights now, I could pick up and work there. -DePiep (talk) 23:46, 9 March 2017 (UTC)

Biological role of nitrogen[edit]

Biological role of nitrogen is a redirect that currently points at a non-existent section of the main Nitrogen article. I've nominated this redirect for discussion at Wikipedia:Redirects for discussion/Log/2017 March 9#Biological role of nitrogen where your comments are invited. Thryduulf (talk) 00:48, 9 March 2017 (UTC)

Reclassifying the nonmetals[edit]

This is not a formal proposal from me. I'm only requesting views from you on the alternative classification scheme set out below. There is no rush as I have some more research to do. It's therefore a slow time project. As well, R8R is engaged on getting lead to FA status; and YBG is on a wikibreak until mid-April, so I'm happy to let this sit on the bench top for quite a while.


We currently colour code non-metals as polyatomic, diatomic, or noble gas.

Ever since we adopted these categories I've wondered (partly prompted by R8R) if that was the right decision, and if there was a better alternative.

This has been hard. Categorisation of nonmetals in the literature—aside from the halogens and the noble gases—is shabby. Metalloids complicate the situation. Some authors recognise such a category; others don't. The one that don't have to divvy them up between the metals or the nonmetals.

Authors usually throw up their hands and simply look at the leftover nonmetals—the ones other than the halogens and the noble gases—on a group-by-group basis. So you might have separate sections in a book on, say, hydrogen; carbon; nitrogen and phosphorus; and oxygen, sulfur and selenium. And some or all of the metalloids might get added to the applicable sections.

Otherwise there is the question of what to call these leftover nonmetals. The category names I've seen in the literature are "biogen", "CHONPS", "organogen" or "other". The first three of these categories tend to get tripped up by what to do with selenium. The last category—other nonmetals—is the "I-give-up-it's-too hard-I-need-to-get-published-so-I'll-treat-them-as-leftovers" category.

For our own classification scheme the "halogen" category is unavailable since we count astatine as a metalloid (as we should—it's either that or a post-transition metal).

Alternative scheme[edit]

In this context, an alternative scheme I've been considering has the following nonmetal categories:

Noble gas – He, Ne, Ar, Kr, Xe, Rn
Corrosive – O, F, Cl, Br, I
Intermediate – H, C, N, P, S, Se
Weak nonmetal (metalloid) – B, Si, Ge, As, Sb, Te, At

The corresponding legend looks like this:

Alkali metal Alkaline earth metal Lan­thanide Actinide Transition metal Post-​transition metal Weak nonmetal (metalloid) Intermediate nonmetal Corrosive nonmetal Noble gas

There is no change to the colours we use; the only change is to the three category names between post-transition metal and noble gas.

Corrosive nonmetal. These are are corrosive, and highly electronegative (> 2.6) and are, or their species are, capable of acting as relatively strong oxidising agents. Here are some examples from the literature as to other similarities between oxygen and fluorine, oxygen and the halogens, and oxygen and chlorine:

  • "Fluorine tends to bring out the highest valence of the element with which it combines. In this its shows a strong resemblance to oxygen. In combination with metals, oxygen appears to be the best for the highest valences, e.g. OsO4 and KMnO4, but fluorine appears best if the highest valence is relatively low, e.g. for CoF3, CuF3, AgF2, TbF4, and BrF5. With non-metals the difference between oxygen and fluorine is less apparent." (Phillips & Williams 1965, p. 446)
  • "…oxygen, like fluorine, forms strong covalent bonds, and there are a number of similarities between covalent oxides and fluorides." (Emeléus & Sharpe 1973, p. 318)
  • "Simple anionic chemistry is limited to oxygen and the halogens, although polyanions and polycations can be formed by many [nonmetals]." (Cox 2004, p. 145)
  • "Chlorination reactions have many similarities to oxidation reactions. They tend not to be limited to thermodynamic equilibrium and often go to complete chlorination. The reactions are often highly exothermic. Chlorine, like oxygen, forms flammable mixtures with organic compounds." (Kent 2010, p. 104)

Intermediate nonmetal. The more temperate nature of the intermediate nonmetals is relatively self-evident, situated as they are between the corrosive nonmetals and the weak nonmetals. This is so-called "other nonmetal" territory. There is a magic thread (my name for it) that binds the intermediate nonmetals, and it goes like this: H → C → P → N → S → Se:

  • Chemical similarities between H and C were discussed by Cronyn (2003) in the Journal of Chemical Education. They include proximity in ionization energies, electron affinities and electronegativity values; half-filled valence shells; and correlations between the chemistry of H–H and C–H bonds.
  • C and P represent an example of a less-well known diagonal relationship, especially in organic chemistry. Spectacular evidence of this relationship was provided in 1987 with the synthesis of a ferrocene-like molecule in which six of the C atoms were replaced by P atoms (Rayner-Canham 2011, p. 126). Further illustrating the theme is the extraordinary similarity between low coordinate P compounds and unsaturated C compounds, and related research into organophosphorus chemistry (Dillon, Mathey & Nixon 1998).
  • P and N are in the same group. Although N and its oxides are gases whereas P and its oxides are solids, the two elements "show many similarities in their compounds" (Malati 1999, p. 83). Despite these similarities "the chemistries of nitrogen and phosphorus are very different" (Wiberg 2001, p. 686). However P and N form an extensive series of phosphorus-nitrogen compounds having chain, ring and cage structures; and the P-N repeat unit in these structures bears a strong resemblance to the S-N repeat unit found in the wide range of sulfur-nitrogen compounds (Roy et al. 1994, p. 345) discussed next.
  • N and S have a less-well known diagonal relationship, manifested in like charge densities and electronegativities (the latter are identical if only the p electrons are counted; see Hinze and Jaffe 1962) especially when S is bonded to an electron-withdrawing group. They are able to form an extensive series of seemingly interchangeable sulfur nitrides, the most famous of which, polymeric sulfur nitride, is metallic, and a superconductor below 0.26 K. The aromatic nature of the S3N22+ ion, in particular, serves as an exemplar of the similarity of electronic energies between the two nonmetals (Rayner-Canham 2011, p. 126).
  • S and Se are in the same group: "As in the case of the halogens, the chemical similarities, at least for sulfur and selenium, are abundantly obvious" (Scerri 2007, p. 49).

Weak nonmetal (metalloid). Some authors count metalloids as nonmetals with weakly nonmetallic properties rather than having a discrete metalloid category, and that's the approach I've taken here. For a recent example see Cox (2004, pp. 26–27), who treats B, Si, Ge, As, Sb, Te and At as nonmetals, but notes that Si, Ge, As, Sb, Se and Te are sometimes called metalloids.

Precedents, approach, and features[edit]

In terms of precedents, the literature certainly refers to (a) O, F, Cl, Br, and I as having corrosive qualities; and (b) to the weakly nonmetallic chemistry of the metalloids. As far as I can see nobody has ever referred to an intermediate nonmetals category but then the literature is a terminological wasteland when it comes to a collective name for this part of the periodic table. Just about anything would be better than "other nonmetal".

I had to apply some violence and abstraction of detail to the alternative scheme in order to keep it simple. So there may arguably be some discontinuities and boundary overlaps. For example, counting iodine in the same league as O, F, Cl and Br may raise an eyebrow. Then again, iodine is corrosive, has a pretty decent electronegativity (2.66), and its periodate ion is a formidable oxidising agent (stronger than the perchlorate ion, for example); even the iodate ion is a stronger oxidant than elemental bromine. And pragmatically speaking it makes more sense to keep iodine with its lighter halogen congeners. As another example, nitrogen has a high electronegativity of 3.04 but all of its chemistry is essentially covalent, and the average oxidising power of nitrogen and its species, in aqueous solution, is less than that of both iodine and of sulfur.

On the question of discontinuities and boundary overlaps I turn to Jones (2010, pp. 170–171): "Though classification is an essential feature of all branches of science, there are always hard cases at the boundaries. The boundary of a class is rarely sharp…Scientists should not lose sleep over the hard cases. As long as a classification system is beneficial to economy of description, to structuring knowledge and to our understanding, and hard cases constitute a small minority, then keep it. If the system becomes less than useful, then scrap it and replace it with a system based on different shared characteristics." I feel that the similarities within each of the categories of weak nonmetal (metalloid), intermediate nonmetal, and corrosive nonmetal outweigh their differences, and blurry edges, sufficiently to establish them as discrete divisions.

The alternative scheme is better than what we have now because it's easier to get your head around, the new category names are more natural, and the new categories themselves fall into place quite naturally.

It maintains a good job of showing the progression in metallic to nonmetallic character as you go from left to right across the periodic table. And it facilitates a symmetry that is more rewarding than the contrast between metals and nonmetals, or between the alkali metals and the group 17 nonmetals, as shown in the following side-by-side match-up:

"Reactive metals"^
Groups 1–3, Ln, An
Corrosive nonmetals
O, F, Cl, Br, I
Transition metals (the "mundane" ones)
Most of 'em
Intermediate nonmetals
H, C, N, P, S, Se
Post-transition metals
Ga, Bi etc
Weak nonmetals (metalloids)
B, Si, Ge, As, Sb, Te, At
Noble metals^
Ru, Rh, Pd, Ag, Os, Ir, Pt, Au
Noble gases
He, Ne, Ar, Kr, Xe, Rn
^  I am not proposing that we have colour categories for reactive metals, and noble metals. The current colour categories for metals (alkali, alkaline earth, transition, lanthanide, actinide, post-transition) are fine.

There are references in the literature to this pattern. For example:

  • "Between Groups I and VII there are gradations from active metals (Col. I) to less active metals to moderately active nonmetals to volatile nonmetals (halogens Col. VII)." (Perlman 1970, p. 439)
  • "A period represents a stepwise change from elements strongly metallic to weakly metallic to weakly nonmetallic to strongly nonmetallic, and then, at the end, to an abrupt cessation of almost all chemical properties." (Booth & Bloom 1972, p. 426)
  • "By the end of 8th grade, students should know that…there are groups of elements that have similar properties, including highly reactive metals, less reactive metals, highly reactive nonmetals (such as chlorine, fluorine and oxygen) and some almost completely unreactive gases (such as helium and neon)." (AAAS 1994, p. 78)
  • Between the "virulent and violent" metals on the left of the periodic table, and the "calm and contented" metals to the right are the transition metals, which form "a transitional bridge between the two" extremes. (Atkins 2001, pp. 24–25)
  • "Describe how groups of elements can be classified…including…highly reactive nonmetals, less reactive nonmetals, and some almost completely nonreactive gases." (Padilla, Cyr & Miaoulis 2005, p. 27)

Overall, I find the combination of:

  • symmetry (as in the four complimentary metal-nonmetal categories) and asymmetry (many metals/few nonmetals);
  • the natural fit of the intermediate nonmetals between the weak nonmetals and the corrosive nonmetals;
  • the thread that links the nonmetals in this category (i.e. the intermediate nonmetals); and
  • the balanced 5-6-6-6 distribution of the nonmetals across their four categories

to be especially pleasing.

† The fact that there are 4 + 4 = 8 symmetry components is cool, too.

The wisdom of YBG[edit]

YBG suggested that any new categorisation scheme should be:

  • clear—"The criterion for division should be easily explained";
  • unambiguous—"It should be relatively obvious which category each element fits into"; and
  • meaningful—"The categories should have significance more than just dividing for the sake of dividing. There should be enough within-group similarity and enough between-group dissimilarity so that each group could be the subject of a separate encyclopaedia article."

Clear. In this case, the criteria for division go something like the following:

  • The noble gas category is self-evident (so to speak).
  • The corrosive nonmetals are, well, corrosive.
  • The weak metal (metalloid) category corresponds to the elements commonly recognised as metalloids. Astatine, which is included here, is irregularly recognised as a metalloid but we decided quite a while go to recognise it a metalloid. It has since been predicted to have the band structure of a full-blown metal.
  • The remaining nonmetals, which are neither corrosive nor weak (metalloid), are the intermediate ones.

This is so easy it almost explains itself.

Unambiguous. It is relatively obvious which category each element fits into.

Meaningful. As discussed earlier in this section, the resulting categories have enough internal similarity, and between category dissimilarity, to make them meaningful. We already have separate articles for noble gases and metalloids; I'd be inclined to have separate sections for intermediate nonmetals and corrosive nonmetals in the nonmetal article, as we do now for the polyatomic nonmetals and diatomic nonmetals.


  • AAAS (American Association for the Advancement of Science) 1994, Benchmarks for science literacy, Oxford University Press, New York
  • Atkins PA 2001, The periodic kingdom: A journey into the land of the chemical elements, Phoenix, London
  • Booth VH & Bloom ML 1972, Physical science: a study of matter and energy, Macmillan, New York
  • Cox PA 2004, Inorganic chemistry, 2nd ed., Bios Scientific Publishers, London
  • Cronyn MW 2003, "The proper place for hydrogen in the periodic table", Journal of Chemical Education, vol. 80, no. 8, pp. 947–950, doi:10.1021/ed080p947
  • Dillon KB, Mathey F & Nixon JF 1998, Phosphorus: The carbon copy: From organophosphorus to phospha-organic chemistry, John Wiley & Sons, Chichester
  • Emeléus HJ & Sharpe AG 1973, Modern aspects of inorganic chemistry, 4th ed., Routledge & Kegan Paul, London
  • Jones BW 2010, Pluto: Sentinel of the outer Solar System, Cambridge University Press, Cambridge
  • Kent JA 2007, Kent and Riegel's Handbook of industrial chemistry and biotechnology, 11th ed., vol. 1, Spring Science + Business Media, New York
  • Malati MA 1999, Experimental inorganic/physical chemistry: An investigative, integrated approach to practical project work, Woodhead Publishing, Oxford
  • Padilla MJ, Cyr M, Miaoulis I 2005, Science explorer (Indiana Grade 6), teachers's edition, Prentice Hall, Upper Saddle River, New Jersey
  • Perlman JS 1970, The atom and the universe, Wadsworth Publishing, Belmont, California
  • Phillips CSG & Williams RJP 1965, Inorganic chemistry, vol. 1, Principles and non-metals, Clarendon Press, Oxford
  • Rayner-Canham G 2011, "Isodiagonality", Foundations of Chemistry, vol. 13, pp. 121–129, doi:10.1007/s10698-011-9108-y
  • Roy AK, Burns GT, Grigora S & Lie GC 1994, "Poly(alkyl/aryloxothiazenes), [N=S(O)R]n : New direction in inorganic polymers", in P Wisian-Neilson, HR Alcock & KJ Wynne KJ (eds), Inorganic and organometallic polymers II: advanced materials and intermediates, American Chemical Society, Washington DC, pp. 344–357, doi:10.1021/bk-1994-0572.ch026
  • Scerri E 2007, The Periodic Table: Its story and its significance, Oxford University Press, Oxford
  • Wiberg N 2001, Inorganic chemistry, Academic Press, Berlin
  • Wisian-Neilson P, Alcock HR, Wynne KJ (eds) 1994, Inorganic and organometallic polymers II: advanced materials and intermediates, American Chemical Society, Washington DC


Some first responses. Added bullets to allow for subthreading. -DePiep (talk) 12:07, 12 March 2017 (UTC)

  • Interwiki RfC? Is it possible to enlarge support for our enwiki categorisation by making this an interwiki RfC? Our 2013 changes did not pick up well even in large wikis like de:, zh:, ru:, fr: (halfway only); ja: did. Note: checking the At category is enough, I can't blame any wiki for not using the poly-/di-atomic category names. -DePiep (talk) 12:07, 12 March 2017 (UTC)
    • And guess who made it stick for ja:! ^_^
    • BTW, there is some regional variation as well: in Japanese アルカリ土類金属 is a direct calque of "alkaline earth metal", but it only refers to Ca, Sr, Ba, and Ra (not Be or Mg). Double sharp (talk) 13:20, 12 March 2017 (UTC)
I'd like to contain the discussion to our own project, for now. Kudos to Double sharp san. Sandbh (talk) 23:42, 12 March 2017 (UTC)

Changing name from "metalloids" into "weak nonmetals"[edit]

  • Changing name from "metalloids" into "weak nonmetals". This is a surprise, and I'm not yet convinced. I didn't know there is a problem with 'metalloids'. First, the new name introduces a relative term in "weak": weak compared to others I must understand. And these others are: nonmetals (a de-classifying name by itself). So I must understand that there are elements that are stronger nonmetals, that are opposite of metals altogether. Then, seeing their position next to metals in the periodic table, if they are 'weak nonmetals', are they almost 'strong metals'? Or almost 'weak metals'? This is not playing, this is what words do. -DePiep (talk) 12:07, 12 March 2017 (UTC)
This is truly excellent and thoughtful feedback, thank you DiPiep.
To clarify, I am proposing to change the category name from "metalloid" into "weak nonmetal (metalloid)". I will discuss a problem with the term "metalloid" a little later in my response but for now it has its uses, which is partly why I am proposing to retain this word in the new category name. Yes, I agree, if you see the expression "weak nonmetal (metalloid)" it suggests there are stronger nonmetals, which is indeed the case.
If you see that they are positioned next to metals in the periodic table, and this prompts you to wonder if they are almost weak metals or almost strong metals then this is a good thing. The name piques curiosity, rather than bewilderment. I did not know what a metalloid was the first time I saw that word but I sure knew what metals and nonmetals were. The proposed category name is more informative than our other category names, which rely on some level of familiarity to be able to work out what they mean.
The scenario of wondering if a weak nonmetal would almost be a strong metal seems unlikely. Things in general get named for their dominant character rather than any subsidiary character, don't they? Some 'strong' transition metals are capable of forming oxyanions, which is nonmetal-like behaviour, but nobody calls these elements weak nonmetals. Crudely put, if an element was one-third nonmetallic (i.e. weakly non metallic) and two-thirds metallic (i.e. almost strongly metallic) it would be classified as a metal rather than a nonmetal. An analogy can be drawn with water that is weakly acidic (pH 6). This does not mean the water is almost strongly alkaline (pH 12). It means the water is almost weakly alkaline (pH 8).
I am sorry to say that one cannot work this out from the name "post-transition metal". If we had used something more descriptive like "poor metal" that would likely have answered the general reader's question. They would see that weak nonmetal was next to poor metal. But we decided to use "post transition metal", which is fine for the more technical reader. Alas, I feel that I should only work with the metal category names that we have. Sandbh (talk) 04:32, 20 March 2017 (UTC)
You raise very good points here, as always.
Yet I have a bit of an objection to this. Is it not the case that all of these descriptive terms need some knowledge of chemistry to understand? You can't understand "alkali metal" or "halogen" without knowing what an alkali or a salt is (in the latter case you also need to have a little knowledge of Ancient Greek, namely ἅλς "salt, sea"). You need some of this knowledge to get anywhere, and often you only get it retrospectively. If the name of the category does not suffice to tell us how Au or Te behaves, then maybe we should look more at the chemistry of their groups as a whole. Double sharp (talk) 04:16, 21 March 2017 (UTC)
I may have misunderstood the basis for your concern but will plough on anyway.
Largely I would say yes you do need some knowledge of chemistry to know what these terms mean. Alkali metals and alkaline earth metals may be somewhat of an exception. I can recall a high school science demonstration of calcium skipping about on the surface of water so that may have given me some idea of the nature of an alkali or alkaline earth metal. I can't remember when I learnt what "alkaline" meant but it may have been when I was introduced to the idea of acids and bases and phenolphthalein—and that was only in general science. And I'm pretty sure I knew by them what quicklime was and its causticity, although I can't remember if I would have recognised this as being due to its alkaline nature. I can't remember if I saw what happened when a few drops of iodine were added to powdered aluminium. Nor can I remember when I was taught about the contrast between sodium and chlorine. And I don't know if high school chemistry these days includes real experiments designed to show the properties of such reactive metals and nonmetals.
There are some category names that I would recognise without needing a background in chemistry. Base metals, precious metals or coinage metals. Maybe even refractory metals. Ferromagnetic metals. Gaseous nonmetals. Crystallogens perhaps.
It's good thing to have simpler descriptive category names, is it not? Sandbh (talk) 10:12, 21 March 2017 (UTC)
It is a good thing, but there are always some prerequisites. And we shouldn't forget that some of the names are misleading, even though they are intuitively obvious. Pb may be a crystallogen, but it does not even have a diamond cubic allotrope, and while Sn does have one it's not really that attractive to look at. Some of these have artificial breaks: Re is a precious metal (of sorts) but Tc is not, because no one wants to have radioactive things (never mind that Tc has a long half-life and decays to clean stability, eliminating most of the problems with things like Pu and Cm which have comparable half-lives).
Well, yes, there are always some prerequisites; equally, category names are labels of convenience not labels of absolute truth, are they not?
There are several other category names that fall short under pressure. The alkaline earth metals category is a pretty good label for most of the group II metals but wobbles a bit in accommodating beryllium and its amphoteric oxide, and its capacity to form beryllates. (What is an amphoteric element doing hiding among a bunch of supposedly alkaline metals? Hoping no one will notice? ^_^). Pnictogens is an IUAPC approved name for group 15 but has zero intrinsic relevance to anything after nitrogen. Much of the chemistry of the early actinides is quite different from that of actinium. The crystallogen example is interesting. C, Si and Ge form diamond cubic allotropes at room conditions; tin only does so at low temperatures and the result is an amorphous grey power unless special care is taken to prepare it in its crystalline form (which looks metallic). A diamond cubic form of lead has not so far been prepared but there has been speculation that this might be possible under the right conditions.
Labels of convenience, as such, don't necessarily care or need to worry about artificial breaks. Sandbh (talk) 04:47, 22 March 2017 (UTC)
And it seems to me that "polyatomic" and "diatomic" have fewer prerequisites, as they do not require us to get an idea about what could be intermediate about nonmetals (which means learning the chemistries of many of these elements, which are fantastically complicated, so we don't tend to go past the vital C and H in high school): they just require us to learn a few simple terms. Later we can marvel about how these two are correlated with other important properties as well. Double sharp (talk) 14:37, 21 March 2017 (UTC)
Maybe "polyatomic" and "diatomic" require as much pre-knowledge as intermediate nonmetals i.e. you need to at least know what polyatomic means, and for intermediate you need to know they are neither as extreme as corrosive nonmetals (like Cl) nor as weak as weak nonmetals (like Si). If you happen to have a wikipedia table in front of you with the proposed new colour categories you will at least be able to appreciate straight away the relevance of the intermediate label (just like the transition metals are transitional). I don't think there is much between the two schemes in this sense, nor is there is much need to, at first, go past C and H in either of them. Sandbh (talk) 04:47, 22 March 2017 (UTC)
re you do need some knowledge of chemistry to know what these terms mean. Alkali metals .... Is true, but is not the primary issue in categorising. Category name and meaning is secondary, and that's what links are for. Primary is: group together what belongs together, and do not include stuff that does not include there. Just give the group a unique color and name, that's enough. The name can be as fancy or invented, no intrinsic meaning required, no prior knowledge for the reader required. Exactly what the grouping criteria is, should be described in the namelink or in an overview article. OTOH, a name should not be misleading by ignoring existing prior knowledge (is why we cannot reuse and redefine 'rare earth metals' for lanthanides). In this case, I think that 'metalliods' (and 'de:halbmetalle') are fine, unloaded words. Once we would use '.. nonmetal' somehow, it can not be made to mean 'but not really a nonmetal' any more. -DePiep (talk) 08:57, 22 March 2017 (UTC)
Hmm, I don't like the term 'halbmetalle' or its English relation 'semi-metal' given confusion with the physics-based terms half-metal, and 'semimetal'. In the latter case only carbon, arsenic, antimony and bismuth are semi-metals yet few would regard carbon and bismuth as metalloids.
I no longer like the term 'metalloid' (but I acknowledge it is still used in the literature). It means 'resembling a metal' but since metalloids generally behave at least chemically like nonmetals, the term 'metalloid' tells less than half the story. Even physically, metalloids mostly act like semiconductors and are brittle, which are properties associated with nonmetals, so although they have the appearance of faux-metals, the 'metalloid' nomenclature is rather misleading.
I prefer the term 'weak nonmetal (metalloid)' as it captures the best of both worlds, without misleading anyone, and it does not involve a change of meaning. Sandbh (talk) 12:28, 22 March 2017 (UTC)
I used 'halbmetalle' only as an illustration of a not-misguiding word, as is 'metalloid' (or so I thought). If there is no descriptive available spot on, and if 'halfmetal' or 'metalloid' etc. are mainly wrong (I doubt), I'd prefer a new, unloaded wording. Definitely not 'semi-conductor' for that is only one property. I don't think the combination wording 'weak nonmetal (metalloid)' will stick; others will pick one ~randomly, so we'll be introducing a future confusion, also since we (enwiki) cannot control their definitions when used off-site. "Category Q", "MNM", "HM"? (kept as unspecified lettercodes). -DePiep (talk) 16:59, 22 March 2017 (UTC)
That's all good. I intentionally suggested the term "weak nonmetal (metalloid)" to avoid the controversy of coming up with something more novel. What is likely to happen is what happens now. Some people will like our categories; some won't. Some will keep calling boron a nonmetal; others will call it a metalloid, or a semimetal, or a halbmetalle. Others will ignore our polyatomic and diatomic categories and use e.g. halogens (counting At as a nonmetal), and refer to the rest of nonmetals as "the rest of the nonmetals" or "the other nonmetals" or the "less reactive nonmetals" or "less electronegative nonmetals" etc. instead. And that is OK: "weak nonmetal (metalloid)" is unlikely to get anyone's nose out of joint, as it accommodates nearly all preferences.
On using an unloaded category name I strongly fear that the result will be too much of neologism. "Chemical phenomena are highly complex" and hard enough to "treat rigorously from a fundamental theory" (Scerri 1996, p. 171) as is, let alone trying to derive unloaded wording from the existing literature. Since the language of qualitative and descriptive chemistry is reasonably replete with the adjectives "weak", "strong" and e.g. "low" and "high" I figure that "weak nonmetal" as a corollary to "corrosive nonmetal" should be nothing to write home about. (I thought it was bold enough already to claim and implement polyatomic and diatomic as discrete nonmetal categories yet there were no screams. The new scheme is not half as scary; the contours of each category are already present in the literature). [Ref: Scerri ER 1996, "Stephen Brush, the Periodic Table and the nature of chemistry," Die Sprache der Chemie, P Jannich, N Psarros (eds), Könighausen & Neumann, Würzburg, pp. 169–176 (171)] Sandbh (talk) 03:44, 23 March 2017 (UTC)
I think the reason why there were no screams for "polyatomic" and "diatomic" is that those are pretty standard terms. Everyone knows what a nonmetal is; everyone knows what a diatomic molecule is; so no one would bat an eyelid if you said "hydrogen is a diatomic nonmetal". It is a true statement using two standard terms: "diatomic" and "nonmetal". (It is also nice that it extends naturally to the noble gases as "monatomic nonmetals", but that's not the main reason.)
The trouble with "weak nonmetals", "intermediate nonmetals", and "corrosive nonmetals", on the other hand, is that while the contours are certainly there, the terms are not. For example, what exactly does "weak" mean when it modifies a nonmetal? As a counterweight to corrosiveness, it sounds like the noble gases belong there too. And this problem comes from the fact that "weak" is not a standard chemical term for elements (it of course is one for acids and bases, but that definition doesn't carry over). You would have to define what it means here newly, and that is more like original research than "polyatomic" and "diatomic" as standard modifiers.
(I'd also argue that while the contours are there, the exact boundaries are continuous, so it's not clear what should be in what category. We currently have an amicable disagreement on astatine, for example. But this is not the main point that I am making here.) Double sharp (talk) 15:28, 24 March 2017 (UTC)
I thought the screams might have arisen for any of the following reasons.
  1. Sure, H is referred to as a diatomic nonmetal but there no concept in the literature of diatomic nonmetals having other shared properties. There are a few references to some nonmetals being "polyatomic" but there is no concept of other shared properties among them. Now, I thought we had good reasons for getting rid of other nonmetals. And diatomic nonmetal was certainly a recognised term, and "polyatomic" by itself was reasonably objective, and found in the literature, and the distinction between diatomic and polyatomic nonmetals happened to bring out a reasonable difference in properties.
  2. The categorisation of S as a polyatomic nonmetal, while structurally correct, always worried me from the point of view of attempting to bring out its commonalities with C, P and Se. There is a very large and interesting note about this in the nonmetal article. Truly, sulfur is the Pluto of the polyatomic nonmetals.
  3. The existence of ozone, a polyatomic nonmetal, bugged me. Sure oxygen is a diatomic nonmetal but you would hardly call it a borderline diatomic nonmetal, so why does ozone exist? I'd expect this sort of thing for an iodine-like nonmetal, not oxygen.
  4. The division was based on a structural distinction, unlike any of the other categories. Chemists don't think of the elements in these terms, not immediately, anyway.
  5. I worried a little about calling iodine a diatomic nonmetal. It is as a gas, but in its standard condensed state it has some polyatomic character. There is evidence for significant intermolecular coupling between the individual iodine molecules (each iodine atom forming weak bond with its two next nearest neighbours, as well as a stronger bond with its molecular partner) implying a bulk coordination number of 1+2, rather than 1. There is a note along these lines in the metalloid article. The funkiness of the crystalline structure of iodine is further borne out by its similarity to that of gallium.
  6. Only S is a polyatomic nonmetal in the sense that one can point to an S8 molecule. There are no discrete molecule-equivalents for C, P and Se in their most thermodynamically stable forms.
It feels odd to critique a proposal that I put a lot of effort into at the time. However I thought it was the best of the alternatives then available to us.
I agree "diatomic nonmetal" is a plausibly standard term. As noted, that's about all one could say about it i.e. that some of the chemical elements are diatomic nonmetals. End of conversation. No further discussion in the literature, direct or indirect on their other shared properties.
Except for the noble gases, the literature on classifying the nonmetals is quite challenging, as is the case for metals in, or in the vicinity of, the p block.
This doesn't mean we have to come up with newly defined terms for categorising the nonmetals. The literature has already done our work for us. Unlike the polyatomic and diatomic nonmetal scheme, which took nine unsuccessful proposals to surface and is not readily evident in the literature, this one almost fell into place by itself once I thought about the contrast between the so-called "other nonmetals" and their neighbours to either side. (I have to say it was a mental challenge to have to once again contemplate the construct that is the other nonmetals).
We can look at how the nonmetals of concern are described in the literature and seek to apply the same, or equivalent, terminology.
Before I continue:
  • The terms "metal" and "nonmetal" are composites. There is no single measure of metallicity or non-metallicity. Given this, when "weak" modifies a nonmetal or a metal, it means that while the nonmetal or metal involved suffers from a deficiency of the attributes that we associate with being a nonmetal or metal, it still flops over the line sufficiently to make the grade as a nonmetal or a metal. In the literature, the terms "strong" and "weak" are relatively commonly encountered when describing metals. This is much less the case with the nonmetals.
  • I don't believe I've advanced or suggested the position that "weak" is a counterweight to "corrosive", so I see no issue here.
  • I'm puzzled by your observation that it's not clear what should be in what category. The corrosive nonmetals are clear. The weak nonmetals (metalloids) are clear presuming, as we have done, that it is acceptable to include here only the elements most commonly classified as metalloids, plus astatine as per a prior decision of this project. The intermediate nonmetals are what remain. These allocations are consistent with the way the elements involved are described in the literature.
  • I agree there is some overlap between the categories. For example, the line between carbon and, say, boron and silicon; and a little bit between iodine and, say, selenium or tellurium; or between selenium and, say, arsenic and antimony, or tellurium. However, I observed the same kinds of overlaps between the polyatomic and diatomic nonmetal categories. Indeed, our nonmetal article says, "The distinction between the three categories of nonmetals, in terms of receding metallicity is not absolute. Boundary overlaps occur as outlying elements in each category show (or begin to show) less-distinct, hybrid-like or atypical properties." More generally, we have a section in our periodic table article addressing instances of this kind occurring in the rest of the periodic table. In our categorisation decisions it is perhaps preferable to minimise such overlaps but, then again, I'm not so sure that to do so would accurately reflect periodic table reality, which tends to be more messy. So, on the question of overlaps, I cheerily say, yes, that's an interesting feature of the periodic table: there are always hard cases at the boundaries.
Getting back to looking at how the nonmetals of concern are described in the literature and seeking to apply the same, or equivalent, terminology:
  • We know that the literature talks about strong metals, weak metals, weak nonmetals and strong nonmetals. Sometimes the language used is more specific and it gets couched in terms of e.g. "less reactive nonmetal", "moderately active nonmetal", or "more strongly electronegative nonmetal".
  • We know that oxygen and the halogens regularly attract superlatives in terms of e.g. their electronegativity, ionisation energies, and oxidising power. Indeed, the word "oxidation" arose from oxygen's formidable combining power. We know they are all described as being corrosive.
  • We know that metalloids generally behave chemically as nonmetals; that they show less tendency to anionic behaviour than other nonmetals; that they always give compounds less acidic in character than the corresponding compounds of (regular) nonmetals; that they have the lowest electronegativities and ionisation energies of the nonmetallic elements; and that they are wimpy enough—nonmetallically speaking—that they can be cajoled into forming alloys and organometallic compounds.
  • We know from the literature that the remaining nonmetals are generally not described in terms of these extremes and that they are most commonly termed as "other nonmetals". We know that nobody really likes this term, and that the meaning of "other" includes, "existing besides, or distinct from, that already mentioned or implied".
  • Given the moderate nonmetallic character of these other nonmetals, and since they are found on the periodic table besides the weakly nonmetallic metalloids and the extreme or corrosive nonmetals, calling them "intermediate nonmetals" seems like the most literature-consistent and descriptively detached label available to us, in light of the meaning of "intermediate". Here is that meaning: "coming or occurring between two things, places, etc.; holding the middle place or degree between two extremes; interposed, intervening…in spatial position: situated in the middle place, or between two things or places".
In summary, the proposed division is more natural than that of the current scheme, in terms of the way the nonmetals involved, and their properties, are described and discussed in the literature (anomalies and all). Sandbh (talk) 01:35, 27 March 2017 (UTC)

"metalloids" as "nonmetals"[edit]

  • Another issue with the proposed name is, that it claims the metalloids as 'nonmetals'. Sure there will be some arguments for this in literature, but to me it is a surprise major change. Our 192 sources on metalloids did not float what must be such an obvious characteristic. And, seeing the property comparing overviews on multiple characteristics, also does not push them into nonmetals (in some properties they are plain metals even). So naming them 'nonmetal' is a change of concept difficult to get. In other languages they are still called de:halbmetalle. Some naming history is nicely described in metalloid (the lede for starters). All in all, I gather that it is bad practice to use a single characteristic description for the overall category name. -DePiep (talk) 12:07, 12 March 2017 (UTC)
This is why I am proposing to change the name to "weak nonmetal (metalloid)". Our lists of metalloids article was only a list. It's purpose was to show which elements were classified as metalloids in the literature. This explains why it does not say anything about metalloid chemistry.
At this point I feel it would indeed be helpful to summarise more of the state of the literature.
  • There is consensus that some elements, in the vicinity of a line running roughly through boron to astatine, can be tricky to classify as either metals or nonmetals.
  • Some authors classify these elements as metalloids. Others make a call and classify each of these elements, on a case-by-case basis, as either a nonmetal or a metal. In this scenario, there is consensus that boron, silicon and tellurium are nonmetals. Germanium is sometimes classified as a metal and sometimes as a nonmetal; the same goes for antimony and polonium. In light of astatine's status as a halogen, its unimportance, and seeming uncertainty as to its properties, the lowest common denominator consensus is to assume that it's a nonmetal.
  • Each of these classification decisions are open to challenge. This rarely occurs, as long as the rest of what the author writes is not completely inconsistent with their original classification decision (and presuming this decision was not blatantly erroneous on the first place).
  • There is consensus in the literature that metalloids have properties that are intermediate between those of metals and nonmetals or properties or a mix of metallic and nonmetallic properties. There is consensus in the literature that metalloids look like metals and are semiconductors. (The fact that arsenic and antimony are not semiconductors in their most stable forms is a common oversight).
  • Much of the literature stops at this point. Any further discussion of elements that are considered to be metalloids generally occurs only in the parts of texts giving the general descriptive chemistry of the elements in each p block group.
  • There is consensus in the more considered literature that metalloids or nonmetals in the vicinity of the dividing line between metals and nonmetals generally behave chemically like weak nonmetals (and that metals in the same vicinity generally behave chemically as weak metals e.g. Tl, Pb, Bi).
In summary, the proposed term "weak nonmetal (metalloid)" captures the complexity of the situation with a high degree of consensus—much higher than we have now with "metalloid". By itself the term "metalloid" is misleading, since it means 'resembling a metal', which is less than half the story. But the term weak nonmetal (metalloid) expresses the nature of these elements much more comprehensively, and accurately.
As to our own content, the metalloid article, the properties of metals, metalloids and nonmetals article, and the origin and use of the term metalloid article each note that the overall chemical behaviour of metalloids is nonmetallic. Yes, some of their properties are halfway between metals and nonmetals, but none of these properties are beyond the scope of what you could reasonably expect to find in nonmetals on the borderline between metals and nonmetals. And most of the metallic properties of metalloids are physical properties, and physical properties are trumped by overall chemical behaviour. Sandbh (talk) 04:32, 20 March 2017 (UTC)
I am not sure that they should be trumped so completely though, especially in such a borderline region when no one's arguing over which elements are in the group. If you look at any p-block group going down, you will start with mostly nonmetallic character (notwithstanding that boron is an "honorary metal atom" and vice versa), and you will gain more and more further down. If we look at group 16, for instance, O and S are insulators, Se and Te are semiconductors, and Po is a metal; and while chemistry lags behind a little, we do see some emerge, as Se is not attacked by dilute HCl, while Te dissolves slowly and Po dissolves enthusiastically to give pink PoII and self-oxidises to yellow PoIV. And tellurium even looks like a metal: there's a reason why Müller, who discovered Te, called it metallum problematicum.
I don't think it's a far stretch to say that the metalloids look like metals but act like nonmetals, although of course we have to take this with a pinch of salt (I suppose that needs to be Na2Te for this post), but in contrast to humans where beauty is no guarantee of moral character, the shiny lustre does say something about an element's physical properties, which we shouldn't neglect. There's even fuzziness on the other side; Ge dissolves slowly in hot concentrated H2SO4 and HNO3, proving that some of the metallic character that we had expected is there, if muted by the d-block contraction. But even Sn is amphoteric, as is shown by the following reaction:
Sn + 2 KOH + 4 H2O → K2[Sn(OH)6] + 2H2
It also has a clearly nonmetallic allotrope to boot, while Si and Ge transform to metallic tetragonal β allotropes like Sn if you just subject them to pressure along the c-axis (~200 kbar for Si, ~120 kbar for Ge). Even the SnX4 are tetrahedral volatile solids and liquids and the obvious comparison is with the tetrahalides of Si and Ge. So tin appears to be fair game for being a "weak nonmetal" as well in this sense, which is funny because if you go one step to the right(!) you obtain a more clearly ionic compound in SbF3 and even a clearly intermediate SbI3! Furthermore, looking at group 15, As, Sb, and Bi have very similar electronegativities, so that cannot be the only guideline for grouping them as metals, nonmetals, or metalloids; the difference is structures is more reflective of how the octahedral gaps in the hcp iodine lattice are filled by each of these elements, and only Bi is large enough to fill them symmetrically.
Of course, we can wave away the question of why such elements are separated by appealing to the literature, but this is science: if the literature says something, we should critically examine it to figure out why something is being said.
And even with clear if intermediate nonmetals like phosphorus, even the halides are getting close to the ionic-covalent divide: PCl5 is ionic [PCl4]+[PCl6] in the crystalline phase and covalent in the gas phase, and PBr5 (which is [PBr4]+[Br]) outright decomposes in the gas phase. Indeed, phosphorus is so common an element that I remember being told this one in high school. ^_^ The solvent itself also influences the choice between ionicity and covalence, and even tweaking the substituents can lead to massive changes: PhPCl4 is molecular while MePCl4 is ionic. There is metallic character even here, and focusing on the nonmetallicity fails to show it clearly.
But the main point is that "metalloids" is way more common in the literature than "weak nonmetals". The "metal" tells you that at face value, that is what they look like: the "-oid" suggests that appearances can be deceiving. What is there more to ask for a term? Double sharp (talk) 14:59, 22 March 2017 (UTC)
After a long stint at responding to Double sharp, Sandbh reaches for his tequila and lime, when he suddenly sees the label on the salt shaker...
Right then, I'll give this a go.
I recall reading in an old chemistry reference that metals and nonmetals used to be distinguished according to their physical properties but that "these days" (back in the 1800s, I believe) chemical properties are more important in making the distinction. So when I said chemical properties "trump" physical properties I may have gotten a bit carried away. It's probably better to say that chemical properties are more important, and that this does not mean physical properties have no relevance.
Metallic character is present even in the extreme nonmetals i.e. the halogens, never mind the intermediate nonmetals, in the right (extraordinary) circumstances. Same goes for nonmetallic character in the alkali metals. Yes, focussing on the nonmetallic character, of e.g. the intermediate nonmetals, will fail to show their metallic character. Same goes for the metals generally and their nonmetallic character, but these are matters of detail added to first order generalisations and category assignments, as you have noted.
My recollection is that cationic behaviour or the acid/base nature of oxides are important considerations in distinguishing between metals and nonmetals. Tin forms a cation in aqueous solution ergo it is near universally regarded as a metal, albeit a weak one. If the case was any stronger for calling tin a weak nonmetal, I'd expect to see it be predominately called as such in the literature yet no one does.
Germanium is curious. Mendeleev predicted it would be a metal but I don't know what he meant. Was he referring to its appearance or its chemistry? Why would he predict it would have a metallic chemistry if he was presumably familiar with the weakly metallic chemistry of tin? As it turns out, when Winkler announced the discovery of germanium in 1886 he called it a new non-metallic element, on the basis of its chemistry!
None of the other metalloids or nonmetals are capable of forming a simple cation in aqueous solution, as far as I know. Yes, astatine can, but then I reckon we should classify it as a post-transition metal. Neither do any of them, as far as I can discern, form oxides that are predominately basic i.e. metallic.
The term metalloid is certainly more common in the literature than weak nonmetal. That is why I am proposing to keep it. I suppose that the "metal" prefix tells you at face value what they look like (but it doesn't tell you that they mostly don't conduct electricity like metals). As to the suffix "-oid" this means "resemble" or "like" doesn't it? When I read "metalloid" I think of something resembling a metal. So I'd expect it to be almost but not quite a metal. What I get instead is something that is brittle, with no structural applications, and that has the general chemical behaviour of a nonmetal. Ripped off! Sold a pup! ^_^ The term "metalloid" tells less than half the story.
We know that IUPAC tried unsuccessfully to get rid of the term metalloid in 1959, and again in 1971, and to replace it with the term "semimetal". These were poor efforts given the physicists had already appropriated the term, "semimetal". The Encyclopedia of Chemistry (Hampel & Hawley 1973, p. 727) had another go: "The term "metalloid" is becoming obsolete in that the term nonmetal is more precise. The term semiconductor now is also applied to such nonmetal elements as silicon, germanium and selenium...". The Condensed Chemical Dictionary, which has been around since 1919 and is currently in its 15th (2016) edition, is closer to the mark. In the entry for "nonmetal" (p. 988) they count metalloids as nonmetals that more nearly resemble metals than then rest of the nonmetals, but say of the term "metalloid" that "it is no longer used by chemists", which is not quite right. (I'm not sure what is happening here since their nonmetal entry has read this way for at least their last six editions, since 1977, under different sets of editors).
Whatever the goings on with the term metalloid, it still has its place in the literature hence I'm proposing to keep it but to use the much more representative term "weak nonmetal (metalloid)".
PS: I expect I'd politely decline having Na2Te metalloid salt with my tequila and lime, given the toxicology of Te. Sandbh (talk) 02:41, 24 March 2017 (UTC)
Well, the suffix of the term "metalloid" is originally -ειδής in the Ancient Greek that it comes from, which comes from εἶδος "form, likeness". So when I see "metalloid", I expect something that has the form and likeness of a metal, but is not actually one; for if it were one in all of its behaviour, then why not just call it a metal? So to me it does not really feel like the term rips us off; rather, it gives us a warning that while these elements might look like metals, they don't act like metals.
I wouldn't call PCl5 particularly extraordinary. I mean, it is in the sense that it is right next to the ionic–covalent divide, but it is not in the sense that it really is quite a common compound. If not, we should have to call water extraordinary, as you do not see its characteristic properties in H2S, Li2O, or F2O. Looking at things that way seems to be along the lines of Novalis talking about making the familiar strange and the strange familiar, because it is only on account on the singular properties of the elements we use the most that they can support something as complicated as life. So although they do not really show the most common behaviour among the elements, they show the most common behaviour if you factor in how often we see certain compounds.
What this means, of course, is that there are certain elements for which the first-order rationalisation of their properties is so far off the mark that it is essentially worthless. I reckon that this happens to a great extent with the first-row anomaly, which is why H, B, C, N, and O are so different from the rest of their groups, and even Li, Be, and F, while visibly allied, have some qualitative differences that can't be overlooked. If we tried to predict the properties of H2O in a first-order sort of way from those of H2S, H2Se, and H2Te, I suspect that our first prediction would be a rather foul-smelling gas (though it would presumably be relatively tolerable compared to its congeners) that condensed at around −100 °C. Clearly, the first-order prediction overlooks so many things that it's not very useful, and to make valid predictions for a case like this you need to know the answer already by knowing the singular chemistries of H and O.
Now, the trouble I find with the idea that forming a cation in aqueous solution is that important as a factor (instead of being merely suggestive) is that many of the transition metals do not form such a cation in their most common oxidation state. Indeed, when one considers how anion formation tends to stabilise high oxidation states, one sees many similarities between the middle of the transition metal group and the nonmetals that they are placed with in the short form of the periodic table. Consider silicon, the most electropositive of the metalloids (EN 1.90). This value is equalled by Cu, Tc, and Re! Consider hydrogen at EN 2.20: this is equalled by Ru, Pd, Ir, and Os (with Mo not far behind at 2.16), and even surpassed by Rh and Pt (2.28), W (2.36), and the champion Au (2.54) which almost equals carbon (2.55), and shows a suggestive proclivity towards catenation and even forms a simple anion in Rb+Au.
If we consider tungsten, for example, its most characteristic oxidation state is +6, a very high one characteristic of clear nonmetals. And we get molecular polytungstate anions with oxygen in this oxidation state, and the most well-characterised halides are the hexahalides, which are practically covalent compounds (WF6 is famously a gas, and the other two are pretty volatile). And the only electrode potential listed by NIST for a W cation (W3+/W0) has a footnote to (a) saying "This half-reaction contains at least one doubtful chemical species"; since no one doubts the existence of W metal, it must be the putative W3+ that is being referred to. Finally, what of the tungsten oxides? They are mildly acidic, dissolving in aqueous alkali to form tungstates!
Is any of this what we would have expected for the chemistry of a metal? Absolutely not! The arguments for calling W a metal are mostly physical; it is hard, strong, dense, and a good conductor of electricity. Except that even then we can't say that for sure, because W is also brittle! If we are agreed that tungsten should count as a metal, then I would imagine that these criteria need serious revision or supplementation. Between the elements that define what the public considers to be standard metals – the transition metals – and the criteria that most people think of as chemically defining metals – there is a great conflict.
Admittedly there is not much family resemblance here
Note that I do not believe that this detracts from the power and effectiveness of the periodic table. In all of its supreme principles and rules, there are always exceptions, and the rules and the exceptions strengthen each other in an unceasing spiral until they unite and become indistinguishable from each other. A good example of this is how Greenwood and Earnshaw, on p. 27, include "anomalies" and "anomalous" properties in their list of "periodic trends which occur in the chemical properties of the elements", because it is impossible to get a good understanding of period 2, the lanthanides, and the heavy p-block elements without taking these into account so consistently that they harden into a new set of rules! These necessary insights for reading the periodic table are a concealed, self-destructive assault on the principles that it is itself founded on, and it paradoxically gives it the power to cast its net wide and apply the periodic law to every element we know of, from the beautiful simplicity and singularity of hydrogen to the scintillating complexity and radioactivity of oganesson. Its imposing castle-like structure, radiating the confidence of a mathematician, masks this self-destructiveness, and yet the two are simply two sides of the same coin. I realise that this makes Mendeleev sound like Mozart (and I got this idea from reading Charles Rosen's The Classical Style, which describes Mozart's art as subversive and uncompromising in similar terms to what I have done for Mendeleev's, but of course in an even better way), but the more I think about this comparison the more valid it seems to me.
As for germanium and tin, I would imagine that if Mendeleev saw no reason to deny Sn metallic status on account of its amphoteric aqueous chemistry, then he probably saw no reason to deny it to what he predicted for Ge as well. It should also be remarked that Si was originally thought of as a metal, because its oxide was analogous to aluminium and beryllium oxide, so much so that Sir Humphry Davy originally named it silicium; this could also have been a major factor in considering germanium to be metallic. In any case, Si, Ge, Sn, and Pb all have amphoteric oxides: yes, even the clearly metallic (if soft and low-melting) Pb. But I am not surprised that the proposed criteria runs into this problem in the p-block, if it is already straining in the d-block with tungsten. I have no doubt that they came from a great deal of thought, and might summarise how people mentally think for some portions of the table, but I believe that we need to press on and find something even better.
I am looking forward to your reply after you enjoy your tequila and lime: after all, I did plan this reply for over an hour, and then proceed to edit and expand it another nine times. ^_^ I must say that this conversation has been wonderful so far at making us both think and reexamine the amorphous blob of principles and connections between them that constitute our understanding of chemistry. Double sharp (talk) 04:16, 24 March 2017 (UTC)
Broadly, we seem to agree about the suitability of the term "metalloid". Regardless of our differences in interpreting the term, a metalloid is not a metal. I presume a metalloid can therefore be regarded as a "non"-metal—a distinctive kind of such, but a nonmetal nevertheless.
On the "cation in aqueous solution" consideration, my impression of the literature is that this only tends to become important when attempting to sort out what happens when the metals meet the non-metals in the p block.
I remember reading in C&W that some of the transition metals had no simple cationic chemistry but the fact that tungsten has no basic oxides was something that hadn't occurred to me. It happens that tungsten is listed as forming a "green-yellow" trivalent monoatomic cation in The Aqueous Chemistry of the Elements (Schweitzer & Pesterfield 2010, p. 306) but the same authors show germanium as forming a colourless monatomic divalent cation (p. 190)! No supporting citations are given apart from a general reference list at the back of the book. Now, in the germanium section of our metalloid article there is a note on whether Ge can form such a cation and I stand by my assessment that the evidence is unclear. I haven't done the same degree of research on tungsten but I suspect the same may apply to the notion of a tungsten cation. Having said all that I don't think the fact that a few d block metals may or may not be able to form cations matters. Everyone acknowledges they are metals, wrinkles and all.
(Looking back at your comments it seems to me that your reference to the "cation in aqueous solution" criterion as being "suggestive" was prescient, in the context of what I've written above).
So, no, I think there is no need to come up with a more rigorous way of distinguishing between metals and nonmetals. For our purposes, the literature has already done the work for us in distinguishing between metals and nonmetals (including metalloids).
Continuing in this theme, I'm not so sure that I would describe the irregularities and anomalies in the periodic table as a concealed self-destructive assault on the principles that the PT was founded on, since these were loose in the first place. I like this quote attributed by Scerri (1996, p. 174)^ to Peter Nelson, "who often writes on matters of conceptual chemistry": "The application of periodic law is not simply a case of making logical deductions from basic principles as in the case of thermodynamics. It involves rather a fairly thorough knowledge of how the principles work out in practice - of exceptions, trends and patterns [e.g. the first row anomaly, anionic gold, rogue transition metals etc] - so that a particular deduction can be appraised and if necessary adjusted. In the limit the process becomes virtually intuitive". I think however, that you arrive at essentially the same conclusion!
^ details as previously noted
Not that it matters much but your thoughtful speculation as to why in 1869 Mendeleev predicted germanium would be a metal doesn't seem to mesh with what was known about silicon at the time. Borrowing from our article on silicon, Thomas Thomson, in 1817, renamed silicium as "silicon" since, in his words (and with my best Scottish brogue engaged): "there is not the smallest evidence for its metallic nature, and as it bears a close resemblance to boron and carbon, it is better to class it along with these bodies, and to give it the name of silicon." I presume Mendeleev would've been aware of the nonmetallic nature of silicon by the time he made his prediction. On silicon having an amphoteric oxide, I gather from the literature that silicon is normally regarded as having a weakly acidic oxide. There is a note about this in the silicon section of the metalloid article. I had a look at the Davy paper on silicon but could not make sense of what the similarity was with oxides of aluminium and beryllium.
I did not have any tequila at hand so I settled for some refreshing whites, three walks, watching the AFL Women's grand final on TV, and some study time with musical accompaniment. And yes, this is a most absorbing and rewarding thread. Sandbh (talk) 10:19, 25 March 2017 (UTC)
Yes, I think that would be a fair portrayal. The fact that the metalloids need the "-oid" suffix suggests immediately that they're not metals. The need for an intermediate classification comes more, I suspect, from the fact that both the metals and nonmetals have their own characteristic properties, and if you use those characteristic properties instead of defining "nonmetals" simply as the complement of the set of "metals", then it's difficult to group cases like the standard nonmetals in one group or the other. If we consider the listing at Properties of metals, metalloids and nonmetals, for example, we see that the bulk of the properties of metalloids are reasonably distinct from the characteristic ones of the metals and the nonmetals. This would seem to suggest the validity of regarding them as a class apart, instead of taking the term "nonmetal" at face value.
On the "self-destructiveness" of the amendments to the periodic law, I tried to go for the most colourful and eye-catching language possible, and apparently went too far. ^_^ But this is what I was going for here: the periodic law is loosely constructed as applying to every element from H to Og, but then you have localised exceptions to it, and then the localised exceptions at some points harden into another loosely constructed law, except that these new laws also have exceptions. So every level of the "law" has an exception until the laws and the exceptions become indistinguishable. I did not mean to say that this reduces its efficacy, and indeed I do not believe it does. Like Mozart's art, Mendeleev's magnum opus turns into a conception of pure intuition, as one cannot consider so many factors at once: one merely feels it. To continue that comparison, surely Mozart did not learn the musical language of his time that he mastered so well by reading simplified accounts and principles and building them on each other; he must have learned it through listening and developing an intuitive sense of balance. Much the same sort of intuition seems to hold when "power users" use the periodic table as a predictive tool, exactly as you say. ^_^
This intuition does have a few consequences, because some of those amendments to the periodic law are pretty localised. It seems that when chemists think of criteria like cation formation to distinguish metalloids, they generally only think of the contentious region in the middle of the p-block. The fact that some of these criteria would literally exclude some of the heavier d-block metals from metallicity then gets overlooked.
So, what I would then ask is: if we are very sure that tungsten is a metal, which of its properties give us this confidence, given that its chemistry does not inspire us with this confidence? ^_^ We can then consider these other properties and use them to test the well-known behaviour of the metalloids (ignoring At, which we colour as one mostly because of the absence of experimental evidence of clearly metallic behaviour, instead of evidence of its absence).
What I was thinking of was the fact that Davy considers the oxides of Al, Si, Be (and, looking it up for real now, also Zr) in the same breath. He writes "Had I been so fortunate as to have obtained more certain evidences on this subject, and to have procured the metallic substances I was in search of, I should have proposed for them the names of silicium, alumium, zirconium, and glucium." I suppose Mendeleev may have noticed the somewhat borderline behaviour of Si (which does have a metallic lustre, after all), and expected metallicity to become more evident at Ge as an interpolation between Si and Sn. In fact, since there were no A and B groups yet, he may also have been swayed towards a metallic prediction for Ge by looking at Ti and Zr, also in group IV of his table. But ultimately, I suppose we can't know what he thought unless he wrote it down somewhere. Double sharp (talk) 13:44, 25 March 2017 (UTC)
  • The difficult part is to fomulate the question right, so I'll build it up. Sandbh proposeds to name this category "weak nonmetal (metalloid)" ['non' added, DePiep], because it 'captures the complexity of the situation with a high degree of consensus'. Now if we accept this new better, more descriptive(!) name, the question is: Would this name change the membership of the category?
Note that this is a different approach, namely from the category definition, not from individual element's properties. Most if not all of above arguments depart from an individual element's properties (set). While, when we have a better category definition, a membership test can and shall be performed from that definition.
Examples. Looking at the arbitrary dividing line between metals and nonmetals (very arbitrary of course, if only because it reduces the number of categories from three to two; not unlike the current proposal btw), it is likely that Ge and Sb are considered metals, and so will not be nonmetals (of whichever subspecification). So, if the new name is better and descriptive, the membership could change? (A curiosity then to be explained is, that these two elements score safely high on the 192 sources list). A contrary example I do not have, or maybe At while likely being an current-name 'metalloid' does not qualify as a nonmetal? -DePiep (talk) 15:50, 26 March 2017 (UTC)

Leftover nonmetals categorising[edit]

  • Leftover nonmetals categorising. This is more problematic, in that it does not succeed in making a (YBG-)convincing subclassification. It could be an improvement in that at least one class (into corrosion) has a wiki article compared to zero for both of todays categories. The magic thread Sandbh points out (I'd call a magic chain) links the elements by pairs only. This breaks an other ground rule for good classification: class elements should all be strongly tied within the category, and badly tied to outside the category. The chain does not do this (no. 3 is not tied to no. 1).
Maybe this is happening: if sub-categorisation of the leftover nonmetals is that difficult, absent or sought after, there probably is no subcategorisation. Keep them one category? Of course, all descriptions and notes made in the reasoning still are valid, and should have a place in the article leftover nonmetals (working title). -DePiep (talk) 12:07, 12 March 2017 (UTC)
Looking at all the elements that are not metals, there are three categories that almost populate themselves; noble gas; corrosive nonmetal; and weak nonmetal (metalloid). We know that the corrosive nonmetals are strong nonmetals and that the weak nonmetals (metalloids) are weak nonmetals. The remaining nonmetals are neither as strongly nonmetallic as the corrosive nonmetals nor as weakly nonmetallic as the weak nonmetals. They are in the nonmetal goldilocks zone—not too chemically hot, not too chemically weak. Effectively, they are intermediate nonmetals.
The magic chain of the intermediate nonmetals is not the primary thing that links them. What links them is their temperate or moderate nature, compared to the corrosive nonmetals and the weak nonmetals. Sure, rather than reading the chain as H → C → P → N → S → Se, you could read it as H → C; C → P; P → N; N → S; S → Se. I think this is cutting off your nose to spite your face. Even across groups 1 to 18, only the elements in groups 1, 3, 17 and 18 are strongly tied together, and group 1 becomes a little wobbly if it includes H. There are more groups that are not tied together strongly than there are groups that are tied together strongly. But I don't think this matters since organising the groups according to their valence states is what revealed the deeper relationships among the elements in the first place, including the diagonal relationships and patterns which become particularly prominent in the p block.
Yes, it is hard to subcategorise the leftover nonmetals i.e. those that are not noble, corrosive, nor weak (metalloid). There are only a few names that have been given to them in the literature and none of these are widely used. That does not mean we necessarily need to adopt an unfortunate category name like other nonmetal, when—consistent with the literature—I like to think that there are more generic, helpful, descriptive and preferably plain English words that we could try. I may be deluded though given how hard it seems to be to come up with something better than "other".
The proposed scheme fine-tunes our nonmetal categories in way that is more consistent with the literature, and retains and enhances the work of the founders of the Wikipedia periodic table colour categories. (The first scheme, from February 2002, was based on the LANL table, and the Environmental Chemistry table, here.) Sandbh (talk) 04:32, 20 March 2017 (UTC)
The trouble is that I do not see why the thread should stop there. If it can go from S to Se, why not from Se to Te? They're not all that different chemically. Simply put, there does not seem to be a very clear demarcation as to why the thread should contain some elements and not the others, that isn't rationalised from already having selected which elements to exclude from the other categories and put in the thread. Double sharp (talk) 04:16, 21 March 2017 (UTC)
The thread can be traced back as far as H → Li → Mg → Be → Al → B → Si → C → etc. It alternates between diagonal links (which are often forgotten) and vertical links, and since going onto Te will result in two vertical links, S is a natural place to stop. (It did not occur to me at the time to keep going onto Te, since S was an intermediate nonmetal and Te was a weak nonmetal). The fact that there is this thread that runs through all of the intermediate nonmetals wasn't something I realised until later. The natural part—like falling off a log when compared to the poly/di-atomic marathon—was dividing the nonmetals into the corrosive nonmetals (they speak for themselves) and the weakly nonmetallic metalloids (they speak for themselves, too). All that was left after that were the goldilocks nonmetals most of which have been labeled with the "it's-too-hard-to-work-out-what-to-call-them" category name of "other nonmetals". I didn't realise that they happened to be linked by diagonal and vertical links until I thought about what else they had in common, aside from being intermediate nonmetals. If I have made too much of a song and dance about the magic thread, I can only reiterate that the main game is about the contrast between the weak nonmetals (metalloids) and the corrosive nonmetals, and the remaining bystander or sandwich nonmetals (relatively speaking). Sandbh (talk) 13:13, 21 March 2017 (UTC)
Okay, but surely not all the diagonal links are relevant: surely no one thinks that fluorine has anything to do with argon. I don't think the diagonal links are often forgotten: the ones between Li and Mg, as well as Be and Al, are really quite well-known. What is forgotten is that the analogy may be pushed further (though there are limits; comparing Ce and Pa is rather nonsensical).
More to the point, the trouble I find is that the distinction is very fuzzy, and there is nothing like the precipitous drop in metallic behaviour that occurs when you cross past group 11 (there is another drop between very strong metals and milldy strong metals past group 3 and the Ln and An). There is a reason I raised selenium and tellurium: we split them into two separate categories, even though they are very similar chemically and are in fact very often found together: we shouldn't forget that selenium was named after the moon because it resembled tellurium, which was discovered earlier and was named after the Earth! They even form a continuous range of solid-state solutions in which Se and Te atoms alternate in the helical chains (since hexagonal Se resembles the main modification of Te in its structure).
So why do we put selenium apart from tellurium in the first place? Because Se has a number of properties that are somewhat uncharacteristic of a nonmetal which are only brought to maturity in the next member of the group. But already Se shows an advance on S in incipient metallic properties, so that if one is willing to call S–Se a valid link, then disallowing Se–Te immediately after that seems to show symmetry being taken too far (after all, S forms S8 molecules while Se and Te form long chains). S–Se–Te is quite a good triad for this "no man's land" between clearly metallic and clearly nonmetallic territory and it does not sit well with me to get rid of it for reasons like this. It is fine if we are applying some artificial sharpness in our categorisation by going for a hard line based on how many atoms happen to be in a molecule of the pure element (never mind that S and I seem a little out of place), but with fuzzier categories like this I am slightly more ill at ease. Double sharp (talk) 14:09, 21 March 2017 (UTC)
Indeed, there is no diagonal link between fluorine and argon and I did not allude to this. In my experience, the diagonal links are ignored---certainly in high school where the focus is on vertical similarities. Maybe things have changed since I went to HS; I tend to doubt it.
On the fuzziness of the proposed categories, and I hope I am not belabouring things, I did not make the distinction between (a) corrosive nonmetals and (b) metalloids, and in consequence of those extremes, the leftover nonmetals. The first two "categories" are the way they elements involved have been described in the literature whereas the leftover nonmetals don't attract such language. In this case I tend to think that the wisdom of the masses and its delineation of the nonmetals into "naturally sharp" categories is as at least as good as the artificial sharpness in our hard line categorisation based on how many atoms happen to be in a molecule of the pure element.
As noted, extending the magic thread to Te, breaks the alternating diagonal-vertical sequence. I'm not sure what joining a diagonal-vertical-repeat thread (an orange) onto a group-based vertical-repeat thread (an apple) would achieve (apart from causing me to raise an eyebrow).
We currently call Se a polyatomic nonmetal and Te a metalloid even though Te is a polyatomic "not-metal", because the literature tells us that Te is counted as a metalloid 4 times more frequently than is the case with Se (98% v 24% according to lists of metalloids). The fact that Se happened to be polyatomic was a fortuitous bonus that enabled us to keep it grouped with its other polyatomic colleagues. There is no distinction between Se and Te in an atomic structure sense, since both are polyatomic.
Nothing essential changes with the proposed scheme. Te becomes a weak nonmetal (metalloid) and Se becomes intermediate, consistent with distinctions made in the literature. Sure S forms rings whereas Se forms chains but S does form long chains in plastic sulfur. The S-Se-Te triad is still there, and its natural category division is in the same place. The fact that there is a cross-cutting thread fortuitously and curiously running through the intermediate nonmetals is something I happened to notice afterwards. (And the presence of the thread does not mean there is no vertical link between Se and Te), just like the polyatomic-diatomic line between the same two elements doesn't.) Sandbh (talk) 04:47, 22 March 2017 (UTC)
Postscript: A possible alternative way of mentioning or considering the other relationships among the intermediate nonmetals is to merely observe H → C; C → P; and N → S, and not say anything about P → N and S → Se, since the latter two relationships are already part of the periodic table furniture. Sandbh (talk) 06:51, 22 March 2017 (UTC)
Okay, but surely most of the diagonal relationships come with the periodic table furniture as well: they are one of those things that you learn as you read the thing, sometimes to the point that you intentionally break some things to show some points. I have sometimes naturally wanted to talk about aluminium as if it were above scandium instead of above gallium, for example (and when R8R Gtrs writes the Al article I am looking forward to reading about that similarity). We also know of the linkages between groups n and groups n + 10 that used to be considered A and B subgroups; the "knight's move" relationships stretching across groups 11 to 13; the way the early actinides "pretend" to be transition metals; and so forth. Essentially, you start with a first-order rationalisation, and then keep adding adjustments to it, like some chemical analogy for a Taylor series. Just because not all of this is taught in high school (although I remember that Li–Mg and Be–Al were emphasised, and B–Si was mentioned but not treated in so much detail) does not mean that they are ignored totally: it just means that we're not ready for it then, but might be later.
Certainly the extremes exist, and similarities between the elements that are not clearly in the extremes also exist, but because there is a commonality leading through all of them as you pass through adjacent cells on the periodic table I am not so sure that the "magic thread" is particularly convincing as an argument. It actually feels more convincing to me to say that these elements are just somewhere in the middle, without appealing to the thread. So I see the only difference between the current and the proposed classifications is where we put hydrogen and nitrogen, right? I would note though that both are actually vigorously reactive, except that you have to heat them up first, whereas sulfur does not even need the heating: whereas diamond is extremely unreactive and graphite needs rather violent temperatures to react even with fluorine. Meanwhile black phosphorus is not very reactive at all (the element's reputation comes mostly from the white allotrope) and is even a semiconductor. The difference is mostly whether you can find simple anions as a matter of course, but even then, At would form At, and with the reactive metals even Po forms "quasi-salts" like K2Po. Double sharp (talk) 09:50, 22 March 2017 (UTC)
I think we are in agreement on diagonal relationships, and adjusting from first order rationalisations as one goes. I do not claim the magic thread as a supporting argument, only as a nice to observe. You refer to the same thing that occurred to me which is that the extremes in nonmetal behaviour are already evident in the literature, and that naturally leaves the rest in the middle. (I don't know why this didn't occur to me when we were doing option 10 etc---in retrospect it was staring me in the face all along).
Yep, H and N are the only movers except that, by default and without doing anything, they end up in the middle. The decision outcome as to which nonmetallic element goes into which category is a composite of the many factors that contribute to, or are aspects of, a nonmetallic element's nonmetallic character, including some that you mention such as reactivity, and anion forming capacity. In the case of the corrosive nonmetals, it happens that their corrosive nature correlates quite well with their overall nonmetallic character. The work on assessing all of the factors that contribute to the nonmetallic-fu of the rest of the nonmetallic elements has already been done via a wisdom of the masses consensus in distinguishing between the metalloids and the remaining nonmetals. (I have a reservation about At as a metalloid, as mentioned, but there you go). Sandbh (talk) 04:14, 23 March 2017 (UTC)
The reason why I still think At deserves to be called a metalloid, even if it might form cationic At+ in solution, is because it is also homologous with iodine, which is important for its medical use. I should think that this kind of paradoxical, mixed chemistry of a combination of the halogen and the metal should deserve the name "metalloid". After all, we get halogen-like bonds to organic molecules like PhAt, that oxidises like iodine compounds to PhAtCl2, and the usual species like HAtO3 and H5AtO6 that remind me very much of iodine. I don't think cationic chemistry is enough to prove it as a metal; look at tungsten, for example. Double sharp (talk) 06:57, 24 March 2017 (UTC)
(Please excuse me by the way for not mentioning anything about H and N in this reply. There are two main reasons: the less important one is that I am still pondering what you have said here, and the more important one is that I do not wish to overwhelm you with so many long replies at once. ^_^) Double sharp (talk) 15:30, 24 March 2017 (UTC)
Thank you!
Regarding astatine, I think that's somewhat of a sideshow at least for now. I guess I can only blame myself since if I'm going to raise it I ought to do it well. But then I though it worth mentioning in passing.
Perhaps the key things to say about it might be (1) the recent prediction that the condensed form would be expected to be a (ductile) full blown close-packed metal; (2) the At ion is relatively easy to oxidise back to At (0) and there is some speculation that +1 might be the most stable state; (3) the extremely low concentrations at which its chemistry has been observed, and the possibility of reactions with all manner of contaminants (impurities, walls and filters, radioactivity by-products) and other unwanted nano-scale interactions; (4) “since the trace chemistry of I sometimes differs significantly from its own macroscopic chemistry, analogies drawn between At and I are likely to be questionable, at best" (I recall that's attributed to Kirby, writing in the Gmelin handbook on astatine); and (5) as a putative p block metal it could be expected to show significant nonmetallic character. (I caveat all of this lot by saying I haven't looked at the more recent literature on astatine.) Sandbh (talk) 11:26, 25 March 2017 (UTC)
The way I would see it is that while our picture of the chemistry of astatine is definitely fuzzy, it is the best we have at the moment: we shouldn't treat the theoretical predictions as being on a par with experimental results until they are verified (we're not giving astatine a "predicted" colour). So far, the properties known (admittedly on trace quantities) tally reasonably with that of a weird cross between a metal and a halogen, so calling it a metalloid seems reasonable given the current state of limited knowledge. When better data comes in, we can of course revisit this, like we did for polonium. All right, now off to formulate a reply to the bigger block of text. ^_^ Double sharp (talk) 12:08, 25 March 2017 (UTC)
Yes, no worries about astatine. I suspect it needn't impact on the merits of the proposal. It'll be ironic if the 1940 decision of Corson and his colleagues (the discovers of astatine) to classify it as a metal on the basis of its analytical chemistry, turns out to be true. Some 77 years later and we still don't know! That doesn't seem right to me but I have too many other things to do right now. Sandbh (talk) 01:49, 27 March 2017 (UTC)

The future of periodic table categorisation[edit]

(Paragraphs were moved here from [1] and [2]). -DePiep (talk) 19:31, 22 March 2017 (UTC)
  • I'm sorry to say what I'm about to say next. Considering how important classes are in chemistry, one would think that IUPAC would have gotten its terminological and conceptual house in order but no, it hasn't, so we're stuck with the mess that is metalloids and semimetals, while the physicists happily apply these words to their own, much better defined, physics-based uses. Poor show, IUPAC. Sandbh (talk) 12:28, 22 March 2017 (UTC)
Sure this is a task waiting for that IUPAC initiative. After groups, periods and blocks I think categories (we needed to invent this word too) are the 4th graphic structural visible in the periodic table (the rest is text in element cells). BTW, we could filter & funnel our threads into some Categorisation of elements by metallic-nonmetallic characteristics. -DePiep (talk) 16:59, 22 March 2017 (UTC)
  • Closer at home, there is this same classification issue in Wikidata. In short, an element hydrogen (Q556) does not have wikilinks at Wikidata, instead it has eg a property called "PT group". Then Wikidata structures this all by formal claims like "is a ..", "is part of ..", etc. This is called classification, or orthology, or categorisation. These class relations are being defined right now at Wikidata ("is the H atom: also element H, or a manifestation of element H?"). AFAIsee, our infoboxes are way more precise & correct. -DePiep (talk) 21:30, 22 March 2017 (UTC)

Other comments[edit]

  • Symmetry, and " 4 + 4 = 8 symmetry components is cool". I get the feeling, but please take care not to use an outcome as an input argument. "It is symmetric so it must be something good" is not sound reasoning. The periodic table is full of exceptions, we'd not want to miss one this way (by being misguided seeing a non-existent regularity). -DePiep (talk) 12:07, 12 March 2017 (UTC)
I partly agree. I guess that is why I put it in as a footnote. I suspect Mendeleev would have approved :)
While there are irregularities, the overall pattern in the periodic table of strong metals to weak metals and weak nonmetals to strong nonmetals is very well recognised (not forgetting the noble metals and noble gases). Sandbh (talk) 06:09, 15 March 2017 (UTC)

DePiep's comments are brilliantly insightful. I would add one more thing, though: I find our current use of "polyatomic" and "diatomic" quite felicitous, because it relates the molecular structure of the pure elements to their chemical behaviour (even in compounds), thus linking together two slightly different things. Double sharp (talk) 13:22, 12 March 2017 (UTC)

Hmm. I like the polyatomic, diatomic and monatomic division too. This nomenclature is a bit different from the other categories, which tend to be more descriptive. That is, alkali, alkaline = caustic. Ln/An = like lanthanum/like actinium (these two are a bit more obscure); transition = going from one thing to another; post-transition = coming after the transition metals (these metals used to be called B metals, which at least conveyed the impression that they weren't as buff as the rest of the metals); metalloid = metal-like (mind you, the first time I saw this word I didn't know what it meant); halogen = salt former; noble gas = relatively inert. Polyatomic and diatomic describe the structures involved but don't tell you anything about the character of these nonmetals. And showing the progression in metallic to non-metallic character going across the periodic table is a bit harder to grasp with these two categories than it is with the alternative proposal. It certainly is easier to find references in the literature to this progression based on a first order observation of the nature of the elements as strong, transitional, moderate or weak metals or nonmetals, than it is to have to drill down and find explanations of the same progression based on crystalline structures.
It might come down to comparing the two options…
  • polyatomic nonmetal ✦ diatomic nonmetal; or
  • weak nonmetal (metalloid) ✦ intermediate nonmetal ✦ corrosive nonmetal,
…and working out which one is more information-friendly?
If our role is to summarise topics with an eye to toward broad-based, literature-supported consensus on subjects, then I think the alternative proposal has more consensus credibility. There is consensus that:
  • relatively weak nonmetals are sometimes called metalloids
  • oxygen, fluorine and chlorine are the most buff nonmetals; bromine and iodine are not quite so bad-ass but still in the same ballpark
  • the remaining nonmetals (setting aside the noble gases) are described, more or less, using more moderate language.
There is consensus that diatomic and polyatomic nonmetals are diatomic and polyatomic but it is quite hard to find discussion in the literature on their other shared properties.
And if the way chemists tend to write is a reflection of the way they think, then they will think about the chemistry of different nonmetals in terms of where each nonmetal approximately lies on a composite scale (or spectrum) of nonmetallic "liveliness" or "intensity" rather than in terms of polyatomic or diatomic nonmetals. At one end of such a scale lie the corrosive nonmetals; at the other end are found the weakly nonmetallic metalloids. Sandbh (talk) 04:32, 20 March 2017 (UTC)
  • Let's not forget, before we get used to these appearances: this is once again a wonderful great overview of element property & behaviour by Sandbh. Well researched & sourced, good to read, scholarly level. -DePiep (talk) 18:51, 12 March 2017 (UTC)
Thank you!, DePiep and Double sharp. Sandbh (talk) 04:32, 20 March 2017 (UTC)

Antimony: Most stable oxidation state[edit]

The antimony article says that, "The +5 oxidation state is more stable." Is that right? I do know that the +3 oxidation state in Bi is more stable, and that Sb(V) shows less tendency to revert to (III) than P and As. Sandbh (talk) 02:58, 16 March 2017 (UTC)

Sb shows no reluctance to be oxidised to the +5 state, unlike As (another side effect of the 3d contraction). I would note that SbCl5 exists and is reasonably stable, while AsCl5 leads only a fugitive existence and even BiF5 is vigorously reactive. I am not sure that I would say that +5 is "more stable" but I would have no problem with saying that like P, both +3 and +5 are important oxidation states for Sb. Double sharp (talk) 03:42, 16 March 2017 (UTC)
I am trying to do two things at once, so don't have time to to respond yet other than saying thank you and that I found a quote from an old piece of work I did once: "The higher valency state (+5) is the more stable in arsenic and antimony… The reverse is true of bismuth… Bi+5 compounds are uncommon." (Steele D 1966, The chemistry of the metallic elements, Pergamon, Oxford, p. 69). Perhaps that is where it comes from. More later. Sandbh (talk) 04:34, 16 March 2017 (UTC)
A more important question for that article: why isn't there anything about Sb cluster compounds, since the As and Bi ones are pretty well-known and get a mention in the articles for those elements? (Because they don't fit the chosen classification into Sb(III) and Sb(V) compounds, maybe?) Double sharp (talk) 03:47, 16 March 2017 (UTC)

Recoloring our PT[edit]

We had a great discussion on changing colors of the current categories about a year ago. I re-discovered my suggestion a few days ago and found it great. DePiep also had a draft for one but it at the moment didn't comply with the web design color criteria he introduced me to (DePiep acknowledged this, saying this would be worked on in a later revision).

Is anyone still interested in developing a new color scheme? I'd want to reignite the discussion for now if that's possible.--R8R (talk) 11:16, 17 March 2017 (UTC)

It's on my todo list for over a year, I'd very much like to redesign that. Colors we had have drawbacks, maybe full restart needed. Colorbrewer is a nice inroad. (Just finished an excercise with YBG for this). -DePiep (talk) 12:19, 17 March 2017 (UTC)
Could you explain to the amateur that I am what's wrong with them? I'd see if I could contribute some ideas for improving this if I knew what to do.--R8R (talk) 16:33, 19 March 2017 (UTC)
User:DePiep/pt-2016 is where we were at last year. I just edited them with Sc-Y-La-Ac and the new element names for Nh, Mc, Ts, and Og. Double sharp (talk) 16:09, 19 March 2017 (UTC)
That version is still far from it in terms of color contrast according to the standards DePiep introduced me to. For one, red on orange for nonmetals gives a contrast of 2.97 whereas we aim for 4.5. I resolved this at the price of having a not-so-red red, which I find to be a poor solution. I just got a new idea and I'll give it a try. Want to hear from DePiep on bad colors, though.--R8R (talk) 16:33, 19 March 2017 (UTC)
@R8R Gtrs and Double sharp: My late reply on these individual points, and why I think we should restart design from blank (so skip January 2016 results, mostly). My general reply is already below, aiming at a grander approach. We know that category-coloring has nothing to do with the categorisation itself, it's just putting a color on a category 1:1 (unless science requires more categories than current 10—we'd be in trouble).
Hmm, that might be a problem when element 121 gets discovered in the near future and we start the superactinides row... ^_^ Double sharp (talk) 05:27, 23 March 2017 (UTC)
Yes, that's right. I've had this in mind. It's only two elements away from what we already have so may be expected to appear in a reasonable period of time.--R8R (talk) 19:21, 23 March 2017 (UTC)
Sort of spoiler. But yes, that should be covered indeed. Will try to add this as an outlier (instead of treating as a straight color #11 to be available: too many restrictions). Or maybe ask Sandbh to rewrite the category scheme by then ;-). -DePiep (talk) 09:52, 24 March 2017 (UTC)
We would have a bit of time before it appears on the standard (non-extended table), because surely it wouldn't be chemically characterised for a while longer: meitnerium was synthesised in 1982 and is still waiting for a chemical characterisation 35 years later. Of course everyone is sure that it will be a good eka-iridium, but we do not know it yet.
Something I would wonder about, though, is how the PT would look as the eighth period is illuminated. I think we can just put in elements that have been discovered, even if it leads to a "gap" in the brick wall like tennessine was for a while. It'll be nice to see the bricks being put on, one by one. (There is no conflict between the various models from Fricke, Pyykkö, and Nefedov until we get to element 139; by the time that happens, presumably there will be a better model. ^_^) Double sharp (talk) 15:21, 24 March 2017 (UTC)
We'll add one category color (11th) for this, with reduced requirements. Once we get to element 139, we don't need more elements but we would need more colors (extended rainbow). Come to think of it, is it correct that actinides and lanthanides are a different category when categorised correctly? (And having no border exceptions at that). Or is this an old split into some indiscriminate 'series', e.g. from discovery history? (If they are actually the same category, we could give them the same color. And, of course, superactinides too). -DePiep (talk) 15:37, 24 March 2017 (UTC)
There is a name for the union of lanthanides and actinides ("inner transition metals"), but it is not used very often. Almost all descriptive inorganic chemistry textbooks I know of classify them separately, and I think the main reason for that is that the average Ln behaviour is quite different from the average An behaviour. (If you actually took all the actinides into account it would be a lot closer, but because the only actinides people usually work with are Th and U, and those do not behave at all like lanthanides, things are set further apart. In general, Th–Am and borderline Cm don't act like lanthanides, and all the rest do.) Double sharp (talk) 15:53, 24 March 2017 (UTC)
We all know what is wrong with the 2013 colors, those we show today. About every proposed January 2016 scheme would be an improvement that kills those bads (text contrast, five reds one green one yellow one blue, extreme teinte changes, grey and brown used, color blindness effects). If we propose a change to say any latest Jan 2016 proposal, that could win an RfC. Then we'd loyally change our 100+ colored templates & legends & images accordingly of course. But please do not propose that. Because: our Jan 2016 proposals have flaws too, they are not perfect. Not perfect.
And I want a perfect color scheme. One that is sound for our enwiki categories of course. But not only that. Also for other lang:wikis (who sometimes happen to copy our images & templates, blindly in trust). Wikidata should follow (!). Our commons images are used published outside of Wikipedia. They should be perfect in this. en:wiki should be the goto-site.
Also: there is off-internet. Printing a PT from us? Should be perfect (print the large-cell form as classroom wallpaper, why not). CMYK colors, for professional printing? Go enwiki. Beaming a WP page with a PT? Needs attention (beaming makes colors lighter compared to screens). Did you know today a table is not rendered into a Wikipedia-into-pdf at all?
And not in the least: in the professions. Broadcasting a perfect color scheme, will make our colors the standard colors. Think scholars both chemical and physics. IUPAC needs guidance in this. -DePiep (talk) 21:01, 22 March 2017 (UTC)
I understand your noble goal and want you to keep on. You have my greatest sympathy in this.--R8R (talk) 19:21, 23 March 2017 (UTC)
Quick re (playtime ahead): Recently I worked with YBG towards this result. Intense, and very fruitful. See especially this table for the development. (Notes: only two editors were involved; just nine colors to solve—the PT needs ten+unk; bg-colors only, not yet text-contrast issues; and in between we did solve that Nixon-Cheney 36/46 issue).
I found some new improvements I want to introduce here: (and its background research book) is very smart (in map coloring, so it is closely related. Nice to play with the options, read about colorblndness, and missing option 10...).
I also came up with my "al-most opposite color" theory. Will arrive here!
Will reply to (describe carefully) the topics mentioned (including contrast rules, and degrees-of-freedom).
Want to set this standard once and for all (though maybe not for ever). -DePiep (talk) 20:52, 19 March 2017 (UTC)
I'm already excited to see what you have there. Waiting for updates from you!--R8R (talk) 21:38, 19 March 2017 (UTC)

Periodic table form for first introduction[edit]

I think it worth considering, to introduce the periodic table initially in the left step form, and have all classrooms wallpapered with it. Instead of the crumbling-castle like figure.

  • The advantages are, that the structure showing is much more regular (or just: regular). It nicely shows the shell-filling, inviting some pupils to ask: "why is that step two-rows high?" (send that pupil to a university for chemistry or physics).
  • Of course the valence-ordering by I-VIII-0 groups is lost, partly. Ordering the groups by valence (1–8) of course was the original building principle for periodicity, being the discovering's entrance before electrons and orbitals were known (in 1871 and ever since the columns were already split into series A and B by M.—but is that really helpful for a first introduction btw?), but that does not say today we must introduce the periodic table that way. And for those who adhere to and 18-column form (any 18-column form): the f-block still nicely can be moved to the bottom again, where the irregular "f-block filling" can be avoided by omitting column headers.

How is the periodic table introduced anyway? I am open for teachers who say that one form or the other is way too complicated to teach. Note: this post is to datastamp this idea, although I probably am not the coining one. -DePiep (talk) 11:55, 21 March 2017 (UTC)

Well, the chemist's answer is that helium behaves nothing at all like beryllium. I would also add that this form makes the most salient divide look like it is between groups 2 and 3, when in reality it is at group 18 as the eye of a hurricane that envelopes groups 17 and 1. But the main reason is of course valence, since the standard format (if you ignore the d-block) lets you show a cute little picture like:
H He
Li Be B C N O F Ne
Na Mg Al Si P S Cl Ar
K Ca
which shows you the first 20 elements lined up under the number of electrons in their outermost chemically active shell, taking 0 for the inert cases of He, Ne, and Ar. (Yes, yes, I know Ar can form a compound, but we're not going to tell them that yet.) We don't teach people to run before they can walk, so we start here first before going on into other elements and trends. Even then, we gingerly thread down the groups with the most similarities (I, II, VII, and 0) or the one with elements that everyone's heard of (IV), and then gingerly show the first row of transition metals to see what lurks beyond the first 20, enlarging it to 30. In fact, in some cases we might not even teach spdf until after the basics of the octet rule are finished (and first handwaving it as "there are exceptions and you'll learn them later"). It's only much later that we go to the rest of the 4th period, and most of the 5th, 6th, and 7th lie in darkness even when high school is finished. Everything needs to be progressive, after all! Double sharp (talk) 09:26, 22 March 2017 (UTC)
Ah, so that's how to introduce it all. Then at some point, when in a classroom I'd like to get more familiar with the physical trends and shell filling and aufbau (I probably skipped that in highschool, this is my punishment). DePiep (talk) 17:09, 22 March 2017 (UTC)
IIRC Aufbau and the shell filling came the next year. With the first 20 elements (and a few examples of homogeneous groups like I, II, VII, and 0) we were made aware of the trends, along with the metal–nonmetal divide and the admittedly fuzzy but mighty useful division of bonds into ionic, metallic, and covalent. Then we later revisited them in more details (I remember going through group VII in way more detail in the last year of high school; I'm not saying it was as comprehensive as I've made the WP articles for Cl, Br, and I, but a significant amount of what you see there I do remember, IIRC. ^_^) But this is from my own memories, so it might not be done now, and might not be done the same way elsewhere. Double sharp (talk) 01:56, 23 March 2017 (UTC)

On the standard atomic weight[edit]

I am working to understand and improve the standard atomic weight and its related/unrelated quantities. Aim is to publish the right values with the right quantities (definitions, names, numbers, units, values, nonconfusing).

Main terms
  • relative atomic mass (r.a.m.): physical quantity of a sample (any sample), reflecting isotopic composition of a single element in that sample. Dimensionless ('no unit', aka 'unit=1').
  • standard atomic weight (s.a.w.): a r.a.m. for specific samples, as published by CIAAW. Specific samples: must be terrestial, natural, stable wrt radioactivity. Most commonly published value, especially in periodic tables. 84 elements have one.
Ar, standard (quantity symbol, my way of writing following SI). Examples:
helium: Ar, standard(He) = 4.002602(2). The "(2)" notes the uncertainty in the last digit (so read: 4.002602 ± 0.000002.
hydrogen: Ar, standard(H) = [1.00784, 1.00811]; an interval. No uncertainty added. Is not the same as an uncertainty range (rather, the sources differ systematically, eg H from ocean water and from air).
Not published by CIAAW. Mainly used for the 34 instable elements that therefor have no s.a.w. (Tc and heavier metals from and beyond 84Po).
Well, the full list to be precise is Tc, Pm, Po–Ac, and everything from Np onwards. Double sharp (talk) 14:36, 22 March 2017 (UTC)
Mass number: 210Po = 210.
  • Terms to be avoided: relative atomic weight, standard atomic mass, atomic weight (confusing)

Derived from standard atomic weight:

  • conventional atomic weight: single s.a.w. value for interval. CIAAW published. Example:
Ar, standard(H) = [1.00784, 1.00811]Ar, conventional(H) = 1.008
  • abridged atomic weight: s.a.w. value, rounded to 5 significant figures maximum, has uncertainty or is again an interval. CIAAW published.
Ar, standard(He) = 4.002602(2)Ar, abridged(He) = 4.0026(1) (note: "(1)" is usually omitted, but not "(2)").
Work in progress
Main: Use "standard atomic weight" and its value where possible, and not where not applicable. The CIAAW value is sacred. Root out wrong or misunderstood usage.
Standard atomic weight: new article, should cover CIAAW aspects. Started with text copied from Relative atomic mass, to be teared apart.
In Wikidata: trying to have properties accepted. proposal:Property s.a.w., proposal:Property r.a.m.. Not an easy exercise.
Published table: "Standard Atomic Weights". Commission on Isotopic Abundances and Atomic Weights (CIAAW). 2015. Retrieved 2015-09-20. 
Technical Report 2013: Meija, J.; et al. (2016). "Atomic weights of the elements 2013 (IUPAC Technical Report)". Pure Appl. Chem. 88 (3): 265–91. doi:10.1515/pac-2015-0305. 
CIAAW update 2017: s.a.w. of ytterbium was formally changed into 173.045(2).
Questions, todo
  • Improve core articles
  • Use {{val}}, produces spaced numbers: 4.002602(2).
  • Clarify which term to use for non-s.a.w. (instable elements).
  • What to have in our infoboxes? Which labeltext, data values. s.a.w.: always when exists (84×), Conventional: sure; abridged: no. What with instable elements (that have no s.a.w)?
    • Unstable elements use the mass number of the isotope with the longest half-life in square brackets. So Tc gets "[98]" and Pm gets "[145]". Although what I might want to do is to put a little table in the article (not the infobox) giving the Ar of the most common ones (Po would have 207.9812, 208.9824, and 209.9828 for 208,209,210Po respectively). Double sharp (talk) 09:15, 22 March 2017 (UTC)
So 'Mass number' should be the labeltext in these situations, definitely not s.a.w. Maybe a minor explanation with it (isotope symbol).
I think the three values you give are not an Ar (dimensionless, and from actual samples) but are an isotopic mass or a relative isotopic mass (per specific isotope, no averaging done). Could be a true mass (in Da) or relative (dimensionless when divided by 1 Da). btw the isotopic mass (in Da) is in Isotopes of Po#table. -DePiep (talk) 09:46, 22 March 2017 (UTC) -DePiep (talk) 14:03, 21 March 2017 (UTC)
Good catch, though in this specific case I do mean Ar. This is because a realistic sample of Po will usually only contain one of the isotopes: you select for them by deciding what to bombard your bismuth target with, and at what energy, and then one will dwarf all the others in the amount produced. I agree with you, though: we should probably not label them as such in general. Double sharp (talk) 14:36, 22 March 2017 (UTC)
(Of course, I was not out for a 'catch', I am studying the topic and testing my understanding ;-) ). -DePiep (talk) 19:44, 22 March 2017 (UTC)
About the new isobox, developing. For isotopes of oganesson I made and added {{infobox oganesson isotopes}} (a full copy from {{infobox oganesson}}). Intention is to add this to oganesson#isotopes section too (and remove table from main infobox). When done, those isotopes masses could be added to that box (Oganesson is a trivial example with just one isotope). -DePiep (talk) 09:56, 22 March 2017 (UTC)
Hmm, but I do think that I would want some consistency in what the infobox contains, and it seems to me that the visible elements should set the tone, no? The isotopic abundances are an important properties of the element as it naturally occurs and so I think the major isotopes should be in there.
One thing I do want is an extra column squeezed in for nuclear spin, which is very important for NMR. Double sharp (talk) 14:32, 22 March 2017 (UTC)
Will be back on this later. I am making this much to complicated, raising too many issues at once. DePiep (talk) 17:16, 22 March 2017 (UTC)

Proposed changes[edit]

technetium,  43Tc
Current Red XN
Standard atomic weight (Ar) [98]
a. Mass number (A) 98
b. Mass number (most stable isotope) 98
c. Mass number (most stable isotope) 98Tc: 98
d. Mass number of most stable isotope 98
  • About the instable elements, and their infobox. These elements do not have a standard atomic weight, so we should not use that label (label=the lefthand text in infobox). Also, the bracketed notion [98] is not clear of what it says, may be a own invention(?), and square brackets are also used formally for the interval notation of other elements (hydrogen: [1.0084, 1.00811]).

The new label should have "Mass number", which has symbol A and is dimensionless. Adding "(most stable isotope)" would be the clearest descriprion IMO. Quite trivial, but also very clear is adding the isotope: "98Tc: 98". (a, b, c are added for reference here only).

Comments? Other variants? -DePiep (talk) 13:20, 23 March 2017 (UTC)

It can't be our invention if Greenwood & Earnshaw also use it for Po and At, but yes, "mass number of most stable isotope" would work, and it is exactly what they use for the actinides (except of course Th, Pa, and U). Double sharp (talk) 13:27, 23 March 2017 (UTC)
Added d. Admit, 'own invention' is a bit charged but it is not a regular format and requires description somehow. The proposed labels solves this. In a periodic table, things are different. -DePiep (talk) 14:01, 23 March 2017 (UTC)
Yes check.svg Done b., See technium. -DePiep (talk) 22:20, 26 March 2017 (UTC)

magnesium,  12Mg
Standard atomic weight (Ar) 24.305 (24.304–24.307)[1]
a. Standard atomic weight (Ar) [24.30424.307][1]
conventional 24.305
b. Standard atomic weight (Ar)
  • [24.30424.307]
  • conventional: 24.305[1]
c. Standard atomic weight (Ar)[1]
  • [24.30424.307]
  • conventional: 24.305
d. Standard atomic weight (Ar)[1] [24.30424.307]
conventional: 24.305
e. Standard atomic weight (Ar) conventional: 24.305
f. Standard atomic weight (Ar) N conventional: 14.007
  • Twelve elements have an interval as their standard atomic weight value (which is not an uncertainty range). For these, CIAAW also publishes a "conventional" atomic weight for more practical use like in trade and commerce.

The interval is written, by SI convention, with square brackets, comma, and a space. We should follow that. Then, I think it clarifying to add "conventional" to the second value. The source ref (CIAAW) could go with the label? Example e. has the order reversed. Examples b and c are using {{unbulleted list}}, d and e have a <br>. The data being in two rows, one should also check mobile view (makes a. look bad). f is nitrogen. Comments? -DePiep (talk) 15:38, 23 March 2017 (UTC)

FYI: The sqaure brackets is the mathematical notation to show its an interval. [a, b] means from a to b including a and b. ]a, b[ means from a to b excluding a and b. (a, b) means the same as ]a, b[. Christian75 (talk) 15:52, 23 March 2017 (UTC)
Is what I tried to say. So current writing (a–b) is wrong. CIAAW uses this because there is variation between sources, not uncertainty. btw, any preference? -DePiep (talk) 16:09, 23 March 2017 (UTC)
a-b is ambiguous because you do not know if a and b are included or not. But most people understand the notation. Not many knows the meaning of [a, b], and I think the current infobox for Mg: "Standard atomic weight (Ar) 24.305 (24.304–24.307)" will confuse most people because its an atypical way to express Ar. Maybe some "see text", or an explaining footnote. Christian75 (talk) 16:44, 23 March 2017 (UTC)
Agrees. The problem for a reader is not the notation, but it being an interval at all. "Whhhat is that supposed to be?". So once we want that interval value showing (and here we want it), we better stick to the formal notation (and as CIAAW publishes it). The reader is helped by the nearby 'conventional' plain number, and the wikilink standard atomic weight. I note that that article was split from relative atomic mass only a week ago, to improve this explaining (for myself at least). A footnote I don't think applicable (is not article-specific enough). -DePiep (talk) 18:12, 23 March 2017 (UTC)
Yes check.svg Done b., see hydrogen, helium. -DePiep (talk) 22:20, 26 March 2017 (UTC)



Why I periodically write about the elements on Wikipedia[edit]

Thought you might be interested. A couple of weeks ago, I was contacted by Wikimedia staff member Ed Erhart; he asked me if I would write an article for Wikimedia blog about why I keep writing articles about the elements, to which I agreed. Just very recently, it went live and you're free to give it a read :)

BTW sorry for not being able to take part in the discussion of the scheme proposed by Sandbh. I have a lengthy reply in mind but I am unable to write it down for the next couple of weeks (or more) due to lack of spare time.--R8R (talk) 08:17, 22 March 2017 (UTC)

Ye gods, the fluorine article really did suck back then, didn't it? ^_^ And thanks for the mention!
I am very much amused by your description of why you chose F. In fact, it was for much the same reason that I went for alkali metal (well, that and the thought process "why write on all six individually, when I can just put them all together?"). Then I kept finding more stuff and more stuff and it's become incredibly long. I would note by the way that doing what we did has become a lot harder for the element articles (too many have been done), except maybe in the lanthanides (which are really boring to do: maybe I'll finally dust of praseodymium, which is mainly notorious for its impossibly long name). The ripest pickings for those who want to do this kind of thing are probably group articles: alkaline earth metal and halogen are good choices (but certainly not any of the others). I should also take this opportunity to clamour for the much-anticipated Sandbh rewrite of transition metal. ^_^
And indeed, despite being the closest thing we have to a resident Stakhanov at the moment (who nevertheless sees fit to disappear for a few months at a time), I fully appreciate that we can't possibly do everything all by ourselves. Double sharp (talk) 09:12, 22 March 2017 (UTC)
Having worked on some of your (R8R's) articles it is remarkable how much your Wikimedia article adds to you as a real person, rather than someone I usually think of as R8R Gtrs. That was a good read. I shall wait patiently until you are ready to provide some comments on my proposed scheme (and for the return of YBG). Meanwhile I think I'm still learning about the scheme (= sharpening my approach) in the course of my chat with Double sharp. The transition metal article rewrite is still there in the background, as is some more work DS and I have been asked to do on the Group 3 question, and then there is the lead FAC to get ready for again, and I have another RL commitment coming up soon, followed possibly by further university studies, so who knows when I'll get round to everything. No matter. Sandbh (talk) 11:49, 22 March 2017 (UTC)
Indeed, please take my comments as an attempt to focus the spotlight on the issue! Regarding group 3, I have also been mulling about the block-definition thing (and how there does not seem to be an agreement as to exactly how the blocks are defined): you will see this when I feel it is ready for your comments (which is a status it is not anywhere near yet ^_^).
P.S. to R8R: "Soviet-styled" is now in my mental dictionary as a synonym of "the highest quality of pure scientific scholarship". ^_^ Double sharp (talk) 12:12, 22 March 2017 (UTC)
LOL! Sandbh (talk) 12:32, 22 March 2017 (UTC)

I like how you mention Au as a far-future project, BTW. It is another of those things I have wanted to tackle in a group, but have a phobia of doing alone.

Silver is a good example of this sort of phobia: the first three sections, dealing with the properties of the element, are fine, but doing the rest of the article absolutely terrifies me. I got over it for iron mostly because there was already a good base there and all I had to do was to find citations. I will definitely need a PR for this to continue further into applications, history, and the cultural background. Double sharp (talk) 15:15, 22 March 2017 (UTC)

  • A pleasant, interesting and inspiring reading, R8R. -DePiep (talk) 14:41, 23 March 2017 (UTC)

Calcium, strontium, and barium[edit]

Has anyone actually ever seen any of these three elements (except Sr in the fabulous Wikipedia picture) without an oxide layer? Only in the aforementioned fabulous WP picture do you see the "pale yellow" colour that Greenwood and Earnshaw refers to on page 112. (Same goes for europium and ytterbium.) Double sharp (talk) 15:10, 22 March 2017 (UTC)

  • "Nodules of very high-purity, completely un-oxidized calcium metal in an argon-filled ampule", here
  • "About the purest barium you'll ever see", here

That what you had in mind? Sandbh (talk) 02:36, 23 March 2017 (UTC)

Yes, yes, ten thousand times yes! But we'll have to ask him for permission to use them... Double sharp (talk) 03:08, 23 March 2017 (UTC)
Oh, I see. There is one other source I can check. Sandbh (talk) 04:31, 23 March 2017 (UTC)

Pictures for synthetic elements[edit]

Obviously, it is currently impossible to take a photograph of a sample of one of those violently radioactive elements past 100. I went through some of them and added some relevant pictures: most of them already had some (namesakes), but for the few that are named after places instead of people there are other solutions. (The reason I did this is that without some cool and relevant pictures like this the article seems very dense and impenetrable to the lay reader.) R8R already gave Db the map (though I'd have liked to see something in the city, like the JINR building); I put a Hessian festival of culture for Hs, and because the JINR pioneered cold fusion I chose the Russian chapel in Darmstadt for Ds. ^_^ Nh has the picture of the discoverers; Mc now has Red Square; Ts now has the main campus of Vanderbilt (Hamilton's institute).

Now what I am curious about: is there a better picture of Oganessian to use for Og? The one we have is currently a few years old and isn't particularly high resolution. Meanwhile, Theodore Gray has a much better one in colour on his poster. Double sharp (talk) 04:45, 23 March 2017 (UTC)

WP:CHEMISTRY/WP:CHEMICALS shorcut updated[edit]

Note that per this RFC, the shortcuts to WP:CHEMISTRY/WP:CHEMICALS have been updated.

Old discussions have had their shortcuts updated already. If I have made a mistake during an update, feel free to revert. Headbomb {talk / contribs / physics / books} 15:57, 23 March 2017 (UTC)